Harold Garfinkel: Parsons' Primer [1st ed. 2019] 978-3-476-04814-1, 978-3-476-04815-8

Table of contents :
Front Matter ....Pages i-v
Introduction to Parsons’ Primer (Anne Warfield Rawls, Jason Turowetz)....Pages 1-108
Parsons’ Primer (Harold Garfinkel)....Pages 109-326
The Program of Ethnomethodology (Harold Garfinkel)....Pages 327-338
Back Matter ....Pages 339-393

Citation preview

CONTRI BUTIONS TO PRAXEOLO GY

Anne Warfield Rawls (Ed.)

Harold Garfinkel: Parsons’ Primer

Beiträge zur Praxeologie / Contributions to Praxeology Series Editors B. Karsenti Paris France E. Schüttpelz Siegen Germany T. Thielmann Siegen Germany

Die „Beiträge zur Praxeologie / Contributions to Praxeology“ setzen sich zum Ziel, die Praxis allen anderen Erklärungsgrößen vorzuordnen, und die theoretischen Grundbegriffe aus dieser Vorordnung zu gewinnen, zu klären oder zu korrigieren. Sowohl die Arbeiten von Wittgenstein als auch die von Schütz und Garfinkel verweisen auf eine gemeinsame mitteleuropäische Genealogie der „Praxeologie“, die bis heute allerdings weitgehend unbekannt geblieben ist. Die Reihe will sich daher in drei Stoßrichtungen entfalten: durch philosophische Theoriearbeit, durch empirische Beiträge zur Theoriebildung und durch Beiträge zur Revision der Wissenschaftsgeschichte. Die Bände der Reihe erscheinen in deutscher oder englischer Sprache.

More information about this series at http://www.springer.com/series/15976

Anne Warfield Rawls Editor

Harold Garfinkel: Parsons’ Primer With an introduction by Anne Warfield Rawls and Jason Turowetz

Editor Anne Warfield Rawls University of Siegen Siegen, Germany

Supported by the German Research Foundation (DFG) through the Collaborative Research Center „Media of Cooperation“ (SFB 1187). Beiträge zur Praxeologie / Contributions to Praxeology ISBN 978-3-476-04814-1 ISBN 978-3-476-04815-8 (eBook) https://doi.org/10.1007/978-3-476-04815-8 J.B. Metzler © Springer-Verlag GmbH Germany, part of Springer Nature 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Cover Illustration: Finken & Bumiller, Stuttgart (Foto: © M-image) This J.B. Metzler imprint is published by the registered company Springer-Verlag GmbH, DE part of Springer Nature. The registered company address is: Heidelberger Platz 3, 14197 Berlin, Germany

Contents

Introduction to Parsons’ Primer . . . . . . . . . . . . . . . . . . . . . . Anne Warfield Rawls and Jason Turowetz

1

Parsons’ Primer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Harold Garfinkel The Program of Ethnomethodology . . . . . . . . . . . . . . . . . . . . 327 Harold Garfinkel The Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339

v

Introduction to Parsons’ Primer Anne Warfield Rawls and Jason Turowetz

Harold Garfinkel gave the title Parsons’ Primer to a manuscript of nine chapters that he wrote from 1959 to 1963 to explain the importance of Talcott Parsons’ social theory and the significance of his move toward social interactionism in his later work. The manuscript is important for at least four reasons: First, written by Garfinkel in response to discussion and correspondence with Parsons (while teaching a seminar on Parsons in 1959 and again in 1963 at UCLA), it reflects a sustained but unknown dialogue between the two scholars about culture/interaction in their respective positions (Rawls and Turowetz 2019); Second, throughout Garfinkel maintains that Parsons was misunderstood because he put interaction (and culture as interaction) at the center of his social theory, while few scholars realized that he had done so; a problem that was exacerbated by a wartime turn against the study of social interaction in US sociology (Rawls 2018); Third, the analysis challenges conventional readings of both scholars, showing Parsons to be more sophisticated and oriented toward interaction, and Garfinkel more indebted to him, than generally understood; and, Fourth, in the manuscript we find Garfinkel offering his own research on “ethno-methods” as a demonstration of Parsons’ position. In essence, Garfinkel credits Parsons with taking a social contract position that inverted Hobbes. Whereas Hobbes and the social contract philosophers, who followed him, had argued that rational beings would see the necessity of making a social contract as the only way to avoid living in a perpetual state of war, Parsons proposed that only in the context of a prior and implicit social contract can humans function as the “rational individuals” that Hobbes’ position assumed. Parsons was introducing Durkheim’s “constitutive practice” References to parts of the text within this book are indicated only by page number. © Springer-Verlag GmbH Germany, part of Springer Nature 2019 A.W. Rawls (Ed.), Harold Garfinkel: Parsons’ Primer, Beiträge zur Praxeologie / Contributions to Praxeology, https://doi.org/10.1007/978-3-476-04815-8_1

1

2

A. W. Rawls and J. Turowetz

position to US sociology (Rawls 2009, 2012)1 with a twist that incorporated Weber’s conceptions of legitimacy and mutual orientation, while putting implicit social contract at the foundation of a theory of social order in which “systems of interaction” ground an approach to culture as “assembled stable social structures of interaction”, with a focus on the assembly. As Garfinkel explains it (Chapter IV), Parsons’ initial contribution to sociology in the 1930’s was to posit the need for a moral commitment to an underlying social contract and its guarantee of sanctions, as a precondition for the possibility of rational actors, meaningful objects and meaningful actions. Parsons’ move toward culture as an independent dimension of social action in 1949 and his formulation of interaction in terms of “double contingency” in 1951 were moves toward a comprehensive theory of interaction. Garfinkel’s elaboration of the “Trust conditions” that are required to ground this conception of “culture as interaction” built on Parsons’ foundation.2 While Garfinkel was also inspired by Schutz’ conception of “taken-for-granted” practices and Wittgenstein’s notion of “language games”, and made extensive use of both – he used them to supply filler for the theoretical architecture of the interactional system derived from Parsons – as neither treated constitutive practices in terms of a “system of interaction” that was fully sociological in the sense that Garfinkel required. This is not to say that Garfinkel adopted Parsons’ position, but rather that he tried to modify and improve it – and to do so with Parsons’ approval. The 1

A constitutive practice is one in which meeting constitutive criteria is required to accomplish the practice. Durkheim introduced the idea in the Division of Social Labor (1893) that constitutive self-regulating practices could enable modern work and science to escape the boundaries of belief. The legacy of this idea was passed from Durkheim through Parsons to Garfinkel (Rawls 2019). But, it has been essentially lost to mainstream sociology. 2 What Garfinkel meant by Trust is not the ordinary meaning of the word. For that reason the word will be capitalized whenever it bears Garfinkel’s meaning. Garfinkel had been sketching what became known as the “Trust argument” and Trust conditions in various forms since 1946 and had presented it in various forms in unpublished manuscripts and conference talks (Turowetz and Rawls 2019). The first published version of the argument in 1963 was titled “A conception of and experiments with “trust” as a condition of concerted stable actions”. The argument maintains that participants in social settings must meet three constitutive criteria in order to make sense together. Constitutive criteria are necessary and defining criteria of an action, meaning or object. The three constitutive criteria are 1) that participants/players orient a set of basic constitutive rules that they expect to use regardless of personal preference; 2) the participants/players expect that the same set of constitutive expectations are binding on the other participants/players as are binding on themselves; 3) the participant expects that as they expect conditions 1 and 2 of the other participants/players, the others expect 1 and 2 of them (Garfinkel 1963:190). That state of mutual sharing is called Trust and meeting Trust conditions. It does not mean that we trust one another in the ordinary way. You can deeply mistrust someone and still Trust in Garfinkel’s sense that they are meeting the constitutive conditions of sensemaking. Participants in social situations must Trust that others are meeting the constitutive Trust conditions as the others must Trust that they are. When Trust conditions are not met, meanings cannot be established and, in the absence of remediation, the situation becomes unintelligible. As such, Trust – as an implicit set of constitutive contractual conditions – is the cornerstone of social order and a lack of Trust leads to disorder and senselessness.

Introduction to Parsons’ Primer

3

shortcomings Garfinkel called attention to in Parsons’ theory were, he argued, the result of Parsons’ failure to fully comprehend “constitutive” aspects of the social contract position he had taken: with particular emphasis on his failure to appreciate constitutive aspects of language/concepts. To say that practices are constitutive means that they must meet the criteria actual participants are using to achieve them in actual instances of use. This involves contingency and variation. Parsons like most other scholars adopted a simplistic conceptual and referential view of language and social practice that was inadequate to the task. This led to contradictions, including Parsons’ adoption of external authority to limit variations (exigencies), his acceptance of inequality and his emphasis on consistency; all of which conflict with the flexibility and mutual orientation required to achieve meaning in actual situations. It was Garfinkel’s view that if the question of how meaning and other social objects are assembled/achieved in actual interactions was addressed, the problems with Parsons’ position could be remedied. From the very beginning of their relationship in 1946 Garfinkel had been trying to convince Parsons that this was possible. Even without remedy, however, Parsons had, in Garfinkel’s estimation, gone farther than any other social theorist in positing an underlying moral commitment and the possibility of further clarification to situated “systems of interaction” and their corresponding sanctions, as the foundation of social order. The necessary preconditions for this interactional system and its situated interactional framing, which Garfinkel conceptualized as “Trust conditions” were, in Garfinkel’s view, most successfully worked out by Parsons in his 1960 revision of the pattern variable argument3 and other writing after 1960, although his continued acceptance of authority and inequality and his insistence on consistency kept Parsons from ever getting it quite right. Durkheim had been the first to try modernizing sociology in the 1890’s by setting it on an implicit social contract footing that eliminated individualism and positivism. Unfortunately, early critics had misunderstood the argument, and WWI brought the Durkheim School of sociology to an abrupt end. Parsons had re-introduced Durkheim in the 1930’s with the idea of modernizing US sociology on similar terms. Once again war intervened, as the urgent demands of WWII for fast and efficient “scientific studies” worked against Parsons’ attempt at modernization; fostering a narrative that elevated the older 3

The “pattern variables”, which Parsons first introduced by name in the 1940s and spent the next two decades revising, were meant to capture all of the normative ways participants could orient the assembly of social objects in interaction. The pattern variables were of particular interest to Garfinkel because they were a key way that Parsons centered interaction in his theory. While he felt that they ultimately fell short of their purpose, Garfinkel argued that the pattern variables represented an advance over other contemporary social theories, which ignored interaction entirely.

4

A. W. Rawls and J. Turowetz

individualist and positivist approaches that Parsons was challenging (Rawls 2018). Treating social objects as if they were “natural” and measuring them as such – counting them as if they were natural “things” – is much easier and quicker than establishing their social parameters. Without studying how they are “assembled” as social objects in particular situations of practice, however, the “scientist” will naively incorporate taken-for-granted cultural assumptions into the categories of things they count. By the early 1960’s retrograde approaches based on naturalism were starting to overwhelm Parsons. Garfinkel was concerned for the future of sociology in the face of this backward trend and saw in Parsons’ theory the best foundation for grounding an approach to interactional systems with the potential to stop social theory’s slide into irrelevance. The Parsons of the Primer is not the structuralist, idealist, or functionalist we have come to expect, and Garfinkel, its author, emerges as a dedicated sociologist and social theorist.

Garfinkel’s Effort to Promote and Defend Parsons’ Social Theory For many years a mimeographed version the first five chapters of the Primer (dated 1960) circulated with the mistaken idea that its purpose was to twist Parsons’ position to fit Garfinkel’s own perspective. However, materials in the Garfinkel Archive, including letters to and from Parsons (Appendix #1), make it clear that the manuscript constitutes a serious and important scholarly effort by Garfinkel to explain why Parsons’ social theory is important to sociology, and that the effort was coordinated with Parsons. The manuscript explains important aspects of Parsons’ argument that are not well understood, while also providing essential grounding for Garfinkel’s own approach. The circulation of the manuscript may initially have been Sacks’ idea. In an undated letter to Garfinkel, Sacks wrote: “Since you’re not going to be doing anything much with the Primer for a while, it might well do to publish a barely revised version as ‘working volume’. It is too important to let lie around”. Most treatments of Garfinkel’s position that have considered its relationship to social theory, have discounted Parsons’ influence and maintained that Garfinkel drew primarily from Alfred Schutz. Parsons and Schutz are described as holding opposing positions, and Garfinkel is said to have chosen between them, with most critics claiming he chose Schutz. In the Primer, however, Garfinkel paints a very different picture. Rather than choosing be-

Introduction to Parsons’ Primer

5

tween Parsons and Schutz, it becomes clear that Garfinkel considered both important for how they treated interaction/culture as the key to social order, albeit in different ways. Misunderstandings about the relative contributions of Parsons and Schutz to Garfinkel’s position, along with a lack of knowledge of his indebtedness to Wittgenstein, have combined with a general lack of awareness of Parsons’ own developing interactionism to obscure both Garfinkel’s general position and the argument he was making about Parsons. Consequently, the relationship between Garfinkel and Parsons has remained unknown, the Primer has not been taken seriously as a scholarly treatment of Parsons, the overall point has been missed, and the manuscript is essentially unknown beyond a small circle of those interested in Garfinkel’s work. This is consequential because the Parsons connection is an important missing piece of insight into Garfinkel’s argument. Parsons provided a social contract foundation for an approach to a theory of social interaction, conceived as systems of rules grounded on an underlying social contract. Without it Garfinkel’s position has been interpreted as individualist. While Schutz suggested a preliminary way of conceptualizing the assembly practices that are used to create social objects and meaning within those interactional systems, and Wittgenstein offered a way of talking about culture/interaction as a language game, neither supplied an overall architecture needed to ground Garfinkel’s conception of a system of interaction. Garfinkel saw the positions of Parsons and Schutz as fitting together and often referred to his indebtedness to Parsons and Schutz in combination (see Appendix #2). In another manuscript, “Notes on Language Games” written in 1960–1, but not published until 2019, Garfinkel grounded the effort to adequately articulate the kind of constitutive rules involved in “interactional systems”, and the Trust conditions involved, on Wittgenstein (Garfinkel [1960]2019). He was not taking sides. In Garfinkel’s view, the contributions of Parsons, Schutz and Wittgenstein were all necessary to ground an adequate theory of how social order and meaning are achieved in a diverse modern society. When Garfinkel wrote the Primer he was hoping to craft a synthesis of these positions (and others) with the addition of his own research on members’ methods of making sense (which he named ethnomethodology in 1954) as both a demonstration and a transformation of Parsons’ argument. At the time, Parsons was still hoping to satisfy his critics, Wittgenstein scholarship was enjoying a period of popularity, and phenomenology had not yet fallen under the wheels of the post WWII “objectivity” train. Later, as they all fell out of favor, Garfinkel would have little

6

A. W. Rawls and J. Turowetz

choice but to let ethnomethodology stand on its own, and this important book was not published. From 1958 to 1963, however, still optimistic about the prospect of effecting a theoretical synthesis, Garfinkel worked on the Primer with the idea that a better understanding of Parsons would strengthen social theory, while also making it clear that Garfinkel’s own work addressed issues at the center of both Parsons’ theory and sociology more generally. In particular, Garfinkel sought to explain how Parsons’ conception of culture as an independent domain of social order had the potential to modernize social theory by moving social interaction to center stage, treating actor and action as “objects” that exist only, as Parsons put it, “in-their-course”, in the context of a prior implicit social contract/agreement, and then framing the question of social order such that “culture” could be specified in empirical details, through studies of interactional assembly practices and the systems of constitutive expectation that define them.

The Importance of Parsons’ Position on Culture as Interaction It was Garfinkel’s position that in theoretically centering culture as interaction, Parsons had made an invaluable contribution to social theory, but that the move had been misunderstood precisely because of its emphasis on interaction. Garfinkel was worried about this and expressed concern that, while Parsons’ theory was essential to a modern sociology, important interactional aspects of his argument, like his embedding of interaction in the variables in his 1960 revision of the pattern variable argument, were not being recognized, which was leading to additional criticism. In a January 22, 1959 letter to Parsons, Garfinkel expressed the view that, properly understood, Parsons’ work had the potential to sort out serious problems that he saw developing in post-war sociology as quantitative methods pushed aside general social theory and important theoretical domains like social interaction. Referencing his own department at UCLA as a case in point, Garfinkel described a troubling disconnect between theory and methods: “Methods are stressed while students and faculty alike operate with an impoverished knowledge of the nature, tasks, uses and procedures of sociological theory. Where general theory is concerned, the paradoxical combination of interest and fear is an added ingredient. I propose to demonstrate that the intent and effect of your work is the clarification of the important theoretical domains...”

Introduction to Parsons’ Primer

7

Letters between Garfinkel and Parsons (Appendix #1) show that Garfinkel’s effort to promote Parsons’ social theory as an antidote to these problems was coordinated with Parsons from the summer of 1958 to 1964 (and probably through at least 1968) and enjoyed Parsons’ support and encouragement. These letters also show that Garfinkel’s effort to explain Parsons in the Primer was closely connected to his development of ethnomethodology, as well as to several versions of his “Trust argument” and his studies of language/culture and games, for which he drew on Wittgenstein: It is important that all of these efforts were taking place at the same time and do not indicate a change of position. Unfortunately, this important collaboration between Garfinkel and Parsons was overwhelmed by the turbulent politics of the 1960’s (war getting in the way again, this time the American war with Vietnam), which characterized Parsons as a conservative “insider”, while casting Garfinkel as a radical “outsider”. The idea that Garfinkel abandoned Parsons and sociology generally, along with the belief that his position was built on Schutz, contributed to a false narrative of Garfinkel as an “outsider” who rejected sociology and focused on the “individual” experience of social interaction, rather than on society and social order. While being Jewish had marginalized him socially (and was an important source of insight, see Turowetz and Rawls 2019), academically Garfinkel was not an outsider, and he did not abandon either sociology or Parsons. Rather, it was sociology that abandoned the study of social facts, social selves and social interaction during and after WWII, and in its quest to be more “scientific” abandoned the Durkheim/Parsons vision of a social order grounded on implicit social contract that Garfinkel was building on. That Parsons was collaborating with Garfinkel to defend the importance of both social interaction and general social theory was left out of the narrative. That intellectually Garfinkel was more of an “insider” (having been at North Carolina with Howard Odum, at Harvard with Parsons, Robert Bales and Jerome Bruner, and at Princeton with Wilbert Moore), and that his innovative ideas dated from the 1940’s not the “radical” 1960’s, was also left out. However, because Garfinkel preserved the relevant materials in his Archive, we have been able to reconstruct the collaborative effort of Parsons and Garfinkel to revive sociological theory through an explication of culture as social interaction. Both Parsons and Garfinkel were radical thinkers in this sense. Parsons had initially centered social theory on interaction/culture to remedy theoretical shortcomings in the 1940’s, and so had Garfinkel. Parsons’ embedding of interaction in the pattern variables in 1960 was another revolutionary move, and in writing about money in 1963, Parsons referred to money as “interaction”, to emphasize the point.

8

A. W. Rawls and J. Turowetz

Garfinkel emphasized this interactional side of Parsons, and for his own part, focused on how “members” of society assemble mutually intelligible meaning and identity in interactions that are grounded on implicit Trust conditions. He noticed early on that some categories of person (Black, Jewish, transgender, mental patient) had to work harder than others to “pass” as “normal”. Studies of such people, he argued, were “natural experiments” that could reveal the assembly practices and underlying “Trust conditions” that any member of society has to engage in to achieve meaningful social objects and identities. For most members of society, the work of achieving “normal” identities and meanings remains invisible and taken-for-granted. But, Garfinkel argued, those persons inhabiting marginal identities that require more work develop an awareness of that work which can be studied to reveal, in turn, what the rest of us are doing.4 Developing this awareness of the work involved in being seen as “normal” is much like the self-awareness of being treated differently by Race that is involved in W.E.B. Du Bois’ “Double Consciousness” thesis, aligning Garfinkel with Du Bois in significant ways.5 While Garfinkel’s debt to Schutz is large, Schutz did not offer a way of theorizing interactional systems sociologically. Neither did Wittgenstein. Parsons did. This is one reason why interpreting Garfinkel primarily in terms of Schutz (and as a phenomenologist) has created the false impression that Garfinkel was focused on the individual and not on interaction as a social system. Materials in the Garfinkel Archive document Garfinkel’s substantial and frequently acknowledged debt to Parsons and his theory of interaction as a system of sanctioned rules and expectations grounded on implicit social contract, along with a dedication to sociology and social theory generally. Sections of the Primer manuscript (especially Chapters VI–IX), also make it clear that Garfinkel was using his own research in ethnomethodology to defend Parsons and build on his theory. In transcripts of the 1962 ethnomethodology conference at UCLA 4

Garfinkel and Goffman had planned to publish a book together in 1962 to address this issue. Tentatively titled On Passing, it was to include Garfinkel’s chapter on Agnes and Goffman’s monograph on Stigma. After Garfinkel got additional information about Agnes that required revisions to the chapter in February/March 1962. Garfinkel told him he should try to publish without him and not let the opportunity get held up waiting for the Agnes revisions. Goffman was able to secure a contract to publish his own contribution as a monograph. Goffman’s Stigma was subsequently published without Garfinkel’s contribution. 5 The taken-for-granted attitude of “normal” members of society is analogous to the White American lack of awareness of tacit racism (Rawls 2000, Rawls and Duck Forthcoming). The awareness that research in ethnomethodology develops about how marginal identity and other social categories are achieved in everyday life is like a double consciousness. The sociologist engaged in Ethnomethodological research learns to see how social categories are being achieved in ways that are invisible to most others. It is an important resource. For further discussion of the relationship between Garfinkel’s argument and Du Bois see Rawls and Duck, Forthcoming and Turowetz and Rawls 2019.

Introduction to Parsons’ Primer

9

(in the Garfinkel Archive), Garfinkel talked about Parsons and the Primer in the context of ongoing ethnomethodological research. Garfinkel consistently attributed his interest in sociology to his early reading of Parsons (Garfinkel 2002), and his admiration did not wane. For a celebration of the 50th anniversary of Parsons’ Structure of Social Action (1937), at the American Sociological Association meetings (ASA) in 1987, even as he reiterated his disappointment that Parsons never accepted the need to examine actual concrete interaction (the principle unresolved difference between the two scholars), Garfinkel wrote that “Ethnomethodology began with Parsons” (Garfinkel 1988:3).

The Relationship between Parsons and Garfinkel The relationship between Parsons and Garfinkel began in 1946 when Garfinkel arrived at Harvard and became a PhD student of Parsons. Then after Garfinkel completed his PhD in 1952, there were several years of less contact until the relationship intensified again from 1958 to 1964 while Garfinkel was reading and teaching Parsons in seminars at UCLA, preparing the Primer, meeting with Parsons at Harvard in 1958, working with him on the revision of the pattern variable argument in 1959, and corresponding and sharing materials. The relationship continued through the late 1960’s with Garfinkel arranging for Parsons to visit UCLA in 1964 for an extended meeting at the Suicide Prevention Center and again in 1966 to give a “Chair’s” talk (transcripts in the Garfinkel Archive), and spending his sabbatical at Harvard in 1968. The two managed to coordinate their efforts to situate interaction at the center of social theory for over two decades – without anyone noticing. That they did not immediately succeed in rescuing social theory is due more to the post-WWII turn away from social interaction and back toward an earlier individualism and naturalism, and the various misunderstandings that resulted from that, than to the soundness of Parsons’ interactional social contract position, or its relevance to Garfinkel and ethnomethodology. The effort was certainly not derailed by attention to it. It should not be surprising that Garfinkel admired Parsons and was inspired by his work, or that Parsons in turn was influenced by Garfinkel. Garfinkel was after all Parsons’ PhD student and had gone to Harvard to work with him, both were important social theorists, and their working relationship lasted over 20 years. What is strange is that we don’t think of the two bodies of work as related. They are related.

10

A. W. Rawls and J. Turowetz

In the Primer Garfinkel explains why Parsons’ theory is important for an adequate approach to culture/interaction and says that Parsons was misunderstood because of this important contribution. In essence, Garfinkel is arguing that Parsons specified a system of social interaction that his critics simply did not acknowledge. This is strange in itself as it was Parsons who had first singled out culture/interaction as the central feature of a modern approach to social theory in 1949 (Parsons 1950), and again in 1951 (while Garfinkel was with him at Harvard). The Primer makes it clear that this much overlooked interactional aspect of Parsons’ work was the basis for Garfinkel’s interest in Parsons. In Garfinkel’s view, Parsons’ theory wasn’t perfect. It needed changes. In particular, it took for granted essential aspects of language and interaction that led him to incorporate authority and inequality into his system. This was a problem for Garfinkel. One of the first things he had done when he got to Harvard in 1946 was to lay out a blueprint for the changes he said were necessary. He did this in 1946, 1947 (Garfinkel [1948]2006), and again in his 1952 PhD dissertation (Garfinkel 1952). But, he expressed hope that Parsons would be open to resolving the problems.6 While Garfinkel had taken guidance on some of these issues from Schutz (starting around 1946), in the Primer he maintains that Parsons continued revising his position to accommodate the criticisms. In Garfinkel’s view, the development of Parsons’ work after 1958, and, in particular, the revision of the pattern variable schema in 1960 (on which Garfinkel worked with Parsons), resolved most of the problems with the earlier argument, resulting in a model that fully embedded interaction, bringing Parsons closer to what Garfinkel considered an adequate theory of modernity. It is an indication of his seriousness about the importance of Parsons’ theory, that in mounting his defense of Parsons, Garfinkel used material from his own “studies” of “members’ methods” for assembling culture to explain aspects of Parsons’ social system. Garfinkel did this to explain the revisions to the pattern variable argument in 1960 (Chapter V), and in defending Parsons’ conception of norms and values (as constitutive rules) from his critics (Chapter VI). On page three of a January 14, 1963 letter to Parsons, and in the Primer itself, Garfinkel indicated that he considered instances of “action-in-its-course”, that are recognizable as references to his own studies of ethno-methods (e. g. 6

In his 1946 seminar paper, “Some Reflections on Action Theory and the Theory of Social Systems”, written while he was a PhD student at Harvard, Garfinkel (1946:2) remarked on a “very promising kind of unsettledness in Parsons’ thinking” and expressed his hope that Parsons would make a “breakthrough to a full phenomenological position” that would allow him to achieve “a fully rational theory of social action” (1946:3). Here, a “fully phenomenological position” would be one that combined Parsons’ insights about social contract and moral order – his theoretical architecture – with insights from Schutz about the assembly practices through which that order is assembled.

Introduction to Parsons’ Primer

11

on “accounts” and “accounting”), to be “representations” or “symbols” of the “social system” as Parsons’ conceived it. The point made is similar to C. W. Mills’ (1940) famous argument that institutional order can only be actualized through “vocabularies of motive” (a sociological elaboration of Kenneth Burke’s (1936) approach to motives/accounts through language). Parsons, for his part, was not indifferent to Garfinkel’s assessment of his position, and confirmed their mutual understanding in letters and audiorecorded meetings, which show Parsons contributing ideas, time, and encouragement to the effort. In a remarkable audio recording of a session with Parsons, Garfinkel, Goffman and Sacks that took place at the Suicide Prevention Center in Los Angeles during the spring of 1964, Parsons demonstrates a surprisingly nuanced understanding of what Sacks and Garfinkel were saying about situated meaning, at times even explaining their position to Goffman, when he was not following the argument.7 Unfortunately, the urgent demands of the 1960’s for a macro political focus increased the already strong pressure on social theory to move away from social interaction where inequality actually plays out, driving a wedge between the mainstream social theory that claimed Parsons as its champion, and the social interactionism that Parsons was himself moving toward (that would subsequently be led by Garfinkel, Goffman and Sacks). It did not help that this was all happening just as the alleged “structural-functional” Parsons was coming under increasingly heavy criticism by those who did not realize how much his argument had evolved in an interactional direction and how that could have eliminated its problematic adoption of authority and inequality and led to an interactional demonstration of the need for equality of the sort Garfinkel produced.

An Invitation from Parsons – 1958 The Primer manuscript itself has a complicated and interesting history. The collaborative discussions between Garfinkel and Parsons that were the immediate inspiration for its production were initiated by an invitation from Parsons to Garfinkel in the summer of 1958. In a letter dated July 14, 1958, the day before Parsons left the Center for Advanced Studies at Stanford to drive back to Cambridge Massachusetts, Parsons wrote inviting Garfinkel to attend a conference of Parsons’ most valued associates at Harvard that September, 7

There is a transcript and audio recordings of this meeting in the Garfinkel Archive, an excerpt from which appears at the end of this introduction.

12

A. W. Rawls and J. Turowetz

at which the group would discuss Parsons’ recent work. The conference was planned to extend over six days, four working days with a weekend in the middle. Comprised of five former students and colleagues, Garfinkel, Kaspar Naegele, Jim Olds, Robert Bales and Robert Bellah (Neil Smelser couldn’t make it), the conference was to be devoted entirely to discussion of Parsons’ work in the context of a manuscript he had just completed that summer at the Center.8 Letters make it clear that Parsons had sent out a call for those of his students with serious theoretical interests to return from far flung places to advise him, and at fairly short notice.9 Naegele was in British Columbia, Garfinkel in Los Angeles and Smelser at Berkeley (Olds, Bales and Bellah were already at Harvard). Parsons included Garfinkel in this group of close and valued associates and wanted him by his side.10 That Parsons included Garfinkel in this group will be a surprise to many. That Garfinkel took this invitation seriously and prepared for the conference with diligence at a time when he was focused on his own studies in ethnomethodology may surprise them even more. The seriousness with which Garfinkel took this invitation is evident in a letter he wrote to Parsons on August 19, 1958, less than a month before the meeting, in which he described his preparations for the conference: “I’ve put everything aside in order to work through your manuscripts. The reading is going slowly because I am trying to manage several tasks in the reading: to grasp the intended sense of your formulations; to sort the corpus into its analytic parts and to re-read and interpret these parts in light of my own concerns. Also, I’ve taken the conference as an occasion to re-read Durkheim. The whole effort has turned into an ‘experience’”.

Garfinkel described his preparations much the same way in a letter he wrote to Naegele the following day, on August 20, 1958, saying that he had “put everything aside” to prepare: 8

Robert A. Scott was Assistant Director at the Center for Advanced Study while Parsons was there in 1958. Peter K. Manning reports that Scott would tell the story that the other residents at the Center made daily bets on how many pages Parsons would dictate each day. Apparently, he dictated his manuscripts into a Dictaphone and was legendary for how much he could “write” in a day with little correction. It was one of those stories you “dined out on”. 9 It is clear from the July 14th letter that there had been a prior inquiry about this invitation. But the earlier letter has not yet been found, so we don’t know how far in advance Parsons began planning for the conference. 10 In a letter to Schutz written August 20, 1958, Garfinkel explains that his schedule is flexible enough to allow him not only to attend the conference, but to schedule a meeting afterward with Schutz in New York because he has been awarded a five year fellowship from the National Institute of Health, which also explains his availability to work with Parsons over the next several years; an availability that Parsons also refers to in his letters to Garfinkel.

Introduction to Parsons’ Primer

13

“I’ve put everything aside in order to work through the large number of Talcott’s manuscripts. The reading is going slowly because I am trying to manage the double tasks of reading for the intended sense of Talcott’s formulations, and interpreting his formulations in light of my own concerns”.

Then, in a letter to Martin Martel, also written on August 20, 1958, Garfinkel announced that he was planning to give a seminar on Parsons during the spring semester of 1959 at UCLA, a project that appears to have been inspired by his preparations for the Harvard conference. He asked Martel to send him a copy of a manuscript, titled “Primer for Parsons”, that he said Dave Gold had told him Martel had prepared, offering to pay for the cost of sending him a copy, and telling Martel he would like to assign it to his students, and also share it with those attending the Harvard conference (it does not appear that Martel sent the manuscript). We know that Garfinkel discussed the proposed course, and the idea of preparing a manuscript on the basis of it, with Parsons (probably at the Harvard conference), because on January 22, 1959, Garfinkel wrote Parsons asking him to send copies of four pieces of his work for students to read in that seminar, mentioning the Primer as if Parsons already knew about it, and promising to tell Parsons more about his plans soon. The UCLA seminar in 1959 was SO 172 (in 1963 it was renumbered SO 251), titled Contemporary Sociological Theorizing, and we see from the syllabus (Appendix #3) that, while focusing on Parsons, it also covered related materials, including articles by Parsons’ students (like Naegele). Parsons had already given Garfinkel the new materials and manuscripts discussed at the Harvard conference. The additional materials Garfinkel requested were: “Introduction to Part I, Section A of the Theory Reader: The General Interpretation of Action”. “General Introduction to Part II of the Theory Reader: An Outline of the Social System”. “An Approach to Psychological Theory in Terms of the Theory of Action”. “The Role of General Theory in Sociological Analysis: Some Case Material”.

These materials are in the folder of materials for that course in the Garfinkel Archive, so we can assume that Parsons sent them. In these letters there are many requests for materials, as well as requests to return materials given previously.

14

A. W. Rawls and J. Turowetz

This exchange of materials is in itself important. Because this work was being done in the era before Xerox, sharing unpublished work required maintaining relationships. Everyone made mimeos of their work. But, a mimeo master would print only a limited number of times. When copies ran out the mimeo masters needed to be retyped at great cost. Copies were valuable, to be given only when it mattered, and only to trusted people. To get copies scholars needed to write to each other and ask for them. Copies were given and returned as treasured items in a kind of gift exchange without which scholarly collaboration in that period would have been severely limited.11 Parsons and Garfinkel were engaged in such an exchange (Garfinkel and Martel were not, hence the offer to pay for the copy and Martel’s failure to send it). Garfinkel wrote to Parsons again one week later, on January 29, 1959, fulfilling his promise to tell him more about how he was going to teach the course. His plan, he said, was to teach the materials backwards as a way of keeping the discussion current, and avoiding the inevitable focus on the development of Parsons’ ideas that Garfinkel worried would come from dealing with them chronologically. Garfinkel wanted to emphasize the relevance of Parsons’ mature approach to the “current situation of the discipline”: “I’ve decided to teach your materials by using the most recent writings as the precedent for reading the earlier ones. Reading the corpus “backwards” lends to the earlier work, its sense of what you have been up to “all along” or “after all”. The criticisms of such a rereading procedure, i. e., that it produces an “interpretation” of Parsons, or that it reads Parsons for something more, less, different, better, worse than he intended, are easily met. A chronological reading would be difficult to carry off without making the development of your ideas a central theme. I feel that the seminar, and a “Primer” based upon it, should be directed to the uses to be made of your materials in the current situation of the discipline”.

The important thing was to explain what Parsons had been “up to” and how that was relevant to the situation post-war sociology found itself in. That “current situation”, as exemplified by his own department, Garfinkel wrote, was an “impoverished knowledge... of sociological theory”, combined with an imbalance in favor of statistical methods and against social interaction, that was leaving students and faculty alike in a paradoxical state of “fear” and “interest” with regard to social theory that is still familiar today. Garfinkel proposed to rectify the situation by explaining Parsons, and in particular why Parsons’ emphasis on social interaction is a necessary foundation for a modern 11 Anne Rawls was given a copy of the index to Sacks’ lectures as a wedding present by Alene Terasaki as late as August 1978.

Introduction to Parsons’ Primer

15

sociology, and also offering much needed empirical demonstrations based on his own research. For Garfinkel, in spite of their differences, the way Parsons had positioned social interaction/culture at the center of his social system, grounding it on mutual commitment and insisting on its independence from social structure, was an essential platform for Garfinkel’s own studies of ethno-methods. If Parsons was positioning social interaction in his theory the way Garfinkel says he was, sociology somehow missed that important theoretical development. Furthermore, if Parsons was proposing that objects and identities exist only within the context of an interactional system grounded on an implicit social contract, then the fact that his critics continued to treat his position as positivist, structuralist, and/or idealist, is a huge mistake that has left a big hole in social theory. Garfinkel was still hoping in 1958 that a revised Parsonian theory that approached interaction in terms of self-organizing practices, on the basis of studies of actual concrete interaction, could improve the situation. That he gave up on the attempt to theorize about this publically at some point does not mean that he ever stopped believing in its importance. Ethnomethodology is a theoretical commitment to approaching the problem of social order in concrete empirical terms – not a method – and it built in essential ways on Parsons. We hope that this publication of the Primer manuscript will revive and reground this important collaborative effort in contemporary social theory.

The Parsons’ Primer Manuscript This book reproduces the last and most complete copy of the Parsons’ Primer manuscript that resulted from this collaboration between Parsons and Garfinkel. As it appears in the Garfinkel Archive, the version of the manuscript published here was assembled as a “book” (in a blue cloth three ring binder), consisting of nine chapters, and bearing the date April 1962 (although Garfinkel worked on it extensively after that date). Garfinkel produced at least three different versions of the manuscript between 1959 and 1963. The first five chapters were completed in 1960 (the part of the manuscript that has been circulating, although some editing was done to Chapters II, III and V in 1963). The four additional chapters (VI to IX) are in the April 1962 binder, but evidence suggests that most of what they contain was not written until 1963 and was certainly edited in 1963.

16

A. W. Rawls and J. Turowetz

Because the original mimeographed version of the manuscript – the first five chapters – continued to circulate and Garfinkel continued to teach from it, most people who have seen the manuscript have seen the 1960 version of the first five chapters and may be surprised to find that those chapters are slightly different here, as Garfinkel continued editing them. They may also be surprised that there are nine chapters. The manuscript as we reproduce it contains all nine chapters from the 1962 binder. We know from their letters that Parsons commented on Chapter V in January 1963, and that Chapter VI seems to have been written in response to those comments from Parsons. Chapters VII to IX build on Chapter VI and represent a further elaboration of Parsons’ later position, so they also must have been written after 1963. These later chapters all have in common a more explicit positioning of Garfinkel’s own research with regard to Parsons’ theory than the first five: with Garfinkel using descriptions of his research to demonstrate theoretical points in Parsons’ argument. In preparing the manuscript for publication, we have attempted to incorporate all of Garfinkel’s edits so as to publish the most up to date version of each chapter. Garfinkel produced an entire new mimeo of Chapter V (which discusses the pattern variables) in 1963, with a new introduction and many corrections which we have incorporated. It appears that he was preparing that chapter as a stand-alone article which was never published. There are also several edited versions of Chapters VI to IX all of which appear to be from 1963. In making corrections to the manuscript we found that some ends of pages had been typed too close to the bottom on the original mimeo masters and had consequently never been reproduced. Other corrections to mimeo masters made early on had also never been reproduced. We worked from the original master sheets for these corrections. Those who are familiar with the original version, and have been looking for missing parts, will hopefully find them included. Garfinkel’s corrections were fairly clear on the manuscripts from which the mimeos were prepared and the versions were often dated. His corrections of the mimeo masters themselves, however, could have been done at any point after the masters were typed, making it impossible to date those changes. Working from the latest versions that could be identified in all cases, the intent has been to publish the manuscript in a form as close to the final one produced by Garfinkel as possible. This has also meant leaving his footnotes in the middle of pages bounded off by lines, as he presented them, and adding citations and clarifications only in [Editor’s Notes] at the bottoms of pages.

Introduction to Parsons’ Primer

17

The nine chapters are published together with a paper that has been positioned at the end, titled “The Program of Ethnomethodology”, and bearing the date October 16, 1961 (when it was apparently delivered orally to an audience of persons listed on the manuscript’s last page). Garfinkel kept several versions of this “Program” manuscript in folders with Parsons’ Primer materials (there are several file drawers of these in the Garfinkel Archive) along with handwritten directions for how sections of the Primer should be edited to accompany it. He obviously intended the Primer and the Program to be published together. In one case the Program was to be the opening section of a relatively short selection of materials from the Primer. In another it was positioned as the introductory chapter of a binder titled Essays in Ethnomethodology (1961). In Appendices, along with the letters (Appendix #1) and Garfinkel’s situated discussions of his indebtedness to Parsons and Schutz (Appendix #2), we include a syllabus (Appendix #3) for the Parsons’ seminar as Garfinkel taught it in 1959, which mainly consists of a bibliography. Much of Parsons’ work as listed in this bibliography was unpublished in 1959 and was still unpublished in 1963, the second time Garfinkel taught the course. Some of this work was subsequently published, but too late to temper the criticisms that would undermine Parsons in the late 1960’s.

Points of Agreement and Difference between Parsons and Garfinkel Because one of the most serious misreadings of Garfinkel is that his work proceeded in stages from a point where he was more conservative and embraced Parsons, to a later more radical phase that has been associated with Schutz, it seemed instructive to include the Program manuscript from 1961 – which announces Garfinkel’s Ethnomethodological studies – along with his careful exegesis and defense of Parsons in the Primer, which was completed two years later. The argument that there was a transition is wrong. Garfinkel did not abandon Parsons for Schutz. Garfinkel was not criticizing Parsons either. The Primer is not so much a critical reading, as a close reading of Parsons’ argument and in the later chapters, which were written as he got closer to the publication date for Studies in Ethnomethodology (1967), Garfinkel used more of his own research to support Parsons’ ideas – having been encouraged to do so by Parsons’ responses to reading the first five chapters of the Primer (expressed both in person and in his letters). In 1961, in the Program manuscript, Garfinkel had already

18

A. W. Rawls and J. Turowetz

laid out the program of research he would pursue for the rest of his career. Chapters V through IX of the Primer were written after this announcement. There is extensive overlap between the Program statement and Garfinkel’s use of Ethnomethodological research to defend and explain Parsons in the Primer. Parsons and Garfinkel agree that categories of actor, action, and objects must be achieved in systems of social interaction that are grounded on an implicit underlying social contract. They agree that evidence of the underlying commitment and descriptions of the values/rules must come from empirical observations of sanctions. They agree that social order is constantly made and remade through the cooperative use of interactional assembly practices. They agree that a theory is only valid if the assembly practices of interaction/culture that it describes would produce the actions and interactions that actors are actually engaged in. In other words, they agree that a theory can be falsified by empirical research. The main difference between them is how they handle the exigencies/contingencies involved in interaction. Because he did not grasp the full implications of constitutive practices, Parsons thought it was necessary to analytically control the exigencies, while Garfinkel was able to document that contingencies are “routinely” turned into orderly/meaningful phenomena in actual concrete interactions by participants in ways that can be documented systematically. For Parsons these variations are problems, while for Garfinkel they are part of the solution. Parsons maintained that in order for meaningful categories to be accomplished in interaction, what he called “environmental exigencies”, the endless situated variations in interaction, must be brought under control by external social authority. For Garfinkel, by contrast, the contingencies are important resources that people use to make order and meaning. Garfinkel rejected the idea that external authority and inequality are necessary, instead embracing self-regulation in a context of equality; which he maintained was a necessary condition of sensemaking. Because organizing the contingencies must, in Garfinkel’s view, remain a flexible process, he argued that how contingencies work cannot be specified analytically, but must be discovered through empirical study of actual concrete practices in use. In insisting on this focus on the concrete details of actual practices Garfinkel was taking a different variant on Durkheim from Parsons: taking up Durkheim’s (1893) argument that in diverse modern societies contingencies can become a resource with which participants who are engaged in the “double contingency” of social interaction, can self-regulate from the “bottom up” to create certainty from uncertainty through the cooperative use of constitutive

Introduction to Parsons’ Primer

19

practices (Rawls 2008, 2019).12 Durkheim had argued that this needs to happen in diverse modern societies, not only because it is the only way of making sense across diversity, but because it is the only way that science and technical progress can break free from the constraints of top down authority and shared belief. This self-regulation, in turn, enables the creation of new meanings and new constitutive practices. Garfinkel was already echoing this Durkheimian position in 1946–47, writing that a way of making sense that does not depend on shared beliefs or analytics needs to develop in any society that is not bounded by what he called “tribal” reasoning.13 Although they differed over how to handle interactional contingencies, however, both Garfinkel and Parsons treated contingencies as a central concern, and agreed that the processes involved in bringing order to contingency in interaction needed to be taken seriously. In the Program manuscript, Garfinkel aligns with Parsons in criticizing the discipline for ignoring contingencies, and concludes that, as a consequence of this neglect, sociology had abandoned itself in favor of the “biological and physical aspects of the events of human conduct”, leaving it mired in the individualism and naturalism that Parsons had been warning against since 1938, and negating the purpose of sociology as a discipline. 12

Parsons (1951:6) introduced the term “double contingency” to refer to the fact that in interaction, meaning is contingent on what an actor does and says and on the response it receives. Meanings are made and remade sequentially, over the course of an interaction, a sequential process Garfinkel (1967:7) would come to call reflexivity. 13 In his 1947 “Notes on the Information Apperception Test” Garfinkel contrasted “Tribal” with “Cartesian” perception, a distinction that he would maintain for decades. He also discussed tribal reasoning in his PhD proposals. The text from 1947:25 is as follows: “We pointed out earlier in the paper that communicative effort was conceived of as the expression of an “inner dialogue.” Insofar as the subject engages in the personalization of the objects, we have a clue as to the way the subject has his relationships with others. Depending then on how he relates the objects to each other and how he relates himself to them we have a picture of what is crudely approximative of the Rorschach concept of the feeling for closeness, but which we shall conceive of as the subject’s modes of participation with others. The reasoning does not go, however, that the objects are representative of other persons, though such reasoning would be legitimate where it could be shown that the objects are the express representations of latent objects which are persons. (See the protocol of DJS for an example of this.) Rather, the reasoning goes that personalization is a phenomenon peculiar to the discourse of gemeinschaft relations. It occurs as a peculiarity of something we shall call “tribal perception”, and does not occur in “cartesian perception.” It bespeaks a mode of perception peculiar to a set of relations where one depends for a check on the accuracy of one’s view of an object on the consensus of other persons to whom one is related as a member of primary group. What one sees is given in a context in which another person or persons are particularistically given. Such a set of conditions placed upon the object seen is to be compared with the Cartesian view in which one is made accountable to a formal body of apersonal rules of procedure for the accuracy of things seen; one knows “for himself ”; the rules are given in a context in which the other person is represented according to some formal schematic model, as for example, “the rational man”. We shall have to be satisfied temporarily at least with these few notes, and will make whatever use we can of them as they stand. The problem of the relationship between perception and the tribal and Cartesian modes of understanding is badly in need of investigation. The area has been treated in scattered writings but it lacks systematic investigation”.

20

A. W. Rawls and J. Turowetz

In essence, in the Primer manuscript Garfinkel describes a deep convergence between his own core positions and those of Parsons. They both maintained that the exigencies/contingencies of interaction/culture stand at the heart of social theory and that ignoring them was undermining sociology. They both focused on the assembly of social objects. What they diverged sharply over was how the exigencies/contingencies are rendered meaningful in specific instances of social interaction and this difference had implications for how contingencies should be approached theoretically. Parsons’ analytic approach left him needing to accept authority and inequality, while Garfinkel’s focus on the constitutive conditions of concrete practices enabled him to demonstrate that authority does not organize interaction (except indirectly via accounts) and that inequality at the level of constitutive assembly practices of interaction/ culture prevents meaning. The difference over exigencies/contingencies makes a big difference. But, it is only one point, which explains how Garfinkel could agree with so much of Parsons’ theory, and still be the author of an innovative theoretical position that ends up being so different that it has appeared to many critics to be opposed to Parsons. It is not. The theoretical implications of Garfinkel’s position with regard to Trust, equality, and how to study social order and social interaction, all turn on this difference.

Working on the Primer Manuscript from 1959 to 1963 With the initial lectures from spring 1959 in manuscript form, Garfinkel secured a Fellowship at the Harvard Law School in the fall semester of 1959. It is interesting to note that Goffman and Sacks were also in Cambridge that semester (at Harvard and MIT respectively), and that this is where Sacks met Garfinkel and Goffman.14 It is also likely that Garfinkel introduced them both 14 Garfinkel and Sacks both attended a sociology of law seminar given by Talcott Parsons in Spring 1960. Audio for two meetings is in the Garfinkel Archive. In a letter to Edwin Schneidman on March 22, 1962, Garfinkel says the following about Harvey Sacks: “Sacks is 26 years old, has an undergraduate degree from Columbia, a law degree from Yale Law School, a year of graduate work in the political science department at MIT, sitting in on whatever looked bright a MIT and Harvard particularly in linguistics and sociology, and is currently finishing his second year of graduate work at Berkeley. He has been working under Phil Selznick’s sponsorship in Phil’s Center for the Study of Law and Society. I met Sacks at Harvard when I was there 1959–60. By the time the work at the Center would begin, Sacks will either have finished his degree or will be close to it. For the past two year he has been shuttling back and forth between Berkeley and UCLA to work with me on my ethnomethodology stuff. I’m talking with full responsibility when I say that Sacks is the best graduate student I’ve ever had a hand in training, and one of the two or three best ones I’ve ever seen in my travels.”

Introduction to Parsons’ Primer

21

to Parsons that fall. Although Garfinkel and Goffman had exchanged work and corresponded as early as 1953,15 this was the first extended face-to-face time they had spent together. Sacks would subsequently become a graduate student of Goffman’s at UC Berkeley, while traveling frequently to UCLA to do his graduate research work for his PhD with Garfinkel in Los Angeles at The Suicide Prevention Center. The three would collaborate on studies of social interaction for the next 15 years. During that fall semester of 1959 Garfinkel not only shared and discussed a draft of the Primer with Parsons, but the two engaged in extended discussions of Parsons’ intended revision of the pattern variable argument for his 1960 “Reply” to Dubin’s “friendly” critique, which Garfinkel helped Parsons to edit. A new introduction at the beginning of Chapter V of the Primer (added in 1963), explains that while the original pattern variable argument had changed several times, at the time Dubin wrote his critique in 1959 Parsons had not yet published a revision of it. Parsons decided to take the occasion of his “Reply” to Dubin’s article to do so. Parsons’ 1960 revision uses terminology like “oriented object”, “oriented actor” and “member”, which had first appeared in Garfinkel’s work around 1948, and through their collaboration was now appearing in Parsons’ writing. Evidence for their collaboration on the revision consists of a citation in the article (Parsons 1960), Garfinkel’s own narrative of the time, the revised introduction to Chapter V, and several hours of audio recordings made in the fall of 1959 of meetings between Garfinkel, Parsons and Winston White, as they discussed their plans for the Reply/revision of the pattern variable argument (audio in the Garfinkel Archive). Parsons and Garfinkel saw each other again at Stanford in 1959 at a conference on “deviance” for which there are audio recordings of the presentations and discussions in the Garfinkel Archive (Goffman and Bittner were also present). In a letter from Parsons to Garfinkel on January 30, 1963, he also mentioned seeing Garfinkel at Berkeley (no date) and expressed regret that they had not found time to talk about their work in any detail on that occasion because Garfinkel had been busy with Evelyn Hooker. In that letter, Parsons expressed a desire to find time to discuss in person issues raised in their letters about his reading of the Primer and suggested that he might be able to make it out to California in the spring of 1964. 15

Reported to Anne Rawls by Garfinkel in the 1980’s: Garfinkel said that Saul Mendlovitz told him that Goffman had been getting copies of papers written by Harvard Graduate students and circulated in mimeo form for seminars for years. These included several papers of Garfinkel’s. Then in 1953 Goffman sent Garfinkel comments on his [1948]2006 manuscript – which talked about the presentation of self – and encouraged him to publish it. Garfinkel did not.

22

A. W. Rawls and J. Turowetz

An extended meeting with Parsons did take place in Los Angeles in 1964 at The Suicide Prevention Center. Sacks and Goffman also attended (along with Ed Rose and Edwin Schneidman) and the meeting was audio taped (audio and transcript in the Garfinkel Archive). In the transcript Garfinkel introduces Parsons as their theoretical resource (Reel 3 of 4, Side A1, trans. Pgs. 68–9), saying: “As I see it Talcott was invited to spend the two days with us as a resource on the problem of social order”. Garfinkel had been actively working on the Primer manuscript until several months before this meeting, and Sacks would complete his PhD thesis on calls to The Suicide Prevention Center later that year. Parsons was an important resource for both. Garfinkel produced many versions of the manuscript between 1960 and 1963 all of which bore the title Parsons’ Primer. In all versions the first five chapters are similar and correspond closely to the mimeo version that was circulated in 1960 (with the exception of the new pages added to the beginning of Chapter V). The additional four chapters written between 1960 and 1963 continue to present a serious exegesis of Parsons’ later position, even though in the interim Garfinkel had prepared the 1963 version of the Trust paper for publication, most of the research for Studies in Ethnomethodology (1967) was already completed, and Garfinkel and Sacks had begun working together on the studies of conversation that would generate new insights into how orderly the constitutive assembly practices of conversation actually are and how closely conversational preference orders orient trust conditions.

Parsons’ Revised Pattern Variable Argument Parsons’ (1960) revised pattern variable argument is referred to by Garfinkel in the Primer as the “revision”. The first four chapters of the Primer, written during the period when Garfinkel was collaborating with Parsons on the revision, make it clear that Parsons’ new position was very important to Garfinkel. These chapters also make it clear that in his view much of the criticism of Parsons is due to the fact that scholars have not understood that Parsons’ move toward social interaction changed the epistemological grounding of his theory and the criteria it must meet, such that debates over idealism/realism, culture, functionalism (and other classic questions), on the basis of which Parsons is often criticized, no longer apply. Because Parsons’ critics missed his move toward interaction they missed the fundamental changes in his theory that followed. It seems to have been literally the case that Parsons was putting a kind of social interactionism at the core of his position, while almost no one realized that he had done so. In the Primer Garfinkel’s objective is to highlight not only the

Introduction to Parsons’ Primer

23

fact of Parsons’ move toward interaction, but its overall theoretical significance. He also wants to bring Goffman and Sacks into the picture, and make sure that everyone with an interest in social interaction understands what Parsons had achieved. Garfinkel considered Parsons’ move toward interaction to be particularly clear in the revision, which he reviews in detail in Chapter V. He has spent the first four chapters laying a foundation for this extended review. It is still problematic to Garfinkel that even in the revision Parsons approached interaction in analytic terms, rather than allowing himself to be persuaded by Garfinkel that concrete details of interaction are essential to the model. Nevertheless, Parsons was still the first theorist to argue in his 1949 presidential address to the American Sociological Association (Parsons 1950) for culture as an independent level of social order, still the first in 1951 to incorporate double contingency into the heart of his social theory, bringing actual social interaction into the frame (Parsons 1951), and Garfinkel was still hoping to persuade Parsons that order can be documented in the actual concrete details of cultural assembly practices. In the context of Garfinkel’s discussion of the pattern variables in Chapter V, and in the description of values and norms in Chapter VI, Garfinkel illustrates Parsons’ argument with his research in ethnomethodology, a pattern that continues through the later chapters. Without understanding the collaborative nature of their work on the pattern variables this has been confusing, and there has been a strong sense that Garfinkel was trying to put his own reading over on Parsons here – a thing he did sometimes do and referred to as a purposeful “misreading” of a text. But, that is not what he is doing here. Garfinkel uses the same examples and terminology in his letters to Parsons, and it is clear from Parsons’ responses that he has no objection to Garfinkel’s characterization of his work in these terms; even suggesting further topics for similar treatment, like norms and values, which become the topic of Chapter VI (see Garfinkel’s January 14, 1963 letter and Parsons’ response). Although Trust conditions and accounts are discussed in relation to Parsons in earlier chapters, it is in the review of the pattern variables that it becomes clear that Garfinkel is proposing a relevance for Trust conditions, accounts, and other studies of “normal troubles” in conjunction with Parsons’ social theory. It also becomes clear that Parsons must have seen this relevance as well. It matters a great deal that Garfinkel produces an extended and explicit articulation of ethnomethodology in relation to Parsons in a chapter on which we have comments from Parsons. It is also significant that it happens in a chapter addressed to the pattern variables, because there seems to be an as-

24

A. W. Rawls and J. Turowetz

sumption that Garfinkel would not have had anything positive to say about the pattern variables, whereas we can see that he has a great deal positive to say about that argument, worked on the revision with Parsons, and even prepared Chapter V as a stand-alone article on the strength of its importance. Garfinkel was obviously pleased with the progress Parsons had made in the revision on moving toward interaction and he hoped for progress on other fronts. In his January 14, 1963 letter to Parsons Garfinkel also wrote that “Via the rumor chains I heard that you had revised or were in the course of revising your uses of “function”. The rumor aroused my curiosity and hopes”. Not only was Parsons moving farther in the direction of interaction, but he might also be revising his functionalism.16 Garfinkel had expressed hope in 1946 that Parsons would revise his functionalism, a hope that seems to have persisted until he completed his dissertation in 1952. The change had not come to pass. Now, suddenly, here was another chance. Given the respect Parsons enjoyed in that moment it just might have been possible for Garfinkel to have successfully launched the study of interactional systems as a legitimate sociological enterprise in alignment with Parsons. This would explain Goffman’s presence at meetings between Garfinkel and Parsons, and why Sacks was attending. But, the functionalism would need to change. Parsons’ original argument had been problematic in assuming an untenable form of means-ends rationality as a given, and treating motivation as a characteristic of natural/epistemic individuals, who are oriented toward natural/epistemic objects and goals, also as givens. In other words, in spite of having identified individualism and naturalism as the big problems in American social theory as early as 1937–8,17 Parsons had managed to retain both problems in earlier versions of his own argument, treating reason and motivation as properties of individuals, rather than as characteristics of “courses-of-action” defined by a system of interaction grounded on an implicit social contract. As a consequence, Parsons’ conception of the “unit act” has been treated as an arbitrary attempt to define the boundaries of events that relies on a structure versus individual dichotomy. The revision solved most of these problems and Garfinkel explains how. Among other changes, in the revision, Parsons built motivation and rationality into the rules of the system of interaction, no longer treating them as 16 This would have happened well after the September 1958 conference, if Garfinkel was only hearing about it through “rumor chains” in 1963. 17 For example, in his “The Role of Theory in Social Research”, Parsons (1938:19) emphasized the distinction between the objects of the natural and social sciences. Sociologists, unlike physicists, deal with facts that are cooperatively assembled in interaction, so that it is necessary to specify how they are made – which, for the early Parsons, meant “fit[ting] them to analytical schemes” – before they can be measured or counted.

Introduction to Parsons’ Primer

25

individual attributes, or choices. For Garfinkel, who always approached interaction as a system of rules and assembly practices that members use to make sense, this was an essential move. Parsons even used Garfinkel’s terms “members” and “oriented action” in the revision. In other words, instead of treating reason and motivation as characteristics of the individual actor, and treating goals as independent ends that the theorist and actor would both need to know about, as he had earlier, Parsons’ argument now specified sets of rules and assumptions, each of which included (and displayed) the motivation, goals, and values that belong to an action-in-its-course in a particular social setting. The definition of a recognizable course-of-action in the revised pattern variable schema included motivations, goals and values as characteristics of the action. The theorist using the revised argument would use these to describe the actions of actors and to identify research problems. An action would be described as motivated in a particular way insofar as it could be seen to orient a particular course-of-action and the ends/values/norms it includes, regardless of what went on in any actor’s head. In summary: In the revised pattern variable argument, actions are themselves “oriented” toward goals (values/norms) in ways that include motivation as a factor. Motivation is no longer a characteristic of the actor, or the end goal, but rather of the action-in-its-course. Actions that an actor can engage in are specified in such a way that to engage in such a course-of-action implies that the actor has a certain motivation: Motivations are included by definition in actions defined by the pattern variables, and the motivation will be imputed to the actor who takes that action. The same is true for rationality. Actions are “oriented” toward values/norms in ways that include rationality as a factor. Actions are rational or not as defined by the system of interaction they occur in. To be seen as engaging in action rationally an actor must orient values/norms in ways that the pattern variables define as rational. It is not enough to actually be rational in one’s thinking (whatever that might mean). The action-in-its-course must adequately and appropriately orient rationality (as defined by the interaction system) as a property of the action in question, as a recognizable part of the stable social system. Whatever is in a person’s head, rationality can only be seen by others (other members) as a display of assembly practices against a background of constitutive rules and assumptions. To be meaningful an action must be recognizably oriented toward the constitutive values/norms of the interaction system in play such that the social theorist and the ordinary members the actor interacts with can “see” that their action is rational. Otherwise, it is not rational regardless of how much “reason” the actor might have in their heads.

26

A. W. Rawls and J. Turowetz

Thus, in the new pattern variable schema, rationality and motivation are characteristics of action not actors. This embedding of reason and motivation in the norms/values of the system of interaction (its rules and expectations) is important to Garfinkel, who wants sociology to focus on the cooperative empirical making of social facts, and not on individuals – the reverse of how he has usually been interpreted. One can see in Parsons’ letters (Appendix #1) and recorded conversations with Garfinkel from 1959 to 1964 that he had begun to focus squarely on interaction and the importance of interaction, just as Garfinkel says he did. Whereas formerly, Parsons’ focus on interaction had been reduced to abstract talk about “double contingency”, he was now referring to actions-in-theircourse that he said only have coherence and meaning as a result of “orienting” situated rules and expectations (that relate to norms and values). Parsons (1963) was also starting to use detailed examples that involved “money” as interaction, and “influence” as interaction. Even though, as Garfinkel is careful to point out, Parsons is talking about rules as the social theorist has formulated them – and not the rules that the actor is actually following – he is now thinking of the social theorists’ rules as an abstract version of the rules that actors must actually be following. As a consequence, Parsons holds the theorist’s version of the rules fully accountable to empirical evidence: The theorist’s rules can be falsified by empirical evidence. Parsons is coming very close to trying to describe the rules for the assembly practices of actual interaction in his revision of the pattern variables, and over the next several years, as he elaborated his argument that money and influence are interaction, he would come even closer. Garfinkel took the position that sociology would be improved by an understanding of Parsons’ revised argument. The problem remains, however, that because Parsons continued to treat the exigencies as a problem that needs to be controlled in order to produce social order, he never gave up the need for authority and inequality, and never accepted Garfinkel’s argument that the actual assembly practices of concrete social interaction hold the key to social order. Because, as Garfinkel says, the rules of the actual assembly practices are very complicated and cannot be imagined, Parsons’ attempts to settle on them analytically always fell short.18 For Garfinkel the exigencies/contingencies are themselves important resources in the process of making social order, which can be seen in the “normal troubles” and sanctions that display, reference, and 18 The work that Karen Knorr-Cetina has done on currency trading could be seen as an extension of Parsons’ argument about money as interaction into the realm of actual concrete practices (e. g. Knorr Cetina and Bruegger. 2002. “Global Microstructures: The Virtual Societies of Financial Markets”, American Journal of Sociology 107(4):905–950).

Introduction to Parsons’ Primer

27

symbolize social order, and therefore no attempt should be made to posit them analytically.

Parsons’ Position as a “Point of Departure” for Ethnomethodology For six decades the conventional wisdom has been that ethnomethodology has roots in the phenomenology of Schutz (and Husserl), the Gestalt psychology of Aron Gurwitsch, and possibly the philosophy of Wittgenstein, but that it has little or nothing to do with Parsons. The way Garfinkel portrays Parsons’ system of interaction in the Primer, however, as a “radical” breakthrough with characteristics much like his own studies of Trust conditions and accounting practices, and as grounded on a constitutive commitment to implicit social contract, combined with evidence from their letters of sustained communication between Garfinkel and Parsons about these arguments, suggests that Parsons played a much larger role in the formation of Garfinkel’s ideas than previously known. That Garfinkel uses descriptions of early “breaching experiments” to illustrate the sanctions that he says demonstrate Parsons’ theory suggests that these “experiments” may even have been designed with that purpose in mind. After all, Parsons had begun developing his conception of an architecture for conceiving of interaction/culture as a system with stable features while Garfinkel was studying with him from 1946 to 1951, and we now know that Garfinkel continued to develop these ideas through letters, discussions, and visits with Parsons through at least 1968. Garfinkel did build on aspects of Schutz, Husserl and Gurwitsch. There is no question about their importance. He also incorporated aspects of Wittgenstein, particularly in developing a conception of culture as a language game (Garfinkel [1960]2019; Rawls 2019). The point is not to diminish the importance of Schutz and phenomenology, but rather to emphasize the fact that Garfinkel also built on Parsons, and that essential aspects of Garfinkel’s conception of interaction, specifically aspects that involve constitutive rules, reflexive double contingency, and the idea of interactional systems, align with Parsons and not with Schutz. The significance of Parsons’ conception of systems of interaction, and the way he formulated actors and action in those systems for the development of ethnomethodology, not only challenges the conventional view of Parsons’ social theory, which was Garfinkel’s main reason for writing the Primer, it also challenges the popular depiction of Garfinkel’s own work as lacking a relationship, or even a relevance, to central theoretical problems in sociology. The

28

A. W. Rawls and J. Turowetz

relationship with Parsons evident in the Primer and in their letters suggests that Parsons himself took Garfinkel and his development of ethnomethodology seriously and considered it important to sociology and sociological theory. In a 1960 letter that he wrote sponsoring Garfinkel for a job at the University of Minnesota, Parsons wrote: “It is my view that the work which is now nearing the publication stage, is going to make an outstanding contribution to the field, and I think he will be a man of steadily growing stature in it”.19 Due to a series of misunderstandings, however, when scholars do think of Garfinkel and Parsons together, they tend to focus on the contrast rather than the similarity, or to connect them through Bales and small group studies, rather than seeing a direct connection to Parsons’ position on interaction (which they typically do not acknowledge). Although Garfinkel had studied with Bales as a PhD student at Harvard (and worked on the Jury Project with Fred Strodtbeck, who carried on the small group studies at Chicago), he was not a small group researcher. For Garfinkel, groups are defined by their “constitutive practices”, not by the dynamics of their people. While Garfinkel respected Bales and Strodtbeck, their work had little in common.20 One factor that likely contributed to the many misunderstandings is that in various writings and letters Garfinkel would emphasize his indebtedness either to Schutz or to Parsons, sometimes emphasizing Parsons, and sometimes Schutz, depending on what suited the situation, the audience, or the reception of his work (Appendix #2). Then there are the misunderstandings created by the Trust argument. Garfinkel built that argument more directly on Schutz in 1956, and in the version of the article published in 1963 it can seem as if Schutz was the most important inspiration for it. But, Garfinkel began to lay out the Trust argument in 1947, and in the early versions in which Schutz does not play the significant role he plays in the 1956 and 1963 versions. One from 1960–1 builds the argument on Wittgenstein (Garfinkel [1960]2019). If versions of the Trust article began moving away from Parsons in 1956, the balance would have shifted back when the relationship between Parsons and Garfinkel intensified again in 1958, as Garfinkel prepared for the Harvard conference by reading Parsons’ newer work and corresponding with other participants (Turowetz and Rawls 2019). 19 This letter is in the Parsons’ Archive at Harvard University, with acknowledgments to Helmut Staubmann for bringing it to my attention. 20 Strodtbeck attempted to hire Anne Rawls to be his replacement at Chicago when he retired. But, his effort, which was supported externally by Harold Garfinkel and internally by William Julius Wilson (department chair), Gerald Suttles, Don Levine, and Strodtbeck himself was blocked by James Coleman – the incoming Chair – whose efforts to make social theory individualistic in the style of economics Anne had joined other prominent theorists in opposing in a special issue of Theory and Society in April 1992 (Rawls 1992).

Introduction to Parsons’ Primer

29

It is doubtful, however, whether changes in how Garfinkel presented his argument ever reflected changes in the argument itself, or in his indebtedness to Parsons or Schutz. The many submissions of versions of this argument to journals, and the rejection letters he received, show that he was experimenting with what was going to be acceptable to the “gatekeepers” of sociology journals and books. As the many rewritten and resubmitted versions of the Trust paper and papers related to it attest, he was more successful in sociology with the version of the paper that built on Schutz, so other versions that he had already written remained unpublished. There is also evidence from Garfinkel’s own hand that his position had not been altered, even by his developing work with Sacks. In a January 14, 1963 letter to Parsons about his plans for teaching the course on Parsons for a second time at UCLA that spring, Garfinkel said that he had found no reason to change his argument: “In the two and a half years since I drafted my mistitled “Parsons’ Primer” I have found no reason to revise my argument that the problem of social order is equivalent to the concept “adequate account of the stable social structures of everyday activities”. During the course I hope to make more of this equivalence than time permitted when I wrote the original draft of the “Primer”.

Garfinkel is telling Parsons that he considers the problem of social order, as Parsons built it (on implicit social contract) to be the same, with a few twists, as an adequate account of the stable social structure of everyday activities, and announcing his plans to make more of the relationship between adequate accounts and social structure in subsequent drafts of the Primer. Garfinkel’s correspondence and interactions with Parsons about the Primer, the manuscript itself, and the way the writing of the different versions of the manuscript overlap with versions of the Trust papers, as well as with the research by Garfinkel and his students that in 1967 became Studies in Ethnomethodology, all suggest that there was no radical change in Garfinkel’s position at this time (or at any other time). Furthermore, Garfinkel vehemently maintained that any assertion of a change in his position that turned him against Parsons, or against sociology, or against Sacks, or aligned him too closely with Schutz was false.21 Garfinkel never chose between Parsons and Schutz. The idea that he changed his mind at different points is contradicted not only by the content of his work, but also by the timeline. Garfinkel wrote the 1963 version of the Trust paper, which heavily references Schutz, and the 1960–1 version which 21

Personal communication from Garfinkel to Anne Rawls.

30

A. W. Rawls and J. Turowetz

takes a Wittgensteinian approach, at the same time that he was working on the Primer, which analyzes Parsons’ approach to interaction in terms that are reminiscent of the Trust argument. During those years he was also working with Sacks on studies of conversation and with Bittner on the research for his 1967 book. Garfinkel worked on them all at the same time. He was always building on both Parsons and Schutz and always insisting that he had built on both.

Garfinkel’s “Fourth Theorem” as a Response to a Suggestion from Parsons In a January 14, 1963 letter to Parsons, Garfinkel introduced what he called his “fourth theorem”, which makes references to his research in ethnomethodology, and said that he was: “Taking as my point of departure your suggestion to me, after you read the Primer, that I consider the social system as an object for the orientation of action”. This is evidence that Parsons had suggested a direction for Garfinkel to take and that he was taking it. In the letter, Garfinkel presented his fourth theorem in the context of three other theorems that he associated with Parsons. This letter is important for several reasons: It gives us evidence of how Parsons responded to his reading of the Primer, and it shows that he made suggestions to Garfinkel for further writing and research; it documents how Garfinkel positioned ethnomethodology in response to that reading; it shows that the suggestions Parsons makes depart from any conventional expectation of how he might have responded to Garfinkel’s position; it reveals a special meaning of “norms” that Parsons shared with Garfinkel; and it suggests that what Parsons meant by oriented action and oriented actors was much closer to what Garfinkel meant than is usually supposed. In spite of these points of alignment, however the letter also highlights differences between Garfinkel and Parsons over their treatment of constitutive conditions and temporality (deriving from the initial difference over exigencies/contingencies). After writing that he was having difficulty working out Parsons’ suggestion (to “consider the social system as an object for the orientation of action”), Garfinkel wrote: “How I wish we could talk about this. May I tell you what I have so far...” and then referenced his ongoing research on accounts (January 14, 1963): “A prevailing theme in my reflections is the social system as authoritative grounds of everyday actions. My reflections on this have ranged through such

Introduction to Parsons’ Primer

31

considerations as different types of accounts, different types of accountability, member claims to their competence in managing actual and practical circumstances, time, symbols, non-contractual elements, and the like. I think I am looking for the essential features of “reasonableness”, “good reasons”, the structural meanings of “with good reason” – in short, recognized rationality”.

Garfinkel is making the point that accounts both orient and symbolize the social system and that his own research on ethno-methods, as it involves accounts and good reasons, might be doing exactly what Parsons is suggesting: showing how the social system is an orientation for these actions. In his research Garfinkel found that accounts often “name” rules of the social system as the “authority” for an activity in a way that is consistent with Parsons’ focus on “authoritative grounds” and sanctions (it is also consistent with Burke and Mills; see also Weider 1974). All of this, Garfinkel says, is something like “reasonableness” or “good reasons” seen in the context of the “social system”, as an orientation toward legitimate (i. e., sanctioned) grounds for action. In line with what he wrote in this letter, Garfinkel (1967: Ch. 6) would title his chapter in Studies on the accounting practices of the clinic, “Good Reasons for Bad Organizational Records”. On January 30, 1963, Parsons wrote a two page single spaced response that identified the first three theorems Garfinkel described in his January 14th letter with his own work, and then addressed Garfinkel’s “fourth theorem” in terms of the problem of incongruity in interaction (and anomie), with an approach to norms that Parsons shared with Garfinkel, and with the question of contingency, which he and Garfinkel both treated as a central question, but on which their solutions differed. He also agreed with Garfinkel that it had to do with “reasonableness” and proposed a connection with the idea of the “reasonable man” in legal theory. Parsons (January 30, 1963:2, italics added) writes: “I therefore suggest that the [4th ] theorem has to do with perception of the system – of norms, that is, not normative order in general, but specifically the component we refer to as norms – as a sufficiently stable object so that orientation in terms of this perception is meaningful. I would connect disorientation in this context very definitely with the concept of anomie”.

Parsons correctly identifies the incongruity Garfinkel is talking about as a problem at the level of the “definition of the situation” that is different from deviance or motivated compliance. The essence of the issue for Garfinkel is that incongruity is not deviance: It is a problem of sensemaking versus nonsense, of anomie (literally a lack of constitutive rule orientation). The argument is that

32

A. W. Rawls and J. Turowetz

in the absence of competent practices that orient a commitment to constitutive rules, actions have no shared meaning. Parsons demonstrates his understanding of this. Notice also that Parsons writes “the component we refer to as norms” (to which we have added italics) indicating a special use of that concept that he shares with Garfinkel. The discussion of norms “as a sufficiently stable object so that orientation in terms of this perception is meaningful” is more situated and interactional in this letter, than is usually associated with Parsons. What Parsons means by norms/values is not what most scholars think he means, as Garfinkel makes clear in Chapter VI. Parsons finds Garfinkel’s conception of “reasonableness” and what a reasonable person could be expected to do important. According to Parsons (January 30, 1963:2): “It seems to me that this is quite different from the problem of motivated compliance. Disturbance at this level does not imply alienation in the usual sense of incapacity to accept normative obligations. It lies rather at the level of definition of the situation. What in my paper on influence I have called the “dimension of justification” seems to me to be the connecting link looked at from the point of view of awakening of trust, rather than having it. That is a justification as a part of the complex of persuasion in a reference to a meaningful context of behaving in the way the persuader wants behavior to occur”.

In referring to disturbances at the level of “definition of the situation” as the object of the fourth theorem, and comparing it to his own work on influence/persuasion and the awakening of trust, Parsons demonstrates both understanding and misunderstanding. He has read the Primer and identified aspects of his own position in Garfinkel’s rendering of it. But, in this discussion of incongruity and disturbance, the differences between Parsons and Garfinkel emerge over the idea of trust as gradually acquired (Parsons’ view), versus Trust as a prerequisite for coherence (Garfinkel’s view). Parsons proposes a connection between the issue of trust and the doctrine of the reasonable man in legal thinking, treating trust as something that “awakens” with persuasion, by which he intends an interactional process (January 30, 1963:2): “Mutual trust between is the premise on which this type of persuasion can be expected to operate. This seems to me to fit with your treatment of the references you make to the problem of rationality and good reasoning. I very much think it is connected with the concept of reasonableness and that for example the part played by the doctrine of the reasonable man in legal thinking is a very important example of what you are talking about”.

Introduction to Parsons’ Primer

33

While there are points of convergence here, for Garfinkel, incongruity can only be avoided by participants who meet Trust conditions (who orient the same rules/expectations in the situation, who are competent to use them, and who expect the same of others that they expect of themselves, and expect the others to expect it of them, etc.). It is interesting in light of this difference that Parsons identified Garfinkel’s argument about incongruity and its role in ethnomethodology with his own position on the “dimension of justification”, not seeing that its prospective focus on the constitutive criteria of assembly practices makes it quite different. Garfinkel was so encouraged by this response that he sent a letter to Egon Bittner (February 5, 1963) along with a copy of Parsons’ letter and said: “Enclosed is a copy of Parsons’ reply. I know it will make you as happy as it makes me... with Parsons’ reply in hand, and with him so deadly serious, and us so encouragingly on the right track, I want to make more of what got started there and soon... Notice the last paragraph.” Overall, however, in this letter Parsons formulates the central theoretical issue in terms of the temporal dimensions of action “in-its-course”, in relation to the general “medium” of interaction, highlighting the convergence between their positions on interaction and the degree to which Parsons was actually talking about objects, actors, and actions as the creations of constitutive practices – actions-in-their-course – defined by an underlying social contract: “I think the discussion of influence and the general paradigm of sanctions as presented in Figure 3 of the technical note to the Power paper may give some suggestions of how this [4th ] theorem should be formulated. I note on page 3 of your letter, bottom paragraph, two particular themes that seem to be crucial. The first is the relation of trusted signification, to which you refer. The question is that of the basis of trusting the apparent meaning of expectations in interaction. The other is your reference to the temporally extended course of action. The problem of trust—that is to say, arises with reference to the temporal dimensions and the definition of expectation with reference to possibilities and probabilities of change. The relation of this in turn to the whole conception of the generalized medium seems to me to be crucial. Moreover I think that the bearing of temporal extensions on such concepts as contract (as for example Leon Mayhew develops it) is also very crucial indeed”.

Parsons is referring Garfinkel to aspects of Parsons’ own position that he feels are relevant to Garfinkel’s development of the fourth theorem. In doing so, however, Parsons highlights two points on which he and Garfinkel actually differ: 1) “trusted signification” as the “basis of trusting”; and 2) that coursesof-action are temporally extended. In Parsons’ analysis the focus is on achiev-

34

A. W. Rawls and J. Turowetz

ing and maintaining trust, and the fact that action is temporally extended makes this more difficult, because trust must be maintained across change and contingency. For Garfinkel, by contrast, Trust is a pre-condition for meaningful interaction that is built into the mutual commitment to constitutive conditions underlying interaction. Without Trust interaction fails. Temporal extension for Garfinkel affords the possibility of layering a back-and-forth sequence of moves that can effectively display whether Trust conditions are being met and whether they confirm or disconfirm meaning. In other words, for Garfinkel temporality (sequentiality) is a solution, whereas for Parsons it is a problem. Garfinkel is making better use of double contingency as a resource than Parsons, who introduced it. But, it is important to note that Garfinkel’s use of temporality aligns with Parsons’ and that Parsons is making important suggestions that end up in Garfinkel’s work.

General Overview of the Primer Argument Ultimately, what a close reading of the Primer shows is that Garfinkel and Parsons share a core approach that is similar in many respects. They are similar in positioning social interaction/culture at the center of social theory as an independent level of social action that is organized as a system of interaction; in their treatment of both actor and action as situated matters evident only “intheir-course”; and, in embedding temporality in the context of an underlying constitutive agreement or social contract. Where they differ is mainly over how certainty relates to contingency, and how that difference effects things like Trust, authority, equality and stratification. The temporal character of double contingency, which for Parsons is an unavoidable problem because change over time produces more contingency, is for Garfinkel a solution because the layers of actual contingency over time and change, and their orientation toward sanctioned expectations, make it possible to use sequential positioning as a resource for making meaning and achieving certainty on particular occasions (an aspect that he and Sacks would develop in their studies of conversation). Garfinkel’s position is similar to the mathematical insight that three uncertainties can make a certainty – a counterpart to the idea of triangulation. Garfinkel is arguing that the ordered contingencies of talk work like this when, and only when, all participants are orienting the same set of expectations about the order properties they are “triangulating” so to speak. When Parsons says that the relationship between temporal extension and contingency is “very crucial indeed” he is referencing this issue.

Introduction to Parsons’ Primer

35

Parsons is obviously receptive to Garfinkel’s ideas, understands them, sees their relationship to his own, and makes constructive contributions to the development of Garfinkel’s thinking. In response to comments from Parsons, Garfinkel would write Chapter VI and produce a new introduction for Chapter V. He would also situate his own ongoing research in the theoretical context of Parsons’ advice in a way that defended Parsons’ overall theory (minus his position on exigencies/contingency and the tolerance of inequality that it entails). The conflict between Parsons’ attempt to limit and control pattern variables and Garfinkel’s multiplication of “indexicals” has this origin. Garfinkel argued in one of his first papers at Harvard (Garfinkel 1946:8) that they differed on this issue of contingency because Parsons had made unwarranted assumptions about how language works instead of subjecting it to examination, which Garfinkel would subsequently do.22 Parsons’ assumptions about language led him to ignore how language, concepts and meaning have order properties in interaction. As a consequence of taking language for granted, Parsons thought that contingency needed to be controlled by authority, rather than being handled by rules and sanctions within the interaction. This led Parsons to embrace stratification and inequality, and kept him from fully embracing the constitutive social fact approach that he had adopted in theory. It also opened Parsons to the political criticism that overwhelmed him in the late 1960’s. Garfinkel’s approach, by contrast, treated contingency as an interactional resource that required equality and reciprocity for its meaningful coordination; grounded as it was in constitutive practices for sensemaking, about which there needed to be an implicit social contract that he called “Trust conditions”. As a consequence, Garfinkel’s approach treats inequality as destructive of meaningful interaction, which inspired his studies of inequality, Race, mental patients, stigma, passing, transsexuality, etc. His approach has the potential to show how authority structures that override the equality and Trust required 22 Garfinkel made this argument in an unpublished 1946 paper he wrote during his first semester as a PhD student at Harvard (see footnote 7). The paper, which was for a seminar Parsons participated in, makes a number of important criticisms of Parsons that Garfinkel would develop in later writings, including the Primer. Among other things, Garfinkel calls attention to the “pragmatic assumptions” by Parsons and his contemporaries of (1) “the fact of communication” and (2) that “the method of understanding works for the actors” (Garfinkel 1946:8), such that they assume or take for granted precisely what needs to be explained: that actors manage to make sense together at all. These assumptions, says Garfinkel, are only necessary because Parsons and his contemporaries focus on the actor’s point of view, raising the problem of how disparate points of view can ever be reconciled and mutual understanding achieved. The problem disappears, however, if “we try backing off a little so as to bring into view not only the original actor, but the fellow he was tangling with” (1946:7); in other words, if we focus on what actors are doing together – interaction and its assembly practices – rather than the private intentions, calculations, and interests of discrete actors.

36

A. W. Rawls and J. Turowetz

by the implicit social contract that grounds constitutive practices not only create exclusion and incongruity/alienation in interaction, but also undermine the social institutions that modern societies depend on (Turowetz and Rawls 2019). As the Trump administration has demonstrated, the “rule of law” simply does not work when the implicit norms/values on which it rests are violated. It is a demonstration of Durkheim’s ([1893] 1933) implicit conditions of contract. Macro structures, as Garfinkel argued, depend on interactional systems for their appearance of solidity. Neglecting this argument has left the formal social institutions of modern society dangerously vulnerable.23 It is to those interactional systems that we will need to turn our attention in order to effectively address the moral, institutional, and social needs of a stable, diverse and democratic modern society. In contrast to the general criticism of Parsons, popular since the 1960’s, that paints him as a rigid and conservative functionalist, Garfinkel crafts a spirited defense of Parsons, depicting him as an innovative and even radical pioneer. In particular, Garfinkel credits Parsons with setting sociology on a modern footing by carving out an independent domain of culture, specified as the assembly practices of “social objects in-their-course”, and guaranteed by the sanctioned legitimacy of an implicit social contract. In so doing, Garfinkel says, Parsons’ established the empirical study of interaction as an essential domain of social theory, with the understanding that conceiving all this in the context of a social contract that undergirds meaningful interaction is a necessary element of any sociological approach to theory or research. There are important points of difference between Garfinkel and Parsons. But, these have mainly to do with the question of whether a modern social system can work well under conditions of inequality, which is a consequence of the difference in their respective treatments of contingency/exigency. The contingency that for Garfinkel is an indexical resource that people can use to make order on the spot, Parsons treated as problematic exigencies, that is, variations in the environment of interaction that the system had somehow to bring under control. Mainstream sociology still takes this problematic position – valuing consistency and generality over the situated resolution of contingency. Garfinkel offered a solution, arguing that the assembly practices required to make certainty from contingency require a degree of reciprocity 23 Durkheim had warned that if social science cannot tell us what the necessary foundations of a free democratic society are then we will not know what to protect and strengthen. In neglecting the implicit conditions of social contract and the tacit rules and assembly practices of both interaction and formal institutions sociology has failed to deliver on either score.

Introduction to Parsons’ Primer

37

(Trust conditions) that Parsons’ treatment of stratification (and hence inequality) as a normal and necessary part of any social system could not provide for. Garfinkel’s position echoes Durkheim ([1893] 1933) in maintaining that inequality is a threat to the solidarity of a modern social system. This is one of the central tenets of Durkheim’s position that Parsons did not fully grasp (Rawls 1996, 2019). The differences between Parsons and Garfinkel on contingency, and hence on Trust and equality are important, not only for understanding the relationship between them, but also for understanding misconceptions in social and cultural theory more generally. These differences would eventually get in the way of the theoretical fusion Garfinkel was trying to effect, particularly as Parsons came under increasingly heavy criticism for the inherently conservative character of this aspect of his position. But, for at least 20 years, Garfinkel and Parsons both seem to have worked with the idea that Garfinkel’s interactionism might be able to fill out the empirical details of the cultural/interactional level of Parsons’ social system, and Garfinkel continued to hope that Parsons would see the important implications of actual constitutive aspects of social practice for an overall conception of the social system, and modify his problematic position on the exigencies, with corresponding changes on Trust, stratification, and inequality. Without understanding the innovative character of what Parsons was proposing for sociology, it has been impossible to understand and appreciate either Parsons or Garfinkel. Without understanding how and why interaction – rather than structure and function – became the central theme of Parsons’ position after 1949, or how Garfinkel saw his Studies in Ethnomethodology as taking up and elaborating the interactional dimension outlined by Parsons from 1949 to 1963, it has been impossible to appreciate either the radical and transformative character of Parsons’ theory, or the serious theoretical and sociological character of Garfinkel’s empirical transformation of it. Ethnomethodology addresses sociology’s central problem of social order, and does so in a framework that is built on Parsons’ conception of culture/interaction, in very specific ways. In the Primer, as well as in letters to Parsons, Garfinkel explains that in his studies of accounting practices and accountability (describing the “good reasons” clinic study he did with Bittner), he was showing how the assembly practices of culture are symbolic of – i. e. represent – the broader social structure. The way Garfinkel uses his studies to explain Parsons has points in common with Burke (1936) and Mills’ (1940) on vocabularies of motive, both of which are important theoretical arguments. The Parsons’ Primer manuscript has sometimes been considered a “respecification” of Parsons from Garfinkel’s perspective. However, we find it to be a systematic exegesis of Parsons’ focus on social interaction, and the

38

A. W. Rawls and J. Turowetz

importance of culture as an independent domain of action, in the context of Parsons’ overall social theory and his revised pattern variable argument in particular. In an effort to clarify the innovative character of Parsons’ position – on behalf of both Parsons and himself – Garfinkel decided to write and teach from the Primer. While the manuscript does not elaborate on Garfinkel’s own position as such, it makes it clear how ethno-methods were positioned to play a key role in the specification of the assembly practices of social objects/facts as conceptualized by Parsons and Garfinkel, and does so in terms that would correct weaknesses in Parsons’ approach. These weaknesses left Parsons’ conception of culture out of touch with actual members’ practices, making it look as if Parsons had pictured the social person – or actor – as a “cultural dope” who acts in puppet-like fashion on the basis of rules and categories that the social theorist had defined for them. This is how most critics have understood the position. In the Primer, however, Garfinkel explicitly defends Parsons from this charge, saying that while it was true for the early Parsons, and for most other sociological theories, it is not true of the later Parsons. According to Garfinkel, the later Parsons understood that the actor needed to be situated in the actual world of ongoing contingent social practices to explain double contingency and handle the “exigencies” (even though Parsons still thought exigencies needed to be defined and controlled by authority). In Garfinkel’s view, Parsons’ position was almost fully interactional by 1963. The difference was that Garfinkel located the accountable conditions for ordinary conversation in the actual empirical conditions of interaction – the “Setting of Conversation” that he and Sacks elaborated together in 1967 – treating contingencies as useful resources that can display the orientation of actors to rules, social structures, and underlying social conditions in meaningful self-regulating ways – while Parsons continued to treat variation as a problem. The difference over exigencies/contingencies was never resolved.24 24 In 1967 the collaborative research of Garfinkel and Sacks was prepared for a talk they gave at the ASA meetings that year titled “On Setting in Conversation”. Garfinkel placed five drafts of the paper along with transcripts of his discussions with Sacks (which had been audio-taped) into a binder with that title. The ASA session had been organized by John McKinney, who secured a book contract for the papers. The book edited by McKinney and Edward Tiryakian, titled Theoretical Sociology: Perspectives and Developments, was published in 1970, with an article prepared by Garfinkel and Sacks from the 1967 presentation titled “On Formal Structures of Practical Actions”. According to Peter K Manning who attended the session, Sacks and Garfinkel both attended and presented at the session. Nevertheless, rumors have grown up around the published version of the paper that allege that Sacks had nothing to do with the project and that he and Garfinkel were not working together at that time. As their co-presence at the session, and the audio of their preparation of the talk attest this is false. Currently a book presenting and discussing the importance of these papers, to be titled On Setting in Conversation is being prepared for publication. The origins of Sacks work on conversation in his own research with Garfinkel, which both he and Gail Jefferson insisted on, have for some reason been erased from the history of conversation

Introduction to Parsons’ Primer

39

Discussion of the Primer Manuscript in Detail – Chapter I In the first chapter of the Primer, which is the shortest, Garfinkel states his objectives: “It is the task of this book to recommend Parsons to researchers by teaching the reader how to make use of his work for theoretical and empirical sociological researches” (see p. 111). Acknowledging that Parsons is difficult to read, Garfinkel aims to show readers “how to read beyond what Parsons writes to see what he is talking about” (see p. 111), so that they can appreciate the “originality, rigor, scope, and usefulness” of his “exceptional solution” to the problem of social order. This solution, as Garfinkel will explain it in Chapter IV, is to invert Hobbes’ social contract argument such that all rational action and actors are defined by, and occur within, an implicit social contract, an orientation toward which gives them meaning. The point with regard to an underlying social control, or commitment, is reiterated in Chapter VI. A similar conception of implicit social contract grounds the arguments of Durkheim, Garfinkel, Goffman, and Sacks, and Garfinkel offers his own research as a demonstration of the point. Garfinkel then proceeds to enumerate criticisms that have been made of Parsons’ work, sorting them into four categories: “true but trivial”, “true but irrelevant”, “would be important if true, but false” and “important and true”. Dismissing the first two classes of criticisms in a single paragraph, and spending several pages listing the third, “important if true, but false” criticisms, but not discussing them, Garfinkel devotes the majority of the chapter to a fourth set of criticisms: “important and true”. He does intimate, however, that by the end of the book, the reader will be able to understand and evaluate these criticisms for themselves. In listing the true and important criticisms, Garfinkel first notes that Parsons’ pattern variables “are not an exhaustive list” (see p. 114). This is a serious problem, as these variables are meant to provide for all of the different ways actors could orient situations and objects, including other actors. However, the assumptions made by Parsons about language and the need to control exigencies/contingencies, led him to artificially control and limit the number of contingencies that arise in any interaction. Therefore, although these variables are meant to incorporate interaction into Parsons’ theory, their finite character ultimately limits his ability to explain how actors actually make sense together analysis. Restoring the intrinsic connection between Sacks and Garfinkel is essential to recovering the original import of Sacks’ work. In one letter to Schegloff (March 1974) Sacks referred to himself as “a methodologist for ethnomethodology”.

40

A. W. Rawls and J. Turowetz

in situated interaction. Trying to control and limit contingencies is the main point on which Parsons and Garfinkel differ. The second criticism, that Parsons doesn’t adequately conceptualize the role of time in the structuring of action and situations, and which Garfinkel observes is hardly limited to Parsons, is related to the first: the meanings of objects, situations, and identities change constantly over the course of an interaction – an action-in-its-course – and these contingencies must, somehow, be worked out by participants on an ongoing basis. Because the pattern variables do not provide adequately for contingencies, they have some difficulty dealing with temporality. While Parsons does acknowledge the importance of time, he sees it as a problem because it introduces changes and thus produces more contingency. The pattern variables themselves don’t provide for time, nor does what Garfinkel calls Parsons’ “ingenious” but “unconvincing” solution to the time problem. It is not entirely clear what Garfinkel means by this. But, he invokes time in his letters to Parsons in terms of the “double contingency” of action-over-its-course in a way that suggests that what he calls the “fat moment” (see p. 114) of double contingency was the solution in question. Furthermore, while Parsons does take action-response sequences into account in his theory, he does not specify how they operate in any detail. Garfinkel’s third criticism builds on the first two: Parsons’ structuralfunctional method cannot address the “time of occurrence, pacing, duration, sequence, phasing and the like” of social interaction, instead reducing these elements – the very ones through which meaning and order are assembled – into the sort of “fat moment” characteristic of historical epochs. The upshot of these first three criticisms, all of which involve trying to limit contingencies, is that Parsons has difficulty with contingency and change. Because Parsons has understood that a focus on interaction is the solution to social order, and acknowledged contingency and change as inherent features of interaction, it is a big problem that he still tries to put formal controls on contingency and change, instead of dealing with them as they occur in interaction – and Garfinkel will elaborate the contradictions that result over the course of the manuscript. Nevertheless, Garfinkel argues that Parsons’ insights about the importance of interaction, situated within an implicit social contract, make his theory the one to build on. Moreover, in Chapter VI, he argues that Parsons is moving in a direction that allows him to deal with change. While Parsons recognizes the centrality of interaction to any adequate theory of social action, he still treats it as a problem to be solved, rather than a potential locus of order in its own right – and this Garfinkel also highlights as a problem. As Lidz and Staubmann (2010:11) note, a constant theme in Parsons’ work is that there is “always a need for social control” given that, “In

Introduction to Parsons’ Primer

41

every situation where social interaction takes place, there is a risk of deviance on the part of some or all actors with respect to the expectations of others”. Garfinkel also recognizes the need for control and sanction. But, unlike Parsons, he situates sanctions and controls within the interaction and its rules and underlying commitments. This is consistent with Durkheim who also treated the problem of order and sanction in cases of constitutive practice as one of self-regulation, not deviance; a situation where noncompliance creates normlessness – anomie – for everyone. By contrast, Parsons continues to treat order and sanction as a top down problem of social order in problematic ways, in spite of embracing interaction as the center of social order, and in spite of Durkheim’s (1893) warning against maintaining a top down position. Fourth, Garfinkel writes that “The empirical grounds to justify Parsons’ assumptions that a single integrated set of value premises is a necessary condition of stability of a social order and that this consists in the United States of ‘instrumental activism’ remains to be demonstrated” (see pp. 114–115). In other words, Parsons assumes the need for external control of a particular kind with no empirical basis for making the assumption. He will take up the point again in Chapter VI. Collective orientations, Garfinkel argues, need to be demonstrated, not stipulated, and he will argue that his Ethnomethodological research can supply the needed demonstrations. However, his acknowledgement that criticisms of Parsons’ analysis of American values are justified evidences continued disagreement with Parsons’ insistence on this kind of value-uniformity. While Garfinkel takes up this issue in more detail in Chapter VI (on values), and alludes to it also in Chapter VIII (on stratification), we note that the postulate of a single set of values, which, for Parsons, delineates correct, preferred ways of acting, requires a standardization across situations that Garfinkel argues ultimately defeats Parsons’ own objective in focusing on situations of interaction. It also risks subordinating the interactional preferences of specialized and minority social groups to those of the dominant group, which is problematic not only for interaction across cultural diversity, but also for questions of justice and equality. The knowledge practices of a modern society require a freedom from constraining comprehensive/standardized beliefs and values that Parsons tried to, but ultimately did not, provide for. It is Garfinkel’s aim to remedy this shortcoming through studies of ethno-methods that show how participants handle variation and contingency through actual concrete assembly practices and the underlying values they orient, in each particular situation. If Garfinkel can do this, he can free Parsons’ theory from the need to impose value uniformity (and hence authority and inequality), leaving what remains as an adequate foundation for his own work.

