Wilfrid Sellars, Idealism, and Realism: Understanding Psychological Nominalism 9781474238939, 9781474238960, 9781474238953

Table of contents :
Cover
Half-title
Title
Copyright
Dedication
Contents
Abbreviations of Sellars’ Texts
Introduction: Psychological Nominalism and German Idealism
Part 1: Psychological Nominalism and Realism
1. “Psychological Nominalism” and the Given, from Abstract Entities to Animal Minds
2. Hegel and Sellars’ “Myth of Jones”: Can Sellars Have More in Common With Hegel than Rorty and Brandom Suggest?
3. The Metaphysics of Sensation: Psychological Nominalism and the Reality of Consciousness
4. Language, Norms, and Linguistic Norms
Part 2: Psychological Nominalism and Idealism
5. On the Pittsburgh School, Kant, Hegel, and Realism
6. Reading Sellars’ “Philosophy and the Scientific Image of Man,” with Robert Brandom at One’s Side
7. A Kantian Critique of Sellars’ Transcendental Realism
8. Psychological Nominalism and Conceptual Relativism: An Idealist’s Take
Index

Citation preview

Wilfrid Sellars, Idealism, and Realism

Also available from Bloomsbury A Critical Introduction to Scientific Realism, Paul Dicken Evidentialism and the Will to Believe, Scott F. Aikin Humanism and Embodiment, Susan E. Babbitt

Wilfrid Sellars, Idealism, and Realism Understanding Psychological Nominalism Edited by Patrick J. Reider

BLOOMSBURY ACADEMIC Bloomsbury Publishing Plc 50 Bedford Square, London, WC1B 3DP, UK 1385 Broadway, New York, NY 10018, USA BLOOMSBURY, BLOOMSBURY ACADEMIC and the Diana logo are trademarks of Bloomsbury Publishing Plc First published in Great Britain 2017 Paperback edition first published 2018 © Patrick J. Reider and Contributors, 2017 Patrick J. Redier has asserted his right under the Copyright, Designs and Patents Act, 1988, to be identified as the Editor of this work. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. Bloomsbury Publishing Plc does not have any control over, or responsibility for, any third-party websites referred to or in this book. All internet addresses given in this book were correct at the time of going to press. The author and publisher regret any inconvenience caused if addresses have changed or sites have ceased to exist, but can accept no responsibility for any such changes. No responsibility for loss caused to any individual or organization acting on or refraining from action as a result of the material in this publication can be accepted by Bloomsbury or the editor. A catalogue record for this book is available from the British Library. ISBN: HB: 978-1-4742-3893-9 PB: 978-1-3500-7008-0 ePDF: 978-1-4742-3895-3 ePub: 978-1-4742-3894-6 Library of Congress Cataloging-in-Publication Data Names: Reider, Patrick J., 1974– editor. Title: Wilfrid Sellars, idealism and realism : understanding psychological nominalism / edited by Patrick J. Reider. Description: New York : Bloomsbury, 2016. | Includes index. Identifiers: LCCN 2016029679 (print) | LCCN 2016034170 (ebook) | ISBN 9781474238939 (hardback) | ISBN 9781474238953 (epdf) | ISBN 9781474238946 (epub) Subjects: LCSH: Sellars, Wilfrid. | Nominalism. | Realism. | Idealism. Classification: LCC B945.S444 W55 2016 (print) | LCC B945.S444 (ebook) | DDC 191—dc23 LC record available at https://lccn.loc.gov/2016029679 Typeset by RefineCatch Limited, Bungay, Suffolk To find out more about our authors and books visit www.bloomsbury.com and sign up for our newsletters.