42

A. W. Rawls and J. Turowetz

Finally, Parsons asserts this value unity on analytic grounds, not empirical ones. The way he sets up his theory as what Garfinkel calls “a meaning transformation system” assumes that it must work as stated. In Chapter VI, which he wrote in response to suggestions from Parsons, Garfinkel argues that the way to investigate values in U. S. society, and elsewhere, is to examine which preferences actors orient in actual practice and, more importantly, which ones they enforce as maxims of conduct, i. e. by sanctioning violators. Neither Parsons nor his critics do this, arguably because, as White Protestant males, they take the existence (and basic correctness) of uniform U. S. values for granted. By contrast, Garfinkel’s position as a Jewish minority in 1950s academia made him acutely aware that there was more than one order of preference, and that none was intrinsically more valid, legitimate, or valuable than the others. In Du Bois’ (1903) terminology, Garfinkel possessed a “double consciousness” that enabled him to see beyond what White male academics took for granted as American Values. Thus, when Garfinkel writes in Chapter VIII that in Parsons’ society, “some people plainly stink” (see p. 301), he means that the people at the top, who hoard opportunities and resources, think this way. In their view, those at the bottom and on the margins – the “Blacks, Jews, Reds, Criminals” (Garfinkel [1947]2012) and so forth, including Garfinkel – “stink”. He wrote about this extensively in his unpublished papers of 1946–8. What Garfinkel saw was that those who the rich and powerful say “stink” have developed their own preferred orders of interaction that are viewed as illegitimate, or deviant, by the dominant social class. But they are considered legitimate in their own local, situated interactions. Parsons could not conceive of multiple constitutive orders of preference, at least not in non-hierarchical terms. This is not to say that Parsons ruled out the existence of conflicting general and specific values corresponding, respectively, to the whole society and to individual groups within it, but that he viewed their relationship hierarchically, which is problematic. Fifth, Garfinkel suggests that “readings of Durkheim and Weber” that differ from Parsons’ reading “could have serious implications for [his] work”. Since Garfinkel does not elaborate here, it is difficult to be sure what he has in mind. We do know that many scholars read Weber in a more interpretative way, and that Durkheim at the time was widely considered a rather simpleminded positivist-functionalist – as Robert Merton (1934) famously painted him. Readings of both have subsequently emerged that align more closely with the interpretation of Parsons that Garfinkel proposes, even suggesting that Durkheim had a better grasp of the importance of the details of spontaneous and self-regulating constitutive practices in modern science and occupations than Parsons (Rawls 1996, 1997, 2008, 2019).

Introduction to Parsons’ Primer

43

Sixth, Garfinkel agreed with Parsons’ critics that there needed to be more empirical testing of his arguments. In Chapters V and VI, Garfinkel lays out a method for doing this empirical testing, which is exemplified by his Trust paper (1963), and various of his research projects as reported in Studies in Ethnomethodology (1967). This testing consists of inducing, or observing, norm violations and seeing how strongly a society’s members sanction those violations, thereby revealing value-commitments that are ordinarily taken for granted and invisible in everyday life. It is a method for artificially producing the awareness (that Du Bois ([1903]2004) called double consciousness) that people in marginal identities acquire naturally through the experience of exclusion. This procedure of triggering sanctions also aligns with Durkheim’s ([1893]1933) proposal to treat sanctions as the empirical sign of values and of changes in values in modern society. In sum, while Garfinkel works in the Primer to show significant convergences between Parsons and the method and theory behind his research in ethnomethodology, Garfinkel’s review of “important and true” critiques of Parsons also indicates points on which they would never converge. In particular, Parsons did not find an adequate way to deal with contingencies – positing external controls to the end. As a consequence, Parsons refused to look at members’ actual assembly practices, insisting that they must be formulated analytically. Parsons could not see that interaction had its own inherent order properties because he didn’t look. His system reconstructed a version of order properties analytically. But because his conception of order properties was not based on actual practices, his argument ultimately glossed those order properties in abstract and misleading terms and failed to see the damage that inequality did to them. By contrast, Garfinkel always treated concrete order properties as the essential element in social order that everyone was overlooking, beginning with his ethnographic research on accounts and presuppositions in the 1940’s. In the early 1960s Garfinkel and Sacks found that they could locate the order properties of interaction in the details of recorded conversations, leading to a shift away from looking at sanctions and accounts as evidence of underlying values, and looking instead at naturally occurring repairs and realignments, turn-by-turn, that succeed and therefore don’t get sanctioned.

44

A. W. Rawls and J. Turowetz

The Primer in Detail – Chapters II through IV In Chapters II to IV of the Primer, Garfinkel focuses on Parsons’ formulation of the problem of social order, contrasting it with other conceptions and arguing for its originality in terms of its empirical verifiability and how close it stays to interaction. In Chapter IV he says the problem of social order as defined by Hobbes is the starting point for understanding Parsons because he arrived at his formulation of the problem of order by inverting Hobbes to put social contract first. This formulation of the problem is important because as Garfinkel writes at the start of Chapter II, “The problem of social order... embraces almost all of the problems to which Parsons’ work is directed” (see p. 115) and that the whole of Parsons’ work “can be pulled together by conceiving it as a solution to the problem of social order”. Garfinkel distills this “solution” into two concise theorems, which he spends the rest of Chapter II, and indeed of the Primer, unpacking (see p. 115): “(1) The real social structures consist of institutionalized patterns of normative culture”; and, “(2) The stable properties of the real social structures are guaranteed by motivated compliance to a legitimate order”. All theories of social organization propose a solution to the problem of social order, and all follow a set of rules for transforming objects of everyday concern into theoretical constructs. What distinguishes Parsons’ approach from others, according to Garfinkel, is not that it proposes a solution to the problem of social order, or that it transforms commonsense objects into theoretical ones, but how it does these things, by putting an implicit social contract first and transforming events into social objects through the assembly practices of an organized system of social interaction conceived as culture.

Chapter II in Detail – Sociology as a “Sense Transforming” Operation The bulk of Chapter II of the Primer is concerned with the logic of social theory, or how what Garfinkel ([1948]2006:101) elsewhere calls “seeing” in new ways, both sociologically and ethnomethodologically, differs from other ways of seeing, laying out its essential features, and considering the basis for choosing between competing theories. Drawing on what Garfinkel calls the “analytical philosophy” of the phenomenologists (not to be confused with the “analytic” approach to philosophy that became prominent in the U. S. and Britain after the Second World War), Garfinkel observes that the objects of

Introduction to Parsons’ Primer

45

sociological theory are assembled through “sense transforming” operations (see p. 117): The sociologist takes an everyday object from the lifeworld (which can include people’s own theories and beliefs about their actions or environment) and, through sanctioned disciplinary procedures, such as counting, coding, measuring, comparing, typifying, contrasting, etc., transforms the everyday object into a theoretical construct. Garfinkel is not talking about phenomenology here. He is referring to Parsons’ theory as a sense transforming theory. Such sense transforming operations are not confined to social scientists, either, but to anyone who steps outside the attitude of everyday life and orients the world according specialized criteria; although the criteria will differ depending on the ground rules, of the “science”, or bureaucracy, they orient. Garfinkel offers the example of his uncle trying to persuade a bureaucrat to increase the allotment of fuel oil to heat his home during WWII (see p. 118). From the uncle’s lengthy story about his circumstances (including an angry wife, sick children, and a cold house), the clerk selected only those features relevant to making a decision that was accountable from the perspective of the bureaucracy he worked for. The square footage of the house, how many rooms it had, the make of its heater, etc. are relevant. By attending only to what his office protocols treated as relevant, the clerk, acting as a government official, oriented Garfinkel’s uncle as a “homeowner”, who during the war, was entitled to a certain allotment of fuel based on specific measures of his home’s characteristics. Other ways of orienting him – as a husband with a complaining wife, the father of sick children, etc. – were not relevant for deciding what, “from the standpoint of the administered [bureaucratic] form, was ‘the case’”.25 Parallel operations are at work in Garfinkel’s other examples: a social worker transforming a client’s story about what brought him to a psychiatric clinic into a set of “presenting complaints”; Garfinkel and colleagues speeding up a tape of a psychiatrist talking to an experimental subject in order to focus on unintelligible “bursts of talk” and their relative frequencies; and a sociology professor transforming a student’s account of an armed robbery into an example of the illegitimate transfer of property rights between persons (see pp. 119–121). In each case according to Garfinkel, someone follows the procedural rules of a bureaucracy, legal institution, formal organization, or scientific discipline, to transform one description of something into another description, changing its sense in the process (see p. 118).

25

This example appears for the first time in Garfinkel’s writing in 1946.

46

A. W. Rawls and J. Turowetz

The outcome of the sociological sense transforming operations that Garfinkel attributes to Parsons’ theory is the “real social structures” that make up sociology’s subject matter (see p. 117). These “real social structures” are a theorist’s construct. Like the story constructed by the clerk about fuel requirements, the transformation that results from the use of Parsons’ theory is treated by Parsons as the “real” state of the world from the theorist’s perspective. Not only does social theory perform sense-transforming operations, but different theories result in different versions of the “real” world. The particular rules a sociologist follows will depend on how they solve the problem of social order, and compliance with these rules defines what, according to Garfinkel counts for them as an “adequate structural analysis” (see p. 116). The outcomes will differ based on the theory, not based on the real world they are claiming to describe. For example, a “rational action” theorist’s proposed solution to the problem of social order might be a state of equilibrium among actors with competing interests. This solution would require that the theorist’s society be populated by certain kinds of actors (homo economicus) with particular types of interests (rational) whose behavior could be modeled in terms of utility functions. Given such a society, an adequate structural analysis would specify how equilibriums could be achieved in different settings. In other words, ordinary accounts and events would be transformed into a rationalized account of instrumental behavior by such a theory. And such an analysis would be correct by the theorist’s criteria, since, as Garfinkel says: “The correctness of an assembly is defined in no other way than by listing the rules of the assembling procedures that are treated by the assembler as correct procedures” (see p. 121, emphasis in original). There can be no comparison between the outcome of the procedure and the actual world. Since the procedure has transformed the world there is nothing to compare. Accordingly, the way to select among competing theories of this type is not to ask which is “correct”, in the sense of best representing an independent social reality, but to specify their respective internal criteria for correctness, as well as the assembly procedures they specify and the features of the “real social structures” these produce, and then see whether they would produce the world we live in. According to Garfinkel: “This permits different claims about matters of fact... to be compared” (see p. 123). But, from his perspective it is not ideal. All social theories, Garfinkel observes, necessarily begin by assuming empirical objects, since the sense transforming operation needs something to operate on. And all of these theories produce descriptions of “theoretically” possible societies. However, they do not all yield equally rigorous, verifiable empirical

Introduction to Parsons’ Primer

47

propositions. Contrary to Parsons’ critics, who commonly allege that his theory is too abstract, Garfinkel argues that it is “at every point in its development controlled or controllable by empirical knowledge” (see p. 116). According to Garfinkel Parsons’ system “is astonishing in the extent to which its correct use as a method involves the user in literal and rigorous procedure, and in the empirical propositions that such literal procedure produces” (see p. 123). In later chapters, Garfinkel will elaborate on those propositions that direct the researcher’s attention to the actual practices that a society’s members use to assemble stable social objects, the courses of action they treat as correct, and those they treat as incorrect and sanctionable. Garfinkel liked to say that Parsons was a better phenomenologist than Schutz, and Parsons’ awareness of the dependence of social objects on assembly practices illustrates this point.26 However, while Parsons knew that this process of sense transformation needed to be specified empirically, and he understood that it occurred within the definition of the situation and its implicit and sanctionable social contract, he fell short of appreciating the need to base his system on an examination of actual assembly practices. What Parsons does not do is to consult concrete practices to get the actual rules, rather than formulating rules analytically. What Garfinkel wants is for the rules that the sociologist uses to make this transformation to be the same rules, or assembly practices, that the social member is using to make order and sense in their everyday life. Garfinkel concludes Chapter II with a discussion of uses to which Parsons’ theory can be put by researchers, a point that he will take up again in Chapter VI. He begins by distinguishing systematic uses of Parsons’ theory, from “ad hoc” uses (Garfinkel subtitled the Primer “ad hoc” uses). Systematic uses, which involve relating different parts of his theory, he says, are not constrained by the logic of the system as a whole, but by “literal and rigorous procedure” (see p. 123). Garfinkel then describes three ad hoc uses of Parsons’ theory. First, researchers can use Parsons’ “structural elements” – culture, personality, collectivity, roles, and organism – to transform any description of social structure into a form that distills its essential features. Garfinkel observes that this process does not involve selectively abstracting from the original description, but rather reconstructing it. This reconstruction changes the description’s sense, and therefore changes the theoretically possible occurrences it provides for, as well as the empirical methods that are relevant for investigating them. In Garfinkel’s example of speeding up a tape recording, which reconstituted an ordinary conversation as “a set of possible serial bursts of utterances” (see p. 125), the original object was changed into one exhibiting rapid bursts of 26

Personal comment to Anne Rawls.

48

A. W. Rawls and J. Turowetz

talk as one of its observable features. It could then be investigated in terms of these features, which were not available as part of the initial conversation observed in “real time”. What we have is a new object, not an abstracted version of the original object. It is not a question of deciding, Garfinkel says, how closely the new object approximates the old object, but rather of investigating the new object on its own terms and with reference to the possible empirical occurrences it entails (see pp. 125–126). Of course, whether these possible occurrences actually happen must be determined through observation. This is a step that users of Parsons’ theory often overlook, but which, according to Garfinkel, Parsons himself recognized as essential (see p. 126). A second use of Parsons’ theory is to use “rules of interpretation” to transform, and relate, his theoretical objects. This means, for Garfinkel transforming objects that were already previously transformed according to the rules of Parsons’ theory, and then specifying their relationships (see p. 128). For example, the researcher could specify relationships between concepts like culture, personality, and collectivity, etc., as Parsons does in his functional analysis of society. In contrast to social structures, these rules of interpretation are not observables. Instead they relate observables. Nonetheless, Garfinkel argues in Chapters V and VI that these rules are still amenable to empirical control, given detailed observation of actual concrete practices. Finally, Garfinkel proposes a third use of Parsons that combines uses one and two to generate hypotheses about social structures. Garfinkel closes the chapter by stressing that Parsons’ theory is not about the social system: rather, he says “the social system is the theory” (see p. 131). Because Parsons doesn’t think that contingencies can be self-ordering, from his perspective, there is no social system without the theory. It is, as Garfinkel would say later (Garfinkel 1988), a case of “Parsons’ Plenum”. As a set of rules for theorizing, he says, it constitutes a method for determining the logically necessary relationships between observable actions and objects, and how the theorist can “arrange his thoughts about a set of merely possible phenomena” (see p. 131, emphasis in original). However, logical necessity does not imply empirical necessity. Garfinkel notes that people do not have to act as they do, or as the theorist says they do (see p. 132). Rather, according to Garfinkel, Parsons’ theorist is concerned with the conditions under which stable social structures are theoretically possible (see pp. 131–132). Empirical inquiry is required to determine which possible phenomena are in fact actual phenomena. Or, in structural-functional terms, whether logically related conditions and consequences are also empirically related.

Introduction to Parsons’ Primer

49

Chapter III in Detail – The “Discovery of Culture” In Chapter III, Garfinkel expounds on the key features of the problem of order, as Parsons conceptualizes it. For Garfinkel the assembly practices and procedural rules of situated interaction, taken together, comprise culture. Thus, when Garfinkel says that social order consists in the relationship between social organization and culture, he means that society’s members accomplish order through the reflexive use of assembly practices in a context of implicit commitment to a system of interaction. In contrast to those who treat the durable products of assembly practices over time in the abstract as culture, Garfinkel is focused always on the ongoing making of culture in situations. He says that this is Parsons’ goal as well. The theorist’s (Parsons’) goal, he says, is to “lay out as a program for actors how they must go about acting in concert with each other” in order to produce the social order that is an empirically observable feature of the society being investigated (see p. 147). The way to formulate such a program, he says, is through structural analysis, so that “the problem of social order is another way of talking about structural analysis” (see p. 148, emphasis in original). According to Garfinkel, every theory provides a version of “correct structural analysis” and each in turn results in a different possible society, and therefore in a different account for social order (see p. 148). But, in Parsons’ view (and in Garfinkel’s) these are always accountable to the observable society, without the excuse that the observable society is just a messy version of some ideal. People must live and make sense in the observable society. It is the researcher’s task to decide which of the many possible theoretical analyses would produce the recognizable features of the actual empirical society that is under study. Starting with the Structure of Social Action, Parsons (1937) worked to set out the problems that any theory of social order would need to solve and the criteria it would have to satisfy. These criteria are, according to Garfinkel, pre-theoretical, and he differentiates theories according to how they solve the problems these pre-theoretical problems pose (see p. 134). Garfinkel observes that Parsons’ work is “exhaustive” in two senses: it provides a method for systematically relating any set of social actions and, on a more fundamental level, it specifies the necessary properties of social action and its products. It is in the second sense that Garfinkel says that Parsons’ project is “radical” because the properties of social action he specifies include the dependence of social objects on the definition of the situation and an implicit social contract (see pp. 134–135). Social objects are created in a constitutive context or they don’t exist. However, it is precisely the radicalness of Parsons’ theory that Garfinkel says, “Among American sociologists... has gen-

50

A. W. Rawls and J. Turowetz

erally been overlooked” (see p. 134). It is, as a consequence of this oversight, he argues, that Parsons has been misunderstood and criticized. Because of its exhaustiveness, according to Garfinkel, Parsons’ theory allows for the identification of elements of “ideology” in other theories. A theory is ideological when it assumes the truth of propositions that must be empirically demonstrated. Parsons is also able to identify which of his own propositions require evidence, as well as the type of evidence they demand. This is not to say that he always succeeded. As Garfinkel points out Parsons doesn’t succeed in his analysis of American values or in his assumptions about language. What is important is that Parsons furnishes criteria by which theories can be judged, his own included. Much of Chapter III is organized around what Garfinkel calls “the discovery of culture” (see p. 135), a discovery that he attributes to the joint efforts of anthropology and sociology (see p. 136). In a passage that would reappear five years later in Chapter III of Studies in Ethnomethodology (1967:76–77), Garfinkel describes the discovery of culture as “the discovery from within society by its members of the existence of commonsense social structures...” (see p. 136). As Garfinkel says: “In a word, it consists of the grounds of further inference and action that persons as conditions of the enforceable character of their claims to bona fide membership in the group expect each other to employ and subscribe to” (see p. 136, emphasis in original). This is essentially a description of Trust conditions, and in attributing this description to Parsons, Garfinkel is attributing to Parsons a discovery that is essential to ethnomethodology. Garfinkel describes culture as consisting of commonsensically known social structures, the rules for their assembly, and the ways they are enforced. Accordingly, culture constitutes a domain of analysis that allows the sociologist to explain how social order is produced. Parsons had been arguing for the autonomy of culture since at least the late 1940s when he made it the focus of his 1949 ASA Presidential Address (Parsons 1950). By conceiving of culture as constitutive of social order, rather than as a repository of folkways, customs, habits, and traditions, as earlier scholars (e. g. Comte, Giddings, Taylor, Boas, Malinowski) in both sociology and anthropology had done, Parsons provided sociology with a method for explaining how social objects, including actors themselves, are assembled during interaction through the cooperative and sanctioned use of culture. This discovery of culture as assembly practices was of monumental importance. According to Garfinkel “It was a further distinguishing feature of this vast discovery that social reality is conceived of and treated as an object of scientific study” (see p. 136). In other words, culture is not to be assumed as a background of folkways and values, but is to

Introduction to Parsons’ Primer

51

be discovered as the actual sequential, time relevant assembly practices of social interaction. It is this treatment of social reality as a scientific object that distinguishes sociologists from what Garfinkel calls the “natural critic”. All members of society make critical observations about the social world: they doubt, question, complain, speculate, and the rest. What differentiates “seeing sociologically” from other ways of orienting society is that the sociologist takes members’ own accounts and descriptions as objects of inquiry. Sociologists are interested in how the commonsense world is constituted as an objective environment toward which members can take up attitudes, critical or otherwise, in the first place. This is as true of Parsons and other sociologists as it is of Garfinkel. According to Garfinkel (see p. 138), in transforming descriptions of society that are produced by and for its members into objects of scientific analysis, the sociologist subjects them to what Parsons calls “instrumental criticism”, identifying the practices through which they are assembled. Garfinkel argues that instrumental criticism must solve four problems, which he terms sensibility, objectivity, warrant, and causal texture. (1) The problem of sensibility he describes as one of defining the nature of the objects under investigation and their relationships: for example, what constitutes a “group”, and how are groups related to one another? Different theories answer these questions differently, furnishing different propositions, empirical observables, and criteria for what qualifies as a correct inference. For example, if we conceive of groups as aggregates of pre-existing individuals, as Hobbes did, we are led to ask about the conditions under which members will join, cooperate, and leave groups. If, however, we understand self and group as co-constitutive, already defined by an implicit social contract and its rules, such that the situated identities and interests of group members do not precede the group’s formation, we would ask a different set of questions, such as: how do actors orient themselves and others as group members and why? What are the procedural rules they follow in their interactions, and how do these rules work? Which behaviors are treated as sanctionable and why? In other words, rather than taking the individual as the primary unit of analysis, the social contract approach accords primacy to the Interaction Order (Goffman 1983, Rawls 1987) of the group (the rules of its assembly practices and the objectives of achieving meaning, objects, and self they orient), which are inseparable from its concrete operations (a discussion of these matters can be found in the discussion of bridge players in Chapter VI). (2) The sociologist, according to Garfinkel, solves the problem of objectivity by using procedures capable of producing findings that hold irrespective of who evaluates them; the results, that is, should hold not only for sociologists,

52

A. W. Rawls and J. Turowetz

but for “Anyman”, so long as that person subscribes to the theorist’s rules of observation and inference. Put this way, the problem of objectivity is basically that of how to achieve universality, at least within a community of inquirers bound by a set of procedural rules, and its solution, Garfinkel says, furnishes “the meaning of the criterion of communality” (see p. 139). (3) The problem of warrant, Garfinkel explains, is to determine if “a proposition can be correctly used as grounds for further inferences and research” (see p. 139). It is up to the theorist to decide which rules of observation will define what counts as a fact. Once a decision has been made the theorist’s criteria can be consulted to evaluate their findings. In practice, however, theorists often accord factual status to propositions that seem to fit with prior observations but which have themselves not been observed. On this point, Garfinkel refers the reader to his paper, Commonsense Knowledge of Social Structures, for further elaboration (see pp. 139–140). (4) Finally, the problem of causal texture involves ordering events in relationships of condition and consequence, a challenge made all the more difficult by what Garfinkel calls “a disconcerting irascibility that persons exhibit” (see p. 140), meaning that people are more complicated than the theorist provides for and don’t often do what the theorist predicts. These four problems distinguish sociologists not only from natural critics, but also from other scientists, particularly those in the natural sciences. Natural scientists do not deal with social objects, or at least they do not believe they are dealing with social objects. As such, natural scientists in the 1950’s and 1960’s (when Garfinkel wrote the Primer) typically thought they did not need to attend to the practices through which their objects were assembled. Later work in social studies of science (e. g. Lynch 1993) that was inspired by Garfinkel would reveal problems with this assumption. By contrast, sociologists are concerned with social facts that are assembled in interaction by actors who have their own solutions to sociological problems, which are different from the solutions of the social theorist. As Garfinkel says: “The crux of the difference resides as the fact that any member of society has some solution to these same problems” (see p. 140). Accordingly, a truly empirical sociology must treat members’ own solutions as targets of analysis, asking which problems ordinary members solve and how their solutions are socially organized. Here, Garfinkel is reiterating a point that Parsons had made as early as 1938: sociologists go astray when they confound social and natural facts. Social facts do not exist independently of social definition and human action, and therefore they cannot be properly understood without taking the specifics of how they are produced, and the social contract they orient, into account.

Introduction to Parsons’ Primer

53

For sociologists, according to Garfinkel the four problems are both topics of, and resources for, analysis (see pp. 140–141). They are topics, in that the sociologist’s focus is on members’ solutions to these problems in their everyday practical affairs. They are resources because sociologists must themselves grapple with these problems in their work. It is important, Garfinkel stresses, that sociologists acknowledge that their investigative procedures constitute a solution, not the solution, to the four problems, while at the same time not conflating these solutions with those of members. Treating their solution as the only solution would not only commit sociologists to a variant of naïve positivism, but would also condescend to members of society, implying that their solutions are at best approximations of the theorist’s. On the other hand, failure to distinguish the sociologists’ solutions from those of members, according to Garfinkel, renders “the procedural meaning of real social structures... unintelligible” (see p. 141), because the real structures, in Parsons’ theory, are theoretical, while the member’s solutions are practical and empirical. Garfinkel concludes Chapter III with a discussion of how Parsons retained enough critical distance from members’ solutions to avoid this problem while still avoiding positivism. He argues that in so doing Parsons is able to explain how practices are assembled, and how this assembly constitutes the real social structures that are sociology’s topic. Parsons’ position is similar to Garfinkel’s in the care with which he maintains this distinction. Parsons partitions sociology’s subject matter into three analytic domains: social organization, culture, and their related character. Garfinkel notes that “The problem of social order is found in the third set” (see p. 141); or, in other words, in how stable social structures are achieved in interaction (or the assembly practices of culture). For Parsons, the problem of social order can also be specified in terms of social organization, which is the ongoing process of assembling real social structures in a context of social contract. But, for Parsons these structures, Garfinkel notes, have the property of being uniform (see p. 141). They consist of observable features of the social world that have a typical, potentially repetitive character. While Parsons uses the concept of social structures in various ways, it is insofar as they are empirical observables related to a set of orderly expectations that Garfinkel is interested in them. Parsons’ social order is one that his actors treat as maxims of conduct. In so doing, their actions become events-in-thatorder, just as a move in chess has the status of an event-in-a-chess-game: the game furnishes the move with its specific sense, while each move in turn constitutes the game as a game of chess. Similarly, a move-in-a-social-order derives its sense from its position in that order, while it simultaneously constitutes and renews that order.

54

A. W. Rawls and J. Turowetz

Actions-in-an-order exhibit both that order and its normative regulation, and at the same time produce standardized social structures. It is this standardization (the rendering of actions in recognizable forms), Garfinkel says, that allows social scientists to produce the statistical descriptions of society (via means, medians, and modes), and the descriptive typologies, that feature in many of Parsons’ writings (see p. 142). However, the social sciences commonly make the mistake of reversing this relationship; treating standardization as a product of statistical operations, rather than a prerequisite for them (cf. Chapter VI). Making the mistake of treating objects as natural also leads social scientists to overlook the interactional assembly processes through which standardization is achieved by members of society. This is an oversight that prevents quantitative approaches from being scientific. Parsons, by contrast, treats all social objects as constituted within a social commitment and its assembly practices. In analyzing social structures, Garfinkel argues, sociologists following Parsons must confront the problem of “social reality”, which consists in distinguishing the “real” social structures that are the topic of their inquiries, from the “perceived social structures” that exist from “the actor’s point of view” (see pp. 143–144). Garfinkel points out that Parsons complained that this distinction is not being made by sociologists. The point of this distinction is not to suggest that the actor’s social structures are somehow less important than those of the sociologist, but rather to clarify the sociologist’s task, which is to explain how it is that actors are able to concertedly produce structures they perceive as real, rational, and commonsensical, and to distinguish that process from what the sociologist does in creating the “real” social structure theoretically. As such, the sociologist’s rules of interpretation, though different from the actor’s, are adequate only insofar as they correspond to, and provide for the reconstruction of, what Garfinkel calls, the “demonstrable features of the actor’s real perceived environment” (see p. 144). Parsons’ social system, accordingly, is a device for trying to specify the real features of actors’ environments analytically in terms of their assembly practices and the procedural rules for using them. What Garfinkel refers to as the “notorious pattern variables” represent one attempt by Parsons, that Garfinkel suggests, may have been “premature” (see p. 146), to specify these rules in terms of intelligible ways that actors can orient an environment. According to Garfinkel, the variables were meant to specify the ways that social objects could be constituted, as well as to specify the preferred practices for constituting them (see p. 146): to specify the rules of the game, so to speak. In delimiting the meanings an object could have, the pattern variables were also intended to define the boundaries of intelligibility: the line beyond which sense making efforts devolve into nonsense.

Introduction to Parsons’ Primer

55

Garfinkel (1963) would solve the problem of intelligibility by appealing to situated “rules” and their underlying Trust conditions. However, while Garfinkel maintained that people can only make sense together when they share constitutive expectations, he allowed that the assembly practices could vary endlessly, requiring only that everyone in a situation orient the same practices for varying them and that those practices themselves not damage the equality and reciprocity necessary for the process to continue working.

Chapter IV in Detail – Parsons: Implicit Social Contract Garfinkel opens this chapter with a description of Hobbes’ social contract theory, in which natural individuals arrive at a social contract through the exercise of reason. Garfinkel argues that one of Parsons’ principal contributions to sociology was the inversion of Hobbes’ position, instead grounding sociology and the possibility of human reason on the need for prior implicit social contract. According to Garfinkel, Parsons developed his solution to the problem of social order in dialogue with Hobbes’ Leviathan (1651), in which the conception of “correct structural analysis” yielded a society characterized by a war of all against all. According to Hobbes, without a social contract it would be every person for themselves. In Hobbes’ society, actors, conceived as rational individuals, are driven by “passions” and “desires” that are intrinsic to them as individuals. Those passions and desires are forever finding new objects to pursue. Hobbes’ rational individuals use reason to anticipate chains of cause and effect, selecting the best means for achieving their ends without regard for the wellbeing or security of others. Indeed, Hobbes argues that the only way his egotistical rational actors can preserve what they have (the capacity to achieve desired objects) is by continually acquiring more power. In Hobbes’ version of society, since each actor evaluates objects differently, they do not share a common meaning. Hobbes argues that each actor is “equal in ability”, each is insatiable, and no actor’s personal standards of right and wrong are binding on the others. Without any rule of law or social agreement, they will inevitably resort to force or fraud to get what they want. Consequently, an endless war will occur unless members of the society come up with a solution to the problem of social order. For Hobbes, the solution involves the development of a social contract by means of which each person cedes his freedom to a higher authority – in Hobbes’ case a monarch (“the Leviathan”) – who alone is given the right to define and enforce a single moral standard

56

A. W. Rawls and J. Turowetz

on society. In Hobbes’ view, this is the choice actors must make if they are behaving rationally, because doing otherwise will result in a permanent state of war (and conflicts of meaning). In the Structure of Social Action, Parsons argues that although social theorists have rejected Hobbes’ solution to the problem of order, along with other social contract variants proposed by Kant and Rousseau, they have not come up with a satisfactory replacement. The problems that Hobbes called attention to are important. Utilitarian individualism, according to which rational actors operate without a social contract, simply on the basis of self-interest, cannot explain how individual utilitarian actors can achieve shared meaning or social stability. J. S. Mill’s conception of a “general will” in the context of which self-interested actors take into account the good of the whole on which their existence depends was promising. But, without a prior social contract (or conception of constitutive practices) it is not clear how the idea that actors’ depend on the state of society would arise in the first place. Through a critical engagement with Hobbes and other social theorists, Garfinkel tells us, Parsons arrived at his own solution to the problem of order. At the center of Parsons’ solution, according to Garfinkel, is a “conception of action governed in its course by an order of moral expectations binding and enforceable upon all as constituent features of their common situations of action” (see p. 153). Garfinkel is saying that Parsons’ solution to Hobbes’ problem is to reverse the argument, putting social contract first as the tacit grounds for rational action. Hobbes assumed that rational actors would make a contract. Parsons recognized that the contract is constitutive of rationality and therefore must precede any agreement between rational actors. The results of the two approaches (social contract 1st versus social contract 2nd ) are quite different. Whereas in Hobbes’ society all individuals have different interests and passions and an arbitrary authority is required to create coherence, Parsons’ society is regulated by its own internal rules and values that members incorporate into themselves. As a consequence they create a shared world of objects and desires as they interact. This is what makes them rational beings for Parsons, and what makes it possible for them to create social order without being forced by a King to conform to the definition of things.

Seven Problems that Parsons’ Theory Solves Although problems remain in Parsons’ conception, including the idea that an authority is still required to sanction violations of the social contract and the values and norms of its assembly practices, it is an important breakthrough

Introduction to Parsons’ Primer

57

in the 1930’s in American social theory, which had previously been heavily influenced by individualism, naturalism, and utilitarianism. Parsons’ inversion of the social contract argument was inspired by Durkheim, who also put the social contract first, embedding in his sociology Rousseau’s idea that social and moral things and ideas don’t exist before social contract, and arguing that implicit conditions of contract ground society. It is also close to Garfinkel’s own position on Trust conditions. Garfinkel describes seven problems that, he says, Parsons’ theory solves. In expounding these solutions, Garfinkel also points out shortcomings in Parsons’ analysis. Garfinkel’s method is to evaluate Parsons’ solution against Parsons’ own criteria, arguing for the cogency of Parsons’ analytical framework and using it in this analysis while continuing to point out shortcomings, and developing it in a direction that, he argues, better satisfies its own requirements. These requirements, he says, comprise “the constituent tasks of structural analysis for the purposes of empirical sociological inquiry” (see p. 159). In other words, the theory must be falsifiable by the empirical data. (1) The first problem that Garfinkel says Parsons solves is to identify the general, constituent features of the concept of action. In contrast to the pragmatic approach to social theory popular in 1930s U. S. sociology that tended to treat individuals and objects as natural and not social, or if social as durable folkways, Parsons’ conception of action was influenced by the legacy of Rousseau, Hume, Kant and Durkheim. Therefore, he sought to logically formulate the necessary (“apodictic”) features that any concept of action would need to include. Parsons identified four basic structural elements – means, ends, norms, and conditions – in terms of which any action could be described. He developed what he called the “action frame of reference” to explain how these elements are to be related to each other in different instances of actions-in-their-courses. The action frame also had an inherent temporal dimension, of which Garfinkel would make much via his argument that an action’s meaning changes over time depending on the responses it receives. According to Garfinkel, Parsons also emphasized “that categories of action as contrasted with behavioral categories are inherently subjective in reference” (see p. 154). In other words, Parsons recognized that individual perception and social meaning could be different and would therefore need to be coordinated, and he argued that the required relationship between them needed to be sanctioned by social authority. (2) A second problem that Garfinkel says Parsons solves concerns the definition of rationality. Parsons defined rational action as the selection of adequate means for achieving one’s ends, with the understanding that both means and

58

A. W. Rawls and J. Turowetz

ends are defined by implicit social contract. Parsons defines “adequate” as what a scientific observer would judge as optimal. The big problem with this standard, Garfinkel notes, is that it invokes a criterion external to the actions to which it is applied. The scientific observer has different objectives (and operates in a different implicit social contract) from the acting member. So, their perception of what is reasonable and/or adequate will be different. This is especially problematic since Parsons’ inversion of Hobbes embeds reason in the social categories. This means that activities like magic, ritual, and so forth are supposed to be seen as rational from within the social contract that defines them, but will seem irrational when viewed in terms of the criteria of scientific observation. Merton’s famous distinction between the “manifest” and “latent” functions of an activity is one example of the false theoretical problems that result from defining reason as a social object, and then forgetting to distinguish between the definition of the situation of the scientific observer and the definition of the situation of the members. (3) The Third problem that Garfinkel says Parsons solves is that of the actor as agent (see p. 156). In everyday situations, questions of causation and agency are typically answered through the assignment of blame and responsibility: “In their everyday use, such categories or procedures are the morally equivalent categories of cause”. But, the theorist is concerned with matters of causation as it pertains to actors, not persons. By this Garfinkel means that social actors within a social contract, or definition of the situation, are not natural individuals in the natural world operating with natural reason. Social actors are not persons for Parsons in that sense. That is, they are not the individuals assumed by utilitarian theory. For Parsons, actors are courses-of-action oriented to an environment of objects within a specific definition of the situation, or social contract. As such, the same person can project multiple actors, or selves, even in a single situation. This complicates questions of causation. The self is a social object that changes in relation to other objects (including other actors) across time and sequence. Thus, Garfinkel explains, the meaning of causation will vary depending on the kind of actor, or course-of-oriented-actions – e. g. a role, role set, collectivity, subsystem, etc. – in question (see pp. 156–157). What Parsons calls the “voluntaristic” character of action provides for a specification of how actors themselves handle attributions of causation. But, the theorist needs to talk about causation differently. (4) Fourth according to Garfinkel Parsons has solved the problem of the logical status of subjective categories. This involves producing a description of action that does not judge the adequacy of action by invoking scientific criteria that are external to it. This problem is related to the first two. The adequacy of action is defined by criteria that are internal to the action. The constituents

Introduction to Parsons’ Primer

59

of action and grounds for its rationality are defined by the implicit social contract and definition of the situation (and its assembly practices). According to Garfinkel, Parsons solves the problem of subjective categories by means of his conception of culture as interaction, which consists of practices used in common by members of a society. Even subjective categories depend on, and are only knowable in terms of, such shared practices. To reconcile these subjective categories with the categories used by an observer, Parsons proposes the pattern variables as a solution, as these delimit the possible relevances afforded by an environment and the ways objects can be oriented in it. The idea is that if both observer and actor use the same criteria the problem is solved. Garfinkel will agree in principle, while arguing that it only works if the pattern variables specify the actual assembly practices that actors orient. Developing a critique that he alluded to at the end of Chapter III (see p. 146), Garfinkel invokes Schutz’s observation that the pattern variables attempt to codify what is known in common without explaining how it is known in common, or exploring what it looks like (see p. 157; cf. 1946:8). Parsons, Garfinkel says, assumes the variables that need to be discovered. This is one of the things Garfinkel would have had in mind when writing that the solution provided by the pattern variables “might have been premature” (for which we can read “was premature”) (see p. 146). (5) A fifth problem that Garfinkel says Parsons solves is that of the role of ideas in action. In the Structure of Social Action Parsons argued that a utilitarian orientation toward individual ends/interests was but one among many that actors could adopt. Actions could also be oriented to “ultimate values”, and Garfinkel explains that Parsons devoted attention there and in subsequent works (e. g. Parsons 1951, 1963) to theorizing the nature of “‘common systems of value’ and courses of common evaluative action” (see p. 158). In Parsons’ view, a stable society is only possible on the basis of normatively controlled actions and shared evaluative standards that define desirable social states that are common across a population of actors (committed to an implicit social contract). In Chapter VI, Garfinkel will identify such norms and values with assembly practices, and the evaluative standards and preferred orderings that regulate their use. Garfinkel is describing Parsons as taking a social contract position in which values/norms and the meaning, rationality, objects, etc. that they define, are true only by agreement. (6) The sixth problem that Garfinkel says Parsons solves, social causation versus naturalistic causation, harkens back to the third problem (conceptualizing the actor as agent), and the discussion in Chapter III (see p. 140), of the differences between the social and natural sciences. Any sociological claim about causation has to deal with the fact that its subjects – actors – are them-

60

A. W. Rawls and J. Turowetz

selves social facts who have their own theory of society, which they will treat as maxims of conduct. While the fact that actors are complying with some theory provides for the possibility of that theory’s description, it also presents the challenge to the sociologist of needing to produce a descriptive, rather than prescriptive, account of what the theory entails. In other words, the theory is only adequate to the extent that it describes the theory that actors are actually using (implicitly) as a maxim of conduct. Garfinkel says that although Parsons experimented with removing subjective categories from his theory, he eventually settled on a framework in which actors’ environments consist wholly of oriented objects: all objects are thus meant objects, such that subjective categories are fully incorporated and shown to derive their meaning in every case from empirically observable object-orientations (solving the problem of how to empirically observe “subjectivity”). Socialization therefore consists of learning to competently orient meant object environments via courses of action. This, in Garfinkel’s estimation, was an important breakthrough by Parsons (see pp. 158–159). (7) The seventh and final problem that Garfinkel says Parsons solves concerns the logical status of the concept of motivated action. Garfinkel doesn’t say much about this here. Instead, he refers the reader to his paper, A Comparison of Decisions made of Four ‘Pre-theoretical’ Problems by Talcott Parsons and Alfred Schutz. But, he will have a great deal to say about motivation in Chapter VI.

Rules for Pursuing Research Based on Parsons’ Solutions After this analysis of the problems that Parsons’ theory solves, Garfinkel follows up with a set of rules for how to pursue a research agenda based on these solutions. Garfinkel refers to these rules as defining the sociological attitude, or the rules and procedures for “seeing in new ways sociologically”. This way of seeing has nothing to do with the theoretical jargon one selects, he says; it comprises rules of procedure, not terminology (see p. 168). Garfinkel distinguishes two sets of rules: those that characterize the “scientific sociological attitude” and those that are “particular to the theoretical study of human actions and their objects” no matter how they are approached (see pp. 160–161). The first set includes (1) the rule of unqualified doubt that the world is as it appears; (2) the rule of observation, which stipulates that the only warrant for a proposition is empirical observation; and, (3) what might be termed the rule of clarification, which holds that problems must be formulated through the procedures of scientific inquiry, without which it cannot be determined

Introduction to Parsons’ Primer

61

whether a solution has been found. This rule set holds, in principle, for the social and natural sciences alike. It is the second set of rules that distinguishes the natural from the social sciences. This second set of rules, Garfinkel says, are (1) the postulate of relevance, (2) the postulate of subjective interpretation, (3) the postulate of adequacy, (4) the postulate of rationality, (5) the postulate of sociological empiricism, (6) the rules of the “action frame”, and (7) the praxeological rule (see p. 161). Those using these rules are doing seeing sociologically, according to Garfinkel. These are the rules that govern the normative assembly practices through which distinctively sociological objects are produced by social science, as well as the preferred criteria for using them. This second set of rules are: the 1) postulate of relevance, which has to do with assigning meaning to observed actions and events. To assign meaning, the sociologist draws from a set of interpretive rules that provide a sort of “grammar” of social action for a society of typical persons, engaged in typical courses of action, in typical environments. In other words, “They furnish him with what Max Weber referred to as the ‘motivationally relevant’ grounds of behavior” (see pp. 161–162). In addition to Weber, what Garfinkel says here also has parallels in Mills (1940) on “vocabularies of motive”, and Burke (1936) on “grammars of motive”. The 2) postulate of adequacy says that the sociologist’s “grammar” of social action must correspond to the motivationally relevant grounds of behavior that the persons under investigation actually use/experience. This does not mean that the person under investigation is the best authority on their actions, according to Garfinkel (see p. 162). Rather, it means that the theorist should attend to the analyses of the person’s circumstances as displayed by members, producing sociological explanations that capture the member’s orientations toward their own values/norms, rather than those of the researcher. Crucially, this allows for an empirical check on theory. Through this rule the theorist can determine when their analysis is wrong. If members do not orient situations as the theorist has predicted, and particularly if they do not treat violations of putative norms as sanctionable, then the theory is falsified, forcing a reconsideration of the operative “grammar of action” that the theorist has assumed. The 3) postulate of subjective interpretation requires that the theorist ask what type of actor or action would be necessary to reconstruct the assembly practices that would constitute the observed social structures under the rules that the theorist has postulated and whether such actors exist. This does not mean imagining oneself in “the role of the actor”, or trying to adopt their point of view; an approach advocated by some in the Weberian tradition of verstehen

62

A. W. Rawls and J. Turowetz

sociology (e. g. Beer 1963). Rather, it means that the theorist should specify the ways in which a given type of actor can recognizably exhibit something defined by the theory as subjectivity, and do so as an accountable feature of the society in question. Such specifications have nothing to do with what is “really” in other people’s minds. Unlike those operating in what Schutz calls the “attitude of everyday life”, the theorist is not concerned to ascribe subjective states to others (beliefs, intentions, dispositions, etc.), but only, Garfinkel says, to explain how such ascriptions can accountably occur; how they can be available and relevant for members, and what kind of knowledge and reason an actor would need to employ to do as the theorist describes (see p. 163). For the sociologist, “mind” is not an empirical phenomenon. But the fact that members treat it as existing and having relevance for their actions is a phenomenon that can be investigated empirically. The 4) postulate of rationality says that an actor is to be constructed so that they would, given sufficient information, act according to the cannons of scientific rationality. Contrary to appearances, this postulate is not actually about actors, but about the theorist who constructs them. According to Garfinkel: “the most ‘irrational’ of activities is to be treated rationally by the theorist” (see p. 164). All activities follow some grammar according to which they can be construed as rational, and it is the theorist’s job to specify these grammars. But, the postulate of rationality only applies to the sociologist’s actors when those actors are themselves operating in the scientific attitude: when their purposes and orientations are identical with those of the theorist. Otherwise, Garfinkel says, (scientific) rationality is a problem for theorists, not members (see p. 164). For members the problem is achieving recognizability and coherence according to shared assembly practices. For the theorist the problem is explaining this achievement in terms of scientific (means-ends) rationality. The 5) postulate of sociological empiricism holds that the sociologist should “permit actual persons to teach him” how they produce stable social structures. It follows from this rule that Parsons’ theorist should refrain from judging the truth, importance, etc., of actors’ claims. Instead, they should restrict themselves to accounting for how these claims are assembled. What Garfinkel says here is similar to Durkheim’s defense of Totemic religion. The critics called such religions false and those who believe in them deluded. By contrast, Durkheim (1912, Rawls 1996, 2004) looked for what was socially true and valuable in these religions. The 6) rules of the action frame, according to Garfinkel, advise the theorist to treat actors’ theories about their actions as data, and not to confound the actual person’s perspective with the theorist’s rules for organizing that perspective (see pp. 166–167). Parsons’ sociologist should be a “stranger” to those whose

Introduction to Parsons’ Primer

63

actions they describe or, as Garfinkel (1967:9) would later put it, they should “treat the rational properties of practical actions as ‘anthropologically strange.’” Insisting that sociologists take the perspective of their subjects (as Weber did) would make no more sense than telling physicists to take the perspective of an atom. What one wants to describe are the “rules” that the atom could be said to be following: Not the “perspective” of the atom. Finally, 7) the praxeological rule, Garfinkel says, requires sociologists to translate any statement about the causes or determinants of action into a “point for point” description of how persons must act to produce whatever feature of social structures is being considered (see p. 167). In other words, instead of saying that A causes B, the sociologist needs to describe an “actionin-its-course” as a course of action orienting a set of rules or assembly practices that would have produced the sequence of actions in question. This is what the pattern variables are designed to do. Obviously, if there are no actions in the real world, and no actors that correspond to what the theorist has proposed, the theory is falsified. This explains what Garfinkel means by saying that Parsons’ theory is subject to empirical refutation.

The Primer in Detail – Chapter V: An Illustrative Reading of Parsons’ Pattern Variables The fifth chapter of the Primer is devoted to explaining Parsons’ pattern variable argument. It was the concluding chapter of the mimeographed set that Garfinkel began circulating in 1960. In the 1962 version it occupies a different, but also pivotal, position as the middle chapter of nine. Thus, Chapter V plays an important role in both versions of the manuscript. The second paragraph explains that it was intended as a summary illustration of the characteristics of Parsons’ work as presented in the first four chapters. The fact that it is devoted exclusively to the pattern variable argument indicates that the first four chapters also relate to that argument, even as they cover broader ground, and that in Garfinkel’s view, Parsons’ overall position culminated in his new 1960 revision of the pattern variable argument. In the longer 1962–3 version of the Primer, in addition to being a culmination of the first four chapters, the chapter stands as a precursor to Chapter VI, which was written in response to Parsons’ suggestion that Garfinkel write an additional chapter to address the relationship between what Garfinkel called his “fourth theorem”, and Parsons’ conception of norms and values and the “reasonable man”. That subsequent chapter positions Garfinkel’s fourth theorem (which describes Ethnomethodological studies of accounting and other

64

A. W. Rawls and J. Turowetz

practices), in relation to Parsons’ theory as it stood in 1962–3, with particular emphasis on the pattern variable argument, and supported by Parsons’ developing conception of interaction/culture as an orderly system of assembly practices. When Garfinkel completed the version of Chapter V reproduced here, Parsons’ pattern variable argument had been evolving for twenty-five years (1937–1963), moving from an earlier version that did not fully incorporate interaction and social contract, to a later version which had resolved the problems with individualism, motivation, choice and naturalism, in favor of a more fully interactional model of a system of social interaction/culture based on an implicit underlying social contract (as Garfinkel explains in Chapter IV). Garfinkel explains this progression in Chapter V, elucidating the problems solved by the revision of the pattern variable argument, and arguing that in his estimation Parsons was coming close to specifying an adequate theoretical approach to social interaction as a system: which was also Garfinkel’s objective. As Parsons overcame earlier problems and moved closer to Garfinkel’s position, his new work became more important to Garfinkel. In spite of a widespread belief that Garfinkel would have rejected Parsons’ pattern variable argument, Chapter V makes it clear that this was not the case. In fact, Garfinkel is recommending aspects of the argument and connecting them to ethnomethodology. What Garfinkel liked about Parsons’ approach included the fact that it specified interaction as a system of rules/expectations bounded by an implicit social contract. This was Garfinkel’s own position.

The New Introduction to Chapter V Garfinkel wrote a new two-page introduction to Chapter V in 1963, apparently preparing it for publication as a separate article (which he never published). The new introduction, which we reproduce here, offers additional information about Garfinkel’s involvement in the writing of the revised pattern variable article in 1959–60, while also explaining Parsons’ reasons for publishing a revision in response to Dubin’s article. According to Garfinkel, Parsons understood that Dubin had intended to produce a friendly explanation of “how to” use the pattern variables. But, the argument had changed so much that Dubin’s account was wrong. Given the extensive changes in Parsons’ argument over time, Parsons felt that the only way to correct the misunderstandings in Dubin’s article was to publish a revision of the argument as a whole. As Garfinkel points out, the revision not only replaces what Parsons calls the first pattern variable paradigm, but also what Dubin calls the sec-

Introduction to Parsons’ Primer

65

ond pattern variable paradigm. Parts of the first two paradigms are contained within the revision. But, the revision transcends and transforms them. This is the problem with Dubin’s friendly version. It treats the second paradigm, which had also been superseded, as if it was Parsons’ mature position. While Garfinkel was at Harvard in the fall of 1959 he would not only have discussed the Parsons’ Primer manuscript with Parsons (as it would have appeared in a pre-1960 draft): they also discussed the revision to the pattern variable argument (audio in the Archive). This discussion, along with Garfinkel’s responses to Parsons’ letters, would subsequently shape the substance of Chapter V. There were several meetings to discuss the revision (with Winston White attending) for which there are audio recordings, and Garfinkel is credited by Parsons in the first footnote of the published revision with helping edit the article. Chapter V is thus about matters that the two scholars had quite recently been engaged in working on together closely. However, that Garfinkel had worked with Parsons on the revision in 1959 only partially explains why Chapter V is addressed rather exclusively to Parsons’ revision of the pattern variable argument. There is also the question of why Garfinkel devoted so much time to working on the argument with Parsons in the first place. He obviously felt the argument was important enough to merit his time and attention. The importance he accorded the argument in this “summary” chapter, and its presence as an underlying theme in the Primer as a whole, suggest that Garfinkel considered it of particular importance. Garfinkel argues that Parsons solved several important problems with his revision that brought him closer to what Garfinkel considered to be an adequate position on social interaction. Some of this Garfinkel has already explained in Chapters I–IV. But, in Chapter V he addresses more directly the way the revision embraces a more systematic approach to interaction, which builds practices, rules and underlying commitments into the interaction system, and takes them out of individuals. This is important to Garfinkel, who always focused on social interaction as a system (and not on individuals). It was also important to Parsons, who rejected individualism and had never meant to embed it in his position (even though some individualism remained in earlier versions). Garfinkel also reiterates that the revision eliminates the problem of presuming a “cultural dope”.

66

A. W. Rawls and J. Turowetz

The Chapter – Explaining the Revised Pattern Variable Argument Because the revision of the pattern variable argument puts the emphasis on the rules and assumptions that the social theorist uses to define the variables that in turn define the meaning of what social actors are doing, Garfinkel devotes Chapter V to a description of all the rules and assumptions that are involved in Parsons’ reconceptualization of the pattern variables. It is pretty heavy going. There are lots of rules and assumptions. To make things more difficult, Garfinkel assumes a lot of familiarity with Parsons’ argument, which the average reader will not have. He has done this to some extent throughout the Primer. But his assumption of familiarity is particularly problematic in Chapter V and it is easy to get lost. While it makes the text difficult, we treat this as another clue to the importance Garfinkel accords to the revised pattern variable argument. Another problem, both with Parsons’ revision and Garfinkel’s discussion of it, is that the language is difficult. Because Parsons has now embedded motivation and rationality in the norms and values of action, the sentences that must be used to discuss the revised pattern variables often don’t use nouns and verbs in the usual way. This means that important explanations are invariably long and difficult. Parsons’ text is closer to how Garfinkel writes about ethnomethodology than it is to Parsons’ earlier writing. And the difficulty is happening for the same reason: Because Parsons has made the actions, actors and objects in his interaction system all conditional on rules, presuppositions, and relationships between them; insofar as, and only insofar as, these can be recognized by other members of an interactional system as normal, expected and competent, ordinary grammar cannot capture it. Getting all of that into a sentence requires turning nouns and verbs into conditionals. It is awkward, but necessary, because ordinary grammar treats action and objects as just existing and ignores the cooperation and orderly assembly work that goes into making them. “Garfinkel’s oriented actor” makes an appearance. In order to include motivation and rationality in the parameters of an action, long clauses develop that make sentences difficult. For instance, in Parsons’ words (1960:471): “The very discrimination of different bases of orientation of actors to objects implies that actors are conceived as systems; they are never oriented to their situations simply ‘as a whole’, but always through specific modes of organization of independent components”. That is a confusing sentence. Actors are systems, objects are systems – but only actualized insofar as they are oriented

Introduction to Parsons’ Primer

67

– and the meaning of every particular thing is a result of the mutual orientations of actors and objects in particular situations of action in which they have made a mutual commitment to such an interaction system and acted on it. Objects, like actors, also perform. As Parsons (1960:472) says: “At their ‘interface’, then, an especially important property of objects is their probable performance in respect to the actors oriented to them. Recall that the prototype of the actor-object relation is social interaction, in which the ‘object’ is also in turn an actor who does something”. The conception of an oriented actor, which Garfinkel introduced in 1948, is essential to the revision. In Parsons’ (1960:473) words the assumption: “that all action involves the relating of acting units to objects in their situation... is the basis for the fundamental distinction between components belonging to the characterization of orienting actors and those belonging to the modalities of the objects to which they are oriented, that is, between the two ‘sets’ of elementary pattern variables”.

The incorporation of orientation (motive/goal) into actors, objects and actions, tends both to make Parsons sound more like Garfinkel and to make the argument more difficult to understand. But, this is the change that overcomes individualism and finally treats interaction as a system grounded in an underlying constitutive contractual commitment. This is a significant change, and Parsons would not have made it if he had not understood – as Garfinkel understood – that a social theory that centers on interaction is not going to succeed unless it can get out of individual heads and focus on the interaction that occurs between actors as an legitimate and sanctioned system of interaction. Because Garfinkel uses few examples as he is elaborating the rules of the revised pattern variables, his explanation (as in the original pattern variable argument) is also difficult to comprehend even though it is actually quite clear and straightforward. We might say that it is too clear and straightforward. It assumes too much familiarity. Because of this we have taken particular care in trying to explain what is going on in this chapter. Ultimately, the revision does not solve all of Parsons’ problems and Garfinkel knows this. But, Garfinkel is encouraged – even excited about it – because of the degree to which Parsons is trying to put all the rules for interaction into the rules for action itself and get them out of the actor. In his January 14, 1963 letter to Parsons, Garfinkel had described himself as “optimistic even confident” that he would be able to work out a fourth theorem on the basis of his understanding of Parsons’ revision and that his 4th theorem would be adequate to support the needed empirical demonstrations of the argument.

68

A. W. Rawls and J. Turowetz

Summary Analysis of Chapter V In this chapter Garfinkel says that Parsons’ revised paradigms of action systems in the revision “consist of rules for looking at everyday activities... as assembled social structures of interaction”. What Parsons means by “an adequate description of social structure” involves the use of these rules. Garfinkel announces the chapter as a description of Parsons’ method in detail. The presentation consists of seven parts 1) the unit act, courses of action, courses of treatment; 2) pattern variables, possible courses of action and possible objects; 3) rules for assembling stable courses of action; 4) rules for relating those assemblies; 5) temporal processes; 6) presuppositions for the rules in 3, 4 and 5; and, 7) the multi-celled pattern variable table. Garfinkel says there is a theorem that corresponds to each of 3, 4, and 5, and to each theorem there corresponds a set of empirical propositions. We believe these are the three theorems referred to in his January 22, 1963 letter to Parsons. The fourth theorem in that letter is Garfinkel’s and represents ethnomethodology. Part I in Detail: “An Exposition of Parsons’ Method”, explains the relationship between the several versions of the pattern variable argument with a focus on (1) the “Unit Act”. Before explaining anything else, Garfinkel deals with misconceptions of the unit act: It is not a unit in any conventional sense and it is not an act. His point is that the unit act is not the problematic conception it is usually taken to be. It may have problems, but they are completely different from what the critics think they are. According to Garfinkel “Unit does not mean single or singular” (see p. 172). The unit act is a course of action and a course of treatment – over a temporal course – within a system of interaction, the orientation toward which defines the boundaries of the action and its meaning. Unit acts are interactional orientations toward objects. They are sometimes actors, sometimes objects, and sometimes actions. This is significant because the conventional critique of the unit act treats it as if Parsons had tried to define the unit act conceptually and then points out the impossibility of doing so. Among other problems, there would be a problem defining where the unit act begins and ends. But, Parsons’ unit act as Garfinkel tells it is something very different. It is a course of action, defined by a set of expectations/rules that are situated within a social contract, which is defined by how it is treated within a system of mutually oriented values and norms characterized by the process of double contingency on a particular occasion of its use. If the actors recognize

Introduction to Parsons’ Primer

69

the unit act as a particular course of action and treat it as such within such an interactional system then it is what they treat it as. Garfinkel explains that this “usage of the pattern variables” in the 1960 revision “departs from usage in Parsons’ previous writings where they referred to the intended properties of actions and objects” (see p. 173). Intention has been replaced. In the revision all treatments of objects are motivated treatments, which are called orientations. The intention is now built into the rules defining the action. According to Garfinkel, “The term object or object modality includes in its meanings the reference to motivation” (see p. 173). The “choices” are also imbedded in the action types in the new schema. The individualism has been eliminated. As Garfinkel explains it: “the course of treatment is the treater” (see p. 174). Here Garfinkel is attributing a very sophisticated argument to Parsons, much more sophisticated than the pattern variables are usually taken to be. Garfinkel also says that “‘Unit’ is a grammatical and not an empirical way of talking” (see p. 172). The unit act is part of a larger whole – a complex of rules, expectations and value orientations that defines it – and it is just one manifestation of a system of interaction, none of the parts of which are separate: like a move in chess. The overall pattern variable schema is intended to define the domain of possible courses of treatment of actions in their course. Each parameter, Garfinkel says, is accompanied by its set of rules (see p. 175). The degree to which these are institutionalized – i. e. can be counted on as a matter of course – is important. According to Garfinkel, “The more are these standards institutionalized, the more does there exist the correspondence from Ego’s point of view between courses of treatments and the behavior of objects under these courses of treatments”: i. e. stability depends on this (see p. 175). Any set of activities that exhibits “stable properties” in this way are solutions to what Parsons calls the four functional problems: goal attainment, adaptation, integration, and pattern maintenance. Garfinkel also states a praxeological rule that any observed actual activities are solutions to the four problems (see pp. 175–176). This is because they would not be observable as meaningful activities unless they had solved the four functional problems. Garfinkel says that “Several conditions must be satisfied if the stable social structures are to be theoretically possible, and thereby to be empirically likely (see p. 176). The remainder of the chapter is concerned with tasks of enumerating and explicating these conditions”. Having addressed the overall issues in the context of a discussion of the unit act in the first section, Garfinkel devotes the remainder of the chapter to setting out the rules and assumptions that are involved in the interaction system as the revised pattern variable argument specifies it. He has already

70

A. W. Rawls and J. Turowetz

stated the seven assumptions, and four functional problems, followed by a statement of the praxeological rule (see p. 176). According to Garfinkel the difference between versions of the pattern variable argument is that: “Parsons’ early argument was concerned with the problem of what these evaluative standards were. They were decided by using the pattern variables to furnish directly the properties of such standards. In the revision ‘Pattern Variables Revisited’, the pattern variables are used instead to generate those sets of standards that are capable of being institutionalized as well as the set of those that are not” (see p. 176).

The result, Garfinkel says, is that “Objects consist only and entirely of motivated features of ‘meanings’” (see p. 177). Everything, he says, is handled by the system without the need for recourse to individuals: “The pattern variables define for the theorist the alternatively possible features that the environment of objects can possibly consist of...” (see p. 177). 2) Possible Objects: Garfinkel introduces a table that he says shows the “grammatical structure of objects” (see p. 177). The discussion draws on Wittgenstein’s conception of a “grammar of use” and Burke’s grammar of motives. Any naturalism in Parsons’ earlier schema has been eliminated. An object in the revised schema is defined not by its intrinsic features – but by how the actor orients/treats its features in a particular course of action in the context of the situated rules of relevance that are in play. As Garfinkel says: “The labels for these assemblies depend for their meanings on the combinational rules and not the other way around” (see p. 177). Objects have “performance features” in relation to the actor rather than intrinsic attributes. 3) Possible Orientations: These categorize the possible treatments of objects – treatments being motivated orientations. By building the motivation into the treatment, the conventional teleological approach and the individualism and naturalism it rests on, are overcome. According to Garfinkel “an object is not called a goal object because of features that are inherent to it (see p. 180). It is spoken of as a goal object by reason of the manner in which action and object are combined”. It is in the recognizable features of the assembly practices, in Parsons’ revision, as they are controlled for by standardized and sanctionable “oriented actions”, that the motivation and goals, as well as the rationality of actions and objects lies. 4) Rules for Relating these as a Set of Stable Concerted Actions: Because all actors and objects have many possible orientations, and manifest as many different “things”, there need to be rules for relating these. In this section Garfinkel explains those rules. According to Garfinkel they involve the use of

Introduction to Parsons’ Primer

71

the “concept of the integrative standard” (see p. 182). There are objects that cannot be simultaneous. For instance, there are things we can expect from a mailman and things we can expect from a friend. The mailman cannot orient both sets of expectations at once. They cannot withhold mail because we didn’t give them a friendly greeting yesterday. So, we need to decide which actor we are dealing with, a friend or a mailman (elsewhere Garfinkel contrasts policeman and husband to make the same point. The expectations one has of a husband conflict with the demands one can make on a husband who is a policemen who has stopped you for speeding. Even though the actor can legitimately claim both roles, they cannot stand in both roles at once, and a husband who tries to stay in the policeman role to give his wife a ticket may find his wife adopting the role of a stranger when he returns home). Garfinkel introduces the notion of “proper balance” between such competing identities and talks about the role of Parsons’ “integrative standards” in achieving such a proper balance (see p. 183). The “higher integrative standards” are also where Parsons located Garfinkel’s work in his letter of January 30, 1963. Proper balance is not a subjective idea. It is institutionalized in the standards to which all actors/members are committed as the defining features of the interactional system that they are interacting within. These are members’ methods. These standards, according to Garfinkel, define the stable combinations such that “alternate combinations are unstable” (see p. 184). 5) Temporal Processes: Stable social structure is created by, and maintained over, temporal courses of action. To add to what Garfinkel has said in this chapter about unit acts, possible objects, and rules for integrating objects, he now introduces the problem of temporality. He asks “How are these stable reciprocities guaranteed over the course of their occurrence as temporal processes” (see p. 185, emphasis added)? The problem of temporality also invokes the problem of the “environmental exigencies”, the recognition of which Garfinkel says, is one of Parsons’ most important contributions to social theory (although they do not fully agree on the solution) (see p. 185). For instance, in a Chess game it matters when a move is taken. Is it an opening move? Or a move later in the play? The meaning of the move will depend on this, which means that the temporal dimension of sequentiality must be built into the standards being used to confer meaning on action in the game. According to Garfinkel: “for the theorist to recognize a move as a move-in-the-game he would have to refer to both sets of rules, basic and preferred, as conditions that a player’s actions would have to satisfy in order for these actions to be recognized by the

72

A. W. Rawls and J. Turowetz

theorist as a move-in-the-game, as well as to be responded to, and recognized by the player, as a move-in-the-game” (see p. 186).

Garfinkel is invoking his own Trust argument and his conception of basic and preferred rules in talking about Parsons and temporal processes. The point here is that Parsons understands this. 6) Presuppositions and Exigencies: To say that the contingencies/exigencies are “‘outside’ the ‘real game’ is not satisfactory” according to Garfinkel (see p. 187). The kinds of contingencies he describes go on all the time and must be taken into account. They are a part of the play. Here he is explaining that we need to know what actual players do. We cannot just describe the ideal game and think we have described what real people do. By adding in the exigencies Parsons is no longer treating the game as an ideal that can only be played by a “cultural dope”. According to Garfinkel Parsons’ critics have made the mistake of thinking he has this problem: “If a theorist who used Parsons’ system was required to use it within such assumptions – several of Parsons’ critics are incorrectly convinced that this is the case – then we should call Parsons’ actor a cultural dope, or better a version of a cultural dope” (see p. 189). What this misreading of Parsons leaves out, Garfinkel argues, is “the work of actors” in handling the exigencies over a temporal course of action: “the conditions that motivate this work Parsons refers to as the environmental exigencies of the system of action. The work that actors direct to the constitution, definition, repair, test, review, maintenance, elaboration, codification, criticism, historicizing, etc. of the relevances, Parsons refers to as the “adaptation” of the system of action to its environing exigencies” (see p. 190).

This real time attention to handling the exigencies makes all the difference to Garfinkel, because Parsons is now including in his schema the actual work done by real actors in concrete social action to produce these stable features of the system of interaction across temporal courses of action: i. e. across temporal sequences of action. And this is so, for Garfinkel, even though Parsons still formulates that work conceptually and wants to control the exigencies in ways that Garfinkel objects to. What matters here is that what actual actors do can falsify his theory. Previously there had been three sources of instability in Parsons’ system. They had been, according to Garfinkel: the compliance of actors, relevant object features, and relevant concerted matchings (between elements of action) (see p. 191). “The exigencies” Garfinkel says, “furnish a fourth source of instability” (see p. 191). Moreover, it is a source of instability that applies

Introduction to Parsons’ Primer

73

separately to each of the prior three because it has to do with the actor’s commitment to each of them. It is this instability that requires cooperation and commitment from actors. This is the concern of Garfinkel’s fourth theorem. According to Garfinkel, “The actor’s knowledge of these exigencies may or may not be identical to the theorist’s knowledge” of them (see p. 192). Obviously, it is not often the same. And it is the actor’s knowledge of the exigencies that the theorist needs to know about. Even when the theorist gets this right, however, the theorist’s interest in this knowledge will never be the same as the actor’s interest because, the actor doesn’t care about theory. The actor just uses their knowledge, whereas the theorist wants to analyze and understand it. Garfinkel will talk a lot about this later in his studies of work, pointing out that no matter how competent a researcher becomes in the work they are studying, their interest in that work as an analyst will never be the same as the interest in the work of the non-sociologist. That is why he named those studies “Hybrid Studies” of work. The important thing in terms of Parsons’ theory is that the system needs to adapt to the exigencies across temporal sequences of action – and that Parsons understood this and included it in his model. 7) The Adaptation of the System of Action to the Exigencies According to Garfinkel: “Parsons’ actors require a knowledge of the real world, and enforce this knowledge upon each other, as a condition of their moral rights to manage their concerted affairs without each other’s interference” (see p. 195). The real world involves endless exigencies. This section on adaptations to the exigencies is important because this is the big point of disagreement between Parsons and Garfinkel. Variations in the exigencies, Parsons treats as a source of discrepancy that he seeks to limit. Garfinkel does not. Garfinkel says that Parsons postulates four sets of exigencies that “In use they have the feel of concepts in search of their contents” (see p. 195). Then there are four rules for dealing with the exigencies. According to Garfinkel: “Parsons refers to the first two as the actor’s symbolic “knowledge” of the real world (see p. 197). He refers to the second two as the actor’s “categorical ‘knowledge’ of the real world”. Garfinkel turns to games to explain what Parsons means, saying that: “Game-furnished conditions are to games what exigencies are to Parsons’ system of action. The analogy is an exact one; the properties of game-furnished conditions and exigencies are identical” (see p. 199). In other words, players play games in a real world of contingency and variation that become part of the play and therefore must be taken into account by the system of interaction.

74

A. W. Rawls and J. Turowetz

Conclusion to Discussion of Chapter V It is this fifth chapter that Parsons was mainly referring to in his letter to Garfinkel of January 30, 1963, and it is after he receives that letter that Garfinkel adds the new introduction and revises the chapter as a stand-alone article. This is important information about the communication between Parsons and Garfinkel about this chapter. They not only worked together on the revision at Harvard in 1959, we also know from his letter that Parsons had read and recognized his own position in Garfinkel’s characterization of the revised pattern variable argument. In addition, we know that Parsons identified what Garfinkel called the first three theorems with specific parts of that argument, while identifying Garfinkel’s fourth theorem with his own “higher integrative functions”. It is the values and norms that actors/actions orient that (in addition to the rules and assumptions defined by the pattern variables) determine what a particular oriented action means. This is so for Parsons as it is for Garfinkel. The difference remains that for Garfinkel this can happen in two different ways. Actions that are accountable to social institutions need to orient the shared “values” of the institution. But, in ordinary everyday interactions the values that govern the use of rules are the interactional ground rules/expectations presupposed by the system of interaction; they are like the basic and preferred rules of games, and not an institutionalized value standing outside of the system of interaction. The two are institutionalized in very different ways, institutionalized sets of assembly practices versus formal institutionalized rules that are only accountable and not “followed”. Parsons’ commentary in the letter is important because it is one of the few places where we have Parsons directly commenting on either what Garfinkel makes of his efforts, or on Garfinkel’s own work and his attempt to relate it to Parsons’ theory. Chapters V and VI already stand out from the rest for the importance of their summary and synthesis of Parsons’ ideas and their relationship to ethnomethodology. Parsons’ commentary makes Chapters V and VI even more important, because those comments not only give us Parsons’ response to Chapter V, but Garfinkel writes Chapter VI after receiving this response. In his January 30, 1963 letter, in the fourth paragraph, Parsons comments on the three theorems that Garfinkel had outlined in his letter of January 14, 1963. The comments are cryptic and assume a great deal of knowledge of Parsons’ system on Garfinkel’s part. Parsons (January 30, 1963) identifies the three theorems specifically, saying: “they are clearly identifiable as pertaining

Introduction to Parsons’ Primer

75

specifically to three of the basic functional categories: namely, 1. 2L; 2. On motivated compliance to G; and 3. On satisfying functional problems as conditioned to A”. In addressing the fourth theorem proposed by Garfinkel, Parsons (January 30, 1963) goes on to say: “This suggests, by elimination, of course, that the new theorem should be looked for in the area of integrative problems and functions”. That is, Parsons not only identified Garfinkel’s first three theorems with specific aspects of his own work without any difficulty, but went on to suggest that Garfinkel’s fourth theorem (which deals with accounts, accounting, situated reason, etc., in a recognizably Ethnomethodological fashion) can be located in a part of Parsons’ argument that he refers to as “integrative problems and functions”. Whereas many scholars have characterized Garfinkel’s preoccupations as micro and individualistic, Parsons identifies Garfinkel’s studies with his own integrative functions, which in the 1960 article are described as “higher” functions. According to Garfinkel, Parsons dealt with the reductionism of his earlier model of means-ends rational choice by including motivations within the defining parameters of actions and their orientations in the revised pattern variable argument (see p. 173). Meanings are now motivated features of objects as they are oriented within a legitimate and sanctioned interactional system to which participants are obligated by commitment. A course of action within such a system is a process that relates objects (which are themselves courses of action) in ways that are recognizable to all actors who are committed to the system. In its outlines this description of Parsons is very close to what Garfinkel was doing with ethnomethodology: identifying “courses of actual action” and seeing how they relate to the “objects they treat” within a set of underlying expectations and commitments that define a domain of interaction. The difference is that Garfinkel was committed to finding the actual rules that concrete persons were actually using on concrete occasions of interaction to make the meaning of their actions evident to other persons within the system of interaction within which they were interacting. This turned out to be a huge breakthrough for Garfinkel because he was able to demonstrate preferred rules that explained how contingencies are used as resources to achieve certainty. Parsons rejected this approach because he did not believe that it was possible to make order from contingency. Garfinkel (1988) would later say that Parsons never realized that actual interaction had the ordered characteristics he said it needed to have, so Parsons fell back unnecessarily on an analytic set of criteria to describe what was going on.

76

A. W. Rawls and J. Turowetz

Parsons accepted that real courses of action could falsify his schema. But he did not accept that the schema could be built in the first place on observations of an actual order that was evident in concrete interaction. He thought actual interaction was too messy and full of contingencies for this to be possible and therefore Parsons remained committed to the necessity of proposing analytic criteria as a universal to bring the contingencies/exigencies under control.

The Primer in Detail – Chapter VI – On Norms and Values Written with Parsons’ encouragement after he had read and then commented on the first five chapters, Garfinkel pulls everything together in this sixth chapter. It is here that he most explicitly connects his own studies in ethnomethodology with Parsons’ position, arguing that the principle weakness in Parsons’ position was his lack of an empirical demonstration with which to answer his critics, and offering his own Ethnomethodological studies as the needed demonstrations. In so doing, Garfinkel makes it clear why he took such an interest in explaining the development of Parsons’ theory. Garfinkel is essentially proposing ethnomethodology as the missing piece of Parsons’ argument. At the same time it could be said that the explanation of Parsons in Chapter VI is a missing theoretical piece of Garfinkel’s ethnomethodology. It is a mystery why the argument of this chapter did not figure in the published volume of Studies in Ethnomethodology in 1967. Written only four years before the publication of that volume, after most of the research for that book was complete, this chapter demonstrates a strong connection between Garfinkel’s empirical studies and Parsons’ theory. There are many references to Garfinkel’s “Trust Conditions”, and the discussion of values and norms overlaps with a paper titled “Notes on Language Games”, in which Garfinkel works out the Trust argument in terms of Wittgenstein’s constitutive rules.27 In that paper, Garfinkel treats Wittgenstein’s treatment of rules in language games as parallel to his own distinction between constitutive and preferred rules, which in Chapter VI he treats as a version of Parsons’ distinction between values and norms. This attribution of a sophisticated, phenomenological, and Wittgensteinian position on rules/expectations to Parsons is quite different from the conventional reading of what Parsons meant by values and norms. 27 Another critical missing piece of Garfinkel’s theoretical argument, the previously unpublished paper “Notes on Language Games” was published in The European Journal of Social Theory in May 2019.

Introduction to Parsons’ Primer

77

There are many surprises in the chapter. Parsons’ theory is, Garfinkel says, a “calculus of significance”, i. e. it can be used by the theorist to prioritize research questions. By this he means that because it is a theory grounded on the idea of commitment to a set of constitutive rules/values (and their assembly practices), which in turn depend on the legitimate expectation that violations will be sanctioned – the theory defines a number of research topics in a useful and relevant way. This positions Parsons’ theory as not only focused on the legitimate assembly practices of interaction – as culture – but on sanctions themselves as the empirical evidence that these assembly practices are part of a legitimate, sanctioned, and institutionalized social agreement between members. To get at sanctions empirically, Garfinkel proposes to focus on “trouble”. Again, this is a familiar aspect of Garfinkel’s work. But, here we find it proposed as a way of getting evidence to support Parsons. After introducing this idea of a “calculus of significance”, with the vague suggestion that the chapter will deal with empirical research, Garfinkel sets out to explain how Parsons’ later argument, based on the conception of values, norms and commitments, differs from his earlier iterations of his position in ways that set a foundation for the empirical demonstrations of his position that Garfinkel will propose. Unless the difference between Parsons early and late theory is understood, none of what Garfinkel proposes here makes any sense. Garfinkel explains this difference with a particular focus on changes in Parsons’ treatment of culture and interaction that he maintains constitute the essence of Parsons’ new position. In all, Garfinkel describes a progression through five distinct theoretical positions, with Parsons finally adopting what Garfinkel calls a “Paradigm of Social Interaction”. Given the many misconceptions and criticisms of Parsons’ work, this discussion is eye opening. Garfinkel maintains that Parsons’ inability to satisfy his critics was not due to inadequacies in his mature theory, but rather to a lack of empirical demonstrations of his position, which he says Parsons is well aware of. After surveying the relationship between Parsons and his critics, Garfinkel devotes the last five sections of this chapter to a discussion of what research is needed to supply the empirical demonstrations of Parsons’ theory, and offers a “method” for doing so: which turns out to be his own studies in ethnomethodology and his Trust argument. If Garfinkel is right about the role of values/norms, commitment, and interaction in Parsons’ mature position, it is clear that the critics never came close to appreciating either the sophistication of Parsons’ argument, or his vision for developing a modern sociological theory centered on culture as interaction. We find a refreshingly different Parsons presented here. Garfinkel does not elaborate on earlier versions of Parsons’ argument or why they changed.

78

A. W. Rawls and J. Turowetz

As he told Parsons in his 1959 letter, he does not want to get caught up in the chronology. Instead, Garfinkel lays out the progression succinctly in a twopage section titled “Decisions that Preceded the Action-Situation of Objects Schema to Conceive Observables”. The sparseness and clarity of the explanation results in a depiction of Parsons’ position that is strikingly different from the conventional view. Yet, we know from their collaborative work on the revised pattern variables, and from their correspondence, that Parsons does not disagree with Garfinkel’s depiction – as startling as it is.

The Stages of Parsons’ Position – Chapter VI: Sections II to IV As Garfinkel presents the development of Parsons’ position it went through five stages. The first, in The Structure of Social Action (1937), depicted rational persons who were directed toward goals in a means-ends schema that retained elements of both individualism and naturalism. Around 1940, Parsons dropped the means-ends schema and adopted a second two part schema consisting of an “actor-situation of objects framework” and a “cultural situation of objects” that resolved some of the contradictions in the first schema. Then Parsons dropped that two part schema and “in its place” adopted a third “three-part schema of actor, action, and situation of objects”. But, even though, according to Garfinkel, this resolved some of the contradictions in the first two iterations, it retained the problematic feature of a “cultural overlay of meaning”. In The Social System (1951), Parsons presented a fourth iteration in which, Garfinkel says, “The actor was dropped out leaving only courses of action”. It became a two-part schema again: the actor as a set of possible actions in their course, and an environment of possible treatments of those objects. The problematic cultural overlay idea was replaced by a conception of the situation as a cultural rather than a physical environment. Many problems were overcome, but interaction was still not sufficiently centered in the scheme. In the fifth iteration, the action-situation model was retained, but Garfinkel says that Parsons added “a third term” which underscored the importance of interaction, reintroducing “the concept of a relationship between actor and situation” in which both are conceived as “systems of action” that have effects on each other. In that fifth iteration, Garfinkel says, the problems of idealism, rational individualism, structural functionalism, and the problematic idea of a cultural overlay were all overcome. The Parsons revealed in Garfinkel’s sketch of the progression of his work from 1937 to 1963; the Parsons who was collaborating with Garfinkel from

Introduction to Parsons’ Primer

79

1958 through the 1960’s, and who was also taking an interest in the work of Goffman and Sacks, is a very different Parsons from the one the critics took apart in the later 1960’s, and is completely unrecognizable as the subject of so many contemporary criticisms. The Parsons portrayed by Garfinkel would have satisfied most of his critics. Having summarized how Parsons revised his position to meet both external objections and internal inconsistencies, Garfinkel quickly moves on to a detailed explication of Parsons’ position as of spring 1963 (when Chapter VI was written). Garfinkel’s explication begins with an analysis of Parsons’ mature position, which he calls a “paradigm of social interaction”. According to Garfinkel, the paradigm relies heavily on Max Weber. While there are problems with Weber, including the fact that “events” in Weber’s formulation are not “of particular sociological interest”, and that Weber introduces the problem of “subjective” meaning in a problematic way, he also furnished a conception of a sociological environment for action that is critical to the development of a sociological formulation of action. It is, Garfinkel says, Weber’s incorporation of the response of the other into the situation of action that makes the difference. “Social action”, Garfinkel says, “is a type of action in which the actions of two or more persons are ‘oriented’ to each other in this fashion”. This mutual orientation defines a “relationship”. But, according to Garfinkel the thing Weber did that “really put sociology on its feet” was to introduce the idea of “legitimate order”. Garfinkel maintains that what he refers to as Parsons’ “paradigm of interaction” elaborated on these three essential ideas of Weber’s, providing for a set of actors, rules and mutual commitments that “provided a normative order of activities and objects, to which each of the actors held himself and the others” accountable, and employed “sanctions... to minimize discrepancies between legitimate and actual states of interactional events” (see pp. 209–210). In contrast to those who have interpreted Parsons’ position as idealist, Garfinkel says that this “paradigm is a definitional procedure” and a “sense transforming operation” with two rules, referred to earlier as Parsons’ “two theorems”, for how the theorist uses the paradigm. According to Garfinkel: “The first rule states that the real social structures consist of institutionalized patterns of normative culture. The second rule states that the stable features of the real social structures are the products of motivated compliance to a legitimate order” (see p. 211). Garfinkel emphasizes that there are many legitimate orders, and that motivated compliance is always to a legitimate order – not the legitimate order. This matters because for Garfinkel and Parsons each situation of action can have its own legitimate order, secured by commitment

80

A. W. Rawls and J. Turowetz

to constitutive practices, whereas it is often assumed that there can be only one legitimate order. According to Garfinkel, what defines an order of expectations as legitimate for Parsons are the values and norms that constitute the social agreement, or commitment, within which social action is taking place. Since, according to the first rule, “the real social structures consist of institutionalized patterns of normative culture”, Garfinkel says, “Culture consists entirely and exclusively of values and norms”. By this he means that culture – like Parsons’ social structure – does not exist (see p. 211). Rather, it is a set of procedures (grounded in a social commitment) that define a field of possible occurrences (like a game) and their expected sanctions. It is in this sense that culture is institutionalized. As such, Parsons’ values and norms define a set of procedures for producing (assembling) culture within a particular context of legitimacy. Understanding this definition of culture, Garfinkel says, requires understanding “what Parsons means by normative”, which he maintains is so central to Parsons’ later writing that it is not possible to understand his mature position without a clear grasp of what Parsons means by normative – which includes what he means by value. Parsons himself alluded to this problem in his 1963 letter when he wrote “what we mean by norms”. Understanding what Parsons means by normative also settles the question of whether he is “in sight of the real world”, because, as Garfinkel says, “Parsons keeps the real world in sight” through “the events intended by” the term normative. Garfinkel’s explication of what Parsons means by norms, normative and values makes up the remainder of this very long chapter.

Parsons’ Innovative Conception of Norms and Values – Chapter VI: Sections V through X According to Garfinkel, the various parts of Parsons’ schema are all tied together through his use of the two concepts “values” and “norms” (see p. 214). “In Parsons’ conception of society”, Garfinkel maintains, “there is an intimate tie between the sociologist’s actor’s environments that have for actors the features of being actually known or potentially known in common, i. e. of institutionally guaranteed environments-known-in-common, and the production of the real, stable society” (see p. 214). “Both terms”, values and norms, according to Garfinkel “refer to sanctioned, expected environmental events”, which are “institutionalized” in the sense that they are guaranteed by a shared commitment to them as legitimate that will be sanctioned when violated (see p. 214). These events are called values when they exhibit one set of

Introduction to Parsons’ Primer

81

features, and they are called norms when they exhibit another (the difference to be explained below). As Garfinkel explains it, both norms and values relate “normative contents to the actor’s courses of action” (see p. 219). They are, he says, relational concepts rather than “thing” (or idea/belief ) concepts. And according to Garfinkel, the way they relate normative contents to courses of action is a functional property: The term normative (as a collector for both values and norms) means “sanctioned expected ways in which environmental events should occur” (see p. 214). “Sanctioned, expected... environmental events”, Garfinkel says, are called values when they have the property of relating normative contents to the actor’s courses of action in a way that transcends the actor’s particular roles. They are called norms when they vary across social roles. But, it is incorrect to say that values are “shared” (see p. 220). Rather, values define the situated (game like) parameters of social action such that the commitment to them as rules for action does not vary across actors. They belong to situations, not actors. “Values”, Garfinkel says are “used to name the empirical matters of interest” (see p. 221); Norms, by contrast, are “shared” features of the environment of action that the values call attention to. This, he will go on to say, is something like the distinction between the constitutive rules of a game and the preferred rules of play. Values are not ideas or things. Rather, they set constitutive expectations for a particular type of social situation. In the revision Parsons uses the pattern variables to stand for the complex of values and norms that define the sanctioned legitimate expected environment of events and their assembly practices. Tacit choices are provided for as tacit institutionalized alternatives – similar to moves in games. “A value”, Garfinkel says, “is not definable except in terms of a rule that defines a correct choice”. There must be an expectation of sanction in the case of incorrect performance in order for values to define a legitimate system of interaction. Garfinkel refers to this use of the term value as a hallmark of Parsons’ position on social action and social systems, and maintains that not understanding this usage has motivated many misconceived criticisms of Parsons. For instance, the more restrictive and determinist conception of “habitus” (Bourdieu) would likely never have seemed either appealing or necessary if this feature of Parsons’ argument had been understood. The tacit assembly practices of interaction can do everything “habitus” does without the limitations. The same could be said for criticisms of Parsons’ alleged idealism and/or positivism. Values, on this view, are not ideas or goals.

82

A. W. Rawls and J. Turowetz

Parsons himself sometimes referred to values as analogous to the basic (constitutive) rules of games – although Garfinkel is quick to point out that the correspondence is “figurative” rather than literal. According to Garfinkel, such basic (constitutive) “rules provide those expected features of the environment which constitute that environment as a domain of possible events and as a system” (see p. 220). When the basic rules change, the type of situation we are in also changes. What Garfinkel is talking about here is similar to his own distinction between the constitutive and preference rules of a game, as he made it in his various Trust papers. The basic (constitutive) rules are, Garfinkel says, what Parsons means by rules of valuation: “the states that such rules of valuation define as preferred states of the system are values” (see p. 220). The ways in which these values are taken up by various actors in different specific roles are norms. Values in this sense are universal and hold across all instances of a particular type of game or situation for all players. Norms, by contrast, define specific situations of preferred play, in specific role situations of action, within the particular domain defined by the constitutive values.

Viewed this way Parsons’ Theory Provides for both Commitment and Change By grounding the possibility of social stability on the legitimacy of values (as constitutive orders of rule), Garfinkel says, Parsons provided “for the theoretical necessity of the phenomenon of commitment” (see p. 236). In other words, Garfinkel is saying that Parsons posited something like Garfinkel’s own Trust conditions as a mutual commitment that is a necessary condition for stable social action. Critics have sometimes said that Parsons created a model of a system that cannot change. But, Garfinkel says this is wrong. Parsons does provide for change. In fact, Garfinkel says, for Parsons “action by its very character... is directed to a situation which has a contingent character” (see p. 237), that is always changing across time. Values – as Garfinkel says Parsons intends them – are not words, concepts, or slogans and they cannot be inferred from actions. Values have priority over norms. According to Garfinkel, for Parsons “a system of values controls a system of norms” (see p. 239). What Garfinkel calls “good reasons” and “reasonableness” are invoked by actors in ways that reference “those desirable states of the system”, i. e. Parsons’ values, “that define what the procedural norm that ‘governs’ his course of action should have produced” (see p. 239).

Introduction to Parsons’ Primer

83

Garfinkel explains that for Parsons: “Level of commitment refers to the procedures whereby commitments are insured” (see p. 240), and while Parsons talks about the internalization of values and value commitments, Garfinkel insists that, for Parsons, values are not in the head ; that Parsons’ usage of the term values departs from the conventional use that depicts persons as “having” values (see p. 240). Values, Garfinkel says are “nowhere but in the person’s behavioral environment”, going on to say that “In Parsons’ work values are a theorist’s device. Values are not what members are concerned with” (see p. 241). They are a method, a way of talking about the fact that persons can require normative compliance of each other. According to Garfinkel, to talk of values in Parsons’ action frame means to talk about “persons who are acting in conformity to, or are committed to, or who engage in correcting each other with respect to, and thereby produce via their actions, the social structures that the sociologist encounters by means of the theoretical apparatus he employs” (see p. 246). What remains troublesome about this for Garfinkel, however, is that Parsons is still not talking about “concrete actions” which remains a sticking point for Garfinkel with regard to Parsons’ approach. Garfinkel wants to find the values/procedures that actors are actually using to assemble their actions, rather than having the sociologist posit values/procedures. The discussion of action “in its path”, or over its course, that ends section XXII takes up this issue. According to Garfinkel, in Parsons’ system “Values furnish norms their ‘rationale of legitimacy’” and “norms furnish values the programs for their production” (see p. 247). Another way of putting this is that “Whereas values are defined in terms of rules of preference, norms are not. Instead they consist of rules of procedure”. However, it is possible, Garfinkel says, for procedures to have the status of “preferred procedures” (see p. 247) without being fully constitutive (think of Conversation Analytic preference orders here).28

28

Conversation analysis has shown that actors orient to preferred ways of producing and responding to actions, treating violations of preference orders as both meaningful and sanctionable. For example, the preferred way of responding to a question is to produce a next-turn answer, without delay or hesitation, which satisfies the question’s relevance constraints. Not doing so is a “move” that can convey information. Participants must attend to the developing sequence to find out whether the lack of an answer is a technical problem (like a failure to hear), a problem of mutual reciprocity (like a failure to acknowledge what was said), or a move of another kind (such as a positioning to tell a long story). The need of other participants to figure this out is in itself a resource that the speaker can make use of. Whether the lack of an answer in the preferred position turns out to be a problem or not, and is sanctioned or not, will depend on what it turns out to have been in light of the response. It is in this sense that such preference orders are not fully constitutive: they provide for an ordered relevance of options.

84

A. W. Rawls and J. Turowetz

Answering Parsons’ Critics with EM Studies – Chapter VI: Sections XXIV to XXIX Garfinkel opens this part of the chapter with the judgment that Parsons’ answers to his critics are inadequate in every case because they require the critic to adopt Parsons’ theory, or his conception of ideology, to decide matters of fact. In Garfinkel’s view this is both inadequate – because these matters need to be settled empirically – and unnecessary, because Parsons’ position on culture as interaction, properly understood, can be given the needed empirical demonstrations. Garfinkel maintains that Parsons was “well aware” that his responses to critics were inadequate, giving us a glimpse of the relationship between the two scholars, and the suggestion that one of the reasons they were collaborating was because of the possibility that Garfinkel’s research could provide demonstrations of Parsons’ position – as Garfinkel argues in sections XXVI to XXIX. It is Garfinkel’s proposal to defend Parsons in a way that relies entirely on observable matters, and the approach to the needed empirical studies that he sketches out would, he says, rely on both Schutz and his own Ethnomethodological studies. This is possible, Garfinkel says, because “Alfred Schutz and others have furnished researchers a clarification of the meanings and uses of the critical conceptions of sociological theory that are relevant to this dispute” (see p. 251). After explaining how Parsons’ answers to critics are inadequate (even though his theory is not), and suggesting that the necessary empirical demonstrations can be built on Schutz, and Ethnomethodological studies, Garfinkel spends the remainder of the chapter discussing what this research will need to show in order to defend Parsons, and offering what he calls a “method” for conducting the research that involves a focus on demonstrating the social “commitment” that defines social order for Parsons (and Garfinkel), and on the empirical possibilities offered by “trouble” and “sanctions” for producing these demonstrations to support a theory grounded on the need for commitment. In this discussion, Garfinkel’s Trust argument, and his “breaching experiments” come into play. His studies of social identities whose performance of self is troubled (like Agnes), or who are subject to boundary maintaining degradation ceremonies (like military personnel and Black Americans), and their use for making legitimate expectations visible through sanctions, also come into play. Garfinkel’s studies of trouble are, he says, demonstrations of Parsons’ argument. With regard to the dispute between Parsons and his critics, it is Garfinkel’s objective, he says, to sketch the “fundamentals and to indicate the empirical issues and methods for settling them” (see pp. 251–252). Parsons and his

Introduction to Parsons’ Primer

85

critics, according to Garfinkel, have in common that they “are equivalently committed to the conception that the intelligibility of the phenomena of social structure requires that some provision be made for an actor whose actions occur in a culturally defined environment” (see p. 252). This environment, Garfinkel says, includes for both the “objects of the actor’s interests, known and treated by him in accordance with his ‘stock of knowledge at hand’” (see p. 252). Garfinkel also maintains that: “The idea of the actor as a course of actions, or operations upon an environment of related possibilities which Parsons refers to by his term ‘action-situation of objects’ is a fundamental conception for both Parsons and his critics and defines the elementary set of possible occurrences” (see p. 252). Garfinkel then elaborates a series of differences between Parsons and his critics. However, what ultimately makes the difference, in Garfinkel’s view, are Parsons’ positions on interaction, values, and commitment to a constitutive order. According to Garfinkel: “By following Parsons’ procedure the actor’s cultural environment is transformed in its sense to consist of a set of related analytic phenomenal states of affairs, i. e. relevances. The various functions of these relevances now define for Parsons the presence of cultural premises, values, and norms. Thus: (1) Some set of related analytic phenomenal states of affairs, i. e. relevances, constitutes the meaning of legitimacy. Any related states of affairs that operate in this fashion are called “cultural premises”. (2) Any related states of phenomenal affairs, that define for actors those features of a society that depicted members are depicted as assuming for themselves, assuming the same for others, and assuming that, as they assume it of others, the others assume it for them, are committed to a respect for, and production of – i. e. are equivalently committed to the respect for and production of, are called ‘values’” (see p. 256).

In this exposition of Parsons’ position, Garfinkel began with Parsons’ procedures and ended with a summary of his own Trust conditions. Parsons’ focus on the importance of constitutive commitment and the dependence of legitimacy on the expectation of sanctions with regard to that commitment is, according to Garfinkel, the key factor differentiating him from his critics. In fact, Garfinkel will argue that everything about Parsons’ position turns on this difference. It is the foundation of Durkheim’s sociology as well, and turns up again in Garfinkel, Goffman and Sacks. But, it is missing from most social theory. According to Garfinkel, Parsons makes the Durkheimian argument that empirical evidence of the necessary tacit constitutive commitment can be found in the sanctions. The determining factor is whether violations get

86

A. W. Rawls and J. Turowetz

sanctioned. Here Garfinkel is combining the Weberian idea of “legitimacy” with the Durkheimian focus on “sanctions” as empirical evidence of that legitimacy, and attributing the position to Parsons as an element of a system of constitutive social commitment. According to Garfinkel “It will be seen that “commitment” in the sense of turning over to others the rights to control the actor’s alternative uses of resources is built into the conception of real values as a constituent meaning of real values, and thereby as a constituent meaning of Parsons’ ‘good society’” (see p. 257). As Garfinkel explains it, for Parsons’ “the meaning for the sociologist’s actor of a phenomenal state of affairs as a real value, i. e. its status as a real value from the sociological theorist’s point of view, consists entirely and only in its ‘validity’, i. e. in the ‘guarantees of compliance’, ‘commitment’, ‘degree of internalization,’ ‘degree of institutionalization’” (see p. 257). Garfinkel argues that this can be given empirical demonstration, and explains the basis for the demonstration: “The empirical demonstration of the validity of some good society for actual members of U. S. society (in principle this holds for actual members of any society) can best be made by giving persons a chance to exhibit compliance in situations in which they could if we were wrong refuse compliance. Hence we require empirical situations in which the question of compliance can be made problematical for the actual members” (see p. 257).

In this regard, Garfinkel is arguing for the utility of “trouble”. Trouble, as his own research demonstrated, can be used as a method to reveal the values/constitutive rules that participants sanction (Turowetz and Rawls 2019). Instead of waiting for sanctions to occur, as Durkheim (1893) had proposed, Garfinkel advised the researcher to do two things: First, look for places where trouble naturally occurs (as it does with certain categories of persons, i. e. transgender, Black, mentally ill, or in troubled interactions like calls to the Suicide Prevention Center) and in those places search for evidence of the values (constitutive practices) that are being sanctioned; and, Second, make “trouble” and see whether, how, and by whom it is sanctioned. The passages where Garfinkel discusses the procedure of treating some identities of person as locations for naturally occurring troubles are obvious references to his own studies of Agnes, Race, mental illness and troubled interactions generally. The procedure of producing “trouble” where it is not naturally occurring is a recognizable description of what have come to be known as “breaching experiments” and/or tutorial problems. In essence, Garfinkel maintains that it is not Parsons’ theory that is the problem (with the caveat that Parsons’ emphasis on invariance, controlling

Introduction to Parsons’ Primer

87

exigencies, and his corresponding toleration of inequality must be dropped), but rather his inability to show critics empirical evidence of what he is arguing through studies of actual naturally occurring interaction. In so arguing, Garfinkel offers studies in ethnomethodology as the missing empirical proof of Parsons’ theory. The mystery of why this argument is missing not only from Studies, but even from anything that Garfinkel published during his lifetime will need to remain unanswered, although we can guess at some possible explanations. Parsons was coming more heavily under fire as the 1960’s progressed and the two scholars seem to have kept their working relationship to themselves. Maybe they decided that it was not to Garfinkel’s advantage to identify himself more publicly with Parsons. Then again who knows?

The Primer in Detail – Chapters VII, VIII and IX Having explained that Parsons was misunderstood because his theory was interactional in focus, and that Parsons’ conception of values and norms was focused on the constitutive assembly practices of interaction in ways that could be given empirical demonstration – points for which Garfinkel says that Parsons had been mistakenly criticized – Garfinkel turns attention in the final three chapters to what Parsons referred to as the “higher integrative functions”: economy, polity and the stratification and structured strain that accompany them. Because Parsons located meaningful social action at the level of culture as interaction, these higher integrative functions were more, not less, important in his theory than in theories not focused on interaction. There must be some way in which the myriad separate interactions are coordinated and this was the job of the higher integrative functions. But, they don’t function as macro structures, popularly conceived. Rather, they function as something of an umpire for the constitutive rule following that they oversee. In Parsons’ view, the assembly practices, or rules of cultural interaction, must be institutionalized in order to have any consistency – or integration – and the devices for this integration were located in the mechanisms of the economy and polity. In his 1963 letter to Garfinkel he also identified these higher functions with Garfinkel’s fourth theorem. In keeping with what Garfinkel has been saying about Parsons in the first six chapters, in Chapters VII through IX he describes Parsons as treating economic and political resources as oriented objects in an institutionalized system of interaction. According to Garfinkel, Parsons’ economic and political resources acquire their identities only within a legitimate program of activities

88

A. W. Rawls and J. Turowetz

to which participants are committed: an implicit social contract (see pp. 271– 272). Parsons, he says, is interested in how economic and political objects are made, in the commitments they control, and in how distinctive orders of practice are institutionalized. The value of Parsons’ social system, for Garfinkel, consists in its focus on interaction and how it makes the objects it treats empirically observable. Garfinkel explains in these chapters that Parsons differs from most other social theorists in not beginning with a commonsense object, such as money or power, and then abstracting the most general features from that object to create a concept. Rather, according to Garfinkel, Parsons’ concepts are “in search of contents”, meaning that they direct the researcher’s attention to the actual practices by which their referents are assembled. This is one of the ways in which Parsons’ theory is empirical in orientation in a way that focuses on interaction and can be demonstrated. Garfinkel has been saying throughout the Primer that this interactional aspect of Parsons’ theory explains most misunderstandings of his position. In Garfinkel’s terms, Parsons’ concepts permit members to teach the theorist what these concepts could mean, and that meaning is not reducible to a specific content, but rather consists in the function performed by the activities they designate (see p. 189). But, one cannot locate these functions by asking participants about them. Because they are institutionalized, participants are not usually consciously aware of them. As Garfinkel puts it, “One finds the empirical events he is talking about by looking for the functions that the concept provides for in the rule for looking for the functions that the concept’s definition consists of ” (see p. 281). Methods for revealing the rules that people are orienting include methods for observing sanctions. In Chapter VII Garfinkel attempts a general explanation of the role of economy, polity, money and power in Parsons’ social theory. In Chapters VIII and IX he explains and defends Parsons’ much criticized positions on stratification and structured strain. These are aspects of Parsons’ position with which Garfinkel did not entirely agree, but which he still felt had made progress in ways that Parsons’ critics had not understood and could thus profitably be defended.

Introduction to Parsons’ Primer

89

Discussion in Detail – Chapter VII: Economy, Polity, Money and Power Garfinkel opens Chapter VII with a general explanation of why Parsons is interested in economy and polity. For instance, Parsons’ interest in commodities, according to Garfinkel, “is not that it [a commodity] can be located at a particular place and not that it can be transported from one place to the next (see p. 271). Instead its interest from the “cultural-institutional approach” consists in the fact that there are rights of acquisition, of disposal, of the use of that particular thing”. The interest, Garfinkel says, is not in the thing done, but in the right to do it and in how that right obligates the actions of others. Parsons’ social system is, according to Garfinkel, a device for viewing activities as legitimate or not legitimate and how likely rights are to be enforced. In a society there are many actors, actions and activities being coordinated. In this discussion Garfinkel refers to Parsons’ actors as “units” and says that Parsons’ social system “consists of a set of categories and ways of thinking about the related character of these phenomena that these categories refer to” (see p. 273). The general social framework must provide the rules, values, commitments, and enforcements capable of maintaining motivated compliance to the overall cultural-interaction system. In providing for this integration, Garfinkel says, “The stable features of concerted activities consist of adequate solutions to the functional problems” (see p. 273). The value system must be coordinated with the myriad different interactions of the units. According to Garfinkel “the concepts of economy and polity are designed to aid in conceiving solutions to this problem” (see p. 274). The solutions afforded by Parsons’ conceptions of economy and polity involve the allocation of wealth and power resources and rights and obligations with regard to them. In particular, Garfinkel says, these resources help secure commitments from actors to the constitutive practices that regulate rights to possess and dispose of commodities in the economic sphere, and to role performances in the political sphere (see p. 239). Such commitments ensure motivated compliance with the assembly practices – norms – that are used cooperatively to produce economic and political facts. The imposition of positive and negative sanctions displays their legitimacy (or lack thereof ). In short, according to Garfinkel, Parsons approaches economy and polity as he does other topics, in terms of their relevance to the overall problem of social order, but with an emphasis on making visible the concerted actions (interactions) through which stable economic and political phenomena are produced.

90

A. W. Rawls and J. Turowetz

There are two features of the system of integration that Parsons deals with. Garfinkel calls these the normative problem and the realistic problem. The normative problem involves both maintaining consistency between the rules and their desired effects, and maintaining consistency from one aggregation of “units” to the next. The realistic problem, according to Garfinkel, involves the fact that, “The actual play of the game is not the same as the rules that define how the game should be played” (see p. 274). This discrepancy is an important one. Garfinkel again emphasizes the degree to which Parsons focuses on actual interactions, and compares his perspective to that of Kenneth Burke on this issue. Essentially, he says, this difference between the rules and the actual play can be managed to produce desired outcomes by either manipulating the situation of units or actions. For Parsons, leadership, authority, and regulation perform key functions in the polity. They are treated by Parsons as sets of norms, or preferred practices that represent solutions to the problem of ensuring members’ commitments to role performances that are valued by the collective. Leadership, which involves performance, confers not only the right to direct others, but also to allocate blame when things go awry. This is essential, Garfinkel argues, to achieving what members define as social justice. Authority confers the right to make decisions that are binding on members of a collectivity. Nevertheless, Garfinkel says, this does not absolve members of responsibility: “Normatively speaking they can be held accountable for the decisions that were made ‘on their behalf ’” (see p. 281). Those with authority do not merely represent an office, but instantiate and accomplish it as a practical fact through practices that all members have a stake in. One who holds an office “is” that office when they act in an official capacity. Through the legitimate allocation of responsibility and blame, involving adequate role performances, leadership and authority together contribute to Parsons’ elaboration of “the just society”. Finally, in this scheme, regulation has to do with discretionary actions that are not directly responsive to leadership or authority, but rather based on “unwritten rules”. Such “common law” is prototypical of norms of regulation, with competent, bona fide members of society being expected to take appropriate actions based on their interpretations of the collectivity’s history. The value pattern for the polity (by which Garfinkel, per Chapter VI, says that Parsons intends the same distinction between basic and preferred rules that Garfinkel makes) centers on effectiveness. In the economy, the parallel pattern of basic and preferred rules is based on utility. According to Garfinkel, in Parsons’ scheme three sets of norms regulate the economy: contract, occupation, and property – all of which involve role per-

Introduction to Parsons’ Primer

91

formance and interactional assembly practices (see p. 283). As with the polity, the three sets of norms constitute functional solutions to the problem of integration; in the economic sphere, the goal is to achieve “economic justice”. Contracts are central to such justice, as they specify the obligations of transacting parties toward one another; importantly, their normative features are invariant, despite the greater or lesser clarity in the terms of any given contract. Property involves the right to transfer, use, and dispose of objects. But, property also involves power over others. Enforcing rights to property requires a commitment to action from others. Thus, for Parsons, property is a relational concept: one has ownership of an object only to the extent that it is likely others will enforce one’s claim to it. For Parsons, property only exists within a social order. But, unlike classical liberalism, for Parsons, property rights are created by the social contract, rather than being the basis for it. Property is the right to expect a commitment to property rights from others, and thus entails rights to the behavior of others. Parsons also deals with the difference between occupational rights and rights involved in the employer-employee relationship. Occupational rights consist of rights to the assembly practices and role performances of an occupation by individuals or groups. The employer-employee relationship, by contrast, involves contractual relationships involving wages and labor. Contracts play an important role in Parsons’ analysis of economy and polity. Consequently, Parsons exempts relationships that cannot be contractually controlled, such as parent-child and husband-wife relationships, from his treatment of economy and polity. But in keeping with his emphasis on demonstrating Parsons’ theory through Ethnomethodological studies, Garfinkel recommends treating these family relationships in occupational terms so as to make observable the “structural incongruities” this treatment produces in the sanctions and troubles that follow (see p. 286). Finally, Garfinkel devotes the remainder of the chapter to money (see pp. 286–292). By 1963 Parsons was treating money in surprisingly interactional terms. Money was important to Parsons not only as a measure of value and as a medium of exchange, but for the “immense variety of commitments” it could control. Moreover, as an interactional resource money is “highly mobile” and does not depend on the ascriptive ties that are characteristic of gift exchange and barter in traditional societies. In modern societies money can be exchanged between strangers, and without specifying what the recipient will do with it. As such, it is a generalized medium that is equally accessible to all, as are the constitutive practices that comprise public interactions in a modern society. Such practices cannot function in a generalized way if

92

A. W. Rawls and J. Turowetz

they are not available to all members, and consequently, their predominance requires justice. When Parsons remarked to Garfinkel, Goffman, and Sacks, in 1964, that, “money is interaction”, he would have had these issues in mind. In the polity, interactive forms of power can also serve this generalized function, at least in principle. Power does not depend on ascriptive forces, and transcends any specific set of commitments, although it is capable of controlling and enforcing all of them.

Detailed Analysis – Chapter VIII: Stratification Chapter VIII focuses on Parsons’ theory of stratification. In talking about a system of stratification Parsons is talking about cultural objects and is concerned with them only insofar as they are part of a social system of activities, evaluations and treatments (he is not concerned with “natural” objects). Garfinkel opens by noting differences between two articles by Parsons on stratification, one from 1940 and one from 1953. As he did in his discussion of the pattern variable argument, Garfinkel explains that much of the confusion over Parsons’ position on stratification comes from relying on the 1940 article instead of the 1953 article which – without changing the theory – became much more empirically specific with significant implications for methods. Unfortunately, as Garfinkel notes, the 1940 article is simpler and easier to read. The more interactional and empirical 1953 article is confusing and difficult. The result is that the progress in Parsons’ argument was once again overlooked. One of the obvious reasons for being interested in stratification, according to Garfinkel, is that almost every research study uses it in some way. So, even if it were not important in its own right, it would still be important for that reason. However, as Garfinkel explains, stratification is central to Parsons’ theory, and Garfinkel is interested in his writings about it because of its relevance to the problem of social order as Parsons’ conceives it in interactional terms. This is another source of misunderstanding, because what Parsons was doing with stratification is so different from what others were doing with it. According to Parsons, a system of stratification represents a solution to the problem of minimizing the illegitimate use of power by society’s members. In Parsons’ society people are ranked on the basis of moral worth. The difference between Parsons and Garfinkel on this point is that Parsons believed that these moral rankings must be common to (i. e. agreed upon) all members of a modern society, which means that “inferior” members will embrace their own inferiority, whereas Garfinkel knew from his own personal experience and

Introduction to Parsons’ Primer

93

from his early research on race, that this is not the case. Parsons argues that all members perceive these rankings as natural and commonsensical, and become upset when persons are treated with more or less respect than they merit on the basis of their social position. Garfinkel had already done research in the 1940’s on how ranking systems can clash, such that one group’s ranking system is not embraced as fair by others. In 1946 Garfinkel had accused Parsons of embedding ethnocentrism in his system (Garfinkel 1946:2).29 This is one of the places where that embedding is apparent. The refusal of some groups of people in American Society to buy into the overall ranking system that oppresses them (women and minorities for example) and to insist that they deserve respect even when (or especially when) the overall ranking system denies it to them, is one of the more important things explaining progress toward racial and gender equality in the US, and the backlash against it. Parsons didn’t see it. Nevertheless, in spite of differences on this score, Garfinkel argues that Parsons’ theory of stratification is important and a big improvement over other approaches. On the one hand, he says, Parsons solves problems that confound other approaches to stratification; unfortunately, on the other hand, Parsons works on the assumption that a certain measure of inequality is not only inevitable, but also desirable and desired even by those in “inferior” positions, with the result, says Garfinkel, that in Parsons’ society, “some people plainly stink” (see p. 301). As a Jewish man with a socially marginal identity, Garfinkel was acutely aware that the particular stratification system of the mid-Twentieth Century US was neither inevitable nor desirable to everyone, and that people in inferior positions do not inevitably embrace their inferiority. Rather, in his view stratification is usually the result of power asymmetries that are not based on either desire or merit, which erode the Trust conditions on which modern societies depend. Thus, when Garfinkel cautions readers, “But remember this is Parsons’ version of a society” and says that in Parsons’ society “some people plainly stink”, he is not agreeing with Parsons (see p. 306). According to Garfinkel, “the frequency with which we encounter persons who see through a great deal of the moral order is not convincingly handled with the use of Parsons’ society” (see p. 306 emphasis added).30 Such persons are not the cultural 29 Garfinkel made this argument in a 1946 seminar paper. He argued that while Parsons has been “radical” in his pursuit of the necessary features that any sociological theory of action would need to account for, there were nonetheless “ontological elements in the meanings of the structures he proposed – in this case a scientific ontology – which would mean in turn that the break with ethnocentrism had been incomplete” (1946:2). In other words, Parsons embeds unexamined assumptions about social life in his theory, leading him to overlook aspects of its achieved character. 30 What Garfinkel is talking about here is analogous to Du Bois’ (1903) “Double Consciousness”.

94

A. W. Rawls and J. Turowetz

dopes that sociological theory frequently portrays them to be. The phrase in current use in the US today is “woke”. Garfinkel certainly would have been counted among those not “taken in by” the apparent naturalness of the moral order of White Anglo-Saxon Protestant society. While there are some religions in which believers embrace inferior positions as ordained, and some colonial societies in which the people who had been subjected came to revere the colonizers, revolutions typically followed. In any case this is rare in a modern society. The problem is that Parsons does not subject his taken for granted beliefs as a White Anglo-Saxon Male to the same critical scrutiny to which he subjects other aspects of social order. Nevertheless, Garfinkel maintains that Parsons makes important advances over his contemporaries in the study of stratification. That is, Parsons was able to explain how the moral order of his time was empirically assembled, even as he failed to see its problematic character. Although Parsons treats the inequality he finds as a “natural and inevitable outcome” of social arrangements when it isn’t at all, at least he finds it. One of Garfinkel’s objectives in this chapter is to contrast Parsons’ conception of stratification with conceptions more generally in use. Garfinkel achieves this comparison by laying out the principle problems in the general study of stratification by summarizing Ruth Kornhauser’s critique of Lloyd Warner’s then-influential theory of stratification. A long series of problems are elaborated. Garfinkel follows this with an explanation of how Parsons dealt differently with these issues. A main point of critique that Garfinkel focuses on is Kornhauser’s criticism of Warner for conflating the perception of stratification with stratification actualized in social action. This point goes directly to methods, and involves the difference between asking people about stratification and observing what breaches of stratification norms they sanction. According to Garfinkel, sociologists in general have trouble reconciling the subjective sense of stratification collected in verbal accounts with actual processes of stratification and the resulting rankings. In Garfinkel’s view, Parsons does not have this problem because he does not treat the objective/subjective difference the same way others have. Parsons’ approach is more interactional. Also, for Parsons, reporting on stratification is also producing it. Such accounts are relevant in a way that aligns with the narrative side of cultural sociology and its focus on discourse and stories about processes. Unfortunately, having pointed out the shortcomings in the dominant Warner approach, Kornhauser nevertheless assumes that there is a separation between people’s perceptions and the objective world “out there”, which Garfinkel faults her for. In contemporary sociology it could be said that assuming such a separation has given rise to problems like “false consciousness”:

Introduction to Parsons’ Primer

95

in other words, the separation leads to the question whether perceptions of stratification are reflections of its actual basis, or distortions produced by structural forces that can be seen by the social scientist, but not by those they study? For Parsons, by contrast, stratification is not, and cannot, be subjective because it is social. Everything social is made collectively and is therefore objective. As Garfinkel says, for Parsons, members orient a shared system of rankings not only as a matter of course, but also as a condition of their social competence (see p. 303). The social order is a moral order, and differential rankings are part of its perceived legitimacy for members. Thus, rankings are a pervasive concern for members themselves, and it is precisely because they are a problem for members that they are topic for researchers. If sociologists have trouble reconciling different, sometimes-incompatible bases of stratification, it is because members struggle with this same issue. For the same reason, Garfinkel says, for Parsons, stratification correlates with just about every outcome imaginable, as members are constantly engaged in distinguishing themselves from others, and using positive and negative sanctions to enforce these distinctions (see p. 298). Members can become upset or anxious when they receive better or worse treatment than they expect. Garfinkel illustrates this point by reporting on a study in which he had undergraduates go into stores and treat customers as though they were clerks. Not only were the customers agitated and embarrassed, the students were also uncomfortable. This study, later reported in Studies in Ethnomethodology (1967: Chapter 2) is a further instance of Garfinkel illustrating how ethnomethodology can empirically demonstrate Parsons’ theory of social order. In this vein, Garfinkel also recommends reading Vance Packard’s (1961) description of “status seekers” for how inadequacies in Packard’s approach nevertheless reveal structural incongruities induced by those who do not accept what is conventionally regarded as their “proper place” in the social hierarchy (see p. 305). Garfinkel maintains that the key difference between the 1940 and 1953 versions of Parsons’ article on stratification is that where the former was vague about the specific features on which actors evaluate themselves and others, the latter says precisely “what features [actors] can be evaluated on, what features had to be evaluated, and what these evaluations had to consist of if the concerted actions were to show stable features” (see p. 301). In the later article, Parsons placed greater emphasis on the assembly practices of interaction. The 1940 version is easier to read – probably because it is less interactional – but the 1953 version is more empirically adequate – and in consequence, Garfinkel says, can be empirically tested.

96

A. W. Rawls and J. Turowetz

This development in Parsons’ approach to stratification reflects the evolution of his thinking as Garfinkel described it at the beginning of Chapter V. The later Parsons is more specific about the constitutive assembly practices that would need to be in place for actors to produce a stable social order – in this case, to solve the functional problem of the legitimate use of power. The mature Parsons furnishes Garfinkel with empirical observables that can be investigated Ethnomethodologically by way of various breaching demonstrations. These demonstrations show how actors orient stratified objects, including other selves, in specific interactional settings, thus constituting the stratified social order that their narrative accounts – as elicited, for example, in surveys, interviews, or informal conversation – only loosely describe and gloss.

Detailed Analysis – Chapter IX: Structured Strain One general criticism of Parsons’ theory has been that it could not deal with variation and change within the social system. In Chapter IX, Garfinkel reviews Parsons’ analysis of the American Family in the late 1950’s – which he considered an empirical tour de force – and explains how Parsons’ analysis of the family provides for “structured strain” (and hence change) in ways that his critics have not appreciated. Garfinkel begins the chapter by stating that two “key conceptions” inform Parsons’ analysis of American kinship: first, the by now familiar premise that real social structures consist of institutionalized patterns of normative culture, and second, “structured strain”. According to Garfinkel, structured strains are “Normal Troubles... that persons encounter as collectivity members by reason of committed compliance with routine, valued, institutional practices” (see p. 307). This conception, he argues, allows Parsons to account not only for social stability, but also for social instability and change – something his critics have not understood. Parsons can deal with both stability and instability because he analyzes the function of the assembly practices involved in American family life and shows that there are discrepancies (strains) built into the institution. In the context of the American family, these discrepancies involve gaps between the values (rules) that members subscribe to and the outcomes they realize. What is innovative about Parsons’ analysis, according to Garfinkel, is that it shows how functional practices, which constitute solutions to problems of order, can ironically produce dysfunction among those who adhere to “routine, valued, institutionalized practices” that result in boring unfulfilled lives (see p. 307). For example, for Parsons the conjugal family – or nuclear family

Introduction to Parsons’ Primer

97

– represents one possible solution to a set of problems, including child rearing, management of household resources, access to sex partners, and inheritance. In the U. S., this family type was, at the time of Parsons’ research, considered the normative solution and regarded by members as normal and legitimate, although Parsons did note that it was found most frequently among members of the post-war urban middle class. As such, it could be invoked by members to justify normative household decisions and used to sanction non-normative ones. But, for all its functional advantages, this family form also creates significant strains for its members. For one thing, its independence from the spouses’ families of origin, along with the constant mobility required by a capitalist economy, meant that the urban, middle class family form was structurally isolated. In place of traditional kinship obligations and customs, the only thing holding this family together was the sentiments of husband and wife (note: at the time, only male-female marriages were legal and considered legitimate). Relationships were sanctioned on the basis of romantic love, and “falling in love” and “feeling right” were the conditions couples were expected to satisfy in order to have a stable relationship. However, as Garfinkel points out, insofar as they subscribe to these conditions, spouses must find a way to deal with the routine, disenchantment, and boredom that can come with an isolated long-term relationship (see p. 317). The problem couples face is analogous to what Weber called the “routinization of charisma”: any relationship that begins on a charismatic basis must eventually adjust to the mundane realities that accompany stable social forms. Under these conditions it is hard for spouses to feel “right” or “in love” over time, which can strain or destabilize the relationship. Another source of strain is that while the society values equality, and all persons are brought up to believe that they have the same opportunities, sex roles were rigidly segregated in the normative conjugal family of the American 1950’s. Women were expected to prioritize family and children over formal employment, yet their efforts in the domestic sphere did not translate into social status. It was the husband’s job that provided status and income for the family, with “women’s work” being devalued and subjected to informal, “humanistic” modes of evaluation. Husband and wife were supposed to be equal “partners” in the marriage. But, as Garfinkel points out, women did not have the structural opportunities and resources to actually demonstrate equality with men (see p. 319). This produced resentment and frustration. Then there were the dislocations that come with major status transitions over the life course: when children marry and establish their own household, when the family breaks up, and when parents, especially men, retire from work,

98

A. W. Rawls and J. Turowetz

losing a key source of their sense of being socially competent and valuable members of society. Garfinkel argues that the various strains affecting the American family were known and well publicized before Parsons’ analysis (see p. 313). What is new and important is how Parsons works out “the concept of structural arrangement”, and particularly his notion of “structured strain”. Garfinkel maintains that: “Although there had been a great deal of talk about normative and actual family structures, the use of the concept that the real social structures consist of institutional patterns of normative culture to relate normative and actual family contributions is Parsons’ contribution” (see p. 313). Parsons shows that strain is a direct result of the social structures that members assemble through their constitutive practices. It is not because of deviance, but rather because they subscribe to and actualize the value patterns that are constitutive of the conjugal family that members are vulnerable to the strains they experience. Accordingly, Garfinkel credits Parsons with offering an imminent critique of American kinship “from within”, so to speak, rather than merely observing patterns and generalizing about them, as his contemporaries were wont to do. It is of particular interest to note Garfinkel’s discussion of gender inequality in the family and how the various discrepancies in women’s roles – which led to dissatisfaction with marriage for women – also have consequences for the production of male anxiety and aggression (see pp. 322–326). Through his elucidation of a series of contradictions, Parsons notes that men are forced to identify as “bad” in order to be men because of the inferior position of women. Additionally, he says that men learn from their mothers, whom they love, that women are inferior and not to be emulated. The contradictions involved lead to problems for men as well as women. Given that Parsons was describing the normative White middle class urban/suburban family – insofar as that family still predominates among more conservative White Americans – his analysis of how female inequality in that family can lead to male members developing feelings of inadequacy, anxiety and aggression, which in turn prompt compensatory acts of “compulsory masculinity”, might explain a number of important contemporary social and political problems involving White men in the U.S.

Introduction to Parsons’ Primer

99

Conclusion – What to Make of the Primer One conclusion we draw from reading the Primer and related materials is that the relationship between Garfinkel and Parsons was close – they were even collaborators in trying to develop a comprehensive theory of social systems – and that their relationship is consequential for understanding the work of both scholars. In Parsons’ case, the upshot is that his social theory was much more sophisticated and focused on interaction than most scholars realized and that as a consequence most of the criticisms of his work had little to do with his actual position. In Garfinkel’s case, the upshot is that throughout his career he was attempting to document systems of interaction that were consistent with the broad outlines of Parsons’ theory – with the caveat that Garfinkel never accepted Parsons’ position on exigencies and his consequent acceptance of inequality. The Parsons Garfinkel worked with is not the Parsons of the critics. It is an unknown, interactional Parsons. Garfinkel did not repudiate Parsons – rather he worked closely with him. From the very beginning Garfinkel had criticized Parsons for making assumptions about how language/meaning works and he argued that these assumptions led Parsons to make mistakes about how to treat the exigencies/contingencies. This in turn led to differences between them over whether an analytic (formal analytic) approach could be adequate, or whether sociologists needed to look at actual concrete practices in order to formulate the rules of the assembly practices of social interaction systems. Parsons could never see how the messiness of actual concrete practices could theoretically be brought under control – leading him to accept the need for top-down authority and the inequality that came with it. Garfinkel, by contrast, focused on the idea that if actual people could make certainty from contingency, then a theorist could do it. But, these were differences between scholars who were working together on an approach to systems of interaction that they shared in broad strokes. Another conclusion we draw is that Garfinkel wanted us to understand that Parsons had put interaction at the center of his theory. Garfinkel makes it clear in a number of different ways that it is this particular understanding of Parsons that is needed to rescue social theory, and that it is in not having grasped the interactional heart of Parsons’ position that social theory has been weakened. Garfinkel’s manuscript constitutes a scholarly exegesis of Parsons’ argument, explaining it in great detail, with an emphasis on showing that Parsons’ theorizing was more focused on interaction than his critics understood and explaining how overlooking its interactional focus had been the cause of much of the criticism – leaving sociology without a theoretical center – which remains a problem for social and cultural theory today (Rawls and Turowetz

100

A. W. Rawls and J. Turowetz

2019). In essence, Garfinkel proposed that interaction was more important and central to Parsons’ theoretical concerns than anyone had understood. The Primer was supposed to make Parsons’ treatment of interaction and its centrality to his overall theory clear, and in the process both answer Parsons’ critics and position Garfinkel and ethnomethodology as demonstrations of Parsons’ system of interaction. Garfinkel’s differences with Parsons from beginning to end centered on the problem of how to handle contingency, and whether or not the rules, expectations, and commitments of the assembly practices of social interaction could be specified analytically by the theorist before studying social interaction in ways that would settle those contingencies, or whether they needed to be established through studies of actual concrete practices. Garfinkel insisted on the latter, treating interaction as self-organizing from the bottom up, whereas Parsons remained convinced that, because there were so many contingencies in social interaction, the move to concrete practices that Garfinkel was urging on him, would lose social order and meaning altogether. In Parsons’ view, without a top down authority there would be nothing that held constant across the contingencies. Garfinkel from the beginning thought that Parsons had made this mistake because he had taken language for granted, and not seen that it also needed to be explained in terms of interactional assembly practices (or language games). The acceptance of inequality by Parsons turns on this distinction. From the beginning Garfinkel was convinced that Parsons was wrong about contingencies: People do manage to make sense in the midst of contingency, so there must be some “method” for doing so, and it was not likely in his view to be the analytic method that a theorist like Parsons who had not studied what actual people were doing in real situations would come up with. Studies in ethnomethodology and conversation analysis have shown that Garfinkel was right about this. Durkheim’s position also falls on Garfinkel’s side of this debate. There are detailed order properties that constitute the meaning of social interaction and they can be documented. Not seeing the deep theoretical import of this point, and mistaking this research for an individual level of analysis has encouraged social theory to misinterpret this achievement and overlook its implications for a reappraisal of Parsons’ argument. In spite of the prevalence of critics who have mistakenly insisted that Garfinkel worked at an individual level of analysis, he always focused on an interactional level of analysis, with a focus on what happens between people, in a way that treated assembly practices as collective, as Parsons had. In fact, it was Garfinkel’s critique of Parsons that his earlier pattern variable argument had strong vestiges of individualism that had only been eliminated in the 1960

Introduction to Parsons’ Primer

101

revision. Garfinkel was against individualism and never studied individuals, or “individual interaction”. He took the position in his first criticisms of Parsons (1946, [1948]2006) that there is no such thing as individual interaction. Interaction takes place in the spaces between people and not inside of them (or their heads), and it always takes more than one person to interact and a system of interaction to allow for the coordination of meaning. Furthermore, he argued that interaction does not communicate shared meaning unless it makes use of shared assembly methods and symbolic/coding systems that are undergirded by a set of mutual commitments to an implicit social contract that Garfinkel called “Trust conditions”. As a consequence of these misunderstandings, Garfinkel’s important empirical demonstration that what others see as contingency has order properties, has gone unappreciated. Garfinkel’s wish for Parsons to finally understand that the pattern variables must be based on actual concrete orders of interaction, and his disappointment that not only Parsons, but all of sociology, had failed to appreciate it, would explain why Garfinkel rejected Alexander’s invitation for ethnomethodology to join the fold of Parsons’ followers in 1987–8. Garfinkel was still hoping to convince Alexander to join him in demonstrating that the actual assembly practices of social interaction should be given precedence over practices/rules that are analytically specified by the researcher/theorist, or treated as independent of situations of interaction: Only then, he argued, can Parsons’ position be empirically defended. It is a simple point, but one on which everything else rests and Garfinkel had been making it to Parsons directly since 1946. Parsons had obviously been listening to him and slowly shifting his interactional system in Garfinkel’s direction over the next two decades. But, their differences over how contingencies were to be “controlled” and standardization achieved, were never resolved. Parsons continued to think that they needed to be controlled analytically in advance. Garfinkel, however, was already demonstrating how actual people engaged in social interaction were handling contingencies in actual situations in ways that made order out of what appeared to be infinite variation; and showing also that this required equality. The difference between Garfinkel and Parsons was never about being at different “levels” of analysis: one individual and the other structural. Both had specified interaction as a system. The difference was that Parsons, believing that contingencies needed to be controlled and minimized, tried to specify the variables analytically and in advance, to minimize their number, and make them logical and manageable. Garfinkel, by contrast, believing that social interaction was orderly and workable on its own terms, trusted that if concrete practices were examined in sufficient empirical detail they would give up their

102

A. W. Rawls and J. Turowetz

secrets. In other words, what appeared to Parsons (and most other sociologists) to be messy and disorderly, with infinite and therefore unmanageable variations, was in Garfinkel’s view an “interactional system” in its own right that was being ordered second-by-second over the course of sequences of interaction, by actors who were displaying their orientation to the expectations embedded in legitimate and sanctioned systems of interaction. Actors displayed this orientation by producing orderly assembly practices for one another, and responding to the assembly practices of others turn-by-turn in expected and orderly ways. In 1962 and 1963 the first two conferences in ethnomethodology were held at UCLA and researchers were already documenting how this process works; a research program that continues today, but without much understanding of its relationship to sociological theory. If the mistake of treating Garfinkel’s position as focusing on the individual had not been made early on, scholars might have read his work differently, and realized that he was not only working with Parsons, but using Parsons’ idea of an interactional system; the difference being that Garfinkel focused on actual concrete interaction and its self-organizing practices to fill out the schema. When Alexander (1988) proposed a fix to Garfinkel’s position to bring about a convergence between the positions of Parsons and Garfinkel he proposed a move in the right direction. But, he did not realize that what they had achieved together actually went farther toward a rapprochement than what he proposed. Why the initial mistake about individualism? It may have to do with the association of Garfinkel with Schutz and phenomenology. The classical phenomenologists were philosophers, not sociologists, and thus they were essentially individualistic. Certainly Garfinkel built on Schutz and phenomenology in important respects. But, Garfinkel was beginning with the basic architecture of Parsons’ social system, and filling in the “interactional/cultural system” by studying actual concrete interaction in its course – not focusing on individuals. Garfinkel considered Schutz’ “typifications” and other phenomenal characterizations of the actor’s problem helpful in this regard. However, Garfinkel NEVER adopted the actor’s point of view. In fact, from first to last he was explicitly critical of those who did. The problem was always how – from within a field, or system, of interaction – two or more actors (members of a practice) were making meaningful objects and actions evident to each other: what rules and presuppositions they were using and how they could be sure that they were using the same ones. This is a focus on an organized system of interaction, not on the individual. It is clear from Parsons’ commentary on Garfinkel’s work that he understood this.

Introduction to Parsons’ Primer

103

There is substantial support for Garfinkel’s view of Parsons, not only in Parsons writing, and in his letters to Garfinkel, but also in their interactions. For instance, in 1964, on arriving at a meeting with Garfinkel, Erving Goffman, and Harvey Sacks in Los Angeles, Parsons said that he had just returned from “telling people” that, “money is interaction” and then laughed. The laughter plays on how surprising his critics would find the statement, and echoes a letter to Garfinkel the year before, in which Parsons recommended a paper of his on “influence”, in terms which equate influence and money with interaction, suggesting that “interaction” was becoming increasingly important to him (January 30, 1963:1): “I think it carries the analysis of the generalized medium operating in the process of interaction rather further than any previous attempt—though you will recognize the line of succession”.

It was in the published version of this talk on influence that Parsons (1963) made the argument that “money is interaction” in some detail. Money as interaction requires cooperation and reciprocity, and Parsons also referenced Marcel Mauss in this regard. The funny thing from Parsons’ perspective seems to be that if money is interaction then almost everything is interaction, and it is Parsons saying so, not what most people expect from him. But, he intimates that Garfinkel will understand. These letters and jokes assume significant familiarity and mutual understanding. Later in the same 1964 meeting, there is an interesting exchange between Garfinkel and Parsons that demonstrates their degree of mutual understanding. Parsons had been talking about money as interaction, and Goffman and Sacks were chiming in with examples of suicide. Then Garfinkel reformulated a question Sacks was asking Parsons, positing a “central banking system” and a “system of market exchanges” as equivalent to “suicide”, with regard to how a recognizable action in both related to structures of accountability and enforcement. In asking this question, Garfinkel displayed how Parsons’ treatment of money as interaction was like their discussion of suicide (1964, Reel 4 of 4, Side A): Garfinkel: Is it possible to consider the various properties of the media... to consider those... since these are the mediators of interaction... consider these as the terms in which it would be demonstrable to the parties to those activities that an event had in fact occurred that was clear, or intelligible, or usable, or recognizable... or doable in a typical fashion like that, alright, under circumstances grasped by them, ... consisting, really, their real circumstances then would

104

A. W. Rawls and J. Turowetz

consist of, let’s say, of such things now as a system of market transactions mediated by money controlled by a central banking system as compared with a system of market exchanges... which consisted of a series of barter relationships or gift exchanges, or any of the rest. So in that sense your suicide it could be said – it seems to me that I can detect in your argument that you’re saying, “Therefore, as we examine any society for the way in which it’s stage, so to speak, in the development of the structures of...” One could therefore ask “what were the conditions under which someone making himself dead would have to be counted a suicide by...” and then we would also get it by the variety of agencies that would be empowered, personages so to speak, that is, the agents in the society who would be required to recognize it, who would be required to do anything but recognize it, for whom it would be a matter of their direct concerns. Parsons: What kind of enforcements... Garfinkel: Right, which is to say, the production and recognition of that event would be a matter in which you could not be indifferent. It would be an enforceable matter that you see that this is what happened. Note that Parsons’ statement “What kind of enforcements...” completes Garfinkel’s prior utterance. It is not a question, but a confirmation of that utterance that also reiterates the main point of the theoretical position they are both taking. Garfinkel’s proposal is that accountable structures determine the conditions under which agents in a society would be required to recognize both suicide and money as achieved social objects. Parsons confirms that it is an issue of “What kinds of enforcements...” actions are accountable to. If actions are legitimate and sanctionable there must be enforcement. Garfinkel treats it as an agreement and says “Right”. This exchange exhibits their shared understanding of money and suicide as media – mediums – of interaction that involve assembly practices, sanctioned accounts and enforcement. By placing money and suicide sequentially “side-by-side” in this interaction and then offering an explicit formulation of that placement, Garfinkel shows that his own interest in accounts is similar to Parsons’ interest in enforcement. Garfinkel is treating money and suicide as exchangeable terms in this talk about mediums of interaction/conversation, and Parsons had been doing so himself, which is what made the conversation seem weird in the first place.31 31 This point of agreement contrasts with parts of the discussion where Parsons disagrees with Garfinkel, so we know he is not just agreeing to be nice. This sequence contrasts with an earlier attempt in the same

Introduction to Parsons’ Primer

105

As late as 1966 Garfinkel arranged for Parsons give a “Chairs” talk at UCLA (transcript in the Garfinkel Archive). This suggests that he was still hoping in 1966 that an interactional specification of members’ assembly practices could make their combined approach to social theory viable. The big progress in Garfinkel’s approach had come in 1954, when he formulated the jury study in terms of concrete ethno-methods. His discovery of preferred orders of selforganizing detail in conversation with Sacks in 1962 was another move in the same direction. Neither involved a theoretical change, or a repudiation of Parsons, and both occurred before, or during, Garfinkel’s work on the Primer. The insight that came in 1962 while Garfinkel and Sacks were working together on the Suicide Prevention Center data was rather, that having discovered the self-organizing detail in the assembly work of the practices they were studying, Garfinkel and Sacks realized they could proceed without theoretical framing; that their empirical work was in some important sense solving theoretical problems on its own. Given the problems that Parsons and social theory in general were starting to encounter a decoupling likely became expedient. For these and many other reasons the several theoretically oriented binders/books that Garfinkel had so carefully prepared were never published. Parsons was coming increasingly under attack. Garfinkel’s efforts to defend Parsons had not been taken seriously by anyone, and it must have seemed that there was no point in continuing the effort. Garfinkel’s analysis of Parsons was still relevant to his own work, however, and he kept circulating the manuscript.32 Although the political atmosphere of the late 1960’s, with Vietnam War protests and the beginning of the Post Structural and Post Modern critique, was a very uncomfortable time for Parsons, and it probably would not have helped Garfinkel’s case at the time to have been identified more closely with him, it is still a shame that this book about Parsons was not published. In retrospect, the manuscript helps explain both of their perspectives, which seems to be what Garfinkel had hoped for. Certainly, Garfinkel continued to consider Parsons’ perspective important enough to teach it, and circulate the original 1960 mimeo-version of the Primer for decades. If the manuscript had been interaction by Garfinkel to state an equivalence between money and language exchange – that was not accepted by Parsons as stated – but at least had the right parts – which suggests that Garfinkel knew in detail what was coming. 32 There were many personal conversations between Anne Rawls and Garfinkel, Schegloff, and others about the futility of continuing to have theoretical discussions. Schegloff said in 1977 that he and Sacks had decided to get on with the empirical research because they would be long gone before anyone woke up to the theoretical implications of their work. He had previously tried to articulate his position theoretically for over 20 years without success. It is not surprising that he stopped doing so the empirical work started to become persuasive on its own.

106

A. W. Rawls and J. Turowetz

published we might have learned to read Parsons differently by now – we might still – and who knows what that could change. The Parsons of the Primer isn’t the Parsons whom critics were hounding, although problems remain. It is also not the Parsons whom critics claim Garfinkel repudiated. It is instead a Parsons close to Garfinkel’s own heart: The Parsons we can see Garfinkel writing about in the Primer and writing to in his letters. But we don’t have to take Garfinkel’s word for it. The evidence of their relationship and convergence of ideas is in Parsons’ own letters, his own words, and his own publications (Parsons 1963).

References Alexander, J.: “Action and Its Environments”, J. Alexander (ed.): Action and Its Environments: Toward a New Synthesis. New York: Columbia University Press 1988: 301–334. Beer, S. H.: “Causal Explanation and Imaginative Reenactment”, History and Theory 3(1) (1963): 6–29. Burke, K.: Permanence and Change. New York: New Republic 1936. Du Bois, W.E.B.: The Souls of Black Folk. New York: Routledge [1903]2004. Durkheim, E.: The Division of Labor in Society. New York: The MacMillan Company [1893]1933. Durkheim, E.: The Elementary Forms of Religious Life. New York: Free Press [1912]1955. Garfinkel, H.: The History of Gulfport Field, 1942, Volume II, Part II: The Aircraft Mechanics School. A. W. Rawls (ed.) University of Siegen [1943]2019. Garfinkel, H.: “Some Reflections on Action Theory and the Theory of Social Systems”, unpublished manuscript, Garfinkel Archive, Newburyport, MA 1946. Garfinkel, H.: “Notes on the Thematic Apperception Test”, unpublished manuscript, Garfinkel Archive. Newburyport, MA 1947. Garfinkel, H.: “The Red as an Ideal Object”. Ethnographia e Ricerca Qualitativa (1) ([1947]2012). Garfinkel, H.: Seeing Sociologically. Boulder, CO: Paradigm Publishers [1948]2006. Garfinkel, H.: “A Further Note on the Prospectus for an Exploratory Study of Communication and the Modes of Understanding in Selected Types of Dyadic Relationship with Particular Reference to the Jew as an Object of Social Treatment”. Unpublished manuscript, Garfinkel Archive, Newburyport, MA, October 21, 1948 Garfinkel, H.: Perception of the Other: A Study in Social Order. PhD Dissertation, Cambridge, MA: Harvard University 1952. Garfinkel, H.: “Notes on Language Games as a Source of Methods for Studying the Formal Properties of Linguistic Events”, European Journal of Social Theory, 22(2) ([1960]2019): 148–174.

Introduction to Parsons’ Primer

107

Garfinkel, H.: “Aspects of the Problem of Commonsense Knowledge of Social Structures”, Transactions of the Fourth World Congress of Sociology 4 (1959): 51–65. Garfinkel, H.: Letter to Edwin Shneidman, Garfinkel Archive, Newburyport MA, March 22, 1962. Garfinkel, H.: Letter to Egon Bittner, Garfinkel Archive, Newburyport, MA, February 5, 1963. Garfinkel, H.: A Conception of and Experiments with Trust as a Condition of Stable Social Actions, O. J. Harvey (ed.) Motivation and Social Interaction: Cognitive Determinants. New York: Ronald Press Company 1963: 187–238. Garfinkel, H.: Studies in Ethnomethodology. Englewood Cliffs, NJ: Prentice Hall 1967. Garfinkel, H.: “Evidence for Locally Produced, Naturally Accountable Phenomena of Order, Logic, Reason, Meaning, Method, etc. in and as of the Essential Quiddity of Immortal Ordinary Society (I of IV): An Announcement of Studies”, Sociological Theory 6(1) (1988): 103–109. Garfinkel, H.: Ethnomethodology’s Program: Working out Durkheim’s Aphorism. Lanham, MD: Rowman and Littlefield 2002. Goffman, E.: “The Interaction Order”, American Sociological Review 48(1) (1983): 1–17. Knorr-Cetina, K. and U. Bruegger: “Global Microstructure: The Virtual Societies of Financial Markets”, American Journal of Sociology 107(4) (2002): 905–950. Lidz, V. and H. Staubmann: “Introduction”, T. Parsons: Actor, Situation and Normative Pattern: An Essay in the Theory of Social Action, V. Lidz and H. Staubmann (eds.). Piscataway, NJ: Transaction Publishers 2010: 5–30. Lynch, M.: Scientific Practice and Ordinary Action. Cambridge: Cambridge University Press 1993. Merton, R. K.: “Durkheim’s Division of Labor in Society”, American Journal of Sociology 40(3) (1934): 319–328. Mills, C. W.: “Situated Actions and Vocabularies of Motive”, American Sociological Review 5(6) (1940): 904–913. Packard, V.: The Status Seekers. New York: Pocket Books 1961. Parsons, T.: The Structure of Social Action. Glencoe, IL: The Free Press 1937. Parsons, T.: The Role of Theory in Social Research”, American Sociological Review 3(1) (1938): 13–20. Parsons, T.: “The Prospects of Sociological Theory”, American Sociological Review 15(1) (1950): 3–16. Parsons, T.: The Social System. Glencoe, IL: The Free Press (1951). Parsons, T.: “The Pattern Variables Revisited: A Response to Robert Dubin”, American Sociological Review 25(4) (1960): 467–483. Parsons, T.: “On the Concept of Influence”, Public Opinion Quarterly 27(1) (1963): 37–62. Rawls, A. W.: “The Interaction Order Sui Generis: Goffman’s Contribution to Social Theory”, Sociological Theory 5(2) (1987): 136–149.

108

A. W. Rawls and J. Turowetz

Rawls, A. W.: “Can Rational Choice be a Foundation for Social Theory?” Theory and Society 21(2) (1992): 219–241. Rawls, A. W.: “Durkheim’s Epistemology: The Neglected Argument”, American Journal of Sociology 102(2) (1996): 430–482. Rawls, A. W.: “Durkheim and Pragmatism: An Old Twist on a Contemporary Debate”, Sociological Theory 15(1) (1997): 5–29. Rawls, A. W.: “Race as an Interaction Order Phenomenon: W.E.B. Du Bois’ ‘Double Consciousness’ Thesis Revisited”. Sociological Theory 18(2) (2000): 239–272. Rawls, A. W.: Durkheim’s Epistemology: The Elementary Forms of the Religious Life. Cambridge: Cambridge University Press 2004. Rawls, A. W.: “Harold Garfinkel, Ethnomethodology and Workplace Studies”, Organization Studies 29(5): 701–732, 2008. Rawls, A. W.: “An Essay on Two Conceptions of Social Order: Constitutive Orders of Action, Objects and Identities vs. Aggregated Orders of Individual Action”, Journal of Classical Sociology 9(4) (2009): 500–520. Rawls, A. W.: “Durkheim’s Theory of Modernity: Self-Regulating Practices as Constitutive Orders of Social & Moral Facts”, Journal of Classical Sociology 12(3) (2012): 479–512. Rawls, A. W.: “The Wartime Narrative in U. S. Sociology, 1940–1947: Stigmatizing Qualitative Sociology in the Name of ‘Science.’”, European Journal of Social Theory, 2018; DOI: 1368431018754499. Rawls, A. W.: La Division du Travail Revisited: Vers une Théorie Sociologique de la Justice. (The Division of Labor Revisited: Toward a Sociological Theory of Justice). Philip Chanial (ed.), translated by Francesco Callegaro and Philip Chanial. Paris: l’Bord de l’eau 2019. Rawls, A. W.: “Introduction to Garfinkel’s ‘Notes on Language Games’: Language Games as Cultural Events in Systems of Interaction”. The European Journal of Social Theory 22(2) (2019): 133–147. Rawls, A. W. and W. Duck: Tacit Racism. Chicago: University of Chicago Press, forthcoming. Rawls, A. W. and J. Turowetz: “‘Discovering Culture’ in Interaction: Solving problems in cultural sociology by recovering the interactional side of Parsons’ conception of culture”, The American Journal of Cultural Sociology 2019. Sacks, H.: Letter to Garfinkel about circulating the Primer manuscript, (undated), Garfinkel Archive, Newburyport, MA. Turowetz, J. and A. W. Rawls: “The Development of Garfinkel’s ‘Trust’ Argument from 1947 to 1967: Demonstrating how Inequality Disrupts Sense and SelfMaking”, Journal of Classical Sociology, 2019. Weider, D. L.: Language and Social Reality. The Hague: Mouton 1974.

Parsons’ Primer Harold Garfinkel

April 1962

VOLUME 1: “AD HOC” USES

Harold Garfinkel University of California, Los Angeles

© Springer-Verlag GmbH Germany, part of Springer Nature 2019 A.W. Rawls (Ed.), Harold Garfinkel: Parsons’ Primer, Beiträge zur Praxeologie / Contributions to Praxeology, https://doi.org/10.1007/978-3-476-04815-8_2

109

110

H. Garfinkel

TABLE OF CONTENTS 1. The Book’s Task 2. Introductory comments to the problem of social order 3. The problem of social order 4. “Adequate” descriptions of social structures 5. An illustrative reading of Parsons’ revised paradigm of the components and processes of action systems of rules; a theory for assembling stable social structures 6. Partial draft and note for a chapter on values 7. Economy, polity, money, and power 8. Partial draft of materials on 1940 article on stratification: ad hoc Parsons 9. Partial draft for illustrative use of “structured strain”: ad hoc Parsons

111

Parsons’ Primer

CHAPTER I THE BOOK’S TASK It is the task of this book to recommend Parsons to researchers by teaching the reader how to read Parsons in order to make use of his work for theoretical and empirical sociological researches. His work, which is one of many available solutions to the problem of social order, is an exceptional solution in its originality, rigor, scope, and usefulness. But because Parsons is so difficult to read his work has been used less and criticized more than it deserves. The task of this book is to show how to read past what Parsons writes to see what he is talking about. I hope that by the conclusion of the book the reader, having learned to do this for himself, will have seen for himself that these are not empty claims. Parsons is generally esteemed as a leading sociological theorist in the United States. At the same time his work is the object of scandalous criticism. This book would fail if it did not set itself the task of answering these criticisms. But it would fail too if it were directed to this task as a polemical one. Given an interest in the possible uses of Parsons’ work, a more appropriate method for answering the criticisms is to confine attention to a description of his work. Convincing answers to the criticisms would then be obtained if the reader was able to decide for himself whether the criticisms adequately describe Parsons’ work. In this chapter I shall list the more common criticisms as “true but trivial”; “true but irrelevant”; “would be important if true, but false”; or “important and true” but without furnishing a justification for the way I have classified the criticisms. At the conclusion of the book the list will be reviewed again, and the classification will be justified by consulting the materials of the book’s arguments. At the outset several criticisms may be put aside as true but trivial: that Parsons’ writing style is formidable; that he uses strange, ungrammatical, ponderous terminology; that he does not furnish enough illustrations. It is true but irrelevant that Parsons’ theory of a stable society can be read as a defense of the status quo. The following four groups of criticisms are the heart of the matter because they would be important if they were true, but they are false. A first group of criticisms is concerned with Parsons’ theory as one among competing alternatives. The criticisms run as follows: that his theory amounts for the most part, to an obscure way of saying what is well known and used by sociologists via more familiar terminologies; that little or nothing original

112

H. Garfinkel

is to be found in Parsons’ writings; that his claim to have gone “well beyond what is now current and even in the best literature of the subject” is untrue; that Parsons’ claims to originality, e. g. in Towards a General Theory1 and in The Social System2 consists of the “discovery” of the interdependence of personality, culture, and society; that the large expenditures of effort that are necessary in order to come to terms with his difficult writings yields little more than familiar concepts and commonplace findings; that Parsons is unfamiliar with the literature of American sociology and by reason of this has merely rediscovered common coin. A second group of criticisms that would be important if they were true but are in fact false describe the lack of research utility of his work. These criticisms assert that Parsons’ theory is neither intended for nor useful in research; that his theorizing is speculative; that it has only occasional and highly selected contact with and is not controlled nor controllable by investigative procedures or empirical knowledge; that there is little discernible and certainly no obvious correspondence between his theoretical constructions and the observable events that these constructions are intended to order; that he provides no acceptable method whereby hypothetical empirical propositions may be derived from his theoretical constructions; that predictive pay-offs must be aided by ad hoc concepts and post hoc interpretations; that an exceptionally large input of concepts yields an exceptionally small output of predictions. A third group that would be important if they were true, but are false, are of general methodological import. They assert that Parsons provides an interminable number of cross-classifications the purposes of which are either obscure or non-existent; that column and row headings are arbitrarily assigned while cell entries are little more than metaphorical rubrics; that the method for constructing and using such classifications consists of the creative search for and more wit with words; that if these paradigms exemplify Parsons’ conception of the method of formal analysis then he has substituted word play for formal analysis’ central task of clarifying the essential structures of problematic phenomena; that Parsons does not distinguish between matters of fact and matters of definition; that characteristically his definitions are either exceptionally complex and thereby obscure, or make much ado about obvious matters; that his major concepts lack operational definition; that in his repeated references to “patterns” he has substituted a concern for the logical consistency of observed events for causal chains; that he has developed 1 [Editor’s Note: Parsons, T., E. A. Shils, & N. J. Smelser (eds.) (1965): Toward a general theory of action: Theoretical foundations for the social sciences. Piscataway, NJ: Transaction Publishers. Garfinkel likely used a pre-published manuscript of this book.] 2 [Editor’s Note: Parsons, T. (1951): The social system. Glencoe, IL: The Free Press.]

Parsons’ Primer

113

a classification and not a theory; that his constructions are not demonstrably consistent with each other; that his claim to have deduced types of social structures from “lower order constructions” is not defensible; that his scheme is too massive; that it does not permit piece-meal grasp or piece-meal test; that something important and vital to the progress of sociological theorizing and inquiry is lost through constraints imposed by the discipline of a systematic position; that a sufficient amount of rigorous data does not exist in sociology today to support a strong theory; that progress in the discipline is best guaranteed by theories of the middle range. A fourth group of criticisms that would be important were they true but they are false deals with the substantive arguments within Parsons’ work. These run that Parsons’ use of structural functional theory subjects him to the criticisms of functionalist interpretations; that his brand of “structural-functional theory” involves him in the fallacy of functional teleology; that Parsons’ social system is an ontological conception and involves him in the fallacy of misplaced concreteness; that he has failed to derive or to incorporate conventional sociological concepts like associations, in-groups, and the like which, in the words of one critic, “are the very ones that tend to require some uniting of organisms with the environing conditions set by other organisms”;3 that in postulates of equilibrium, and inertia, and in his references to self-maintaining systems, cybernetic controls, balances of forces, mechanisms, and the like Parsons is using physical concepts; that sufficient weight is not given to the way in which the social structures are part of the cultural map; that the concept of a strain towards consistency in culture involves all the criticisms that can be directed to the rationalistic fallacy; that the pattern variables cannot be appropriately regarded as dichotomous; that the systems of classification generated by the pattern variables do not “fit” the “needs” of personality, social, and cultural theory; that with his emphasis on perceived situations Parsons does not have a defensible way of reconciling his actor and his actor’s society with the real social structures; that Parsons’ depictions of actor, situation, and orientations of action are almost exclusively intra-organismic; that actor, situation, and orientations permit some events occurring within the organism to be described, but do not permit these occurrences to be related to the events “that set them in motion and determine their direction;” that if the properties of actors, situations, and actors’ orientations to situations are examined it will be seen that they are not three categories of properties but one – the orientation of actors to situations; that Parsons’ concepts of meaning and expectations are 3

[Editor’s Note: The critique referenced here is by G. E. Swanson. See Swanson, G. E. (1953): “The approach to a general theory of action by Parsons and Shils”, American Journal of Sociology 18(2):125–134.]

114

H. Garfinkel

obscure; that the categories of cognizing, cathecting, and evaluating are the same categories as those of psychology and are subject to the same criticisms; that in his conception of an actor who “chooses” and “decides” Parsons has committed the familiar intellectualistic fallacy; that his actor is a very rational fellow indeed; that the subjective categories and the “actor’s point of view” are obscure in import or neglected or both; that too much is made of the subjective categories; that the subjective categories are problematic for Parsons’ actor only after the actor has developed sufficiently to have developed a symbolized world; that the actor’s meaningful world is an amalgam laid upon a physical core; that the actor is hedonistically conceived with respect to motivations and values that govern his choices; that the social system is little more than an intricate socially structured schedule for administering rewards and punishments; that Parsons has retained the means-ends frame and with it the difficulties that attend its use; that for Parsons social change and disorder are synonymous; that he does not have an adequate theory of social conflict; that for Parsons stable features of social structures and social conflict are contrary if not contradictory conceptions; that Parsons has at best a classification of types of change of social structures, but does not have a theory of such changes; that with respect to the state of a system, Parsons’ problems arise only with respect to what it takes to change the state whereas the maintenance of that state is not problematic; that Parsons makes little if any provision for the individual as a source of change; that the individual is sacrificed to the good of the stable system; that Parsons’ empirical descriptions of American society are trite, strange, uninformed, incorrect, uncorrected, and uncorrectable by the “facts;” that in forever finding “convergences” Parsons exaggerates the actual unity and coherence of the main bodies of sociological thought, findings, and problems. In sum, most of the common criticisms of Parsons are trivial, irrelevant, or false. Are there then no criticisms that are both important and true? There are many, although surprisingly few are to be found in published reviews of his work. To cite a few, it is both true and important that (1) the pattern variables are not an exhaustive list. (2) The role of time in the structuring of an action and actor’s situations is handled ingeniously but unconvincingly by Parsons although the same is so for much, and perhaps all of contemporary sociological theory. (3) When the related social structures are conceived according to Parsons’ structural functional method their features of time of occurrence, pacing, duration, sequence, phasing, and the like are handled as properties of a “fat moment” much in the manner of an historical epoch. (4) The empirical grounds to justify Parsons’ assumption that a single integrated set of ultimate value premises is a necessary condition of stability of a social order and that

Parsons’ Primer

115

this set consists in the United States of “instrumental activism” remains to be empirically demonstrated in a convincing way. (5) Within Parsons’ own program different readings of Weber and Durkheim are possible, with serious consequences for the interpretation of Parsons’ work. (6) Although Parsons’ program of theoretical researches has been going on for many years, empirical researches motivated by the theoretical program are of markedly uneven quality and overall are in short supply. Before all others, one criticism stands out as both important and true: Parsons is difficult to read. This criticism is frequently accompanied by the erroneous belief that Parsons’ best work, the work that is most useful for researchers, and the work that bids to last, is his early work. This book has been written with the aim of furnishing a corrective.

CHAPTER II PARSONS’ THEORY AS RULES OF SENSE TRANSFORMING PROCEDURE In order to use Parsons’ work for research a grasp is required of the kinds of problems that his theory is designed to solve. The problem of social order, about which we shall talk more exactly in later chapters, embraces almost all of the problems to which Parsons’ work is directed. The entirety of his work can be pulled together by conceiving it as a solution to the problem of social order. A solution; not the solution. There are many solutions. When Parsons’ solution is compared with alternatives, most particularly with those available in text books as well as with those of other leading sociological theorists, his solution shows many contrasting and original features. Parsons’ solution consists of two theorems: (1) The real social structures consist of institutionalized patterns of normative culture. (2) The stable properties of the real social structures are guaranteed by motivated compliance to a legitimate order.4 As they stand these two theorems are much too formularized. They convey little more than a sketch of the matters that the remainder of the book will finally have spelled out. They may serve us in the way that a prologue serves the 4

[Editor’s Note: There is a note in the margin beside the theorem that reads: “Worked out in lecture for Soc172 spring 1962.”]

116

H. Garfinkel

remainder of a play, furnishing us at the beginning with the outcome which we may use as a precedent and a context for understanding the steps that lead up to it and which are the outcome’s explication. The usefulness of Parsons’ work for researchers consists in the specific ways in which his theorizing elaborates one facet or another of these two theorems as a solution to the problem of social order. It is the purpose of the next several chapters to show (a) that every theory of social organization is a solution to the problem of social order, and (b) that when such solutions are treated by the theorist as rules of theorizing and inquiring procedures to which he seeks to comply, that these rules define in the work of the theorist the concept “adequate structural analysis”. It is by reason of this equivalence as well as in the particular ways in which it is so in Parsons’ work that his theorizing not only takes empirical knowledge of social structures as a necessary point of departure but is at every point in its development controlled or controllable by empirical knowledge.

I Sociology as Procedures for Transforming Commonsense Knowledge of Social Structures Sociology can be characterized in many different ways: by describing its attitude, its problems, its methods, its practitioners, its historical development, its body of fact, its theories. For an introduction to Parsons it is useful to begin by considering how the activities of sociological theorizing and investigation are characterized from the standpoint of analytical philosophy. I am referring particularly to the type of philosophical analysis that is found in the work of the German phenomenologists.

Their work abroad is represented by phenomenology’s founder, Edmund Husserl. In this country leading figures are Alfred Schutz, Aron Gurwitsch, Marvin Farber, Herbert Spiegelberg, and Dorion Cairns. Two other figures who, early in their careers were members of the Vienna Circle, but later departed radically from their early doctrines and who are to be considered contributors to the methods of analytic philosophy, are Felix Kaufmann and Ludwig Wittgenstein.5 5

[Editor’s Note: Garfinkel placed his footnotes in the text between lines like this. We have kept them as they were in the original manuscript, both to preserve his text and to facilitate the use of editor’s notes.]

Parsons’ Primer

117

From the standpoint of analytical philosophy, the theorizing and investigating activities of sociologists consist of sense transforming operations. According to analytical philosophy, sociology is in the business of transforming available descriptions of life-in-society in accordance with the rules of sense transforming procedure found in the methods that sociologists actually employ and to which they actually subscribe as correct ones. Descriptions of life-in-society are treated by societal members as objects of their practical everyday concern. Such descriptions depict for members the socially organized and operating features of a society in which the members live and with which they must somehow come to terms for the management of their everyday affairs. Of course such members’ descriptions are not confined to descriptions that describe only features of practical interest. There are for members descriptions of possible as well as actual societies. Their descriptions exhibit the full range of logical modes, and include conjectured, hypothesized, imaginary, dreamed, fantastic and similar societies. Descriptions of life-in-society, whether they be initially available and known with respect to theoretical scientific, practical, imaginary, religious, esthetic, moral, or playful interests – i. e., in any way whatsoever – are transformed in their sense through the methods of sociological theorizing and inquiry. The product of these methods, which is a transformed description, sociologists refer to with the term “real social structures.” For the moment we shall not make anything more of the point that such an exceptionally important term as “real social structures” refers to whatsoever transformed descriptions, are products of whatsoever methods, sociologists actually use. Later we shall examine this point with more exactness. The following notation may help to make clear what I am talking about in saying of sociological activities that they are sense transforming procedures: K ˆ ! T. Let K stand for any descriptions whatsoever of life-in-society that the sociologist might treat as a point of departure for a theoretical or empirical sociological inquiry. K might consist of descriptions of persons, financial arrangements, territorial distributions of persons, attitudes, motives, kinship arrangements, the after-life, future states of the society – anything at all. The term ˆ designates some set of operations that the sociologist performs upon K, operations like bringing K under a definition, describing it, counting some of its features, judging it, matching it with something else. ˆ consists of the operations of criticizing, classifying, rearranging, neglecting, comparing, revising, testing the materials of K. ˆ consists of activities of sociological theoretical and empirical investigation. Methodwise, ˆ consists of the rules of

118

H. Garfinkel

theorizing and inquiring procedure that sociologists actually use in deciding that between alternative means of classifying, neglecting, comparing, counting, revising and so on, the correct one was chosen. The product of these activities is another description of the society. When the product is compared with K the product shows change of sense. When K has been transformed by the operations ˆ we shall speak of the result as the product, T, of the procedures ˆ on K, or notationally, of the procedure K ˆ. Because the notions of sense transforming operations and their products are critical, I would like to furnish them some illustrations. Example 1. During the war my uncle had occasion to go to a government office because he wanted an increase in his allotment of fuel oil. There he complained to a clerk that his allotment was insufficient.6 He had a long story with which to justify his request for an increase. He described his circumstances at home. It was cold in the house; his wife was unpleasant because it was so cold; there was that large dining room which was always hard to heat even when you could buy as much fuel oil as you could afford; he was living in a particularly cold part of town; the children were down with one illness after another, one giving what ailed him to the next, so there was no rest for anyone; and so on. After several minutes of this the clerk stopped him, “How large is your house?” The story started again describing how large the house was; how it had always been a burden; that his wife and not he had wanted the house. The clerk interrupted again. “Excuse me, how many rooms do you have in the house? How many square feet?” My uncle told him. “What kind of heater do you have?” and “What was your allotment last year?” And so it went. Out of the flow of material with which my uncle described his situation the clerk established about four or five points. The clerk understood of course that the situation as my uncle described it was a fix in which a person could be. But the clerk consulted the rules of office operation, and in terms of these rules, exemplified in the information that was asked for on the form that the clerk filled out, the clerk undertook the process of selection, of classification, and the rest such that the clerk came up finally with what from the standpoint of the administered form was “the case.” There was the description of the social structures that my uncle furnished the clerk. The transformed description of my uncle’s circumstances found in the form described a world which did not include complaining wives, or a house whose size and expense were regretted. Such features, though known to the clerk, were not relevant. Instead the clerk described a social situation which included instances of houses with certain square footages, with certain 6

[Editor’s Note: This narrative first appears in a Black Notebook of Garfinkel’s papers from 1946–1947.]

Parsons’ Primer

119

typical numbers of rooms, with certain typical inputs of oil to certain types of heaters, that would on the average produce certain units of heat over a unit period of time, with the expected result that some expected amount of a categorized scarce commodity would have been used up by one instance of a “home owner.” When we ask how does one produce the description on the form from the description that my uncle gave, we shall have described how the former is a transform of the latter, i. e. how an operation of ˆ on K produced T. Example 2. A person walks into the waiting room of the U.C.L.A. Psychiatric Outpatient Clinic and is asked by the social worker why he has contacted the clinic.7 In reply to the social worker’s question the person produces a story in which various troubles, troubling circumstances, and persons are implicated. After attending to the account and with the aid of questions to formulate the person’s remarks as answers to these questions, the social worker writes on a card an account that thereafter will be consulted as “the presenting complaints”. When a tape recording of the conversation is compared with the account on the card the story on the card consists of a more or less unified picture of that person conceived as a potential psychiatric patient. The person, his circumstances, his remarks, are not described by the social worker as those of a customer, or a fellow-conversationalist, or as a client, or a friend, or even as a fellow human, but rather as a potential patient. Some “theory” of social psychiatric and clinic functioning is employed by the social worker and serves her as a set of rules of relevance in terms of which the patient’s characterization of the social structures within which the patient operates is more or less systematically transformed so as to produce the features that are known via the account that the social worker writes on the application form and to which she refers as the “patient’s presenting complaints.” Example 3. We once arranged and recorded a series of experimental interviews which involved a psychiatrist and several adult experimental subjects. The experimental subjects had not been told beforehand that the psychiatrist would simply sit there and say nothing. When the subject came into the presence of the psychiatrist a one-sided conversation began. The experimental subject tried, for the most part in vain, to learn from the psychiatrist what was expected of him. “Well, what do you want me to say, Doc? Why am I here? What’s this about?” To all this the psychiatrist tried and in large part succeeded in saying nothing. He sat. Occasionally he nodded. Sometimes he 7

[Editor’s Note: This example refers to research Garfinkel was doing at the time with Egon Bittner, and which was eventually published as Garfinkel, H. (1967): “Good organizational reasons for bad clinic records”, Chapter 6 in Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice Hall.]

120

H. Garfinkel

said a few words, but for the most part he sat with his chin in his hand and stared at the person. Subjects showed a great deal of disturbance. They tried to elicit a reply, any reply; they tried to return the stare; they tried to play the game with the psychiatrist that they felt the psychiatrist might be playing with them. When the sessions were finished we decided that for the while we were not going to be interested in what the two persons said to each other. Instead we would try to characterize the presence of status considerations by attending only the possibility that there might be a rule that operated in such “situations of interaction” that regulated the sheer reciprocal exchange of bursts of talk. We supposed that irrespective of what was said, just the fact that persons were talking for a length of time might itself be a feature of the exchange that is socially controlled. Counting these as the events of interest, we imagined a world in which interactions between persons consisted only in the fact that persons were engaged in exchanging bursts of utterances. To describe the actual events in such a world the following procedure was used. Instead of playing back the tape at 3 34 feet per minute which was the speed at which the conversation had been recorded, we played it back at 7 12 feet per minute. This produced bursts of Donald Duck talk separated by easily heard silences. Then we counted the bursts of talk as they occurred in their serial production. Procedurally speaking, the transformation consisted of taking the conversation in the way it was known by the conversationalist to have proceeded and systematically disregarding, overlooking, “counting out” certain features of it which were known and relevant within one perspective, adopting instead another perspective in which only the frequency with which bursts of talk were delivered were the possible occurrences of that conversation. After that, we were interested in obtaining the likelihood that if one of these persons, for example a patient, talked that the next person to talk would be the patient instead of the psychiatrist. That was the kind of conversation that got produced under this procedure for transforming the sense of one set of events into another set. Example 4. It frequently comes as a surprise to beginning undergraduate students in sociology to hear an instructor talk for the first time in class about armed robbery. If some student was ever held up and is asked to tell what happened, his story is likely to be delivered with considerable detail and to carry an undercurrent of indignation and denunciation of the thief who had frightened him and taken his money under threat of physical injury. If a sociologist were to listen to the student’s story and particularly if that sociologist were Parsons he might be inclined to say that he was going to set up a general definition of

Parsons’ Primer

121

theft, count robbery an instance of it, and so conceive it as to be able to talk about this case as one among a set of comparable ones. He might conceive it like this: speak of theft as an instance of the transfer of rights to control the other person’s use and disposal of commodities in which the transfer occurs in violation of the legitimate order of market arrangements that governs such transactions. For a while at least, the victim’s complaints, the hardships that the loss incurred, the deceit and the loss of dignity, the indignation and so on is being overlooked. The sense of the event would literally have been transformed so that its essential features consisted of a particular type of interpersonal transaction involving the transfer from one person to another of rights to acquire, use, transfer, and dispose of a commodity where this transfer of rights had been accomplished outside of the ways in which from the point of view of a member this type of market transaction “normally” and “naturally” occurs. Students are frequently a little disconcerted to find that a crime can be treated as a type of market transaction. When sociologists refer to this kind of an operation they are likely to talk about it as “viewing facts from a particular frame of reference”, or “structuring a situation” in accordance with sociological interests. The “structuring” consists of processes of selection, discrimination, abstraction, definition, of relating, classifying, and the like. The result is the sociologist’s description of a possible society. But to say that one has “selected” or “discriminated” or “abstracted from” or “defined” or “related to” or “classified” tells us only what has been achieved. It does not tell us how one went about the business of producing the end result. The trick therefore is to specify the “structuring” as a definite set of procedures. The result consists of the end-point of an assembly. This end-point is defined by the procedures for accomplishing it as an assembly. The correctness of the assembly is defined in no other way than by listing the rules of the assembling procedures that are treated by the assembler as correct procedures. If it is true that scientific sociological activities are a recent and new historical development, the truth must consist in sociologist’s peculiar rules of assembling social structures. It consists of an historically new definition of the concept “real society” and “real social structures.” By that very fact the meaning of the real social structures consists of the products of theorizing and investigative activities. Thus its conceptions are operational by reason of the fact that their meanings consist of how they are used. The important question is not whether or not sociological methods are operational but rather what do its operations actually consist of.

122

H. Garfinkel

The analytical philosophers propose that sociological activities can be conceived to consist of procedures whereby a society that is known to its members in a more or less common sense way is idealized in accordance with the rules of idealizing procedure known as “the rules of scientific theorizing” and “the rules of scientific inquiry”. The question is always at hand: how does the theorist idealize the K; not whether or not he does so. Of course one need not start with a society that is known in the manner of common sense. One may begin with a description of a society known to sociological discourse as its accepted body of fact. Like most sociologists, Parsons starts with both of these. Parsons’ entire theory of action as well as the particular parts of it – as the social system is a part of it – consists of a set of procedures for idealizing, abstracting, selecting, relating those features of a social world that is to begin with known in any manner and that has any status of warranted knowledge about the society that sociologists might be expected to subscribe to as matters of fact and of producing a description of the society as a product of the transformation performed upon this K. Call the product (T), of methods that sociologists actually use and actually sanction as correct methods, by the term that sociologists conventionally employ, the “real social structures.” Real social structures are the products of sociological criticism. The exact meaning of the term “real social structures” is obtained by examining directly the characteristics of the theorizing and inquiring procedure that sociologists in fact count as correct ones. In this way we see that the empirical references of the “real social structures” are found by consulting the rules that govern the use by sociologists of a description of the society as warranted grounds for their actual further inferences and inquiries. “Real social structures” consist of descriptions that are governed in their use by sociologists as grounds of further inferences and inquiries (i. e. as fact) by accepted rules of sense transforming procedure. The “real social structures” do not consist of the contents of the descriptions themselves. Instead they consist of those descriptions of the society that are controlled in the use that sociologists may make of them by the rules of procedure that define for those who subscribe to them the conditions under which the use of these statements in the manner of fact is correct. If we are to understand how to use Parsons’ work for research, it is necessary at the outset not only to stress the concept of “real social structures”, but to stress its meaning as we have just assigned it. Such stress is necessary for two reasons: (A) Parsons’ general theory of action and social system consists only and entirely of a set of rules of sense transformation. His “system” consists only and entirely of a method for delineating essential and related sets of ob-

Parsons’ Primer

123

servable features of a society, i. e. the social structures, given any account of the social structures as an initial point of departure, K. The result of applying Parsons’ sense transforming rules yields a possible description of the real social structures. “Possible” and “essential” of course have their meaning within the methods’ terms as well as within the terms of the program, attitude, subject matter, and accepted findings of sociology. Insofar as any of these are different, or are changed from what Parsons has taken them and takes them to be, so too must their product, the real social structures, change as well.

Parsons’ own intellectual biography affords an excellent illustration of these changes. In their rate and scope these changes have been disconcerting, amusing, awesome, and resented.

But if Parsons’ general theory of action consists only of a set of rules of sense transforming procedure, the same is true for any theory of socially organized activity. A sociologist who uses Parsons’ or any other theory accomplishes with any the same product i. e. the real social structures. (B) As we shall soon see there are features of the real social structures that Parsons’ rules of procedure produce that distinguish them from the real social structures that are produced by using alternative theories. This permits different claims about matters of fact between Parsons and others to be compared. The more literal and the more rigorous the alternative interpretative methods are, the more is the researcher provided the chance that an actual research may be used to decide between them. (C) Finally, as we shall see in later chapters, “the system” is astonishing in the extent to which its correct use as a method involves the user in literal and rigorous procedure, and in the empirical proposition that such literal procedure produces. From the preceding we obtained the first rule of procedure which is that we begin not with “sense data” but with a society that is known in any way in which it is known. This includes the society that is known in Parsons’ terms as well. Parsons’ theory consists of a set of rules whereby a society that is initially known in any way whatsoever is progressively transformed into a description of a theoretically possible society. How is Parsons’ theory to be used as a set of sense transforming operations? We have already let the term K refer to any description of social structures that the sociologist chooses to use as a point of departure for his inquiries. Thus the term K can refer to a description that the sociologist finds in the

124

H. Garfinkel

professional journals. K may refer as well however to any knowledge of the society that the sociologist addresses as a problematic state of affairs for the purposes of formulating and conducting his inquiries and whose clarification he takes as the project of his inquiries. Any description whatsoever of social structures is placed in the position K by reason of but only by reason of being addressed by the investigator as a sociologically problematic state of affairs. In accordance with the general procedural rule that the sociologist, as a matter of principle, may call any description of society into doubt as to its warranted character, any description of the society is eligible for this treatment. Thus there is the important consequence for characterizing sociological methods that the sociologist in fact does not undertake his research under the auspices of the assumption that he knows nothing about the matter that he is investigating. Instead the rule consists in this: that whatsoever is known about a societal state of affairs, regardless of who knows it or the manner in which it is known, may be brought under the rules of criticism which consist of the procedures of sociological theorizing and inquiry. The term K stands for the knowledge of the society that the sociologist uses as the object which he addresses as a problematic state of affairs. It consists of something known in any way whatever about the society. There are many distinguishable uses to which Parsons’ writings may be put as procedures for criticizing K. They can be classified as “ad hoc” and “systematic” uses. No invidiousness is intended by the distinction. Rather, the particular contributions to sociology by Parsons consist of the systematic uses. The choice of one use or the other is dictated by the usual considerations of researcher interests and practical circumstances. Parsons refers to a number of matters as “structural components”. Sometimes he refers to them as structural elements; sometimes as structures; sometimes as social structure; sometimes as components; sometimes as “they”. There are five “structural components.” He refers to them by the terms “culture,” “collectivity,” “roles,” “personality,” and “organism.” Later in the book their meanings are explained. For the moment it will suffice only to talk of their use as sense transforming operations performed upon K. Their use in this respect can be represented in Figure 1. K ( )k , Any description of social structures as a problematic point of departure.

Figure 1 ˆ Parsons’ definition of culture or collectivity or role or personality or organism.

! !

T Essential features of possible observables.

Parsons’ Primer

125

One use of Parsons for a researcher consists in reviewing K in light of any one, or any combination of, Parsons’ definitions of culture, collectivity, roles, personality, or organism. The product of such a review is the essential features of a set of possible observables, T. Among users of Parsons’ work this procedure is the one most frequently used. Because no further conditions are imposed upon the consequences of these definitions beyond the fact of their use to conceive the materials of K, and because the theorist is free to elect accompanying contexts as they suit the purposes of his research this use may be appropriately called “ad hoc Parsons.” When definitions are considered procedurally, they consist of a set of rules of relevance in terms of which the original objects are reconstructed as another set of possible events. It is necessary to think of the process of applying a definition as a process of reconstruction rather than as one of “abstracting from” or “selecting from” or “coding” certain features of the “empirical world” and disregarding “the rest.” Thus, in Example three above, the conversational exchange between psychiatrist and patient was reconstructed as a set of possible serial bursts of utterances. The actual empirical occurrence of the possible events provided for in the definition was then decided by a different set of operations entirely. It is necessary to make these points and to insist upon the reconstructing use of definitions in order to see how Parsons’ theories are related to a domain of empirical events. It would be misleading to say that we played back the recorded conversation in light of our definition of the conversational exchange as serial bursts of utterances while “disregarding” what the persons were talking about, even though it is true that the clear meaning of the situation defined as serial bursts of utterances permits us to say specifically what was disregarded. To say that the features of the original recording were disregarded is misleading because many features of that recorded conversation that might later be brought to our attention as features of it but which we have no knowledge of at the time could also be said to have been “disregarded.” Therefore, to characterize the definition procedurally as a device whereby “selective attention” is paid to the phenomenon as it was “actually encountered” is an elliptical way of describing what we are actually doing when a definition is used. Unless we can propose the set of all things that we could have attempted, we have no way of saying in any empirically definite way what it is that a definition selects from or how a user should proceed to select.

Later we shall see that the conception that a definition consists of rules of selection is based upon a conception of the relationship between “fact” and

126

H. Garfinkel

“theory” whereby something called the “concrete situation” is referred to a set of physical stimuli or physical “sense data” and these are accorded privileged status in conceiving the meaning of adequate evidence.

For these reasons we propose that a definition is not correctly described as a rule of selection or as a rule for paying selective attention. It consists instead of a depiction of a set of related possible events which is used in this way: the user transforms the events in K in such a way as to bring their features under the jurisdiction of the definition which is used as a rule of interpretation to portray the essential features of the events in K. That a comparison of K and T reveals a “selection” to have occurred is a consequence of having permitted the definition the status of a procedure in terms of which the user decides whether or not T is an adequate description of K. Obviously T as an empirically adequate description of K in terms of the procedures ˆ is decided, at least in the ideal, by asking whether or not the intended meanings of the events in T correspond to the meanings of actual observation of the events in T in terms of the rules of literal observation. When this correspondence occurs, we say that the sociologist has produced a sociologically correct description of the social structures. That the social structures described in T turn out to differ from the social structures described in K is therefore no surprise. Thus, K is naturally the point of departure for an inquiry that employs Parsons’ definition of the structural components; it is also the point to which the researcher returns to find the events that he is interested in describing. Critics of Parsons who have asserted that his theory takes the researcher out of sight of the world of empirical events are therefore in error. We have obtained as our first result that the researcher may start with any description of a society as an investigatively problematical state of affairs. This description is called K. When ˆ consists of a definition that Parsons furnishes the terms culture, collectivity, roles, personality and organism as structural components, the procedure has as its product the essential features of a set of possible observable occurrences. These structural components describe a possible empirical world of activities. They are one set of procedures that Parsons uses to come to terms with social action as empirical occurrences. The procedures consist of bringing the empirical world under the jurisdiction of these definitions as descriptions of a set of necessarily related possible empirical events. The important point to be taken from these remarks is that these definitions are applied to some description or to some knowledge of the society at hand.

Parsons’ Primer

127

By applying these definitions certain essential features of the phenomena are yielded. These yielded features then remain to be addressed by the question of whether intended and observed features correspond in their meaning. When Parsons’ critics complain that he furnishes no operational definitions they sometimes mean that he has not been concerned with methodological problems of demonstrating that intended and observed appearances are identical in meaning. In this respect Parsons is not distinguishable as a sinner from most sociologists. Even those who observe scrupulous care in investigative procedure, found for example, in the closely reasoned procedures of survey techniques, most frequently assign correspondence by letting the reported observation stand on behalf of the event that they mean to be describing. This problem is general to sociological investigation. It would be out of place to discuss it further at this point. Figure 1 describes one of several ways in which Parsons’ theorizing may be used. It happens to be the most frequent way in which Parsons’ writings have been used. And misused. For example in some recently published instances of such use (Scott, Etzioni)8 possible observables that the definitions produced were treated by these authors as if what the definitions proposed had been observed. The error consists in having treated merely empirical possibilities as established matters of fact. There is nothing about Parsons’ definition that requires a researcher to do any such thing, no more than a researcher who uses Weber’s definition of bureaucratic organizations of activities to conceive observed activities in a government agency is required to say that these activities are bureaucratically organized just because Weber has provided a way of conceiving a government agency as a set of bureaucratically organized activities. In each case the step of observation is required. Descriptions must accord with the rules of adequate evidence. This is a commonplace. Otherwise they are of little more interest than one finds in any case where the sociologist changes a definition into a matter of fact and proceeds to write his account accordingly.

8

[Editor’s Note: This may refer to Scott, F. G. (1959). “Action theory and research in social organization”, American Journal of Sociology 64(4):386–395. While Etzioni published several articles in the late 1950searly 60s’ on organizational analysis that drew on Parsons, it is not possible to determine which of these Garfinkel had in mind.]

128

K ( )k , (Same as Figure 1)

H. Garfinkel Figure 2 ˆ Any of Parsons’ concepts used in a strong fashion. 1. “Parameters” of action-in-general 2. Paradigm for the analysis of social systems: AGIL 3. 3. Structural components 4. Interchange paradigm 5. Sanction paradigm 6. Change:

! !

T An account of “K” which consists of the essential features of possible observables. “Essential” is defined with respect to the possibility of stable social structures as a controlling consideration in accomplishing a strong application of these concepts to K

a. Action as process and product b. Segmentation c. Differentiation d. Structural change

A second use of Parsons, shown in Figure 2, consists of starting with culture, collectivity, roles, personality, and organism in the K position. With this operation already having been performed, one would have at hand either a knowledge of some observed state of affairs or a set of possible states of affairs construed with the use of Parsons’ structural components. To this there is applied a set of rules for conceiving the related character of the possible occurrences construed at K under the definitions of the structural components. For convenience call these “rules of interpretation.”

Parsons occasionally refers to one or some combination of them as his theory of action; others occasionally mean to refer to one or another of them when speaking of “the system,” but both are loose usages and should not be taken seriously.

These rules of interpretation go under the names of (1) the system parameters or, the functional problems of a system; (2) the rule of the nesting series;

129

Parsons’ Primer

(3) the rule of hierarchical order. In the Working Papers9 five additional rules of interpretation are discussed as the “postulates” (4) of inertia and (5) of equilibrium, and the concepts of (6) of acceleration, (7) of system maintenance, and (8) of phase movement. We shall of course say more about these “physical” conceptions later in the book. The terminology is particularly unfortunate for in several cases it has motivated serious misunderstandings, for which Parsons must be held responsible. It is necessary to stress that although Parsons uses terminology from classical mechanics their references are not physical events. This stress is introduced in the hope of forestalling criticisms directed to targets that are not there in the first place. In Economy and Society,10 where it was first worked out in detail, and since then has been greatly elaborated in thus far unpublished writings and which constitute the heart of the systematic use of Parsons’ theory, there is a set of rules that Parsons refers to generally as (9) the set of boundary exchanges. Unlike the structural components these rules do not have the status of a set of observable occurrences. Instead they define the meaning of “correctly conceived relationships among observables.” They thereby generate possible observables, but not by themselves. However, that they do not themselves have the status of a set of observables does not mean that their use is not subject to empirical controls. This will be seen in Chapter Five. To summarize the second use, a set of structural components are used to present an empirically possible world of data which is prefigured via the ways in which the concepts of structural components are used to construe a domain of possible occurrences. The rules of interpretation furnish a set of operations performed upon these available possible observables to yield another depiction of a possible society in terms of its essential features. K ( )k , Corpus of sociological “fact”; “Present state of a problem”; “What is ‘known’ so far”

Figure 3 ˆ (1) Theory of stable actions and objects (= PV Revisited)11 (2) Theory of boundary exchanges (“solvency”)

! !

T Hypotheses that formulate essential features of stable social structures as possible observables.

9 [Editor’s Note: This almost certainly refers to Parsons, T., R. F. Bales, & E. A. Shils (Eds.). (1953). Working papers in the theory of action. Glencoe, IL: Free Press.] 10 [Editor’s Note: Parsons, T., & N. Smelser. (1956). Economy and society. Glencoe, IL. Free Press.] 11 [Editor’s Note: This refers to Parsons, T. (1960). “Pattern variables revisited: A response to Dubin”, American Sociological Review 25(4):467–483. See Chapter 5 for a discussion of the pattern variables.]

130

H. Garfinkel

A third use of Parsons is shown in Figure 3. If we start with any knowledge of society and use the combined “structural components” and “rules of interpretation” as the operator ˆ the product is a set of hypotheses about observable social structures. Such combinations occur in the illustrative set of paradigms shown in Figure 4. K ( )k , Corpus of sociological “fact”

Figure 4 ˆ Three theorems that summarize Parsons’ solution to the problem of social order

! !

T Hypotheses that formulate essential features of stable social structures

The dotted brackets in Figure 3 enclose what Parsons is talking about when he uses the term “the system of action.”12 The “system of action” consists of the set of structural components and the rules for conceiving the related character of the observables that are intended by them. A part of the set of “related structural components” Parsons calls “the social system”. The social system is shown in Figure 5. K Structural components

Figure 5 ˆ Use of interchange paradigm = Parsons’ “structural-functional paradigm”

! !

T Essential features of stable changes of social structures as possible observables: (a) structural differentiation (b) structural change

Figure 6 states that an inquiry would be the relevant operator that would produce actual findings. The step of inquiry in Figure 6 is possible after each of the three uses depicted in Figure 1, 2, and 3. Uses 1, 2, or 3 are in no sense identical with the step in the process of inquiry depicted in Figure 6. Figure 6 describes a next and necessary step in uses 1, 2, and 3 that Parsons rarely discusses. But then the step is not one that is particular to Parsons, nor are the difficulties that are involved in carrying it off.

12 [Editor’s Note: There are no dotted brackets in the original manuscript or any of the figures accompanying it, so it is unclear what Garfinkel is referring to here.]

131

Parsons’ Primer

K Any theory of social organization whatsoever, wherever it is found, howsoever it is encountered (ethnosociological or sociological)

Figure 6 ˆ (1) Theory of action as a method of criticism (2) Theory of social system

! !

T Interpreted theory. “T” is an analysis of ideology when it does not satisfy the conditions of stable social structures.

It cannot be stressed strongly enough that the social system consists only and entirely of a set of rules of theorizing procedure. It is a method for sociological theorists. With its use the theorist decides the logically necessarily related character of a set of possible observable courses of action and their objects, called related social structures. To introduce some measure of clarity into a situation where important issues have been lost in a welter of troubled talk, the social system is the theory. It consists of a method whereby the theorist may collect and arrange his thoughts about a set of merely possible phenomena. The way in which this is done so as to achieve a depiction of the related set of possible phenomena as necessarily related conditions and consequences is the structural-functional feature of the theory. Its structural-functional feature consists in this: that it takes the stable features of the social structures as the condition to which its judgments of necessary inference are referred for their correctness. Parsons’ method differs from those of Malinowski, J. W. Bennett, Melvin Tumin, Kingsley Davis and others, in many essential ways, but in this respect all are alike. By speaking of the method as “structural-functional,” attention is called to the fact that the method is designed to define certain decisions of necessarily related social structures as correct decisions. Parsons uses one among several alternative structural-functional methods, all of which are alike in that they furnish the theorist some solution to the question of the necessary relationships in the data. While other authors’ methods differ in specific procedure, their methods are alike in this important aim (Malinowski, Marion Levy, Bennet, Tumin Davis). “Structural-functional” means in general, necessarily related conditions and consequences. The tasks of ordering events of conduct to relations of conditions and consequences are different for sociology than they are for the physical sciences. This is so by reason of sociology’s having to come to terms with problems of causal explanations and determinate theorizing given the problematic phenomena of perspectival view, the exercise of choice, motivation, compliance, the person’s knowledge of his own circumstances, and the rest. That is to say, it permits the theorist to solve the problem

132

H. Garfinkel

of saying that it occurs in this way while prohibiting him from saying that it occurs in another way. It solves the problem for the theorist of whether or not what he proposes in an empirical proposition necessarily follows from what he has been proposing or is prepared to propose in the related set of propositions. Put in yet another way it permits him to solve the problem of finding the logically necessary relationship in a set of data. I stress logically necessary. The necessity has nothing whatsoever to do with empirical necessity. Persons do not have to act in the way they do. The necessity is imposed on the sociologist’s inferences if and only insofar as he wishes or seeks to abide by canons of consistency and wants his accounts to be critically assessed in such terms. Thus if the sociologist makes a set of statements about familial activities in the United States, then within these rules of interpretation as rules of correct inference, he is required to say certain things about the set of industrial practices. For example, the nuclear, isolated, conjugal family when interpreted according to Parsons’ method requires of a sociologist using Parsons’ methods to say that a Chinese form of family life is logically incompatible with the set of industrial practices as they occur in Western society. That does not mean that the two practices cannot occur and would not occur. It does mean that the sociologist is on record as saying that if the combination occurs in this logically prohibited way that his statements are to be counted as empirically unwarranted. To summarize: when Parsons’ work is viewed with respect to its use, it consists of a systematically related set of rules of sense transformation. The system in Parsons’ “system” resides in this and only this. These rules are used to transform some initially given description of social structures. The transformation consists of the essential features of phenomena in exactly the same way that a definition of crime might reconstruct a report of robbery as an instance of an illegitimate market transaction. The method consists of applying the definitions of structural components whose features are related by the “rules of interpretation.” These rules of interpretation tell the interpreter how he must relate observable social structures that the structural components compose as the phenomena under review. In that the related social structures are logically necessary relationships, this procedure is called Parsons’ version of structuralfunctional method. Like any structural-functional method it solves for the theorist a very important problem in sociological theorizing: the problem for the observer of wanting to say that some set of social structures are necessarily related. Parsons’ systematic method and his major contributions are incorporated in the notation of his diagrammed paradigms. The tabular notation is an elected convention, nothing more. But the paradigms as a systematically related set

133

Parsons’ Primer

of transformation rules are in no way “elected conventions.” The paradigms are used by the theorist to generate propositions about necessarily related sets of possible social structures. These rules presuppose as controlling conditions something that is held to be empirically the case about stable social structures. As a result there corresponds to each rule a theorem about the conditions under which stable social structures are possible. Five sets of such rules are discriminable on the grounds of the different purposes that are served in the use of each set: (1) There is the set of pattern variables as the rules with which possible courses of action and possible objects are initially conceived. If one thinks of Parsons’ theory as a game, then the rules of the pattern variable combinations of courses of action and objects have the status of the game’s basic rules. (2) A second set of rules are used for assembling courses of action and objects as stable and unstable assemblies. (3) A third set of rules relates the assemblies of (a) courses of action, (b) objects, (c) integrative standards, and (d) exigencies of stable actions and objects as a concerted set of necessarily related social structures. This is done from the point of view of a theorist who seeks to account for the stable features of the concerted set of activities. (4) A fourth set of rules defines the stable social structures as the maintained product of the processes of activities within the set of related social structures and between this set and its environments. (5) A final set of rules are used to define the rules in sets 2, 3, and 4 as a system of rules. Parsons’ theory as a solution to the problem of social order consists in the content and use of these rules. With them Parsons has furnished sociology a theory which provides the specific conditions that a theorist must claim to have satisfied if he is to account for the existence of a stable society. Parsons’ two theorems summarize these conditions. It is time now to consider the problem of social order, the procedurally equivalent concept of structural analysis and the conditions of adequate description of social actions. These topics are discussed in the next two chapters.

CHAPTER III THE PROBLEM OF SOCIAL ORDER Parsons’ systematic theory is one among many different solutions to the problem of order. But whether a theorist is free to elect it or not as it suits his interests is at least by Parsons’ intent another story. Early in his work Parsons set for himself a radical program whose accomplishment could not be satisfied if he either proposed one more election of theory, or if he proposed to soci-

134

H. Garfinkel

ological colleagues that if this particular way of thinking about social action and societies did not suit their interests that they were free to choose another theory that did. The feature of Parsons’ task that made the difference was this: that along with his theory as a solution to the problem of order he attempted to delineate as well the constituent topics of the problem of social order about which theoretical decisions would have to be made by any sociological theorist. This enterprise began with The Structure of Social Action. There he formulated and attempted the answer to the following question: with respect to the subject matter, attitude, methods, and accepted findings of sociology – i. e., within sociology’s program – what were those problems that any theory of social action, Parsons’ theory included, would necessarily represent a solution, some solution to. The contributions to sociological theory made by The Structure of Social Action consisted in its having accomplished two major tasks: (1) It delineated the critical pre-theoretical problem in terms of which the program of sociology as a scientific study of concerted rule governed human action was itself defined. The book formulated the constituent topics of the problem of social order. (2) It proposed a solution to these problems. Both the formulation of and solution to the problem of social order were elaborated and progressively altered in Parsons’ subsequent writings. In short, over the course of his work Parsons has been concerned with providing the program of sociology its definitive features as a scientific discipline. It is with respect to these two tasks that one may speak of the “exhaustive” character and intent of Parsons’ work. There are two meanings of “exhaustive.” First, Parsons’ theoretical conceptions are intended to furnish the sociological theorist with a method of conceiving any set of social actions and their objects as systematically related. Procedurally speaking, his conceptions may be used to do exactly that. That this is so furnishes one way in which it is correct to speak of the “exhaustive” character and intent of the social system and the theory of action. Obviously this does not mean that social actions might not be conceived differently as a set of related social structures. There is a second, and radical, meaning of “exhaustive” in the philosophical sense of the term, radical. This meaning is found in Parsons’ claim to have delineated the formal analytic structures of social action and to have furnished sociological researchers thereby with the constituent and thereby the necessary or required ideas comprising the concept of social action and its products, i. e. the real social structures, within the program of scientific sociological theorizing and inquiry to which practicing sociologists actually subscribe. Among American sociologists the latter contribution has been generally overlooked. Or when seen it has for the most part been misunderstood.

Parsons’ Primer

135

Although the two tasks can be treated separately, they are intimately related. Parsons’ delineation of the pre-theoretical problems and the constituent ideas of the concept of social action and the real social structures furnished Parsons the grounds for justifying his theoretical elections as correct elections. It is the combination of the two that afforded Parsons the position from which in a succession of papers he has appraised prospects and progress in sociological theory, not only for himself but for the work of others. Above all, the combination furnished Parsons a conception of the real society from which point of view he has found in his examination of competing researches and theories the phenomena of partial descriptions and ideological emphasis. On many occasions Parsons has stated that these properties in the work of various other sociologists are not matters to be decided through intellectual debate. Instead their presence is established by describing the features of these alternative theories as empirical features. It should therefore be a matter of considerable interest to researchers that the grounds that Parsons himself employs are matters whose correctness is to be decided by the tests of evidence rather than by subscription to one or another philosophical authority. For our purposes the problem of social order is best introduced by first considering the discovery of culture. In Part I of the Reader: Theories of Society Parsons reviewed what he called the sources of contemporary sociological theory. Parsons began with what he referred to as secular general knowledge of the society, and chose a selection from Machiavelli to illustrate its character. More plainly, perhaps, Parsons started with the phenomenon of the discovery of culture. What is this “discovery of culture?” It consists of the discovery from within the society by its members of the existence of common sense knowledge of social structures and of the treatment by the societal members of this knowledge, and of the procedure that members use for its assembly, accumulation, test, management, transmission and the like as objects of mere theoretical sociological interest.13 It is the discovery from within the society, which is to say the discovery from the perspective of a person concerned with knowledge of his society for the management and mastery of his everyday affairs, that he and others, in a more or less similar and known in common way and as members of the same society, are engaged in an orderly and concerted treatment of social reality. The discovery “from within” of social reality is the heart of the matter. By social reality is meant the knowledge that persons, as members of a society, known to each other 13 A similar formulation appears in Garfinkel, H. (1967). Studies in ethnomethodology, Chapter 3: 76–77. Englewood Cliffs, NJ: Prentice Hall.

136

H. Garfinkel

and knowing of each other in this fashion, have about the ways in which their social environments are organized and operate. This is knowledge that they take for granted, that they use in the manner of “of course” as grounds of their further inferences and actions, with the distributions, uses, tests, and the rest having a legitimate and sanctioned character. It is knowledge of the society, subscription to which, and evidence of use, is taken for granted by the members, as a condition under which they themselves and others, in a more or less identical way, are accorded and retain the rights to manage their own and each other’s affairs without interference. In a word, it consists of the grounds of further inference and action that persons as conditions of the enforceable character of their claims to bona fide membership in the group expect each other to employ and subscribe to. It is a further distinguishing feature of this vast discovery that social reality is conceived and treated as an object of scientific study. Obviously, as an object of interest social reality may be conceived and treated differently. Somewhere in the course of his development, any child upon encountering a frustration, for example, from those who are “the bearers of the tradition” may doubt whether or not the social world as his parents say it is, is in fact that way at all. But the sociologist is to be distinguished from “natural critics”. Although the two have abiding interests in the procedures and categories of “truth”, and even though each has well worked out ways of deciding the presence of fact, conjecture, hypothesis, lies, ignorance, error, prejudice, and the like, the sociologist’s interests are discriminable from those of the “natural critics” in that the sociologist restricts his interest in the “natural critic’s” knowledge to their technical features as social phenomena. A variety of attitudes can be taken with respect to the knowledge of the social world that is brought under question. One can doubt with respect to religious interests, or to interests of furthering personal or class advantages, or to the interests of political policy, or with the hope of living one’s life as a work of art. The distinguishing feature about the sociological attitude is that it elected the vision of a theoretical scientific contemplation, criticism, revision, codification, and clarification of the social world known from within. Together with the anthropologists, sociologists discovered social reality and named it with the term, culture. That discovery is what Parsons is concerned with. Obviously, he is not alone in this concern. The members of every department of sociology are concerned with the identical thing. Buy why? To what end within the sociological enterprise? Sociologists want the real world, the real social structures. Getting it consists of criticizing and reconstructing the members’ culture, while respecting the condition, that the work of reconstruction must be carried on according

Parsons’ Primer

137

to, and that its products be subject to criticisms of, canons of scientific inquiry. Therein is the difference. For it is this condition that transforms the problematic knowledge of his society that any member can have into a problematic object of mere theoretic interest. Insofar as this condition is respected the sociologist addresses his subject matter under the auspices of a very peculiar rule of interpretation. The rule is: actual persons live out their actual lives, have the children that they have, earn the income that they earn, treat each other in the ways that they do, getting married, divorced, engaging in the political activities that they engage in, developing the psychiatric symptoms that they develop all in order and the better to permit sociologist to solve his theoretical problem. From the point of view of theoretical sociological interests, and speaking literally, people live as they do in order to permit the sociologist the better to collect the data to solve his theoretical questions. Not to make life better, or worse, not to gain an advantage, not to improve international relations, not to make persons more loving or hateful but rather, as the euphemism goes, to “understand society a little better.” Actually, the sociological perspective and its accompanying program assume jurisdiction over all social reality with respect to its clarification in terms of scientific procedure. It seeks to say what the social world really is and to say therefore from the standpoint of the sociologically conceived member of the society, what that membership and that society really consist of, how that society is really organized and operates, what the member’s aspirations and fears are really aspirations for and fears of, what his aspirations really amount to and will really have amounted to. When sociological activities are themselves viewed sociologically, sociology in its visions of the real society and the conditions under which communally governed concerted actions are possible, and in the very vastness of the range of human actions that are subject to the jurisdiction of its concepts, can be counted an authentic instance of a universal church. With respect to every practical and ideological interest, sociologists are both consultants and observers. Because scientific sociology is a theoretical practice of this sort there is no reason why sociology should not get along well with any practical discipline and no reason why sociologists and practitioners of the practical and humane arts should not come to like each other instead of quarrelling as they sometimes do, because except as the practical practitioners are themselves objects of sociological interest, the two have nothing whatsoever to do with each other.

138

H. Garfinkel

In his preface to The Theory and Practice of Hell14 Eugen Kogon furnished an illustration of this point. Kogon, a survivor of a German concentration camp, had been trained in economics and sociology. Following the downfall of Nazi Germany, he said, he decided to write in as detached and as accurate a way as he could manage, what had gone on. He wrote one of the few adequate and excellent sociological accounts of life in the German concentration camps. In the preface Kogon expressed misgivings about publishing the book because, he said, to the extent that his account was accurate, future students of concentration camp affairs would be able to learn from his book how to avoid the mistakes that were made first time around so that the second time around the job would be done more effectively. On the other hand, Kogon proposed, the very accuracy of his account might perhaps prevent a second time from occurring since similarly to the extent that his account was accurate he would have provided opponents a knowledge of the conditions under which concentration camps could arise and thrive. Kogon’s view which he treated as an ethical dilemma but only with respect to the choice of publication or not, is sometimes presented in ethically flatter tones in the commonplace truth that no canon of scientific procedure exists that a sociologist can consult to decide what it is that he is required to study, but once having elected the problem he is required, if he is to anticipate that his solution will be treated as an adequate one, to seek the solution within what is called “the discipline.”

This is the sense in which sociologists, in their work as a scientific discipline – or, to use Parsons’ phrase, as an “analytic science of action” – teach each other that the solutions to their problems are to be sought within a respect only for those criteria that define an adequate solution, i. e. the canons of scientific procedure. It was previously pointed out that the work of sense transformation consists of criticisms that sociologists direct to descriptions of the society as these are provided by and known to the members of the society as the arena of their practical everyday interests and activities. Parsons talks of this criticism as “instrumental criticism” which is directed to the clarification and codification of what is known about the “real society,” treated as an object of theoretical interest. Sociology’s theories and methods, as definitions of correct rules of inference, are solutions to four problems that furnish the constituent mean14 [Editor’s Note: Kogon, E. (1950). The theory and practice of hell: The German concentration camps and the system behind them. New York: Farrar, Straus, and Giroux.]

Parsons’ Primer

139

ings of “instrumental criticism.” I shall refer to these problems by the names “sensibility,” “objectivity,” “warrant,” and “causal texture.” 1. The problem of sensibility consists for any sociological theorist in the work attendant to two decisions; first, deciding the nature of the event he has under study, which is to say deciding that he knows what he is talking about. In sociology there are problems of deciding clear and consistent usage of such terms as action, group, meaning, member, class, and the like. In short the first problem is to decide the events-in-the-discipline that he purports to study. The second is to decide the necessary relationships of meaning among these intended events. This means to decide among various rules of inference those that permit the theorist to talk about the necessarily related character of intended events. 2. The second problem, objectivity, also has two parts. First it consists of the work attendant upon the theorist’s decision to use investigative procedures in such a way that their products can be said to hold irrespective of who it is that recommends the product as findings. Any actual procedures, by virtue of actual acceptance, solve this question and thereby actual procedures alone define the meaning of the criterion of the anonymity of the observer. Second, it consists of the work attendant upon the decision to treat findings as holding for Anyman. Again, any actual procedures by reason of actual acceptance solve this question, and thereby themselves explain the meaning of the criterion of communality. 3. The third set of problems are those of warrant. Its tasks consist of those that attend the theorist’s decision as to whether or not a proposition can be correctly used as grounds for further inferences and research. The theorists’ tasks are those of deciding the rules of observational procedure that will be used to define the status of fact.

Rules of procedure that define factual propositions may be called “rules of observation” because they assign to what the proposition proposes its feature of “having been observed.” Nevertheless rules of procedure that are actually followed in deciding whether or not a proposition is to be admitted to the status of a sociological finding may not require of what the proposition proposes that it have been observed, despite the fact that the sense “having been observed” is nevertheless assigned. That this is so may be seen in the frequency with which propositions that accord in meaning, with what has previously been accorded the status of fact by virtue of this meaningful accord, frequently will be assigned factual status. Other discrepancies are common as well. These are

140

H. Garfinkel

discussed at greater length in my paper Common Sense Knowledge of Social Structures and later essays.15

4. The fourth problem, causal texture, consists of the theorist’s tasks of ordering events to relationships of conditions and consequences. In sociological research this problem is a particularly nasty one because the ordering must be accomplished given a disconcerting irascibility that persons exhibit. This irascibility has been encountered in two extremely important sociological phenomena: habitual action, and the exercise of preferences. The problem of causal texture in the fabric of social life consists of the tasks that the investigator must attend in order to decide how to handle the phenomena of habitual action simultaneously with the fact that persons are capable of exercising and altering their own “definitions of a situation” in obstinate disregard of the fact that sociologists’ predictions consist of instructions that persons must comply with if the sociologist is to have succeeded in describing how persons are making the real society for each other that the sociologist seeks to reproduce. Solutions to the problems of sensibility, objectivity, warrant, and causal texture are of interest for sociologists in ways and for reasons that do not occur in the natural sciences. The crux of the difference resides as the fact that any member of a society has some solution to these same problems. As a matter of the sociologists’ empirical concern, it is not a question of whether or not a member has a solution, but what his solution consists of. Just as the sociologist has his conception of truth, falsity, warrant, fact, objectivity, adequate evidence, rules and categories of proof, and the like, so too does the member of the society. The member of the society does not consult the scientist or the philosopher in deciding that he knows what he is talking about, or in deciding that what he proposes to be the case with the social world, is the case. Thereby solutions to the problem of sensibility, warrant, objectivity, and causal texture are of interest to the sociological investigator in two contrasting but related ways. (1) As solutions that members of a society employ, i. e. “socially employed categories and canons of evidence,” because they are of interest as empirical phenomena they must be described and accounted for as phenomena of socially organized activities. (2) As solutions that sociologists use they are of interest as categories and canons of evidence upon which the claims of sociological investigators to have described the real society are based. 15 [Editor’s Note: Garfinkel, H. (1961). “Aspects of common sense knowledge of social structures”. pp. 51–65 in Transactions of the Fourth World Congress of Sociology, Milan and Stresa, 8–15.9.1959, Vol. 4. Louvain: International Sociological Association.]

Parsons’ Primer

141

Thus these four problems are both problematic topics and problematic features of sociological investigation. If one urges that the sociologists’ results are insured if he elects the canons of scientific procedure, the possibility that a society that is known in the manner of everyday interest can be reconstructed by the sociologist and furnished guarantees as products of social science activities loses its sense as the reproducible products of definite procedures. If the election is urged but the procedural distinction between the two types of solutions is abandoned as a topic of inquiry in its own right, the procedural meaning of the “real social structures” becomes unintelligible. It is with respect to this critical problem that Parsons proposes the subject matter of sociology to consist of three domains of sociological phenomena: (1) the phenomena of social organization; (2) the phenomena of social reality, or culture, and (3) the related character of these two sets of phenomena. The problem of social order is found in the third set. 1. The phenomena of social organization consist only and entirely of related social structures. “Social structures” is the key concept. It formulates the general class of observable phenomena that Parsons is talking about whenever he insists, as he does in many places, that the empirical world has always been the point of departure for his work and that he intends his theorizing always to refer back to the empirical world. For Parsons observables consist of social structures. And “organization”, or speaking more literally, “an organization of social activities” consists of a set of related social structures. What is meant by a social structure? Readers of Parsons are well aware that the term has many uses in his writings; that it does an enormous amount of work: that frequently it is used to stand for one or several of its constituent meanings. We shall attend here only to its use by Parsons to formulate the meaning of empirical observables. Speaking generally, the term involves the idea of a uniformity. But because uniformities of action are involved, the idea has several particular features to it. Parsons once introduced the term “structure” to an undergraduate class by drawing a waterfall, sketching in a cliff and a flow of water going over the cliff. He proposed that any of the various ways in which the path of the flow might be described would be referred to as a structure in the sense of the term intended by him. Thus the term structure refers to the properties of the typical, potentially repetitive, uniform character that observables acquire by reason of the observer’s interest and methods of description. Three further properties are meant that are perhaps less conventionally understood. Those additional properties are: (A) The occurrence is not an “isolated” occurrence but is more correctly an occurrence-in-a-context. The hyphenated phrase is critical. Just as it is the case for the linguist that a linguistic

142

H. Garfinkel

event like a phone, a phoneme, a word, or a sentence, is a linguistic-eventin-a-language, or in games a move as an occurrence is a move-in-a-game, so an event of conduct consists of the occurrence of an action-in-a-social-order. The linguist presupposes a language that the subject uses as a set of maxims of conduct within which the utterance acquires its status as an event in that language for the user as the linguist conceives the user and his language. The linguist uses this entire construction in order to recognize the occurrence of a linguistic event. The student of games presupposes the game that a subject treats as maxims of conduct in order to produce the move within that game. In an identical way Parsons presupposes for his subjects a social order which the subject treats as maxims of conduct within which the action has its status as an event in that order. The properties of typicality, potential repetitiveness, and uniformity are assigned to occurrences by reason of the normative regulation of the production of these occurrences. This usage contrasts with usage in the natural sciences where the “role of ideas in action” is specifically disregarded as a feature of an event of action that the theorist is required to consider in order to recognize an occurrence in the first place. (B) The property that depends upon the feature of an action-in-an-order is the property of a standardized occurrence. This property is of interest and necessarily of interest insofar but only insofar as it is empirically correct for the theorist to assume on behalf of the subjects that their actions consist of actions-in-an-order. This property consists therefore of a social standardization. The occurrence acquires this standardization by reason of the subject’s action occurring in compliance with a normative order of activities. (C) Finally, the term structure refers to one further property that events acquire but only as a matter of the investigator’s procedural convenience. This additional property is occasional and consists in using a mean, or a mode, or a typical, or figurative feature of the event to stand for the event’s distributed features, or for its features as they would be described in fine structure. Thus it frequently occurs that for the purposes of empirical description fine descriptions may be deliberately disregarded by Parsons, and instead a model occurrence or some presupposed major though unspecified percentage frequency may be used to refer to this distributed set of frequencies, with lesser frequencies disregarded for the proposes of accomplishing a description. Thus Parsons’ empirical talk may consist of talk about husbands and wives, of occupations, of beliefs in occupational advancement, of an urban population, of a legal order, and the like. In this respect Parsons’ empirical talk is not distinguishable from that of most sociologists. Social structures at the outset and throughout their use refer to typical courses of action, typical persons, typical social relationships, comprising a typical society. For Parsons the study of social organizations of concerted ac-

Parsons’ Primer

143

tion consists of the study of the ways in which the social structures which consist of numbers of sociologically typified persons in sociologically typified territories with typified distributions and typified relationships between them all governed by typified rules are so conceived by the theorist as to permit him to attend their related character. A theory of social organization consists of and is nothing else than a set of rules that a theorist follows in deciding the related character of a set of social structures. In these ways a social structure is an observable. 2. Social reality, or culture. The concept of social reality is more troublesome. The trouble is this. The sociologist has his ways of typifying courses of action between persons, numbers of persons, territories, rules and the like; but the persons that he is studying also have their ways of performing typifying operations. Thus not only are there social structures from the sociologist’s point of view, but also from the sociologist’s point of view there are the social structures from the subjects’ point of view. This second domain of sociological events, constructed of course by the sociological theorist, is conventionally though elliptically called “the society from the actor’s point of view”, or “perceived social structures”. The problems for the sociological investigator of deciding sensibility, warrant, objectivity, and causal texture with respect to this domain of events are conventionally referred to by speaking of the problem from the actor’s point of view. There are numerous specific problems in this regard. Without enumerating them they may be generally referred to by calling attention to the commonplace observation that when the sociologist describes the social structures, the social structures he attends and describes characteristically do not reproduce the features of the environment that the persons, whose environments are being described by the sociologist, themselves attend and describe. This distinction between the social structures described from the sociologist’s point of view and the social structures described from the sociologist’s-point-of-view-of-the-subject’s-point-of-view, is not the same as the distinction that is more conventionally spoken of, in which provision is made for the society as persons see it on the one hand, and the society as the sociologists see it on the other hand. In this latter view the sociologist’s society is treated as the real society while the subjects of that society are somehow enmeshed in various mixtures of prejudices and realism. Parsons has proposed a radical reconstruction of the relationship between the sociologist’s point of view, the actor’s point of view, and the conception of the real society. Insofar as the concerted actions of the members of a society show stable features, their environments consist of, and their treatments of these environments are responsive to, those social structures which the sociologist, by using the theory of social systems, provides as the features of their

144

H. Garfinkel

situations and actions, which assemble the set of concerted actions of those of the stable society. In this sense the system’s features consist of those which, detected by the sociologist using Parsons’ social system as rules of interpretation, are the society’s real structures. Thereby, the appropriate distinction is between the sociologist’s conception of an actor treating his environment, which consists of the real social structures seen from within, and the sociologist’s conception of the social structures as assembled products of actors, treating the sociologist’s society “seen from within.” The theorist then describes two sets of phenomena: (A) That set of real social structures which consist of assemblies of concerted actions; (B) That set of real perceived social structures of the environment that the set of real concerted actions confronts the members with. The adequacy of the sociologist’s rules of interpretation that make up the social system consists only and entirely in the fact that there corresponds to each of these rules demonstrable features of the actor’s real perceived environment and that the social structures are the assembled products of their concerted treatment by members who know their society only in the manner of their perceived features. “Real” therefore means the sociologist’s procedures that he treats and subscribes to as the definition of adequate description of social structures and perceived environments. The real social structures consist therefore of descriptions of the society as the product of the sociologist’s theorizing and inquiring activities. They are the products of his transformations of the society known in the first place in any way whatsoever. Properly understood, the real social structures are nothing else than a sociological description of the society. What descriptions of society would contrast with the real social structures? Would these contrasting descriptions appropriately be called “unreal” descriptions or unreal social structures? The appropriate contrast to the real social structures would be social structures ethnosociologically described. Ethnosociologically described social structures consist of common sense knowledge of social structures. Sociologically speaking, common sense knowledge of social structures, “common culture”, refers to the socially sanctioned grounds of inference and action that persons use in their everyday affairs and which they assume that other members of the group use in the same way. Socially-sanctioned-facts-of-life-in-society-that-any-bona-fide-members-ofthe-society-knows depicts such matters as the conduct of family life; market organizations; distributions of honor, competence, responsibility, good will, income, motives, frequency, causes of and remedies for trouble, and the pres-

Parsons’ Primer

145

ence of good and evil purposes behind the apparent workings of things.16 Such socially sanctioned facts of social life consist of descriptions from the point of view of the collectivity member’s interests in the management of his practical affairs. Alfred Schuetz describes such knowledge of socially organized environments of concerted action in his classical works on the attitude of everyday life and “common sense knowledge of social structures.”17 Thus far we have considered one problem that the sociological theorist is confronted with in dealing with the data of social reality: the problem of distinguishing the so-called “sociologist’s” point of view from the “actor’s” point of view. Parsons’ social system discriminates those and relates them under the concept of the real social structures on the one hand and the real perceived environments on the other. The sociological theorist is confronted with the task of achieving an adequate description of the data of social reality. Stated prototypically, the theorist’s problem is illustrated in the multitude of sociological and social psychological researches where an interminable list of perceived “this-and-thats” have been described: perceived parents, perceived friends, perceived income, perceived opinions, perceived organizational hierarchies, and the like. The very enumeration confronts the researcher with the procedural problem of where the line might be drawn, or whether within the theories that inform these procedures a line can or needs to be drawn. In this respect there are two problems that the use of Parsons’ social system furnishes procedural answers to. (1) What kind of a limit is to be placed on an enumeration of properties that a perceived environment can show under any of the devices that are available to sociologists for encountering and describing them? (2) How are the categories of objects and courses of judgment and perception that the subject employs to be made comparable to the categories of objects and courses of judgment and perception that the sociologist wants to handle? Parsons’ theory of a social system solves these problems by making the categories that are used by the sociologist to construct the possible events of the social system definitive of the members’ environments, of their treatments of those environments, and of the social structures which are the assembled products of members’ treatments of their environment. Parsons’ theory of a social system is designed to permit the theorist to say, indeed it is correctly used only insofar as the theorist is required to say, within its terms what features of per16

[Editor’s Note: A similar formulation appears in Garfinkel, H. (1967). Studies in ethnomethodology, Chapter 3:76. Englewood Cliffs, NJ: Prentice-Hall.] 17 [Editor’s Note: A slightly different version of this sentence appears in Garfinkel (1967:76): “Basing our usage upon the work of Alfred Schutz, we shall call such knowledge of socially organized environments of concerted actions ‘commonsense knowledge of social structures.’”]

146

H. Garfinkel

sons’ perceived environments these persons can attend, and how the features of these perceived environments are organized according to the set of categories of the social system that define for the sociological investigator the real society. The social system is designed to delineate real environmental features, real treatments of these real environmental features, and real assembled products of these real treatments of real environmental features. The social system delineates their essential properties. Another name for these essential properties are real properties. To speak of these essential properties as real properties is to delineate their procedural meaning. Their substantive meaning consists in their status as courses of action, perceived environmental features, and assembled products of these courses of treatment of perceived environments as those which the investigator is required to predict will occur if the conditions of the stable features of social structures are to be satisfied. Conversely if these conditions are not satisfied, the investigator is required to predict that these stable features of social structures cannot occur as observables. With the device of the social system Parsons offers his solution to the two problems that were just enumerated. In passing it may be remarked that the notorious “pattern variables” are one set of devices whereby for Parsons’ actors and for Parsons’ environments of objects, perceived environments can no longer have any meaning. There is a definitive class of possible features that objects in perceived environments can have and a definitive class of treatments that these possible features can be afforded.

Although a great amount of criticism has been directed to the pattern variables the question has been neglected as to whether the solution that Parsons undertook that is represented in their use might have been premature.

The pattern variables were introduced in order to permit the theorist to say for the person knowing and treating the society “from within,” what on the one hand the alternative possible objects could display in their motivated features (i. e. their meaning) for him. They propose on the other hand what alternative treatments of these objects the person could “elect” to give them. These “pattern properties” as Parsons sometimes refers to them, have by reason of their use by the theorist two important properties: first, the theorist’s actor – not the actual person for he is not a member of the system – but the theorists’ actor who makes up the membership of the conceived society, can elect to give a priority of relevance to one feature rather than another of the objects in his

Parsons’ Primer

147

environment, and he can elect to treat these features in one motivated fashion or another. For the theorist’s actor these elections are elections from a set of dichotomous choices. The actor cannot make the election of relevance or, as Parsons says “give the accent of primacy” to both alternatives simultaneously if the theorist is to propose a stable course of treatment of the objects. This much of the discussion of the pattern variables is introduced to explicate the above point. Their use for the purposes of assembling stable courses of action and stable social structures is discussed in detail in Chapter V which furnishes an illustrative analysis in these terms. 3. The problem of social order. Thus far we have discussed social organization and social reality. The relationship between the two is called the problem of social order. This is the third domain of sociological subject matter that Parsons delineated. The problem of social order may be stated as an empirical problem and as a theoretical one. We begin with the fact, as the sociological investigator begins with fact, that there are stable, persistent, reproducible, more or less standardized, continuing features of social structures that remain invariant to changes of the personnel by whose actions these stable social structures are somehow assembled and of whose actions the social structures consist. At the same time persons populate this society who encounter the various environments that that society confronts them with, who “know” or “see” those environments “from within those environments.” The question is this: how to reconcile the fact that on the one hand there are these stable, persistent products of activities, which are on the other hand the activities of persons who are directing their activities to environments “seen from within.” That is the theoretical problem of social order. The empirical problem of social order consists of all the tasks that a scientific investigator must come to terms with in order to achieve reproducible, credible, clear descriptions of these related social structures, given the fact that he has two large classes of these social structures to describe. The theoretical problem of social order consists of those tasks of conceiving the inhabitants of a theoretically possible society in such a way that the theorist can lay out as a program for actors how they must go about acting in concert with each other with the result that their concerted programmed actions would produce a society having features that conform to the empirical features of the society that the sociologist in the first instance has encountered in his researches and is interested in accounting for. One must remember that a social system is not inhabited by persons. Not only are there no persons in the social system; but the social system does not exist. A social system is inhabited by puppets, dummies, actors who are some-

148

H. Garfinkel

thing else again than persons since they will think the thoughts they think, take the actions they take, encounter environments that they will encounter, produce the assembled structures that their concerted actions assemble, only because the theorist says so and only insofar as the theorist permits them by virtue of a stock of information that he has drawn on to begin with and has distributed among the actors and features of the social system to produce a society for the purposes of accounting for the observed actual social structures. The social system is therefore merely a device that is used to solve the problem of the necessary connections in an actual or possible array of observations, i. e. actual or imagined data. These observed phenomena show the peculiar properties of stability, reproducibility, persistence, and the rest. Along with this there is the fact for the theorist that – again as observables – persons treat environments of the society but their environments do not conform to the environments as the sociologist describes them “from the outside.” The theoretical problem of order consists of the tasks of bringing these two features under a single conception in such a way that the theorist can put a set of imagined actors running around in an imagined territory in concert with a number of imagined friends and enemies, in accordance with specific though imagined rules and in such a fashion as to reproduce the actual empirical features that he has taken as the problematic phenomena at hand. Then, if a reproduction is possible, whatever the theorist might have had to have his actors do constitutes an explanation of how these empirical features occur in the first place. Once the problem of social order is seen in this way there is an easier way to talk of it. The problem of order is another way of talking about structural analysis. The concept of the problem of social order is equivalent in all respects to the procedural concept, structural analysis. Thus the empirical problem of social order consists of the tasks of making a structural analysis. The theoretical problem of social order consists literally and entirely of the very concept of structural analysis itself.18 Any theorist and any inquirer has a solution of sorts to both problems. Procedurally speaking, therefore, any theorist by putting forward his theory presents the inquirer a conception of “correct structural analysis.” It is the investigator’s task in electing one solution as compared with another to learn what its actual rules of procedure consist of and what its consequences consist

18 [Editor’s Note: Garfinkel made margin notes here that seem to read, “Reconstruct these as ethnosociological (or ethnomethodological) ‘cognates.’” Perhaps this represents an early version of his argument that ethnomethodology and formal analysis are “asymmetric alternates”. Cf. Garfinkel, H. (2002). Ethnomethodology’s program: Working out Durkheim’s aphorism. Lanham, MD: Rowman and Littlefield Publishers.]

Parsons’ Primer

149

of for the depiction of a possible society and the ways in which actors by their actions assemble that society. In constructing his own solution Parsons took as the point of departure the formulation and solution to the problem of social order of Thomas Hobbes. Parsons considered Hobbes the first theorist to have formulated explicitly and clearly the problem of social order for sociology, and thereby to have formulated the problematic character of the concept, “correct structural analysis.” In The Structure of Social Action Parsons asserted on intended grounds of evidence that any concept of correct structural analysis that used a Hobbesian conception of adequate description of social structures would reproduce the society, as Hobbes demonstrated, as a war of each against all, and not as the recognizable society. Parsons sought to describe the topics that made up the problem of social order as its constituent tasks that any sociologist would necessarily have to have a solution to insofar as the theorist allowed that the solution, to be counted correct, had to satisfy the condition of sociological seeing as a scientific procedure. This involved two tasks: (1) to describe the constituent topics of the problem of order; (2) to tell how one proceeds when he is “seeing” sociologically. In short, Parsons attempted the radical clarification of the logical foundations of sociology as a scientific “discipline” by attempting to clarify the concept, “adequate description of social action.” The next chapter discusses Hobbes’ statement of the problem of order, enumerates several of these constituent tasks, and formulates the rules of procedure that Parsons came to employ as instructions that a person is following if he is seeing sociologically.

CHAPTER IV “ADEQUATE” DESCRIPTIONS OF SOCIAL STRUCTURES Parsons considered Hobbes the first theorist to have formulated explicitly and clearly the problem of social order for sociology. Hobbes’ statement was as follows. Conceive a man as Any Man. Let this Man be impelled by a variety of “passions.” Let these passions have the properties of hierarchical organization such that the desire for self-preservation is the one to which all other passions are subordinate and reducible. For every passion there is an object to which the passion is directed for satisfaction. These objects are of two kinds. One

150

H. Garfinkel

class consists of those internal to the Man; the second consists of objects that are environmental to him. Internal objects are faded reproductions of external objects. External objects are given to the Man through the immediate evidence of the senses. All of what environmental objects are for the Man are knowable by him through the procedures of science. The desires are consistently “in motion.” They exhibit a “constant plan” such that “the object of a man’s desire is not to enjoy only and for an instant of time but to assure forever the way of his future desire and therefore the voluntary action and inclinations of men tend not only to the procuring but also to the assuring of a contented life.”19 Let the Man be a learner in a world of objects and events which he can know only through sensory engagements. The Man encounters various environmental regularities which Hobbes called antecedents and consequences. These he will store in the form of memories, which memories he will “recon”, i. e. call back into view in imagination as a “trayne” of thought. Hobbes now introduces the very important concept of rationality. One particular form of imagining, or reckoning of the train of thoughts, consists of a train of thoughts of antecedents and consequences, the train being regulated by the degree of constancy of a desire. There is a method of manipulating or reckoning this train which consists of drawing the proper knowledgeable conclusions from a process of “adding and subtracting” antecedents and consequences. The Man’s success in arriving at successful knowledge depends upon proper naming. Much depends also upon the fund of experiences that the Man can draw on. The conclusion of these calculations presents the Man with his design. The design guides the Man in his attempts to satisfy his desires. The instrumentalities, or means, or steps that he can then follow will be appropriate to the fulfillment of his desires, to the extent that they are rational, i. e. that they were arrived at by the use of proper ways of calculating antecedents and consequences. The test of rationality is a pragmatic one. Expectations that are proposed by the plan will be frustrated to the extent that that Man had been in error in attending or sensing or conning20 the flow of events in the sensorily given world. The Man cannot meet the criteria of rationality, as Hobbes proposed them, and find a world that runs counter to his expectations. But the Man does not desire to be rational. The thoughts are to the desires as scouts and spies to range abroad and find a way to the thing desired. All steadiness of the mind’s motions proceeds from this. Reason’s method is 19 [Editor’s Note: Garfinkel is quoting from Chapter 11 of Thomas Hobbes’ (1651) Leviathan. As it is not clear which edition of Hobbes’ book Garfinkel used, we indicate the chapters where relevant quotations appear, but not specific page numbers.] 20 [Editor’s Note: Older English usage for reconning, reckoning, or knowing, from the Scottish.]

Parsons’ Primer

151

that of conceiving a sum total from an addition of parcels, of conceiving a remainder from the subtraction of one sum from another, and wherever there is a place for additions or subtractions there is also a place for reason, and where these have no place, there reason has nothing to do. Reason consists entirely of reckoning the consequences of general aims agreed upon for the marking and signifying of Man’s thoughts. Reason therefore is instrumental to action, and for the Man is a condition for his survival. Whatever is the object of the Man’s desire, the Man will call good. The object of his aversions he will call evil. Good and evil are relative to the person that uses them, there being nothing simply and absolutely so, “nor any common rule of good and evil to be taken from the nature of the objects themselves, but from the person of the Man there is no commonwealth, or, in a commonwealth from the person who is encountered there, or from the arbitrator or judge who, men disagreeing, shall by consent set up and make his sentence the rules there.”21 The present means that a Man has to obtain some future good is called the power of a person. Given the postulate of constant motion, and given two or more Men, then says Hobbes, in the first place “I put for the general inclination of all mankind a perpetual and restless desire for power after power that ceases only in death and in the cause of this it is not that a man always hopes for a more intensive delight than he has already attained to, or that he cannot be content with a moderate power, but because he cannot assure the power and the means to live well which he has at present without the acquisition of more.”22 A society then is an aggregation of men so conceived. A few more characteristics complete the model. First, all persons are objects to each other, and according to the postulate that there is no rule of good to be taken from the object itself, and that “motion” is incessant the other person for any Man is at any point no more than a means for that Man’s end. Second all Men are equal in ability. Thus there is nothing more put into the Man that will require him to defer to the desires of others. That the Man will not defer, Hobbes refers to as the equality of Hope. A social relationship then is one in which two or more Men enter into each other’s regard as elements of each one’s design. There are several ways in which they can enter. First, where two or more actors require the same objects, the possession of which by one entails the loss of it by the other. Second there is no standard to which all men stand in equal awe, each man being different to the others. Hence each must provide for his own security, and an instrument for this is anticipation 21 22

[Editor’s Note: Hobbes, T. (1651). Chapter 6.] [Editor’s Note: Hobbes, T. (1651). Chapter 11.]

152

H. Garfinkel

which leads to the tactics for the removal of potential threat. Finally they can come into a relationship because each man will look to his companion “that he should value him at the same rate that he values himself and upon signs of contempt they are undervaluing naturally endeavors to extract a greater value from the condemner.”23 Inasmuch as rational calculation alone insures successful engagements with a world depicted in this way, engagements by which the objects desired are obtained, the actors can insure that they obtain their ends and ultimately insure their own survival by a process of constantly gaining access to and employing the means that are most efficient to the task. Otherwise, one Man becomes subject to another. The direction in which an issue is settled by this formula is a function of the opposing amounts of power, or the opposing accessibility of resources. The society consists of activities in which each Man contends with each of the others in a constant process of acquisition and denial. There being no awe of death of the other but only a fear of death for oneself, and insofar as the Man acts according to the norms of rationality, force is the most efficient means of survival. There being no agreements except as the Man choose at any point to honor them as it suits their ends, and insofar as the Man acts according to the norms of rationality, fraud is the most rational tactic. The logical consequence of these conditions, says Hobbes, is a state of war in which every man is pitted against every other. Hobbes achieved the stable society by having the Man of his society hand over to a sovereign as a matter of rational self-interest the rights to set the conditions under which survival would be insured for all and to adjudicate disputes. As Parsons pointed out, Hobbes solved the problem of social order with an account of how the institutional order is possible. It is possible because Hobbes’ Men are motivated out of self-interest to establish rules of action binding and enforceable upon all in a more or less equivalent way with the likelihood of their enforcement depending upon the respect for the sovereign. The problem of stable social structures consisted therefore of the problems of controlling force and fraud. The tasks of insuring the stability of respect for agreements were the functions that the sovereign served, i. e. the functions of authority. Major parts of Hobbes’ conception of and solution to the problem of social order are to be found in contemporary sociological theories and researches. For example, the conception of empirically adequate means is prominently employed as a definition of rationality, the difference between passions and 23

[Editor’s Note: Hobbes, T. (1651). Chapter 13.]

Parsons’ Primer

153

thoughts with thoughts ranging about as scouts for the passions’ satisfaction; the war of all against all in which when a legitimate order dissolves the deliberate and calculated use of force and fraud is unleashed in the interest of self-preservation as a conception of the terminal state of society. In The Structure of Social Action Parsons treated Hobbes’ statement of the problem of order as one that formulated with a force that, Parsons proposed, had never adequately been answered by those that conceived the possibility of a stable society in utilitarian terms, the consequences for any conception of a society that attempted to depict a society as consisting of men who acted only in satisfaction of the Hobbesian rule of rationality. Parsons sought on both empirical and theoretical grounds the systematic factors that the Hobbesian formulation had overlooked. The search terminated with the conception of action governed in its course by an order of moral expectations binding and enforceable upon all as constituent features of their common situations of action. This was one product of the book’s arguments. In the course of arriving at it The Structure of Social Action provided a definition of the constituent tasks of the problem of social order and proposed their solutions. The following is a partial enumeration of the many tasks that Parsons discussed: (1) The constituent ideas of the concept of action-ingeneral; (2) The problem of rationality; (3) The problem of the actor as an agency of action; (4) The methodological status of the subjective categories; (5) The role of values and knowledge in action; (6) The methodological distinction and consequences of the distinction between naturalistic causation and social causation; (7) The problematic relationships between “freedom” and “necessity”; (8) The empirical reference of religious ideas; (9) The logical and methodological status of the concepts of motivation, motivated actions, and social control. To give these tasks the attention they deserve, we should have to depart too far from the main purpose of this book. Nevertheless some remarks about them are in order. (1) The constituent ideas of the concept “action-in-general.” In looking for the constituent ideas of the general concept of action Parsons followed a strange course as far as the interests of American sociologists in theory were conceived, at least at the time that the book was written. Parsons’ direction had much more in common with German idealistic philosophy than it had with the pragmatic interests in theory and research that were prevalent in the early and middle thirties when Parsons worked out these problems. Parsons was looking for the necessary conceptions that any sociologist would have to invoke if he were to conceive social action in the first place. He sought define in the general case the necessary conditions of adequate description of social

154

H. Garfinkel

actions. In the terminology of philosophers, Parsons sought the conceptions that were apodictic to the general concept of social action. Hence he sought in the first instance the constituent ideas of the general conception of social action. If he had followed a conventional course he would have elected a definition of the term, social action, as a theorist’s option. Parsons sought instead to (1) clarify the essentially presupposed ideas about action as a fundamental occurrence as it was referred to in established use within the program, the attitude, the methods, and the intended findings of sociology; (2) to disengage these presuppositions; and (3) to formulate them as criteria of an adequate conception of action as an empirical occurrence. The purpose thereby was to furnish the conditions that needed to be satisfied for the adequate description of social actions as empirical phenomena. Parsons proposed the ideas of ends, means, norms, and conditions as constituent ideas. Their intended significance was this: that whatever particular theoretical election a theorist made, he would necessarily have furnished an explication of this set of ideas. A most important intended use to which such a set of constituent ideas was to be put was to furnish the terms in which alternative theories could be compared regarding the question of their adequacy as descriptions of social actions. In the early sections of The Structure of Social Action Parsons used them for exactly this purpose. Parsons attempted to formulate the constituent ideas of ends, means, norms, and conditions into a coherent notion of “an act.” The “action frame” was the device whereby a coherent formulation “an act” was presumably achieved. The action frame consisted of four conceptions that Parsons said theorists must of necessity presuppose in bringing a definition of social action to clarity as an empirically possible event. First, said Parsons, the ideas of ends, means, norms, and conditions were minimal “structural elements” of the action. By minimal was meant not only that the four were exhaustive but that they were not reducible to each other in meaning nor were they reducible to another set of terms without either altering in an essential way their meanings or by failing the requirement of “meaningful”, i. e. adequate, description. Second, Parsons argued, there is implied in the relationships of these elements the purpose governed character of all actions. Third, he said, there is inherent in the notion of motion its sense, or its meaning, or its feature as a “process in time”, i. e. an inherent temporal reference. Finally, he proposed that categories of action as contrasted with behavioral categories are inherently subjective in reference. Few published criticisms of Parsons’ work in America attempted to come to terms with these important statements even though they are vulnerable to criticism.

Parsons’ Primer

155

With a few important exceptions, these features of Parsons’ work have never been criticized. Sorokin proposed that Parsons’ claim as to the essential teleological character of action, formulated in the conception of ends and means, did not meet the test of evidence. He argued further that Parsons’ method for arriving at these constituent ideas was faulty. Alfred Schutz in an unpublished review also criticized Parsons’ method and thereby Parsons’ results. Schutz argued that Parsons had obtained these ideas while failing to suspend his belief in the presuppositions of the natural attitude. (Some brief further comment here on the researches that are possible and the research consequences for Parsons as well as others who have proceeded on similar grounds).24 (2) The problem of rationality. In delineating the conceptions that are binding upon the theorist in conceiving an action in the first instance, Parsons proposed that the theorist had to come to terms with the problem of rationality. The “problem of rationality” can be thought of as consisting of five tasks for the theorist: (A) Discriminating and clarifying the various referents of the term “rationality.” These tasks include stating the behavioral correlates of the various “meanings” of rationality as features of the individual actions, as well as features of the concerted actions, such as those which might comprise “the system;” (B) Deciding on the ground of an examination of experience rather than by an election of theory which of these designata go together; (C) Deciding an allocation of designata between definitional and empirically problematic status; (D) Deciding the grounds for justifying any of the possible allocations that the theorist might finally elect to make; (E) Showing the consequences of alternative sets of decisions for sociological theorizing and investigation. Parsons’ solution to this problem was to elect the rule of adequate means as a definition of rationality in conduct. This rule states that action will be called rational if, in the selection of the means for manipulating or altering a present state of affairs so as to bring it into conformity with a future goal, the selections accord in content and procedure with those that empirical science would propose as adequate both for formulating an accomplishable goal and for deciding whether or not it has been accomplished. This does not mean that Parsons’ actor was a rational actor, although it did mean that Parsons’ actor could not achieve a stable society if he was empirically speaking unrealistic. It meant too that in order to come to terms with the important phenomena of religious ceremony, magic, and ritual Parsons made use of a standard drawn from outside these activities in order to define ceremonial, magical, rit24

[Editor’s Note: Throughout the manuscript Garfinkel added these notes to himself, presumably because he was teaching from the manuscript. We have left them as they were.]

156

H. Garfinkel

ual and similar activities as observable occurrences. The outside standard was furnished by Parsons’ belief that the procedures of “logico-empirical science” guaranteed the descriptions of the world as empirically adequate ones. The rule of empirically adequate means was Pareto’s solution to the problem of rationality as well. Although Parsons has claimed that Weber and Durkheim used identical solutions it is not clear from readings of Weber by other scholars (cf. Bendix,25 Schutz) that this is demonstrably so. The same criticism would be made for Durkheim although the claimants on this score have neither been as numerous nor as vocal as they have been for Weber. It is quite clear that one has to follow this rule in reading Pareto who announced this rule explicitly.

Marion Levy in The Structure of Society26 followed the work of early Parsons and Pareto in elaborating the categories of logical, non-logical, a-logical and illogical, as descriptive categories.

(3) The problem of the actor as the agency of an action. Methodologically this consists of the problem of how to conceive the person as the author of an action. The problem concerns the appropriate use of causal categories to deal with motivated actions and their consequences. The substantive problem consists in the fact that in everyday activities the problem of a person as the agent of an action is handled with the use of notions of and procedures for assigning responsibility and blame. In their everyday use such categories or procedures are the morally equivalent categories of cause. Parsons employed several devices in handling those problems. One was his emphasis on the “voluntaristic” character of social action. A second device was to distinguish between the person and the actor. Parsons never tired of underlining the point that an actor is not a person, and that actors and only actors populate Parsons’ possible society. Sometimes the term actor as Parsons employs it refers to a course of action governed by a set of expectations about which the sociologist may learn by looking to a person. In this case “the person” consists of a methodological device to aid the investigator in locating a field of actual observation. Authorship in this usage means only that there is a place for the observer to look in order to find the actual occurrences that he wishes to study. This usage is different from the procedure of imputing authorship 25

[Editor’s Note: Garfinkel is probably referencing (among other works): Bendix, R. (1960). Max Weber: An intellectual portrait. New York: Double Day.] 26 [Editor’s Note: Levy, M. J. (1952). The structure of society. Princeton, NJ: Princeton University Press.]

Parsons’ Primer

157

in order to decide causal source or assignments of responsibility. Later in his work the referents of the term actor were extended so that the term could refer variously to a person, to a role, to a set of roles, to a collectivity, to a functional sub-system. The common meaning of these different referents consists of an organized course of action directed to an environment of objects. The problem of causal accounting varied with each particular usage. (Disregard. Needs redoing.) (4) The logical status of the subjective categories. The problem consists in the question of whether adequate description of a course of observable action is possible for the sociological theorist if he does not presuppose some reference to the interests of persons, their knowledge of their environment, their treatment of these environments, but without the sociological theorist making reference to or invoking the scientific adequacy of these interests in order for the theorist to decide what the real features of the environments are to which the person whose actions are being described is responsive. In talking about an adequate description of the real features of the person’s environment while putting aside the reference to scientific adequacy, reference is made to the actor and to the questionable theoretical procedure of invoking criteria of scientific adequacy that the actor may be presumed by the theorist to be employing when the theorist describes the actor’s real situation. Parsons handled the problem with his concept of the normative situation of objects, i. e. culture. Parsons’ depiction of a person’s society with the device of the subjective categories consists of a depiction of the common sense features of that society and of the interest in, the knowledge of, and the treatment of these features by persons who are directed to and treat such environments in the management and mastery of their everyday affairs and as members of a society. Another troublesome feature of this problem consists of providing for actors’ interest while retaining the property of this knowledge as knowledge of the society held in common. (Revise and refer to problem of “cultural relativity.”) Earlier we mentioned Parsons’ concept of the real society as the device that he employed to solve this problem (among others). Particular attention should be called to the use of the pattern variables as a construction that permitted the reconciliation of the subjective categories with those that the observer employed. Alfred Schutz, the burden of whose work as a sociologist was directed to the description of the actor’s environments as an organization of constituent common sense features, criticized Parsons’ use of the pattern variables as a “mere” theoretical decision. Schutz argued that for Parsons the conception of a world known in common, “common culture”, was problematic only with respect to the question of what was known in common or whether it was known in common but neglected the question of how a

158

H. Garfinkel

community of understandings was possible in the first instance. This task, said Schutz, is that of clarifying the constitutive expectancies whereby a situation acquired its features “known in common with others.” It was a specific theme throughout Schutz’s work which he handled with the methods of constitutive phenomenology. Schutz pointed out that what Parsons had treated as a point of beginning, Schutz considered the point at which his work would have terminated. (5) The problem of the role of ideas in action. This problem received considerable attention from Parsons in The Structure of Social Action in his discussions of the role of ultimate values. The utilitarian conception of rationality and the conception that actors acted to maximize utilities was treated by Parsons as one type of action rather than as a scheme for conceiving action in the first instance. Parsons directed attention to actions which were presumed by him to be governed in their course by the expected accomplishment of “non-empirical ends.” Intimately related to this problem was the problem of the nature of these “common systems of ultimate value” and courses of common evaluative action. Throughout his work Parsons has maintained that a stable society is neither conceivable as a realistic possibility nor will one be empirically observed unless the concerted actions of which that society consists are normatively controlled actions with a single more or less unified set of standards of evaluation that define the desirable states of the concerted action for the members, to the realization of which their actions are committed, and which govern the actions of all subordinate systems. (6) The problem of social causation versus naturalistic causation. This problem consists of the theorist’s tasks attendant to the decisions both to accept a definition of, as well as to accept an empirical report as, an “adequate explanation.” Given the property of the normative regulation of action, any attempt to relate conditions and consequences in the manner of natural causal sequences has to confront the troublesome fact that persons could, did, do treat theories of action as maxims of conduct. Because this is the case, causal explanations must presuppose the compliance of persons with some theory of their society as a set of rules of conduct. This compliance furnishes one important guarantee of the adequacy of the theorist’s explanation. Hence the obstinate difficulty is encountered by the theorist of discriminating normative from descriptive theories of conduct and of furnishing empirical guarantees that a theory of conduct is a descriptive one. The problem is complicated by reason of physical and biological conditions of action. Following The Structure of Social Action and pending its statement in The Social System Parsons was uncertain as to whether or not to retain the subjective categories as required features for his actors in a social system. In an interim formula-

Parsons’ Primer

159

tion Parsons attempted to handle the problem for the sociological theorist by providing an actor for whom the problem for the sociological theorist of the subjective categories did not arise unless and until the actor had been in an environment long enough to have placed upon a physical and biological environment a cultural overlay. The social environment for Parsons’ actors could be compared in structure to a golf ball with immensely intricate “cultural meanings” built up in successive layers upon a core of physical and biological facts of life. In The Family, Socialization and Interaction Process27 this formulation was abandoned. Actors’ environments of objects now consisted entirely of motivated features of objects (meanings). Such environments are conceived by Parsons to be both the conditions as well as the products of socially controlled actions. Indeed the neonate’s environment from the theorist’s point of view is an environment of meant objects. The socialization of all actors can be read as the processes of socially controlled “growth” of object environments. (7) The logical status of the concept, motivated action. (See the paper on some pre-theoretical decisions28 of Parsons for remarks to this topic. Also consult the notes for this chapter pages 51, 52.) Since their initial formulation in The Structure of Social Action tasks and solutions have been progressively elaborated. Some of these tasks remain unchanged; others were changed considerably and many others have been added. It must be counted one of Parsons’ services to sociological theory that his concern for systematic theory yielded a delineation of these constituent topics of the problem of order. If we recall the equivalence that was made before between the problem of order and the concept of structural analysis, one of Parsons’ contributions consists in this: that by examining the conditions of a stable society conceptually as well as empirically, Parsons delineated the constituent tasks of structural analysis for the purposes of empirical sociological inquiry. Not only did Parsons delineate the constituent tasks of the problem of social order; he provided a progressively developing solution to them. The enumeration of these constituent tasks defined the set of decisions, an open set as it happens, in terms of which the meaning of structural analysis is constituted in the first instance. Solutions to these tasks – any solutions – constitute the meaning of structural analysis, or more literally, “adequate description of the 27

[Editor’s Note: Parsons, T., & R. F. Bales. (1956). Family socialization and interaction process. London: Routledge and Keegan Paul.] 28 [Editor’s Notes: Unpublished manuscript in the Garfinkel Archive. There are plans to publish it in Doug Maynard and John Heritage’s forthcoming collection on Garfinkel: Maynard, D. W. and J. Heritage (eds.), Harold Garfinkel: Praxis, social order, and the ethnomethodology movement. Oxford: Oxford University Press.]

160

H. Garfinkel

real social structures”. The list of topics and their solutions were developed by Parsons with a respect for the following three conditions which together define “the discipline.” These conditions are (1) the program of sociology, by which is meant the corpus of sociological findings taken in terms of the intended events of social action as the fundamental problematic phenomena. Parsons’ concern for the necessary structures of social action were directed to the clarification of this corpus as a corpus of observed occurrences. One might say that Parsons’ concern for the necessary structures of social action provided sociology with a definition of its “fundamental occurrences.” (2) The methods of sociological inquiry; and (3) the sociological attitude, by which is meant the rules of procedure that define the meaning of “seeing sociologically.” According to Parsons, adequate descriptions of related social structures and adequate descriptions of social reality, i. e. culture, were to be achieved while respecting two conditions that defined the meaning of “adequate description.” One condition consisted of the problem of social order. The second has to do with the sociological attitude. I have commented at length on the problem of social order, showed its equivalence to the concept of structural analysis and thereby to the methods of sociology insofar as both are referred to as the discipline within which the solutions of sociological problems were to be counted as adequate ones. The second condition concerns the rules of procedure that define how one proceeds when he is said to be “seeing sociologically.” Two sets of rules need to be discriminated. There are first the rules that define the scientific sociological attitude. These consist of rules to which reference is made when we talk about the unity of science and say that this unity is to be found in its methods. These rules are features of the attitude of scientific theorizing and inquiry in general. They consist for example of the rule of unqualified doubt, i. e. the right that is claimed and accorded an investigator to doubt that the world is at it appears to be. Another such rule is the rule of observation, i. e. that nothing is to be warranted or treated as correct grounds for further inferences in inquiries unless the proposed features, of a society, for example, have actually been observed and that only the outcome of the procedures of observation constitute adequate grounds for assigning to empirical propositions warranted status. The rule that tasks of scientific inquiry are to be governed by the ideal of clarification of the problem at hand, in accordance with the procedures of scientific inquiry, which are the sole grounds for deciding whether or not a solution has been achieved is another. And so on. On these grounds there is no distinguishing, ideally speaking at least, between canons of procedure that are found in sociology from those found in any science. These rules define the attitude of scientific theorizing and inquiry that

Parsons’ Primer

161

hold, ideally speaking, across all domains of inquiry and that explains what is meant when one speaks of them as “theoretical sciences.” An additional set of rules are found that are particular to the theoretical scientific study of human actions and their objects. They may be named as (1) the postulate of relevance; (2) the postulate of subjective interpretation; (3) the postulate of adequacy; (4) the postulate of rationality; (5) the rule of sociological empiricism; (6) the rule of the “action frame;” and (7) the praxeological rule. These rules state how one is to proceed in order to “see” sociologically. They define the characteristics of the specifically theoretical sociological scientific attitude. They may be regarded also as the definitive procedures of the theoretical social science attitude in general. These rules of procedure govern the way in which the sociologist goes about constructing and using his theories in the course of ordering the data of human conduct. They serve to define the sociological perspective in operational terms. A person who, in his observation of human conduct and in his use of language to name what he witnesses, observes and interprets within these rules of procedure is to be counted as observing and theorizing sociologically. Deciding matters of sensibility, objectivity, warrant, and causal texture with respect to these rules defines the term “seeing sociologically.” (1) The postulate of relevance This postulate states that the theorist’s imagined typical person, typical courses of action, typical persons’ environment, typical organization of activities, typical society always implies an accompanying set of types in terms of which the particular type’s meanings are decided and decidable. These types re-present the problematic phenomena for the clarification of which the types were constructed. The theorist’s problem determines the domain of all possible types that he may have constructed as members of a set of types. This holds whether the theorist confronts an empirical or a theoretical problem. The set of related types within which the particular type acquires its definition of “proper usage” has the character of, i. e. consists in the use of a set of rules of interpretation. The enumerated list of these rules defines procedurally the meaning of “correct decisions of interpretation.” The scheme is used by the sociologist to “assign meaning” to witnessed behavioral occurrences. The set is used in the manner of a grammar that the theorist employs in assigning to witnessed actual occurrences their status of sociological events of conduct. They are employed by the theorist to decide correspondence between witnessed actual occurrences and categories of intended occurrences. They consist of rules that the theorist employs in “understanding behavior.” They furnish him with what Max Weber referred to as the “motivationally relevant” grounds of

162

H. Garfinkel

behavior. For the theorist the types and their grammatical use make up the solution to his tasks of clarifying the phenomenon at hand whose clarification is precisely the project of the theorist’s action. It is in the service of this project that the types are constructed. This raises the question of the adequacy of the types as solutions to the tasks of clarification. (2) The postulate of adequacy This rule states that each rule that makes up a grammar, with the use of which correct correspondence of meaning between intended and actual observations is defined, must be so constructed that an action that is described by the type is an experienceable one for the actual persons whose actions the type describes. If the typified action is not experienceable from the point of view of an actual person, it is not adequate from the point of view of the theorist. The adequacy of the type thus consists in the reproducibility of the actual person’s actions with respect to the features described in the type. This is not to say that the actual person is his own best reporter of his own conduct and circumstances. Such would only be the case if the purposes that the actual person engaged in managing his affairs and reporting on them for were identical with those of the scientific sociologist’s. It is the sociologist’s task to state the conditions under which an observed actual course of action is to be counted an instance of the intended course of action that his type formulates. The postulate of adequacy controls the theorists’ statements of these conditions. It controls these statements in the way the rule of adequacy is used. Its use is this: that it provides for the sociological observer a reference to the conditions under which he is to be counted wrong. It furnishes these conditions by imposing the requirement upon a description that its correctness is decided by consulting the experienceable character of the action for an actual person. It is this that makes the observed person’s actions and circumstances from that person’s point of view the final authority in deciding the correctness of the theorist’s interpretation. The typical relevant possibilities of the conceived person in his typical environment are correct if they reproduce the structured relevances of the actual person’s environment to his treatments of that environment and on no other grounds. (3) The postulate of subjective interpretation. This rule states that the theorist in constructing a typical course of action or a typical actor or a typical organization of activities is required to ask what type of actor or what type of action must he construct in order that the social structures, which consist of the assembled products of the conceived actor’s treatments of his environment, may be reproduced as recognizable assemblies that conform to the observed social structures. The rule of subjective interpre-

Parsons’ Primer

163

tation does not say that the observer must “take the role of ”, or must “adopt the position of ”, or that the observer must “put himself in the place of ”, the person whose action he seeks to describe. Instead the rule of subjective interpretation says only this: that the sociological theorist is obligated to put into his theoretical society those features and only those features which that society can exhibit given the type of actor that has been assumed. This rule states that the theorist has to ask what type of individual mind can be constructed and what typical thoughts must be attributed to it to explain the fact in question as the result of this typical mind’s typical activities within a typical and understandable relationship for the actual person whose actions and circumstances the type is intended to reproduce. This formulation may be compared with the way in which the formulation would occur within the theories constructed according to the interests of everyday life. Within the attitude of everyday life the rule would run by contrast that the theorist had to ask what kind of a person is the other person really – or what would he have to be really to explain the fact in question. A moment’s reflection shows that the rule of subjective interpretation within the attitude of scientific theorizing provides the theorist with a type whose sole function it is to allow the theorizer to relate statements describing the conditions under which an event of conduct would occur. It says nothing about what is “really” in the mind of the other persons. The “real” other person, indeed, is for the theorist nothing more than a field of observable behaviors. What is “really going on in his mind” is not known, need not to be known, and indeed can never be known, for the reason that for the theoretical sociologist “the other person’s mind” is not an empirical entity. It consists instead of a set of rules that the sociological theorist employs in “making sense of ”, i. e. in assigning to the status of the events of conduct, whatever are his actual observations. The fact that mind is treated as existing in the world by actual members of a society and that interests of daily life take such an imputation for granted is itself a phenomenon of theoretical sociological interest. This is so because the sociologist attends the properties of “ontological minds” as objects in person’s environments in much the same way that other objects like property rights, tables, gods, ancestors and the rest are of similar interest. As such the sociological theorist would expect such data to show regular relationships with the data describing the structures and operations of the society in which such minds are found. (4) The postulate of rationality. This rule states that typical social action must be constructed in such a way that: (A) the actual person in the observable world would perform the typified act if he had a clear and distinct scientific knowledge of all the elements that bear upon the future course of his action; (B) that he treated the alternative

164

H. Garfinkel

courses as alternatives from which he could choose; and, (C) that he exhibited the constant tendency to choose the most appropriate means for the realization of his ends. This rule is not saying what it might appear at first glance to be saying. It is not saying that unless the actual person in the observable world acts rationally that sociological theories can have nothing to do with him. Nor is it saying that the typified actor is conceived to be acting rationally. The rule is talking about the rationality of the theorist in constructing and using the type and nothing else. The rationality of the actual person’s actions is found not in the type but in the procedure for constructing and using the type that the scientific theorist follows. In this sense the notion that the actual person in the observable world would perform a typified act if he had a clear and distinct scientific knowledge of all the elements relevant to his choice and would have a tendency to choose the most appropriate means for the realization of his ends makes no reference whatsoever to the fact that the actual persons in the actual world act in a manner and according to the conditions and with recourse to whatever means that he sees are sufficient to retain the rational character of his endeavor. But for the scientist such matters are data and not construction. If the scientist desires a scientific picture of the layout of Los Angeles he does not construct the map by consulting taxi drivers. Nor if he wants on the other hand “the-city-of-Los-Angeles-as-an-object-for-action-fortaxi-drivers” can he settle for a cartographically portrayed Los Angeles except perhaps as a standard that he might use as an arbitrary point of comparison. That the actual person might act without regard for the norms of rationality is a matter of great empirical import, and is an empirical question that can never be far from the investigator’s attention. But the most “irrational” of activities is to be treated rationally by the theorist. This is the point. The grammar of the dream, for example, is a different grammar from the grammar mathematicians use to define an adequate proof. The fact that this is so, however, has little if anything to do with the fact that the “irrationalities” of the dream may be rationally understood, and its construction reproduced with a rational grammar appropriate to dream objects as objective occurrences. The type is constructed according to the precept that defines the activities of theorizing and inquiry as rational ones. This is what the rule of rationality is saying. In this sense and only in this sense – i. e. from the standpoint of the sociological theorist employing the type in order to make the observed actions of actual human beings understandable, rationally understandable – one can speak of the “typical actor acting rationally.” In this sense and only in this sense the actual actor would be said to act as he does if he had a clear scientific knowledge of all the elements bearing upon the future course of his actions,

Parsons’ Primer

165

which is to say that the actual actor would act in that manner if he were to treat his circumstances as the social scientist treats them. Only within such a perspective would an actual person faced with a catastrophe be faced with the prospect of selecting between the alternatives of feeling horror or of not feeling horror; of fleeing or not fleeing; of shrieking in panic or not shrieking; and of selecting among the alternative actions to take, thoughts to think, feelings to feel, while selecting on the basis of a knowledge of the consequences for the future state of his howls or the consequences for facilitating the panic of others and the like. It is exactly in this fashion that the theorist treats the experiences of his typical actor and his typical environments and the typical conditions under which typical treatments of typical environments occur – to the extent of course that the scientist adheres to the norm of rationality in his use of the typical society and its typical members. (5) The rule of sociological empiricism. This rule states that insofar as the theorist takes as his object for investigation persons’ actual treatments of their actual environments, the adequacy of the sociologist’s description is furnished by the fact that he permits actual persons to teach him how that actual person is conducting himself in these environments to produce the empirical social structures that the sociologist takes as problematic and wishes an accounting of. Given this, the rule of sociological empiricism proposes that with respect to the sociologist’s treatment of the entire content of the actual person’s activities, and the descriptions that he furnishes of these environments, that the observer abstain from passing any judgment as to the truth, falsity, triviality, importance, worth, valuelessness, of the subject’s assertions and actions and that all of the sociologist’s theorizing and descriptions must be such as to respect the limits imposed by this abstention. This rule presents the following procedures as the ideal which a critic would be allowed to cite as justifiable grounds for accepting or rejecting the adequacy of a description or an interpretation of the events of conduct: (A) With regard to description, the observer’s descriptions must be such that the matter described is reproducible in its entirety by any other observer who establishes the conditions that the descriptions provide as the conditions under which the event is reproducible. Where matters of fact are concerned this rule stands without qualification as the only scientifically legitimate grounds for establishing consensus as to the correctness or not of the decision to admit or reject from the body of sociological knowledge any statement purporting to describe the real social structures. (B) With regard to interpretation, the observer is required so to construct his interpretative rules that taken in themselves they contain no concepts and no statements that depend for their meaning upon the user’s subscription to a sociological theory of existence. This is to say that

166

H. Garfinkel

these rules of interpretation must be so treated as to furnish only a set of rules for relating statements of fact but make no statements of fact themselves. Put otherwise, the theory is allowed to say only what the world means, i. e. how factual statements are to be related to each other, but never what the world consists of in fact. Such statements define the meaning of “correct inference;” they are not themselves statements of fact. (6) The rules of the action frame. As an object in an investigation the actual person treats objects around him in terms of the features they present to him. Thus he is said to entertain various expectations about his environment; he is said to operate with various presuppositions, he is said to employ theories that are relevant to and govern the course of his treatments of these environments in much the same way that the sociologist’s theories are instrumentally employed for the purposes of ordering the events of his environment. Thus the actual person’s theories are as much objects for empirical description for the theorist as are any other activities that the actual person may exhibit. It is thus that an important part of the theorist’s task of constructing a typical, factual world consists of his constructing theories about the typical person’s typical theories. There arises then the distinction between environments depicted from the observer’s point of view and environments depicted from the observer’s-typical-actor’s point of view, a distinction that must be maintained in order that matters of data – and all of what the actual person thinks and does including his theories of what he’s doing and thinking is data and only data – and matters dealing with the observer’s attempts to more or less systematically order this data, not be confounded. It is frequently and conventionally understood that the notion of “taking the point of view of the actor” consists for the observer of his somehow “putting himself in the place of the other person” using the procedure then of asking, “What would things look like to me, if I were in his position?” Such a method is sometimes referred to as “the method of sympathetic introspection.” Such a procedure undoubtedly furnishes the theorist with legitimate hypotheses, and for economy in investigation such a device is no doubt useful. But such a procedure will fulfill what is required of it only under the condition that (1) both observer and subject are assumed to be engaged in the activities of scientific theorizing and (2) that the other person is assumed to have the identical knowledge of his circumstances that the observer, as an outsider to it, has, and to act in strict regard and only in strict regard for this depiction of his objectives as the conditions for obtaining them. These conditions are satisfied however only in the case of a person engaged in scientific theorizing.

Parsons’ Primer

167

Thus the rule of the action frame cannot mean and does not mean that the observer must put himself in the place of the other person. It refers instead to the way in which the theorist is committed to construct the typical actor’s environment once he has constructed the typical actor. The rule consists of this: that the observer in assigning features to the actor’s environments must assign them in such a fashion as to abstain from any judgments about these environments except that these judgments are possible within the meaning of the construction that defines the typical actor and except as the procedures for deciding these consequences are employed with an approximation to the canons of scientific demonstration. There is no more subjectivism about such a procedure than there is when the physicist adopts “the point of view of his model” to represent the behavior of falling bodies within the assumption of classical mechanics and according to the rules of scientific procedure. Hence it can be said that just as the physicist is a “stranger” to his objects just so is the sociological theorist a “stranger” to the persons under observation. Just as it is procedurally speaking, obscure to say and to mean in any literal way that the physicist can place himself in the position of a typical atom and see things as a typical atom sees them, so too is it procedurally obscure to say and to mean in any literal way that the social scientist can do this with his typical persons. (7) The “praxeological” rule. As the enterprise of praxeology was described by Kotarbinski29 it consists programmatically speaking of the search for similarities of successful methods in many different domains of activity. Praxeology seeks to formulate statements of methods and to extend their generality seeking as wide a domain of applicability as possible. The technical values of action, i. e. efficiency, efficacy, economy, productivity, coherence, and the like are praxeology’s principal concern. The theory of games and scientific methodology are instances of praxeological disciplines. Insofar as the praxeological rule may be said to constitute a definitive feature of the sociological attitude, the following is being referred to: that whatever the sociologist treats as the determinant of a phenomenon, or whatever he treats as the conditions of an effect, or as a causal account, may be translated point for point without altering any part of the sense of an adequate sociological description, into a series of statements that tell him how the person engaged in a set of activities, if these persons acted with full appreciation for the rules of the empirical adequacy of their actions and its conditions as means (one of the things meant by rational action, in this case of course, is the rational 29

[Editor’s Note: Probably referring to: Kotarbiński, T. (1960). “The concept of action”, The Journal of Philosophy 57(7):215–222.]

168

H. Garfinkel

activities of the investigator), must act in order to produce whatever property of the social structures it is that the sociologist is interested in accounting for. When the sociologist is interested in the given rate of output of an industrial plant, a praxeological description of this output seems to be a fairly familiar way of talking about it. However the identical thing holds for distributions of persons by class or income, for exhibited voting preferences, for the characteristic features of the division of labor, for differential risks by class of illness and death, for the stability of a suicide rate, for the way symptoms cluster into discriminable clinical entities, and so on. This is to say that the sociologist’s accounts of the conditions under which a phenomenon occurs may be translated point for point into the terms of the strategies that persons could follow under the conditions of rational game play, in order to achieve the pay-off represented in the value of the variable under study. This rule for adopting the sociological attitude runs: for the sociologist, any and all properties whatsoever of a set of related activities are to be treated as technical values which the personnel of the system achieve by their actual modes of “play.” This rule then describes all features of social structures as how-to-do-it features. Several applications of the rule can be cited to make clear how it is being used. (1) Merton’s theory of anomie: To multiply the anomic characteristics of systems of action, socialize the parties so that there is uniform respect for all valued goals while routinely and differentially depriving the parties of the legitimate measures for realizing these goals.30 (2) Levy’s rule regarding the disparity of the locus of authority and responsibility: to multiply the instability of a system of activities, deprive persons of the means for controlling the conditions of an outcome, while holding them to account as the causal agents of that outcome.31 (3) From Lindeman’s findings on drug addiction: to increase the rate of drug addiction, teach persons which symptoms of distress are due to the withdrawal of the drugs.32 Contrary to conventional opinion in the matter, the attitude of sociological theorizing does not require that the theorizer employ a systematic terminology. The fact is that one may theorize and see sociologically with terminologies that are not only unsystematic but that are drawn from fields whose jargon is unfamiliar to sociologists. In fact it is impossible to say what theory one must 30

[Editor’s Note: Probably referring to: Merton, R. K. (1957). Social theory and social structure (Rev. ed.). New York: Free Press.] 31 [Editor’s Note: Probably referring to: Levy, M. J. (1952). The structure of society. Princeton, New Jersey: Princeton University Press.] 32 [Editor’s Note: Erich Lindemann was a psychiatrist who worked on integrating drug-based therapy with psychoanalysis. It is not clear which publication(s) Garfinkel is referring to.]

Parsons’ Primer

169

use by way of language in order to be counted as “seeing sociologically.” The essential feature that marks off the sociological view does not consist of the particular theory that is being employed but does consist of the rules within which both the construction and use of the theory proceeds. These rules in effect govern usage; they are rules of usage. They thereby explain the meanings of the terms of the particular terminology. It is this important property of the relationship between meaning and perspective that makes Parsons’ theory an instance of sociological theorizing rather than a plea for the preservation of the status quo, or a secular theological tract, or an exercise illustrating how difficult the English language can be made if only one makes the attempt, or any of an interminable list of possible things that rules of usage would turn it into if one uses any other rules than those that define the sociological perspective.

CHAPTER V AN ILLUSTRATIVE READING OF PARSONS’ REVISED PARADIGMS OF THE COMPONENTS AND PROCESSES OF ACTION SYSTEMS OF RULES (A THEORY) FOR ASSEMBLING STABLE SOCIAL STRUCTURES33 In the Spring of 1960 Robert Dubin submitted an article to the American Sociological Review in which he attempted to explain and “operationalize” Parsons’ paradigmatic accounts of stable courses of concerted actions and their objects. Charles Page, the editor, invited Parsons’ reply with the offer of simultaneous publication. Because Dubin’s account, obviously a serious and friendly one, nevertheless consisted of an elaboration of original and serious misunderstandings, Parsons was somewhat disconcerted. His chagrin was multiplied by the thought that even if there had been nothing to correct in Dubin’s account, it referred to an outdated version of the “pattern variable schema”, although Parsons had not yet published his thoughts about its revision. Parsons used the occasion to publish his revised views, and wrote a most important paper, “Pattern Variables Revisited: a Response to Professor Dubin’s Stimulus”. Both articles were published in the Review in the issue of August, 1960.

33

[Editor’s Note: Chapter V was updated February 8, 1963 with the addition of the first two pages as they appear here and with a new title: “Parsons’ Solution to the Problem of Social Order as a Method for Making Everyday Activities Observable ‘From the Actor’s Point of View’”. A Numbering system for organizing the Chapter was abandoned. See Ed. Note 34.]

170

H. Garfinkel

In this most important article Parsons presented two paradigms. Together they comprise a theory of stable social structures of concerted activities. By presenting these two paradigms as sets of rules for assembling stable social structures I hope that many of the criticisms of his work may be answered, and many of the characteristics of his work that were discussed in previous chapters may be illustrated. Parsons’ paper furnished revised paradigms of the “components and processes of action systems”. The revised paradigms consist of rules for looking at everyday activities so as to make observable their essential features as assembled stable social structures of interaction. In exactly the sense of the discussion in the previous chapter, “The Problem of Social Order and the Concept of Adequate Description of Social Structures”, these rules of interpretive procedure have the methodological status of a solution to the problem of social order. This is to say that the usage of Parsons’ concept, “adequate description of social structures” is explained by the orderly exposition of these rules. These rules are for the use of the sociological investigator to reconstruct observed activities in treatment of, and members’ knowledge of, everyday environments of interaction so as to produce an account of socially ordered activities that consists of procedurally warranted answers to the questions: What is the member of stable everyday social groupings really looking at? What is he really doing? What is he really talking about? What are his real circumstances? What are the real consequences of his actions? Their character in these reports is most easily seen if the rules are read with some actual scene, or activities, or monograph close at hand. When addressed as a method for making everyday activities observable, Parsons’ theory is of interest in several respects. (1) As a procedural definition of the real social structures it can be and is intended by Parsons to be read in a strong way. (2) The method is

By “strong” reading I do not mean that with respect to the matters that rules are talking about that they mean only what they say and say only what they mean. (I know of no case where such a description is satisfied in an actual practice.) I mean by strong reading that rules are read by presupposing a unity of meaning as an overriding project in reading. I mean also that the relevance of a “heterogeneity of cognitive standards, and the relevance of practical vision” (cf Egon Bittner, Radicalism: a Study of the Sociology of Knowledge, mimeographed manuscript, 1962) is suspended in favor of reading what the rules are about for their own unequivocally stable meanings, of reading the rules under the auspices of the reader’s aim that an internally consistent inter-

Parsons’ Primer

171

pretation is possible. By contrast, one could read them, to cite Bittner again, by using “practical judgment as an instrument of intuition and as a standard to recognize the meaning and in order to judge the conduct” under examination. A strong reading is one that is undertaken with the aim that an internally consistent reading is possible, that intends to reduce the “horizon” of imaginable alternatives of interpretation to a preferred alternative, that seeks to satisfy this intention in the actual matter that the meanings of the rules are about, and that satisfies this intention in this actual matter as a source of authority and in the fashion of evidence (i. e. Husserl’s “essential intuition”).

intended by Parsons to be read as a principled social scientific version of a way of looking at stable courses of everyday activities so as to make observable their essential features with respect to the programmatic problem of social order, while providing for the grasp of and production of these stable courses of everyday activities on the part of the member of the society and from his point of view. In this essay I shall describe Parsons’ method in detail. The presentation consists of seven parts. Part I discusses the “unit act” which consists of a course of action directed to an object. The term actor refers to courses of action; the term object refers to an ensemble of environmental features that are relevant to the courses of their treatment. Part II treats the pattern variables as categorical conceptions with the use of which “possible courses of action,” and “possible objects” are conceived. The remaining parts, III to VII, present rules for assembling and relating stable sets of social structures. Part III provides the set of rules for assembling stable courses of action (orientations), stable objects, stable integrative standards, and stable knowledge of environmental exigencies. Part IV provides the rules for relating these assemblies as a stable set of concerted actions. Part V presents the rules that define the stable social structures as the maintained products of temporal processes. Part VI furnishes the presuppositions with the use of which the sets of rules in part III, IV and V are defined as a system of rules. Part VII discusses the multi-celled tables as a system of notation and provides the syntactical rules of that notation. It will be shown that to each of the assembly rules of Parts III, IV, and V there corresponds a theorem, and to each theorem a set of empirical propositions.34 34

[Editor’s Note: Garfinkel cut this paragraph from both the 1962 and 1963 versions and removed section numbers, but kept section divisions corresponding to Roman Numerals I and II. We kept the paragraph and added back the missing numerals and divisions as described in the paragraph to facilitate the sense of the text].

172

H. Garfinkel

I An Exposition of Parsons’ Method Conceive of a set of actors each of whom attends and treats an environment that includes himself as well as other actors among its objects. For temporary purposes of exposition, and until further notice, think of these actors as persons. Think of treatments as actions-in-their-course, and let them be covert and overt; covert, like judging, perceiving, comparing, classifying, imagining, inferring, wishing, hating, loving, reflecting, choosing; and overt, like running, sitting, buying, selling, etc. Call both sets of actions courses-of action or courses-of-treatment. When they are classified by Parsons as cognitive, cathectic, and evaluative actions Parsons refers to them as orientations to objects. Think of objects as ensembles of related environmental features, i. e. as ensembles of remembered, recollected, expected, anticipated, inferred, lovable, hateful, imagined, judged, desired, valued, chosen, etc. features. Think of them as ensembles of the constituent actual and possible appearances of the environment that the environment displays to an actor under the actual and possible course of the actor’s treatments of those ensembles. When these environmental features are classified, Parsons refers to them as modalities of objects. Courses of treatment are to be thought of always as treatments of objects. Parsons uses the term “unit act” to refer to this presupposition. “Unit” does not mean single or singular. It means irreducible or parametric. “Unit” is a grammatical not an empirical way of talking. Grammatically speaking, the “unit act” specifies the constituent ideas that make up the meaning of the general concept, “an event of conduct.” It furnishes the two sets of terms that a description of an empirical instance of an “event of conduct” must provide for if it is to be counted an adequate description. Courses of treatment can be described without describing objects, but not without presupposing the parameter of objects. Usage is similar to the case where an instance of a sound can be described by furnishing only the values of intensity and duration but not without presupposing that the instance has some frequency. Similarly for the “unit act.” An instance of an action can be described with respect to one parameter or the other but not without presupposing that the action consists of some course of treatments of some ensembles of environmental features. The question is never whether or not, but which ones among the set of possible courses of treatments and the set of possible objects. To each parameter there is a domain of possible determinations that the parameter can assume. With respect to each domain of possibilities, there is a

Parsons’ Primer

173

rule for deciding a possible selection. The pattern variables in the 1960 article are used to define the domain of all possible courses of treatments and all possible objects in an environment. This usage of the pattern variables departs from usage in Parsons’ previous writings where they referred to the intended properties of actions and objects.35 Hence, the pattern variables are used to define the domain of possible values of each parameter. Performance-quality and universalism-particularism define the domain of possible determinations of objects, specificity-diffuseness and neutrality-affectivity define the domain of possible determinations of treatments. Each parameter is accompanied by its set of rules. Thus the set for courses of action defines stable combinations of courses of action; all other combinations are unstable. A second set of rules for objects defines stable combinations of properties of objects; all other combinations are unstable. After a few additional remarks we shall return to this point to see how these conceptions are to be used. The actor’s treatments of an environment are motivated treatments. Motivated treatments of an environment are called orientations. The term orientations includes in its meaning the reference to motivation. The environment’s features are also motivated features. The term object or object modality includes in its meanings the reference to motivation. Objects consist of ensembles of motivated features, or “meanings.” Among the various ways that Parsons uses the term “meanings”, he uses “meanings” as synonymous and interchangeable with motivated features of objects. If the reader will recall that the actor’s actions are directed to objects as the actor “sees them”, i. e. objects in what Koffka called a “behavioral environment” this usage will not appear strange.36 For each actor with respect to other actors, treatments, and environments of objects can be specified as choices – theorist’s choices – among a set of possible treatments (or orientations – occasionally Parsons relaxes his usage and calls them attitudes) and objects. Environments of objects consist of choices of relevance to the actor’s courses of treatments of the possible features that ob35

[Editor’s Note: As Parsons states in his “Reply to Dubin” (1960:467): “The pattern variables first emerged as a conceptual scheme for classifying types of roles in social systems, starting with the distinction between professional and business roles”. His earliest published discussion of role differentiation in terms of institutionalized “patterns”, particularly “functional specificity” and “diffuseness” and “universalism” and “particularism” can be found in Parsons, T. (1939) “The Professions and Social Structure”. Social Forces 17:457–467. In his Reply to Dubin (1960:467), Parsons also reports that pattern variables were “substantially revised” in Parsons, T. and E. A. Shils (1951) Toward a General Theory of Action, where they were generalized beyond role analysis to “analysis of all types of systems of action”.] 36 [Editor’s Note: Probably referring to Koffka, K. (1935). Principles of gestalt psychology. New York: Harcourt, Brace, & Company.]

174

H. Garfinkel

jects can exhibit for him. Exercised choices consist of actions. Thus a course of action is by definition a process of relating one or more acting agents to one or more objects that comprise their environments. Although Parsons frequently speaks of actors being “related” to objects via their treatments of these objects, this usage is somewhat misleading because there are no phenomena that correspond to “an actor” apart from a course of action. The actor consists of a course of actions directed to an environment of objects. Action in its course is the actor in the identical sense that James said that the thought is the thinker.37 Similarly the wish is the wisher, and the course of treatment is the treater. For the moment, the two terms actor and course of action will be permitted to stand. The corrective is pedagogically easier to introduce later in the exposition.38 Say of the actor that he experiences the treatments he affords objects over the course of these treatments. Speak of such structures as “I” or “the actor”. When the theorist uses these structures as a reference point, this usage is referred to by speaking of “Ego” or “Ego’s point of view” or “the point of view of the acting agent.” From Ego’s point of view there are persons as objects. Persons are social objects. Social objects from Ego’s point of view are called “alters” when they refer to other actors, and are called actor-as-object-for-himself or actor-as-object when these objects from Ego’s point of view consist of Ego’s accomplished-courses-of-treatments that he has accorded and can accord objects. The identical terms for speaking of “personality” as a system of action are used to describe the constituent features both of actor and the social objects of his environment, i. e. alters, and actors-as-objects for themselves. Nothing is thereby necessarily entailed about awareness. There is entailed, however, that all actions involve a “consciousness.” “Consciousness” or more correctly “consciousness of ” an object, is identical in meaning with responsiveness to an object, irrespective of awareness, deliberation, and the like. Consciousness means relevance both to value and to motivation of an object’s features regardless of whether the object is actor-as-object or alter-as-object. One parameter of this consciousness is the parameter of evaluation. With respect to the parameter of evaluation there are several assumptions. (1) Assume the evaluated character of courses of actions and objects from the point of view of the acting agents as the theorist conceives them. (2) From Ego’s point of view, if Ego is to treat the environment as one whose features 37 [Editor’s Note: This quote appears at the end of Chapter 10 (“The Consciousness of Self ”) in James, W. (1900). Principles of Psychology, Vol. I. New York: Henry Holt. “If the passing thought be the directly verifiable existent which no school has hitherto doubted it to be, then that thought is itself the thinker, and psychology need not look beyond”.] 38 [Editor’s Note: The corrective Garfinkel refers to may be found in Chapter 6, Section III:204–206.]

Parsons’ Primer

175

are known in common with others, it is to be assumed by the theorist that evaluations cannot be particular to each of the interacting parties. (3) It is assumed that environments whose features from Ego’s point of view are known in common with others, consist of objects whose motivated features consist of relevant features of objects that are evaluated by common standard of evaluation. (5)39 By reason of these common standards, environments, from Ego’s point of view, exhibit the properties of a related set of objectively relevant features. (6) The more are these standards institutionalized, the more does there exist the correspondence from Ego’s point of view between courses of treatments and the behavior of objects under these courses of treatment. Thereby, the more are these standards institutionalized the more does the chance exist from Ego’s point of view that actual and normative features of the environment will correspond, e. g. that his legitimate expectations will be fulfilled. (Expand this to provide for exigencies and to qualify this by providing for “external” considerations to the system of action.) (7) In the case of a sufficiently institutionalized set of evaluative standards there must be conditions that are inherent to the processes whereby one or more acting agents are related to one or more objects of their environments such that only under such conditions will their activities show stable characteristics. Any set of actual activities which exhibit the properties of uniformity, standardization, persistence, continuity, reproducibility – call them stable properties – under exchanges and turnover of the system’s actual personnel may be analyzed as solutions to four problems which are spoken of as “functional problems.” By referring to them as “functional problems” the nature of the solution is indicated as activities that satisfy the conditions of stable effectiveness in the attainment of goals (“goal attainment”); of stable production and allocation of resources, (“adaptation”); of stable concerting of expectations and sanctions and guaranteeing of compliance with agreements (“integration”); of stable “respect” for objective desirable states of activities as solutions to the problems of goal attainment, adaption, and integration (“pattern maintenance”). The four functional problems provide the theorist with the conditions that his description of related actions with stable properties must satisfy if his descriptions are to be counted “adequate descriptions,” i. e. if his descriptions are to satisfy the conditions of “logical exhaustiveness,” i. e. meaningful adequacy with respect to sensibility, objectivity, warrant, and causal texture, with respect to reproducing the empirical social structures in the manner of the processes for their production that the theorist’s actors are following. 39

[Editor’s Note: There is no number 4 in the original manuscript. It is possible that Garfinkel incorrectly numbered his points, but we are not sure. So, we have left the text as it was originally.]

176

H. Garfinkel

According to the praxeological rule any observed actual activities are in effect solutions to the four problems. Any observed actual activities consist of the effective production of social structures regardless of whether the conditions of the stable social structures are satisfied or not. Stable social structures and terminal disordered states are, equivalently, states that the actors achieve by their actual actions. The four problems are devices to aid the theorist in conceiving the observed social structures, stable or not, as the achieved products of steps taken by actors to produce them. Actual activities are related to social structures as products, as steps in their production. The relationship consists in conceiving the observed activities as solutions to these four problems. Actual activities that show stable properties are to be conceived as solutions to the four functional problems. Several conditions must be satisfied if the stable social structures are to be theoretically possible, and thereby to be empirically likely. The remainder of the chapter is concerned with tasks of enumerating and explicating these conditions. Stable properties are possible and thus, empirically speaking, the theorist can expect them to occur if, and only if, the standards of relevance that the actor consults in his “choices” of treatments and in his “choices” of object features are used by the actors according to (a) an institutionalized rule, (b) that assigns a priority of relevance to orientations and object features, (c) and assigns priority of relevance to those orientations and object features which are solutions to the four functional problems. Hence there are to begin with several important conditions under which the theorist must predict unstable features of observed activities. These are: (1) if there is free variability in actors’ “choices” of orientations and object features; (2) if no priority of relevance among courses of treatment and object features is demonstrable; (3) if the priority of relevance in choices of either courses of treatments or object features or both is not institutionalized. Thus, stable social structures are not possible if the priority of relevance is governed exclusively by considerations of instrumental efficacy. The pattern variables are particularly concerned with problems of evaluation and thereby with the tasks of conceiving the conditions under which the systems of activities can be characterized as stable sets. Parsons’ early argument was concerned with the problem of what these evaluative standards were. They were decided by using the pattern variables to furnish directly the properties of such standards. In the revision, Pattern Variables Revisited, the pattern variables are used instead to generate those sets of standards that are capable of being institutionalized as well as the set of those that are not.

177

Parsons’ Primer

Modalities of objects provide for alternative recollected, remembered, anticipated, expected features of objects which the actor addresses as matters of motivated interest. Speaking literally, these features are motivated features of objects. Properly understood, little is lost by speaking of them as “motivated meanings of objects.” Objects consist only and entirely of motivated features or “meanings.” A separate classification is required for motivational orientations to objects. The pattern variables define for the theorist the alternatively possible features that the environment of objects can possibly consist of, just as they define for the theorist the alternatively possible intentions that an actor can possibly address to objects. For the tasks of developing a classification of objects in the actor’s environment, the pattern variables of performance-quality, and universalism-particularism provide the theorist with the constituent or categorical ideas in terms of which the general concept “possible object in the actor’s environment” is defined. The four pattern variable alternatives are not themselves classified features of objects. Instead a classification of possible objects is defined in terms of constituent alternatives which the actor employs as rules of evaluative relevance in deciding whether or not the attended features comprise his real environment.

II Possible Objects The following table shows the grammatical structure of objects as stable assemblies of motivated real features for the theorist’s actor and suggests labels for the elementary set of four stable assemblies. It will be noticed from the following discussion that the labels for these assemblies depend for their meanings on the combinational rules and not the other way around. TABLE I TYPES OF MOTIVATED FEATURES OF OBJECTS Universalism Particularism Performance Objects of Utility Objects of Cathexis Quality

Objects of “Generalized Respect”

Objects of Identification

The four types of relevances of motivated features of objects are to be understood as follows: 1. When any actual features whatsoever of an object are attended under a rule of relevance that assigns priority of relevance to expected use of these

178

H. Garfinkel

features as facilities in the attainment of some goal of the actor, this object is spoken of as an object of utility. The pattern variable characteristics of performance-universalism specify the “choices” of possible significations of the object’s actual features in this way: primacy of relevance is assigned to features in terms of what these features can do for the actor’s purposes, as contrasted with the alternative relevance of features as qualities that are independent of this teleological intent. Utmost stress must be made for the point that the features that define an object, any object – in the case at hand, an object of utility – are determined only and entirely by a rule that governs the actor’s treatment of these features and not at all by an object’s “substantive” or “intrinsic” features. For example, insofar as any features of an object whatsoever are attended according to their expected use as facilities in the attainment of the actor’s goal, this fact and this fact alone constitutes the object as an object of utility regardless of the fact that the “identical” features might on another occasion be attended differently. 2. When any actual features are attended under a rule that assigns priority of relevance to the establishment of a particular relationship to the object as a goal state of the actor’s treatment of the object, the object is spoken of as an object of cathexis. The pattern variable characterization in terms of performance and particularization may be interpreted to mean that the object’s significant features are “chosen” in terms of what the object can do, but in this case as part of a set of performance features that stand in a particular relation to the actor. Insofar as any actual features of an object are attended by the actor with respect to their personality in the step-wise accomplishment of a goal state, the object is spoken of as an object of utility. Insofar as any actual features of an object are attended by an actor with respect to the actual or potential status of the object in a terminal or consummatory relationship to the actor’s course of action, the object is spoken of as an object of cathexis. In alternative terminology, the object’s features are said to have cathectic meanings for the actor. 3. When an object’s features describe for the actor an inclusive system in which the actor seeks membership, i. e. its features have this meaning to him, such an object is referred to as an object of identification. The social collectivity, from the point of view of the actual or potential member, is a typical object of identification, but it is not the only case. A “spiritual body” such as an invisible church or totem are other cases. An actor does not come to be included in an object by cathexis. He does come to be included in it by identification, but he does not thereby necessarily lose his own “separate identity.” In pattern variable terms this relation to objects is particularistic. In the collec-

179

Parsons’ Primer

tivity case the relationship, once it is established, may be said to be ascriptive. Membership status is something one “has” rather than something one “does,” though of course what one does may alter it. 4. When an object’s features are normatively evaluated by the actor independently either of its particularized relation to him, or of its usefulness in the sense of utility, it is spoken of as an object of “generalized respect.” To the actor the object’s features are definable independently of anything the object can “do” for him, and independently of any particular relation to him.

III Possible Orientations The attitudinal pair of pattern variables, specificity-diffuseness and affectivityaffective neutrality, provide the categorical ideas for constructing the set of possible treatments accorded by actors to objects. Motivated treatments are called “orientations.” The classification of orientations to objects in Table 2 matches the classification of features or meanings of objects to actors in Table 1. TABLE 2 TYPES OF MOTIVATED ORIENTATIONS Affective Neutrality Affectivity Functional Specificity Instrumental Utilization Consummatory Needs Functional Diffuseness

Needs for Commitment

Needs for Affiliation

1. By saying that the actor treats the environment of objects as a field for instrumental utilization is meant the actor’s disposition to seek out the potential utility of object features and to seek ways and means to exploit this potential utility. To do so effectively it is necessary to keep affective “attachments” of either the consummatory or the affiliative types, in abeyance, i. e. the attitude should be affectively neutral, and it is necessary to evaluate objects in terms of their functional specificity, in terms of particular abstracted aspects rather than in terms of their diffuse totality. 2. By speaking of the consummatory orientation is meant that the actor seeks “gratification,” i. e. that he seeks states which will terminate (a) goalseeking activity, (b) “motivation” for such activity, and (c) the corresponding motivated features of objects. In pattern variable terms this orientation is specified by relaxing the restraints on “expressiveness” that are imposed in instrumental phases of activity, and “enjoying” the consummatory state. Call

180

H. Garfinkel

this “affectivity”. At the same time the “need” involved is specific to this consummatory situation. It does not consist of a diffusely general state of “satisfaction” with a complex set of relations to an object-world. 3. Needs for affiliation share with consummatory needs the termination of a “search.” Hence the relevance of affectivity is stressed as a categorial criterion. The object’s features cannot be restricted in their relevance to any one specific context, unless the specificity is itself defined as relevance to the affiliation, in which case the system reference has been shifted. Hence affiliative orientations are both affective and functionally diffuse. 4. Finally, there is a type of orientation which is diffuse in the sense of not being dependent on situational specification, but which is affectively neutral in the sense of independence from consummatory or affiliative particularity and immediacy. Call this a need for commitments. Commitment to values as such, independent of their contexts of spelling out in particular contexts, is a prototypical case. Conventional usage, which speaks of the “teleological” nature of action often formulates such action simply as its orientation to the attainment of goals. The relation between classified orientations and objects corrects and elaborates this usage in two ways. First, the conception of a simple one-to-one relation between “interest” in an object and its consummatory significance, for example as a goal object, is not adequate for many of our research interests. All objects if they are significant at all for actors have multiple significations for them. Conversely, all actors if they have any basis of interest in a given object have multiple bases of interest. Second, an object is not called a goal object because of features that are inherent to it. It is spoken of as a goal object by reason of the manner in which action and object are combined. Thus the identical object has many potential features whose describable character depends upon taking the nature of the relationship of action and object specifically into account. Finally, one cannot say without introducing confusion that an actor is simply a “consummation-seeking” entity, if by that is meant that he will treat all objects only as goal objects. True, he seeks goals, but even this much is unclear in meaning without presupposing at the same time that while doing so he also seeks affiliations, commitments and means for the achievement of goals. All objects may potentially exhibit their features to a given actor in any rank order of their modalities. And irrespective of the particular object, all orientations of actors may vary with respect to the place in a rank order of relevance of alternative orientations with respect to any object. Independent variability is imaginable on both sides of the pair. But independent variability

Parsons’ Primer

181

raises the question of how to conceive imagined combinations of independent possibilities. The rule with which this question is decided is as follows.

IV Rules for Relating these Assemblies as a Stable Set of Concerted Actions If actions are to exhibit stable properties, actors cannot choose among orientations and significations according to a rule of their independent occurrence. The appropriate model for their orderly combination is not that of two eightfaced dice. What then is the appropriate model with which to conceive the order that is established between the two sides? This question is answered with the use of the concept of integrative standards. Integrative standards define the conditions that pairs of motivated features and motivated treatments must satisfy if actions are to exhibit stable features. When any actual set of observed activities are analyzed as solutions to the four system problems of stable social structures, the analysis cannot assign equivalent relevance to each of the four conditions of adaptation, goal attainment, integration, and pattern maintenance if the theorist is to predict stable features of the observed social structures. Instead, stability can be stated as a required inference if and only if the solutions are themselves arranged in a hierarchical priority. For example, from the standpoint of the theorist’s actors, the theorist must say that the actor cannot simultaneously assign equivalent value to, and attend in equivalent fashion, the exigencies in the production and allocation of facilities, together with the maintenance of motivation and meaning, if the theorist is to say that the observed activities will exhibit stable features. Instead evaluated priorities of relevance in evaluation stand to each other in relationships of control. Terminologically the evaluated priority of relevances is spoken of as the “phase” in which the system is operating. The phase is designated by the evaluation standards that occupy “top” priority. An example. It has often been remarked that in the large scale occupational settings familiar to our society employees may and do give each other “therapy on the job,” but there are severe difficulties in institutionalizing reciprocal therapeutic demands and satisfactions while fulfilling the normative order of “the days’ work” as an evaluatively equivalent order. Alternative priorities are institutionalized. In other settings of concerted actions other priorities are institutionalized, like pattern maintenance standards in familial settings, integrative standards in settings of psychiatric treatment, and adaptive standards in the labor market.

182

H. Garfinkel

However one must not start by presupposing that a “functional” arrangement of standards corresponds to each concrete course of action. Instead the correct point of departure is the observed phenomenon of stable activities regardless of what else one might be prepared to say about that setting by way of concrete description. The functional paradigm is to be used as, and consists only and entirely of, a method to be used by the investigator in deciding the essential and related set of phenomena of activities that determine the stable characteristics of the observed concerted activities regardless of how the theorist might otherwise choose to conceive them, e. g. as family arrangements, voluntary associations, labor unions, folk societies, primary groups and the rest. When the previously stated pairs of pattern-variable alternatives are treated by the theorist as evaluative standards of interactional states to which the theorist’s actors are said to be committed as standards of evaluative relevance of treatments and object features, their use by the theorist in this regard is tagged by calling these pairs “integrative standards.” To each pair there corresponds an alternative pair. Now Parsons’ actor is not such an unthinking, unremembering, valueknowing, unequivocally committed, cultural game-playing dope, that when he is confronted with the conceived environment that the theorist confronts him with for the theorist’s problems of deciding what the actor will do, that the actor, because he is said to attend features of environmental objects as instances of typical features, thereupon banishes from the world the respects in which the object has simultaneously singular and particular features as well.40 He may know just as I may know of this mailman who delivers mail to my home, the respects in which this mailman is “like a mailman”, even while I know at the same time the details of his particular biography. When the doorbell rings I can wonder if he will say, “I have mail for you but only on the condition that you explain your failure to give me a friendly greeting yesterday. If you do not give me a satisfactory explanation, I shall deliver your mail again only after you have decided to change your ways.” I can wonder about this. But I shall consider it an unlikely interactional sequence, and wrong, and consider it unlikely precisely because it is wrong. If he should treat me like this, I can insist that he deliver my mail whether we like each other or not and expect to be socially supported for insisting that one thing has “nothing to do” with the other. Parsons’ actor is capable of the same. So the theorist must make a “choice” for his theorist’s actor among the alternative pairs of pattern variables with respect to their possible simultaneous 40 [Editor’s Note: Garfinkel maintains that the later Parsons’ actor is not a cultural dope, whereas the actor of other sociological theory and of the early Parsons is. See section VI.]

Parsons’ Primer

183

or variable relevance over a particular interactional sequence. The “integrative standards” are the grounds for the theorist’s choice. If the condition of stable particular sequences of action is to be satisfied there are four and only four rules of choice. These rules are as follows. 1. Stability in the adequacy of adaptive activity is dependent upon the “proper balance” (the meaning of “proper balance” will be explained in a moment) between the actor’s objectivity in his empirical assessment of his situation (universalism) and his “single-mindedness” in attending their relevance to his motivation for the goals he is seeking to achieve (specificity.). 2. Stability of activity directed to the attainment of goals depends upon the “proper balance” between the actor’s active mobilization of facilities (performance) and his mobilization of motivational support for this activity (affectivity). Parsons refers in the latter point to the actor’s “investment” in the particular goal, relative to all other goals that might be relevant. 3. The stability of activity that is primarily integrative in its significance for the entire set of related activities is dependent upon “proper balance” between the place that any particular object feature holds in the total system of his orientations to objects as teleologically significant to him (particularism), and its integration in the plurality of motivated features that are important to him (diffuseness). 4. The stability of commitment to the legitimate orders of orientation requires a “proper balance” between clarity or definiteness in the cognitive definition of the object constituted by the act itself (quality) (actor relating himself to situation) and the “seriousness” of motivational commitment to this definition (neutrality). What is meant by “proper balance”? Among the set of possible choices of orientations of actors to objects is a sub-set of choices which, while made by individual actors, must satisfy the conditions of stability in the choices that the other actors must make if the theorist is to say of the concerted set of choices that the concerted set will show stable features. These choices which are made by individual actors are what the integrative standards “integrate.” Integrative standards are formulae that the theorist employs on behalf of his actors which define the ways in which choices of orientations of actors to chosen objects can be characteristics of the same system of choices that the set of actors employ. The concerted set of choices are said to be governed by integrative standards to which the theorist must say that the actors are committed to comply with, if the theorist is to say that the choices of any one actor fulfill the conditions of stable concerting of choices. That set of actors’ choices that satisfy the conditions of a stable concerting of choices is spoken of as choices in “proper balance”.

184

H. Garfinkel

It hardly needs to be pointed out that “proper balance” involves no more value judgment than does the mathematicians’ notion of a “proper” sub-set of elements.

The following rules define choices of orientations and objects in “proper balance.” Commitment by actors to compliance with these norms of choice is required if the theorist is to say that the conditions of the stable concerting of choices have been satisfied. (1) If a set of treatments of an environment are to show stable features through temporal and circumstantial variations in actual appearances of that environment, then the motivated features of that environment must be categorized universalistically and the actor must be directed to the treatment of universalistic features in terms of specific motivational interests. (Revise and clarify). Alternative combinations are unstable. (2) If activities are to remain stably effective in the successive step-wise accomplishment of steps in goal attainment, the actor must assign relevance to performance features of objects, and his courses of treatment must be directed to the accomplishment of these steps in affective terms. (Revise and clarify.) Alternative combinations are unstable. (3) If activities are to exhibit stable properties in concerting the commitments among actors in their choices of courses of treatments and choices of object features, despite variations that are possible in these respects by virtue of differentiations of potential treatments and interests in objects by collectivity members, object features must be categorized by assigning relevance to particularistic object features, and treatments must assign priority of relevance to diffuse motivations. Alternative combinations are unstable. (4) If the normative orders of desirable states of the system of activities, and desirable measures for the production of such desirable states, is to exhibit stable features, object features must be categorized so that primacy of relevance is assigned in terms of their qualities, and orientations of actors must assign primacy of relevance to the affectively neutral character in choices of treatments. Alternative combinations are unstable. The phrase, “alternative combinations are unstable,” means that any attempts to enforce commitments to compliance with the alternative norms to those that define “correct” choices will anomicize the actor’s environment of objects and multiply the disorganized properties of the social structures.

185

Parsons’ Primer

V Temporal Processes Thus far we have discussed stable assemblies of orientations, objects, and pairings. The task has been identical with defining stable reciprocities of performances and sanctions. The conditions for the assembly of stable reciprocities are satisfied when the actor’s actions are governed by the norms represented in the set of assembly rules. A large question remains: How are these stable reciprocities guaranteed over the course of their occurrence as temporal processes? If Parsons provided nothing more the theorist could use this much of the paradigm to construct stable performances and sanctions if he could assume that actors made the choices they did only because the alternatives were there to be chosen. This would be like saying that a sequence of recognizable moves in chess occurred because the basic rules of play provided the players with a set of legally possible moves, preferred rules of play provided them with ways of defining legal moves to achieve a legal play of the game, and players were committed to the basic rules and preferred rules of play as maxims of conduct. The partial system of action that is described with the use of the rules that define stable orientations, stable objects, and stable matchings of orientations and objects is analogous to a game composed of players whose play is governed by only two sets of considerations: basic rules of play, and rules of preferred play. The game analogy is useful because it permits us to demonstrate why this formulation is inadequate, to show Parsons’ solution to these issues which is found in his concept of “the environmental exigencies,” and to clarify Parsons’ use of this concept. Parsons’ concept of the exigencies is a unique and most important contribution to sociological theory. Like many of his other contributions, this contribution resides in his handling of the “reality” problem. The partial system of actions thus far described is analogous to a game composed of players whose play is governed by only two sets of considerations: the basic rules of play, and the rules of preferred play. In order for the theorist to recognize a move as a move-in-the-game he would have to refer to both sets of rules, basic and preferred, as conditions that a player’s actions would have to satisfy in order for these actions to be recognized by the theorist as a movein-the-game, as well as to be responded to and recognized by the player as a move-in-the-game. For example, from the standpoint of the theorist’s chess player, the displacement of a piece on the board may be an “opening move,” a “knight fork,” a “capture” in-chess by reason of the fact that the theorist thinks of the actor as attending an environment of chess-possible-events, while

186

H. Garfinkel

committed both to comply with the basic and preferred rules of chess play as maxims of his conduct, as well as to the use of these rules as “knowledge” of his real situation. Such an assumption would enable the theorist to treat the conditions of stable production of an actual play of the game as consisting entirely of the set of legal possibilities provided by the basic rules, and of the strategic goals and procedures provided by the rules of preferred play. (The terms “move,” “play,” and “game” follow the usage of Von Neumann and Morgenstern.41 ) With this much accomplished the question for the theorist would remain of how a stable and recognizable play of the game can be guaranteed as a production of two persons playing chess over the temporal course of their elections of moves. As matters stand, the theorist would have to assume that the persons made the choices they did only because the alternative choices were available. The theorist would then be limited in his account to saying that a sequence of recognizable moves “in chess” occurred because basic and preferred rules of play provided players with a set of legally possible moves and a set of effective, economical, aesthetic, conventional, or whatever ways of combining the legal possibilities to achieve a recognizable (i. e. “real” or legal) play of the game. The guarantee that players would produce a play of chess over the temporal course of their elections of moves would then be handled by saying that players were committed to compliance with basic and preferred rules of play as maxims of one another’s conduct. Thus if the preferred rules define effective play one would have players committed to compliance with a model of effective action to govern their elections. If the preferred rules provided for conventional play, one would conceive that players were committed to compliance with models of conventional players to govern the temporal course of their elected moves, and so on. The difficulties that have been attributed to the preference for models of rational play (in communication, machine translation, and betting studies for example) are at least in principle resolvable under this procedure. Nothing is obviously wrong with it. For example although Reinfeld in his book, How To Play Winning Chess, discusses effective procedures for achieving short range goals, he could have written “advice” on how to play so as to satisfy any type of game that a theorist might want a player to observe. Hence one can imagine alternative titles, like How To Play Schlemiel’s Chess or How To Play Aesthetically Satisfying Chess and Never Mind Who Wins, and so on. The difficulties do not reside in the fact that a model of effective action is employed since in 41 [Editor’s Note: Von Neumann, J., & O. Morgenstern. (1944). Theory of games and economic behavior. Princeton, NJ: Princeton University Press.]

187

Parsons’ Primer

principle one could employ a model of ineffective, or foolish, or aesthetic, or conventional, or any combination of such actions. Nor do the difficulties reside in the fact that theorists would have to conceive of players acting under these models as instructions and consulting them to decide the correctness in their choices of a course of play. The difficulties instead consist of this: that an actual player must be thought of as a person who knows the basic and preferential rules whatever they might be, never loses sight of them, never distorts in his recall of them, is concerned only with whether or not he remembers them correctly, and is motivated to act only and entirely under their jurisdiction for all the occasions under which he might consult his model of preferred play to decide his real situation and the actions to take in that real situation. The difficulties reside in the fact that when the production of an observed play of a chess game as an instance of a normatively regulated production of the event “a play of a game of chess” is accounted for (a) by consulting the rules of play and (b) by assuming that persons are committed to these rules as maxims of conduct, i. e. that persons acted like players, the result is a normative instead of a descriptive theory of rule-governed conduct. This may be shown as follows.

VI Presuppositions/Exigencies Various conditions immediately come to mind that would have to be satisfied in order to make this assumption a realistic one, conditions that are “outside” the game in the sense that neither the basic nor the preference rules, no matter what they consisted of, would necessarily not provide for. Consider some examples. While a person awaits his turn in chess he might be “away” on a wool-gathering journey, which may be much more pleasurable than the nasty prospects that confront him on the board. With more pleasant prospects elsewhere, why should he “return to the game”? Or, a player may conceive a modification of the basic rules and envisage his situation as a state in the modified game. Then why not act on it? Hunger pangs, nature’s call, sleepiness, and boredom would bear directly on the likelihood that he would contribute to the production of a play of the game. Should he become convinced that a win could not be accomplished through the temporally extended actual course of play, he could withdraw his interest, upset the board, go into the kitchen and refuse to return. He might promise to make his move, but delay indefinitely in making it. He might invoke some modification of chess and insist that this was the game that he had been playing all along, although it might not have occurred to him to take this line until things started to go badly.

188

H. Garfinkel

It is possible to imagine an interminable array of contingencies all of which would determine the likelihood of the person’s commitment to comply with the basic and preferred rules of play. To say that after all players have agreed to play chess does not solve the problem, since there are many actions that a person could take that would not violate the basic rules of chess and still be a play of the game that would not be recognizable as a play of chess that would be produced if the basic rules were “just understood.” For example a player might insist that his chances were aided by interchanging pawns before he made each move. Even if one asserts in reply that after all players who just understood the basic rules of chess could see that such a maneuver in no way altered the material positions of the pieces, the question would nevertheless remain whether actual instances of players would treat the matter in this way.

Several adults with whom I played chess and interchanged pawns and several other adults with whom I played double solitaire and interchanged piles were startled by this action and refused to continue until I had “explained myself ” and promised not to do this again.

To say that such contingencies are “outside” the “real game” is not satisfactory since such an insistence makes of the basic and preferred rules of chess a normative theory of chess play, whereas a descriptive theory is needed. If the game the actual players are playing is the object to be described, then who else but the players by their actual modes of play is to say that we have to be describing the game of chess that rule books describe? If we say that whatever the players are actually doing is to be described by some set of basic and preference rules, then we can only learn the rules they are playing by, by having the players, by their actual play, teach us what these rules are. If, then, whatever are the rules they are actually playing by is the game, then anything that we cannot handle as part of the game is not “outside” the game; it is outside the inquiry. Either the rules must be specific to what the players are actually doing, or we must frankly acknowledge that we can describe actual play only if the players will follow the instructions of the game that the theorist wants them to follow, in order that the theorist can describe their actions as actions in the game that the theorist knows and is talking about. In such a case, every incongruous event would be “outside” the game; there can be no end to such incongruities; and no satisfactory answer would exist to the problem of choosing between nor-

Parsons’ Primer

189

mative and descriptive theories except to admit that the theory is normative and that is that. (See notes on structural incongruities in inquiry.) These difficulties are particular to the study of rule-governed actions. For the purpose of coming to terms with them so as to permit a descriptive theory of normatively regulated conduct to be constructed Parsons furnishes his version of the exigencies.

We have used the game to aid our presentation of critical issues that arise with respect to the question: how are stable reciprocities guaranteed over the course of their occurrence as temporal processes? Parsons’ formulation is directed to clarifying the character of normatively regulated actions of everyday life, and not of games. Nevertheless as we shall see later we can consult games to clarify the nature of the exigencies for the stable social structures of everyday actions.

The argument was proposed that if nothing more were provided to the theorist than the rules for assembling stable reciprocities of performances and sanctions that have been discussed thus far in the paradigm of social action, stable reciprocities would be constructed if the theorist could assume that actors made the choices they did only because the alternatives of orientations, objects, and matchings were there to be chosen. The realistic character of this assumption depends upon the possibility that actor’s definitions of relevant courses of action and relevant features of objects (as these definitions are contained in the theorist’s assembly rules) are maintained by the actor as distinct from neighboring possibilities. Such an assumption is treated as though its conditions were satisfied when the theorist thinks of an actual player as a person who knows and never loses sight of the definitions of relevance that the assembly rules describe, that the player’s environment does not contain events that these rules do not cover, and that the player acts under their exclusive jurisdiction. Call such a player a game-dope. If a theorist who used Parsons’ system of action was required to use it within such assumptions – several of Parsons’ critics are incorrectly convinced that this is the case – then we should call Parsons’ actor a cultural dope, or better, a version of a cultural dope. Other sociological theorists have different versions though they are alike in this respect. Cultural dopes produce the observed stable social structures by acting out the possibilities that are provided in the legitimate orders of integrative standards. Or, to use the more familiar terminology of cultural emanationism, the observed social structures “express” the values and norms of the society. The imagined

190

H. Garfinkel

members of the possible society accomplish what they accomplish in the way of stable concerted actions by reason of their motivated compliance with the possibilities that the culture “provides.”

There correspond to cultural dopes other versions of game dopes like learning dopes who act out the possibilities provided in their conditioning histories; or personalistic psychological dopes who act out the possibilities provided by the rules of the game that correspond to psychiatrically conceived biographies and “inner dynamics.” (Describe the dopes that satisfy the procedures of X2 , r, Q-techniques, etc.)

If the matching rules are sufficient to define the conditions of stability of the structures of performances and sanctions over the course of the choices that actors must exercise to produce these stable structures, this can only be because actors, by reason of the work that they devote to the establishment and maintenance of definitions of relevant treatments and relevant object-features in the face of conditions that make it possible to go astray, make it possible for the theorist to bring actual empirical features of “rule-governed actions” under the jurisdiction of these matching rules as realistic descriptions of these rule-governed actions. The distinctions of relevance are the work of actors. The conditions that motivate this work Parsons refers to as the environmental exigencies of the system of action. The work that actors direct to the constitution, definition, repair, test, review, maintenance, elaboration, codification, criticism, historicizing, etc. of the relevances, Parsons refers to as the “adaptation” of the system of action to its environing exigencies. The partial paradigm of a system of action thus far described provides three sources of instability of structures of performance and sanctions. These sources consist in the problematic compliance of actors in choices of relevant orientations, relevant object features, and the relevant concerted matchings of orientations and objects. Choices of relevance are depicted in the rules that define stable assemblies of orientations, objects, and concerted matchings. The exigencies furnish a fourth source of instability. With respect to each of these previously named sources there exists a set of conditions that guarantee motivated compliance with normative choices of relevance in orientations, objects, and concerted matchings. There is the possibility with respect to each that these conditions may or may not be satisfied.

Parsons’ Primer

191

The fourth source of instability consists of conditions which uniquely determine the likelihood that commitments to compliance with the normative orders of orientations, objects, and matchings can be guaranteed. These conditions are Parsons’ “exigencies.” Not only must objects and orientations be matched with each other, but all three sets of assemblies, i. e. objects, orientations, and integrative standards, must be adapted to environmental events that are “external” to the actions that are conceived to be related under the systematic set of assembly rules, i. e. “the action system”. Such “external” events, says Parsons, occur according to conditions that are not reducible either to the terms of any one or any combination of structurings of orientations, objects, and integrative standards. Such conditions impose a contingent character upon the system of actions as an identifiable, bounded, and boundary maintaining system. Because such a system is the product of the committed compliance of actors with the norms that define stable choices of orientations, objects, and matchings as a system of controls, the guarantee of the stable social structures consists in the likelihood that compliance with this order as a system of controls can be enforced. The exigencies are one set of conditions that determine the “magnitude” of this likelihood. The researcher’s usage of the concept exigencies is controlled by their two distinctive features. (1) The exigencies consist of conditions that are external to the system of activities that the system of activities must satisfy if it is to retain its identity as a system. (2) To each set of commitments to compliance with a normative order of orientations, objects, and integrative standards, and by reason of and only by reason of these commitments, there corresponds to the system of action its exigencies. How are the stable reciprocities guaranteed over the course of their occurrences as temporal processes, given that the choices of relevance must be made in the face of contingencies that make it possible to go astray? Parsons formulates and answers this question with the aid of the concept of the exigencies, as follows. These contingencies are known to the theorist under the auspices of the theorist’s concept of an order of nature. Sociological theories, in contrast to Parsons’ usage, conventionally refer to these contingencies as the features of the “real society”, or the “real environment”, or as “non-social,” i. e. biological, physical, geographic, genetic conditions of action, or as social and non-social “determinants” or “causes” of actions. In contrast Parsons treats these contingencies as conditions that determine the likelihood that commitments to compliance with a legitimate order can be enforced, and calls such conditions “exigencies.”

192

H. Garfinkel

By the nature of the theorist’s enterprise of accounting for the production of stable social structures by actions undertaken in treatment of environments “known from within”, the theorist must necessarily take these contingencies into deliberate account. The question is not whether or not he does, but which ones does he take into account, with what methods of theorizing and inquiry. Like any scientific knowledge, the theorist’s knowledge of these conditions is provisional. It is subject to the usual qualifications attached to empirical description: “so until further notice.” It is subject as well to qualifications (and reconstructions) that are incurred by the fact that empirical knowledge of exigencies is the product of actual (as contrasted with the ideally portrayed) procedures of sociological inquiry whereby questions of sensibility, objectivity, warrant, and causal texture are actually not supposedly decided. (Insert comment here on ethnomethodology.) With these qualifications understood, the theorist’s knowledge defines the exigencies. They include but are not exhausted by, or exhaustive of, the real social structures that are defined by the theorist’s theory of the society and his methods for deciding the correctness of his descriptions. The reference to the theorist’s knowledge, and this alone, is what Parsons means by speaking of exigencies as “external” to the system of action. Parsons uses the terminology “objects in the system of the action’s environment” to refer to the events in the theorist’s order of nature. When Parsons says that these “objects” may or may not be “included”, or be “internal” to the system of action, he means that the actor’s knowledge of these exigencies may or may not be identical with the sociological theorist’s knowledge. If the two are identical, then either the actor attends his society with the procedures of scientific sociological criticism, or the sociologist attends the society with the aims of producing and using a knowledge of the society in the management of everyday affairs and under the constraints that the attitude of everyday life imposes upon sociological criticism. The two domains of knowledge of the social structures are distinct; they do not shade into each other. Parsons intends the distinction and generally means by the “actor’s” knowledge of the exigencies, common sense knowledge, not theoretical sociological knowledge. (The term actor of course covers the cases of any perspective, including that of theoretical sociology. But when this usage is intended Parsons usually says so. Otherwise “actor” and “sociologist” are intended as contrasting perspectives as is the case the present argument.) The point is important enough to warrant some further comments. To say that the exigencies are “internal” to the system of action means that the actor’s knowledge of the exigencies as sanctionable grounds of the actor’s further inferences and actions, as a member in good standing of the system of action, is identical with the theorist’s knowledge of these exigencies as the

Parsons’ Primer

193

grounds of the theorist’s further inferences and actions. For the theorist their sanctionable use is decided entirely (presumably) on grounds of the technical efficacy of his theorizing and inquiring procedures as solutions to the problem of achieving a description of the social structures that is adequate according to the canons that comprise the program of scientific sociology. The theorist is prepared to say only what knowledge of exigencies is sufficient to account for the production of the stable social structures. If the actor in the system acts in the identical way, then he must treat his situation in the manner of the scientific sociologist. That means he must assign to the alternative courses of his thoughts to think, feelings to feel, object features to attend, loyalties to pay, agreements to form and fulfill, sources of authority to invoke, and the rest, a priority of relevance as objects of sociological theoretical interest. If however the actor “lives in” his society and assembles its features while acting differently than the sociologist does, whose “projects” consist of conceiving and describing the society as a merely empirically possible one, then the actor must be said to act under the auspices of perspectives that contrast with those of sociological investigation. The actor acts in contrasting “statuses” and “roles”. But the exigencies that the theorist knows of, says Parsons, are the conditions that determine the enforceable character of the actor’s commitments to comply with the legitimate order of concerted action. Further, says Parsons, if there is stability in the normatively regulated production of social structures, the exigencies that the theorist knows must be satisfied in the actions that the actor takes regardless of whether or not the actor, while acting under the auspices of his membership in the society, is capable of furnishing the theorist a description of the exigencies that conforms to the exigencies as the theorist would describe them. Parsons speaks of this important effect by referring to the actor’s actions as “solutions” to the exigencies. Although the actor-as-member’s knowledge and interest in the exigencies will not be identical to the theorist’s knowledge and interest, nevertheless the actor-as-member will have a knowledge of these exigencies. This knowledge consists of sanctioned grounds of the actor’s further inferences and further actions. These grounds define for the actor the real world to which his actions are subject and with which he must come to terms while complying with legitimate orders of concerted action that define his membership in a social system. This is a knowledge which he requires of himself and of others as a condition under which he may claim and enforce his own rights and the rights of others to manage everyday affairs without interference. This is a knowledge of the real world of practical activities that is assembled and used within the legitimate orders of concerted actions with others, where the conditions that guarantee committed compliance to these legitimate orders define for the actor

194

H. Garfinkel

the real world of physical, biological, and social events as a world of practical, everyday affairs. Parsons expresses this by saying that although the exigencies are not “internal” to the system of action, “meanings of the exigencies” (to use his terms) are. That is to say, the features that the exigencies have, as these features are portrayed by the actor’s practical knowledge of the real world, are known to the sociological theorist as features of the exigencies that the theorist knows of (i. e. are “internal”), on the basis of the theorist’s rule that a set of exigencies corresponds to every legitimate order to which compliance may be sought. Only those legitimate orders that satisfy the exigencies that the theorist knows can be stably enforced, i. e. to only those orders can commitments to compliance be guaranteed. Thereby, only those legitimate orders can define a real world for actors that satisfy the theorist’s knowledge of the conditions for the production of the real, stable social structures. This does not make the theorist out to be God, however, since this interpretation is to be understood according to the additional rule of method that if the observed social structures are stable ones, then the actor’s actions, because they are conceived by the theorist as solutions to the exigencies, imply that the actor has at least as good a “knowledge” of the exigencies as the theorist has, even though the actor, if he is asked to furnish this knowledge, may be at a loss to do so. The actor “knows” the theorist’s exigencies better than the actor can say; his actions demonstrate it. Then if the theorist cannot account for this stability by the theorist’s knowledge of what actions can satisfy the exigencies as the theorist knows them, then the theorist’s knowledge is at fault since he then must necessarily argue the impossibility of the stable social structures that are nevertheless empirically observed. (Elaborate the converse and related possibilities.) In sum, in the case of observed stable social structures, the actor’s knowledge of the exigencies that the theorist knows, consists of legitimate orders of knowledge at hand. The actor is committed to the use of this knowledge as a condition of his competence; he is continually required to furnish evidences of his competence by its use; and the uses of this knowledge of exigencies, as well as the signs of his uses of it, are evidences for himself and others not only of his grasp of the real world, but that he is engaged in realistic action in the real world. (Comment here on socially defined types of actors, actions, collectivities etc. and attitudes toward them that act in disregard of this knowledge: daring, heroic, childish, fanatic, foolish, irresponsible, enthusiastic, psychopathic. Think this through.)

Parsons’ Primer

195

VII The Adaptation of the System of Action to the Exigencies Parsons’ actors do not merely populate the real world that the theorist describes with the use of the social system, but from the theorist’s point of view, when the system is stable, Parsons’ actors require a knowledge of the real world, and enforce this knowledge upon each other, as a condition of their moral rights to manage their concerted affairs without each other’s interference. This real world is conceived by Parsons with the use of four sets of exigencies: (1) exigencies of adaptation; (2) exigencies of goal attainment; (3) exigencies of compliance with agreements; and (4) exigencies in the recognition of the society as an objective fact of life (which Parsons refers to as exigencies in sources of authority.) These concepts, like so many of Parsons’ concepts, are “relational” ones in Cassirer’s sense and according to his description (See Chapter IV).42 In use they have the feel of concepts in search of their contents. 1. For any stable system of action there will be found a set of events which can be analyzed as conditions that the activities must satisfy for the efficient production and allocation of resources as this is a condition for committed compliance with actor’s choices of stable orientations, objects, and pairings. Such events are called exigencies of adaptation. 2. For any stable system of action there will be found a set of events which can be analyzed as conditions that the activities must satisfy for the effective temporal arrangement of actions to occur in alteration of present states of affairs so as to bring them into conformity with some terminal state of affairs as these are conditions for committed compliance with actors’ choices of stable orientations, stable objects, and stable parings. Such events are called exigencies of goal attainments. 3. For any stable system of action there will be found a set of events which can be analyzed as conditions that activities must satisfy if there is to be motivated compliance with, knowledge of, recognition, interpretation, and enforcement of agreements, as conditions of committed compliance with actor’s choices of stable orientations, stable objects, and stable pairings. Such events are called exigencies of agreements. 4. For any stable system of actions there will be found a set of events which can be analyzed as conditions that activities must satisfy, if here-and-now actual appearances of the social system are to be set into correspondence with the social system as an object, as authoritative evidence of the social system as object, as these are conditions for committed compliance with choices of 42

[Editor’s Note: Cassirer, E. (1953). Substance and Function. New York: Dover.]

196

H. Garfinkel

stable orientations, stable objects, and stable pairings. Such events are called exigencies of real sources of authority. Each of these exigencies furnishes a set of conditions whose variable character is from the theorist’s point of view a source of discrepancies in the actors’ situation between the actor’s legitimate expectancies and the actual fall of events. The actor’s committed compliance to the legitimate order of social structures must depend upon the manner in which the actor “comes to terms” with these exigencies. His “choices” of orientations, objects, and reciprocal matchings occur over the temporal course of choice in which these exigencies as sources of discrepancies between legitimate and actual states of affairs can “lead his choices astray.” Thereby the exigencies motivate the actor’s work in maintaining the normative orders of “choices” as a system of relevances. Parsons refers to this work as the “adaptation” of the system of actions to the exigencies. This work consists of both the processes, as well as the products, of the actor’s “knowledge” of the exigencies. From the theorist’s point of view there are the theorist’s actor’s types of knowledge of these exigencies. Parsons provides for four types of such knowledge. Each of the four types is spoken of as an “adaptation” to the conditions under which committed compliance can be guaranteed, i. e. the exigencies. Since the work of assembling and using this knowledge is directed to exigencies that are “external” to the system, Parsons locates these four types in the adaptation cell of the overall paradigm that defines the components of a system of action. An actor’s “knowledge” of the exigencies is a constituent characteristic both of legitimate orders and actual productions of the social structures that constitute the system of action. These four types of knowledge serve from the standpoint of the members of the system to define that system for them as a real world. Parsons provides four rules that actors must comply with in representing as an order of nature the theorist’s exigencies, if that knowledge is to provide actors with stable grounds of action in the face of discrepancies between their legitimate expectations and the actual fall of events. Parsons uses the pattern variable combinations to explicate the features of the exigencies that each of the four types of knowledge would have to choose as relevant features if that knowledge as grounds of the actor’s treatment of the exigencies is to remain stable, i. e. if that knowledge is to remain enforceable as a legitimate definition of the social system as a real world. These rules are as follows: 1. If the relevance of exigency features to the production and allocation of resources (adaptive relevance) is to be stably represented, features must be represented in terms of what they actually or potentially “do” (performance), and the actor must treat the features without respect for their potentialities

Parsons’ Primer

197

for gratifying the actor (he must “orient the feature with affective neutrality”). Only this combination is enforceable as a stable one if the actor is required to “maximize objective understanding” of the exigencies. Parsons refers to this legitimate order of knowledge as “empirical cognitive symbolization.” 2. If the relevance of exigency features to the step-wise temporal arrangement of steps in the accomplishment of future states of affairs is to be stably represented (goal attainment) features must be represented as projects of action (specificity) and the actor must treat these features as actually or potentially belonging to the system of action as a definitive feature of that system and the actor’s membership in it (particularism). Parsons calls this “the generalization of particularistic meanings to a universalistic level of significance”, and refers to this legitimate knowledge as “expressive symbolization.” 3. If the relevance of exigency features to compliance with knowledge of, recognition, interpretation, and enforcement of agreements is to be stably represented (integration), features must be represented as norms that are “external” to the system in the sense of an objectively given order (quality) and the actor must attend these features “with affectivity”, i. e. it cannot be a matter of indifference to the actor as to whether or not he feels committed to the objectively given character of the order (affectivity). Parsons calls this legitimate order of knowledge “moral-evaluative categorization.” 4. If the relevance of exigency features to the authoritative evidential relationship between witnessed actual appearances of the social system and the social system as the object that is presented through the witnessed appearances (the social system as real source of authority) is to be stably represented, it is necessary to define the feature as one whose properties are independent of any reference to the system of action as an object that includes it (universalism) and to orient the feature with diffuse activated interests (diffuseness) such that this significance is treated as invariant to the continually changing temporal course and circumstances of actual appearances. Parsons calls this legitimate order of knowledge “existential interpretation.” Parsons refers to the first two as the actor’s symbolic “knowledge” of the real world. He refers to the second two as the actor’s categorical “knowledge” of the real world. The first two define the real world in the manner of the conceived appearances it presents; the second two constitute the real world in the manner of its appearances as possibilities “in the first place.” Distinctive logics and methodologies correspond to each of these four types of knowledge. There are several more distinctive features of the exigencies that require comment. Their exposition is perhaps more easily made if we once more turn to games.

198

H. Garfinkel

It is fairly obvious that if an action is to be recognized by both theorist and player as a move-in-a-game it must satisfy the basic and preference rules of play. There is another set of conditions that actions must satisfy if they are to be recognized by both theorist and players as moves-in-a-game. Call these conditions “game-furnished” conditions. Game-furnished conditions consist of events to which players’ moves are subject, which are features of every present state of a game although, in the sense that the basic and preference rules make no provision for them, they are “outside” the game. In this sense they are part of the game’s environment. If the theorist is to construct a descriptive theory of the game he must necessarily take them into deliberate account. The player however can not take them into deliberate account in the way the theorist does while he attends the game. Nevertheless the player’s actions as moves-in-the-game are necessarily subject to these conditions both for their recognition by him as well as for their effects. Only if the player treats his situations of play as objects of theoretical scientific interest can he take these game-furnished conditions into account in the way in which the theorist does. Short of this he must address them in the manner of practical interests. This is to say that a player’s compliance with the basic rules of the game is accorded priority by him over compliance with the rules of theoretical inquiry when the player is directed to describing the game that he is really engaged with. The basic rules of the game are employed by the player as a scheme of interpretation for describing his real situation. This contrasts with the theorist’s procedure of employing the rules of theoretical inquiry as a scheme of interpretation with which to constitute the game as an object for description. Game-furnished conditions thereby are treated by the player as events with which he must come to terms. His theory of the game as a definition of his situation is governed by these considerations, rather than by considerations in which his real situations of play consist of theoretically possible states whose reality consists entirely in, and is guaranteed only and entirely by, the methods used for achieving their consistent, coherent, clear ordering as empirical possibilities. Game-furnished conditions describe characteristic features of moves-inthe-game and thereby describe characteristic features of any play of the game. Their effects operate independently of a player’s chances for success or failure in the game. They are invariant to the changing present states of the game; they hold for every present situation in which a move must be made; and hold over every play of the game. They hold insofar, but only insofar, as the player treats the basic rules of the game as maxims of his own and the other player’s conduct. They hold insofar

Parsons’ Primer

199

as the player seeks to act like a player of the game, i. e. insofar as he subjects his actions to definitions, interpretation, and sanction, to the legitimate order of possible occurrences that are defined by the basic rules of possible play. Players’ choices of moves must satisfy these conditions if the theorist is to recognize the move in a developing or accomplished play of the game, and if the theorist is to recognize the play as an instance of a play of the particular game. Players’ choices must satisfy these game-furnished conditions too if the play is to acquire and retain its identity for players as a play of the game over the temporal course of the players’ attempts to accomplish a play of the game. Finally, a play of the game must satisfy these conditions if the play is to be a stable one. Game-furnished conditions are to games what exigencies are to Parsons’ system of action. The analogy is an exact one; the properties of game-furnished conditions and exigencies are identical. When one proceeds from one game to the next, as for example when one compares chess and kriegspiel, the game-furnished conditions are exceptionally variable in their specific content, but the formal properties of these gamefurnished conditions that have been described before remain invariant to the variations in content. Parsons’ definitions of the exigencies depend upon this important fact. Thus for example the following are examples of gamefurnished conditions in chess and kriegspiel. 1. In chess each present state of the game (in both chess and kriegspiel the “present state of the game” is defined by three parameters: What are the pieces? Where are they? Whose move?) is for either player one of perfect information. In kriegspiel each present state of the game is one of imperfect information. 2. In chess each present state of the game is altered only by the occurrence of an actual move. In kriegspiel each present state of the game is altered by the occurrence of an actual move only for the player whose move it is, while the present state of the game may be or may not be altered by an actual move of the player’s opponent. In kriegspiel as contrasted with chess each present state of the present state of the game can be altered by a remembered, imagined, supposed, anticipated move. This of course is not the case in chess. 3. In both chess and kriegspiel each present state of the game endures until an actual move occurs upon which, and by reason of which, the present state of the game flops over into a succeeding state. In both, the temporal structure of a course of present states of the game consists of a set of discrete successive states. But only in chess can these states be arranged in strict chronological sequence at any present state of the game. In kriegspiel one has not a chronology but a “history.”

200

H. Garfinkel

4. In chess each present state of the game can be constructed as an ahistorical state of affairs. It can be completely described without respect to either the procedures whereby the present state came about, or to the procedures that describe how it will have developed. In chess the player may exercise the choice of treating the present state of the game either as an historical event or as an ahistorical state of affairs without in either case impoverishing his grasp of the present state of the game as a definition of his real circumstances. In kriegspiel the present state of the game is identical with the game-as-ithas-developed-thus-far. The present state of the game thereby consists of a retrospective-prospective reconstruction of an accomplished course of play whose parameters at best can be assigned only approximate and probabilistic values. The player who attempts to exercise the choice of treating the present state of the game as an ahistorical state of affairs can only do so at the necessary cost of impoverishing his grasp of his real situation. Many more features might be enumerated. The point to be taken from this much is that player’s decisions are necessarily governed by these gamefurnished conditions as conditions that sustain the players’ commitments to compliance with the basic and preference rules. Basic rules, preference rules, and game-furnished conditions provide the theorist with the conditions that players’ actions must satisfy if the theorist is to describe the game as a set of activities with definitive features that discriminate it from an environing set of activities. (Furnish examples of game-furnished conditions from linguistics and traffic studies.) Although physical and biological events make up the exigencies, in no sense are exigencies synonymous with physical or biological events. For example the properties of the present state of the game – perfect information in chess or double solitaire, imperfect information in kriegspiel or bridge – are neither physical nor biological events. Neither are the events in the statistical structure of language. The fact is that these exigencies are such that to every system of action there corresponds its environment by reason of that system’s features as a set of normatively governed actions. The system’s actions are related to the exigencies as their solutions. Parsons talks of these solutions as the “adaptation” of the system to its exigencies. Think of system’s actions as “coming to terms with” the exigencies. How is the adaptation, or “coming to terms with”, to be described? Parsons rejects the prevailing conception which formulates these exigencies as determinants of the system’s features. Examples: textbooks which speak of the biological basis of culture; economic, political, geographic, biological, territo-

201

Parsons’ Primer

rial, climactic determinisms. The “of course all these are important” treatment is a mere truism. The question remains how is this importance to be conceived while providing for the fact that the actor and the theorist encounter these exigencies in the identical way, i. e. through their knowledge of the conditions of their own actions. The problem is not one of naturalistic causation, but of strategies of action whereby the consequences of action are whatever they are. Parsons’ solution is to provide for actors who “symbolize” the meanings of these “objects.” That is to say, that actors encounter these exigencies only in the manner of the “mechanisms” of adaptation; through knowledge of these exigencies, i. e. through symbolic representations of the system’s environment, i. e. the real world for the system of action as its irreducible grounds of action. Parsons’ actor must have knowledge of the real world and must be oriented to its treatment, whether he is capable of saying so or not. His actions show it. UNFINISHED

202

H. Garfinkel

LIST OF TITLED NOTES FOR THE CHAPTER ON VALUES I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI XVII XVIII XIX XX XXI XXII XXIII XXIV XXV XXVI

Parsons’ delineation of the tasks of the problem of order as a “calculus of significance” How Parsons’ usage of “social system” handles difficulties with the concept, “group” Decisions that preceded the action-situation of objects schema to conceive observables Paradigm of social interaction Norms and normative Normative environment, (E): norms and values Rules and maxims of conduct Values Values explicated with the use of basic rules of games Values stability and an answer to some criticisms Values versus slogans Values are not inferred from actions Values control norms “Level” of commitment to values Values are not “in the head” Values and the subjective categories Real perceived social structures Empirical observables intended by the term, values Empirical meaning of values Goals Values furnish norms their “rationale of legitimacy”; norms furnish values the programs for their production The two theorems Review of Parsons’ answers to critics in “The Societal Value System” A rationale for a method for deciding between Parsons’ and his critics’ arguments on American values More on a method

Parsons’ Primer

203

CHAPTER VI PARTIAL DRAFT AND NOTE FOR A CHAPTER ON VALUES43 I PARSONS’ DELINEATION OF THE TASKS OF THE PROBLEM OF ORDER AS A “CALCULUS OF SIGNIFICANCE” In the seminar Richard J. Hill distinguished between a calculus of truth and a calculus of significance.44 By pointing to a calculus of significance, he meant to ask “if there were any rules that the sociologist could construct with which to decide that one problem was worth investigating whereas another one was not.” We shall not dignify the material in the previous chapters by calling it a “calculus of significance”, but the elaborate theory that Parsons constructed is directed exactly to this point. Parsons’ work is to be considered a “calculus of significance” insofar as his metrics have defined the nature of the problem of social order as a constituent task, and insofar as the problem of order is in fact equivalent to the task of structural analysis.

II HOW PARSONS’ USAGE OF “SOCIAL SYSTEM” HANDLES DIFFICULTIES WITH THE CONCEPT “GROUP” In sociological discourse, terms like group, organization, system are commonly employed. What are these terms good for? What is a group good for? What is a group? Such concepts are of exceptional importance; if sociologists are deprived of their use, their interests would have to become those of other professionals, like clinical psychiatrists or physicists. In an important sense a person is in the sociological business because he uses the terms. In order to avoid the difficulties of having to clarify its usage, it is sometimes said that these terms are primitive terms of sociological discourse. Does that mean that a term like group is really undefined, and that the sociologist who uses it must always wait to find out what he means by it by first talking about it for a little while? It cannot possibly be that “primitive”. This is why. 43

[Editor’s Note: This chapter and the three that follow were written after the first five. Only the notes for this chapter were included in the original Primer mimeo. The last page of Chapter V in this volume.] 44 [Editor’s Note: This would be the seminar for which Garfinkel was producing these lectures/chapters. Richard Hill was presumably a student in the seminar.]

204

H. Garfinkel

Suppose that a foundation representative offers an investigator a great deal of money in order to study four people playing poker, and requires that he study them as a group according to any conception of a group that is available. Is the term, group, such a primitive one that the various ways in which a sociologist would analyze and describe four such persons playing poker would not differ as he proceeded from one conception of these persons playing poker to another using some theory of a group or some conception of a group to organize his thought? And is the concept so primitive that he would not know before he began that the descriptions would differ and would know how they would differ? For example, consider a very empirical notion of group, in fact as empirical a notion as one is likely to find. Consider the Stephan-Mishler45 notion in which preceptorial groups at Princeton ranging in size from two to seven persons were observed according to the following procedure. An observer was placed in the room with these students and their instructor. The observers conceived the talk that was going on according to the following rule for recording this talk, namely, that when a person talked the fact was marked by writing “1” opposite his name. No other mark was made opposite that person’s name until someone else had talked before he talked again. Such an action consisted of the fact that someone continued talking until someone else talked. The set of possible observable occurrences was very clear. The procedure for counting these actions was built in to the very conception of what it was that was under study. Using this procedure, an investigator could observe four persons playing poker and count the remarks as “distributed among these persons”. Then the observer would plot the relationship between the rank order of frequency of bets on talk, and the comparative percentage of times that persons made bets and talked. That would be one way of describing four persons playing poker, that is to say, of analyzing the activities of poker-playing as a StephanMishler group. As an alternative, one could describe the identical poker-game with the use of Bavelas’46 conception of a communications net. If one did not want to do it in Bavelas’ way one could use Hemphill’s47 group dimensions. 45 [Editor’s Note: Likely referring to: Stephan, F. F., & E. G. Mishler. (1952). “The distribution of participation in small groups: An exponential approximation”, American Sociological Review 17(5):598–608.] 46 [Editor’s Note: Alex Bavelas published a paper in 1942 on individual and group ideology and another in 1944 on leadership training. He also worked on communication networks at MIT during the late 40s and 50s, and on group behavior and measurement in the 1940s and 50s. See, for example, Bavelas, A. (1952). “Communication patterns in problem-solving groups”. pp. 1–44 in Von Foerster, H. (ed.) Cybernetics: Circular causal and feedback mechanisms in biological and social systems. Oxford: Josiah Macy, Jr. Foundation. He also seems to have been connected to the Ohio State University, where Garfinkel was employed for a brief stretch in the early ‘50s.] 47 [Editor’s Note: John Hemphill was also connected to the Ohio State University (Bureau of Business Research). He worked on group processes, and devised an instrument, the Hemphill Questionnaire, to measure groups along 13 dimensions, including autonomy, control, and intimacy. See, for example,

Parsons’ Primer

205

These would furnish one more way of analyzing the action of poker-playing. Each would produce a description of poker activities conceived in accordance with each of the different ways of describing them as group activities. Nevertheless there is considerable justification for talking about the primitive character of the group concept if one considers the touchy problems that any theorist is likely to have to come to terms with in making a description of something like poker-playing under the auspices of the concept of the group character of this activity. Some of these problems can be briefly reviewed. To begin with, there is the questionable empirical meaning of the duration of the group. Investigators would like to ask, do ask and interestingly enough answer such questions as, when is the group formed? When is the group “there”? When does the group cease to exist? How long did the group last? (From the remarks here on solvency, bankruptcy, life expectancy of businesses, to the Wichita experience when the jurors had their first “party” in the jury box, clock time compared with phenomenal time in talking about duration, Wittgenstein on linguistic usages in language games.) A second problem concerns the questionable empirical meaning of constancy or identicality of the group. Is it the same group today as it was five years ago? Is it a different group if there has been a turnover of the persons who are members of it? If a family was interviewed on Sunday, is it the same family that is being observed on Monday if the children are away at school? If all the persons in the group die and are replaced, is the group to be counted one group? (Cf. Felix Kaufmann.)48 A third questionable empirical meaning is concerned with counting groups. How is a group to be conceived as an entity such that its theoretical meaning permits arithmetic operation on groups. Are we dealing in literal usage when we say there are ten groups, one dissolved, nine groups were left. Is talk about the existence of many groups in a town or of many more groups in one town than in another to be understood as a quantitative comparison within the rules of arithmetic? (Correction: consult the note from Social Control lectures for an unelaborated enumeration of another dozen problems.) (With respect to each of the enumerated difficulties tell briefly what the empirical meaning would be within Parsons’ usage of the concept social system. For example, a system does not exist where Parsons is concerned. A system Hemphill, J. K. “Group dimensions: A manual for their measurement”. Research Monograph 87, xi, 66. Ohio State University: Bureau of Business Research, and Hemphill, J. K., and C. M. Westie. (1950). “The measurement of group dimensions”, The Journal of Psychology 29(2):325–342.] 48 [Editor’s Note: Possibly referring to: Kaufmann, F. (1936). “Remarks on methodology of the social sciences”, The Sociological Review 28(1):64–84.]

206

H. Garfinkel

consists of a set of procedures that Parsons uses to define a field of possible observable occurrences. Thus the only thing that can happen to a system is that it can fall into disuse.49 A system does not grow, it is not born, it doesn’t die, and the rest. It is a procedure that the sociologist can use to collect thoughts about the related character of activities and their objects. It is by reason of this however that clear empirical meanings of statements like “The system exists” or “The family was established” or “There are many more groups in this small town in than in that one” can be furnished by the systematic use of Parsons’ method. Do the same briefly for each of the enumerated difficulties).

III DECISIONS THAT PRECEDED THE ACTION-SITUATION OF OBJECTS SCHEMA TO CONCEIVE OBSERVABLES Parsons has gone through several stages in conceiving the empirical nature of the matters that are under observation. There is every prospect that there will be further changes. Parsons started in The Structure of Social Action by conceiving that the thing he was describing consisted of the activities of persons directed towards the accomplishment of goals and constrained by various conditions. One set of these conditions he called physical conditions; another set he called normative conditions. The problem of the relationship of action to its goal was handled with what Parsons called a means-end framework. This changed. Somewhere around 1940 Parsons substituted instead what he called an actor-situation of objects framework. The construction of means-ends was dropped. Parsons tried instead to conceive a set of observables by referring to persons via the term actors and putting them into environments that he called cultural situations of objects. This environment consisted of a physical environment with a cultural overlay. Running into difficulties with this construction he changed once again. In its place he used a three-part schema of actor, action, and situation of objects. This schema was a familiar one to the preferences in social psychology both at that time as well as today. Parsons populated his society or his group with persons; made it very definite that they were going to be depicted as actors; that they were invested with certain possibilities of activity; and this activity was directed to an environment of other persons similarly conceived. The situation of objects retained the feature of a physical environment over which there was a cultural overlay of “meaning”. The actor49 [Editor’s Note: On this point, cf. Garfinkel, H. (1946). “Some reflection on action theory and the theory of social systems”. Unpublished manuscript, Garfinkel Archive.]

Parsons’ Primer

207

action-situation-of-objects scheme had the additional feature that it conceived the persons to be related to the situation via the action that the actor directed to the objects that made up the situation. Thus, for example, the sociologist might have as a domain of observables a set of persons scattered around a room which is defined with respect to its physical characteristics. Imagine that there is a meal on the table at one end of the room with a person at the other end. Then the person is related to the meal via his “orientation of action” as for example via the fact that he desires to bring himself into proximity with the dish of food the better to satisfy his hunger. Then the course of his actions would be described as a changing relationship between the actor and the table with this dish of food. Empirically, it would have been recorded as a changing linear relationship between the actor and the dish, the changing distance being under the control of the actor. This formulation was used in The Social System. Parsons changed again in the succeeding formulation. The actor was dropped out leaving only courses of action or treatments of an environment of objects in its place. Any reference to the actors that started with the person as an analytic notion was put aside. There was no longer an actor as a substantial unit loaded up with potentialities for altering a relationship between “himself ” and “his” situation. Instead the actor consisted of a set of activities. Thus the model to frame these observables had two terms: actor and situation of objects. The term actor meant a set of possible activities in their course that are directed to an environment of social objects and consists of the treatment of these objects. The change that is involved on the situation side consists of this: that the situation was no longer a physical environment with a cultural overlay but was entirely a cultural environment. The facts of physics and biology nevertheless came in as conditions of activity that were handled as exigencies to which structures of action directed to cultural objects adapt as environmental to the structures. The physical environment was of interest only with respect to the problem of how persons were conceived to go about the business of exhibiting socially controlled and concerted actions to Parsons who was then only interested in accounting for the production and features of concerted activities: features like stability, duration, initiation, termination, their motivated and socially controlled character. This scheme continued to be used through the material of the book on the family. Then it changed again. The most recent formulation starts with Economy and Society50 and is elaborated in unpublished materials since then far beyond 50

[Editor’s Note: Parsons, T., & N. J. Smelser. (1956). Economy and society. London: Routledge & Keegan Paul.]

208

H. Garfinkel

anything Parsons had written before. The action-situation model is retained, but Parsons has provided a third term which reintroduces the concept of a relationship between actor and situation. This relation is one between courses of action and situations of objects both of which are conceived as system of action. Cultural systems are included as systems of action. The relationship between the systems consists of exchanges of effects between systems of activities. Sets of action-directed-to-objects are to be analyzed as exchanging effects that these structures have for each other, once different sets are conceived to be related in the way that Parsons prefers. The relationships are spelled out between these activities in Parsons’ recently highly developed paradigm of boundary exchanges. The exchange effects are handled with the categories of input and output. This schema is elaborated later in the book. There are at least five closely discriminable conceptions that Parsons has used, all of them directed to solving the problem of how to conceive the activity of persons under the auspices of their “group” character so that one can describe in a literal way such events that are of obvious interest to sociologists as the duration of activities, the numbers of members, the work that determines their identicality over time, in fact all of the sociologically interesting properties of conduct which sociologists in their aim of rigorous description have chronically stumbled over.

IV PARADIGM OF SOCIAL INTERACTION A convenient place to begin the explication of Parsons’ system is with his “paradigm of social interaction”. Weber furnished most of the essential matters the “paradigm” is talking about; Parsons’ formulated them as a “paradigm.” In Part I of Weber’s theory of Social and Economic Organization51 he started with the notion of the behavior of a person. This term refers to the movement of a body with respect to its surroundings. It is a physical event and as such permits precise and literal description. (See for example the elegant descriptions of structural linguistics, or Dodd’s work to illustrate difficulties in confining the sociologist’s intention to the description of the behavior.) Behavioral events for Weber were not events of particular sociological interest, although it was necessary to refer to behavioral events in order to construct a general category of action which does construe the matters of sociological 51 [Editor’s Note: Weber, M. (1947). The theory of social and economic organization [T. Parsons, trans.]. New York: Free Press.]

Parsons’ Primer

209

concern at least as Weber intended to be concerned. Weber talked about “behavior with a subjective meaning attached to it and governed thereby in its course.” What is this subjective meaning? Its empirical sense can be defined by the use of the following demonstration. (Insert excerpt here from the paper “classroom demonstration of Sociological Concepts”.)52 Weber then proceeded to define the concept “social action” which he did by elaborating this subjective meaning to include in the actor’s environment the expected replies by the other person to the actor’s actual and anticipated courses of action. Weber now had a course of action governed by an environment which includes for the actor the anticipated reply of the other person. Social action is a type of action in which the actions of two or more persons are “oriented” to each other in this fashion. Weber then used the term social relationship to refer to the likelihood assessed from the standpoint of the actor that the actions of persons would be conducted in this fashion. A social relationship for Weber consists of the “chance” that actions would actually be governed in this fashion. Then in the particular manner of his genius, Weber introduced the concept that put sociology on its feet: the legitimate order. The legitimate order is also a way of referring to environmental features of the acting parties. With its use Weber proposed that so far as [actors’] expectancies were concerned, these expectancies would be referred to as legitimately expected environmental features, i. e. to the correct or to the sanctionable character of expected interactions as assessed by interacting parties. The legitimate order provided for the sanctionable features of persons, notions, numbers of persons, territories, relationships, and the like all of which had already been incorporated into the initial conception of action in the first place. In his theory of (insert excerpt here from my psychiatry article).53 In his Theory of Social and Economic Organization Weber rang the changes on the various types of legitimate order, social relationships, activities, goals to which the ac52

Garfinkel is referring to his paper, “Some Demonstrations of Sociological Concepts: An Essay in Sociological Theory”, which describes activities he devised to explain Parsons and ethnomethodology to his students. A note accompanying the paper in the Garfinkel Archive says that it was read at the 1956 meetings of the American Sociological Society in Detroit. The note continues: “This paper fell between two stools; it was not enough of a demonstration article because the emphasis was too theoretical; it was not enough theoretical because it talked about classroom demonstrations. It was submitted for publication to the Journal of Educational Sociology and rejected on the grounds that it ought to cut out the theory; it was submitted to Social Forces and rejected because it was too “applied”. I always thought of it as an article that combined theory and methods. The demonstrations were developed and used in my courses at U. C.L.A.”. 53 [Editor’s Note: The reference is either to Garfinkel, H. (1961). “Reflections on the clinical method of psychiatry from the point of view of ethnomethodology” (unpublished manuscript, file drawer 20, Garfinkel Archive); or to Garfinkel, H. (1956). “Some sociological concepts and methods for psychiatrists”. Psychiatric Research Reports 6:181–195.]

210

H. Garfinkel

tivities might be directed, and with his architectonic method generated the various types of social organization for which he is famous. Parsons’ paradigm of social interaction elaborated these essential ideas and formulated them as a rigorous definition of social structures of interaction. Parsons’ paradigm provided for a set of actors, whose actions were directed to some future state of affairs, the actor having access to various resources and instrumentalities employed in the manipulation of a present state of activities so as to bring it into conformity with a future, the use of instrumentalities being governed by various rules that defined for who complied with them “correct procedure”, technical, ethical, moral, aesthetic – provided a normative order of activities and objects, to which each of the actors held himself and others, to a respect for and for the actualization of, and that as departures occurred between legitimate expected states of affairs and actual affairs sanctions were employed to minimize discrepancies between legitimate and actual states of interactional events. The essential ideas of the paradigm and their empirical meanings can be shown in the following classroom demonstration. (Insert here the section from my paper on classroom demonstration of sociological concepts.) (Discuss briefly institutionalization and internalization.) (Briefly consider the question: when the theorist seeks to incorporate into the features of the social system the possibilities of change with respect to its environment, one of the environing systems consists of the activities that are analyzed with the use of the concept of “personality” as a system of activity. The question then is, how to reconcile the possibility of stable institutionalized practices with the fact that not all of these institutionalized practices are “wellinternalized.”) The paradigm can be used as a sense-transforming operation. Applied to the activities of four persons playing poker, for example, it would delineate the sense of those activities as events of interaction provided for by the paradigm. Because the paradigm is a definitional procedure it would merely define a domain of study. Of course this is all that the paradigm of social interaction is intended to serve. As it stands it generates only those theorems about social structures of interaction that one can obtain by an ad hoc consideration of the consequences of this definition. (An example. Use the classroom demonstration to give examples of empirical content.) By furnishing the paradigm two critical rules of interpretation, the paradigm formulates the essential ideas of Parsons’ systematic theory. Important and interesting theorems can then be generated from it. Parsons refers to the two critical rules as “fundamental propositions about the social system”. They are not meant or used as empirical propositions, but

211

Parsons’ Primer

are instead used as rules to govern the way in which the theorist would use the paradigm of social interaction to collect his thoughts about observable notions. The first rule states that the real social structures consist of institutionalized patterns of normative culture. The second rule states that the stable features of the real social structures are the products of motivated compliance to a legitimate order. That does not mean motivated compliance to the real society, nor does it mean motivated compliance to the legitimate order. It means exactly what it says: motivated compliance to a legitimate order. (Later we shall examine the relationship between stable possible legitimate orders and the concept of the real social structures.)

V NORMS AND NORMATIVE In order to explicate the first rule, we shall start with the concept of the culture as Parsons defines the term. Its meaning can be easily summarized. Culture consists entirely and exclusively of values and norms. The definition can be stated that simply, if we understand what Parsons means by normative. The concepts of norm and normative are so central in Parsons’ work and are so thoroughly at the heart of it in his later writing that if one opens a page of his writings at random and selects at random any word on the page there is a very high likelihood that one has landed in the near neighborhood of that term in the text. A grasp of what Parsons means by normative should settle the question of whether or not Parsons is in sight of the real world because the way Parsons keeps the real world in sight is through the events intended by this term. How is the term normative used? First, it refers to some set of possible events in the sociologist’s actor’s environment. This environment is a behavioral environment. It consists entirely and exclusively of cultural events, not physical or biological events. The features of the events it contains are exclusively motivated features – “meanings”. That the term normative refers to events in a behavioral environment cannot be overemphasized. Second, the sociologist’s actor treats these possible events in the behavioral environment as a standard. Third, these possible events in the sociologist’s actor’s environment, which the actor treats as a standard, depict for the actor conditions under which sanctions may be employed. They consist of the conditions under which the actor, in applying either rewards or punishments, for applying them will expect to be supported by others. They depict the

212

H. Garfinkel

conditions under which the actor counts himself correct for taking corrective, restorative action. Finally, these events are treated by the sociologist’s actor as maxims of conduct to compliance with which the actor is committed. By “committed” is meant by the theorist the conditions under which the actor’s motivated compliance with these events as maxims of conduct are guaranteed. To speak of an actor’s commitment to a normative order is to tag as a matter of the theorist’s empirical concern the problematic conditions under which the actor’s motivated compliance is insured. The real normative events of Parsons’ actor’s environments are the ones that Parsons’ actor actually, not supposedly complies with. The question is not whether or not but which ones. No motivated compliance, no normative environment. This rule holds for all cases except the case of the ideal theoretical terminal state of complete anomie. In Part II of the Introduction to The Reader on Sociological Theory54 Parsons invited the reader to consider that a game incorporates all of the “main structural components” of a system of interaction. One of these structural components is a normative order of environmental events. Since Parsons invited us to consider the game, let us consider several rules of play from bridge and see what he’s talking about. In a bridge game a rule runs that players play in serial order, A, B, C, D, A... After we had observed an actual play of the game we would have two sets of events that we could compare. One set would consist of the actual normative order of play that players “subscribe” to; a second set would consist of the actual order of play that actual players exhibited in their playing conduct. The set of normative data might be displayed in the manner of Tables 1 and 2. TABLE 1 ACUTAL NORMATIVE ORDER OF PLAY FOR BRIDGE FOR THE RULE A, B, C, D, A... Observed percent of all subjects who play bridge who showed disapproval and corrected their fellow players or apologized to the other... Upon the occurrence of the sequence 0% A, B, C, D, A... All other sequences 100%

54 [Editor’s Notes: “The Reader in Sociological Theory” was likely the title of a manuscript that was eventually published as Theories of Society: Foundations of Modern Sociological Theory (2 Vols.)., edited by Parsons, T., E. A. Shils, K. Naegele, and J. R. Pitts, Glencoe, Il: The Free Press, 1961. Garfinkel would have received the pre-published version from Parsons.

Parsons’ Primer

213

TABLE 2 OBSERVED PERCENT OF ALL SUBJECTS WHO SHOWED APPROVAL OF SEQUENCE A, B, C, D, A... AND DISAPPROVED OF ALL OTHER SEQUENCES WHO INVOKED AS JUSTIFICATION FOR THEIR ACTIONS THAT ANY BRIDGE PLAYER WOULD BE ENTITLED TO ACT AS THEY DID In the face of a challenge to propriety... Invoked the jusInvoked other Treated any tification that justification justification as any bridge player irrelevant would be entitled to act as they did Approved 100% 0% 0% the sequence A, B, C, D, A... Disapproved all 100% 0% 0% other sequences

The set of data for the actual order of play that players exhibited in their playing conduct might be displayed in the manner of Table 3. TABLE 3 OBSERVED PERCENTAGE FREQUENCIES OF POSSIBLE SEQUENCES Percent Sequence A, B, C, D, A... 100% All other sequences 0%

And a final table: TABLE 4 OBSERVED CORRESPONDENCE BETWEEN SUBJECTS WHO PLAY BRIDGE WHO “SUBSCRIBE” TO THE BRIDGE RULE A, B, C, D, A... AND SUBJECTS WHOSE ACTUAL PLAY CONTRIBUTED TO THE OBSERVED SEQUENCE A, B, C, D, A... Played in sequence Played in some other A, B, C, D, A... sequence Approved the sequence 100% 0% A, B, C, D, A... AND disapproved all others Approved any other se0% 0% quences than A, B, C, D, A...

Parsons provided for an actor who is directed over the course of his actions to an environment whose features consist of the expected ways in which numbers of persons, who they are, how they dress, how they talk with each other, what they talk about, and the rest, from the point of view of the actor can occur as

214

H. Garfinkel

a sanctioned set of occurrences. Parsons provided as well for the actual way in which these events can occur, and included among these events the occurrence of norms. Finally, Parsons provided for actors who are responsive to differences between actual normative and actual observed environmental events, and who under certain conditions of this difference take corrective action. When Parsons talks about normative patterns of culture, he is talking to the following points: (a) that actors, by their actual activity, attend an environment which consists of the sanctioned expected ways in which events should occur, (b) that by their actions persons produce some approximation to the very events which the normative order provides in the first instance as correct events. The real social structures are the consequences of persons operating in this fashion. More literally, the real social structures consist of institutionalized normative orders of action and their objects. To say that a society “exists” means that a set of persons treat each other under the auspices of, or, their treatments of each other are governed by, a normative order of social structures which they seek by their actual activities, to hold each other to a respect for, and to the production of. Thereby the realistic character of the sociologist’s categories like age, sex, occupation, class, family practices, and the like depend on the fact that persons are so engaged in normatively governed activities as to make the sociologist’s categories realistic descriptions of their actions. For example, the realistic character of the census categories depends not upon physical or biological facts of life, but depends instead and exclusively on the extent to which persons, within the terms of normatively regulated actions, engage in those activities which are described with the census categories. We have defined what Parsons means by normative. In brief, the term refers to sanctioned expected occurrences of environmental events. The production of the real social structures, their “existence”, depends upon the probability that compliance with a normative order of social structures can be enforced. Real social structures that are of particular interest to Parsons exhibit stable properties, i. e. they describe the stable society. In Parsons’ conception of society, there is an intimate tie between the sociologist’s actors’ environments that have for actors the features of being actually known or potentially knowable in common, i. e. of institutionally guaranteed environments-known-incommon, and the production of the real, stable society. Parsons ties the two together with the use of two concepts: “values” and “norms.” Both terms refer to the sanctioned, expected environmental events. Sanctioned expected environmental events are called values when they exhibit one set of discriminating features; they are called norms when they exhibit a different set of discriminating features.

Parsons’ Primer

215

To explain Parsons’ usage of the term “values” we need to list the definitive features that environmental possibilities can show, for it is these features to which the sociologist’s attention is being called, and it is these features that are being tagged by the term “value” as observable events. The same is true for the term “norms.”

VI NORMATIVE ENVIRONMENT, (E); NORMS, VALUES 1. The constituent ideas in the concept of the actor’s normative environment are: a) A set of possible events of social structure. b) They are used by the actor as a standard. c) They define for actors oriented to them the conditions under which the actor expects to be supported for taking corrective action. d) The actor is committed to compliance with them as maxims of conduct. Call these constituent ideas the “parameters” of the general concept, normative environment. Nothing is being said or is necessarily implied about the clarity, consistency, validity, definiteness, operativeness, empirical “adequacy”, etc. for actors of these possible events, their status as standards, or their function as conditions of corrective action. 2. Every rule, without exception, can be translated into a set of expected possible features of socially structured “behavioral environments” (to use Koffka’s term for referring to the environment “seen” from the actor’s point of view.).55 3. The result is that some rules – e. g. the rules of some kinship systems, the rules that define rational play in some games – can be mapped with respect to the mathematical relationships between the events they propose, and thus the logical properties of these normative orders can be rigorously explored. Many rules, however, cannot be mapped in a literal mathematical fashion with present mathematical resources. 4. The normatively depicted environment is always an environment for some actor. This holds according to the doctrine that every intended object 55

[Editor’s Note: this likely refers to Koffka, K. (1935). Principles of gestalt psychology. London: Routledge.]

216

H. Garfinkel

is an object of someone’s intention. More simply: no humans, no human culture.

There are more actors than humans, and an actor is not a human let alone human, but this does not contradict these statements.

5. Since the reference to an actor is necessarily involved, the sense of the rule as a regulator of conduct is retained by saying that a set of expected possible events that an actor seeks to produce, or that he respects as the productions of others, defines the term, “maxim of conduct.” 6. Every system of rules thereby can be used by the theorist to construct a normative order of actions, relationships, numbers of persons, territories, social types, communications, communicative paths, feelings, motives, biographies, futures, etc. all having the property that they are maxims of conduct. One of the very powerful points and unique contributions of Parsons consists in the fact these normative orders are ordered according to the structural functional paradigm. Thereby they define the actor’s environment literally as a system of relevances. 7. All normative states are given a scenic reference with respect to the actor since they are by definition objects of his actions. By the terms of action theory an actor consists entirely of a related set of courses-of-action. From the theorist’s actor’s point of view the terms “subjective” and “objective” refers to the grounds on which an actor decides the meaning of an object. As theorists we say that if the actor’s decision is made by consulting rules that he expects others to know and employ in a more or less similar way, and if he refers this similarity to a matter of required public knowledge then we say that the meanings are objective meanings (e. g. the meanings of words for those who subscribe to Webster’s dictionary as the set of rules defining correct correspondence between terms and referents). If the correctness of the correspondence between the appearance of the object and the object that is intended through the appearance is referred to matters of personal biographical considerations, i. e. if the actor employs “personal” rules, we say the meanings are subjective. From the theorist’s point of view all objects of his assessment are to be brought to objective status. “Internal” and “external” therefore are theorists’ terms to refer to systems; the skull and the skin are specifically irrelevant. It makes no difference whether we put the entire world in the actor’s head or in the actor’s

217

Parsons’ Primer

environment.56 We still have to go to the real world of activities to find these objects and their treatments

I prefer a scenic reference because it facilitates my inquiries. By contrast Olds57 prefers to put the actor’s objects into the actor’s “skull” as a set of cell assemblies. There is no analytical difference between my actor’s objects and Olds’ cell assemblies. Olds explores the properties of these objects by manipulating the real biochemical, physiological, and neurological conditions of their formation, constancy, organization, etc.; I explore the properties of these objects by manipulating the real social conditions of their formation, constancy, organization, etc.

8. Given that every rule can be translated in the actor’s scenic possibilities, then to every system of rules there corresponds a normatively described environment of cultural events. For ease of exposition a normatively possible environment will be referred to by the term, (E.). 9. Because (E) has the status of a maxim of conduct, there corresponds to (E) a description of the social structures described by the actual activities of actors. We shall refer to these by the term, E. 10. Thus, the frequency statements in the following table of racial intermarriages: Wife Husband

White Negro

White Many Few

Negro Few Many

When treated as hypothetical propositions about the frequency of inter- and intra-racial marriage describe a possible actual behavioral state E. The theorist would decide E’s factual accuracy by consulting the census reports. 56

[Editor’s Note: A similar formulation appears in Garfinkel’s paper on Trust. See Garfinkel, H. (1963). “A conception of and experiments with, trust as a condition of stable concerted actions. pp. 187–238 in O. J. Harvey (ed.) Motivation and Social Interaction: Cognitive Determinants. New York: Ronald Press Company.] 57 [Editor’s Note: Garfinkel is referencing James Olds, a psychologist who was doing research on the brain-behavior relationship in the 1950s and 60s (e. g. using electrodes to stimulate different regions of rats’ brains) and who was associated with the Parsons group.]

218

H. Garfinkel

When these statements are treated as maxims of conduct they describe a possible normative state (E). The theorist would decide (E)’s validity, which is the probability of its enforcement, by consulting some set of persons who are relevant to the theorist’s research because the theorist thinks, they might seek to abide by it as a maxim of conduct, and finding out which among the many related possibilities portrayed in the table evoke which sanctions. The theorist’s findings about (E)’s validity would then be subject to the usual tests for accuracy. Russian propaganda portrays American whites as oriented to a normative environment (E) as the definition of correct interracial marriages such that the upper left and lower right cells contain for U. S. whites “All” and the upper right and lower left cells contain “None” and that any departures from this in the behavioral domain get swift and brutal correction. 11. Obviously the sociologist’s references to the “real social structures” can refer to either E or (E) or to both in their related sense. Parsons’ theorem that the “real social structures” consist of institutionally guaranteed compliance with a normative order of social structures refers to both E and (E) in their related sense. 12. (E) and E stand in a relationship of comparison with a discrepancy between them that can assume some “magnitude.” Obviously the actor is the comparator since we are talking about the environment of objects of his actions. 13. Now a big but obvious point: actors produce by their actions, or respect as the productions by others, social structures E that are some approximation to the structures that are oriented by them as a legitimate order, (E). Thereby we encounter one important meaning of Parsons’ first theorem. Obviously there are other meanings as well. 14. The term “acting in conformity with norms” has as its empirical reference that persons are producing social structures that coincide with the normative structures that are enforced as maxims of conduct. Alternatively, “acting in conformity with norms” is synonymous with “actualization of culture.” 15. A system of stable normatively regulated conduct is by definition therefore one that exhibits the “cybernetic” properties of self-maintenance, selfregulation, and self-alteration. 16. The function of the concept (E) in this theory is to permit the researcher to come to theoretical and empirical terms with the contents of a normative environment. What functions do the concepts of values, goals, and norms serve in this theory? They serve to explicate the “cybernetic” properties so that the notions of self-maintenance, self-regulation, and self-alteration can be used

Parsons’ Primer

219

as something more than invitations to think long thoughts. They explicate the “cybernetic” properties by relating normative contents to the actor’s courses of action. To see how this is so it is first necessary to consider a peculiar logical property of these three concepts: their meanings are grasped not by consulting contents but by consulting functions. A theorist’s decision to use, within Parsons’ theory, as “values” some set of normatively possible occurrences like those described for example in the rules of monogamous marriage is made not by consulting the content of the rule of monogamous marriage, but by asking whether or not the content is an instance of whatsoever normative possibilities define the states that members’ activities are more or less equivalently committed to the production of. The theorist starts with a set of observed or observable activities showing stable properties. If he chooses to conceive these activities as a stable system of normatively regulated actions then he is committed to the notion that some contents of a behavioral environment will stand to the actions that they regulate in a relationship described by the function “members’ actions are more or less equivalently committed to the production of them.” All contents therefore that have the required properties are by virtue of being members of this class of function, “values.” The domain of specific contents over which this function can range is defined by the properties of this function. The peculiar features of this type of conception were described and commented on by Cassirer in his book Substance and Function.58 He calls this type of concept a “relational concept” and contrasts it with a “thing concept” finding the prototype of the latter in Aristotelian logic of classes and finding the prototype of the former in the theory of functions in mathematics. This usage is a hallmark of Parsons’ theories of action and social systems, and is an important source of the power of these theories as methods for delineating the essential structures of the observed phenomena of activities. The failure to see that many of Parsons’ critical conceptions are to be used in this way lies behind many incorrect criticisms of his work, the most prominent ones being that he does not properly discriminate definitional statements from statements of fact, and that his theories are not controlled by or referable to the empirical world. Among alternative states of a system from which an election might be made by an actor we say that a state is preferred in terms of some rule that defines its election as a legitimate one. Rules of preference thereby define the conditions

58

[Editor’s Note: Cassirer, E. (1953). Substance and function. New York: Dover.]

220

H. Garfinkel

under which a desired state is correctly desirable. Thus a value is not definable except in terms of a rule that defines the correct choice. But because all rules and thus rules of preference too must themselves be translated into expected possible features of the behavioral environment we must ask how these features differ from anything else that is found in the actor’s environment. They differ in this respect: the rules provide those expected features of the environment which constitute that environment as a domain of possible events and as a system. With due regard for the qualifications that must be entertained in using games as analogies, the basic rules of games consist of rules of valuation, the states that such rules of valuation define as preferred states of the system are values. Some set of normative environmental possible events are going to be treated by the actors as preferred. “Preferred” is perhaps a euphemism. Like the term “desirable,” it refers to the fact that with respect to the alternative ways in which events can occur one set will require the actor’s compliance as a condition (a) of his grasp of his real environment, (b) of his rights to claim his competence to treat that environment without interference from others, and (c) to engage in stable, concerted transactions with others. Parsons summarizes this by speaking of the choices that are required of him as “an incumbent of a role.” Required choices among alternatives are correct choices, “correct” being defined in terms of sanctioned rules of choice. These rules can define moral, ethical, technical, aesthetic, efficient, etc. choices. The alternative that the rules define as correct is called the valued alternative, or “value.” If the rules are clear, the choice is clear, if the rule is ambiguous, the choice is ambiguous. Of course the thing to be preferred, the “desirable”, may be depicted with a clear rule as something either vague or clear; specific or indefinite, and the rest. Parsons uses the pattern variables to depict the properties that these alternatives can have that the rules provide for in defining the desirable alternative. The election of one among alternative possibilities is a major feature of the “desirable.” The alternatives are inherently comparable. That there are desirable states still does not make the difference between the sub-sets of (E) that Parsons calls values and other sub-sets of (E) that Parsons calls norms. A first feature that makes an important difference is that these patterns of normative culture have, for the actor, the feature that they are shared with others and are treated by the actor as shared with interactional partners. If one elects Parsons’ usage, it is incorrect to say “values are shared” and mean by that phrase that values can be described or are to be defined independently of the existence of “shared conceptions of the desirable.” Rather, “shared conceptions of the desirable” are features of normative environmental events that the term value is intended to call attention to, or better “shared

Parsons’ Primer

221

conceptions of the desirable” is used to name the empirical matters of interest. Indeed, it is incorrect in Parsons’ usage to let “shared conceptions of the desirable” define the term values in more than the figurative way in which one might use the skeleton of the bear to stand for the bear. “Shared conceptions of the desirable” is a device for calling attention to a particular empirically possible feature that normative environmental events can have, a feature that is of interest because of its bearing on the problem of conceiving and investigating the conditions of stable social structures. The theorist’s problem is to find in a set of observable activities “the conceptions of the desirable that are shared.” Parsons proposes that, if the activities show stable features, such “shared conceptions” must “exist”, i. e. the likelihood that compliance with them, whatever they may be, is enforced is high. This empirical feature of the observed actions is related to the stable properties of the social structures as a condition of their stable properties. Another thing that makes the difference between values and norms is that whatever either might consist of content-wise, values are shared by incumbents of different roles. That is to say, if two or more persons in interaction each expect performances of himself and his partner that differ in the character of their contributions to the outcome of their concerted actions – an example would be the coordinated actions of husband and wife whereby the accomplished expenditures of the family income is a produced feature of their concerted purchasing decisions – certain states of possible accomplished expenditures are desirable states that each treats as known to and binding upon both. These would be called values for the pair. Parsons proposes that where there is no difference in the roles that incumbents play, that no distinction between values and norms can (and need not) be employed. Where there is a distinction, then it is necessary to distinguish for each partner the normative environments oriented by each as well as the normative environment that each complies with and expects the other to comply with in a more or less equivalent way; the latter being necessary in order to bring their differentiated understandings under a set of common understandings about the conditions and consequences, as the actors understand them, of their concerted actions. Thus, each “knows” not only what his “part” is, but he “knows” also what is sanctionable by way of the desirable structural effects of their actions “in concert”. The distinction between observable events corresponding to Parsons’ terms “values” and “norms” is dramatically clear and obviously useful when the researcher deals with settings in which divisions by work, age, sex, religion, ethnicity, territorial jurisdiction, and the like are prominent – settings like industrial establishments and cities.

222

H. Garfinkel

Values, in Parsons’ usage, have properties that are almost identical with the normative states of any game that the basic rules of that game provide for. It may be helpful to the reader to think in this way of values as Parsons intends the events that are meant by that term. The analogy can be pushed very far before it begins to raise difficulties. Indeed, in his dittoed paper Societal Value Systems,59 Parsons refers to games in order to introduce the discussion of values by likening values to the “basic rules” of a game. It needs to be stressed however that Parsons is using the notion of basic rules of game play as this notion is commonly understood. The usage is figurative. Therefore the reader should not try to make too much of the term “basic rules” but should instead attend definitive features of normative environments. With this reservation, the following are some of the properties of values that are identical with the properties of states of games that basic rules of play provide. Interestingly, even if the theorist deals with large scale and complex sets of activities like a society, for example, values retain the same properties as basic rules of a game.

For purposes of criticism one consequence of this is that Parsons’ social systems have the features of a table of organization. Later we shall see how this is so, and what its research consequences are, both good and bad.

What are these correspondences between “basic rules” and “values”?

VII RULES AND MAXIMS OF CONDUCT Every rule may be translated into a set of expected possible features of socially organized activities. These expected features the actor seeks through his actions to produce. It is this usage that is intended by speaking of the expected possible environmental events as maxims of the actor’s conduct. Parsons’ actors act to produce by their actions (or to respect the production by others), structures oriented by actors as a legitimate order. In this way we encounter the sense of Parsons’ insistence on the conception that is critical to his entire theory, namely, that the real social structures consist of institutionalized patterns of normative culture. 59 [Editor’s Note: One of the mimeos sent by Parsons for Garfinkel’s course. Located in the Garfinkel Archive.]

223

Parsons’ Primer

Any use of Parsons that ignores this fundamental presupposition results either in the distortion of Parsons or produces a virtual as compared with an actual use of his theory. His structural categories, – culture, collectivity, role, personality, and organism – his rules of interpretation, and the resulting theory of action and theory of social system elaborate his fundamental conception. Failure to grasp it, therefore, is a fundamental source of systematic misunderstanding. Most particularly, the failure leads to the popular but incorrect charge that he has furnished no “reduction statement” that would permit the coordinating of the theory with observable events. When this theorem is correctly grasped, it is obvious that empirical references are intended and demonstrable at every point in Parsons’ work without exception.

VIII Values Parsons uses the following definition of values. He calls them normative descriptions of desirable states of the system in relation to its situation, and shared between incumbents of differentiated roles. The enumerated distinctive features of normative environmental events meets this definition. Parsons uses the term norms to refer to normative descriptions of desirable states of existence which apply to one unit but not to another. Our remarks on this will be made after the discussion of values has been completed. The first point to be made is that the term values refers to some sub-set of a set of expected states of a normative environment. The sub-set consists of normative conceptions of desirable states of the interaction between actors and involving actor and alters as events in the actor’s normative environment. Parsons refers to this as “normative conceptions of desirable states of the system of reference as such.”60 It needs to be mentioned in passing that value refers to a property of the expected states of the system that govern action in the normative not in the predictive sense. A social value system consists of a set of related normative events treated by the actor as a definition of desirable states of the set of interactions between ego and alter. A third enumerated definitive feature consists of this: that the normative events define desirable states of the interactional system between ego and alter in relation to that interaction’s environment, and this normative feature is itself 60

[Editor’s Note: This passage appears in Parsons, T. (1989:557). “A tentative outline of American values” Theory, Culture, and Society 6:577–612.]

224

H. Garfinkel

a part of the set of normative environmental events with which the actor seeks to comply. The relevant sub-sets of normative environmental events can be called values if and only if they are “shared” between the incumbents of different roles. “Shared” is the critical term. The “shared” environmental possibilities depict a normative state that is treated by the actor as known in common with alters. This agreement is an effect that is achieved, maintained, managed, and the rest, by the ability of the actor to act in accordance with three expectancies without incurring irreducible incongruities or “strain.” The three assumptions are these: he assumes that of the possible states, compliance with certain ones are required as maxims of his conduct; he assumes that the required states binding upon his courses of conduct are binding upon alters in a more or less equivalent way; and he assumes that as he assumes it of alters, alters in turn assume it of him. The empirical “measure” of “shared” therefore is only elliptically represented if some measure of central tendency is used. Such a measure would be at best only and “indication” or “index” of the phenomenon that is intended through the term. Appropriate qualifications would have to be understood with such usage. A small standard deviation, for example, would be only one product of these assumptions as they were routinely employed over a course of action whereby desirable states were actualized. Its proper measure would consist of the work that it takes to call forth restorative activities. This is so because of the peculiar role of time in the structure of social action. That subject of normative events to be called values has the additional feature of the terms in which witnessed behaviors are assigned by actors the status of events of conduct. In their terms, the witnessed acts “acquire meaning” which are assigned by actors to behavioral displays and are interpreted by actors with reference to this known-in-common set of normative conceptions. They are the terms in which situations are not only defined in a common way but are constituted in their meanings as “known in common”. The sub-set has also the feature in use of overcoming the particular character of specific responses to specific situational stimuli. Thereby they define the features of actually encountered occurrences in the actor’s environment which remain invariant to their changing actual appearances. By the use of this definitive property, Parsons succeeded in defining a most critical area of empirical researches of fundamental theoretical relevance, i. e. he has equated the phenomenon of object constancy, with the work whereby motivated features of objects are managed, sustained by the concerted work of actors, so as to retain the object as a temporally identical one over the course of the changes in encountered appearances, as evidences of, or specifications, of the object.

Parsons’ Primer

225

Thus, the research problems of “stable values” and “object constancy” are most intimately related. The sub-set of normative possibilities that are called values are universalistically defined within a “reference of relevance.” The sub-set of possibilities are neither specific to function or to situation. This means that their normative validity, that is to say, the likelihood that compliance with them will be enforced, is not a function of the particular categories or personalities of actors of whom compliance is required. The likelihood that is assigned to these possibilities that compliance with them will be enforced is not a function of the nature of or of the technical knowledge available for realizing the desirable states that they describe. When such “situation specificity” is introduced, note is taken of this determination by referring to such possibilities as goals. Of the states that such normative possibilities describe, the description does not take account of the “internal differentiation of the system of interaction” in which they are institutionalized. In their relevance they transcend functional differentiation. (Furnish examples from games and the clinic studies.) The following points are to be made in enumerated fashion by way of explicating the concept of values as a rule of interpretive procedure. 1. The term value is reserved for those patterns of normatively possible events which are on the “highest level of generality which is relevant at all to the orientation of action in the social system of reference,”61 i. e. (a) values are the principles from which the less general norms can be “derived” as legitimate expectations. b) They stand on a level of generality which is independent of any specific object-structure; c) They involve only the exigencies which are generic to the category of social system in question. For example, in the case of a society, these exigencies are (i) that it is a social system at all; (ii) problems of cognitive grounding and symbolic expression of values; (iii) as a system it is subject to the basic exigencies of any system. Such exigencies are analyzed terms of the general paradigm of basic functions of a system of action, i. e. the four system problems. (iv) Values are defined as independent of any specific level of structural differentiation of the system in which they are held. (v) Of any “level” each sub-system of the system that includes them will have its own values. These are “derived” from the more general value system. Included and including values stand to each other in a relationship of priority of relevance.

61

[Editor’s Note: This passage appears in Parsons, T. (1989:579). “A tentative outline of American values”. See Editor’s Note 60.]

226

H. Garfinkel

(2) A value system of a social system is by definition common or shared between the acting units whether they be individual personalities, roles, or collectivities. “The degree” of “sharedness” among all parts, that is to say among “classes” of the population of units of the system of action under review is a problematic point having its priority in the theorist’s concern with the characteristics of “cultural integration.” The difference between normatively possible events that are values of an actor’s environment, and those normative possibilities that are the actor’s ideology consists of the difference between the actor’s treatment of an environment of events with ideological features and an environment of events of value features. (1) In the case of values these features are generic, definitive characteristics of the most general category of that system’s features. (2) In the case of ideology these features are those of types of sub-systems of the main system of reference, and not of the system itself. Thus for example an actor may treat the normatively defined division of labor as a definition of the real society, in which case such a description of the society will be called the actor’s ideology. Parsons would insist that despite the ideological portrait of the society, the actor would nevertheless subscribe to the valued society and the theorist must consider that this is in fact the case if the question of the stable actual society is the question at issue. This would be Parsons’ version of the common phenomenon in which persons justify their actions in the name of an idealized portrait of the society while acting nevertheless with respect to the “real values” and in fulfillment of the conditions whereby stable social structures are possible. (See Parsons’ recent papers on culture and his Milan paper on the sociology of knowledge.)62 The value system of a sub-system of action can be “derived” from the norms of the higher order system, but not directly from it without referring to that higher order system’s situational features. Parsons insists that “at some level of motivation of individuals values imply commitment.” By commitment is meant the conditions of action that guarantee the actor’s compliance with the events in (E) as maxims of conduct. With respect to each of the areas of the functional problems of system maintenance, the conditions that guarantee compliance will vary. For example, consider the area of adaptation. Commitment would mean assigning others the right to interfere with, review, supervise the use of resources in the manipulation of a situation. Thus the decisions for the sanctionable use of resources is no longer under the unit’s sole purview. To say the unit is “committed” to system val62

[Editor’s Note: This would refer to Parsons, T. (1963). “On the concept of influence”, Public Opinion Quarterly 27(1):37–62.]

Parsons’ Primer

227

ues is to furnish the condition under which such an assignment of rights by the unit is, or can be, complied with. Within the adaptation cell, the four sub-system problems would be consulted for the conditions whose satisfaction would be defined [as] the meaning of the concept “stable commitment”. The reader’s attention is drawn to the way in which this notion looks into the concepts of values and situations known as common. It is seen too that the concept of commitment relates expectation to the “interplay of performances and sanctions.” The conventional distinction between “ideal” and “operative” values is handled by Parsons by consulting the property of the commitment. “Operative” values are “conceptions of the desirable”, to the actualization of which the unit is committed. “Ideal” values would have the properties that were enumerated before, of normative environmental possibilities, but would lack the unit’s commitment to the production of social structures on their behalf. Conformity and deviance are defined with respect to the concept of commitment. They involve by definition an explicit reference to motivated compliance with a legitimate order. (Insert a cross-reference here to the second theorem, i. e. that stable features of the real social structures are guaranteed by motivated compliance to a legitimate order). Not only values, but all other components of cultural systems are both institutionalized in social systems and internalized in personalities. When speaking of social systems “level of commitment” is referred to as institutionalized values. When speaking of personalities “level of commitment” is referred to as internalized values. In both cases reference is made to motivated compliance to a legitimate order. The category, value, describes a relationship between the actor and his behavioral environment. It is not a property or an attribute of either, and most particularly it cannot be clarified as to its empirical references, as Parsons means them, if one attempts to conceive the actor or his environment of action independently of each other. It needs to be pointed out again that an object is not a value. A value above all is misconceived if it is thought of in the fashion of a mustard plaster that is attached to an object. Value is instead a product of the process of evaluation. An environmental feature is a valued feature by reason of a rule of preference that the actor subscribes to as a maxim of conduct to govern his use of that feature. The feature is valued in terms of a set of rules that define “correct preferences.” When these rules of preferences are translated into maxims of conduct in the fashion that was indicated before, they describe the character of the relationship between

228

H. Garfinkel

the actor as a course of action, i. e. a course of evaluations, and the matters among which the preference is to be exercised. This describes the relationship. It will be seen, therefore, that in every statement referring to a pattern of preferences, i. e. “a value pattern”, a reference is always presupposed to the actor’s role. In every reference to value, such a reference is at least implicit and necessarily presupposed. A “value orientation”, i. e. a course of evaluation, is always stated from the point of view of an actor whose course of evaluation represents his “orientation” to an environment of possible events, i. e. to objects composing that situation. Such objects may include any of the categories of objects that Parsons uses in the systematic analysis of observable activities. These include social objects, cultural objects, personalities, organisms, physical environments, relationships, and the rest. A social value then may include the evaluation of all categories of objects. It is a matter of major importance for a grasp of Parsons’ concept of values that no objects by reason of whatever features they may exhibit are in principle exempt from treatment as values. The sole exempting condition consists of the theoretical possibility as Parsons furnishes these theoretical possibilities that their treatment as values is compatible or not with the conditions of stable social structures. This may be stated in another way: while any course of interpersonal actions are imaginable as a game, that not every imaginable game can in fact occur as a playable one. Although social values include the evaluation of all categories of objects, only evaluations of social objects are constitutive elements of the structure of a social system itself. Values function in this way as components of systems of interaction. By reason of the distinction between Ego and alter – what is from one point of interest a reference to an actor in a given social system is from another point of interest a situation/object. In order to avoid the difficulties of relativity, the concept of value must be formulated so that its identical sense is retained despite changes of such perspectives. That is to say, the concept of value must be formulated so that one is speaking of the same value whether it be held by Ego or alter. This condition is satisfied by the known-in-common feature of the behavioral environment of events and it is in terms of their known in common features that values are by definition objective. It is this feature that remains identical under variations of particular perspectival appearances of a situation or an acting unit, i. e. the so-called actor’s point of view. For the theorist, social values are a primary point of departure for any analysis of stable social structures and the processes of action whereby social structures are assembled. In the same way the type of value system furnishes

Parsons’ Primer

229

the theorist a principle in terms of which a classification of various types of social systems is generated. The theorist’s procedures in conducting an analysis and classification involve his use of two working assumptions: first, the presupposition that there exists a single, unified value system; and two, that the value system for a society tends to remain stable over a period of time. (These points need more explication, particularly with respect to the question of the different changes in social structures as compared changes in the value system. For example, structural differentiation of a system of interaction can occur without the paramount value system altering. However in the case of such differentiation the values of the included systems, i. e. “sub-systems” must necessarily involve change).

IX THE PROPERTIES OF VALUES EXPLICATED WITH THE USE OF BASIC RULES OF GAMES63 In deciding the character of values for a system of activities it was proposed that some sub-set of the normative related possible events of the actor’s behavioral environment are going to be treated by actors as preferred. “Preferred” is a euphemism. Like the term desirable, it refers to the fact that with respect to the alternative ways in which events can occur, one set will require the actor’s compliance (a) as a condition of his grasp of a real environment; (b) of his right to claim competence to treat that environment without interference from others; and (c) to engage in stable concerted courses of interaction with others. Parsons summarizes this by speaking of the choices that are required of the actor as an “incumbent of a role.” Required choices among alternatives are correct or sanctionable choices. Correct consists of rules of choice that define correct choices or correct preference. These can be moral, ethical, aesthetic, technical, etc. rules. The alternative that the rules of preference define as correct preference is the valued alternative, or “the value.” Where the rules are themselves clear, the choice that the rule provides is clear. However, this is to be distinguished from the point that the thing that is to be preferred, that “is desirable” may be depicted through a rule that is clear in meaning something vague or clear, specific or indefinite, and so on. Parsons employs the pattern variables to handle these 63

Some of the material on the next 2–3 pages repeats verbatim earlier portions of text. However, we have elected to leave it as it was in the original manuscript, rather than guessing at what Garfinkel would have decided to remove or revise.

230

H. Garfinkel

properties. The election of one among alternatives is the important point that is being referred to. Values’ alternatives, by reason of being defined through rules of preferred choices, are inherently comparable. That there are desirable states for an actor still does not make the difference between the sub-set of (E) that the theorist is going to refer to as values, and other sub-sets of (E) that Parsons refers to as norms. A feature that makes an important difference is that the patterns of normative culture have for the actor the feature that they are shared with others and are treated by the actor as shared with interactional partners. It is incorrect usage to say that values are shared and mean thereby that values, as Parsons uses the concept can be defined independently of the existence of “shared conceptions of the desirable.” Shared conceptions of the desirable are called values. The problem is to find in the set of activities the conceptions that are shared, and, says Parsons, if the activities show stable features, such shared “conceptions of the desirable” must exist, i. e. the likelihood that compliance with them is enforced is high, and this empirical feature of the action is related to the stable properties of social structure as a condition of these stable properties. The feature that finally makes the difference, is that whatever they might consist of content-wise, they are shared in common by incumbents of different roles. That is to say, if you have two persons in a situation of interaction with each expecting performances of himself and of the other, which differ in the character of their contributions to the outcome of their concerted action – an example would be the coordinated action of the husband and wife in a family whereby accomplished expenditures of the family income is a produced feature of their concerted purchasing decisions – certain states of the possible accomplished expenditures as desirable states, that each of the interacting parties treats as known to and binding upon both in a more or less equivalent way, would be values of the interacting pair. Parsons proposes that where there is no difference in the roles that incumbents play, no distinction between values and norms can or needs to be employed. Where there is a distinction, then it is necessary to distinguish for each the normative environments that are oriented by each, as well as the normative environment that each complies with, and expects the other to comply with, in a more or less equivalent way; the latter being necessary in order to bring their differentiated understandings under the jurisdiction of a set of common understandings about the conditions and consequences, as the interacting parties understand these conditions and consequences, of their concerted action. Thus each “knows” not only what his “part” is but he “knows” also what is sanctionable by way of desirable structural consequences of their actions “in common.” It is the latter structural effect of their actions in common that each actor orients as a system of performances

Parsons’ Primer

231

and sanctions in terms of which the desirable states of the system, i. e. the values, consist. The distinction between the observable events that correspond to Parsons’ terms “values” and “norms” is grammatically clear and obvious when the researcher deals with settings in which divisions by work, age, sex, religion, ethnicity, territorial jurisdiction, and the rest are prominent. These would be found in settings such as industrial establishments and cities, though obviously not exclusively there. In Parsons’ usage values have properties that are almost identical with the normative states of a game that the basic rules of a game provide for. It may be helpful to the reader to think of values in this way. It means their analogous character can be pushed quite far before the analogy begins to raise difficulties. In his paper, Societal Value Systems, Parsons in fact referred to games in order to introduce his discussion of values by likening values to “basic rules” of a game. It needs to be stressed, however, that in using the notion of the basic rules of a game, Parsons’ usage is figurative. Therefore, the reader should not try to make too much of the term “basic rules” but should rather attend the definitive features of normative environment in order to grasp the empirical references of values in any literal way. With this reservation the following are properties of the states of a game that basic rules of play provide, properties that stand in close analogous fashion to the properties of normative environments of interactional events that Parsons intends. Interestingly, even if the theorist deals with large-scale, complex sets of activities, values retain properties analogous to those of basic rules of a game. One result of this for the purposes of criticism, is that Parsons’ social system has very much the features of a table of organization. Later we shall remark further on how this is so, and remark on its research consequences, both good and bad. What are these correspondences? First, basic rules like values, for the player who complies with them, are used by the player to define situations of possible game events; situations that are known in common with other players. The lesson we can draw from this analogy is that is it is via the conception of values that Parsons provides the possibility that situations of interactional events acquire the feature for actors that they are known in common with other actors. Second. Displays of actual behavior that persons will generate for each other have the character of a continually changing stimulus array. Parsons holds that under the condition that a person must respond to the immediate course of the changing display, stable treatments of those displays are not possible. Parsons invokes a feature of normative environments to solve what he calls the “problem of continuity in action.” Some normative environmental events will

232

H. Garfinkel

serve to overcome for the actors the particular here-and-now specific character of the immediate stimulus array in favor of permitting the actor to respond to features in the changing array that remain invariant to the continually changing appearances. Basic rules, like values, permit the theorist to say how the player avoids being stimulus-bound. The particularity of actually presented characteristics of actual stimuli are “sacrificed” to their “common sense.” The lesson: values provide for the actor a means for recognizing and responding to what Parsons calls the “generalized stimulus character” of the particular actual action that the actor witnesses. In his theory of socialization Parsons is much concerned with the conditions under which “stimulus generalization” occurs, as well as with what “generalizations” occur inasmuch as such processes describe how actions are brought through a course of training under the jurisdiction of a “common normative order.” A third comparison. Basic rules of the game, like values, are used by the player to recognize the occurrence of events of the player’s and of a player’s partner’s conduct. Further, the basic rules, like values, are used by the player to define his own and other players’ witnessed actions “universalistically” – i. e. as instances of typical actions that are provided by a classification of possible types of actions that each actor exhibits and expects the other to exhibit in a more or less identical way. Thus, to each set of basic rules of a game there corresponds a paradigmatic class of possible events that compose the environment of game-possible events. The identical thing is true for values with respect to environments of possible events of everyday life. Further, just as the basic rules of play constitute a game environment as the players’ real environment of play, values constitute the actor’s environment as his real environment of possible events of everyday life. The parallelism holds further. Just as a “new game” is constituted by altering the basic rules of the game, so too is a “new environment” constituted by altering values. If one thinks of the basic rules of the game as defining the social reality toward which player’s actions are directed, then an alteration of the basic rules reconstitutes that social reality. The identical thing holds for values. Just as it holds for the game player, it holds for Parsons’ actor that the actor’s real environment is constituted by values and its stable features, and thereby the actor’s grasp of his real environment is insured by the actor’s commitment, i. e. the condition that guarantees that his actions are taken in compliance with values. A fourth correspondence. The features of the game that basic rules provide hold throughout changes in actual present states of the game that players’ moves accomplish. Compliance with the basic rules is not conditional upon the specific situation in which the basic rules are presupposed, or operate. Nor are the uses of basic rules contingent upon the courses of actual moves, of

Parsons’ Primer

233

players’ assessments of advantages in their compliance, or personal dispensations. Parsons refers to identical features by saying that values are not “specific to the situation.” Their validity, i. e. the likelihood that compliance will be enforced – is not a function of the particular categories of personalities who are available as partners to an interaction. Similarly, the actor’s normative environment contains a sub-set of possible events with regard to which, if the theorist is to say that actions are stable or can be stable, there is a great likelihood that conformity to what they propose, as the desirable states of the system, will be enforced. When Parsons proposes that values are not situation specific, he means that the probability of their enforcement does not depend on the here-and-now features of situations. When considerations of efficacy are introduced, so that the likelihood that conformity will be enforced, to the production of desirable states, is for actors a function of here-and-now considerations, then Parsons proposes to talk of normative environmental events as goals instead of as values. Does this mean that values are “eternal” goals? i. e. that they do not change? The answer is that, from the point of view of the actor, states are called values, if for the actor they exhibit the perceived property which the actor treats as unchanging, and to which the actor seeks to comply regardless of circumstances. Only if the actor’s behavioral environment contains such possible events is stable action possible. (Cite here Marion Levy and his conception of tradition.) A fifth comparison between basic rules and values may facilitate the exposition of another feature of value. In a particularly obscure passage Parsons proposes a most important feature. He writes, “The term value is reserved for those normative patterns which are in the highest level of generality which is relevant at all in the orientation of action in the social system of relevance.”64 Parsons is saying the following. Consider four persons who are playing poker and think of them for the moment as a collectivity, which is to say, think of them as persons whose performances in the game are regulated by the basic rules of play and define their own identities within the exhibited actions of play by consulting the basic rules of poker with which to define proper poker performance and “membership in the game”. If the basic rules also specifically identify or name the related set of members and performances as a poker game then the values of a poker game are those depictions of that state of affairs which describe such a set of activities for the actors as constitutive of their real circumstances. At this point Parsons is no longer talking about basic rules and values as if they were synonymous. He is talking about the 64

[Editor’s Note: This quote appears in Parsons, T. (1989:579). “A tentative outline of American values”. See Editor’s Note 60.]

234

H. Garfinkel

normative features, which in a definitive manner describe for actors the system of transaction, without respect for considerations of particular contributions of the members and that the actors presume to be binding on all the members of the kind of group in which they are participating, as well as the kind of persons that each of them is, as well as the kind of persons all of them are in common within-that-group. In a word, “we.” Clearly Parsons is talking here of a society as a “natural” i. e. a moral order from the standpoint of an actor who is simultaneously a member of that order and who experiences it as the arena of his life as well as an observer of that order as an object in the world. Obviously the analogy of a game is an inadequate one with respect to this essential feature of group governed action. Parsons is providing here the condition that an observer’s investigation would have to satisfy if the observer is to locate the existence of collectivity-governed action. Parsons is saying that such actions are not locatable without consulting the particular features of values that from the point of view of the actor’s normative environment correspond to rules of relevance that the actor complies with in conducting his everyday affairs in concert with others “like him.” A different set of conditions are provided by other “levels of generality”, in which “normative patterns are relevant to the orientation of action in the social system of relevance” – e. g., “levels” like role-relationships, personality, social system as analytic entities, and the rest. If the investigator wants to conceive and to describe an actual poker game as a collectivity as Parsons employs this concept, then there must exist for the observed poker activities some set of normative conceptions that the actual players comply with, that define for each the “good game”, and which are complied with irrespective of the particular qualifications of the particular features of the players, the plans that particular players might entertain in playing, the advantages of compliance, and the rest. That there are a set of persons who will respond to an environment that has such a feature is a matter of required relevance for the sociological observer who seeks to use Parsons’ concept of collectivity, because this feature will be used by the investigator to assign to particular features of the poker game their sense as collectivitygoverned events. Obviously, the investigator might prefer as an alternative method to consider to poker players as an aggregation of economic men and to leave Parsons’ social system considerations aside. In that case, the “good game” would not be an essential feature of the matter he intends to, or is required to, describe. Because “the good game” is relevant to “poker players”, it would have to be relevant to the observer if he wants to relate patterns of actual play to an order which “governs” the production of these patterns of play.

Parsons’ Primer

235

To summarize: a value system by definition consists of known-in-common desirable ways and desirable features of action to which compliance is enforced by the “acting units” of the system. Shared desirable states of the system of action are called values. We have seen that this phrase is not to be taken literally as a definition. It consists instead of a highly formularized way of talking about the definition. The term “system” in the formula reflects upon the meaning of “shared” so that system holds as a general concept whether one is talking about personalities, roles, or collectivities as actors. Parsons in fact proposes that the investigator will find value features in the environments of organisms. This would mean that an organism would respond to its environment as consisting of a set of conditions under which the organism would take corrective or restorative action on the occasion of discrepancies between the standardized features of that environment and an actual course of events. The organism would be said to have a “culture” in the sense in which Boulding65 in The Image says of the paramecium that it is coded to select from its environment certain signals to which it will respond and to take corrective action when, for example, a discrepancy occurs between certain determinations of salinity and certain standards of salinity that it will “accept” before it moves away or takes some other corrective behavior. Parsons proposes in fact that to each organism over the entire range of organisms a cultural environment corresponds. It happens that Parsons is particularly interested in the cultural environments of human organisms. One can take a further step and propose the reasonableness of applying the concept of the valued features of an environment to the environment of mechanical systems as well, so that it would make perfectly good sense to talk about the culture of the thermostat. The notion of values is not specific to a particular system. One can refer to environmental features of organism; personality; roles; collectivity; functional sub-system; and can refer to any of the ways which Parsons has of analyzing these into “sub-systems.” Thus far we have been able to say that actors share with others, ways of treating an environment, such that, not only is it the case that that environment is treated, but the products of the treatment, of activities, have properties of values, and that actors act in compliance with normative possibilities. But Parsons wants to talk about actors – for the moment let’s talk of them as persons – who are doing something more than merely “believing in values.” When we come to describe values, the problem arises as to the empirically appropriate 65 [Editor’s Note: Boulding, KE. (1956). The image: Knowledge in life and society. MI: University of Michigan Press.]

236

H. Garfinkel

way of talking about how persons “believe in” or “subscribe to” a legitimate order. Shall we be describing appropriate events if we say of chess players that they “subscribe” to the rules of alternating moves and seek to demonstrate that this is so by asking chess players to check off for us how strongly they agree that alternating play is correct play? For Parsons something more is meant on the empirical side. For Parsons, compliance is motivated compliance. Parsons uses the term, “commitment” to values to refer to the empirical features of motivated compliance. Motivated compliance, not “belief in values” is the critical empirical phenomenon. (List and discuss briefly various procedures that would yield determinations of the various definitive features of values that have been discussed above. Consult the note below based on the manuscript that I wrote in response to Parsons’ paper, The Societal Value System and delivered in the seminar of the sociology of law.)66

X VALUE STABILITY AND AN ANSWER TO SOME CRITICISMS With the use of the concept of values Parsons is trying to establish a set of conditions under which the theorist is required to infer the stability of a set of activities as a necessary inference. Parsons provides in this respect for the theoretical necessity of the phenomenon of commitment. He talks of commitment at the social system level by referring to the institutionalized character of valued practices. Parsons has to get actors producing their actions for the theorist in a related way and get them “into the system” in such a way that when his actors exercise choices (a) they exercise legitimate choices, and (b) these legitimate choices are the ones that actors are required to make if the system of related choices are to produce a stable set of interactions. Parsons’ actors must make choices that will produce for the theorist a description of social structures whose features will remain stable over time and through the turnover of the incumbents in the positions of acting agents. The criticism is sometimes made of Parsons’ member of a social system that there is never a situation in which he acts in which he is not accountable. The organization it is said consists by definition of members who do nothing except that they do it under the auspices of value-commitments, that is to say under the auspices of the sanctioned character of desirable states of the system that hold over all occasions of possible interaction.

66

[Editor’s Note: We are not sure which “response paper” Garfinkel is referencing here.]

Parsons’ Primer

237

The criticism would stand were not for two important provisions that Parsons is very explicit about. One is the provision for non-contractual elements of contracts. The second is the provision for the “exigencies” of a system of action. If the criticism were correct and Parsons’ members were to act in the fashion that the criticism depicts, Parsons could hardly provide for change. Once the set of activities got under way they would remain with their identical features since actors conceived in the manner of the criticism would be directed only to the maintenance of that system’s features. To get such a system to change it would not be possible under this version of the member, to invoke anything about the motivated character of his action that would say something more than has hitherto been provided about the characteristics of motivated action. When Parsons talks about a value system he says that motivated compliance is one feature of a system of interaction. There are also the actual events of interaction. Because there are these actual events, Parsons has provided the possibility that the actor’s situation may deal him surprises with varying degrees of nastiness. Thus action by its very character, according to Parsons, is directed to a situation which has contingent character. Only very rarely, as for example, in rigorously controlled ceremonials, and perhaps in games can the actor tie up all the ends with a set of certified checks, which he would have to do by providing by prior agreement, that all the possible ways in which in fact events can occur, have been accounted for, and that the actor has a prior agreement to comply with certain sets of these, and complies with them. Even at this, the contingency remains as to whether or not the actor will have succeeded, even if he holds the agreement without question, in complying with the agreement. The question of commitment to values comes in precisely at this point: how is the theorist to get his actors to fulfill agreements, that is, to conform to valued practices, given the fact that there are going to be discrepancies between what should occur and what in fact life will pay them out. Parsons central and abiding concern for the “exigencies” of a system of action, is a concern for exactly this problem. Think of the exigencies as known to and treated by the sociologist as an order of nature. The whole phenomenon of “deviant from” comes precisely to this: that persons cannot count on any guarantees that restorative action will establish the system as the actor requires it.

238

H. Garfinkel

XI VALUES VERSUS SLOGANS If the researcher attends to those definitive features of a normative environment in describing values there is little danger that he will be describing persons’ words or slogans. Indeed, a description that is accomplished with respect to the definitive features would be far removed in the events it described from a concern with words and slogans. One need only take the very simple case of the desired states defined in the game of chess, by the rule of alternating play, to see how difficult it is to describe even this apparently uncomplicated state in words. In fact, if Parsons’ conception is used and, probably to the extent that persons are committed to certain desirable states of the system, they cannot tell the investigator too much about the valued states to the production of which they are committed. In order to learn on behalf of Parsons’ actors the desirable states of the system that they are committed to, the investigator has to learn how to treat the environment in which they operate, with these environments as well as their treatments of them being conceived in accordance with Parsons’ theory of the social system. Therefore, it is not at all an appropriate way to learn about Parsons’ actors’ values by administering persons a questionnaire. If a questionnaire is used it would have to present persons with those statements to the comment that Parsons provides the real state of the system that the person “lives in.” (Furnish examples from Sociology 120) In no case would a procedure for determining operative values be an adequate one if it depended upon persons telling the investigator what they think is valuable. Using Parsons’ conception, the investigator would have to proceed under the rule that persons do not know their real values, and most particularly is this the case where commitments to their production are institutionalized. Most particularly must the investigator treat persons as not knowing their real values if the sociologists must depend upon these persons to do the job for him of deciding what these real values are.

XII VALUES ARE NOT INFERRED FROM ACTIONS Values are no more “inferred” from actions than are the basic rules of chess and the desirable states of the game they provide for inferred from the play of chess players. Nor are values equivalent as occurrences nor do they necessarily consist of actual performances. Indeed if they had to consist of actual performances, the system would not be recognizable unless persons were do-

239

Parsons’ Primer

ing what the normative order provided and only what the normative order provided. Certainly, this is rarely the case. Even in games there are sufficient discrepancies between the two to make the question an interesting one. The investigator, if he treats the two as the same, would only be able to talk of a system of interactions as long as persons were engaged in active movements. When they were silent, the investigator would be up against the problem of whether or not to call the thing that they were “not doing” an action. Parsons provides for a system of reciprocal actual interactions that are specifically governed in their course and in the meanings that persons will assign them by the possibility of corrective action and by the persons’ anticipations of the concerted course of interactional affairs. Therefore from the standpoint of Ego, what is for Ego an expected action, an expected performance, an anticipated performance on his part, is, in Ego’s anticipated response from the other, the sanction of that expectancy.

XIII VALUES CONTROL NORMS When Parsons talks about norms he is tying the desirable states of a system to an allocation of commitments on the parts of actors to the production of these desirable states. This is the key, and explains Parsons’ usage when he says that norms are “function specific”, and are “specific to situations.” Parsons goes on to say that a system of values controls a system of norms and means by this that in the activities of actors, commitments to values have a prior relevance for the actor in his task of defining the mutual situation of interaction and in defining the proper allocation of rights to invoke norms. To say that values have priority is another way of saying that the question of legitimacy is settled in the first instance by invoking the relevant system of values that the actor presumes to be binding on both parties. So, for example, if an actor has to justify his action, that is, to justify his contribution to the consequences of actions that concern himself and his interactional partner, i. e. to give “good reasons” for his action to furnish them their “reasonableness” he will invoke those desirable states of the system that define what the procedural norm that “governs” his course of action should have produced for both of them.

240

H. Garfinkel

XIV “LEVEL” OF COMMITMENT TO VALUES Parsons proposes that the phenomena of “conformity” and “deviance” are definable only with respect to the phenomena of commitment. That is to say, the investigator cannot clearly delineate the phenomena of deviant conduct by invoking only a departure from a normative order. It is necessary as well to make reference to the notion of a departure from motivated compliance to a normative order. The key notion is “level of commitment.” A “level of commitment” is nothing else than the probability that motivated compliance to the production of desirable states of the system i. e. “motivated compliance to the actualization of norms and values,” will be enforced via what Parsons calls the “mechanism of institutional control.” The level of commitment is not a quantitative notion. Only in a metaphorical way is reference made to a “matter of degree.” This is so even though Parsons frequently talks about “institutionalization in degree” and of “more or less” institutionalization. This will be clearer after the discussion of system relationships. There we shall see that this is not at all a quantitative notion. For one thing there is nothing more than a figurative meaning if not a bromide in talking about a “continuum.” Where one is talking of social systems, “level of commitment” refers to the procedures whereby commitments are insured via the mechanism of social control. In talking about personalities as sets of activities, level of commitment refers to what Parsons calls “mechanisms of internalization.” The meanings of internalization are drawn very heavily by Parsons from the work of Freud. Strong stress is placed upon the relationships between id, ego, and superego functions. In both cases, the reference is to motivated compliance to a legitimate order which is retained throughout as a central theme.

XV VALUES ARE NOT “IN THE HEAD” Parsons’ usage of the term values contrasts with conventional usage insofar as conventional usage provides for persons “having” values or for values being “carried around” by persons from place to place and from time to time. Conventional usage frequently provides for values as if they were engrams lurking in the nervous system or found under the skull somewhere. Such usage does not describe Parsons’ usage. One needs only to consider again the conditions

Parsons’ Primer

241

under which an adequate description is possible of the desirable states of a game that are provided for by the basic rules of a game. The methods for their adequate description are those methods for determining the properties of events in a behavioral environment. Values are nowhere but in the actor’s behavioral environment. Nothing is found in the head but brains. If we talk about a business as a viable organization and say that persons “are in business”, we mean among other things, we are talking among other things, of the likelihood that they can enforce upon each other a respect for the fulfillment of certain obligated treatments that each is assumed to owe to the others in some typical situations of interaction. For example, the worker who says to the employer, “It is Saturday night, I have worked, pay me,” by reason of this describes for the sociological observer a feature of the “viable organization.” Its viability consists in the probability that such a claim will be honored. Then what is meant if we say that the business “goes into bankruptcy?” By this is meant that the validity of this normative order is “lost,” i. e. the probability that a person extending the claim to another, one in the role of worker and the other as employer, for payment for the week’s work, even though he worked, is negligible. To talk about the dissolution of the system is to talk of nothing else than the fact that there are now certain patterns of normatively regulated action the probability of whose enforcement is very small. It means that persons are no longer going to “actualize the system of values.” It is as simple as that. Values therefore do not refer to engrams in the head, even though it would appear at times that this is how we think about them, or that “in the final analysis” they are after all in the head. The erroneous nature of this conception cannot be overstressed. In Parsons’ work values are a theorist’s device. Values are not what members are concerned with. A value is a sociologist’s term, a method. “Values” is a way in which Parsons talks about the way in which persons can require of and enforce upon each other the production of certain kinds of social structures. Therefore, if one says that values are “lost” or that the system has “dissolved”, or that the system has “lost its boundaries”, or that “boundaries have dissolved”, or that the system has “changed” we mean by this that one of the changes that can occur is that the properties of the persons’ actions that were previously described no longer are there to be described. The rule therefore is that one must not go looking for values by asking persons to permit one to insert electrodes into their brains to find out what they “really have in mind,” because what they really have in mind is nowhere else to be found than in the observable treatments that they are giving and getting from each other.

242

H. Garfinkel

XVI VALUES ARE THE SUBJECTIVE CATEGORIES Values cannot be recognized and described except by taking the subjective categories into account. It is necessary for the theorist who uses Parsons’ conception of values to invoke an actor (not a person) and a way in which an environment appears to an actor, what an actor sees an environment to consist of as actual appearances. This may be considered from another angle. Consider the question, what is meant by the real social structures? The real social structures consist of a description of these activities in accordance with the sociological theorist’s ways of conceiving them and in the ways that are acceptable to him for describing them. The real social structures are nothing else than that. It can be stated that flatly. “Real” is not an ontological notion. “Real” is what the investigator gets if he analyzes actions with the use of methods that the sociologists actually use, whatever are the methods that they actually use. Call these methods generally the sociologist’s point of view. They consist of any methods at all, as for example those of Stephan and Mishler, Bavelas, Hemphill, Bales, and the rest.67

XVII REAL PERCEIVED SOCIAL STRUCTURES Distinguish between perceived social structures, naive perceived social structures, and real perceived social structures. The phenomena intended by the term “perceived social structures” are in no way the discovery of the exclusive property of the social sciences. Every lexicon wheresoever it is found provides its users a way of talking about how situations may present their appearances to the other person. Every location provides its users, as members of a society, and as a condition of their right to bona fide membership, an appreciation of an appropriate use of the ways of relating the user’s knowledge of his environment to the phenomenon of perspectival appearances and perspectival view. Matters referred to under such usage would be called generally “perceived social structures”, insofar as the matters referred to are the phenomena of social actions. By tacking the term “naïve” onto the phrase, so that we have “naïve perceived social structures”, attention is called methodologically speaking to the fact that the sociologist’s point of view has been incorporated into 67

[Editor’s Note: these researchers all worked on group processes. See Editor’s Notes 45–47 for details.]

Parsons’ Primer

243

the meanings of the perceived social structures, by identifying the sociologist who treats these matters, while subscribing to the perspectival interpretations of the other person as a member of the society, that is for the sociologist both and simultaneously an object for description as well as an arena of his practical activity. The sociologist proceeding in this fashion might then say that a subject proposed that there were ghosts in the graveyard; just so, there are ghosts in the graveyard. It is not necessary to question the subject any further; it is not required that any sociological criticism of the subject’s description be made. Whatever the other person says is so, is so, i. e. “let it be so.” It is entirely possible to accomplish the description of the perceived social structures in exactly this fashion, and to count it an appropriate sociological description. This is said because frequently sociologists will proceed in exactly this fashion and claim this procedure as a correct one. But, although the sociologist may collect these naive structures, ordinarily he does not stop with that, but instead subjects these materials to a transformation of sense, with the result that the real social structures incorporate these naive perceived structures, or rather transform them under the auspices of a procedure whereby “what the other person really perceives” is defined. The criticized naive perceived structures are described, that is to say, from the standpoint of the sociologist’s actor. This produces a different story than, and procedurally contrasts with, the former one. Thus, for example, a psychologist in a department of psychiatry may complain that the psychiatrists get the gravy while the psychologists get next best; that the psychiatrists spend their time talking and making money while the psychologists spend their time trying to think a fresh thought and not making any money at all. The observer may come then to refer to this and say of the psychologist that he was really talking about the division of work, and of the correspondences between allocations of work and responsibility on the one hand, and allocations of prestigeful income on the other. The psychologist might never have used the terms “division of work”, “prestigeful income”, and the rest, but the sociologist in describing what the psychologist was talking about conceived the social structure that he described in these terms. Parsons does exactly the same thing, but does it with a kind of vengeance. I use this exaggeration in order to stress that Parsons provides in the functional problems of a social system a way of describing what the real meanings of perceived environments or perceived social structures can consist of for any actor. Empirically speaking this means that for any person whatsoever no matter what stories he has to tell, his stories can be transformed in sense, in accordance with the transforming rules provided for in the social system, to describe what he is “really talking about”. Thus, for example, if the subject

244

H. Garfinkel

furnishes the investigator a long story in the subject’s own vernacular about risks he must manage for the successful accomplishment of his job, and describes how these risks on the job are managed, the investigator would say that he conceived subject’s situation with respect to the parameter of adaptation.

XVIII EMPIRICAL OBSERVABLES INTENDED BY THE TERM VALUES The empirically observable events that are intended by the term values are encountered by the investigator by asking what the likelihood is that certain states of activity are in fact going to be produced via the concerted actual action of persons. The empirical reference is as direct and as simple as that. Thus the terms values and norms do not go to any other point, than to the point of the kind of activities or treatments of what objects in their environments are, as matters of persons’ “real interests”. Again the example of a poker game may help. Suppose the investigator seeks to make a description of the environments of poker players. The term player is used to refer to any person who seeks to act in conformity with whatever the basic rules of poker are that he has decided to play with his partner. Let me describe the normative environment of possible occurrences for a player in this game. I consult the rule book. Here is a first rule: players will move alternately. This sequence can now be set up and I say that this rule is a description of what the players are going to treat as a sanctionable procedure or a sanctionable sequence of modes of players, themselves included. Next it is necessary to find out whether this describes their behavioral environment. What is to be done? If this rule has the property of the rule of play, then the breach of it should produce corrective action. This is elementary, but it is by definition an important property of the treated environment. The environmental possibility of serial moves has the property too of a standard. It has the property of assigning to the possible state of affairs their typical features. It also has the property that it is presumably binding on persons irrespective of the function-specific or situation-specific character of the changing actual present states of the game, and holds regardless of advantages that might accrue or be lost to players by reason of adhering to the rule. What happens then if this rule is said to go to pot? Consider a game of tic tac toe. Invite a person to make a move. After he makes a move you erase his mark, move it to another cell, and make your mark. One will frequently encounter with adults and will almost invariably encounter with children that this other persons says, “Oh, no! If you’re going to do that, I’ll do this” (he completes three

Parsons’ Primer

245

in a row) “I win!” Let’s suppose then if you say to him “Let’s play again” and he replies “go away. I don’t want to play with you. You cheat.” Now let us suppose that in response to your invitation your partner “pretends” to be playing the game, that he is not “in the game”, that he is “only going through the motions”, what is meant observationally by saying that while he makes a mark he is “really not playing.” If I make a move, and he says, “Look here, it doesn’t make any difference,” even if we both put marks down according to the rule what is he “really” saying? We say that he is saying that he is “not really playing”; that he is “out of the game”, and that means observationally that an order of obligated productions are no longer enforceable. Again it is as simple as that. It is also that simple that the investigator has this test to make to see if this state of affairs is to be described as “the fact of the case”, as compared with treating it as a possibility or as an investigator’s conjecture. Suppose further that you urge your partner, “Play the game whether you like it or not.” “Oh well,” he replies, “I’ll make the mark since you won’t be satisfied unless I do.” But you observe that although he might make the mark he is “not committed”.

XIX THE EMPIRICAL MEANING OF “VALUES DISAPPEAR” To say that values “disappear” is identical, synonymous, and interchangeable usage with the formulation that values are not operative. The “existence of values” is identical, synonymous, and interchangeable with the validity of values. The “existence” of values consists entirely in the probability that actions will be governed by them i. e. that compliance to them is enforceable.

XX WHAT IT MEANS TO TALK ABOUT VALUES “IN THE ACTION FRAME” Talk about the ways in which said activities are organized means exactly the same thing as talking about how a set of social structures are being produced and maintained. This was introduced before when we spoke about the real social structures as consisting of institutionalized patterns of normative culture. Both formulations have exactly the same meaning. It has been discussed in another fashion by pointing out that the social system as a set of observable occurrences consists of activities directed to the treatment of objects. Thus it

246

H. Garfinkel

makes no sense to talk about values as Parsons intends the term unless one puts it in the “action frame.” Then what would be meant by talking of values “in the action frame”? It means that one talks about persons who are acting in conformity to, or are committed to, or who engage in, correcting each other with respect to, and thereby produce via their actual actions, the social structures that the sociologist encounters by means of the theoretical apparatus he employs to collect his thoughts about the nature of the events he wishes to study and their related character. Therefore it is not “concrete action” that the sociologist is concerned with, but the concrete action that appears upon analysis according to these conceptions. This means therefore that any “concrete action” can be analyzed from a variety of different perspectives, i. e. with a variety of different conceptions. Certainly this thought is an obvious one. The interesting questions arise, as they must inevitably, in specifying the different conceptions as orderly procedures of conceptions, analysis, and description. Throughout there remains the troublesome point that persons do not act and do not have to act the way Parsons says they do; only actors do, Parsons’ actors. That there is this distinction between these two states of affairs multiplies the difficulties in research, of the researcher’s being correct when he offers an empirical description of a society. Parsons is no more and certainly no less involved with these difficulties than are other sociologists.

XXI GOALS Any state of an (E) may be treated by actors as the intended product of his own or concerted actions with others. Thus any state of an (E) can be a project for an actor or actors. When actions are related to their outcomes by actors under a scheme of means-ends relationships entertained by actors to which they seek to conform as maxims of their conduct we call attention to this by speaking of the relevant state of (E) as the actor’s “goal.”68

68 [Editor’s Note: Garfinkel’s also used an “E” operator in his theory of information, which he wrote during his time as a postdoc at Princeton working with Wilbert Moore. Cf. Garfinkel, H. and A. W. Rawls. ([1952]2008). Toward a sociological theory of information. London: Routledge.]

Parsons’ Primer

247

XXII VALUES FURNISH NORMS THEIR “RATIONALE OF LEGITIMACY”: NORMS FURNISH VALUES THE PROGRAMS FOR THEIR PRODUCTION Whereas values are defined in terms of rules of preference, norms are not. Instead they consist of rules of procedure. However procedures may have the status of preferred procedures, i. e. valued procedures, and hence will themselves be governed by rules of preference. In such a case it will be seen that values furnish procedures their “rationale of legitimacy.” Rules of preference furnish rules of procedure a rationale of correctness with respect to the four functional requisites of a determinate rule, i. e. the four parameters of legitimacy or correctness in a rule of procedure. These parameters are: (1) That among competing alternatives the fact of its selection itself furnishes the warrant for the one that is selected (i. e. the meaning of correct in and for itself ); (2) That among competing alternatives the selected procedure is compatible with the others that form the program for assembling the valued state. (When referring to a program as a set of propositions, the rule of preference provides the definition of meaningful consistency. When referring to a program as a normative course of action the rule of preference will define “workmanship”, i. e. in the sense that Kotarbinski69 talks of the style of a method as one parameter of “good work.”) (3) That among competing alternatives the selected procedure is compatible with the coordinated pacing, duration, and phasing of the steps comprising the programmed assembly. (4) That among competing alternatives the selected procedure is an instrumentally efficacious one with respect to the accomplished assembly. The value may define any of these as the sanctionable rule of relevance. Call this feature of sanctioned relevance the “primacy” of the value. If values stand to norms as the rationale for their correctness, norms stand to values as the programs for their production, i. e. for assembling the state as the valued product of a possible course of action. Obviously it is possible to describe a valued state without respect to its program. For example, the conjugal feature of the American middle class family, a valued state of this system, can be seen “in a single glance” and its features described without respect to any program for producing it. Similarly, a theorem can be grasped without knowing the steps in its discovery or derivation. Similarly one can halt in the course of one’s labors to consider “what has been 69

[Editor’s Note: Probably referring to: Kotarbiński, T. (1960). “The concept of action”, The Journal of Philosophy 57 (7):215–222.]

248

H. Garfinkel

accomplished” the adequate grasp of which does not necessarily require that one consider how the “net result” was produced. This is not possible however under the condition that the valued state is in the path of the action, i. e. valued states that govern action in its course. From the point of view of the members of the family in the course of “living in” their relationships with extended kin for the benefit of the sociologist, the conjugal feature is the managed product of step-wise work of claiming and enforcing the priority of their rights over the rights of kin to manage the household’s affairs. Similarly, the theorem for the theorist who lives in the action of discovering it, or the sanctionable employment decision for the employer who has to decide between the relevance of technical competence versus common fraternity membership, are the accomplished products of the steps whereby the products are produced. From the point of view of the sociologists’ actor, in the path of his action, value and program are inseparable. Nevertheless, there are some valued states the programs for which are such that the theorist can disregard the programs for the case of action in its course, for example, an additive object like a truck loaded with gravel, which is the product of a gang of shovelers, each of whom takes a bit off a pile and heaves it into the truck. If the theorist relaxes the condition of a time limit, one could say that because of the programmed structure of the object, all roads lead to Rome, and thus, in the rigorous sense, the assembly is independent of the paths to the assembly. Let any hint of timing come into the matter, and one would have to wait for the program to run its course in order for the value to be definite in its final structure. One could of course get a small run along the path, extrapolate the rest, and be done with it. The difficulty consists in the fact, that the domain of valued states that we need to contend with in our sociological theorizing, is not exhausted by values with a simple additive systematic structure, values with a repetitive serial structure, values with a serial structure and a definite terminal step. A most important, extensively occurring, and exceedingly troublesome additional set of them consists of a “single continuous run” of values whose programs essentially lack any provision for a terminal step. In such a case, the value can be grasped only by recapitulating the program that has developed “as of now” while referring simultaneously to “what a future will have brought.” Such an “as of now” future, consists of a set of open and more or less empty anticipations, which function in the Here and Now to lend a provisional character to the accomplished program. The anticipation provides for unknown anticipations, the fulfillment of which will clarify the import of what has transpired in the past. Thus the program provides for a succession of Here and Nows, each carrying along its history and

249

Parsons’ Primer

prospects. The corresponding valued state, consists of a state that has precisely this “it remains to be seen later what it really will have been all along” property.

XXIII THE TWO THEOREMS Any use of Parsons that ignores these theorems as fundamental interpretive rules, results either in a distortion of Parsons, or produces a “virtual” rather than an actual use of his theory. His structural categories, his rules of interpretation, and the resulting theory of action and theory of the social system elaborate these fundamental conceptions. To fail to grasp them therefore is an important source of systematic misunderstanding. Most particularly, the failure leads to the popular but incorrect charge that he has furnished no “reduction statements” that would permit his theory to be coordinated with observable events. Once the theorems are grasped, it is obvious that empirical references are intended and demonstrable at every point in Parsons’ entire work without exception.

XXIV REVIEW OF PARSONS’ ANSWERS TO CRITICS IN “THE SOCIETAL VALUE SYSTEM” In his manuscript, The Societal Value System, Parsons attempted to answer the criticisms to his claims about the value system of American society with the hope thereby of shifting the “burden of proof ” back to the critics of his claims. He wrote: It will be our general contention throughout this volume that (with respect to values and cultural premises) American society has remained essentially stable for the nearly two centuries of its national existence.

In order to furnish the contested claims their common ground, Parsons stated four aspects of the empirical problem presented by American values: (1) the problem of value content; (2) the problem of unity, i. e. “whether we can speak of a single paramount system, or must we speak of value pluralism;” (3) the problem of stability, i. e. whether the data require recognition of crucial changes over the period of the national existence of American society; and (4) the problem of distinguishing between changes of values and changes in the level of structural differentiation.

250

H. Garfinkel

It seemed to me that the methods that Parsons used in making his answers were incapable of carrying the weight of the task of demonstration that Parsons had set for himself. Thus: (1) Parsons treated the problem of content by referring to the preceding section of his paper where the conditions for a “determinate logical structure of the problem of meaning in the U. S.” was said to be satisfied by ascetic Protestantism, while the “good society” as a determinate system of moral expectancies was satisfied by the values of worldly instrumental activism. Parsons was well aware that this method did not permit him to meet the argument that he might have started with ascetic Protestantism and worldly instrumental activism as the goal of his search. Obviously it is not enough to show the meaningful compatibility of ascetic Protestantism with the cultural premises that the theory provides as a formal possibility. An adequate demonstration would require as its minimum feature that the operativeness of ascetic Protestantism be independently demonstrated. (2) The method for answering the question of the existence of a single paramount system versus value pluralism was first to phrase the problem as two questions: (1) whether subsequent infusions of population elements... with traditions different from the predominant early ascetic Protestantism... introduced fundamental value pluralism; and (2) with regard to non-religious foci of value differences, how seriously (a) regional differences, and (b) differences between social classes are to be treated with respect to the existence of “parallel” systems. In each case the data were used as indications rather than as evidence. In addition, and by virtue of using historical materials, the argument got hung up on the time problem, i. e. how to demonstrate the integrity of the value system given that infusions of migrants from other societies would “for some time” pose the problem of whether and when they would have become assimilated or “remain” an “enclave.” In every case Parsons’ critics might say of the period of migration that here was a case of value pluralism. Parsons in each case would have to say “Wait and see,” or “Let’s see what will finally have developed.” This impasse is possible for every situation that shows the features of historical development where it is concerned with any state of affairs that is conceived as an historical process. (3) The problem of stability was handled by first distinguishing values and “implementing behaviors” and then directing attention to the first. Parsons then agreed that though the facts cited by the critics were correct, these facts could be interpreted differently than the critics did, and thereby these facts lacked evidential character about changes at the societal level. Parsons argued further that his critics’ arguments were made at the cost of introducing ideo-

Parsons’ Primer

251

logical distortions. The shortcomings of this procedure consist in the fact that in order to decide questions of fact, Parsons’ theoretical formulations had to be accepted as a necessary way of conceiving the facts. Similarly his definition of ideology proposed that there were empirical consequences of the critics’ position. Obviously then these consequences are to be decided as to their accuracy as questions of fact. Parsons’ argument, however, rested upon the choice of a definition of ideology. (4) The crux of the fourth problem was proposed to consist of the view that the system of values “includes all those patterns which are essential to the functional exigencies of a going social system, (and)... that because a system gives primacy to one functional sub-system does not mean that the remainder are illegitimate, but rather that there is an order of primacy, and that this order is intimately related to the processes of structural differentiation... but need not itself change as the differentiation proceeds.” The method of the answer consisted in effect of explicating a theoretical position. Given these methods it would be altogether proper for a critic to withhold credence by invoking the rule that matters of fact stand independently in their sense and warrant of the particular scheme of inference that might have been used to formulate them as empirical propositions in the first place. Parsons’ arguments therefore were impaired by the failure to have proposed the properties of cultural premises and value systems in U. S. society as matters that are decidable solely as questions of fact.

XXV A RATIONALE FOR A METHOD FOR DECIDING BETWEEN PARSONS’ AND HIS CIRTICS’ ARGUMENTS ON AMERICAN VALUES The analytical philosophical researches of Alfred Schutz and others have furnished researchers a clarification of the meanings and uses of the critical conceptions of sociological theory that are relevant to this dispute. These researches permit the claims of the contenders to be formulated with sufficient rigor to permit their comparison with respect to sense and warrantability. Assuming that agreement can be achieved on these fundamentals the possibility exists of contrasting the empirical claims of contenders in terms that permit a decision between them on the basis of a series of researches. It is the purpose

252

H. Garfinkel

of this note to sketch these fundamentals and to indicate the empirical issues and methods for settling them. Parsons and his critics are equivalently committed to the conception that the intelligibility of the phenomena of social structure requires that some provision be made for an actor whose actions occur in a culturally defined environment. This environment consists of objects of the actor’s interests, known and treated by him in accordance with his “stock of knowledge at hand.” This environment is sometimes referred to as the “society seen from within”. These cultural environments include not only other persons as objects, but the actor as he is an object for himself and others; he finds not only others, but himself as objects in this environment, but he finds himself as well, as an object along with others, and from the perspectives of his own courses of action, or treatment of these objects. This distinction between the actor as a set of actions in their course and the related set of possible events is rendered by the phrase “subject-object dichotomy”. The idea of the actor as a course of actions, or operations upon an environment of related possibilities which Parsons refers to by his term “action-situation of objects” is a fundamental conception for both Parsons and his critics and defines the elementary set of possible occurrences. It can now be demonstrated that the concept action-situation of objects is identical in meaning with the concept action directed to the treatment of an environment as a system of relevances. This notion is neither new nor strange either to common theoretical sociological discourse or to Parsons. (For example, functional analysis.) For another example, “structural differentiation” as Parsons intends this concept, at the level of empirical observations, is identical in sense with the “institutional segregation of relevances”. In talking about “cultural premises” and “values” the contenders are talking about priorities and commonalities of relevances, i. e. “community of understandings” or “common culture”. Within the theories in which they are used, the concepts of cultural premises and values serve the purpose of solving the theorist’s problem of how a common world, i. e. how situations of objects which are manifestly treated by persons as objects that are known in common with others are possible despite the empirical phenomena of perspectival appearance, differing personal histories, the real world which is encountered only over the temporal course of its successive appearances, and the like. Such phenomena are known not only to the sociological theorist but are known to the actor as well.

Parsons’ Primer

253

If this much is correct, then each of the contenders must necessarily provide a solution to at least four important problems: 1. How to reconcile the phenomena of stable concerted actions with the relativity of perspectives? 2. How to provide the actor’s environment with the scientific theorist’s feature of a determinate set of actor’s expectations without turning the actor into God on the one hand or a cultural dope on the other? 3. How to provide the theorist a way of claiming the necessary character of his inferences about the related character of social structures given that his theory, as a scheme of inference, must correspond in its logical structure to the logical structure of the actor’s objects and the courses of the actor’s treatment of these objects? 4. All the contenders have the problem of differentiating the naive phenomenal values from analytic phenomenal values, i. e. “real values”. Parallel distinctions are involved with respect to every phenomenon that the sociologist may choose to bring under review with respect to its status as a “real” versus a “virtual” or “perceived” or “phenomenal”, etc., e. g. perceived persons, motives, biographies, relationships, attitudes, activities, socially organized arrangements, knowledge, common culture, etc., etc. In a word, the problem is that of defining the “real social structures” and the “real community”. Such definitions consist of whatever rules of investigative and theorizing procedures the sociological theorist counts as correct ones, so the problem consists of deciding among alternative theories of social actions and their objects, as contexts for criticizing life in society as the actor portrays it, and methods for meeting the attendant problem of relating expected and actual phenomena in the manner of evidence. Insofar as each of the contenders solves these problems differently there will be different empirical consequences to correspond to these different solutions. On each of these problems there are more differences between Parsons’ solutions and the solutions of his critics than there are between his critics. Space does not permit the further documentation of this assertion. For present purposes, it will have to suffice merely to mention quickly the nature of his solutions. 1. Parsons solves the problem of the relativity of perspectives with two devices: (1) by distinguishing the sociologist’s point of view and the sociologist’s actors point of view (his contenders distinguish the sociologist’s point of view and the actor’s point of view) and (2) by using the “structural-functional

254

H. Garfinkel

paradigm” to transform the culturally defined environment as a system of naive relevances into a set of essential references to those features of the system of interaction that Parsons says must be referred to if the relevances are to have a determinate character for the sociologist’s actor and thereby if the conditions of stable structures of interaction are to be met. The result is a rule of procedure that we might call the “rule of involuted reference”. This procedure is not only unique to Parsons but is one of the most powerful analytical devices that is available to sociological researchers. By the same token, it is a point in his method of formal sociological analysis at which any demonstration that his decision to proceed as he does is based on empirically incorrect grounds could do great damage. 2. The pattern variables are used to solve the problem of the determinacy of actor’s expectations. 3. Parsons as well as his critics solve the problem of necessity by using one or another type of structural functional method. Parsons’ variation employs the structural-functional paradigm and the boundary exchange paradigm as a method. 4. The real social structures, real values, etc. are defined by Parsons by the following procedures depicted in the following figures:

Parsons’ Primer

255

256

H. Garfinkel

In a third way, “analytic”, as it applies to the use of definitions, refers to Parsons’ use of “relational” concepts as compared with “substantive” concepts. Cassirer described the two; the terms “relational” and “substantive” are Cassirer’s. (Cf. Substance and Function.)70 Parsons and his critics are alike in their subscription to and use of the “rule of sociological empiricism” to accomplish the description of a cultural object, e. g. for the “real, good society.” They differ from each other however in the limits that are placed on what this rule suspends with respect to the theorist’s judgments of belief, value, importance, truth, etc. The “good society” for Parsons is the product of treating what the subject is “talking about” (use Peter Blau’s description of the good society to illustrate the subject), transforming its sense according to the rule of sociological empiricism while presupposing that the theory of action comprises the formal invariant features of the ensemble of presentations of the good society in Peter Blau’s account. Thereby, the structural functional paradigm is used to decide what the subject is really talking about. There is thus a distinction between the “good society” which, as an evaluated object, consists of the social system depicted as a related set of systemic or analytic phenomenal values, which are for Parsons real values, and the “good society” as an evaluated object which consists of the society depicted as a set of virtual phenomenal values, which is an alternative “good society” most frequently encountered in classical ethnographies, or which would be produced by reading Blau by using the rule of sociological empiricism without presupposing any particular theory of social organization. By following Parsons’ procedure the actor’s cultural environment is transformed in its sense to consist of a set of related analytic phenomenal states of affairs, i. e. relevances. The various functions of these relevances now define for Parsons the presence of cultural premises, values, and norms. Thus: (1) Some set of related analytic phenomenal states of affairs, i. e. relevances, constitutes the meaning of legitimacy. Any related states of affairs that operate in this fashion are called “cultural premises”. (2) Any related states of phenomenal affairs, that define for actors those features of a society that depicted members are depicted as assuming for themselves, assuming the same for others, and assuming that, as they assume it of others, the others assume it for them, are committed to a respect for, and production of – i. e., are equivalently committed to the respect for and production of, are called “values”. (3) Any related states of affairs that define the conditions under which corrective action must be taken as a condition of bona fide, i. e. competent, collectivity membership are called “norms”. 70

[Editor’s Note: Cassirer, E. (1953). Substance and function. New York: Dover.]

257

Parsons’ Primer

It will be seen that “commitment” in the sense of turning over to others the rights to control the actor’s alternative uses of resources is built into the conception of real values as a constituent meaning of real values, and thereby as a constituent meaning of Parsons’ “good society”. Thereby the meaning for the sociologist’s actor of a phenomenal state of affairs as a real value, i. e. its status as a real value from the sociological theorist’s point of view consists entirely and only in its “validity”, i. e. in the “guarantees of compliance”, “commitment”, “degree of internalization,” “degree of institutionalization”. More exactly the validity of values and norms consists in the probability that compliance with legitimate values and norms can be enforced as maxims of conduct. Thus, the empirical existence of values and norms consists only and entirely in their validity. The method for deciding the empirical existence of real values will thereby have to be based on this sociological sense in which values and norms are said to “exist.” Thus the concept of “validity” furnishes us with the empirical phenomena that such properties of premises, values, and norms as their content, unity, stability, consistency, changes, hierarchical control, etc. refer to. The empirical demonstration of the validity of some good society for actual members of U. S. society (in principle this holds for actual members of any society) can best be made by giving persons a chance to exhibit compliance in situations in which they could if we were wrong refuse compliance. Hence we require empirical situations in which the question of compliance can be made problematical for the actual members. (Remarks on the usefulness of trouble.)

XXVI A Method The fact that the structural-functional paradigm has the logical status of a solution to the problems of relativity of perspective, determinacy of expectancies, necessity in inference, and the reality problem has the very important theoretical consequences that no unique cultural order corresponds to a system of behaviors if one starts with the behaviors and constructs the order from these. But if one starts with the cultural order a unique system of behaviors will be found that corresponds to this order as the behaviors for which the order serves as governing maxims of conduct. This holds however under the condition that the social system does not fall into the terminal states of either total disorganization or total integration of institutional norms. Since neither terminal state has ever been even closely approximated in any large-scale system – on the personality level, recent experiments with sensory

258

H. Garfinkel

deprivation have provided data on “deculturation” – we may safely assume that any system that we choose to study will be suitable as far as this condition is concerned. This theorem permits us to generate a concept that we shall call “the real world” that will make it possible to state in a rigorous way not only how we must proceed empirically to test the properties of premises and values that Parsons proposes but to discriminate in a procedural way between the phenomena that Parsons is calling values and what his critics are calling values without having to subscribe to the theories of either in order to make the comparison. The critical notion that we begin with is Parsons’ notion that structural components as well as functional sub-systems are not checklists but are related according to the rule of the central hierarchy. To use his terminology, in ascending order these sub-systems stand to each other in a relationship of environment; in descending order they stand in relationship of control. I propose that these terms “ascending order” and “descending order” are two parameters of a determinate situation of action. Alternatively stated, they constitute the concept of the sociologist’s actor’s real world. The reasoning is as follows: in saying that premises, values, and norms stand in a relationship of control the meaning is that the relationships of each to its superordinate category is one of institutionalized priority of relevance. In “descending” order each category stands to its subordinate in a relationship of environment. The meaning of this is that the subordinate category stands to the superordinate as consisting with respect to the superordinate category of committed actions as adequate solutions to the functional exigencies. Given their content these two parameters define action in recognition of a “real world” or action “in the real world”. This real world remember is the actor’s situation of relevances which govern actual courses of action. The method for deciding the empirical features of content, unity, consistency, etc. is based on the two properties of institutionalized primacy of relevance and committed-courses-of-action-as-adequate-solutions-to-functional-exigencies, i. e. is based on the establishment, maintenance, restoration, etc., of the real world, i. e. of a determinate field of empirically possible actions. This field is a theorist’s device. Interpreted from the point of view of the sociologist’s actor, this field consists of the correspondence between a legitimate order and the actual occurrences of the events of the social structure. It is a unified way of stating Parsons’ two theorems: (1) the real social structures consist of institutionalized patterns of normative culture; (2) the stable features of the real social structures are guaranteed by compliance to a legitimate order.

Parsons’ Primer

259

Various modifications of the “real world” are possible for the sociologist’s actor. Modifications of content and primacy of relevance produce the alternative fields of play, scientific theorizing, dreaming, drama, games, etc. Each modification begins with the field that is constituted by the institutionalized primacies of relevances of the real society and of the committed actions that constitute it as a viable system. The other fields are constituted as specific modifications of the real world as temporally limited “departures” from it. Each dissolution of a derived field returns the actor to the field of the real society.71 The dissolution of the field of the real society is a deculturated state. The real world of the real society has the further property that it is encountered by the actor only over the temporal course of the real society’s successive actual appearances. Parsons’ claims as compared with the claims of critics can now be stated in the following analytic propositions: A field corresponds to each modification of the fundamental field, “the real world of everyday activities:” i. e., the real world of play, the real world of scientific investigation, the real world of the on-going theatre drama. Living in any of these, actions and objects constituted in the others and interpreted from the perspective of real world at hand will appear irrelevant, arbitrary, contingent, undetermined, accidental. Consider now only the real world of the real society, i. e. the real world of activities of everyday life. Since it is a determinate field of expectations, it must function for the actor as a context of interpretation in terms of which he assigns to actual appearances their sense of appearances-of-the-real-society. We wish to establish the content of values, and most particularly of those which, precisely to the extent that they are institutionally routinized, the actor cannot tell us about them even if we ask him without making of the real world the object that he addresses under reflection and in accordance with the identical interests of the scientific investigator; which is to say that he transforms the real world of daily life activities to furnish it its sense of an object of scientific theorizing. We conclude that he cannot tell us what the real values are when we ask him without altering the real world as an arena of real worldly interests. Thus a procedure is required which permits us to learn from the actor what these values are. The procedure must be such as to permit the actor to teach us what these values are while his actions remain those occurring in and governed by the 71

[Editor’s Note: This anticipates Goffman’s work on “Frames” and “Keying” by over a decade. See Goffman, E. (1974). Frame analysis. Cambridge, MA: Harvard University Press. But it also has antecedents in Garfinkel’s 1947 paper on the Information Apperception Test, written during his student years at Harvard. Garfinkel, H. (1947). “Notes on the information apperception test”. Unpublished manuscript, Garfinkel Archive.]

260

H. Garfinkel

matrix of the real world of the real society. He can teach us what these values are if we introduce into his actual situation events that are “surprising” or “incongruous” when, as scenic appearances, they must be treated as appearances of the real world of the real society.

XXVII MORE ON A METHOD FOR DECIDING ETC. (SEE XXV, XXVI) Researches of analytical philosophy, most particularly those of Alfred Schutz, have furnished researchers a clarification of the meanings and uses of the critical conceptions of sociological theorizing that are relevant to Parsons’ argument. Recent theory is available to researchers to permit the claims of Parsons and his critics to be brought under formulations that permit their rigorous comparison with respect to sense and empirical warrant. Because agreement can be achieved on these fundamentals, the possibility exists of a comparison of the empirical claims of each in terms that permit a decision between them on the basis of the outcomes of a series of researches. Subject-object. Replace “subject” with “course of action” and replace “object” with “situation of objects”. It can be demonstrated that the situation of objects is identical in meaning with an actor’s environment as a system of related, i. e. relevant, possibilities. Equivalently committed to the action-situation of objects (subject-object) which is a fundamental conception for both, Parsons and his critics use this frame to define the elementary set of possible occurrences. (I define an object in the sense of an “organization of meanings” to consist empirically of a set of related, expected, anticipated, recollected, remembered possible events. These are encountered via the sociological phenomenological reduction. See my description of the rule of sociological empiricism. The result of this procedure is the “pure culture”, i. e. a social system as it is meant, and only as it is meant.) When this is done, the action-situation of objects is demonstrably identical in meaning with the action-system of relevances. Actor consists of a courseof-actions, or operations, upon an environment of related possibilities. This notion is neither new nor strange, either to common theoretical sociological discourse or to Parsons. See Schutz on the meaning of function and functional analysis in these terms. Structural differentiation and functional differentiation are identical in sense at the level of empirical observations with “institutional segregation of relevances”. Family and occupation re: structural differentiation of collectivities; functional differentiation in “changing functions of family” vis-à-vis other collectivities. In talking about cultural premises

Parsons’ Primer

261

and values we are talking about the priorities and commonalities of relevances, i. e. the “community of understandings”. Cultural premises and values are concepts that solve the theorist’s problem of how a common world, i. e. how situations of objects treated by actors as known in common is possible despite the empirical phenomenon and the knowledge of the actors of perspectival appearance. If this much is correct then each of the contenders must necessarily provide a solution to at least four problems: 1. How to reconcile the phenomena of stable concerted actions with the relativity of perspectives. 2. Problem of determinacy of expectations. 3. Problem of necessity. 4. Problem of real social structures and real community. Insofar as each of the major contenders solves these problems differently, there will be clearly discriminable empirical consequences. Parsons’ solutions: 1. Given the stress on the actor’s point of view, what solution does Parsons use to the problem of relativity of perspectives? (a) Difference between the sociologist’s point of view and the sociologist’s actor’s point of view. The contenders distinguish the sociologist’s point of view and the actor’s point of view. (b) “Rule of involuted reference” – this is also Parsons’ solution to the problem of idealism vs. realism. This rule is Parsons’ method for defining the real society, real values, real norms, etc. All contenders have the problem of differentiating naive phenomenal values from analytic phenomenal values, i. e. “real values.” Parsons differs from the other contenders in method but not in aim. 2. Pattern variables as solutions to the problem of determinacy of expectations. 3. Different structural-functional methods as solutions to the problem of necessity. 4. Problem of real social structures and real community. Difference between virtual phenomenal values and analytic or systemic phenomenal values, i. e. real values. Society as an evaluated object, i. e. the good society as: (S) naive phi (product of the rule of sociological empiricism); (S) analytic phi (product of the rule of sociological empiricism plus the sociological phenomenological reduction – the sociological epoché while presupposing “the system” as the formal schema of the ensemble of presented features of actions, and using the “system” to decide what persons are really doing and really talking about).

262

H. Garfinkel

1. Related analytic phi states of affairs (relevances) that constitute the meaning of legitimacy – called “cultural premises”. 2. Related analytic phi states of affairs (relevances) that define the conditions, features of a society that members are equivalently committed to a respect for, and the production of – called “values”. 3. Related analytic phi states of affairs (relevances) that define the conditions under which corrective action may be taken, these being conditions that define bona fide collectivity membership, i. e. competence – called “norms.” 4. Commitment is built in as a constituent meaning of real values, i. e. of (S) analytic phi. 5. The meaning for the sociologist’s actor of a state of affairs as a real value, i. e. its status as a real value consists entirely and only in its validity, i. e. “guaranteed compliance”, “commitment”, “degree of internalization”, “degree of institutionalization”. The method for deciding the empirical existence of real values is based on the sociological sense in which values and norms are said to “exist”. “Exist” = “validity”, i. e. the probability that compliance with legitimate values and norms can be enforced as maxims of conduct. The concept of “validity” furnishes us with the empirical phenomena referred to by the properties of content, unity, stability, consistency, changes, hierarchical control, etc. of premises, values, and norms. Object Values Cultural premisesconstitutive expectancies

Modes of sensibility of phi environment Univocality“Clarity in equivocality degree” “Ambiguity” Absurdity ambivalence Anomie Formal nonsense

Expression Conceptual Categoricalconstitution

Now we have to decide the problem of evidence. When we say that cultural premises stand to societal values in a relationship of “control” the meaning is that the relationship is one of institutionalized priority of relevance. The method for deciding the empirical features of content, unity, consistency, stability, etc. is based on this property, i. e. institutionalized primacy (or priority) of relevance.

Parsons’ Primer

263

Clarity in degree – “the book note in the margin; it refers to something in the text but I can no longer remember what.”72 Absurdity – round square. Incompatible within a common domain. In Boston, going in any direction, always you know what street you’re crossing but you never can tell what street you’re on. Nonsense – hue without extension. Sound that has duration and amplitude but no frequency. Unimaginable. Equivocality – faces around Harvard are familiar to me but none is recognizable. The thing in the window of the carriage house – is it a coil of dusty wire; the horn of an old discarded gramophone; a design painted on the window? Ambiguity – rules of correct preference – Hatt-K procedure; bargaining for standard-priced merchandise; stranger in the house.73 Anomie – insult to constituent meanings of legitimacy. The task is how to make an analysis of an observable set of activities so as to produce for the set the “values” of the system. The analysis must produce these as invariants with respect to the specific contents, but when translated into the contents particular to the setting under observation they consist of specific expected features of that scene. Such expectancies must have the property that if they are breached that this operation anomicizes the system, i. e. re: formal logic, it produces formal nonsense; or behaviorally, it anomicizes the actor’s grasp of his situation (makes it senseless); structurally, it multiplies the disorganized features of the interaction, i. e. alters the state in the direction of the ideal terminal state of disorder, its features approximate those of the ideal terminal state of total disorganization, dissolution of the order. However this dissolution is not a dissolution of claims as occurs when we say that a collectivity has dissolved (e. g. a business has gone bankrupt), but is the dissolution of a real world. Begin with Weber’s concept of misfortune, which he defines as the discrepancy between destiny and merit. Use the term “trouble”. Interested in 72

[Editor’s Note: Garfinkel refers to this example again on p. 266, case b, where he is clearer about what it involves.] 73 [Editor’s Note: “Bargaining for standard priced merchandise” and “stranger in the house” are breaching demonstrations Garfinkel had his students do. The studies are reported in Studies in Ethnomethodology, Chapter 2:68–70 and pp. 44–49.]

264

H. Garfinkel

trouble with the following properties. Call it “Important Trouble”: (a) obstinate; (b) objective; (c) essentially senseless; (d) normal, i. e. “arising out of ”, “existing as a possibility by reason of ”, “constituted by”, “invariant condition for the occurrence of ” – commitment to the actualization of the real values. Thus whatever conditions guarantee the commitment, i. e. whatever conditions guarantee a high validity, guarantee the trouble. In each case, where the method of structured strain is concerned, the values are the definitions of the good, natural, true, real, etc. “life”. The cultural premises furnish the constituent meanings of “correct”, legitimate. The method: I propose that an adequate test of the contents of value commitments, stability, unity, consistency, etc. consists of examining the conditions of the presence of structured strain, the selection of remedies, their effectiveness, the assessment of adequacy, and particularly for the case where structured strains are sustained without demoralization of the system. Structured strain, as a method for demonstrating the meaningful and empirical adequacy of the conceptions of “real values of the American Society”, i. e. content, hierarchical control, constitution, unity, stability-constancy, consistency, primacy of relevance, essentiality of structures delineated by the structure-function paradigm, generality of relevances assigned in the American case in defining real values, i. e. analytic phi values. Parsons’ arguments can be tested by considering the problem of stability, unity, consistency, content, hierarchical control, etc., of values in the light of the concept of structured strain. A method to confront persons with the conditions under which anomicizing of behavioral environment should occur. Also review following materials: (1) Responses of middle class housewives in an extension course to descriptions of structured strain, covering family features of conjugality, structural isolation, openness, multilineal lines of descent, monogamous marriage, addition of parents to household, incest.74 (2) Schwartz and Messinger75 materials on managing the hospitalization of schizophrenic spouses. 74

[Editor’s Note: See Chapter 9 on “structured strain” for further details on these topics.] [Editor’s Note: Messinger coauthored at least two papers on hospitalization in the early 1960s. One of these, “The mental hospital and marital family ties”, says in the acknowledgements: “The general perspective adopted [in this paper] owes much to conversations with and published and unpublished writings by Harold Garfinkel”. The paper also references a publication by Charlotte Green Schwartz (1953), “Rehabilitation of mental health patients”. These are probably the materials Garfinkel is referencing here. Cf. Sampson, H., S. L. Messinger, R. Towne, et al. (1961). “The mental hospital and marital family ties”, Social Problems 9(2):141–155 and Schwartz, C. G. (1953). “Rehabilitation of mental health patients”, Public Health Monograph No. 17, Washington, DC: U. S. Government Printing Office. Cf. also Schwartz, C. G. (1957). “Perspectives on deviance: Wives’ definitions of their husbands’ mental illness”, Psychiatry 20(3):275–291.] 75

Parsons’ Primer

265

(3) Reactions of psychiatrists to presentations of Hollingshead and Redlich.76 (4) Bargaining for standard price merchandise.77 (5) Administering features of American society as valued features of the society, via questionnaires. (6) Moral legitimacy of economic rationality. (7) Evelyn Hooker’s materials on homosexuality.78 (8) Lay responses to pictures and biographies of intersexed persons.79 (9) Psychopath materials. (10) Concentration camp and hell ship materials.80 (11) “Deculturation” experiments. (12) Aging and character disorders. Features Content Unity Primacy of relevance Stability Consistency

76

Paradigmatic responses “You describe my situation exactly.” General agreement – “That’s obvious.” “What to do?” And when alternations of values states are suggested, they are ruled out as out of the question and for the theoretically anticipated reasons. Problem is that of object constancy. Work of historicizing. See discussion of “common law” interpretations. Discovery of private life; prevalence of secrets; attitudes toward difference between public agreements and private life; discovery that sin is possible without demoralization.

[Editor’s Note: Hollingshead, A. B., & C. F. Redlich. (1958). “Social class and mental illness: Community study”. Garfinkel wrote an unpublished paper, “Comment on Hollingshead and Redlich: Social Class and Mental Illness”, which can be found in a black binder titled “Unpublished Papers” in the Garfinkel Archive.] 77 [Editor’s Note: These results are reported in Chapter 2 in Studies in Ethnomethodology:68–70.] 78 [Editor’s Note: Evelyn Hooker was a psychologist known for her research on homosexuality, particularly her argument that homosexuality is not a mental illness and not inferior to heterosexuality. She and Garfinkel worked together in the late 1950s’ and ‘60s, during the period when Garfinkel was studying trans-people, most famously Agnes.] 79 [Editor’s Note: Probably a reference to Garfinkel’s research on trans-people at UCLA in the late 1950s’ and early ‘60s. Garfinkel interviewed 15 trans-men and women at Robert Stoller’s clinic, including Agnes. See Garfinkel, H. (1967). Studies in ethnomethodology, Chapter 5. Englewood Cliffs, NJ: Prentice Hall. 80 [Editor’s Note: Garfinkel collected accounts of people in extreme situations where the Trust Conditions needed to sustain social order are absent (anomie) or nearly absent, including concentration camps, the Japanese “Hell Ships” (see Chapter 7 of the Primer), and natural disasters. Garfinkel also wrote a research proposal, which was not funded, titled “Persons in Extreme Situations”, where he reports that he began work on the topic in 1951 while preparing lectures for the course on Social Disorganization he taught at Princeton.]

266

H. Garfinkel

XXVIII COGNATE STRUCTURE OF MODES OF SENSIBILITY Modes of grammatical sensibility Mode Clarity in degree

Level Expression

Formal absurdity

Conceptual

Formal nonsense

Categorical, or constitutional

Modes of sensibility of the phenomenal environment Mode Level Univocality... Object Equivocality (a) Absurdity or Values ambiguity Evaluation (b) Ambivalence Anomie Cultural premises, or, constitutive expectations

EXPLICATION MODES OF GRAMMATICAL SENSIBILITY Mode: Clarity in degree Level: Expressions Rule of relevance or “order”: Word as sign of meaning in terms of a system of rules that assign the word its function as a sign. Cases are generated by manipulating the formula, s-r, R Cases and their illustrations: a) I encounter a word I have never seen before. I assume it is a word in the English language. I look it up in the dictionary. b) The note that I wrote in the margin of my book which I know refers to something in the text, but I can no longer remember what and though I try I can no longer see what it was referring to. c) My number is flashed on a board at the desk of the reference librarian at UCLA. When I answer in anticipation that the librarian has something to tell me about the book I ordered she tells me that there is something wrong with the signaling apparatus. The next time my number flashes I am uncertain about whether or not I am being summoned. d) The page of mathematical notation is an argument that is saying something that I very much want to know but instead of the steps

Parsons’ Primer

e)

f)

g)

h)

267

of the argument I find the author has written “It follows that” and I am unable to provide the intervening steps. The page of mathematic notation describes a technique that I want to know but the author has not explained his notation. When I search the text for the meanings, I find that he is using conventions in notation that mathematicians take for granted but that I have no idea about. The German text which I am reading has a set of recurring words with technical meanings which are not satisfied by the words in the dictionary. Passages where these words occur are obscure to me. I am uncertain that I have grasped correctly the entire argument because of these omissions. A sentence guessing game is being played. The blank spaces where the letters go are partially filled in. I’ll bet heavily that the three letter word that begins with “T” is “The” and if it is then the next word I’d bet is a noun. I try alternately to use the statistical structure to narrow the likely meanings so as to decide the statistical structures; then I start with the possible rhythmic pattern to narrow the meanings; then I start with possible meanings to narrow the set of statistical arguments, etc., etc. I am reading a sociological account in which the author employs a great deal of figurative language usage. I can’t tell at page 3 of the account what the author does not mean by “Dionysian”. By the end of the account I see what was meant “all along” and “in the first place”.

Conditions and remarks: The distinction between “objective expressions” (objective meaning of a sign) and “subjective expressions” (subjective meaning of a sign). By “objective meaning of a sign” is meant the meaning according to a conventional set of rules R, which the user presupposes, and uses under the expectation that others use and expect him to use it as an identical scheme of interpretation. By “subjective meaning of a sign” is meant the meaning according to a scheme of interpretation which the user presupposes is employed by a particular person; the meaning of the sign implies a reference to the person using the sign. An expression is objective if its meaning is assigned “objectively”; it is subjective if its meaning is assigned “subjectively”, i. e. in terms of the objective or subjective references of the scheme of interpreta-

268

H. Garfinkel

tion. The program of sociology provides as a matter of principle that all subjective expressions which are encountered as actual usages of members of the society under study are to be brought to their status as objective expressions while retaining their feature of subjective expressions. The sociologist’s theory of social structures of interaction is the place where this radical ideal of rationality occurs; it provides the sociologist with this objective character of subjective expressions by permitting him to say “what the member is really talking about.” Mode: Formal absurdity Level: Conceptual Rule of relevance or “order”: “Normative logics”, i. e. alternative rules of interpretive procedure among which the user may freely elect with the condition that a unity of sense (coherence, compatibility, clarity, consistency) is the ideal goal of an election. Each set of rules defines a domain of meanings with relationships of relevance that are demonstrable by these rules. Thus to a set of rules of preference there corresponds the domain of preferences; to the rules of inductive inferences there corresponds the domain of warranted assertability; to the rules of fantasying there corresponds the domain of fantasies, etc., etc. Cases of absurdity are generated by manipulating expressions with respect to these orders. Cases and their illustrations: a) When travelling in Boston a driver going in any given direction will always be able to determine what street he is crossing, but will never be able to determine what street he is on. b) The expression “round square.” c) An anthropologist reports to a seminar that a Navaho respondent said that he was convinced of the curative and supernatural power of Navaho curers (“bears”) in 1910 when he saw a curer walk across the ceiling on his hands and knees. d) A tape recording is played for a subject in a psychological experiment which has been recorded at 3 and 34 feet per minute. When it is played at 7 12 feet per minute and he is in the “waking state” he cannot make out what the recorded voices are talking about. Under hypnosis he can. Reactions of the psychologists: It can’t be true but there it is.

Parsons’ Primer

269

e) The time that the steeple clock shows is neither correct nor incorrect. f ) Simultaneous assertion of contradictory propositions. Conditions and remarks: A set of expressions from different domains of meaning are combined within a common domain while the rules of relevance whereby the common domain is constituted as a unified order are retained without alteration. While each expression retains its sense, the “point” is essentially missing. Mode: Formal nonsense Level: Categorical; constitution Rule of relevance or “order”: Pure grammar, i. e. “laws which constitute the a priori relations of adequacy of meaning” (Farber).81 Such laws are constitutive of an order of reality. Cases are generated by breaching these constitutive expectations. Cases and their illustrations: a) Pancakes so thin as to have only one side. b) A color that has a hue but no extension; an instance of a sound that has a duration and an amplitude but has no frequency. c) A person’s mind is in his brain.

XXIX A POSSSIBLE REVIEW STATEMENT ON VALUES A possible review: think of an actor as perceiving, judging, loving, hating, moving, buying, selling, and the rest, objects in his environment. This actor is nothing else than a set of activities in their course. These courses of actions are called orientations. The actor’s situation of objects consists of a set of anticipated, expected, recollected, remembered possible environmental occurrences. Analyzed from the standpoint of courses of action, these objects are specified as physical, social, and cultural objects. All such objects are by definition “meaningful objects”. That is to say their features are motivated features; they cannot be defined without presupposing a course of action directed to their treatment. Behavioral 81

[Editor’s Note: Likely referring to Marvin Farber, though it’s unclear which work in particular.]

270

H. Garfinkel

events to which “meanings are attached”, i. e. action governed in its course by such features is constituted in its meanings in this way. The situation of objects consists thereby and entirely, when analyzed from the standpoint of courses of action, of cultural objects or cultural events. Actors are “oriented” to a scene that is presented through its immediate appearances. Objects thereby consist of the paradigmatic schemata of these appearances. These schemata define for the sociological observer the constituent motivated features of the object that remain invariant under transformations upon its appearances accorded by actions in their course. A set of actors are oriented to environments in which each encounters the others as cultural objects. Given the phenomena of the contingent character of the outcomes of courses of actions, the problem arises to account for continuity of treatment and constancy of an environment. From the sociological observer’s point of view this is a problem of accounting for the stable features of activities as products of interaction. The solution to this problem is accomplished with the use of the concepts of values and norms. Norms may be defined as legitimate courses of performances and sanctions. The complementary character of the two is obtained by reason of concerted actions that are governed by a normative environment of possible structures of interaction, to which Ego assumed and assumes for others a commitment, in a more or less equivalent way. Action then is said to be constrained by such “agreements” despite continual variations in actual appearances, disappointments of expectations, variations in impulse life, and the rest. A central problem in accounting for the possibility of stable structures consists of showing how motivated compliance to these “agreements” is guaranteed in the face of the exigencies whereby discrepancies between legitimate expectations and actual occurrences are introduced. These “agreements” consist of states of the system which each partner to the interaction treats as binding upon both as conditions of their concerted actions, i. e. as real perceived grounds of each other’s actions, i. e. grounds for choosing among alternative thoughts to think, feelings to feel, and features of objects to attend, choices which are otherwise “intrinsically open”. Situations “known in common” are the desirable possible states of the system. Such normative states are called values.

Parsons’ Primer

271

CHAPTER VII ECONOMY, POLITY, MONEY, AND POWER If it is not Parsons’ intent to turn sociologists into economists and political scientists, why does he spend the time that he does on the concepts of economy and polity and why introduce these terms into the sociological vocabulary? The answer is to be found in Parsons’ insistence that the area of integration of a set of activities is the area of peculiar interest to sociology. The good sense of what is being done by introducing the concept of polity and economy has to be related to Parsons’ insistence on the problem of integration. This insistence on the problems of integration accompanies Parsons’ preoccupations with what he calls the normative regulation of action, or what textbooks refer to as Parsons’ “normative approach”. What is the “normative approach”? The “normative approach” consists of using as an interpretative rule for understanding any system of activities – regardless of whether they are in an area of medical practices, government, family, social welfare, or whatever – of using the following rule in order to bring them into view with respect to their problematical character. The rule has been stated several times: the real social structures consist of institutionalized patterns of normative culture. That is the frame of interpretation. Everything that is going to be of interest about polity and economy gets informed by the fact that this is the leading presupposition. In whatever we are going to talk about, by way of economic or political activities, this is going to be of interest. The social system is a device and nothing more than a device for conceiving the control of behavior. Its meaning can be illustrated when Parsons, by way of illustrating what is meant by this notion of – from what Parsons calls the institutionalized point of view, “the cultural-institutional approach” – Parsons argues that the interesting thing about the commodity is not that it is a physical possession and not that it’s a physical entity, though it is a physical entity in many cases but not always, and it is not that it can be moved from one place to another, it is not that it can be located at a particular place and not that it can be transported from one place to the next. Instead its interest from the “cultural-institutional approach” consists in the fact that there are rights of acquisition, of disposal, of the use of that particular thing. Similarly when you come to deal with services of persons, or getting very formal, with their performances, or getting even more formal, their role performances, the crux of the interest is not in the thing that is done. Instead, the interesting matter consists of the right that “exists” to enlist, or direct or use or commit those services.

272

H. Garfinkel

To say the social system is a device for conceiving the control of behavior leads us to view these activities with respect to what occurs with respect to what is right, what is legitimate, and therefore too, what is not legitimate; what is positively sanctioned, what is negatively sanctioned i. e. what is done with respect to a normative order which involves the systematic application of the exercise of rights and their enforcement. Interest is directed to the likelihood of enforcement of these rights and the conditions of socially organized activities which determine this likelihood. A second feature of the social system is that it is a device to help the researchers give an accounting of the stable features of the processes and products of concerted activities. Parsons is continually preoccupied with these concerted activities whenever he talks about a social system. By concerted activity is meant any uniformity, any set of events to which two or more actors (temporarily one may think of them as persons without going far wrong) contribute, i. e. these are effects which come about by virtue of the “contribution” of two or more persons. Example: the usual flow of traffic across a bridge. If the “usual flow” is described, for example, as a rate of movement of cars for some time interval, this rate is such a concerted activity. It is the product of a sequence of related activities, in this case, a series of automobiles driving across that bridge. When I talked earlier about the load of persons in the clinic and the flow of persons through the clinic’s assessment procedures, this too was concerted activity. Talk about familiar activities is talk about the same thing.82 In the stable features of concerted activities is meant that there are some regularities of the products of a set of two or more persons’ actions that remain fairly persistent, fairly continuous, fairly reproducible, etc. over a course of time, illustrated again by the stable flow of traffic across a bridge. A delightful story was published in the New Yorker called “The Night the Law of Averages Failed” that illustrated everything we will be talking about on this score. It happens on one April night that traffic is jammed on George Washington Bridge for miles up and down Manhattan and up and down the Jersey shore. The drivers ask everyone about them what the jam is all about. It happened on that particular night that every person in Jersey and Manhattan who owned an automobile thought that it was a delightful time to take a trip across the bridge. So that was the night the “law of averages” failed. The story winds up with the ominous promise. There had been some evidence 82 [Editor’s Note: references to the clinic concern Garfinkel’s study, with Egon Bittner, of a psychiatric clinic’s record keeping procedures. The study was eventually published in as “Good organizational reasons for bad clinic records”, Chapter 6 in Studies in Ethnomethodology. There is also transcript and audio in the Garfinkel Archive of a presentation of this research, then in progress, that Bittner and Garfinkel delivered at the 3/23/1959 meeting of the Parsons Seminar.]

Parsons’ Primer

273

from a town in upper New York State that the law of supply and demand was beginning to crack. The big feature that we are talking about is this normatively regulated usual flow of traffic across the bridge. The essential phenomenon is action that is motivated to compliance with some legitimate order of concerted activities. The legitimate order of concerted activities is portrayed in the one case via a socalled “scheme of values”. Then, as Parsons proposes, you consider the valued character of the flow and seek to program its production. The rules that you would have to follow to program its production would be nothing else than the norms which specify in effects how that usual flow is to be produced. If you put stock in figuring that it was a condition of membership in the collectivity that one night every year you would get in your car along with all the others and jam up the George Washington Bridge, then it is clear enough that the valued concerted product of your activity would be a really hellishly good jam. The normative element would consist in the rule “Get your car out at 6 o’clock in the evening on April 21st and drive as fast as you can to the entry to the George Washington Bridge as a condition of your own self-esteem.” The social system consists of a set of categories and ways of thinking about the related character of these phenomena that these categories refer to. These categories consist of the “structural components”, i. e. culture, collectivities, and roles; the process categories of function, mechanism, input, and output, and the “rules of interpretation”. The stable features of concerted activities consist of adequate solutions to the functional problems. General criteria of good functioning are laid down in this value system. Now, we have a problem. If we have an aggregation of different “units” – these can be personalities in roles or collectivities, and the rest – the products of their concerted activities show properties of normative regulation. The problem is this: how is this to be conceived for the purposes of clarifying what is involved in the fact that on the one hand there are these aggregations of units and their activities and along with it the fact that the effects of concerted actions seem to be normatively regulated and produced? Parsons talks about the same problem in the following language. He says that “the integration of the society concerns the articulation of the value system with the more concrete, differentiated units of the social structures”. This is a nasty way of saying the same thing. This problematic phenomenon, when it is conceived with the use of a concept of activities as solutions to a functional problem, is called the problem of integration of social structures. To repeat: there is an aggregation of units, of different units, different persons, different activities, different sets of concerted activities. The products of the concerted activities show the properties of normative regulation. When this problematic phenomenon is

274

H. Garfinkel

conceived with the use of Parsons’ concept of activities as solutions to functional problems, the resulting problem is called the problem of integration. The concepts of economy and polity are designed to aid in conceiving solutions to this problem. Therefore the reader is not being asked to become an economist and a political scientist. Rather he is being asked to consider these activities with respect to the processes of social control, so that these aggregations are somehow related to each other in such a fashion as to make clear wherein the phenomena of normative regulation consists. Parsons talks about this relationship with the use of a term that does a tremendous amount of work. The term in institutionalization. Parsons proposes two major features of the problem of integration. He refers to one as the normative problem. The other he calls the realistic problem. The normative problem has two parts to it. (1) The problem of maintaining consistency between norms and “the higher order of values”. It can be stated another way: maintaining consistency between rules of procedure on the one hand and the concerted effects of activities that these rules of procedure produce which are valued effects. (2) The second part of the normative problem is that of maintaining consistency between these rules of procedure, as one proceeds from one aggregated unit to the next, particularly where different sets of rules of procedure may be “assigned” to these different units, with respect to the accomplishment of the concerted effects of their “interaction”. Parsons points out that even if the theorist were able to tell how the sets of activities are related so that these two problems are solved, still he would not have sufficiently accounted for the possibility of getting an aggregation of units related to each other in such a fashion that the resulting product would show the features of normative regulation. This is so, says Parsons, because in addition to the problem of handling the consistent character of the cultural order, there remains the realistic problem. A social system, after all, does not exist only via a set of valued conceptions as to how one should go about getting done what is such a valuable thing as to have gotten done. The actual play of the game is not the same as the rules that define how the game should be played. There remain the actual persons who are going to be playing actual games in accordance with whatever actual rules of procedure they finally come to be playing by. The realistic problem consists of the problem of how the actual actions are going to be influenced with respect to the tasks of insuring conformity with the procedures to produce the valuable effects. Or to use Parsons’ phrasing, there remains the problem of “integrating the action of units with the norms.” The realistic problem is thought of as follows. There are two ways in which it is possible to influence the action of a “unit”, e. g. the action of a person.

Parsons’ Primer

275

One is by affecting what Parsons calls the “orientation” or what we can for immediate purposes refer to as “affecting the courses of actions themselves.” These actions are always directed to, i. e. they consist of, treatments of objects. Action directed to the treatment of an object is called an orientation to that object. A second way to realistically influence this unity is to somehow do something to the unit’s situation. Somehow or other, some manipulation of the unit is made, of the unit’s situation which, as Parsons says, has effects that are relatively independent of any operation performed about the properties of the unit itself, for example, operations that may have been performed upon the courses of action of that unit. The same thing can be said, with only some loss of Parsons’ meaning, with the use of Kenneth Burke’s observation in Permanence and Change: if you want to line up two sticks you can either move the position of the eye or you can move the position of the sticks.83 Parsons is saying the same thing. There are two ways to set up the realistic consequences: either move the position of the eye, or move the position of the sticks. In discussing economy and polity Parsons is addressed to those measures for obtaining, for making secure, for managing, for regulating the commitments of persons to conformity with norms. Two measures are prominently involved. The management, regulation, or insuring of commitments can be done by operating either on the actions or on the other hand by operation on the person’s situation. Economy and polity are used in order to explicate the whole notion of the social control of a set of activities. They involve mechanisms for, and conditions of, regulating commitments of persons, units, actors, and collectivities to conformity with these norms. The discussion of economy and polity involves a discussion for both economy and polity of resources and the way resources are allocated and employed. Now for an explication of the usages of the terms polity and economy. A general definition of an economy. Parsons says it consists of all those activities, any activities whatsoever, which when they are analyzed are viewed as measures whereby performances, services, and goods are produced and allocated. Any such set of activities will be referred to as the economic function of the analyzed activities. Any set of activities, which under analysis are conceived to be directed to the mobilization of resources in the interest of the accomplishment of tasks, are spoken of as the polity. “Resources” can refer to attitudes, commodities, goods, services, performances The term can refer to anything at all as long as whatever it refers to, under analysis can be assigned the status of an instrument 83 [Editor’s Note: Burke, K. (1935). Permanence and change: An anatomy of purpose. CA: University of California Press.]

276

H. Garfinkel

in the accomplishment of some task. One cannot recognize a commodity by asking the commodity if it is a commodity. One can only decide that some entity is a commodity by asking where it fits into a program of activities. The same is true in deciding that something is a resource. Parsons introduces two notions. The first: Any set of activities which can be influenced through the allocation of wealth – and this is especially wealth in a monetary form – is to be called the economy. That is not strictly true, because it is not any set of activities. What he is talking about here is what he calls the functional system. That is to say, if you arrange a set of activities with respect to the production of facilities, then that set of activities so conceived will be called the economy in so far as it can be influenced by allocations of monetary resources. Any set of activities, again functionally conceived, which are accessible to influences though political power, is known as the polity. Put it another way. Any set of activities conceived with respect to the manipulation, the management, the ordering of resources in the interests of the accomplishment of tasks, insofar as it is subject to, or it can be influenced by political manipulation, political power (we’ll talk about it in a minute), is known as the polity. Such usage is known as a concept in search of its content. And that is what is tricky about all of Parsons’ and most particularly of Parsons’ later writing. It is not so clearly the case in The Structure of Social Action although it is there too.84 But starting with The Social System a reader will not know what Parsons is doing if it is not understood that when Parsons talks about an analytic conception, or when he says of his method that it is a method of formal analysis that he means he is in the business of generating concepts which are in search of their contents.85 This makes of “economy” and “polity” general concepts. That does not mean that it is not definite or that it cannot be given what Parsons calls its “specifications”. It does mean, however, that one cannot start with events as they are known in the course of everyday experience and then by abstracting from their substantive features to hope thereby to arrive at what Parsons’ general concepts are. His concepts are much more like formulae than they are like abstract representations of “concrete contents”. They have nothing in common with the procedure whereby “sex” is specified with respect to male or female, and is therefore said to be a more “general”, i. e. inclusive, class. Parsons’ general concepts are not inclusive classes. Because we are in the business of trying to trace out the processes of the control of action, and because we are going to have to deal with units di84 85

[Editor’s Note. Parsons, T. (1937). The structure of social action. Glencoe, IL: The Free Press.] [Editor’s Note: Parsons, T. (1951). The social system. Glencoe, IL: The Free Press.]

Parsons’ Primer

277

rected to the use of various facilities in their situation in the accomplishment of tasks, and in the production of various resources, the large terms “power” and “wealth” are encountered. Parsons proposes flatly that these two notions, in their formal meanings, refer to means whereby the commitments of units are obtained, insured, managed, and the rest. Two classes of commitments are involved. One class consists of the transfer of the rights of possession in physical objects. To these rights of transfer of possession Parsons gives the general name of commodities. The second kind of commitment that he is talking about are commitments of persons or units (use persons for a while) to performances that they are obliged to carry out as a condition of membership in a collectivity. Wealth refers to the transfer of these rights of possession. Wealth represents one set of commitments that can be imposed, controlled, regulated, and the rest. Commitments to role performance are referred to when we talk about power. So to say then of an actor that he is wealthy refers to the control the actor is capable of exercising over the use of property. When we say of an actor that is he is powerful, we are referring to control the actor is capable of exerting over the commitments of persons to membership obligations. Both of these kinds of commitments are known as resources. They may be offered to, or they may be withdrawn from, the control of any unit whether one is talking about actors or collectivities or whatever. Resources operate in two fashions: (1) they operate as facilities; (2) they operate as rewards. In either case their “real character” consists in this, that control over them represents for the units that have control what Parsons calls “real advantages”. Big point: There are various processes whereby control over these resources is allocated within a social system among the units that comprise it. The ways in which these allocations are accomplished solves the integration problem. The realistic problem of integration is solved by modifying the situation in which persons, roles, collectivities, operate. The ways in which this allocation is accomplished represents the ways in which the problem of integration is being solved. If one asks whether this has anything to do with sanctions, it is clear that sanctions have already been introduced. It has to do with the lending and withdrawal of power: withdrawal, withholding, not giving in time, giving in short supply, and the like. One can permit imagination to room a bit over any actual situations in order to start filling it in. With respect to accomplishing the transfer of rights in possessions there are three broad mechanisms. Parsons refers to these as levels of generalization of wealth. The idea is simple. Parsons wants to ask the question of how one can get an economy that is capable of producing and allocating resources at a very

278

H. Garfinkel

high level. The notion is one of sheer magnitude. How much? How much more wealthy can the society be on the basis of its economy’s operation? The three mechanisms for effecting the transfer of rights are intended to give the theorist a way of talking first about some of the conditions under which this output of the economy will be “dampened” and the conditions under which its output can be raised. Parsons distinguishes three mechanisms. First, the mechanism of barter. Parsons defines barter as the procedures whereby a transfer of rights in these possessions is effected with possessions being directly exchanged for each other with no intervening medium that establishes a standard of evaluation whereby the value of the exchanged products can be assessed. This does not mean that no assessment is made of their equivalence. It does mean that there is no standard that stands apart from the actual goods which are exchanged in order for the parties to decide whether or not they have obtained a sanctionable ratio in the exchange. Five blankets are worth what? One horse? Parsons says that the criteria of evaluation are ascriptively tied. “Ascriptively tied” means that parties’ judgments of fair exchange goes to a system of relationships between the exchange partners which provides for their particular relationships in that exchange system in order for the parties to decide that that transaction has been correctly conducted. The thing that is exchanged is tied immediately to the terms in which the partners assess each other’s rights to engage in that particular exchange transaction. A second level he describes as follows. You have stocks of goods that may be exchanged through a system of markets. There may be an abstract medium of exchange evaluation like money, for example. But the factors of production are not involved. The market system operates only to allocate supplies among exchange partners; control over the factors of production remains ascribed. The factors of production are not subject to assessment, control, regulation, and the like in the identical ways that allocations are evaluated in market transactions. The third level is one in which both processes, allocation and production, are similarly tied to an identical series of market operations and are governed by identical standards for evaluating the worth of exchanged products. Parsons asserts that a highly productive economy, one that produces enormous amounts of facilities, is not possible if the rules that govern legitimate exchange transactions are those of barter. Parallel to these three mechanisms that concern the economy are three mechanisms for the commitment of role performances in collectivities. The first mechanism is an equivalent political barter. Political barter consists of an exchange of a specific commitment for another very specific commitment,

Parsons’ Primer

279

for example the quid pro quo where an exchange of specific favors may be involved: “if you do this for me, I’ll do that for you.” The second mechanism, which would be parallel to or “cognate with” a system of allocation procedures while the system of production remains ascribed, would be one in which there are commitments to role performances which are given to a leader – Parsons talks of “support” that is given to a leader without a stipulation in advance as to what specific advantages may be obtained. The incumbency of the leadership is ascribed, but the exchange of commitments between leader and followers is handled in the quid pro quo fashion. The third “level” is as follows. Given the understandings with respect as to the claims that can be made to role commitments, and given the problem of how the leader is going to be assigned a position in which he can enter into agreements, with both of them governed by the very rules that govern the transaction, that the product or result of this is the phenomenon in which specific agreements, the quid pro quo, does not operate in the fashion of “I will do this for you as leader if you in turn as leader will do this for me.” There is a commitment to a role, that is, a commitment to compliance with membership obligations in the collectivity, and therefore the assignment to the leaders of the rights to hold one to causal accounting for actions within this membership. But there does not exist prior to the time the commitment is made, a specific understanding as to what the conditions are under which an appropriate exchange will be considered to have been done. There exists instead a commitment without a preprogrammed set of what one’s rights and obligations are under all circumstances under which the member might be held to the fulfillment of this membership obligation. Parsons talks of this as the generalization of power. Big point: Parsons proposes that the significance of economy and polity, in his type of analysis, does not reside in assessments of the technical level of accomplishment either with respect to how much is produced by way of facilities, or how much is produced by way of the likelihood of controlling the commitments of persons to membership obligations. Instead, he says, the significance of economy and polity consists in the attention they call to the mechanisms whereby these technical effects are produced. Now he addresses two problems: (1) What does the polity consist of as a set of institutionalized norms? and (2) what does the economy consist of as a set of institutionalized norms? These problems are asked as the question, what do power and money consist of as sets of institutionalized norms? (1) What is meant by the institutionalized character of the polity? What is meant by the institutionalized normative character of activities governed by the sanctioned ways of managing resources in the accomplishment of tasks?

280

H. Garfinkel

Parsons begins by telling us what he means by a political institution. A political institution consists of those normative descriptions which define the obligations and the rights of units in terms of collectivity membership with respect to the management of resources and the accomplishment of the tasks of the collectivity or the tasks of the units. By a political institution is meant that set of obligations and rights in terms of a category of collectivity memberships with respect to the management of resources in the accomplishment of collectivity tasks. The institutionalized values, the required desirable states of the political system of activities center on the value attached to the effectiveness in the attainment of goals that are acceptable to the members of the system. Thereby the value system itself defines for the members of the collectivity that it governs what they “really mean” when they talk of political rationality. Or, to put it another way, these valued states of political affairs define what members really mean by justice. Three sets of norms make up the norms in the area of political activities: norms of leadership; norms of authority; and norms of regulation. Leadership. Parsons proposes that there exists in every stable system a normative description of who can hold whom to blame. In Parsons’ language leadership consists of “functional differentiation of a collectivity with respect to responsibility for the affairs of the collectivities as a whole.” Speaking literally, this says that if there are valuable political states of the system, that is to say states which are known as justice, and political rationality that there exists as well an allocation of rights such that when things go wrong, not all persons are invested equivalently with the right to hold someone else to blame. Somebody is going to take the rap. By the very nature of the conception of normative order itself that much is certain; but not everybody is entitled to hold anyone else responsible. Instead there are going to be a set of persons who are peculiarly invested with the right to hold others to a causal accounting for the consequences of their actions for the actualization or the realization of a just state of affairs and politically rational actors. Whatsoever persons are invested with these rights are to be called leaders. This is another instance of a concept in search of its content. One starts with the formula and searches an actual situation under examination with the aid of this definition as an interpretive procedure for the persons who are invested with this right. The step beyond this consists of deciding the likelihood that if persons make the charge they can make it stick. And what are the conditions of corporate life under which they can make it stick if they make the charge? Or, to use Burke’s way of talking, every organized system of activities makes full provision for finding and holding victims. It is an inevitable state of life in organized society; victim-

Parsons’ Primer

281

age is a built-in feature. One needs to ask, therefore, not whether or not, but who can hold whom to what kinds of victimage? As the concept leadership is used here, it is not intended as a description of leadership. It provides instead for leadership as a function. Like many of Parsons’ concepts, his definitions provide for the functions of activities and not for their substantive features. One finds the empirical events he is talking about by looking for the functions that the concept provides for in the rule for looking for the functions that the concept’s definition consists of. Where leadership is concerned, says Parsons, it is “inherent” in the very ways in which leadership rights are allocated that units are ordered to positions of superiority and inferiority. This point was touched on before in emphasizing the existence of a differential distribution of rights to hold others to causal accounting for the consequences of their actions in producing a “just” state of affairs. Necessarily, actors will be distributed with respect to their superiority of inferiority. It is clear enough that superiority and inferiority literally goes to differential moral worth. Authority. Better, an authority function. There are a set of rights which are allocated, institutionalized legitimate rights to make decisions on behalf of the members of the collectivity which are binding on those members. In the case of a leadership function the phenomenon consists of a differential distribution of rights to hold to causal accounting. As a way of talking more elaborately about this, we have a phenomenon called authority, or the function of authoritative actions which consists of the rights to make decisions that bind or commit the members of the collectivity. Normatively speaking they can be held accountable for the decisions that were made “on their behalf ”. So if we say that someone “is” Department Chairman, this is a way of fixing a specific organizational position that provides for this function, for we say that he has the right to make various kinds of decisions without consulting the members, for the consequences of which members can be held to causal accounting with respect to the fulfillment of his decision as their commitments. We say then that he “is” an authority or that his actions are authoritative actions. We speak then of charismatic, rational-legal, traditional authority, and not of charismatic leadership, in order to stress the point of the source of legitimacy. It is the particular way of recommending the legitimacy of the authoritative function that comes in when we refer to differences between charismatic and other types of authority. The authority can make the right decisions about what is expected of collectivity members by way of performances “on the job”. This is not yet dealing with the problem of justice, i. e. we are not talking about what kinds of remedies are available to the members in order to insure the just society. It is the

282

H. Garfinkel

problem of the just society that is under review in all talk about leadership as Parsons uses the concept. Leadership and authority are used to discuss the consequences of members’ actions with respect to effectiveness in the concerted production of desirable states. These concepts provide ways of describing how decisions that commit the members of a collectivity, with or without their consent, with or without their knowledge, are enforceable. When a person performs an authority function, he elaborates the features of the just society. To hold somebody responsible is to elaborate the just society. To be entitled to say, “you are responsible” elaborates whatever the normative order provides as justice, right, legitimate, correct. Valuable states of the system and leadership function is one pairing. Authoritative functions specify the rules of procedure whereby these valuable states are actualized. This is the second pair. Parsons proposes to think of leadership and authority as ways of institutionalizing positive responsibility for the conduct of the affairs of the collectivity. They are ways, too, of institutionalizing the rights to influence the actions of other members in the interests of effectiveness. But every collectivity will of necessity have ranges of actions in which members act “on their own responsibility” and pursue interests which are in some important measure independent of collectivity interests. Hence there is a third set of norms, norms of regulation. These are norms to which the members’ activities are subject by virtue of their status as members, but where the activities do not constitute either an unqualified response to leadership, nor does it involve clear compliance with directives that are issued pursuant of/to the authority. There are ranges of actions for which the just order of the collectivity and the authoritative directives do not make either sufficient or specific provision. Nevertheless, the activities are bound by the considerations of collectivity membership. Two such types of norms are called norms of regulation. One set consists of norms of budgetary allocation. These are norms in which the units receive a budgetary allotment, for example, but are given considerable discretion in how to spend the allotment which is then subject to only general policy review. The second type of such norms Parsons calls common law. This consists of a set of conventional normative practices arising out of the history of that collectivity – but of that collectivity’s own history – which are consulted in deciding the correctness of matters of discretionary action. While one might say “precedent” it is more precise in tying the practices to specific phenomena to refer to their features of common law, including the important features of their being unwritten, and in being “just understood”, of being subject to interpretation within the collectivity itself by those members who are for one reason or another, and in one way or another, empowered with the right to make the appropriate interpretations as to what that collectivity’s

Parsons’ Primer

283

history has “actually” been. To summarize: norms of regulation consist of a set of normatively historically legitimized common understandings that the members of the collectivity are expected to know as a condition of their claims to competence, i. e. as conditions of bona fide membership, which are used in handling discretionary action. These norms govern actions for which no clearcut normative specifications exist, in the sense of the just results, or correct procedural orders, for achieving them. So much for what Parsons called the institutional components of the polity. What are the parallel components for the economy? The value pattern for the polity consists of the valuation of effectiveness. For the economy the parallel collectivity value is the valuation of utility. By this is meant that a unit is committed in the use of facilities to the effective attainment of collectivity goals at minimum cost in the economic sense. Operationally, reference is being made to the ratio of output to input. The value that is being spoken of is the economical management of facilities so as to maximize output with respect to cost. Just as the value of effectiveness defines justice in the domain of political action, economical management defines for collectivity members economic rationality. On the one hand we had a definition for collectivity members of political rationality or justice. Now we consider a value scheme that defines what is for them, i. e. what is for them “really” – where really means in terms of the real social structures which are the social structures that the sociologist describes – economic rationality. At any level of economic reorganization, whether one is dealing with a pre-literate society or with the society that we are so much members of ourselves, members will have and be committed to their models of economic rationality. It is a peculiar feature of economic rationality in U. S. society that the model of economic rationality has peculiar mathematical properties that we recognize, for example, in the “substantively rational” applications and use of the procedures of capital accounting. With respect to the polity, we talked of the norms of leadership, of authority and of regulation. The parallel institutions on the side of economic action are contract, occupation, and property. These are normative economic functions with respect to the problem of integrating these aggregations so as to produce social structures as concerted effects of normatively regulated actions. 1. The economic institution of contract. Just as the institution of leadership is a fundamental mechanism for integrating units of the collectivity for affective goal attainment, the institution of contract is a fundamental mechanism for integrating a plurality of exchanging units to produce the features of a “market”, so as to facilitate the attainment of some level of a utility, that is,

284

H. Garfinkel

the attainment of “economic justice”. The concept “institution of contract” does a lot of work. What does it mean? (A) It does not mean the ways in which an actual settling of terms is accomplished. Nor does it mean the actions whereby a settlement of terms is accomplished. It does mean the rules whereby the actions which accomplish a settlement of terms are governed, the rules to which the processes of the settlement are subject. These rules, says Parsons, can shade off from very legalistic, formally written, explicit understandings on the one hand to the most informal kinds of understandings on the other. Neither feature detracts from the normative character. Such features mean only that these rules may have variable properties. Their normative feature is invariant to whatever these variable specifications may be, e. g. on the one hand of law, on the other hand of tacit common understandings. In this sense one would talk of a thief as a person who engages in an exchange transaction outside of the rules defining an appropriate exchange of such rights. (B) The norms regulate five things: (1) They regulate what can be legitimately subject to a contractual agreement. If transfer is the matter to be agreed upon they define a legitimate, transferable thing. (2) They regulate the activities that may be used in order to secure the assent of the contracting parties. (3) They regulate the consequences of defaulting on promises. An agreement is a promise in that it defines how the persons are going to act on future occasions. (4) They define the consequences of a change in expected circumstances while the contract is in force. They regulate the conditions under which the agreement must be fulfilled, given the contingencies of life. (5) They regulate the consequences of the agreement for “third parties”, i. e. for parties who are not themselves subject to the terms of that agreement. 2. The institution of property. By the institution of property is meant the norms that govern the transfer of rights to make use of, to allocate, and above all to transfer control to somebody else. This last is known as the rights of disposal, of disposing of. Parsons points out that property rights are primarily rights of disposal. If a critical criterion is needed to decide who owns what, then the critical thing to look for is who can get rid of it and expect to have this kind of an action upheld, i. e. have his right enforced. Slavery is a peculiar institution in that it involves the rights to dispose of the services of the person irrespective of, independently of, other ascriptive ties. Parsons confines property rights to rights to the use and disposal of possessions other than human services. If someone can take something of yours, could it be said with this definition that it was never owned because you do not have the right to get rid of it? Consider the arrangement in the Japanese Hell Ships. If you could not hold on to your clothing, or your food, or your watch someone would

Parsons’ Primer

285

come along and snatch it. The way of talking about this system of possessions in Parsons’ terms of a system of property rights is to say that these rights of possession have a very low likelihood of enforcement. You can claim your singular rights of transfer but you can’t enforce them except as you join a group of two or three other persons, each of whom takes his turn in order to see in fact that what you claim as your control of your material will be retained by you during the course of the night. In “more usual circumstances” the conditions that determine the probability of enforcement tend to rule out the explicit or bare use of physical force to effect the exchange. Instead, enforcement turns on something else, like an elaborate stratification system, with recourse to various kinds of adjudication agencies to decide what belongs to whom, i. e. whose rights are enforceable, and in the final analysis, by the courts, that is, by a political system, that can use force to insure a respect for these rights. According to Parsons, property rights in the domain of economic actions are parallel to authority in the polity. The parallelism consists in this. That similarly to the way it holds for the case of the authority function, that persons are entitled to make decisions that are binding upon the collectivity member, it holds in the case of a property function, that the holder of these rights is entitled to make binding decisions with respect to the object of his rights and to enjoy some range of discretion as to what these decisions will be. Conceive a person who has property rights as being in the situation vis-à-vis other persons that his rights literally consist of the rights to control the alternative actions of the other persons with respect to use, disposition, disposal, and the rest of this particular possession. 3. The institution of occupation. An occupation consists of a role performance with respect to the production and allocation of goods and services. The institution of occupation consists of rights to, or rights in, the role performances of persons or collectivities. These consist of rights to legitimate categories of services. That explains the use of the term occupation, in the institutional sense. The term “occupation” is reserved to talk about rights in performances, when emphasis is upon performances. When these roles are spoken of in terms of a specific contract of employment, Parsons prefers to speak of the institution of employment. Employment consists of the voluntary acceptance of membership in a collectivity on negotiated terms which subjects the employed person to the institutionalized orders of the collectivity. It subjects them to the order of leadership, authority, and regulation with respect to the values of effectiveness in attainment of the collectivity’s goals. Therefore, says Parsons, school children, prisoners, patients, and martial partners cannot be properly spoken of as occupational categories, or as employment, because they do not involve services the allocation of which can be controlled

286

H. Garfinkel

through a contractual relationship in which monetary compensation plays an essential part. The voluntary nature may be there, but not the collectivity membership to which obligations are assumed as the terms of employment in return for compensation. That does not mean that one might not analyze a marital arrangement as employment. Indeed the investigator who chooses to use Parsons should as a matter of method analyze it as employment because the incongruities that result would be structural incongruities. Obviously the terms occupation and employment need not be used in this way, but Parsons could not be less interested in terminological choices. Parsons now talks about money, and the institutionalization of money, and furnishes a normative definition of money, i. e. of norms governing its use. His argument is as follows. To begin with, there already exists in economic tradition a way of talking about two aspects or meanings of money; money as a measure of value; and money as a medium of exchange. From the standpoint of the use of these concepts in Parsons’ theory of social systems, money is a special case which involves both the institutions of contract and institutions of property. As a measure of value, the institutionalization of money is a special case of the institutionalization of contract. As a medium of exchange it is a possession and therefore rights to it are institutionalized as property. Parsons explicates the concept of institutionalized features of money as a measure of value. He proposes to follow Durkheim and say that the institution of contract consists of a set of norms which govern, or which define, proper procedures for arriving at agreements, i. e. that define proper agreements. Five points were previously enumerated in this regard. Parsons says that there are two features of contract that need to be discriminated. One is the system of norms as such: the second is the way in which money in use as a measure of value is such that the norms which define this peculiar measure of value control behaviors of persons. The particular kind of behavior that is being controlled or regulated is the behavior having to do with the allocation of resources. With respect to the system of norms, he points out that the normative definition of money is first of all part of the legal system; that these norms are administered in their sense by the courts; that they involve defining the interests of third parties, e. g. through the definition of fraudulent money, for example; it involves definitions of prohibited contractual agreements, of means of securing assent and of the devices for handling consequences of risk and uncertainty, or better, for handling risk and uncertainty. All these are relevant. Parsons gives as an example that an offer of money, for example, is a legitimate means for inducing acceptance of a contractual proposal and that

Parsons’ Primer

287

there is an obligation to accept “good money”, that is to accept money having the status of “legal tender”, that it is a device whereby agreements can be made final, and it is also a device whereby contractual obligations can be legitimately fulfilled. With respect to the institutionalization of money as a medium of exchange, i. e. with respect to the regulating of behavior by controlled allocation of resources, Parsons mentions four features which make up the institution of money as a medium of exchange. There is first a feature that money has in that it is available for contractual commitments. The feature is that money resources are highly mobile in the sense that the status of legal tender does not attach to the person who holds it. This “mobility” of resources, mobile in the sense that it can be passed around from person to person without altering its status as “legal tender”, i. e. without altering in its status as a bona fide measure or value, is morally sanctioned. Persons come to expect that whatever are the devices whereby the presence of money is signalized, whether it be via coins or paper or whatever device, is regarded by users as a morally sanctioned feature of the money. Parsons refers to these features as the institutionalization of the mobility of resources. A second institution concerns the fact that as a measure of value money represents a standard in terms of which different actual services and commodities can be commonly evaluated. Parsons argues that it is a definitive phenomenon of a contractual agreement, that there is established between the two parties, what is called a ratio of exchange. The ratio of exchange consists of the value of the commodities (or services) offered on the part of one, over the value of the commodities that are offered on the part of the other. The resources themselves that are being exchanged may not “intrinsically” be evaluated by a common standard. The institution of money as a measure of value consists of this: that it is morally correct to assign to both sets of resources goods and services, their value in terms of a monetary standard and to treat the ratio of exchange in these terms. All goods and services according to this institution are in principle comparable. They can be compared with each other and the comparison can be expressed in terms of a single monetary unit. These are the features of actions that are institutionalized. The institution of money as a measure of value has another feature that consists of another kind of enforceable moral characterization. This feature is concerned with the responsibility for the management of the monetary unit and with the definition of what a proper monetary unit is. Parsons asserts that it is the case in every society that the responsibility for defining and managing the monetary unit is in every case a matter of public responsibility. In no case is it a matter of private enterprise. Finally, the three properties are governed in turn by another insti-

288

H. Garfinkel

tution which makes the acceptance of the measurement calculus mandatorily legitimate. (See Parsons’ paper.)86 With respect to money as a medium of exchange, what are the things which are allocated? Parsons points out that it is not money that is allocated, but commitments represented via the use of money as the medium that communicates these commitments. He makes a statement that is an odd one to common sense. Because money is a cultural object, the social object that is transferred from one point to another, i. e. the “existence” of the money, is not found in the token but is found in the message. Thus if we talk about the coins in my pocket they consist as money of what Parsons calls a set of contingent promises backed by a political order and that what these contingent promises consist of, or are in effect, are the rights to exchange these coins for a set of “real commitments”. The over-all system has the very interesting feature that the system is itself guaranteed precisely to the extent that the participants use money in this way. To the extent in fact that there is the expectation backed by the actual use of these tokens as contingent promises, the system of reciprocal rights and obligations, with respect to honoring the moral character of these contingencies, is thereby enforced and enforceable. The valuable character of money consists precisely in the likelihood that it is going to be used. The guarantee of money’s status as legal tender consists in the probability that if you offer it for payment, the other person will accept it. The real value of money depends upon the fact that the government is able to enforce the acceptance of it for exchange transactions. The government might say its value is guaranteed by putting gold in a hole in the ground in Kentucky. It would not make any difference for Parsons’ money how much was in the hole in the ground. Instead, the real value of money, at least in our society, consists in the actions of political trust, and that the generalized character of money would hold with or without gold in the vault. At the level of generality of control money is “above any real commitments”. Parsons is saying that of the various objects of possession, money is the most mobile, and the most omni-competent of any of them in that it can control an immense variety of real commitments. Therefore, with respect to the processes of controlling the allocation of goods and services Parsons assigns to money an over-all control function superseding any of the lesser ones, that is, superseding any of the more “concrete”, or any of the more specific commodities and services that it can be used in exchange for. 86 [Editor’s Note: Garfinkel is probably referring to Parsons paper on influence and money, which Parsons recommended that Garfinkel read in a 1963 letter. The paper was published as Parsons, T. (1963). “On the concept of influence”, Public Opinion Quarterly 27(1):37–62.]

Parsons’ Primer

289

Parsons proposes that the person who accepts money makes a commitment. It is a peculiar kind of a commitment because, among other things, the person who accepts money accepts it without meeting the conditions on the part of the one who signs it over to him as to when he will in turn cash it in for any particular goods or services; nor does he specify what goods and services he is going to cash it in for; nor does he specify the time at which he is going to cash it in; nor does he specify the price that he would be willing to pay for any of the transactions that he might use it for. Parsons refers to this singularly important feature as relieving money of its ascriptive features. The model of a generalized medium of exchange is represented in this kind of an institutional arrangement whereby tokens are used in order to obtain from persons unspecified, unspecifiable commitments to the use of future goods, to future transactions that they will engage in, under what conditions, with what persons, and the rest. Money is therefore a very interesting control device because in its status as a set of contingent claims upon others it literally relieves persons of having to specify what the social conditions will be under which they will engage with others in the practices whereby the means of life are exchanged. Parsons says that this feature has led some of the more literary types to talk of the materialistic character of money and to say that it does not represent in the activities of the person a commitment to “values”. Parsons proposes that this formulation is incorrect. The question is not whether or not there are values involved, but what these values are. On this score Kenneth Burke talks about the operations of the Federal Reserve Board in deciding for example such things as interest rates, production rates of new money, feeding new money into the market and the rest as a theological enterprise. The members of the Federal Reserve Board are literally engaged in discussion of the condition of credit, honesty, justice, literally, good membership and the good society. In an exchange transaction in which the commitments of employment are accepted on the one hand in exchange for money payments, the person being employed makes one commitment in that he accepts employment, but thereby in insisting upon money payment, in requiring of the exchange partner to pay him wages, he forces the exchange partner to allow him freedom to decide the circumstances under which he will make his other real commitments. By virtue of entering into a transaction in which services are exchanged for money payments, the employer obtains a real commitment from the employee, but the employee literally obtains a freedom from the employer with respect to the alternative collectivity obligations that he can assume, or that he can enter into. This occurs precisely because the terms of employment provide for money wages. But, says Parsons, this freedom has its “price”: the employee commits his labor by alienating any further commitments in that

290

H. Garfinkel

area. Other commitments remain contingent upon circumstances which at the time of the transaction are not programmable, i. e. that he cannot specify. He accepts money in return for which the employer exacts a priority of loyalty to the membership in the collectivity in which employment is obtained. The institutionalization of power. What money is to the economy, power is to the polity. Just as transactions involved in the flow of money provide a generalized means for magnifying the output of the economy, and for integrating the activities of these economic units, power similarly is a unit that has properties for the polity of a similar sort. To the distinction between money as a measure of value and money as a medium of exchange there is a corresponding distinction for power. Parsons argues as follows. He defines the political function as the mobilization of resources of the system in the interest of effectiveness in the attainment of system goals. The capacity to mobilize such resources is known as power. This requires a definition of effectiveness. Effectiveness is defined as motivated conformity on the part of the units to norms which define the allocated contributions to the members in the accomplishment of collectivity tasks. In the political case, says Parsons, a problem arises of the following sort. Given the imperatives of the accomplishments of tasks, of the system as a collectivity, its performances somehow or other have to be effectively organized by bringing these units to make commitments to the tasks’ accomplishment, i. e. commitments to performance. The problem of effectiveness consists of organizing these units or bringing these units to make commitments of cooperation. The control problem, of ensuring these commitments to cooperation, must go to the problem of how the rights that the units have to make commitments to for the accomplishment of tasks that are allocated to them, can be freed from ascriptive criteria so as to make it possible for these commitments to be disposed of, to be allocated, to be arranged, to be distributed, in light of whatever the collectivity obligations may be. The big problem of ensuring the kind of cooperation that will magnify effectiveness is the problem of producing the mobility of resources. How to achieve the mobility of commitments on the parts of members to do in effect what they are required to do when you can’t or don’t want to say, or be bound beforehand by, what may be required of them. Parsons says the problem is solved if there is achieved an alienation from any ascriptive rights. That’s the trick, but that’s the problem too. If it is to be done one would need some generalized medium, as in the case of money, that would be capable of influencing the way in which these commitments are obtained. Remember, these are commitments to performances, i. e. commitments to honor collectivity obligations relative to collectivity goals.

Parsons’ Primer

291

Parsons sets up a parallelism of these mechanisms to the economic mechanisms, and talks of the different levels of ascriptive of resources, of political barter, and the development of general means of political command of resources. Then he asks, “What are the minimal institutional conditions for the existence of that kind of a political control of resources in which the very exchange of commitments are governed by transactions in which the power to, or the rights to allocate these commitments in any way whatsoever, are also determined?” The following are three minimum conditions which define generalized political control. These are conditions under which a political control which has the same properties as money is possible. First, the measure of effectiveness must be a measure of the relative contributions of performances with respect to the attainment of collective goals. Second it must be a generalized medium which is capable of influencing such commitments without “giving anything in return”. That is, it has to be such that it can be enforced, or can be employed, without engaging in the barter agreement, “If you do this for me, I will do that for you, at this time and that place.” Third, it must communicate the expectation that commitments to cooperation which would be made in the future will be made without either party being specific as to the nature of what they will have done for each other at that time. The following are the conditions under which this generalized political control can exist: First, and similarly to the economic case, resources, in the form of services that persons will perform, must be obtainable in the form of a promise without respect for the particular context under which the services are going to be used. The emphasis is upon functional specificity of a performance with persons being hired for their skills, without respect to any conditions of a particular time or circumstance, i. e. without any ascription built into it as to the conditions under which these skills will specifically be employed. Disposability is another essential condition. The matter that must be institutionalized is the allocatability of skills in accordance with the demands of the task, with the demands of the task remaining unspecified at the time that these commitments to act in their accomplishment is given. It is possible to see from this why Parsons talks of the system of occupations as an adaptation system, for it is directed to the mobilization of resources without respect for here-and-now specifications of time, place, and circumstances under which these resources are going to be employed. Further, there must be some common standard for the evaluation of these performances. Finally, this common standard must be concerned with the assessment of the strategic importance of the contributions of the unit to over-all effectiveness. This is where the stratification system enters in its entirety. Strategic

292

H. Garfinkel

importance means that there must exist some body of common understanding between these persons as to effective use. It is the same as saying that there must exist either an understanding to begin with about how the whatsoever goals to be accomplished are to be programmed in their accomplishment, or a way of enforcing upon the parties a rationale in terms of which the effect, or the strategic character of what they are doing, is defined. Not only must the boss give directives and others must receive them, but the boss must also be in a position to justify the use in terms of a substantively rational account of how these skills have been, are being, or are to be employed. In these terms a person’s power consists in the likelihood that he can obtain from the relevant units a commitment to cooperation in a collectivity with respect to the accomplishment of its tasks, in preference to alternative loyalties and alternative obligations to which that unit’s commitments might be made. Parsons goes further, but we shall not discuss some of his further comparisons between power and money. Parsons points out finally that the activities of voting and “the vote” represent for the polity the use of a medium which has properties that are cognate with money. The recipient of the vote does not receive a real commitment from the voters. Instead he is given the expectation that he will be able to use his power to secure real commitments from them in the future.

CHAPTER VIII PARTIAL DRAFT OF MATERIALS ON 1940 ARTICLE ON STRATIFICATION: AD HOC PARSONS Parsons’ specific discussions of social stratification are found in his two articles, Analytical Theory of Stratification (1940) and Revised Analytical Theory of Stratification (1953). Persons that have used Parsons’ material on stratification generally prefer the 1940 article for the reason that the 1953 article is a model of difficult reading.87 This is unfortunate because the later article is unquestionably the more important and useful of the two. I shall discuss each in turn.

87 [Editor’s Note: Parsons, T. (1940). “An analytical approach to the theory of social stratification”. American Journal of Sociology 45(6), 841–862, and Parsons, T. (1953). “A revised analytical approach to the theory of social stratification”. pp. 92–129 in R. Bendix & S. M. Lipset (eds.), Class, status, and power: A reader in social stratification. Glencoe, IL: Free Press.]

Parsons’ Primer

293

There are several essential ideas to keep in mind in reading either article. (1) In answer to the question, why an interest in stratification in the first place, Parsons proposes that a system of stratification is a theorist’s solution to the problematic phenomena of power and influence. In discussing stratification Parsons is talking about this problematic phenomenon. Stratification systems are devices of social control which in Parsons’ terms minimize the illegitimate use of power. All his arguments work to that point. (2) The social system is in every case a system for the control of activities. (3) A stratification system consists of a system of moral ranks. If one asks the meaning of “moral ranks”, what do they do and why be interested in them, Parsons’ answers elaborate the idea that moral ranks are control devices, whereby questions of entitlement are settled. Later we shall see what these entitlements are all about, why there are brought in and how they operate. (4) All cultural objects, no matter what they are, are parts of a system of evaluation. Therefore Parsons is not talking about or restricting his talk to persons. He is talking about any properties at all that can be assigned to persons or any other objects, properties that come under evaluative judgments. Such judgments are directed to all types of cultural objects. Physical objects are not of interest to Parsons except as they too are brought into the system of socially controlled treatments and evaluations. Insofar as they are of interest in this way but only in this way they enter into Parsons’ discussion of stratification. I shall just enumerate some major difficulties that are encountered with reported stratification studies, then summarize what Parsons promises by way of remedies, and then go through each of his articles in turn to show how he has done it. Ruth Kornhauser discussed seven sets of controversial issues in stratification studies.88 One set is concerned with the usefulness of Warner’s definition of class. The principal criticisms that she lists may be paraphrased as follows: (1) that whereas Weber discriminates three different phenomena that would be handled under the topic of stratification, i. e. class, status, and party, that Warner confounds several of these. Thereby investigators are prevented from determining the effects of each of these “separate factors.” She cites Mills’ criticism that these are three different “bases” of stratification, and that there are differences as well between these as actual and accomplished orders of stratification and persons’ awareness of these orders of stratification. Finally she cites 88

[Editor’s Note: Garfinkel is likely referring to Kornhauser, R. R. (1953). “The Warner approach to social stratification”. pp. 224–55 in R. Bendix & S. M. Lipset (eds.), Class, status, and power: A reader in social stratification. Glencoe, IL: Free Press.]

294

H. Garfinkel

Mills’ big criticism that contrary to Warner’s proposal it would be incorrect to use a person’s opinion of a stratified order as a substitute for conceptual distinctions. As we go through Parsons’ articles it will be seen that Parsons has incorporated these three distinct phenomena into a single more or less unified conception. Although he provides for something like “separate bases” he addresses the question of how, given the fact that members of a society may be differentially stratified on the three, a “generalized continuum of ranks” is nevertheless achieved. Kornhauser’s second set of criticisms is concerned with the fact that Warner’s conception, while it includes several “dimensions” of stratification, emphasizes the prestige dimensions. Some of her criticisms run that Warner has not described the class structure but only a system of prestige classes; that Warner has gratuitously abandoned the role of economic factors; that his discussion is concerned only with a limited number of problems which are conventionally associated with inquiries with respect to class. A third set of issues which Kornhauser cites on behalf of others, center on the point that Warner has not adequately demonstrated the relevance of his studies of prestige classes for the tasks of understanding the general pattern of stratification in the United States. The type of class structure Warner describes for the United States depends upon the ability of members of society to make status evaluations of each other’s conduct. Some approximation to this occurs in a small town, but in large cities there are likely to be several disparate hierarchies, not one overall hierarchy of ranks. In the metropolis, power hierarchies and the formation and activity of groups in the associated pursuit of their interests, do not lend themselves to the kind of analysis that Warner suggests, for it is inconceivable to many students that persons are in the business of living in order merely to obtain the benefits of prestige. A fourth set of criticisms are directed to the point that Warner has not portrayed accurately the status structures even of small towns, that what he has described is the status structure as it appears to upper-middle and upperclass residents, and that his descriptions reflect the perspectives of some of the members of the society, but ignore the complex aspects of class relationships. A fifth set of criticisms charge that Warner has not given sufficient attention to historical data. Two points of criticism on this point are that the question of how class structure has changed over time does not come out of his analysis, and that his descriptions lack any reference to strains and conflicts and to the phenomena of social change that are attendant upon strains and conflict. A sixth set of criticisms deal with Warner’s methods, that the factors he includes overlap in many things they are talking about; that no credible method was proposed whereby differences between the evaluators were reconciled

Parsons’ Primer

295

when they arose; that the method of evaluative participation is impossible to use in the cities; and that it is not clear whether or not the index of status characteristics rates objective socio-economic position, or rates evaluations on the parts of subjects of these objective positions. The seventh set of criticisms are directed to the charge that there are value judgments implied in some of Warner’s work, that in delineating a rank system he stresses those factors which tend to make for the status quo, that these are features of the status quo that he values, and that he tends to endorse the views of the higher classes. It is not my intent to criticize Warner but to follow Kornhauser’s view that these are controversial issues. Therefore the controversies over Warner’s work describe important dissatisfactions in the area of stratification studies. Several more points of general dissatisfaction can be cited. We must assume that whatever Parsons has proposed as a theory of stratification is intended to afford some solution to the points of dissatisfaction. One is the general lack of clarity that is found in the field between the actor’s and the observer’s point of view. This point was touched upon several ways in Kornhauser’s list. A second point concerns the circumstantial character of the criteria that sociologists frequently use in achieving a description of stratification systems. By circumstantial I am talking to the point of the fact that the criteria that are used seem to be highly related to particular historical occasions. Thus they tend to have a rather concrete character with the result that as the investigator in his studies moves from one part of the country to the next, he frequently finds that the criteria are not easily translatable between settings. So it is said of these ranking criteria that they lack “universal” character and therefore the ranking criteria that are used to describe ranks in our society can not be used to describe ranks in other societies. Many further problems arise where perceived criteria are used. An important problem consists of the so-called “mosaic” effect. In administering schedules that ask for persons’ perceptions of rank, a rule is rarely provided that would permit the observer to show how, if at all, criteria in use by participants in the society, are instances of a class of criteria that generalized definitions of class and status would provide for. Another important difficulty comes from the use of perceived criteria where the “empirical” properties of these perceived criteria frequently turn out to be a function of the sociologist’s method for evaluating his information. Most particularly does it occur when scaling methods are used to construct a single, general continuum of rankings. Finally there is a difficulty that plagues investigators, particularly insofar as they must depend upon verbal materials. It consists of the differences so invariably found between criteria of class that are obtained as subjects’ imagi-

296

H. Garfinkel

native productions, and the criteria of class as that are obtained by consulting persons’ everyday activities. These issues can be summarized under a few headings. What Parsons is doing in his theory of stratification is to propose solutions to the problems that would be found under each one of these headings. The first one has to do with the general dilemma in stratification theory of the difference between what we refer to on the one hand as the phenomenal stratified order, and the “real stratified order”. There runs through all of Parsons’ later theorizing, most particularly since The Social System, the notion of the real social structures as consisting of institutionalized patterns of normative culture. We should expect therefore when we come to deal with stratification that the same rule is going to hold there. This rule is used to solve the problem of the difference between the actor’s and the observer’s point of view which is pointed up in this dilemma in the difference between these phenomenal and real orders. A second point that might settle some of these controversial issues is that whatever else the system of stratification consists of, Parsons is not about to treat it as subjective. A stratification system does not “exist” “in the heads” of people. On this point he is very definite. But this is not the same as saying that the features of the real stratification system “exist” independently of the phenomenon that persons are aware of differential rankings. The presence of a stratification system, the existence of a stratification system, is not demonstrated by looking into the heads of people. The observer cannot determine its existence by consulting what persons “have in mind.” It is perfectly true for Parsons that to learn about the stratification system the investigator will have to talk with people, but if for example, no one were to tell him, if they were all to keep their “perceptions” as secrets, this does not mean that the observer would not learn about the stratification system. Indeed, it might be a good idea if persons would keep their secrets from some of the studiers of perceived stratification so as to return the investigator’s attention to the world of activities. The existence of a stratification system is not decided by something that exists in the heads of persons. This is not the same as saying however that its existence can be handled without respect to the fact that persons are aware of the existence of a stratified order. The argument turns on what Parsons means by saying that a stratification system exists. He means exactly what Weber means when Weber says that the existence of a legitimate order means exactly the same thing as its validity. An actual stratification system consists exactly in the probability that a system of perceived ranks as a legitimate order is actually enforceable in the actions of persons. Exactly and entirely in the likelihood that perceived ranking systems are morally enforced in the activities of persons, do we thereby encounter the properties of the real stratification system.

Parsons’ Primer

297

The third set of problems have to do with what Mills proposed as “different bases of stratification.” These “different bases of stratification” in Parsons’ formulation consist of the four functional problems. Fourth, a set of problems can be summarized under the phenomena of incompatible rankings that occur as one encounters “different types of groupings.” Parsons refers to a general phenomenon that he calls “the spread”. It consists of the prevalence with which ranking systems that occur in these four functional areas are incompatible. That this occurs constitutes a set of problems for the sociologist in achieving “a general prestige continuum”. But Parsons conceives these problems differently from the way they are conventionally formulated. Parsons proposes that the problem of achieving a general prestige continuum is the sociologist’s problem because it is the problem of persons in the society. So for Parsons the very presence of the sociologist’s problem of spread is a theoretically interesting and theoretically most important empirical datum about the society that a theory of stratification must come to terms with. Another set of problems consist in this: if there is any single variable that a sociological study pays off on, stratification is it. Let a sociologist consider almost any phenomenon. Should he find himself at a loss as to what he should take as his “independent variable” let him use stratification. Most particularly let him use a list of items like occupation, income, education, friendship arrangements. The disconcerting thing is the sheer number of “effects” that will correlate with differences on this score. This is not the problem of the spread. The problem is this: given a variety of such factors – Kahl89 proposed an eclectic list of them, grouping them as personal prestige, occupation, possessions, interactions, class-consciousness, and value orientations, and said that these form a “conceptual scheme”, proposed that they describe the reality of stratification, that each “variable” can be measured by distinct operations; that each can be therefore used to stratify a population; that these factors are mutually dependent; and that he chose to organize the materials of his book around them because they most efficiently organize the existing empirical data, and that the history of stratification theory is a history of shifting emphasis upon one or another of these six factors. From Kahl’s proposal it appears that despite the variability in these factors, a relationship seems to be intended by one theorist or another. Despite the enumerated list in one way or another, the intention of the sociological theorist is to tie them up together.

89 [Editor’s Note: This probably refers to Kahl, J. A. (1957). The American class structure. New York: Holt, Rinehart, and Winston.]

298

H. Garfinkel

Parsons would agree that some such set of factors must be tied up together. He refers to the “tying” as the problem of “interlarding”. Imagine rankings on occupation, on income, on interaction, on possessions, on value orientations and now interdigitate these ranks so as to produce a related set of ranks in such a way that it is possible by reason of the “reduction” to see through the list of them. Parsons is employing here the same device that he used on the problem of “spread”. He proposes that the very difficulties in achieving an interlarding that the theorist encounters are themselves peculiar to the social system. If the theorist has “methodological” difficulties, it is because the members of the society have the same difficulties. How an interlarding is possible is to be learned by the sociologist by consulting the stratification work in the society itself. So for Parsons the properties of an interlarded system of different criteria that produces different ranks is itself a theoretically important empirical datum about the society and needs to be handled in a theory of stratification. Finally, an issue that summarizes many of the controversies, has to do with the question why be interested in stratification in the first place? Parsons would answer that obviously these rankings correlate with an immense number of phenomena, but he would part company with eclectic and practical understandings of “obviously” and the immense number of correlations. In Parsons view, stratification “variables” correlate with so many phenomena because the phenomenon of evaluative judgments itself is nothing else than, is an integral feature of, the very conception of social action in the first place. Therefore, why differential evaluations should correlate with everything in sight is because persons are continually making, enforcing, modifying, revising, differential moral judgments. Indeed Parsons would go even further and say that accomplished studies for all the variety of factors correlated with “class” have not picked up the least of it. For all the factors that have been reviewed in the textbooks, only the surface has been skimmed. Nor is there much chance of an end to surface picking unless one gets systematic. And this leads to an answer to the question, “Why stratification?” What is the term good for? Parsons would say that stratification criteria and stratification activities – i. e. “the stratification system” – is to be conceived as a solution to “the functional problems of a society.” Our previous discussion of structural-functional method proposed that this method solves for the sociological theorist the problem for him of deciding the necessarily related character of a set of phenomena. The theorist has a choice as to the kind of structural-functional method he may use. Parsons’ solution is a peculiar one since it starts with the conception of a stable set of activities. The so-called functional problems are conditions such that actual activities conceived as solutions to these functional problems provide the theorist with the required statements he has to make about the stable

Parsons’ Primer

299

character of these activities. Thereby the necessarily related character of any set of phenomena are decided. Parsons and others, Kingsley Davis for example, stand in marked contrast in what each regards as the proper procedure of a functional analysis of stratification. This point needs to be heavily stressed. Finally, “Why stratification?” asks what the problematic phenomenon itself is that one should be concerned about. Parsons would say the phenomenon is the legitimate use of power. A stratification system is conceived as a mechanism of social control. As a consequence with respect to the problem of conceiving the possibility of stable social structures, stratification minimizes the illegitimate use of power. Three points are to be made by way of immediate introduction. 1) For Parsons, stratification is a formal property of all social systems. The investigator’s question therefore is not whether there is or is not a stratification system, but only what does it consist of. This is so because social actions, by the very terms in which Parsons conceives social actions, involve evaluative treatments of objects, as a parameter of the “unit act”, as a definition of a possible observable event. Therefore, Parsons argues that the common alternative which thinks of a system of equal rankings as if evaluations were irrelevant would be a theoretical error in his system. Parsons would propose that such a formulation is based on an error of fact. 2) The phenomena of stratification are intimately linked to Parsons’ notion of the integration of a system. Therefore, if one asks “Why stratification?” the important “so what” is a matter that Parsons continually confronts us with when one asks what Parsons altogether is up to: the “so what” of all his intricate theorizing, consists of an answer to the possible stability of a set of activities. (3) If we ask how stratification is tied to the social system, its link lies in the fact that a system of stratification is a solution to the problem of the legitimate exercise of power. Now for the articles themselves. The 1940 article is an excellent instance of “virtual” as compared with “actual” Parsons. By virtual Parsons I mean that a use of the 1940 arguments would not use what Parsons considers to be his unique contribution to sociological theory which is the application of the social system in the interests of a solution to the problem of social order. The 1940 article therefore is virtual because though it contains a collection of very usable ideas, the social system is entirely absent from the 1940 formulation. Parsons begins by proposing that by stratification he means to be talking about differential rankings. He proposes that differential rankings are a fundamental phenomenon. If one says that differential ranking is an essential phenomenon then what would be meant? He answers it by proposing that

300

H. Garfinkel

differential ranking is identical in sense with the differential moral evaluation of individuals. As an invariant characteristic of their activities, persons attend each other differentially as morally worthwhile persons, and this differential moral worth of persons is displayed behaviorally via the attitude of respect. In contrast to the attitude of respect one would encounter the attitude of disapproval and says Parsons, in extreme cases, “indignation.” Parsons says that the stress on moral evaluation is not something that has been invented or discovered by sociologists. Persons in the society are practical experts on these matters and teach the sociologist about this. How do they teach the sociologist? They teach them, says Parsons, by the indignation that these persons themselves show upon occasions in which they detect conspicuous errors in the moral rankings. Not only do they show indignation when persons are misplaced – for example, by being treated either better or worse than they deserve – but they insist that such anger be itself treated as a normal reaction. Persons therefore are entitled to get angry for example if you treat them in the fashion that they are convinced with respect to the order of moral ranks, is either better or worse than they deserve, i. e. that they feel they are entitled to. By way of illustrating this I had undergraduates in my course go into stores, approach customers, and treat the customers as though they were clerks, while giving no indication to the customer’s response that anything that was going on was out of the ordinary.90 Their protocols showed as a massive effect the acute disturbances, anger, embarrassment, anxiety, flight. Interestingly, the experimenters had difficulty in carrying off the procedure, and many preferred to work in pairs because it afforded great relief. Thus not only was the subject disconcerted but the experimenter as well. This would be an instance of treating a person with respect to the scale of moral ranking, and treating him “worse than what he deserves.” In summarizing the points about why stratification is a fundamental phenomenon Parsons argues as follows: First, in the conception of a normative orientation to action, moral evaluation is built in by definition. Not all normative patterns are objects of moral respect, says Parsons. But no system is stabilized, or can be stabilized unless a moral order is an integral feature of the actions of which the system consists. Therefore, every system no matter how trivial it may be, in so far as it shows stable features, must be counted as being stabilized by the presence of a system of moral evaluations of the personnel of that system to which the personnel comply. A word about the phrase “moral evaluations of the personnel of that system.” Speaking literally, for any mem90 [Editor’s Note: This study is reported in Garfinkel, H. (1967). Studies in Ethnomethodology, Chapter 2:68–70. Englewood Cliffs, NJ: Prentice Hall.]

Parsons’ Primer

301

ber of a Parsonian type of social system, that system’s personnel do not count for equivalent value. According to Parsons, there are not only worthwhile members in any system of action, but there are those members who, with respect to the vital interests of that system’s members are not worthwhile. In a phrase Parsons actors live in a society in which some people plainly stink; there are always, for them, good reasons why this is plainly so; Anyone knows it; Anyone can see the objective evidences for himself; and one loses sight of this at the peril of losing a grasp on social reality. The conviction that all men are of equal moral value is believed by no one and acted upon by no one who inhabits Parsons’ societies. Parsons makes a second point about why stratification is a fundamental phenomenon: because of the importance of the human individual in every actual society. There are no systems of activities which are populated, at least for the purposes of empirical examination of actions, with other than individual persons. So for the purposes of describing the phenomenon of stratification the human individual is the site of the activity that goes on from the standpoint of the members of the system. Their environments are populated by human individuals. Given the phenomenon of moral evaluation, and given individuals as features of the social system, then, says Parsons, these individuals are going to be morally evaluated with respect to all kinds of features, none being in principle ruled out as possibilities. In the 1940 article Parsons grouped these possibilities as qualities, acts, and so on. In the 1953 article he got systematic and said very specifically what features they could be evaluated on, what features had to be evaluated, and what these evaluations had to consist of if the concerted activities were to show stable features. Because in any social system there will be an actual ranking system in terms of moral evaluations on the parts of the participants, and given the problematic stability of concerted actions, the inference that Parsons insists is thereby necessarily entailed if the judgment of stability of actions is to be made, is that the standards that are used in these evaluations consist of an “integrated” set. In terms of the integrated set, rankings are accomplished and enforced. Parsons belabors the reader with talk about a “complex” of standards, of a “system” of standards. He was referring to some kind of a related set of such grounds of judgment or standards of judgment, but in the 1940 article despite the insistence upon the related set, little provision was made to permit its description. Parsons then introduced the difference between (1) a scale of stratification and (2) a system of stratification. The distinction is of exceptional importance in the use he makes of these concepts. A normative pattern of moral ranks, that is, the definition for the members of the system of legitimate,

302

H. Garfinkel

sanctionable moral ranks, is called the scale of stratification. It is the normative description of moral ranks that the members “subscribe to.” The actual activities whereby inferiority and superiority are recognized by the observer, but only the observer – because his actors show to the observer the patterns, behaviorally speaking of deference, care, respect, feat, awe, etc. – the actual rankings Parsons proposed to call the system of social stratification. The difference between the scale of social stratification and the system of stratification runs throughout this article, and throughout the revised analytic theory. The scale of stratification is characterized says Parsons by moral authority. From the standpoint of the members of a stable society, the scale’s moral authority is normally, which means obviously as well, a part of the institutional arrangements of his society. Therefore, says Parsons, the proper study of stratification consists of the study of social institutions. The actor treats the morally legitimate ranking systems as an object of his interest. Frequently therefore, he treats it as a goal of his actions. It was already pointed out that the things that the actor may therefore be responsive to, as bearing on the morally respectable character of ranked “things”, includes persons. But the attitude of moral respect includes also attitudes towards the persons, social relationships, and Parsons included in the 1940 article “things.” In the 1953 article the list of objects that were capable of being morally evaluated and thereby of becoming constituent elements of a morally legitimate stratified order, was extended and systematized. In the 1940 article Parsons insisted that evaluative judgments were not confined to judgments of persons but extended as well to the attitudes, or what Parsons later came to call the orientations, i. e. to the courses of action, the treatments, and attitudes, and so on. He points out that obviously any of these objects need not have a moral significance for the persons treating them, and he compares that point, then, with objects that he said were of hedonic satisfaction, objects of affection, objects of hatred, and the rest. At this point he introduced the notion of the institutionalized integration of motives and values. This was done via the concept of the “normal actor as an integrated personality.” Parsons says of this “normal actor as an integrated personality,” who is a member of the society, that the things that he will value are also the things morally that he desires as sources of hedonic satisfaction. This can be illustrated with Harry Stack Sullivan’s introduction91 to his discussion of erotic gratification when Sullivan said there is no society which does not impose controls over the ways in which persons use the genitalia for erotic gratification. 91

[Editor’s Note: Garfinkel may be referring to Sullivan, H. S. (1955). The interpersonal theory of psychiatry. New York: Tavistock Press.]

Parsons’ Primer

303

Parsons would add to this, “And not only that, but he cannot be gratified either except as he achieves these sexual contacts under the jurisdiction of a moral order.” To be called by Parsons a normal person, which is to say an integrated personality, the motivation to the achievement of these gratifications must be carried on under the auspices of, and the control of, a moral order. Thereby, the scale of stratification reaches into the bedroom. Parsons’ actor, not only in the stable society but in the society at every point short of the ideal terminal state of complete anomie, is in no way a man “on his own.” Lonely, perhaps, but not and never “a loner” in anything but a figurative way of talking. The moral order that he attends is very little subject, for the features it has for him, to his singular, particular, personal, subjective interpretation. The moral order has an objective character to it in this sense: that it is out there to be seen by the actor as something that is not a matter of mere personal opinion but is known by Anyman. Therefore the judgments that Parsons’ actors make, and the rank order that they attend, has also therefore this character of being realizable according to a set of rules of judgment that all competent members are expected to subscribe to. Therefore, even though Parsons makes full provision for the “perceived” or the “phenomenal” character of this rank order, it has as a most important characteristic that it is known to the actor as something that is not a matter of his personal opinion, but is objectively there for anyone to see. In Parsons’ later work it comes out with great clarity that the objects in the actor’s environment, which include, among other things, purposes in the world, laws, gods, authorities, and the rest are in the world for the actor. Parsons’ actor consults his own competence and the competence of others in recommending as fact that this is so. Parsons points out that it is a condition of the actor’s self-respect, his competence, of his chances of exercising whatever claims he is entitled to, that he takes for granted as a matter of fact that this is the case. So the actor is not alone even in appreciating the sensible character of his personal and unexpressed feelings. In the 1940 article Parsons talked about moral sentiments. These consist of feelings that are not only governed in overt expression by a legitimate order, if the actor is to recognize their occurrence “in him”, and to appreciate them for “what they really are.” The moral sentiments are shared by others and are treated by the actor as shared by others. Others approve the same normative orders of sentiments, and attitudes of others toward these normative patterns are not matters toward which the actor can be indifferent, Parsons insists and has always insisted upon this with no qualification. I point this out because it becomes a problem of critical theoretical interest in Parsons’ discussions of deviance and social control of deviance.

304

H. Garfinkel

The empirical phenomenon is encountered in the existence of stable activities even though persons are much more indifferent to the binding character of a moral order than Parsons would assume. Parsons subscribed almost from the very beginning of his work to the conviction that the part of normative order which he calls a value system is an integrated, more or less consistent unified set, binding upon and enforceable upon the actions of all members of the system. As a condition for stability, it also is critical to the recognition of deviant action. Along with the notion that the actor is not alone, but instead participates in a social world that is defined and evaluated with respect to normative orders of judgment, there exists a “differentiation” of roles. Parsons asserts that this entails a differentiation of morally approved goals for the different individuals. In his later writings Parsons proposed that given a system of values, i. e. morally evaluated, proper, legitimate, true, beautiful, and good states of the system as products of the actors’ interactions, persons would contribute differentially to the production of these concerted effects. Therefore, there would have to be goals as well that would have to be morally evaluated goals, but which would be different goals as one consulted the round of interactants. The typical and dramatic case of this occurs in the division of labor. But, says Parsons, if society is to be institutionally integrated “they are all governed by the same more generalized patterns”. That means therefore that a system of moral ranks can occur for some sector of the division of work that provides for differentiated goals. But if there is to be a stable set of activities there must be a system of moral ranks that is, from the perspectives of actor’s members, superordinate to their own and other sector’s ranks; all ranks within divisions are subordinate. If there is to be continued stability of the system, the different ways of morally evaluating the worth of persons, cannot and will not be accorded equal priority by actors, nor will the different standards be equivalently at hand for justification and enforcement. The self-respect of actors consists in the “contradicted” conviction that their actions comply with, that they “live up to” the moral order of ranks. The self-respect of actors consists in their living up to the scale of ranks that the actor as a perceived individual himself subscribes to, that the actor as an individual himself approves, and of attaining, protecting, maintaining, a position in terms of the scale of stratification. This possibility, says Parsons, is “reinforced” through what he calls the “interplay” between the scale of ranks and self-interested elements of motivations. Here is what he is talking about. To begin with, it is problematic, i. e. it needs accounting for, that the actor will actually fulfill the obligations that are laid upon him. What motivates him to the fulfillment of obligated actions of his everyday life? Parsons proposes that

Parsons’ Primer

305

the scale of stratification literally defines for him good persons in the society. The actor does not treat this scale with indifference. To say that a scale exists at all, i. e. insofar as the investigator detects any modicum of enforceability, the investigator must expect to find that the actor will be motivated to maintain a position with respect to this scale as a moral order. The actor’s concern for, and his feelings of, self-respect, therefore, are not peculiar to him but are instead integral features of the phenomena of stable social actions wherever they may be found. In terms of the institutionalized character of motivation, the actor acting in his self-interest, both real and perceived, acts in his self-interest insofar and only insofar as his actions are “governed” and “informed” by a respect for, i. e. “under the constraints of ” an institutionalized order. Interestingly enough, these conditions specify the meaning for the observed person, the theorists’ actor, and the theorist of moral actions. On the one hand, the actor is said to act morally insofar as he seeks to establish, to find a place, his proper place, in a stratified order. On the other hand, it will be said that the actor’s self-interests consist precisely in the interests that are provided by, and whose present pursuit is motivated by, the fact that he complies with the demands of the scale of stratification. (From this point of view Vance Packard’s92 descriptions should upon testing be found to consist of structural incongruities that are due to a theory of stratification that does not reproduce the real society, and his “evidence” should consist of documentary evidences obtained from the actual scene of a model of that scene. The entire book should consist of a series of figurative means of showing “what he is talking about.” (See my paper on structural incongruities of literal description.)93 Failure to conform with the scale of stratification, says Parsons, injures the actor’s self-interest by entailing sanctions in the form of withdrawal of help, obstructions to the attainments of his goal, the ruin of reputation, open destructiveness, and the rest. This is for the actor himself. Others experience the actor’s acting in the breach of the scale of stratification as being let-down. Parsons emphasizes that although the scale of stratification can be treated by the actors as an object, he cannot treat it “seriously” as a matter of rational acceptance. (Elaborate “rational acceptance”.) To the extent that the order of rank is institutionalized to that extent the important features of these ranks, and his real grounds of compliance with them in the conduct of his everyday affairs, will not be available to the actor to tell the researcher about if the 92

[Editor’s Note: Garfinkel is referencing Packard, V. (1961). The status seekers. New York: Pocket Books.] [Editor’s Note: We’re not sure which paper Garfinkel is referring to. However, in his paper “Notes on language games”, written in 1960 but not published until May 2019 in the European Journal of Social Theory, Garfinkel talks at length about literal description and incongruities in literal description.]

93

306

H. Garfinkel

researcher asks the actor to treat them as a reporter with theoretical sociological interests. These features will not be recognizable to him as matters of rational explication. This is so because his “respect” is a euphemism for the conditions under which he “grasps his real circumstances.” But remember this is Parsons’ version of a society. The frequency with which we encounter persons who see through a great deal of the moral order is not convincingly handled with the use of Parsons’ society. Such persons are not at all the cultural dopes that sociological theory frequently portrays them to be. It is with regard to the possibility of seeing this through and its consequences for stable concerted actions that Parsons’ insistence upon the internalized, or the introjected orders, furnishes critical research possibilities. Parsons holds that the breach of the scale of ranks is accompanied on the one hand by the certain application of sanctions from the outside and, says Parsons, such breaches must always be accompanied by internal conflict. Wherever the researcher finds one person punishing another for “moving out of line”, the other person must experience the “move” as a matter over which he is internally conflicted, and must regard the sanctions as “no more than he deserved.” (Elaborate and qualify.) In the second part of the 1940 article Parsons turned his attention to a method of analysis. The method of analysis consisted of his tentatively enumerating what he called “bases” of differential evaluation. He listed the following “bases”: membership in a kinship unit; personal qualities; achievements; possessions; authority; and power. He proposed that the status of an individual in a system of stratification was the product of common evaluations underlying the “experiences” with the attributes of status in each one of these respects. With this he proceeded in the remainder of the article to describe the stratification system in the United States. Everything in the 1940 article up to, but short of, the method of analysis is contained in the 1953 article with the qualification that these matters are argued much more systematically. For example the kinds of properties of objects that will be evaluated are systematically defined. Putting that part of the 1953 article aside for the moment there is nothing in the 1953 article that represents a major revision of the theoretical points that were introduced in the 1940 article and that we have just covered. The major changes in the 1953 article consist of the fact that these “bases” of differential evaluation are not around in the same form and in some cases were abandoned entirely. Further, the entire effort explicates the “bases” of differential evaluation. The manner of, not the fact of, this explication, contains Parsons’ contribution.

Parsons’ Primer

307

CHAPTER IX PARTIAL DRAFT FOR ILLUSTRATIVE USE OF “STRUCTRED STRAIN”: AD HOC PARSONS Parsons’ series of articles starting with the article on the American kinship system are informed by two major conceptions and should be read with them in mind.94 One is that the real social structures consist of institutionalized patterns of normative culture. The second is the concept of structured strain. Structured strains consist of Normal Troubles. These consist of discrepancies between “destiny and merit” (Cf. Weber)95 that persons encounter as collectivity members by reason of committed compliance with routine, valued, institutionalized practices. A middle-aged married female student once remarked in my seminar, “I am glad that young men for the sake of the young women of our society haven’t read Parsons’ article on the functions of the family.” She was speaking of Parsons’ conception of family strains as a sociological version of tragedy. He is much concerned with the inevitable, inexorable necessary troubles that occur to the members of a Parsons’ society by reason of their attempts to retain membership in that society. The troubles occur by reason of the fact that persons are not only motivated to compliance with these normative valued states of their familiar activities, but produce as well, in their actual activities, social structures which closely approximate those depicted in the normative order. By the time Parsons gets finished providing a formal description of the American family and relating the strains to these formal social structures, he has produced a catalogue of misery. It has not been unusual for undergraduates to say, following a presentation of these materials, “Well, Dr. Garfinkel, if what you say is true maybe we ought not to get married.” So Parsons can be fun. Parsons begins with four features of the American kinship system. He refers to these features as its conjugal character, its openness, its structural isolation, and its symmetricality with respect to the lines of descent. Before we see what these mean we need two bits of terminology: first, the nuclear family, and second the distinctions between conjugal and consanguinal family. Parsons needs a way of referring to persons who make up a familial unit that does not restrict the meaning of this unit to a legitimate group. Nuclear family or nuclear unit serves this purpose by being used to refer to biological father, biological 94

[Editor’s Note: Parsons, T. (1943). “The kinship system of the contemporary United States”. American anthropologist, 45(1), 22–38.] 95 [Editor’s Note: Weber discusses this in Economy & Society. Weber, M. (1978). Economy & Society (2 Vols.). CA: University of California Press.]

308

H. Garfinkel

mother, and their immediate offspring. The distinction between conjugal and consanguinal family depicts two distinct ways in which the management of the nuclear family’s affairs can be organized. Parsons says of the American families that they are conjugal. What does that mean? Question: With respect to the decisions that the members of a nuclear unit will have to make, decisions that define what the observer is talking about when he says that families will somehow manage their family and household affairs, decisions like where and what kind of place to live in, how household routines are to be run off and duties assigned, decisions of sexual access, eligible partners for erotic gratification, decisions for the acquisition, use, and disposal of family resources, decisions of child care and child training – if the rights to make these decisions are assigned to the nuclear family, this empirical state of affairs is marked off by calling the arrangement a conjugal one. The term conjugal refers only to the fact that the decisions in managing household affairs are legitimately assigned to, accepted by, and enforced upon the nuclear unit, as contrasted with the members of the more extended family group. On the other hand if these rights are assigned to the extended family, as contrasted to the nuclear unit, such a state of affairs is referred to as a consanguinal arrangement. The conjugal arrangement is what Parsons is talking about when he says of the American family system that it provides for a “nucleus of spouses surrounded by a fringe of relatives,” and contrasts this with other systems such as the classical Chinese family which provides a “nucleus of blood relatives surrounded by a fringe of spouses.” “Surrounded by” is metaphorical and does no harm if one does not read the accompanying diagram as a set of overlapping Euclidean circles. The big question to answer is: “where are the rights to manage the household located?” “Conjugal” and “consanguinal” are ways of talking about this and only about this. A most important point occurs over and over again as the unstated theme that this series of articles elaborates in a multitude of ways: Not only is it normatively the case that the family has this conjugal feature (remember that Parsons is describing the American family, and makes appropriate qualifications as we shall soon see), but this normative state of affairs is actualized in the conduct of the members. There is a very close correspondence between family members’ conceptions of “desirable states of the system” – and the conjugal features that are such “desirable states”; namely, that decisions to manage its internal affairs may be allocated and enforced in conjugal fashion – and in the way, in fact, actions in compliance with such “desirable states” are enforced. Or the way in which claims are extended, and the way in which persons operate to maintain some kind of an orderly assignment of rights to engage in the affairs of that nuclear

Parsons’ Primer

309

unit. By no means is it to be taken for granted that such assignments are in any way different than, or separated from, the actual activities whereby such assignments are claimed, extended, enforced, and the rest. As we go through the rest of the family features that Parsons describes, we find that Parsons is engaged in elaborating what he calls the structural features of the family which consist of correspondences between the normative features on the one hand, i. e. the valued states, and the activities which actualize them. Parsons introduces a diagram. (Insert diagram)96 Several points: 1. An inspection of the diagram shows several important formal features of this system. (1) Ego is a member simultaneously of two conjugal units. (2) Any married member of the lineage can be located at Ego’s place without changing the conjugal feature. The two conjugal units are the familiar families of orientation and procreation. The distinction calls attention to the fact that for any person, he will have been raised under one set of conditions and have enjoyed membership in one family group; when he marries he joins a separate family unit and becomes himself a member of a parental family. 2. Each married person therefore is a member of two conjugal families. The system has the further formal feature that there is for Ego as member of a conjugal unit Ego’s spouse who is a member of a conjugal unit that Ego is not a member of. Both of these formal features hold normatively and actually. Parsons refers to this feature as structural isolation. The term structural is to be understood as combining both the institutionalized normative character of this arrangement and the fact that actions of persons reproduce actual familial activities that closely correspond to what the normative order depicts. Several other remarks, conjugal, openness, structural isolation, etc. arrangements by reason of being institutionalized, describe what participants really mean by a normal, natural, good, true, beautiful, legitimate, right way of conducting one’s familial affairs. Members encounter troubles and seek explanations and remedies while taking for granted that this order is not subject to alteration and are literally unalterable conditions of life because they are normal, legitimate, and the rest. 3. The problem that Parsons’ talk about lines of descent is designed to settle is this: There are two exigencies that any family system must provide for with respect to its stability, and the maintenance of the system as a normative order. The first exigency is that of establishing socially defined and legitimate origins 96

[Editor’s Note: there is no diagram fitting this description in Garfinkel’s course materials. However, the diagram in question is probably the one that is published on p.23 of Parsons, T. (1943). “The kinship system of the contemporary United States”, American Anthropologist (45)1:22–38.

310

H. Garfinkel

of the persons in that arrangement. Members’ concerns about the rights of the family name go to the point of establishing legitimate, normal, natural, right expectations with respect to intergenerational, familial succession. It is necessary to watch how the family name is handled between generations, normatively and actually. In the American family the father passes on his name to both the son and to the daughter. The daughter gives up this name when she gets married, and takes the family name of her husband.97 But with respect to how legitimate intergenerational, familial succession is handled, although the wife takes on the husband’s name, she retains the right to identify herself with the family in which she was raised. By watching how the family name behaves we see how legitimate intergenerational succession is ascriptively decided. The second problem of descent concerns the inheritance of property, i. e. in passing on between generations the rights to control, to use, and to dispose of the resources that the ascendant family has with which the succeeding unit may be enabled to reproduce the ascendant family’s status. Parsons makes the following points about the descent of inheritance. First, he says, there is no structural preference either between the ascendant families with respect to the legitimate sources of resources, or between offspring to whom they are to be passed. In Parsons’ terminology there is no structural bias in favor of the offspring either by birth order or by sex. It does happen, he says, that one or another of the offspring will be favored. This occurs in the practice of wills for example. But under such conditions the justification usually goes to matters of personal need. The system is not one in which a morally correct allocation would require preference either by sex or birth order. Such an alternative is not morally enforceable. The big point then is that there is no structured distinction between lines of descent as far as rights and the division of the inheritance are concerned. Parsons points out another feature of decent. He says that there is a relative absence of structural bias in favor of solidarity with either the ascendant or descendent lines in any one line of descent. More plainly, obligated familial loyalties are not institutionally prescribed as due to any family within Ego or Ego’s spouses’ lines of ascent or descent. It occurs then that when this unit gets established a major problem of its stability to be worked out concerns how the relatives and inlaws are going to be handled, with respect to important decisions centering upon the management and maintenance of the new 97 [Editor’s Note: At the time Garfinkel wrote the Primer, American women were legally required to take their husband’s last name. The change happened automatically in most U. S. states when a marriage occurred. It was only in subsequent decades that women were given the option to keep their maiden name.]

Parsons’ Primer

311

household. Although claims of preference might be made by one member of the new unit they are difficult to justify to the spouse and spouses’ family. The absence of structural preference with respect to solidarity or inheritance, Parsons refers to as symmetrical and multilineal lines of descent. Thus the “opportunity” to be on one’s own also consists also of the threat of being structurally isolated from support. (Consider the contrast with systems in which status, nature of local ties, occupational skills, source of economic support, employment is inherited from the father.) 4. The feature of openness. An inspection of the diagram shows that Ego’s spouse is the single point at which one lineage comes together with another. Although they may be related in other ways, Parsons emphasizes that morally obligated ties between the extended units can only be directed and enforced at this point, normatively and actually. The “open character of the system” refers to three features of this particular “linkage”. First, this feature is true for every marriage within the unit. Second, apart from this linkage the two kinship structures are otherwise structurally unrelated. Third, the choice of a marital partner is made from persons whose eligibility consists in their being outside of the kin structure with the choice being independent of any kinship considerations. As Parsons puts it, there is no preferential mating on a kinship basis that has important structural significance. To establish another unit one must go outside these arrangements to find an eligible partner. This is accompanied by a very peculiar institution: the choice of the marital partner, institutionally speaking, is not arranged by parental figures. Responsibility for the choice is institutionally claimed by and assigned to the eligible partners as a matter of personal choice. Personal choice does not mean that they do not consult with or obtain permission of kin, but that in the final analysis the eligible pair consider themselves solely responsible, and others hold them responsible for whatever consequences are socially attributed to the marital choices. The open, isolated, symmetrically multi-lineal, conjugal family is a structurally normal, natural unit, which is to say that these features serve for the persons involved as the “natural” conditions of family life, i. e. “conceptions of desirable states of the system,” valued familial arrangements. These features may be elaborated. The elaborated features are also both normative and actualized. Upon its formation the nuclear family is established as a separate household. The decision as to where the unit will live is claimed by and made by the martial pair. Residence decisions are largely dictated by the limitations placed upon the occupational opportunities of the principle breadwinner who is in most cases the husband. The members of the unit pool their resources for their

312

H. Garfinkel

common support. All the members of the unit, institutionally speaking, have equivalently legitimate claims upon the sources of familial support, principally the breadwinner’s earnings. Equivalently legitimate means that there is no rule which states that a priority in the legitimacy of claims can be recognized. This feature is prominently meant in talk about the “equality” of the members of the American middle class family. No rule defines the priority in the legitimate character of a claim upon the resources of the breadwinner. That there is machinery for differentially allocating money to members’ demands is another story. The husband may get a larger share of his income to spend, or his demands may be met before the demands of others are satisfied, but the justification does not invoke his prior right to get more. The justification lies elsewhere. Neither the household arrangement nor the source of income stand in a relationship of subordination to the family of orientation of either spouse. Parsons points out that this results in a kind of armed truce with respect to the unit’s claim upon or receipt of such things as in-law resources, advice, invitations, as well as quarrels about interference. Thus to the extent that it is conformed with, this feature enforces the structurally isolated character of the unit. There are difficulties that this unit encounters by reason of its structural isolation. Family members by acting in respect for this feature reinforce the structural isolation and guarantee the difficulties that arise from it. Normatively and actually, the unit occupies a home segregated from both pairs of parents. Although Parsons includes this as one of its institutionalized features, he does not make a great deal of it except to point out that by reason of its structural isolation the nuclear unit is very mobile with respect to both class and territorial movements. He points out that separate residential location is one of the factors that facilities the ease with which such movements can be made, and that by and large these movements are highly responsive to changes in the occupational market. Frequently there is large geographic separation between nuclear units of the same lineage. The importance of this, structurally speaking, consists in the fact that if there is to be control that a unit claims and administers over other units, geographical separation attenuates the effectiveness. Parsons points out that the effective exercise of control requires that one be able to put hands upon the person whom you’re going to influence. Finally, the husband holds an occupational status independently of his ties to kinsmen. Parsons makes a lot of this. Parsons qualifies these considerations as features of the “American family”. He proposes that the frequency with which this kind of a conjugal unit occurs is uneven over the class structure. In certain social economic conditions

Parsons’ Primer

313

of rural life there is much more continuity of occupation, and familial honor from one generation to the next than occurs in cities and towns. Parsons talks of “important upper class elements” where there is much more continuity in kinship solidarity because status is closely bound to ancestral status. Multilineal descent is frequently by-passed in favor of a patrilineal system where the inheritance is given over to a son. Also there are many more consanguinal arrangements in upper class families. In lower class situations there are tendencies to unstable marriage and to mother-centered types of family structure. This occurs in both Negro and white families, and is not particular to Negro families. An important correction needs to be made of a common misunderstanding of Parsons’ description. He has not provided an “ideal typical” family. In this essay Parsons is not constructing an ideal type. He has provided instead an empirical description of life in the urban middle class family. In the urban middle class family there occurs with greater frequency than is true in other structural locations the actualization of this kind of culturally defined family life. It is possible to find many specific exceptions, and to cite one or another instance of a family that departs from this as an adequate description of it. Parsons is talking literally of sheer frequency of occurrence. He is saying that urban middle class families exhibit these features with overwhelming frequency. True, he does not use the results of a nose count in order to render the many and intricate propositions that are intended here as factual ones. Instead he uses a structural terminology, employing linguistic categories to refer to empirical regularities. Above all he is not talking about ideal typical families. The way one criticizes Parsons’ description of the family is to show that in fact what he says is not true normatively and/or in actual practices, i. e. to show that what he says is the case about the structural features of the family is not demonstrable. The researcher does not have the choice then of whether to use this description as a method to help him collect his thoughts about family life (although it may help his efforts to do so) because the description is meant as a description of the facts of American, middle class, urban family life. Any quarrels must be quarrels about fact. All these features were well publicized before Parsons’ article. However the concept of their structural arrangement was not previously worked out in the way Parsons did. Nor has the idea of structured strain been included in previous discussions. Although there had been a great deal of talk about normative and actual family structures, the use of the concept that the real social structures consist of institutional patterns of normative culture to relate normative and actual family structures is Parsons’ contribution. No other American theo-

314

H. Garfinkel

rist is as definite as he on that formulation for descriptions of social structures generally, while among family studies, the formulation is conspicuously absent. This emphasis presupposes Parsons’ theory of the cultural character of these activities which in turn is developed from Parsons’ version of what Durkheim intended by “social fact”. The formal sociological emphasis that Parsons gives to the “well-known facts of American family life” comes out with considerable clarity and definiteness when Parsons talks about the structured strains of family life. Structural isolation, says Parsons, is a distinctive feature of the American kinship system. This feature is one among several whereby the kinship system stands in a “functional” relationship to the features of the occupational system. For example structural isolation describes a unit that has relatively little difficulty in picking itself up and moving up and down the class hierarchy or in moving from one territorial location to another in response to changes in the principal breadwinner’s occupational circumstances. Structural isolation and the existence of a mobile labor force that is responsive to changes in the labor market are complementary features. At the same time the family’s structural isolation provides the basis of many severe problems that members of these kinship arrangements complain of. Parsons never tires of stressing the theme of the “institutional segregation” of occupational and kinship arrangements. (In the class discussion of subordination of family life to occupational life, a female student pointed out, by way of argument, that in an agricultural family, you would not find this structure, and would not find it in other than an urban industrial society.) Parsons says that when he was looking for some of these invariant arrangements he finally decided to use this kind of an arrangement as an empirical functional imperative, even though it was not a formally defensible argument at that point that this was the only possible arrangement. Whether it is the “only possible” arrangement could only be decided if this theory was explicit enough to demonstrate the consequences of alternative arrangements. Parsons has not yet achieved that kind of definiteness. He does say that you cannot have a set of industrial practices such as we know in this country and a classical Chinese family type of system. But this is not to say that the present type of family system is the only one that is compatible with present U. S. industrial practices.

315

Parsons’ Primer

I Structured Strain In most kinship systems a person maintains a stable, although changing status in one or more kinship units throughout his life cycle. In the American case this is not so for anyone in the system. Parsons points out that there are at least three severe points of structural “discontinuity”, i. e. in a “normal” course of affairs, and as a consequence of compliance with them, there occur abrupt alterations of life circumstances. Such situations are dramatically stressful. They occur (1) when a person establishes his own family; (2) when the family breaks up, i. e. upon the death of one of the spouses in the nuclear unit or when the children marry; and (3) when the parental figures get old and particularly when the male spouse is retired from his job. After a long period of socialization within the family there occurs a rather abrupt transition in the structure of loyalties. Loyalties that are due to the family of orientation must now be transferred to the new family unit. The normative order of loyalty priorities are re-assigned to the new spouse as an obligated re-assignment. This occurs at the same time that the new family must also attempt independently to establish its class position via the efforts of the male breadwinner in the occupational market. This family now becomes dependent on the newly married male not only for its standard of living but also for satisfying the partners’ status expectancies that are derived from their parental families. Structural discontinuity occurs again by reason of aging when the family members’ principal statuses in the society begin to terminate. Persons then find themselves without family connections or without important connections to other types of associations whereby they can remain competent members of the society. This occurs because of structural isolation and the open system of conjugal units. Parsons’ analysis of the severe and abrupt changes at the time of a marriage runs like this. To begin with, Ego’s marriage segregates him from his family of orientation and imposes prior kinship loyalty to a spouse and these children. But neither of the families of orientation has any structural priority of status within this nuclear unit. Related to this are the efforts of the nuclear unit’s members to insure the independence on the part of the structural arrangement with respect to both sets of in-laws, both spouses being under pressure from each other and from parental units to maintain the in-laws at an equal distance. By reason of the normative order, the rights of either set cannot be enforced in the management of the stresses that this unit is going to encounter. Thereby

316

H. Garfinkel

the mechanisms for ensuring a structurally self-enforcing isolation occur at a period of concerted stress. A large consequence, says Parsons, is that the marital bond is the “structural keystone” in this kinship system. With respect to any influence upon the exercise of a marital choice and again by reason of the feature of structural isolation, any choice exercised by kin is either missing or tends to be minimized. The system therefore tends to the pattern of personal choice without enforceable parental influence. Given structural isolation, what are some of the devices whereby this structural keystone can carry the weight that is imposed on it? One device is concerned with the feature of personal choice. An ideology centers on it that exaggerates the ideal character of the rightness of and necessity for the exercise of personal choice. This is one aspect of the doctrine of romantic love. A related problem deals with the character of the marriage relationship. It is unusual in historical kinship systems, given large numbers of individuals involved in complex relationships of mutual interdependence, to allow personal feelings, or the overt expression of these feelings, to govern the affairs which are of serious interest to not only the nuclear unit, but to the other units with which it is or might become involved. In most family systems the emphasis is to insist upon and enforce the rule that obligations to other kin be fulfilled on impersonal grounds, i. e. that they be fulfilled as a matter of familial or kinship duty. The American system contrasts with this. Instead of sanctions being delivered via the insistence upon fulfillment on impersonal grounds, sanctions are delivered via personal sentiments. One of the consequences of this arrangement for the American family is that the very structural isolation of this conjugal unit, the fact that nuclear members cannot be held accountable to extended kin for the management of the nuclear unit’s affairs by extended kin, frees the affective inclinations of the couple from any hampering restrictions. But the result of this is that expressions of friendliness, love, care, personal affection, and the rest are the principal measures whereby the solidarity of this kinship unit is maintained. Problem: If this is one of the ways in which it is maintained, it would mean that for the parties involved, there are novices’ expectancies at the time the marriage is made that consist of the insistence upon “feeling right” about the partner who has been chosen to establish this unit. Parsons proposes that many severe problems are associated with three structural difficulties that occur by virtue of the fact that the marriage, despite the insistence upon right feelings, must nevertheless be routinized with respect to everyday activities in which personal feelings cannot be invoked as the effective grounds for deciding whether or not mundane family obligations have been adequately fulfilled. Consider the husband who tells his

Parsons’ Primer

317

wife, “You know I love you; why don’t you wash the dishes?” There are first, the problems of how the member can manage the show of affection over the course of a routinized family existence. A second set consists of how the member can achieve the routinized accomplishment of the tasks of everyday family life, given the fact that the pair come rather quickly upon counter indications of right feelings as a source of legitimacy for what is routinely done. Third, how is the member to deal with the inevitable disappointments and above all the boredom that is a common feature of this kind of an arrangement. Parsons proposes that actually this type of institutionalized support for the martial partners is very weak. In response to this weakness there exists the institutionalized expectancy by the spouses of feeling the proper personal sentiment for the other as a condition under which each can be reassured that the natural, normal, good, true, beautiful arrangement is being actualized. Thus there is prominently in this system the obligation to be in love. This obligation is not attenuated, institutionally speaking, as a function of the length of marriage, so that one finds the familiar ideal of aged lovers going down life’s highway into the setting sun, until hand in hand they finally drop over the edge. Among the various ways of entering into love, falling into it is also one of the important sanctioning conditions for the marital union to occur in the first place, being treated as evidence of a right choice by stressing the inability of the partners to have done otherwise. Under the condition that persons fall out of love, this can be invoked as good reason for the dissolution of the marriage. Parsons says that this is to be compared to an alternative in which household arrangements would be maintained while partners look elsewhere for affection, erotic gratification, and the rest. Another line of structured strains concerns sexual equivalence and sexual segregation. Parsons points out that there are very strong pressures in some areas to make the sexes equivalent. He points for example to increasing accessibility of occupational opportunities, of the lack in the legal system of discrimination between males and females, and to equivalent educational opportunities at least through college and graduate school. Above all no premium is placed either upon the masculine or feminine roles as far as maintaining the continuity of kinship roles are concerned. Nevertheless, these features of “sex role assimilation” are accompanied by vigorous efforts to maintain a rather rigid sex role segregation. Parsons enumerates the following features that make for this segregation and during the course of this develops another set of structured strains. To begin with the woman is handicapped insofar as she stands in a competitive relationship with the man. For example she is handicapped by being assigned the rightful care of children. That the woman is the biological agent is important because it is to be understood with respect to the social

318

H. Garfinkel

definition of her being morally better suited. In terms of a normative order she is morally suited both to bear the children as well as to assume the responsibilities for their care. She is naturally the one to care, “naturally”, referring to a rather intricate set of moral obligations. Other factors make for rather sharp segregation. Some of these can be seen by asking how the family maintains solidarity in view of this structural isolation? Parsons points to four factors. First, romantic love is a contributing factor. It exaggerates the difference between the sexes by idealizing these differences. For example, to the extent that “feeling right,” of requiring that one experience right personal sentiments, becomes a condition for deciding the correct or sanctionable character of the martial decision, it makes for this segregation of the sexes. A second factor consists of the various devices whereby intrafamilial competition is restricted. It may be controlled by ensuring that there is no competition in the first place. One way in which this may be done is to provide that males and females are evaluated by different standards entirely. And this he says occurs with the use of at least three institutionalized doctrines: (1) What males and females do and feel are to begin with not comparable, so that even when one finds the man washing the dishes and the woman washing the dishes these are perceived as different actions in the “first” place. Not only are their actions and conduct perceivedly not the same even though they may be engaged in precisely the same behavior, but (2) males and females are also different kinds of persons in the first place, and (3) Finally, the male’s actions are status-determining in a way in which the female’s actions are not. This returns us to the occupational sphere. Parsons points out that the male’s job is not only the primary source of a family income, but it is also the most important single basis of the status of the family. This is meant both in the normative sense as well as in the actual way that persons act out the family’s status. That women are in the labor force would be met by the argument, that the married woman’s job in the majority of cases, and in the normative case, is not one that she engages in and expects to be able to justify as a familial obligation. Nor is it one in which she competed with men who are her own class. That does not mean that she does not work, but it does mean that she does not cash in the kind of work she does for the returns of status honor in the way in which her husband does and is required to do. Another set of discrepancies makes for this sharp segregation. These have to do with the occupational status. Parsons points to the dramatic discrepancy between the role features valued in occupations as compared with those valued in familial performances and attitudes. Parsons points, for example, to the relevance of personal achievement; the emphasis upon opportunities be-

Parsons’ Primer

319

ing distributed in accordance with performance on the job; the emphasis on the importance of mobility in response to technical requirements; the value placed on disinterested devotion to occupational interests, i. e. upon work as a primary duty; the value placed upon economic rationality, and upon the substitutable character of a man’s work, of its standardized character, and the fact that it is amenable to capital accounting. Such features differ drastically from the features of valued kinship relationships. There is general segregation of relevance between technical occupational and what Parsons calls “humanistic” values. Parsons asks whether these two systems could be accommodated to each other. He points out that the kinship system exerts relatively little pressure for the ascription of an individual’s status through familial affiliation. The result of this is that the conjugal unit can be mobile in its status independently of two things: not only of ties to the family of orientation, but of ties to spouses as well. At the same time the unit can be strongly solidary and he points to the matters of restricted competition and romantic love in defining the male and female differences. But these features are accompanied by strains. With respect to the sexual problems these strains are as follows. On the one hand there exist distinct pressures to define the feminine role as a dependent one upon the male, although there are strong pressures which counteract this. One of these is the multi-lineal system which favors equal rights and responsibilities as far as inheritance and family loyalties are concerned. Another is the fact that the marriage relationship is itself made legitimate by the monopolistic rights to the affection of the other that each is supposed to accord to the partner, one of the rules of monogamous marriage. Such pressures can put a premium in the relationship upon mutuality, and equality between the partners. Or to put it another way, there is no clearly recognized and socially enforceable pattern of super- and sub-ordination. Spouses commonly refer to each other therefore as “partners”. This is structurally speaking a euphemism because occupational status has tremendous weight in distributing honor, prestige, esteem, and the rest. Thus the married woman cannot actually demonstrate her equality to the male, which she nevertheless must claim on the basis of the very sentiments that make the relationship legitimate in the first place. If she cannot demonstrate her equality with the male in occupational achievements, neither can she do it through the performance of household duties. Household duties are by and large regarded and treated by both spouses as menial. This peculiar cross ruff accounts for the emphasis in the married feminine role on humanistic achievements rather than technically competent achievements. The housewife must attempt some adjustment to this stress by specializing in

320

H. Garfinkel

matters where good taste is required, engaging in the work of displaying and exemplifying the status of the family. The husband provides the wherewithal that determines the standard of living; the wife is responsible in the different sphere of expressing the status of the family by the arrangement of the house, where they live, how the house is kept, how the children are dressed, where they’re trained, etc. In discussing strain in the adult feminine role, Parsons focuses on the housewife. He begins with the observation that housewife unhappiness is widespread and systematic. It is not an occasional thing; it is not idiosyncratic, it is obstinate and unyielding to remedies, and is invariant to the personality differences between female personnel. Parsons argues as follows. To begin with marriage determines the woman’s fundamental status in society. After her marital choice is made her role performances are no longer determinative of the honor that the family can claim. Her role performances, much more than being status-determining, consist of her living up to certain expectations. She is blocked from any principal areas of status achievement through her own individual performance. There is lack of any clear definition of the adult feminine role. If the possibilities of a career are eliminated, the result is an unstable combination of three patterns of activity: (1) “domesticity”, (2) “feminine glamor”, and (3) the activities of “good companion”. In her marital situation her primary statuscarrying role is that of the housewife. This status consists of three things: (1) her status as that of her husband’s wife, (2) her status as that of the mother to his children, (3) her status as the person who is responsible for the management of the household and for the care of the children. Parsons treats these as socially employed parameters of the status that is going to be accorded to her over the course of the marriage. The housewife by what she does furnishes them their specific determinations. These statuses of the husband’s wife, mother of his children, and household manager are occupied by her by virtue of two circumstances: (1) her husband’s occupational performances and achievements determine the expectations in these respects that she is expected to live up to; and (2) her own specially defined qualities as an individual. But the fact that primary importance is placed upon the husband’s occupational status is an important source of strain in intrafamilial structure of sex roles by depriving the wife of her role as partner in the common enterprise. Thus the common enterprise in effect consists of two things (1) the actual life of the family itself, (2) the informal activities that the husband and wife participate in together. Her occupation, Parsons says, is at best a pseudo-occupation, and is not in any sense comparable to the husband’s occupation at least in struc-

Parsons’ Primer

321

tural terms. The activities of domestic career, feminine glamour, and good companion are modes of adjustment to these circumstances. Parsons defines them and then shows why they are not adequate adjustments. The domestic pattern consists of activities that are organized around domestic and related janitorial services like care of the children, feeding the family, cleaning and straightening the home and the life. It consists also of strong affectional devotion to husband and children. This pattern he says is becoming less common. There is a strong tendency for it to acquire a residual status, i. e. for it to be regarded as a thing which the housewife settles for when other modes of adjustment for one reason or another are no longer possible. The remaining three, career, feminine glamour, and companionship have the status socially speaking of an emancipation from this domesticity. In one way or another they involve some kind of competition on the part of the woman for prestige in other directions. The obvious one is a full-fledged career. This involves following the masculine pattern. The woman seeks a career in fields of occupation in which she is in direct competition with men of her own class. “Men of her own class” is the critical point. Only a small fraction of the women engage in this, or have succeeded with this solution. The second of these, feminine glamour, consists of emancipation from traditional or conventional restrictions on free expression of sexual attractiveness. The third is that of the good companion – Here the emphasis is on common humanistic elements which take several forms. One is a cultivation and appreciation of cultural interests; educated taste for example, becoming proficient in matters of art, music, literature, home furnishings, and the like. The second type of humanistic good companion is the cultivation of humanitarian obligations – service interests, community welfare, and the rest. The strains. With respect to the pattern of domesticity, this pattern ranks lowest in prestige, but because it affords community support, and because of the values of fidelity, and devotion to husband and children, it offers a type of security. But given the features that were enumerated before, for example the course of education when the female may be led to expect equivalent treatment with males, etc. – the pattern of domesticity for the urban middle-class housewife is very difficult to stabilize without resentment and the conviction that life has paid them out with second best. The glamour “adjustment” has at least two structural limitations. One is the resistance encountered with respect to the moral character of the conduct it involves. This mode of adjustment is fraught with ambiguities. Also, there is considerable work and cost involved in managing the show of feminine glamour, the work becoming progressively more difficult as aging occurs. Thus

322

H. Garfinkel

there are age limitations on proficiency as well as age limitations on the full validations of the performances that are undertaken. Parsons says that its primary valuation is obtained in late adolescence when the work and cost is a feature of the marital market itself. The good companion problem has its limitations. These consist first in the fact that the ways in which a woman can be a good companion are not institutionally defined. Although the housewife can become quite proficient in home furnishings, or in community welfare activities, it remains a difficult thing to cash in these activities for moral honor. Such premiums, like scrip, tend to be subject for their worth to matters of local time and circumstances. Support for the role tends to be dependent upon immediate relationships with peers in the immediate environment in which these activities are undertaken. Honor obtained in one sphere or one place is not easily transferred to others. Upon the occasion of a move the woman has the problem of establishing the honorific character of what she is doing by re-establishing new personal ties with peers in the new location, something her husband in a change of occupation is by no means as subject to. Also there are the sheer multiplicity of ways of being a good companion. Choices among them are only conventionally and not institutionally required. Another set of stresses occur in the processes of socialization. Parsons first reviews structural features of the family, i. e. that it is an isolated conjugal unit, dependent for its status and income on the husband and father, that his work is done in a separate place, that its members do not cooperate with the husband on the job, that his status is based very much on the individual qualities that he can exert on the job and his achievements in that domain, that the children upon maturity leave the parental household and must make their own way, that the daughter’s chances are overwhelmingly dependent on the marriage to the right individual man, and that the marriage for both partners is a matter of personal individual choice. Given these patterns of child training the result makes for high levels of structural anxiety and aggression. Parsons’ argument is as follows. The mother is a salient and single source of affection for the children. This creates a high degree of sensitivity on the part of the children to the emotional states of the mother. The children are therefore highly vulnerable to any matters that disturb her, and we have gone over some of the things that can disturb her. Because of the conjugal, isolated features of family life most of the relationships for the children outside of the immediate family involve persons who are in no way parental substitutes. Rather these relationships are “secondary relationships”, which is to say that early in the child’s training it enters into relationships with “outsiders” in which its parents and particularly

Parsons’ Primer

323

its mother, may be implicated, in which the chances of its obtaining adult approval depend (1) upon its meeting certain behavioral standards and (2) upon its display of achievements. Very early it encounters situations where it is treated as a figure who is compared with others; very early in life it is treated as an instance of a behavioral type of person. These secondary relationships are not stressful in themselves.98 The stress consists in this: Rules that define “doing well” in situations are consensually valid rules, and hence involve the mother in attitudes which are shared along with those with whom the child comes into contact. Her love and approval therefore tends to be contingent upon two things: (1) that the child meets her expectations of achievement, and (2) that the child conforms with behavioral standards to which she subscribes along with others. This feature, says Parsons, is a critical meaning of conditional love. Areas in which problems are particularly critical involve the use and respect for property, achievements, self-care, success in play, education, and also and prominently the control and show of aggression. The structural stress in the child’s circumstances consists in the conditions under which, and the way the mother can, give love given the fact that both the mother and the child are involved with a community of other persons who are themselves also engaged in precisely the same tasks, and structurally speaking in precisely the same situation of normative pressures. That these evaluations are administered in situations where the child is compared by others, forces high levels of anxiety, and is one of the primary sources of aggression. The routine structures build up strong needs for affection. At the same time the conditions of approval are such as to make the direct expression of aggression dangerous. This is normally – i. e. normatively – handled via repression. Parsons presents his case with the use of the pressure value theory of aggression. Other factors intensify this insecurity and serve to structure large amounts of aggression. Parsons argument goes like this: The mother is a salient emotional figure; therefore she is a role model. For the girl such a situation tends to be “normal” in a way that does not hold for the boy. This is because the girl serves an apprenticeship to the family, whereas the boy’s apprenticeship is a problem first because the father is not available, being away for much of the time, and second, the father’s job tends for the boy to be abstract and difficult to identify with. The problem of mimicry too is difficult to handle in the boy’s case, particularly in his serving an apprenticeship to the father in which he might learn as a function of the relationship those skills which are useable in 98

[Editor’s Note: Garfinkel used “he” instead of it throughout this section. The editors found this generic use of “he” for “child”, particularly in a discussion of gender roles, problematic. We changed it.]

324

H. Garfinkel

his adult situation. The girl acquires her skills as the “little mother”. Finally, a difficult problem involves father and son in a situation in which it is extremely difficult for them to interchange roles except where the father and son engage in play. As far as this apprenticeship is concerned the father becomes primarily involved with the child as a “buddy”, “advisor”, etc. Girls can more easily make an early and consistent identification with the mother. There is a similar tendency on the part of the boy to feminine identification, but with the important difference that the boy is rejected insofar as he acts like his sister. He learns therefore from his mother that it is bad to act either like her or like his sister. From his relationship with her he learns the distinction between the “bad boy” and the “good boy” in which the “bad boy” is the masculine boy and the “good boy” the feminine boy. In this way he learns from his mother that women are inferior to men. He learns from his mother as well that it is shameful to act like a woman. Hence, during the latency period there develops what Parsons calls compulsive masculinity, marked by the evaluation of athletics and general display of physical prowess; the inhibited display of tender feelings, (most particularly as these are directed to females), and a defense against any feminine identification. These have the earmarks of reaction formation. Whatever he is, he is not a female. These processes produce an adult character structure that has as its features a basic uncertainty or ambivalence in sexual identification on the part of the male with masculine sexuality and sexual accomplishments becoming a chronic source of anxiety. Parsons says it is an easily provoked reactivity. At the same time this ambivalence with respect to his sexual identity is accompanied by a large and insatiable need for affection with the ambivalence showing up in the male’s aggression toward women. Parsons points out that American males are on this score particularly vulnerable to females on two very important points: first with respect to their own feelings of sexual identity and sexual competence, and secondly with respect to managing the display of tender feelings toward women, such management being accompanied by undercurrents of fear and suspicion of “being let down”. Given the abrupt transition to male adult status, Parsons argues the particular stresses that are involved at the particular point, when persons leaving the family must fulfill two socially imposed requirements for adult status: one is that the male get a job and succeed at it, the second is that he get married. Strains are involved in the adult male status occupationally speaking by reason of the different lines of achievement that are valued in occupational roles than those that occur during childhood and adolescence, the change from one to the other being typically abrupt. Not only are matters of personal achievement involved in the new male adult status but there is the requirement of taking

Parsons’ Primer

325

on large elements of responsibility, of being held accountable for one’s own actions both in the occupational as well as in the familial spheres. With respect to the stresses that are imposed by the discipline of the newly encountered standards of achievement, Parsons points out that in some lines of male adult occupational work the transition is not at all a dichotomous one. He points out such occupations as those of small business men, and also salesmen, and points to the prevalence in both these cases of an emphasis on evaluation of masculine activities like sports from which females are specifically excluded, and to the phenomenon of the “businessman-sportsman”. Structured strains occur in the girls’ situation. To begin with there are the opportunities in childhood for continued rewarding identification with the mother. Later the girl must seek her fundamental adult security through the choice of a marital partner. She can only partially control the conditions that will determine the choice, and has to make it while she receives no institutional support from kin. Further, this choice is accompanied by the stresses of the marital market where the choice must be made in competition with other girls. That she must make the choice in a competitive situation with little support from kin, and while assuming personal responsibility for its consequences, are the realistic meanings of the euphemism “free choice”. Her “free choice” means literally that she is required to go about the business of selecting a mate to insure her future familial and status chances, in a situation in which she cannot count on familial support, and in which it literally cannot be assumed that she knows what she is doing. I do not use “know” invidiously: I mean it literally. This important step has to be taken at a time when the occupational prospects of the suitor are in doubt, and to add to her woes, she must compete for the personal favor of a young man – the meaning of personal selection and romantic love – who is himself deeply ambivalent about the role that his future wife is going to play with him. Parsons proposes that these real circumstances permit an understanding of why romantic and right feelings are so heavily consulted by both in managing their convictions that the right choice has been made. It does not mean that they do not consult other feelings or other information. Instead, it means that when the multitude of considerations are “added up” the question still arises, “Am I making the right choice?” to which the structural advice runs, “How do you feel about him?” So, says Parsons, although the girl has been brought up to believe that her adult chances in life will be decided on the grounds of domestic skills or a career, she finds instead when the occasion of choices occurs that companionship and sexual attractiveness are the conditions upon which her chances in the marital market payoff point are decided.

326

H. Garfinkel

For the girl two important frustrations come from such a situation: (1) her discovery of masculine superiority, by which he means her discovery that her security literally depends upon the actions of a man whose chances she cannot directly manage. She learns that her chances are not directly manipulatable by her, but are manipulated only by manipulating her husband. (2) She finds that those ideals which were the focus of childhood identification, the domestic virtues of the mother, are not of primary relevance and value in solving her fundamental problem, which is that of making the marital choice. The transition from the single female to full adult status, which occurs with relative abruptness, and is accompanied by the stresses which we have pointed out before in the establishment of a separate household, is accompanied by the deep-seated feeling of having been deceived. This consists of two convictions: that mothers are not after all what they seem to be, and that men have forced upon women an intolerable situation because the female now has to be two incompatible things simultaneously to the male: mother and sex partner.99

99 [Editor’s Note: It is important to point out that while Garfinkel is describing Parsons’ analysis of the American family here, Garfinkel himself did not come from such a family. His family was an extended kin relationship and he was very close to his parents and brother. There is evidence for this both in Garfinkel’s own descriptions of his family, and in letters home from Georgia in 1939, the summer before he entered graduate school.]

The Program of Ethnomethodology1 Harold Garfinkel

For the management and mastery of everyday affairs, persons are concerned as scientific investigators with the distinction between fact and fancy, truth and falsity, conjecture, hypothesis, personal opinion, and the rest. They do not consult the scientist or the philosopher of science in coming to the conviction that they know what they are talking about or that they have correctly recognized their environments for what they really are, nor in recognizing the adequacy of the grounds of their further inferences and actions from among the many that are potentially available. I wish to treat such methodological interests of the members of a society as objects of theoretical sociological inquiry. I use the term “ethnomethodology” to refer to the study of such interests. The prefix “ethno” may help to fix such interests for our attention. The prefix means “seen from the point of view of the common sense interests of a member of society in the course of managing and coming to terms with his everyday affairs.” The term “ethnomethodology” was suggested by the index of Human Relations Area Files which provides for such topics as “ethnomedicine,” “ethnobotany,” “ethnophysics,” and the like. Such studies are concerned with matters of meaning, fact, method, and causal account about medical, botanical, and physical topics that persons, thought of as members of actual societies, construct, use, test, criticize, but before all – for our interests – that they hold each other to a knowledge and use of within the constraints imposed by a respect for the socially organized character of their everyday activities.

1

Talk given on 10/16/61.

© Springer-Verlag GmbH Germany, part of Springer Nature 2019 A.W. Rawls (Ed.), Harold Garfinkel: Parsons’ Primer, Beiträge zur Praxeologie / Contributions to Praxeology, https://doi.org/10.1007/978-3-476-04815-8_3

327

328

H. Garfinkel

The term “ethnomethodology” is intended in the sense consistent with this usage. “Ethnomethodology” refers to matters of meaning, method, fact, and causal account about meaning, method, fact, and causal account that thought of as members of actual societies – sociologists included – construct, test, respect, use, criticize, and above all hold each other to a knowledge and use of within the constraints imposed by a respect for the collectivity member’s grasp of the socially organized character of his everyday activities. In short, ethnomethodology is concerned with the descriptive study of the procedures, concepts, findings of common sense inquiries. A person will be said to be engaged in an inquiry when he is addressed to the tasks of clarifying the nature and related character of some set of events.2 His theoretical tasks consist of the problems he must solve in conceiving the event that he has chosen to study. His empirical tasks consist of developing to the point of their status of adequate grounds of his further inferences and actions – i. e. of developing to the point of warranted belief – cogent, credible, clear, reproducible, realistic, and where scientific inquiries are concerned, literal descriptions of these events. In accomplishing these tasks a person will subject his efforts in some measure to a “discipline.” Some set rules of procedure will serve as maxims of conduct to govern his decision about the correctness of his procedures and/or the acceptable character of his results of his theorizing and inquiring activities. We shall call such rules of procedure, his methods. I wish however to stress that we need not and should not presuppose that by invoking the notion of the “discipline” as maxims of conduct, that we thereby presuppose some given degree of clarity, or definiteness, nor above all need we presuppose their precoded character. We shall insist however on one feature as a required feature, namely, that in the manner of any legitimate order to which action may be oriented in its course, these maxims serve the person and others as grounds for recognizing the legitimacy of criticisms of his procedures and their products. In a word, we look to the norms of inquiring procedures, of his attitude, and of his claims about findings as the grounds upon which, and in the company of others, questions of competency are advanced and settled. In a word, the inquirer must be able to advance the claim that what he saw was to be seen, and that what he is talking about is understandable and is to be understood, and that thereby he is entitled to manage his affairs without interference. We shall look to this feature as the definitive and central feature in our talk about rules of procedure, and allow whatsoever other features that might be advanced 2

[Editor’s Note: This paragraph was crossed out.]

The Program of Ethnomethodology

329

or encountered, the status of correlated features of the phenomenon that is initially encountered via this central one. A member’s rules of procedure serve him as definitions of “correct inference,” “correct proof,” “correct findings” and the like. Among the many rules of procedure that he might call to our attention, we shall find some set that serve as the actual terms in which he decides the logically necessary relationships in the data, and in terms of which he counts an argument to have been adequately made, or an inference to necessarily follow from, or a demonstration of meaningful consistency, contradiction, contrariness to have been correctly made. In a word, there will be those rules of procedure in terms of which he decides the sensibility of his theorizing. There will be found a set of rules in terms of which the member decides that his findings can be treated as holding, as recognizable irrespective of who it is that recommends them (i. e. the anonymity of the observer) and that his findings hold for Anyman (i. e. communality). Let me refer to these as the procedures whereby the objectivity of his descriptions are decided. These rules of procedure are the terms in which he decides whether or not a proposition or a set of events can correctly be used as grounds of further inferences and investigation, i. e. rules in terms of which credibility of his findings is decided, or his definition of fact. Finally they are terms in which he decides that events been correctly ordered, on the one hand to relations of conditions and consequences, or that they have been correctly ordered to means-ends relationships, i. e. of good advice, or procedural descriptions of good work. I shall use the term causal texture to refer to both the ordered relationships of conditions and consequences, as well as to ordered means-ends relationships of effective conditions. Any set of rules of procedure that an inquirer uses, whether he uses them as ideal or actual maxims, in that they define “correct decisions” serve as solutions to these questions of sensibility, objectivity, and causal texture. I use the term “solutions” in the functional sense. The question is not whether or not but how the person solves these questions. What solutions does he actually use? Any inquirer – whether or not he has scientific interests in his inquiry and whether he be a physicist, a sociologist, a child, an astrologer, or a shaman has some solution to these problems. Scientific methodology, by which I mean the study of definitions of correct scientific decisions, begins its tasks with this elementary fact, for it takes as data for its inquiries, solutions in use. Ethnomethodology, as the study of correct decisions for inquiries of everyday life, also takes as its data solutions in use. A concern for solutions in use is the hallmark of the problematic phenomena of ethnomethodology. More generally, ethnomethodology seeks to

330

H. Garfinkel

describe the essential features of the methods of actions – any actions. With respect to whatsoever the person may be said to be doing, ethnomethodology seeks a description, the “how” of his doing it. Most clear as a matter of concern to the person doing it when he is engaged in inquiring activities, with singular claims in case of activities of scientific investigations and of penultimate concern in sociology which calls all activities under the jurisdiction of its attitude, and methods as ways of answering the two questions: what is a person really doing and how is he really doing it. It follows then that our study is directed to the work of sociologists – lay and professional – and consists in...3 An ethnomethodological study requires that when we address the results of an inquiry, we take into account the actual procedures that actual inquirers used to produce their results, that we examine the contexts of social actions that serve the inquirers as schemes of interpretation and that we look at these schemes of interpretation for their methodological import. The results of an inquiry can be any results whatsoever. They may range from physicists’ descriptions of subatomic particles to Institoris and Sprenger’s descriptions of aweful power in 17th century witches. Similarly, all methods of inquiry are eligible for our examination, regardless of who the inquirers might be and regardless of their claims, from the physicist to the shaman. When the procedures and results of inquiry are attended from an ethnomethodological point of view, we must expect that the wondrous things that procedures of the natural sciences find in the world are matched by the wondrous things found in the world by the procedures of inquisition, jury trial, boards of inquiry, divination, prayer, psychiatric examination, scholarly interpretation, sociological viewing, and guessing games. The remarkable features of the Van Allen radiation belt that are demonstrated by the procedures for their recognition that astrophysicists use are thereby to be compared in an ethnomethodological account with the remarkable properties of witches that are detected by the procedures for their recognition by the Azande. Finally, the features of the society’s inquirers that are used by them in managing the tasks of deciding sensibility, objectivity, fact and causal texture must also come under our examination. Because the features of commonsense inquiries are features of sociological attitude and methods we must be prepared to treat commonsense knowledge of social structures and the methods of common sense inquiries as both a topic of sociological inquiry as well as a feature of these self same inquiries. I mean to be as general as possible in this respect. Therefore while allowing for the 3

There is a last word here that looks like “qual”, which would not satisfactorily complete the sentence. There are also margin notes. One says something about “criticize Naegele”. The other says “Examples: definitions: detecting witches, interpreting dreams.”

The Program of Ethnomethodology

331

possibility that there may be important distinctions between the member of the society acting as a sociologist and the activities of professional sociological inquiries, I mean not to assume because of the claims of either set, or any set, of sociologists, that the distinctions in fact exist. Nothing requires that we assume that methodology in the social sciences is one thing and ethnomethodology is another. Instead, we ask that any beliefs about their identical or distinctive features be temporarily suspended until the features of compliance with whatever are the actual rules that produce a body of commonsense knowledge of social structures has been described. Later I hope to show that there are grounds enough for asserting that common sense knowledge of social structures is both a point of departure as well as a point of return for every portrayal of the social world and for every sociologist’s claim that the social world he describes is a real one, and further, that common sense knowledge of social structures sits in judgment upon every theoretical version of it. This is not to say, however, that common sense procedures in inquiry, and common sense knowledge of social structures differs from the procedures and the results of scientific let alone social activities and products in degree. After we have examined actual features of the procedures of inquiry, in terms of which real societies and real social structures are defined, we shall be able to ask about similarities and differences between the rules that sociological inquirers, most particularly professionals, invoke in describing their inquiring activities, as compared with the rules and procedures that they actually use. There is reason to anticipate the existence of important discrepancies between the two. For example, there is only the faintest chance that the status of findings that are assigned to the results of professional sociological inquiries, can be reproduced if one tries to do so by consulting the rules of procedure that are citied in textbooks of sociological methods if one reads only these rules and reads them literally, which is to say if one imagines a researcher who acts in no other way than in strict compliance with these rules, one encounters at every step gaps and incongruities between the imagined and the actual researcher, and between possible and cited findings. But while this is true, a warning is necessary to avoid abandoning the ethnomethodological enterprise in favor of drawing pejorative consequences. It is no part of the program of ethnomethodology, and in no way the purpose of my remarks tonight, to engage in the tiresome complaint that discrepancies exist between professional sociologists’ claims and their practices. Nor can it be in any way interesting to engage in the equally tiresome defense of the sociologist’s claims to the scientific character of their efforts. Several wellknown publications had as their topic and central argument the problematic

332

H. Garfinkel

possibility, or present actuality, of sociology as a rigorous science. Their methods consisted of little more than assembling exhibits to attest the tragic or comic discrepancies between what is going on and what the interested partisans claim is going on. But no one is more informed of these discrepancies than the practitioners themselves. And indeed, the more careful are practitioners to observe some definiteness and rigor in their researches, and the more are they concerned with and skilled and practiced with rigorous procedure, the more informed they are about these shortcomings, with the result that these recitations deliver news from nowhere. After one takes from these accounts whatever is descriptive about these practices, the reminder consists of little beyond the complaining advice, “Someone ought to do something.” Several efforts which are particularly noteworthy for their vigor and cleverness raise little beyond the suspicion that the author stirred up the waters the better to fish therein. If complaints are not enough, neither is it sufficient to warn the sociological investigator, as methods textbooks routinely do, that departures from prescribed procedures will distort the objective world in a mirror of prejudice. This is not sufficient either as warning, or as good advice, nor is it factually correct. Because the troublesome question remains: what kinds of objective worlds result from the actual decisions that an inquirer must make in coming to terms with the actual circumstances within which his inquiry must be accomplished. Because we are allowing at the outset that every person who addresses inquiries to the events of human conduct operates with some solution to the questions of sensibility, objectivity and the rest, the very variety of human types and interests, actions and circumstances dictates the existence of a variety of solutions in use. To each set of rules that define correct decisions in the tasks of deciding sensibility and the rest there corresponds the products of these decisions in the form of preferred conceptions and terminologies, and sanctionable grounds of further sociological inferences and inquiries. Hence one must anticipate the existence of different portrayals of the world, of different corpuses of knowledge that propose features of social structure that persons will treat as correct grounds inferences and actions. These may consist of descriptions of life in society that some set of members use as sanctionable grounds that other sets of persons, assuming they were afforded an opportunity to judge, would not sanction. If the “may” introduces a caveat with respect to plural bodies of fact, and thereby provides for multiple realities, it introduces a caveat as well with respect to the possibility of unified reality and the unity of the sciences as well. Of this I shall speak further in a moment.

The Program of Ethnomethodology

333

Little information is available on the consequences that different actual methods have in producing a body of knowledge of life in society with properties that are particular to the actual methods that produce this body of knowledge. In the ethnomethodological program we are concerned with characteristics of actually used rules deciding among alternative courses of interpretive and investigative procedure that serve sociologists, lay and professional, as solutions to their tasks of deciding sensibility, etc. and thereby lend to the body of fact the properties that result from such procedures. It is our prevailing task to describe the attitudes, methods, problems, and results of sociological inquiry in general; to describe what they look like procedurally. What does the “discipline” look like in the actual practices of sociologists? How does the work of sociology get done? In a phrase, how are the products of sociological inquiry produced? What are the procedures of sociological procedure, and how are their products thereby produced as a human enterprise? We want to return the study of the activities of sociological inquiries and the products of these inquiries to the natural philosopher – to the one who says: “I would like to see for myself how it is done while being permitted to abstain from any judgments that require a respect for the authority and dignities of sociological procedures and results that accompany their claims as a socially sanctioned professional enterprise”. In no case can or should ethnomethodology pass judgment on whether or not researchers have adequately answered a question. Nor does, can, or should ethnomethodology take as its task the development or the assertion of standards to permit such judgment. Such a task is put aside because the procedures that sociologists actually use, what their accepted procedures are, is exactly the object of our inquiries. We follow closely therefore Wittgenstein’s advice as the factual grounds for this abstention, when he asserts that every move is a rational move in some game. Journals and textbooks are filled with procedural accounts that use the accomplished result in hand as a prologue from the point of view of which the steps whereby it was produced are reconstructed. In a manner very similar to historical reconstruction, the end is known at the beginning of the account, as the event to which the procedural steps had inevitably to lead. As ethnomethodologists we would seek instead to describe the how-it-is-done over the temporally extended next steps, when the result consists at best of an anticipated possible future to be arrived at, by a course of actual manipulations upon each present state of affairs. The adequacy of the answer, then must be contained in the procedures whereby the inquirer must progressively come to terms with the developing actual occasions of his inquiry, directed to an environment of accomplished and following possibilities which he can know only

334

H. Garfinkel

by actual actions that are committed to a more or less clear plan of what his actual investigative activities will finally have come to. For the descriptions of both actual procedures and the actual results of these actual procedures, our findings about rules of good procedure and the criteria of adequate answers are to be described by paying attention exclusively to the procedures and results as they occur from “within” the inquirer’s environment over the course of his step-wise management of that environment in the interests of bringing it to terms. This is not to say, however, that the ethnomethodologist is not interested in the question of the “validity” of the inquirer’s results. By no means and in no way does this procedural emphasis neglect the question of the “realistic” character of any procedure’s results. The problematic correspondence of accepted results of sociological procedures to the real world is very much an ethnomethodologist’s concern. But nothing requires that this relationship be thought of with the use of the conceptions of logical and empirical demonstration that have been received from the physical and biological sciences. Later we shall see that the presumed cornerstone of sociology’s claims to scientific descriptions, the empirical proposition, does not describe the logical structure of sociology’s data sentences. It imposes conditions for warrant that sociologists in the entire history of their inquiries have yet satisfy. Somehow the features of actual observation and the insistence upon actual observation as a condition of warrant have existed side by side with an insistence that the empirical propositions described the data sentences of sociology, when nothing of the sort is demonstrable, except as a perennial and leading ideal. I shall show later that the data sentences that incorporate the sociologist’s observations and that furnish the actual grounds of his claims to concerns with fact and empirical knowledge are production propositions.4 For various reasons we must be prepared to suspend belief that there is a unity in the sciences and that this unity resides in its methods. Apart from the fact that the methods of ethnomethodological inquiry require such a suspension of belief, the circumstances of sociological inquiry themselves dictate the wisdom of such a suspension until the intended events of sociological interest themselves have had a full turn to assert their authority for subscription to this doctrine. Studies of human conduct, as we shall see, have as their fundamental occurrences the phenomena of rule-governed conduct, and therewith is the accompanying and obstinately puzzling phenomena of “signification.”

4

[Editor’s Note: Garfinkel includes a handwritten paragraph here that is largely illegible. From what can be made out, however, it seems to raise issues concerning social scientific procedure.]

The Program of Ethnomethodology

335

Insofar as the student of human conduct seeks in their descriptions to remain faithful to the features of the events that he insistently intends to call attention to in deciding the realistic character of his results, he is continually confronted with the necessity of coming to terms with two problems that physical and biological sciences can neglect: the problem of social order and the problem of the sociology of knowledge. I wish to argue that the acceptance of these problems is definitive of the social sciences as an intelligible enterprise for its practitioners, and from their point of view these problems are essentially implicated in social scientists’ concerns with the literal description and observations of the events of human conduct carried on under the auspices of and subject to the orientations of scientific procedure. These two problems permit us to identify the activities of sociological inquiry in general, and to identify persons acting as sociologists, without requiring a reference to membership in the professional associations in order to say who is acting like a sociologist and who is not. By a person acting like a sociologist – by any person acting like a sociologist – I shall mean a person whose inquiries are addressed to the tasks that make up the problem of social order while directed to or while assuming a solution to the problem of the sociology of knowledge. I shall elaborate on this in a moment. In the ethnomethodological program we are required to put aside the framework of investigation that is usually employed by social scientists. To do so, it required that we either have for ourselves a “platform of evaluation” or allow the devastating criticism that we are after all engaged in a bootstrap operation. This platform consists of having elected to search out the elementary character of institutionalized conduct. The “platform” consists of the problem of how stable courses of treatments of situations that are known in common and taken for granted are possible. Where sociological researchers – lay or professional – are of interest to us, the problem can be posed as follows: what do these persons treat as known in common with others that they do in fact take for granted in deciding that they know what they are talking about and that they know what they have been looking at. The problem arises because characteristically, and perhaps necessarily, sociologists are not limited in their descriptions to here and now appearances, nor do they wish to be limited, nor could they afford to be limited if they were to satisfy the requirements for their descriptions that they are adequate descriptions of rule governed actions. I am talking descriptively, not ironically. I am not saying, “See how far short they fall in their claims to present the facts.” Instead, the question is being posed: What do sociologists’ sanctionable practices show fact to consist of? What do sociologists intend by their claims of adequate description of the real world to be claiming? Or,

336

H. Garfinkel

put otherwise, granted that sociology and the other social sciences are facts in Felix Kaufmann’s sense of fact, then how are they possible as the facts that they actually consist of? The validity of their products then is to be understood then in terms of the procedures for recognizing the occurrence of a case of adequate structural analysis, or, adequate description of social structures. The invariant relevance to the sociologist’s descriptive and explanatory enterprise of the problems of social order and the sociologist’s knowledge comes in exactly at this point. In respect to the problem of social order, for example, the sociologist’s solution in use consists of his theory of organization. In its procedural sense it consists of a solution to the problem of social order and is likewise in its procedural sense identical with the user’s definition of adequate structural analysis. (Insert here material on both problems and equivalence of the theory of organizations as a solution to the definition of adequate structural analysis. See Parsons’ Primer). He shows the equivalence of any theory of organization to a definition of adequate structural analysis when the theory of social organization is given its procedural meaning with respect to the tasks of bringing knowledge at hand about the society under the jurisdiction of the theory as a definition of its essential reproducible, orderly, persistent, continuing, socially standardized features. Given the platform problem, we have elected the position of the phenomenological attitude and the phenomenological method as Edmund Husserl described it. The platform consists in effect of a method of criticism; it is a method as Husserl talked about it for the radical clarification of what the world means. We have borrowed heavily from the work of the late Alfred Schutz. Indeed the entirety of the work can be said to have been inspired by the work of the late Alfred Schutz on the constitutive phenomenology of practical actions in situations of everyday life, work of describing what he called the world of socially structured activities of everyday life known in common with other and with others taken for granted. We have consulted Parsons’ theory of social systems to aid in collecting our thoughts about the nature and conditions of stable structures of normatively regulated interpersonal transactions. Most particularly, however, I have borrowed from Parsons’ work in which he delineated the constituent tasks that make up the problem of social order. This work holds independently of whatever reservations one might have about the particular solutions that he elaborated over the course of his intellectual history. By way of summary, it is a central problem of the ethnomethodological program to make intelligible and to clarify the procedures whereby sociologists recognize sensibility, objectivity, warrant, and causal texture while seeking to abstain from judgment about whether what they say is true or false, important

The Program of Ethnomethodology

337

or trivial, valuable or worthless, correct or incorrect in any other terms than those that are possible within the inquiring games that furnishes for sociologists themselves the reasonable, understandable, realistic character of their own actions and products over the course of their own actions, within their own histories, and within their own social establishments. The ethnomethodological enterprise is directed to bringing their practices, their claims, and their products to intelligibility and to clarity as courses of rule governed conduct while abstaining from any judgments of importance, value, truth, relevance of what they are doing and what they are producing, and to conduct our inquiries and descriptions in such a fashion as to respect this abstention. Overall our purposes are (1) to detect the constituent features of sociological inquiry, lay and professional, and its products, the real social structures, and (2) to clarify the role of common sense knowledge, common sense activities, common sense environments of objects, and the methods of common sense thinking and conduct in the investigative activities, discourse, and findings of sociologists, lay and professional. A word of reservation. It is not the purpose of the ethnomethodological program to recommend common sense knowledge of social structures as “another variable.” If criticism should demonstrate that this is what the program amounts to then it will have failed in its purposes. I will frankly admit that I am after big game. I think it is possible to show that, given the program of sociology the professional sociologist has already presupposed the essential relevance of common sense knowledge and common sense methods to the entirety of their work. And that therefore they have no choice but to address the phenomena of common sense activities and common sense knowledge of society directly as a central, essential, and invariant theme of their inquiries; To explore these phenomena in all their consequences, and to incorporate the results into the methods and findings of the discipline. Insofar as the study of common sense phenomena as a subject matter in its own right is ignored, the radical consequence is that the problem of meaning in all the many puzzling ways it asserts itself remains without a rationally defensible solution, and the aims of rigorous description are everywhere turned away from being satisfied in real events. The result is that for all its concerns and claims to rigor in theorizing, methods, and objectivity of findings – concerns that are exemplified for example in the widespread and spreading insistence upon mathematical models and statistical procedures for gathering and presenting data and for defining adequate inference – the sociological investigator must depend and depends upon the socially organized, well disciplined community of co-believers drawing upon common understandings, employing

338

H. Garfinkel

common courses of understanding of familiar scenes, and drawing upon common sense knowledge of society that “anyone knows” in order successfully to recommend the sensible character of the matters that he purports to investigate, to recommend the objectivity of his methods for investigating them, to recommend the warranted character of what he asserts as his findings. To abandon the concern for common sense activities is to abandon the program of sociology itself in favor of the rigorous use of biological and physical theory and methods in order to study the biological and physical properties of the events of human conduct. Egon Bittner Aaron Cicourel Craig MacAndrew Edward Rose Henry Riecken Eleanor Bennert-Sheldon Harvey Sacks Saul Mendlovitz Evelyn Hooker Elliot Mischler Charles Frake Duane Metzger Valney Steflre Dell Hymes Anselm Strauss Erving Goffman Omar K. Moore Arnold Mendel

The Appendices

© Springer-Verlag GmbH Germany, part of Springer Nature 2019 A.W. Rawls (Ed.), Harold Garfinkel: Parsons’ Primer, Beiträge zur Praxeologie / Contributions to Praxeology, https://doi.org/10.1007/978-3-476-04815-8

339

340

The Appendices

Appendix #1 Letters of Parsons and Garfinkel

The Appendices

Fig. 1: Parsons, T. to Garfinkel, H. – July 14, 1958

341

342

Fig. 2: Parsons, T. to Garfinkel, H. – August 14, 1958

The Appendices

The Appendices

Fig. 3: Garfinkel, H. to Parsons, T. – August 19, 1958

343

344

Fig. 4: Garfinkel, H. to Kaspar – August 20, 1958

The Appendices

The Appendices

Fig. 5: Garfinkel, H. to Martel, M. U. – August 20, 1958

345

346

Fig. 6: Garfinkel, H. to Schutz, A. – August 20, 1958

The Appendices

The Appendices

Fig. 7: Parsons, T. to Garfinkel, H. – August 28

347

348

Fig. 8: Garfinkel, H. to Parsons, T. – Page 1 – January 22, 1959

The Appendices

The Appendices

Fig. 9: Garfinkel, H. to Parsons, T. – Page 2 – January 22, 1959

349

350

Fig. 10: Garfinkel, H. to Parsons, T. – Page 1 – January 14, 1963.

The Appendices

The Appendices

Fig. 11: Garfinkel, H. to Parsons, T. – Page 2 – January 14, 1963.

351

352

Fig. 12: Garfinkel, H. to Parsons, T. – Page 3 – January 14, 1963.

The Appendices

The Appendices

Fig. 13: Garfinkel, H. to Parsons, T. – Page 4 – January 14, 1963.

353

354

Fig. 14: Parsons, T. to Garfinkel, H. – Page 1 – January 30, 1963

The Appendices

The Appendices

Fig. 15: Parsons, T. to Garfinkel, H. – Page 2 – January 30, 1963

355

356

The Appendices

Appendix #2 The Situated Character of Garfinkel’s Comments about Influences on his Work Garfinkel’s letters display the situated character of his statements about how his position was indebted to Parsons and Schutz. It is evident that while he built on both, he emphasized one or the other depending on whom he was communicating with and what the communication was about. For instance, in a letter to Naegele dated August 20, 1958, after saying that he had “put everything aside in order to work through the large number of Talcott’s manuscripts” (for the conference at Harvard that September), Garfinkel mentioned his indebtedness to Schutz: “As you know, many years ago I took Schutz’ work on the constitutive phenomenology of situations known in common and treated in the manner of common sense as a point of departure for conceiving the “problem of social order”. Since then I have laboriously worked through several empirical and theoretical tasks imposed by this re-reading. The result has been a conception of a system of activities made up of persons who are conceived to treat an environment of cultural objects whose features which specify their sense as objects-known-in-common-with-others are assigned to them through the presuppositions of Schutz’ attitude of daily life, and who establish, maintain, restore, defend, etc. the perceivedly normal features of these environments through the methods of commonsense thinking and conduct. I have a series of researches underway to explore the consequences of these actions and a book in the works on the study of norms and normative orders summarizing the work”.

In explaining his position to Naegele, Garfinkel assigns a particular role in his thinking to Schutz, claiming his “commonsense” as a point of departure for considering the problem of social order and citing “research underway”. He does not tell Naegle that he is building on Parsons. He doesn’t need to: Naegle already knew that Garfinkel is building on Parsons’ work. That is why they have both been invited to Harvard. What Naegle might not have known about is Garfinkel’s relationship to Schutz. The letter portrays Garfinkel as reading Parsons’ work in a serious way, while taking Schutz’ notion of commonsense as a point of departure for that reading of Parsons. In a letter to Parsons written on the previous day, August 19, 1958, Garfinkel says much the same thing about his indebtedness to Schutz:

The Appendices

357

“As you know, I have for many years used Schutz’ materials on the constitutive phenomenology of situations known in common and taken for granted as a point of departure for re-reading various sociological theorists for their conceptions of and solutions to the “problem of social order”. The result has been a working conception of a system of activities which is made up of actors who are conceived to treat an environment of objects as objects known in common with others and as such are taken for granted; whose features as objects that are known in common are assigned to them through the presuppositions of the attitude of daily life, with the actors establishing, maintaining, restoring, defending, etc. the perceivedly normal features of these environments through the methods of common sense thinking and conduct”.

This is interesting because in this case Garfinkel is writing directly to Parsons, with whom he is on the point of embarking on a collaboration, and stressing the importance of his debt to Schutz. But, as with Naegle, Parsons already knows that Garfinkel is building on his own work. That is why he has invited him to the conference. What Garfinkel is letting him know is that he is still also building on Schutz, something that Parsons would have known from Garfinkel’s graduate work. Garfinkel again emphasizes the idea that he has used Schutz as “a point of departure for re-reading various sociological theorists for their conceptions of and solutions to the ‘problem of social order.’” In other words, he is telling Parsons that he has used Schutz as a point of departure for reading Parsons. Having told Naegle and Parsons that he has used Schutz as a point of departure, however, Garfinkel then writes to Schutz on August 20, 1958, telling Schutz that he (Garfinkel) is one of a group who have taken Parsons’ work as a “point of departure” for their work, which is why he is being invited to the Harvard conference: “Parsons and several former students and me. All of us, in various ways, have used Parsons’ work as a point of departure. The conference is to review the present state of Parsons’ and our work”.

The letter tells Schutz who will be at the gathering and what they will be discussing. This letter, in which Garfinkel tells Schutz that he will be part of a gathering of scholars who have “used Parsons’ work as a point of departure,” was written the day after Garfinkel wrote Parsons that he had taken Schutz as a point of departure, and the explanation is given in the context of telling Schutz about his plans to attend the meeting with Parsons at Harvard that September, while at the same time trying to arrange a meeting with Schutz in New York after the conference.

358

The Appendices

What appears to matter to Garfinkel is that everyone who is important to him understands that he is building on both scholars. Yet, depending on who he is talking to, he can take for granted that they know he is indebted to one or the other, and he only fills in the information he thinks might be missing. To Parsons and Naegle, he says that Schutz has been an important point of departure for his work. To Schutz, he says that Parsons has been a point of departure. Rarely does Garfinkel discuss his indebtedness to both in the same statement (he suggests the relationship in Chapter IV, page 46). But, in the “Program” manuscript that ends this volume he explicitly attributes his work to both Parsons and Schutz. There is a paragraph in which Garfinkel says that “the entirety” of his work was inspired by Schutz and then follows up in the next line saying that he has borrowed from Parsons. There is not even an “also” between the two statements (Program pg. 15 our draft): “We have borrowed heavily from the work of the late Alfred Schutz. Indeed the entirety of the work can be said to have been inspired by the work of the late Alfred Schutz on the constitutive phenomenology of practical actions in situations of everyday life, work of describing what he called the world of socially structured activities of everyday life known in common with others and with others taken for granted. We have consulted Parsons’ theory of social systems to aid in collecting our thoughts about the nature and conditions of stable structures of normatively regulated interpersonal transactions. Most particularly, however, I have borrowed from Parsons’ work in which he delineated the constituent tasks that make up the problem of social order. This work holds independently of whatever reservations one might have about the particular solutions he elaborated over the course of his intellectual history”.

Statements like these can be a bit confusing and, taken out of context, be easily misunderstood. While it is widely believed that Garfinkel rejected Parsons in favor of Schutz, and that Parsons, for his part, took little or no interest in Garfinkel, it is simply not the case. Letters and documents in the GA show that Garfinkel described himself as indebted to both Parsons and Schutz, in very specific and complimentary ways, that he remained friends with both, and did not make the intellectual choice between them that has been alleged. While it is easy to see how the situated character of Garfinkel’s attributions of indebtedness could have been misunderstood when taken out of context, it also is very clear what he meant. The many references to Schutz created the impression that Garfinkel was a phenomenologist when he was not, while his writing and teaching about Parsons went unrecognized, a problem he complained about to the end of his life.

The Appendices

359

Garfinkel was synthesizing several theoretical approaches to get at the question of how mutual coherence in language and action is achieved in social interaction, and how that relates to the overall problem of social order. Equating Garfinkel with Schutz obscures the fact that the framework of Garfinkel’s approach to interaction has a lot in common with Parsons, and borrows also from Wittgenstein, Mead, Znaniecki, W. I. Thomas, and Charles Horton Cooley, in addition to Weber and Durkheim. The letters between Garfinkel and Parsons about the Primer, the September 1958 meeting at Harvard, and audio recordings in the Archive, make it clear that Garfinkel was one of a handful of Parsons’ most valued and trusted students with interests in social theory in 1958, and that he continued to work on a synthesis of Parsons’ position with his own interactionism through the late 1960’s. If Garfinkel saw himself as working out “Durkheim’s Aphorism” (Garfinkel 2002) via Parsons’ positioning of social interaction as an independent domain of culture, and was doing so in collaboration with Parsons during the crucial years 1958–1964, when he was also drafting the “Program of Ethnomethodology,” writing his paper on Trust and completing the first studies in Ethnomethodology then their relationship was obviously important. If Parsons was not interested in what Garfinkel was doing, then why did he send letters and invitations to Garfinkel? Why did he read the Primer manuscript, apparently more than once, and respond to it both in person and in detailed letters to Garfinkel (see especially January 30, 1963), making thoughtful suggestions and expressing his wish to meet for longer discussions (meetings that took place at Harvard in the fall of 1959, and Los Angeles in the Spring of 1964). If Garfinkel had repudiated Parsons (or had chosen Schutz over Parsons), then why did he spend the years 1958–1964 doing extensive groundwork for this book on Parsons, teaching from the Primer twice in his courses, and continuing to hand out the mimeograph version to his students through the 1980’s? If there was no significant relationship between them, then how was Garfinkel able to spend that time with Parsons, meeting with him four times between 1958 and 1964, once for an entire semester at Harvard, during which they worked on the revision of the pattern variable manuscript together? Finally, and most puzzling, if Garfinkel was working on a project in some kind of collaboration with Parsons, that Parsons was interested in, as materials in the GA document, then why don’t we know about it? The void left by the lack of an answer to these questions has fueled conflicting narratives about Garfinkel and Parsons. Not understanding how committed Parsons was to centering interaction in his theory – and not understanding what Garfinkel was trying to convey in the Primer manuscript – there

360

The Appendices

has been a tendency to interpret Ethnomethodology as phenomenology and Parsons’ later work as continuing his earlier Structural Functionalism. Not understanding the degree to which Parsons’ actors’ actions and cultural objects lived within a circle of socially defined constitutive practice, or how serious Garfinkel was about grounding Ethnomethodology on a modified Parsonian theory, and not at all understanding the depth of the relationship between the two, or their joint commitment to the idea of culture as interaction in-itscourse, various narratives have attempted to distinguish an earlier period in which Garfinkel is alleged to have done more conventional work from a later period in which he is portrayed as having gotten radical and turning his back on Parsons, sociology, and even on Sacks and the Conversation Analysts (CA) (e. g. Wilson 2003).1 The claim that there was a sharp change in Garfinkel’s work, or that he was close to Parsons early on, but changed course later when he began to do the research that became Studies in Ethnomethodology (1967), or that a sharp division emerged in the 1960’s between Garfinkel and Sacks, is simply not supported by the evidence. In fact, the aspects of Garfinkel’s work that are said to have occurred before and after the alleged change were all going on at the same time.

1

Garfinkel’s early work with Sacks from 1960–1962, and the writing and research Garfinkel did before Sacks came to study with him, remain essentially unknown. Sacks’ interest in language developed while he was working with Garfinkel in 1961–2 before Schegloff or Jefferson arrived at UCLA. With the exception of recordings and joint authored papers their relationship has been primarily constructed through hearsay. Even where there are documents and recordings to support their collaborative work scholars have preferred to believe the hearsay. For instance, there are claims that Sacks had not worked on a co-authored paper with Garfinkel in 1967 that was published in 1970. Recordings and documents in the GA show otherwise. Garfinkel began criticizing formal versions of CA that diverged from Sacks early on and never changed his mind. Garfinkel made the point in Ethnomethodology’s Program (2002). Materials in the GA also show that Garfinkel made frequent revisions to arguments. It does not signify that he changed what he was arguing, only that he had found a clearer way of making a point, or was giving a paper orally and trying to cut down the length.

361

The Appendices

Appendix #3 Syllabus for SO 172/251 Department of Anthropology and Sociology University of California, Los Angeles Sociology 251: Selected Topics in the Problem of Social Order: Talcott Parsons Instructor: Harold Garfinkel Spring 1959 #Republished in 1949 edition of Essays *Republished in 1954 revised edition of Essays #*Republished in both editions of Essays

A NOTE ON THIS BIBLIGRAPHY It is my intent to teach you how to read and use Parsons in the interests of research. His writings have been arranged by seminar topics. Within topics, writings are listed with the most recent first. His most recent writings should be read as the precedent for the earlier ones. Reading the corpus “backwards” will lend to the earlier writings their sense of what Parsons has been up to “all along” or “after all.” Some criticisms of such re-reading procedure are that it produces an “interpretation” of Parsons; that it runs the risk of reading Parsons for more, less, different, better, or worse than he intended; that the most recent formulations are a shaky platform since they are still undergoing revision. Nevertheless, I prefer this procedure to a chronological reading since we shall be very little interested in the development of Parsons’ ideas and very much interested in how to use them. I urge that the materials in topics III, IV, V, and II be read in that order and before proceeding with the rest. Relevant materials in topic III should then be consulted in integral parts of the topical bibliographies. It is my hope that many bad listings and topic headings will be remedied in the course of and with the help of the seminar.

362

The Appendices

To the best of knowledge, this bibliography includes all of Parsons’ published and unpublished writings as of September, 1958.

I. Parsons’ early review and criticism of the utilitarian positivistic and idealistic definitions of and solutions to the problem of social order (1928) “Capitalism” in Recent German Literature: Sombart and Weber, Journal of Political Economy 36: 641–661. (1929) Capitalism in Recent German Literature: Sombart and Weber, Journal of Political Economy 37: 31–51. (1931) Wants and Activities in Marshall, Quarterly Journal of Economics 46: 101–140. (1932) Economics and Sociology: Marshall in Relation to the Thought of His Time, Quarterly Journal of Economics 46: 316–347. (1933) Malthus, Encyclopedia of the Social Sciences 10: 68–69. (1933) Pareto, Encyclopedia of the Social Sciences 11: 576–578. (1934) Some reflections on “The Nature and Significance of Economics”, Quarterly Journal of Economics 48: 511–545 (1934) Society, Encyclopedia of the Social Sciences 14: 225–231 (1935) Sociological Elements in Economic Thought I. Quarterly Journal of Economics 49: 414–453. (1935) Sociological Elements in Economic Thought II. Quarterly Journal of Economics 49: 645–667. (1935) H. M. Robertson on Max Weber and His School. J. Political Economy 43: 688–696. (1935) The Place of Ultimate Values in Sociological Theory. International J of Ethics 45: 282–316. (1936) Pareto’s Central Analytical Scheme. J of Social Philosophy I: 244–262. (1936) On Certain Sociological Elements in Professor Taussig’s Thought. Viner, Jacob (ed.) Explorations in Economics: Notes and Essays contributed in honor of F. M. Taussig. New York: McGraw-Hill: (xxi and 539 pp.): 359–379. (1939) Comte, J of Unified Science 9: 77–83.

363

The Appendices

II. Parsons’ initial definition of and solution to the problem of social order (1937) The Structure of Social Action. New York: McGraw-Hill: xii and 817.

III. Parsons’ most recent definition of and solution to the problem of social order (1) THIS GROUP OF WRITINGS IS TO BE READ AS THE PRECEDENT FOR ALL THE REMAINING TOPICS AND THE REMAINING EARLIER WRITINGS. (2) TOPIC IV, “THE GENERAL THEORY OF ACTION” AND TOPIC V, “THE SOCIAL SYSTEM” ARE CONSTITUENT PARTS OF TOPIC III AND ELABORATE ITS MATERIALS. (1959)

(To be Pub.) (To be Pub.) (To be Pub.) (To be Pub.) (To be Pub.) (1958)

(1956)

The Role of General Theory in Sociological Analysis: Some Case Material (Presented to the Theory Section of the American Sociological Society, Washington, DC, August 1957. In a collection of papers presented at the Annual Meeting of the American Sociological Association, 1957 edited by Merton, Cottrell, and Broom. New York: Basic Books, 1958). The General Interpretation of Action. (Introduction to Part I, Section A of Reader on Sociological Theory). General Introduction Part II, An Outline of the Social System (To Reader on Sociological Theory). The Principle Components of the Institutional Structure of a Society. A Tentative Outline of American Values. Durkheim’s Contribution to the Theory of Integration of Social Systems. An Approach to Psychological Theory in Terms of the Theory of Action, Sigmund Koch (ed.) Systematic Theories in Psychology. New York: McGraw-Hill. Economy and Society, Co-author with Neil J. Smelser. The Free Press.

364

The Appendices

IV. The general theory of action A PROPOSED SOLUTION WITHIN THE RULES OF SCIENTIFIC PROCEDURE TO THE FOLLOWING TASKS: ACHEVING A THEORETICALLY AND CAUSALLY ADEQUATE DESCRIPTION OF ACTIONS AND THEIR OBJECTS: DECIDING THE NECESSARILY RELATED CHARACTER OF EMPIRCIALLY OBSERVED UNIFORMITIES OF ACTIONS AND THEIR OBJECTS: (1958)

“Some Highlights in the General Theory of Action”, in Roland Young (ed.): Approaches to the Study of Politics. Evanston: Northwestern University Press. (1956) Elements pour une theorie de l’action, With introduction by Francois Bourricaud. Paris: Plon. (1953) (WP) Ch. 3 The Dimensions of Action-Space. (1953) (WP) Ch 5 Phase Movement in Relation to Motivation, Symbol Formation, and Role Structure. (1951) (SS) I. The Action Frame of Reference and the General Theory of Action Systems: Culture, Personality, and the Place of Social Systems. (1951) “Some Fundamental Categories of the Theory of Action: A General Statement”, by Parsons et al in Talcott Parsons and Edwards Shils (eds.): Toward a General Theory of Action. Cambridge: Harvard University Press: 3–29. Also (listed under other topics): II. The Structure of Social Action, Ch. 2, relevant parts of Chs. 17 and 19. III. (1958) An Approach to Psychological Theory in Terms of the Theory of Action.

V. The social system. (The society as a critical case of the social system) A PROPOSED SOLUTION WITHIN THE RULES OF SCIENTIFIC PROCEDURE TO THE FOLLOWING TASKS: (1) CONCEIVING A DETERMINATE ORGANIZATION OF SOCIAL ACTIONS AND THEIR OBJECTS.

The Appendices

365

(2) ACHIEVING A DEVICE TO PERMIT THE DETECTION AND SYSTEMATIC COMPARISON OF STRUCTURALL REPEATED SITUATIONS WHICH, WHILE DIFFERING IN THEIR EMPIRICAL SPECIFICATIONS, ARE WITH RESPECT TO FORMAL ANALYTIC FEATURES ORGANIZATIONALLY IDENTICAL. (1956)

“The Social System: A General Theory of Action”, in Roy R. Grinker (ed.): Toward a Unified Theory of Human Behavior. New York: Basic Books, Inc.: 55–69. (1951) (SS) Ch. II. The Major Points of Reference and Structural Components of the Social System. The Functional Prerequisites of Social Systems The Institutional Integration of Action Elements The Points of Reference for the Classification of Institutional Patterns Types of Institutionalization Relative to the Social System Outline of Modes and Types of Action-Orientation, Culture Patterns and Institutions The Pattern-Alternatives of Value-Orientation of Definitions of Relational Role-Expectation Patterns (1951) (SS) Ch. III The Structure of the Social System: The Organization of the Components into Sub-Systems. The Structure of the Relational Context of Evaluative ActionOrientations The Modalities of Objects as Foci of Role-Expectations The Solidarity of the Collectivity Types of Social Value-Orientation (1951) (SS) Ch. IV The Structure of the Social System II: Invariant Points of Reference for the Structural Differentiation and Variation of Societies. The Foci of Crystallization for Social Structure The Internal Differentiations of Social Systems I. Categorization of action-units in object roles II. Classification of Orientation-role types III. The “Economy” of Instrumental Orientations IV. The “Economy” of Expressive Orientations V. The Cultural Orientation Systems VI. Integrative Structures

366

The Appendices

(1951) (SS) Ch. V The Structure of the Social Systems, III: Empirical Differentiation and Variation in the Structure of Societies. Some Empirical Clusterings of the Structural Components of Social Systems 1. The Universalistic-Achievement Pattern 2. The Universalistic-Ascription Pattern 3. The Particularistic-Achievement Pattern 4. The Particularistic-Ascriptive Pattern (1946)* Population and Social Structure (of Japan), in Haring, Douglas G. (ed.): Japan’s Prospection. Cambridge: Harvard Univ. Press, 1946 (Xiv and 474 pp.): 87–114. (This book was published by the staff of the Harvard School for Overseas Administration.) (1942)* Democracy and Social Structure in Pre-Nazi Germany, Journal of Legal and Political Sociology 1: 96–114. Also (listed under other topics): II. “General Introduction to Part II of Reader in Theory”. “Principal Components of the Institutional Structure of a Society”. “Tentative Outline of American Values”. IV. “Some Principal Characteristics of Industrial Societies”. “A Sociological Approach to the Theory of Organization”. “Some Ingredients of a General Theory of Organizations”. “The Mental Hospital as a Type of Organization”. “The Principal Structures of Community”. “Economy and Society”. “A Revised Analytical Approach to the Theory of Social Stratification”.

VI. Some organizations of social actions. A. Collectivities 1. Industrial societies (To be Pub.) Some Principal Characteristics of Industrial Societies (In Problems of Soviet Society, edited by C. E. Black)

The Appendices

367

2. Corporate organizations (1956)

A Sociology Approach to the Theory of Organizations, Administrative Science Quarterly I (June 1956): 63–85; II (September 1956):225–239. (To be Pub.) Some Ingredients of a General Theory of Formal Organization (In a volume by the Dept. of Education, University of Chicago, edited by Andrew W. Halpin). (1957) The Mental Hospital as a Type of Organization, Milton Greenblat, Daniel J. Levinson, and Richard H. Williams (eds.): The Patient and the Mental Hospital. Glencoe, Ill: Free Press. 3. “Ascriptive solidarities” (1955) (FSIP) The American Family: Its Relations to Personality and to the Social Structure: 3–34. (1949) “The Social Structure of the Family”, in Ruth Narda Austin (ed.): The Family: Its Function and Destiny. New York: Harper and Brothers: 173–201. (1943) #* The Kinship System of the Contemporary United States, American Anthropologist 45: 22–38. 4. Communities (To be Pub.) The Principal Structures of Community: A Sociological View (Written for a meeting of the Society for Legal and Political Philosophy, Cambridge, Mass. In a volume on Community edited by C. J. Friedrich). B. “Functional sub-systems” 1. The economy as a social system (1956) Economy and Society, Co-author with Neil J. Smelser. The Free Press.

368

The Appendices

2. The polity (1958) Authority, Legitimation, and Political Action, C. J. Friedrich (ed.): Authority. Cambridge, Mass.: Harvard University Press. (1957) The Distribution of Power in American Society, World Politics X (October 1957): 123–143. Review by C. Wright Mills. The Power Elite. (1942) Max Weber and the Contemporary Political Crisis, Rev Politics 4: 61–76; 155–172. Also (Listed under other topics): Part IV, “The Institutionalization of Authority” of Introduction to translation of Weber, in 1949 Essays. VII. Voting and equilibrium of the American political system 3. Religion (1958)

The Pattern of Religious Organization in the United States. Daedalus (Summer 1958): 65–85, issued as V.87 No. 3 of the Proceedings of the American Academy of Arts and Sciences. (1957) Reflexions sur les Organisations Religieuses aux Etats-Unis, Archives de Sociologies des Religions (January-June): 21–36. (1947) Some Aspects of the Relations Between Social Science and Ethics, Social Science 22: 213–217 (Read at the Annual Convention of the American Association for the Advancement of Science in Boston, December 5, 1946). (1944) #* The Theoretical Development of the Sociology of Religion, Journal of the Hist. of Ideas 5: 176–190 (Originally written to be read at the Conference on Methods in Science and Philosophy in New York, November, 1942). Also (Listed under other topics): VIII 1. (SS) Ch. VIII Belief systems of the social system II. Structure of Social Action, Chs. 11, 14, 15

The Appendices

369

C. Other 1. Age and sex roles (1942) #* Age and Sex in the Social Structure of the United States, ASR 7: 604–616 (Read at the Annual meeting of the American Sociological Society in New York, Dec. 1941 and republished in several places notably Wilson and Kolb, Sociological Analysis and Kluckhon and Murray, Personality in Nature, Society and Culture, 1st and 2nd editions.) 2. Social stratification (1953) *

A Revised Analytical Approach to the Theory of Social Stratification, Reinhard Bendix and Seymour M. Lipset (eds.): Class, Status and Power: Reader in Social Stratification. Glencoe, Ill: The Free Press. (1940) #* An Analytical Approach to the Theory of Social Stratification, American Journal of Sociology 45: 841–862.

VII. A system’s relations with its environments (1958)

‘Voting’ and the Equilibrium of the American Political System (In Continuities of Social Research: III The Studies of Voting Behavior, edited by Eugene Burdick and Arthur Brodbeck, Glencoe, Ill: The Free Press). (1956) Economy and Society (Coauthor with Neil J. Smelser, London: Routledge and Kegan Paul; and Glencoe, Ill: Free Press). (1956) “The Relation Between the Small Group and Larger Social System”, and “Boundary Relations Between Sociocultural and Personality Systems” in Roy R. Grinker (ed.): Toward a Unified Theory of Human Behavior. New York: Basic Books, Inc.: 190–199; 325–338. (1955) (FSIP) Ch. 1 The American Family: Its Relations to Personality and to the Social Structure.

370

The Appendices

(1950) The Social Environment of the Educational Process, Centennial. Washington D. C.: American Association for the Advancement of Science: 36–40 (Read at the AAAS Centennial Celebration, September, 1948). Also. (Listed under other topics): VI. A. 3. The American family The social structure of the family, or, The kinship system of the contemporary U. S. VIII. 6. Implications of the (Fiske) study of librarians

VIII. Two critical theorems: (1) THE REAL SOCIAL STRUCTURES CONSIST OF INSTITUTIONALIZED PATTERNS OF NORMATIVE CULTURE. (2) THE STABLE FEATURES OF THE REAL SOCIAL STRUCTURES ARE GUARANTEED BY MOTIVATIONAL COMPLIANCE TO A LEGITIMATE ORDER. 1. Socially sanctioned empirical and non-empirical grounds of actors’ influences and actions. (1958)

The concepts of culture and social system, Co-author with A. K. Kroeber, American Sociological Review 23, no.5 (October 1958): 582–583. (SS) (1951) Ch. VIII. Belief Systems and the Social System: The Problem of the “Role of Ideas”. Existential Empirical Beliefs and the Social System 1. The Institutionalization of Scientific Investigation 2. The Institutionalization of Applied Science 3. The Institutionalization of Ideologies The Relation of Non-Empirical Beliefs to the Social System Religious Belief Systems The Independence and Interdependence of Belief Systems and Value-Orientations

371

The Appendices

(1938) #* The Role of Ideas in Social Action, ASR 3: 652–664. (Written for a meeting on the problem of ideology at the American Sociological Society’s annual meeting. Atlantic City, N. J. December, 1937). Also. (Listed under other topics): III. A tentative outline of American values IX. The link between character and society II. The Structure of Social Action, relevant part Ch. II, Chs. 10, 11, 13, 14, 15 2. Expressive symbolism (1953) (WP) Ch. 2 The Theory of Symbolism in Relation to Action. (1951) (SS) Ch. IX Expressive Symbols and the Social System: The Communication of Affect Expressive Symbolism and Collectivities Role Differentiation with Respect to Expressive Symbolism The Role of the Artist Expressive Symbolism and the Reward System Also. (Listed under other topics): VIII. 5. “Consciousness and symbolic process”. “Social structure and the development of personality” 3. The concept of motivation as motivated compliance to a legitimate order (1951) (SS)

Ch. X. Social Structure and Dynamic Process: The Case of Modern Medical Practice The Functional Setting of Medical Practice and the Cultural Tradition The Social Structure The Situation of Medical Practice A. The Situation of the Patient B. The of the Physician The Functional Significance of the Institutional Pattern of Medical Practice Some Special Problems Some Theoretical Conclusions

372

(1947) #

(1949) (1940) #*

Also. (Listed under other topics)

The Appendices

“Introduction” in Max Weber: The Theory of Social and Economic Organization, translated by A. M. Henderson and Talcott Parsons. New York: Oxford University Press: 3–86. The Rise and Decline of Economic Man, Journal of General Education 4: 47–53. Motivation and Economic Activities, Canadian Journal of Economics and Political Science 6: 187–203 (Originally given as a public lecture at the University of Toronto and also published in Essays in Sociology, edited by C. W. M. Hart, and in Human Relations in Administration: The Sociology of Organization, edited by Robert Dubin, 1951.) II. The Structure of Social Action. Relevant parts of Ch. 4 and Chs. 9, 14, 15, 17 IX. The link between character and society

4. Socialization of motivated compliance to a legitimate order (1955) (FSIP) Ch. 2 Family Structure and the Socialization of the Child (1955) (FSIP) Ch. 4 The Mechanisms of Personality Functioning with Special Reference to Socialization (1954) The Incest Taboo in Relation to Social Structure and the Socialization of the Child, British Journal of Sociology v. V, No. 2 (June 1954): 101–117. (SS) (1951) The Learning of Social Role-Expectations and the Mechanisms of Socialization of Motivation The Socialization of the Child and the Internalization of Value-Orientations Basic Personality Structure: Modal Clustering and Diversity The Situational Role-Specification of Orientations An Example: The “Profit Motive” 5. The individual’s actions as constituent features of the social system Social Structure and the Development of Personality, Psychiatry, November 1958 (1955) (FSIP) Ch. 3 The Organization of Personality as a System of Action (1954) Consciousness and Symbolic Processes in H. A. Abramson (ed.): Problems in Consciousness. New York: MacMillan Company: 67–102. (1958)

373

The Appendices

(1954) (1954)

(1953)

(1950)

Psychology and Sociology, John P. Gillen (ed.): For a Science of Social Man. New York: MacMillan Company: 67–102. The Father Symbol: An Appraisal in the Light of Psychoanalytic and Sociological Theory, Bryson, Finkelstein, MacIver, and McKeen (eds.): Symbols and Values: An Initial Study, 13th Symposium of the Conference on Science, Philosophy, and Religion. New York: Harper and Bros, 1954: 523–544 (The substance of this paper was read at the meeting of the American Psychological Association in September 1952 at Washington D. C.) Psychoanalysis and Social Science with Special Reference to the Oedipus Problem, Franz Alexander and Helen Ross (eds.): Twenty Years of Psychoanalysis. New York: W. W. Norton and Co., Inc., 1953:186–215. (The substance of this paper was read at the Twentieth Anniversary Celebration of the Institute for Psychoanalysis in Chicago, in October, 1952). The Superego and the Theory of Social Systems, Psychiatry: 15–25 (The substance of this paper was read at the meeting of the Psychoanalytic section of the American Psychiatric Association, May 1951 in Cincinnati. Reprinted in Bales and Shils, Working Papers.) Psychoanalysis and Social Structure, The Psychoanalytic Quarterly 19: 371–384 (The substance of this paper was presented at the meeting of the American Psychoanalytic Association, Washington D. C. May 1948).

6. Structured strain (To be Pub.)

(1955)

Implications of the Study (Comment on Marjorie Fiske, Book Selection and Retention in California Public School Libraries) “McCarthyism” and American Social Tension. A Sociologist’s View, Yale Review (Winter, 1955): 226–245. Reprinted under the title “Social Strains in America” in Daniel Bell (ed.): The New American Right. New York: Criterion Books.

374

(1947) #*

(1945)

(1942)

(1942) *

Also. (Listed under other topics)

The Appendices

Certain Primary Sources and Patterns of Aggression in the Social Structure of the Western World, Psychiatry 10: 167–181. Prepared for the Conference on Science, Philosophy, and Religion, at its September 1946 meeting in Chicago, Ill., and also published in the volume published by the conference. Racial and Religious Difference as Factors in Group Tensions, Finkelstein, Louis, etc. (eds.): Unity and Difference in the Modern World, A Symposium. New York, The Conference on Science, Philosophy, and Religion, in their Relation to the Democratic Way of Life, Inc., 1945. The Sociology of Modern Anti-Semitism, Graeber, J and Britt, Steuart Henderson (eds.): Jews in a Gentile World. New York: MacMillan, 1942: (x and 436 pp.); 101–122. Some Sociological Aspects of the Fascist Movements, Social Forces 21: 138–147 (Written as the Presidential Address to the Eastern Sociological Society at its 1942 meeting). VI. A.3. The social structure of the family VI. C.1. Age and sex in the social structure of the United States

7. Structured deviance and its controls (1958) The Definitions of Health and Illness in the Light of American Values and Social Structure (In volume to be edited by Gartley Jaco. Glencoe, IL: The Free Press.) (1953) Illness, Therapy, and the Modern Urban American Family (coauthor with Renee Fox), Journal of Social Issues 8: 31–44. (1951) Illness and the Role of the Physician: A Sociological Perspective, American Journal of Orthopsychiatry 21: 452–460 (Presented at the 1951 Annual Meeting of the American Orthopsychiatric Association in Detroit. Reprinted in Kluckhon and Murray, 2nd edition).

The Appendices

375

(SS) (1951) VII. Deviant Behavior and the Mechanisms of Social Control Interaction and the Genesis of Deviant Motivation The Directions of Deviant Orientation Some Further Situational Aspects of the Genesis and Structuring of Deviance Role Conflict and the Genesis of Deviance The Social Structure of Deviant Behavior Tendencies The Mechanisms of Social Control Appendix (1942) *# Propaganda and Social Control, Psychiatry 5: 551–572 Also. (Listed under other topics) II. The Structure of Social Action, Ch. 10 8. Formal control structures in highly differentiated social systems (1958)

Some Trends of Change in American Society: Their Bearing on Medical Education. Journal of the American Medical Association 167, no. 1. (1958) * A Sociologist Looks at the Legal Profession, Conference on the Profession of Law and Legal Education, Dec.4, 1952, Conference Series Number II, The Law School, University of Chicago: 49–63. (This paper was presented at the first symposium on the occasion of the Fiftieth Anniversary Celebration of the University of Chicago Law School, December, 1952). (1939) #* The Professions and Social Structure, Social Forces 17: 457–467 (Written to be read at the Annual Meeting of the American Sociological Society in Detroit, December, 1938). (1937) Education and the Professions, International Journal of Ethics 47: 365–369.

376

The Appendices

9. The problem of rationality in the social structuring of stable social systems of action (1948) “Max Weber’s Sociological Analysis of Capitalism and Modern Institutions” in Harry Elmer Barnes (ed.): An Introduction to the History of Sociology. Chicago: University of Chicago Press: 287–308. (1947) Science Legislation and the Social Sciences, Political Science Quarterly v. LXII, no. 2 (June 1947); Bulletin of Atomic Scientists (January 1947). (1946) The Science Legislation and the Role of the Social Sciences, ASR 11: 653–666. Also. (Listed under other topics) II. The Structure of Social Action, Chs. 5, 6

IX. Problems and features of structural change (To be Pub.)

The Link Between Character and Society, Coauthor with Winston White (discussion of Riesman). (To be Pub.) Some Reflections on the Institutional Framework of Economic Development (In Problems of Economic Development: A Symposium, Hebrew University, edited by A. Bonne). (1956) Economy and Society, Coauthor with Neil J. Smelser. The Free Press. (1956) A Sociology Model for Economic Development (Coauthor with Neil J. Smelser) Exploration in Entrepreneurial History. Harvard University. (1955) (FSIP) Ch. 7, Conclusion: Levels of Cultural Creativity Generality and the Process on Differentiation.

The Appendices

377

(SS) (1951) XI. The Process of Change of Social Systems The Problem of the Theory of Change The General Nature of Change in Social Systems The Direction of Change in Social Systems Some Specific Examples of Processes of Social Change 1. Institutional Rationalization and “Cultural Lag” 2. The Ascendancy of the Charismatic Revolutionary Movement 3. The Adaptive Transformation of a Revolutionary Movement (1945) #* The Problem of Controlled Institutional Change: An Essay on Applied Social Science, Psychiatry 8: 79–101 (Prepared as an appendix to the report of the Conference on Germany after the War) Also. (Listed under other topics): VI. A. 3. The American Family VI. B. 2. Max Weber and the Contemporary Political Crises

X. General theory and theorizing: the nature of systematic, formal analysis of social actions (1953) (1950) *

Some Comments on the State of the General Theory of Action The Prospects of Sociological Theory, ASR 15: 3–16 (Presidential address read before the meeting of the American Sociological Society in New York City, December 1949) (1948) Sociology, 1941–46 (Coauthor: Bernard Barber), American Journal of Sociology 53: 245–257. (1948) # The Position of Sociological Theory, ASR 13: 156–171 (Paper read before the annual meeting of the American Sociological Society, New York City, December, 1947) (1945) #* The Present Position and Prospects of Systematic Theory. Gurvitch, Georges and Moore, Wilbert E. (eds.): Twentieth Century Sociology, A Symposium. New York: Philosophical Library, 1945.

378

The Appendices

(1938) The Role of Theory in Social Research, ASR 3: 13–20 (An address delivered before the Annual Institute of the Society for Social Research at the University of Chicago, summer, 1937) (1951) Graduate Training in Social Relations at Harvard, Journal of General Education 5: 149–157. Also. (Listed under other topics): II. The Structure of Social Action, Chs. 1, 19

XI. The place of sociological theory among the analytic sciences of action (SS) (1951) XII. Conclusion: The Place of Sociological Theory Among the Analytical Science of Action The Place of Social Systems in the General Theory of Action The Theory of Action and the Natural Sciences The Classification of the Sciences of Action (1955) Die Stellung der Soziologie innerhalb der Sozialwissenschaften, in Wilhelm Bernsdorf and Gottfried Eisermonn (eds.): Die Einhert der Sozialwissenschaften. Stuttgart: Euke. Also. (Listed under other topics) II. The Structure of Social Action, relevant parts of Ch. 19 VII. Economy and Society AR 1–29: 306–309

XII. Parsons’ assessments of other contemporary theorists’ solutions to selected topics in the problem of social order using his solutions as the interpretive position (1949) * Social Classes and Class Conflict in the Light of Recent Sociological Theory, American Economic 39: 16–26. (Read at meeting of the American Economic Association in December 1948) (1957) Malinowski and the Theory of Social Systems. Raymond Firth (ed.): Man and Culture. London: Routledge and Kegan Paul.

The Appendices

379

(1942) Review of Stuart C. Dodd’s Dimensions of Society. New York: The MacMillan Company, 1942. American Sociological Review 7: 709–714. Also. (Listed under other topics): IX. The link between character and society (Riesman) VI. B.2. The distribution of power in American Society (C. W. Mills) VIII. 6. Implications of the study (of librarians) (Fiske) Translations Max Weber. The Protestant Ethic and the Spirit of Capitalism. London and New York: Allen and Unwin, and Sbriners: xi. 292 pp. (1947) Max Weber. The Theory of Social and Economic Organization, Parsons, Talcot editor and translator with Henderson, A. M. Oxford University Press, 1947. # Introduction by Talcott Parsons. Reprinted by the Free Press. Books (1937) The Structure of Social Action. New York: McGraw-Hill, 1937 (1949) Essays in Sociological Theory, Pure and Applied. Glencoe, IL: The Free Press, 1949: xiii and 366 pp. (1951) Toward a General Theory of Action, edited by Talcott Parsons and Edward A. Shils. Harvard University Press (1951) The Social System. Glencoe, IL: The Free Press, 1951: xii and 575 pp. (1953) Working Papers in the Theory of Action. In collaboration with Robert F. Bales and Edward A. Shils. Glencoe, IL: The Free Press, 1953: 269 pp. (1954) Essays in Sociological Theory (revised edition). Glencoe, IL: The Free Press. (1955) Family, Socialization and Interaction Processes (With Robert F. Bales, James Olds, Morris Zeldith, and Philip E. Slater). Glencoe, IL: The Free Press: xi and 422 pp.

380

The Appendices

Economy and Society. Parsons, Talcott and Neil Smelser. The Free Press. (To be Pub.) The Sources of Sociological Theory (Coeditor with Edward A. Shils, Kaspar D. Naegele, and Jesse R. Pitts, 2 vols. Glencoe, IL: The Free Press).

(1956)

Biographical Notes (1949)

(Essays)A Biographical Sketch of Talcott Parsons, by Bernard Barber: 349–352. (Winter 1959) A Short Account of My Intellectual Development, Alpha Kappa Deltan. Department of Anthropology and Sociology University of California, Los Angeles Sociology 251: Selected Topics in the Problem of Social Order: Talcott Parsons Instructor: Harold Garfinkel Spring, 1959 The following list of book reviews, critiques, and discussions of Parsons’ writings are taken form Richard H. Ogles, “A Complete Bibliography of Talcott Parsons and Selected Reviews and Critiques of his Work”, Alpha Kappa Deltan (Winter, 1959): 77–80.

SELECTED BOOK REVIEWS I. Reviews of the Structure of Social Action Bierstedt, Robert: Saturday Review of Literature 17 (March 12 1938): 18–19. Crawford, W. R.: Annals of the American Academy of Political and Social Science 198 (July 1938): 178–179. Kirkpatrick, Clifford: Journal of Political Economy 46 (Aug. 1938): 588–589. Simpson, George: New Republic 96 (Sept. 28 1939): 222–223. Gettys, Warner E.: “A New Approach to Social Action”, Social Forces 17 (March 1939): 425–428. Wirth, Louis: American Sociological Review 4 (June 1939): 399–404.

The Appendices

381

House, Floyd N.: American Journal of Sociology 45 (July 1939): 129–130. Sprott, W. J.H.: British Journal of Sociology 1 (Sept. 1950): 206–263. II. Reviews of Essays in Sociological Theory, Pure and Applied Davis, A. K.: Social Forces (Oct. 1949):90–92 MacRae, Donald G.: British Journal of Sociology 1 (Sept. 1950): 263–265. III. Reviews of Toward a General Theory of Action Massaro, Nick: Sociology and Social Research 36 (March-April 1952):274. Schrag, Clarence: American Sociological Review 17 (April 1952):274. Becker, Howard: Social Forces 30 (May 1952): 464–465. Finch, Henry A.: Annals of the American Academy of Political and Social Science 281 (May 1952): 218–219. Muensterberger, Warner: Psychoanalytic Quarterly 22 (Jan. 1953): 120–122. Smith, M. Brewster: Journal of Abnormal and Social Psychology 48 (April 1953): 315–318. Beals, Ralph L: American Anthropologist 55 (Aug. 1953): 422–423. IV. Reviews of the Social System Hinkle, Roscoe C. Jr.: Annals of the American Academy of Political and Social Science 280 (March 1952): 205–206. Mercer, Blaine E.: Sociology and Social Research 36 (March-April 1952): 264–265. Becker, Howard: Social Forces 30 (May 1952): 463–464. Clark, S. D.: American Journal of Sociology 58 (July 1952):103–104. Riesman, David: “The Fitness of the Social System”, Psychiatry 15 (Nov. 1952): 478–481. Faris, Ellsworth: American Sociological Review 18 (Feb. 1953): 103–106. V. Review of Working Papers in the Theory of Action Swanson, G. E.: American Sociological Review 19 (Feb. 1954): 95–97. VI. Review of Working Papers Demerath, Nicholas J.: Social Forces 33 (Dec. 1954): 193. Bierstadt, Robert: American Sociological Review 20 (Feb. 1955): 124–125. Schmidl, Fritz: Psychoanalytic Quarterly 24 (April 1955): 306–307.

382

The Appendices

VII. Reviews of Family, Socialization and Interaction Process Opler, Marvin K.: Social Research 22 (Winter 1955): 487–491. Neumeyer, M. H.: Sociology and Social Research 40 (Nov-Dec. 1955): 142– 143. Bernard, Jessie: Social Forces 34 (Dec. 1955): 181. Christensen, Harold T.: American Sociological Review 21 (Feb. 1956): 96–97. Blau, Peter H.: American Journal of Sociology 61 (March 1956): 488–489. Marshall, T. H.: British Journal of Sociology 7 (March 1956): 67–68. Spindler, George D.: American Anthropologist 58 (Dec. 1956): 1144–1146. Blank, H. Robert: Psychoanalytic Quarterly 26 (Jan. 1957): 125–127. VIII. Reviews of Economy and Society Bierstedt, Robert: American Sociological Review 22 (June 1957): 345. Kieznets, Simon: Annals of the American Academy of Political and Social Science 312 (July 1957): 175–176. Ayers, C. E.: American Economic Review 47 (Sept. 1957): 686–688. Sinclair, Peter: Social Forces 36 (Dec. 1957): 177–178. Boulding, K.E: American Journal of Sociology 63 (Jan. 1958): 427–428. Spengler, J. J.: Journal of Political Economy 66 (June 1958): 272–273.

SELECTED CRITIQUES AND DISCUSSIONS Bierstedt, Robert: “Means-End Schema in Sociological Theory”, American Sociological Review 3 (Oct. 1938): 665–671. Pinney, H.: “The Structure of Social Action”, Review of The Structure of Social Action, Ethics 50 (Jan. 1940): 164–192. Merton, Robert K.: “Discussion of ‘The Position of Sociological Theory’”, American Sociological Review 13 (April 1948): 164–168. Newcomb, T. H.: “Discussion of ‘The Position of Sociological Theory’”, American Sociological Review 13 (April 1948): 168–171. Francis, E. K.: “American Sociology at the Cross-Roads”, Review of Essays in Sociological Theory, Pure and Applied, The Review of Politics 12 (April 1950): 247–254. Boskoff, Alvin: “Systematic Sociology of Talcott Parsons”, Social Forces 28 (May 1950): 393–400. Crespi, I.: “Comments on ‘The Prospects of Sociological Theory’”, American Sociological Review 15 (June 1950): 432. Becker, Howard: Through Values to Social Interpretation. Durham: Duke University Press 1950.

The Appendices

383

Odum, Howard W.: American Sociology. New York: Longmans, Green and Co.1951. Sheldon, Richard C.: “Some Observations on Theory in Social Science”, in Talcott Parsons and Edward A. Shils (eds.): Toward a General Theory of Action. Cambridge: Harvard University Press 1951: 30–44. Sprott, W. J. H.: “Principia Sociologica”, Review of Toward a General Theory of Action and the Social System, British Journal of Sociology 3 (Sept. 1952): 203–221. Timasheff, N. S.: “The Basic Concepts of Sociology”, American Journal of Sociology 58 (Sept. 1952): 176–186. Levy, Marion J., Jr.: The Structure of Society. Princeton: Princeton University Press. Von Wiese, L: “Ein Neues Amerikanisches Sammelwerk”, Kölner Zeitschrift für Soziologie 5 (no. 1, 1952–1953): 87–98. Swanson, G. E.: “The Approach to a General Theory of Action”, American Sociological Review 18 (April 1953): 125–134. Moore, Barrington, Jr.: “The New Scholasticism and the Study of Politics”, World Politics 6 (Oct. 1953): 122–138. Schnepp, Gerald J.: “The Future of Sociological Theory”, Clement S. Mihanovich (ed.): Social Theorists. Milwaukee: The Bruce Publishing Company 1953: 369–420. Hield, Wayne: “The Study of Change in Social Science”, British Journal of Sociology 5 (March 1954): 1–11. McKinney, John C.: “Methodological Convergences of Mead, Lundberg and Parsons”, American Journal of Sociology 59 (May 1954): 565–574. O’Dea, Thomas F.: “The Sociology of Religion”, The American Catholic Sociological Review 15 (June 1954): 73–103. Hinkle, Roscoe C., Jr. and Gisela Hinkle: The Development of Modern Sociology: Its Nature and Growth in the United States. Garden City: Double Day and Company, Inc. 1954: 64–66. Lindsey, Gardner (ed.): Handbook of Social Psychology, Vol. I. and II. Cambridge: Addison-Wesley Publishing Company, Inc. 1954. Sprott, W. J. H.: Science and Social Action. London: Watts and Co. 1954. Laulicht, J.: “Role Conflict, the Pattern Variable Theory, and Scalogram Analysis”, Social Forces 33 (March 1955): 250–254. Sjoberg, Gideon: “The Comparative Method in the Social Sciences”, Philosophy of Science 22 (April 1955): 106–117. Lundberg, G. A.: “Natural Science Trend in Sociology”, American Journal of Sociology 61 (Nov. 1955): 191–202.

384

The Appendices

Dahrendorf, Ralf.: “Struktur und Funktion: Talcott Parsons und die Entwicklung der Soziologischen Theorie”, Kölner Zeitschrift für Soziologie und Sozialpsychologie 7 (no. 4, 1955): 491–519. Gurvitch, G.: “Le concept de structure sociale”, Cahiers Internationaux de Sociologie 19 (1955): 21–31. Timasheff, Nicholas S.: Sociological Theory: Its Nature and Growth. Garden City: Doubleday 1955: 238–244. Foote, Nelson N.: “Parsons’ Theory of Family Process: Family, Socialization, and Interaction Process”. Review of Family, Socialization, and Interaction Process, Sociometry 19 (March 1956): 40–46. Barber, Bernard: “Structural-Functional Analysis: Some Problems and Misunderstandings”, American Sociological Review 21 (April 1956): 129–135. Lockwood, David: “Some Remarks on ‘The Social System’”. Review of The Social System, British Journal of Sociology 7 (June 1956): 134–146. Lundberg, George A.: “Some Convergences in Sociological Theory”, American Journal of Sociology 62 (July 1956): 21–27. Grinker, Roy R. (ed.): Toward a Unified Theory of Human Behavior. New York: Basic Books, Inc. 1956. Olds, James: The Growth and Structure of Motives. Glencoe, IL: The Free Press 1956. Firth, Raymond: “Function”, William L. Thomas, Jr. (ed.): Current Anthropology. Chicago: The University of Chicago Press. Gouldner, Alvin W.: “Some Observations on Systematic Theory, 1945–1955”, Hans L. Zetterberg (ed.): Sociology in the United States. Paris: United Nations Educational, Scientific, and Cultural Organization, 1965: 34–42. Sorokin, Pitirim A.: Fads and Foibles in Modern Sociology and Related Sciences. Chicago: Henry Regnery Company 1956. Merton, Robert K.: “Role-set: Problems in Sociological Theory”, British Journal of Sociology 8 (June 1957): 106–120. Baldamus, W.: “Sociological Theory of Economic Administration”, British Journal of Sociology 8 (Sept. 1957): 256–262. Becker, Howard and Alvin Boskoff (eds.): Modern Sociological Theory in Continuity and Change. New York: The Dryden Press 1957, esp.: 106–131; 248–258. Merton, Robert K.: Social Theory and Social Structure (rev. ed.). Glencoe: The Free Press. Brennan, M. J.: “Economics and the Theory of the Social System”. Review of Economy and Society, American Journal of Economics 17 (Jan. 1958): 113–122.

The Appendices

385

Kolb, William L.: “The Place of Values in Urban Social Theory: The Clarification of a Theoretical Issue”, Alpha Kappa Delton 28 (Winter 1958): 37–41.

BIBLIOGRAPHY – The critical materials of Talcott Parsons Barnes, Harry E. and Becker, Howard: New preface, Social Thoughts from Lore to Science, 1952. The authors criticize some contemporary sociological writers for not giving “credit where credit is due.” According to Becker (below), this means Parsons.

Becker, Howard: (Review) of Toward a General Theory of Action, Social Forces 30 (May 1952): 463. In a generally unfavorable review, Becker criticizes particularly what he calls “thinking structure and function before process.”

Blumer, Herbert: “What is Wrong with Social Theory”, ASR (Feb. 1954): 3. This is a general discussion in which Parsons is not mentioned specifically, however Blumer’s remarks are apropos. He mentions the difficulty of bringing social theory into a close and self-correcting relation with its empirical world so that its proposals about that world can be tested, refined and enriched.

Boskoff, Alvin: “The Systematic Sociology of Talcott Parsons”, Social Forces 28 (May 1950): 393–400. Presents condensed version of The Social System, little critical comment but excellent presentation of Parsonian framework.

Clark, S. D.: (Review) The Social System, AJS 58:103. Concludes that Parsons’ is not a general theory at all, but an explanation of the relation of theory of personality formation on one side and culture on the other. As such, he thinks it may have merit.

Faris, Ellsworth: (Review) of The Social System, ASR 17 (April 1952): 247. Assails Parsons for his cumbersome terminology and his failure to give credit to other theorists. See also comments by Theodore Abel and John P. Spiegel and the rejoinder by Faris in ASR, (1952): 322–3.

386

The Appendices

Finch, H. A.: (Review) Toward a General Theory of Action, Annals American Academy (May 1952): 218. Hacker, Helen Meyer: (Letter to the Editor) Arnold Rose’s “A Deductive IdealType Method”, AJS 56: 354–356. Believes pattern variables are too general to dictate prediction; she criticizes Parsons’ social system as circular, obvious, and non-verifiable.

House, Floyd: (review) Structure of Social Action, AJS 45:129. Regards it as an important contribution, but suggests it could be set forth in 40,000 well-chosen words. A generally favorable review.

MacRae, Donald: (review) Essays, British Journal of Sociology 1: 264–265. Cites confusion in meaning of “structural functional system” and questions the value of general theory, but believes Parsons’ work is important.

Parsons, Talcott: “The Position of Sociological Theory”, ASR 13: 155–165. Parsons’ view, which may be contrasted with Blumer (above) and Merton. See also discussion by Merton (same issue), who takes strong exception to Parsons’ preoccupation with abstract theory, believes that middle-range theories occupy our time. See also discussion by Theodore Newcomb (same issue), who criticizes specifically Parsons’ distinction between motivated roles which are conforming and those which are not; says the process is the same.

Parsons, Talcott: “Some Comments on the State of the General Theory of Action”, ASR (Dec. 1953): 618. Parsons evaluates Parsons. Generally favorable critique. Article includes answers to Smith and Swanson, indicates latest revisions of the pattern variables, and gives hint of what is to come.

Riesman, David: (review) Social System, Psychiatry 15 (1952): 478–481. Unqualified approval of Parsons’ aims and his methods. Much the same points are raised as in Swanson and Sprott, but Riesman sees the issues in a different light.

Sheldon, Richard C.: “Some Observations on Theory in Social Science”, chapter in Towards a General Theory of Action. The Free Press 1952. Cautions against theorizing at a level too far removed from empirical evidence; theory may integrate, not subjugate facts, he points out.

The Appendices

387

Schrag, Clarence: (review), Social System, ASR 17: 247–249. Believes fundamental overhauling needed before theory can yield testable propositions. Generally unfavorable review.

Smith, Brewster: (review) Toward a General Theory of Action, Journal of Abnormal and Social Psychology (April 1953): 315–318. Presents detailed, incisive criticism of the pattern variables, concludes there is danger in degree to which it carries conceptual elaboration in the absence of data. Generally unfavorable review.

Sprott, W. J. H.: (review) Structure of Social Action, British Journal of Sociology 1: 260. Dismisses Parsons somewhat lightly. Paraphrases Parsons thusly: “Normative beliefs are an essential ingredient of social conduct; look at the works of Pareto, Durkheim, and Weber and then you will see.

Sprott, W. J. H.: “Principia Sociologica”, British Journal of Sociology 3: 203– 221. Sprott voices the same doubts as does Smith regarding the pattern variable – do they form an exhaustive category? Sprott concludes also, as does Smith, that Parsons’ schema must wait to be tested before its fruitfulness can be attested.

Swanson, Guy E.: “The Approach to a General Theory of Action by Parsons and Shils”, ASR (April 1953): 2. Swanson, Guy E.: (review) Working Papers, ASR (Feb. 1954): 95–97. Criticizes pattern variables as neither helping us relate roles to each other in predicted chains of behavior nor permitting us to make general, but refined, descriptions of differences in their patterns. Swanson tends to discount Parsons’ and Bales’ claim that pattern variables are in small group analysis – “material, true, if at all, by definition.”

388

The Appendices

BIBLIOGRPAHY – Reports of Studies Using Parsonian Framework Bales, Robert F.: “The Dimensions of Action-Space”, chapter 3 in Working Papers in the Theory of Action. Glencoe, IL: The Free Press 1953: 63. Shows that Bales’ four steps in action process (1) latent pre-dispositions to act, (2) over instrumental activity, (3) consummatory goal gratification, and (4) re-integration of the actor and Parsons’ pattern variables are different ways of stating the same thing. The flow of motivation through the stages of action as now conceived have four combinations: the object-oriented, achievement, particularism, ascription, and universalism and the attitudinal oriented-specificity, affectivity, diffuseness, and affective neutrality.

Bernard, Jesse: “Sociology of Conflict”, AJS (March 1954): 411–424. Suggests that the theory of games is in line with Parsons’ theory: that is, that it applies only to rational behavior, that given the goals people have, they operate in terms of these rather than anything else. The author also feels this is a serious limitation on the theory of games, and by implication, on Parsonian theory.

Getzels, J. W. and Guba, E. G.: “Role, Role Conflict and Effectiveness: An Empirical Study”, ASR (April 1954): 164. The authors spell out Parsonian view of role-conflict – an excellent in pattern variable terms. Getzels and Guba sought to investigate role-conflict in real-life situation rather than with “made-up” questions (a la Stouffer and Toby) in investigating conflict between teacher and military in officers at the Air Command and Staff School at Air University, Maxwell Air Force Base. Found that empirical data were in line with the theoretical expectations: (1) that the intensity of involvement in role conflict varied as a function of certain individual and attitudinal characteristics and (2) that there was a systematic relationship between the intensity of the involvement in role conflict and role effectiveness. Found theory lacking, however, in that although it indicated relationships existed, didn’t provide understanding of the mechanisms involved in the relationship. Attack emphasis in the study of role effectiveness as congruence of personality needs and role expectations as a necessary condition, but not a sufficient one. Conclude that sufficient conditions (here they leave Parsons) are choice of a major role that is the more legitimate one.

The Appendices

389

Kluckhon, Florence: “Dominant and Substitute Profiles of Cultural Orientations: Their Significance for the Analysis of Social Stratification”, Social Forces (May 1950): 376–393. Uses a schema of classification of cultural value-orientations which Parsons says are compatible to pattern variables. (See Parsons, “Some Comments on General Theory of Action”, ASR (Dec. 1953): 317.)

Olds, James: Psychological Papers in the Theory of Action. Free Press, any day now! Parsons (“Some Comments...”) claims that Olds will show that even the most elementary levels of behavior psychology can advantageously be treated in terms of the conception that such behavior consists in “systems of action.” Olds will show further “how these elementary units of behavior come to be organized in larger systems which approach much more closely to what is usually treated as ‘personality level.’”

Parsons, Talcott: “The Case of Modern Medical Practice”, The Social System. Glencoe, IL: Free Press, Ch. X. Parsons analyzes the medical practice in terms of the value-alternatives which characterize roles and actors.

Parsons, Talcott: “Family, Socialization, and the Interaction Process”. Free Press, Today Parsons (“Some Comments...”) indicates that it “has proved possible (here) to analyze the process of socialization as a process of social interaction in terms of relationship between social structure and personality... and through that analysis to carry the use of pattern variables in the formulation of the structure of personality as a system very much farther than was done before... also true of the classification of the mechanisms of personality process which have been arranged in terms of their relations to the pattern variables.“

Parsons, Talcott: “Propaganda and Social Control”, Psychiatry 5 (1942): 551–572 Here a much younger Parsons makes use of a much less developed pattern variable schema in the analysis of propaganda ad social control. Indicates that pattern variables may be more firmly grounded in empirical observation that is sometimes assumed by critical writers.

390

The Appendices

Parsons, Talcott: in collaboration with Samuel A. Stouffer and Florence Kluckhon, as yet untitled, or title unknown. Parsons (“Some Comments”) refers to a long-range systematic attempt to use and develop his theory in predicting form the characteristics of families, peer groups and schools, and the place of the boy in them, what place in the occupational system he will come to occupy. In this connection the adult occupational system is to be treated as system of reference groups toward which a boy may be thought of as coming to be selectively oriented.

Pieris, Ralph: “Ideological Momentum and Social Equilibrium”, AJS 57: 339–346. Follows much the same analytical framework as does Parsons, although his analysis is not specifically in Parsonian terms. Deals with manifest and latent functions.

Riley, Matilda White, Seeman, Melvin: “Role Conflict and Ambivalence in Leadership”, ASR (August 1953): 373–380. Uses Parsons’ definition of role conflict as referring only to situations in which the observer notes what appear to be conflicting sets of expectations, i. e., to potential sources of difficulty for the actor. Seeman lists “four dimensions of role conflict”, which he says are an exhaustive list comparable to the pattern variables. His four are: (1) status dimension – the conflict between success ideology and equality, (2) authority dimension – values of independence and dependence, (3) institutional dimension – choice between universality as against particularist criteria for social action, and (4) means-ends dimensions – conflict between the job done (expediency) and morality.

Stouffer, Samuel A., and Toby, Jackson: “Role Conflict and Personality”, AJS 16: 395–405. An attempt to move from Parsons’ highly “abstract concepts in social science to empirical generalizations.” On the basis of this attempt, Stouffer anticipates not “quick and easy progress” in such movement, but the “years of patient work which lie ahead.” The study used forced-type of question to determine whether subjects were particularistically or universalistically oriented. The construction of a scale suggested which would measure the latent tendency to be a loyal friend even at the cost of other principles. The jump from the empirical part of the study to the more abstract level is never made, however. The authors cite the importance of a link between theories of institutionalization and theories of personality, but just what link is or how it is to be arrived at is not made explicit.

The Appendices

391

Toby, Jackson: “Some Variables in Role Conflict Analysis”, SF 30: 323–327. Uses Parsons’ definition of role, but further analyzes role conflict in terms of alternatives. This further analysis bears no apparent relationship to pattern variables or Parsons’ other schema.

Vogt, Even Z., and O’Dea, Thomas F.: “A Comparative Study of the Role of Values in Social Action in Two Southwestern Communities”, ASR (Dec. 1953): 645. One of the studies of values in five cultures projects of the laboratory of social relations at Harvard. Hypothesis is that “value-orientations play important part in shaping of social institutions” and “influencing the forms of observed social action.” Study concerns differences in Mormon community of Homestead and that of Rimrock, a community made up largely of dust-bowl immigrants. Differences are due largely to different values, rather than to different geographical or historical factors, the authors conclude.

Theodorson, George A.: “Acceptance of Industrialization and Its Attendance Consequences for the Social Patterns of Non-Western Societies”, ASR 18 (October 1953): 477–483. Analysis of changes which accompany industrialization in terms of the pattern variables.

A BIBLIOGRAPHY OF THE CRITICAL MATERIALS of Talcott Parsons Barnes, Harry E.: Introduction to the History of Sociology: 287–308. Becker, Howard, Review: “Toward a General Theory of Action” and “The Social System”, Social Forces 30: 463. Boskoff, Alvin: “The Systematic Sociology of Talcott Parsons”, SF 28: 393. Clark, S. D., Review: “The Social System”, AJS 58: 103. Faris, Ellsworth, Review: “The Social System”, ASR (Feb. 1953): 103. Comments by Theodore Abel and John P. Spiegel, with rejoinder by Faris, ASR (June 1953): 322. Finch, H. R., Review: “The Social System”, American Academy of Polit. & Social Science Annals (May 1952): 218. House, Floyd, Review: “Structure of Social Action”, AJS 45: 129. Parsons, Talcott: “The Position of Sociological Theory”, ASR 13: 155. Discussion: Robert Merton, Theodore Newcomb.

392

The Appendices

Parsons, Talcott: “Some Comments on the State of the General Theory of Action”, ASR (Dec. 1953): 618. Sheldon, Richard C.: “Some Observations on Theory in Social Science”, a chapter in Toward a General Theory. Smith, Brewster: Review, “The Social System”, Journal of Abnormal and Social Psychology 48 (April 1953): 315. Sprott, W. J.H.: “Principia Sociologica”, III, British Journal of Sociology 3 (Sept. 1952). Swanson, G. E.: “The approach to a General Theory of Action by Talcott Parsons and Shils”, ASR (April 1953), lead article. Swanson, G. E.: Review: “Working Papers”, ASR 19 (Feb. 1954): 95. MacRae: British Journal of Sociology, Vol. 1, 1950. Riesman: Psychiatry, Vol. 15, 1952. Schrag, C.: ASR, Vol. 17, 1952. Sprott, W. J.H: British Journal of Sociology, Vol. 1, 1950.

393

The Appendices

Appendix #4

Fig. 16: Anonymous letter from student