To my mom and dad and the lessons they taught me

Contents Abbreviations of Sellars’ Texts

Introduction: Psychological Nominalism and German Idealism Patrick J. Reider, University of Pittsburgh, USA

viii

1

Part 1  Psychological Nominalism and Realism

1

2

3

4

“Psychological Nominalism” and the Given, from Abstract Entities to Animal Minds  James R. O’Shea, University College Dublin, Ireland Hegel and Sellars’ “Myth of Jones”: Can Sellars Have More in Common With Hegel than Rorty and Brandom Suggest?  Paul Redding, University of Sydney, Australia The Metaphysics of Sensation: Psychological Nominalism and the Reality of Consciousness  Ray Brassier, American University of Beirut, Lebanon Language, Norms, and Linguistic Norms  Willem deVries, University of New Hampshire, USA

19

41

59 83

Part 2  Psychological Nominalism and Idealism

5 6

7 8

On the Pittsburgh School, Kant, Hegel, and Realism  Tom Rockmore, Peking University, China Reading Sellars’ “Philosophy and the Scientific Image of Man,” with Robert Brandom at One’s Side  Joseph Margolis, Temple University, USA A Kantian Critique of Sellars’ Transcendental Realism  Johannes Haag, Universität Potsdam, Germany Psychological Nominalism and Conceptual Relativism: An Idealist’s Take  Patrick J. Reider, University of Pittsburgh, USA

Index

103

121 149

173 195

Abbreviations of Sellars’ Texts “Being and Being Known” in Science, Perception and Reality “Behaviorism, Language and Meaning” “The Concept of Emergence” “Concepts as Involving Laws and Inconceivable without Them” “Empiricism and Abstract Entities” “Empiricism and the Philosophy of Mind” “Foundations for a Metaphysics of Pure Process” “The Role of Imagination in Kant’s Theory of Experience” In the Space of Reasons: Selected Essays of Wilfrid Sellars “Some Remarks on Kant’s Theory of Experience” “Kant’s Transcendental Idealism” “On The Logic of Complex Particulars” “Language as Thought and as Communication” The Metaphysics of Epistemology Philosophical Studies 39: 325–45 Naturalism and Ontology “Phenomenalism” “Philosophy and the Scientific Image of Man” “The Structure of Knowledge: (1) Perception; (2) Minds; (3) Epistemic   Principles” SM Science and Metaphysics: Variations on Kantian Themes SPR Science, Perception and Reality SRI “Scientific Realism or Irenic Instrumentalism: A Critique of Nagel and   Feyerabend on Theoretical Explanation” SRLG “Some Reflections on Language Games” SRPC “Some Reflections on Perceptual Consciousness” SSIS “Science, Sense Impressions, and Sensa: A Reply to Cornman” SSOP “Sensa or Sensings: Reflections on the Ontology of Perception” TC “Truth and Correspondence” TTC “Towards a Theory of the Categories” BBK BLM CE CIL EAE EPM FMPP IKTE ISR KTE KTI LCP LTC ME MEV NO PH PSMI SK

Introduction: Psychological Nominalism and German Idealism Patrick J. Reider, University of Pittsburgh (USA)

This volume is the first study of its kind to address a range of realist and idealist views inspired by psychological nominalism. Bringing together premier analytic realists and distinguished defenders of German idealism, it reveals why psychological nominalism is one of the most important theories of the mind to come out the twentieth century. Although wedded to important aspects of German idealism, Sellars’ theory is couched in bold scientific realist terms of the analytic tradition. Those who are sympathetic to German idealism find his realist’s appropriation of idealism problematic. Wilfrid Sellars, Idealism and Realism thus creates a rare venue for realists and idealists to debate the epistemic outcome of the mental processes they both claim are essential to experience. Their resulting discussion bridges the gap between analytic and continental philosophy.

Nominalism, realism, and a logical space To understand “psychological nominalism” and what is at stake in this theory, one should keep in mind two opposing views: “nominalism” and the specific form of “realism” it rejects.1 Nominalism is a countermovement against “realists,” who claim that universals exist independent of the human mind. In opposition to the realists, nominalists argue that universals are created by and inseparable from the human mind. Psychological nominalism is a sophisticated and linguistically centered version of nominalism. One important aspect of the debate between nominalists and realists concerns the source of human understanding. For instance, in the following excerpt, Michael Loux explains the manner in which the activity of sorting, classifying, and grouping entities together requires the functional use of universals. In this regard, universals play an essential role in our rational thoughts and cognitive experiences:

2

Introduction The objects we talk and think about can be classified in all kinds of ways. We sort things by color, and we have red things, yellow things, and blue things. We sort them by shape, and we have triangular things, circular things, and square things. We sort them by kind, and we have elephants, oak trees, and paramecia. The kind of classification at work in these cases is an essential component in our experience of the world. There is little, if anything, that we can think or say, little, if anything, that counts as experience that does not involve groupings of these kinds. Loux 2010, 18

Universals are essential to human experience (insofar as it is cognitive or rational), because they account for how it is possible for differing entities to be “qualified or characterized in some way by virtue of their all standing in relationship” to the same “quality or characteristic” (2010, 19).2 Sellars’ version of nominalism is a calculated rejection of the mainstream views concerning universals and their operation in language. In what follows, I provide a quick overview of historically popular views on nominalism and realism in order to historically situate Sellars’ claims and appreciate the types of views that Sellars rejects. The first view is a version of “realism” that, for the sake of convenience, I call “universal realism.”3 Universal realism asserts (or assumes) that some hidden power of the human soul enables us to non-­inferentially and immediately “see” universals. The idea driving this view is that universal commonalities shared among diverse objects, occurrences, and thoughts exist in a manner that is independent of the human mind and the individual objects that embody them. Universals, themselves, are thus taken to be individual entities that can be named. If universals exist independent of the human minds that think them, then the standard of truth, objectivity, and knowledge will reside in our ability to observe them. For example, under this view, what makes us aware of someone’s “goodness” is his/her exhibition (i.e., something we can see) of the universal quality of “goodness.” In this context, “goodness” is a universal quality that exists independent of a particular person who is good. Similarly, the universal quality of “red” is believed to exist independent of the individual stop signs, London buses, cherries, and fire hydrants that are red. The second view on universals rejected by Sellars stems from British empiricism. It denies the above view. It claims that, from a sensory particular, the human mind can immediately (i.e., non-­inferentially) “abstract” universal content from sensation. For example, John Locke claims that, from the sensation of a particular shade and tint of a red flower, the mind can subtract all of its

Introduction

3

specific or particular qualities. This operation is thought to endow the mind with an immediate (i.e., non-­inferential) recognition of general ideas, such as redness. He further claims that the “general and universal belong not to the real existence of things; but are the inventions and creatures of the understanding, made by it for its own use . . .” (Locke 1995, 330). The third view we will consider is “rationalism.” It denies the empiricist’s claim that we somehow abstract universals via our senses and argues instead that the human intellect is naturally endowed with a functional grasp of universals. As Descartes argues in his famous wax example, sensation without rational judgment fails to provide knowledge or understanding. In this regard, he claims that, through the intellect, knowledge concerning determinate qualities and characteristics of the appearing world is obtained. According to most rationalists, the mind can perform this operation, because the “natural light of reason” is designed by God to perform this function. Rationalists are typically “realists,” in the sense that they believe the universals implanted in one’s mind bear a true and accurate correspondence with outer reality. Sellars believes that all the above views regarding universals—as well as most other views of his time—make the following mistake: [W]hen we picture a child . . . learning his first language, we, of course, locate the language learner in a structured logical space in which we are at home. Thus, we conceive of him as a person . . . in a world of physical objects, colored, producing sounds, existing in Space and Time. But though it is we who are familiar with this logical space, we run the danger, if we are not careful of picturing the language learner as having ab initio [i.e., from the beginning] some degree of awareness—“pre-­analytic,” limited and fragmentary though it may be—of the same logical space. . . . In other words, unless we are careful, we can easily take for granted that the process of teaching a child to use a language is that of teaching it to discriminate elements within a logical space of particulars, universals, facts, etc., of which it is already indiscriminatingly aware, and to associate these discriminated elements with verbal symbols. EPM, 65

For Sellars, the tendency to assume children start out with a limited functional use of universals is a “myth.” This myth presumes that the ability to employ universals, which requires a capacity to command logical relations, is inherent to the human intellect, regardless of how universals are initially presented to the mind, i.e., regardless if one is a universal realist, British empiricist, or rationalist. As noted above, the universal realists, rationalists, and traditional empiricists all fall prey to the assumptions embedded in the above Sellarsian excerpt. This

4

Introduction

can be illustrated in their various views concerning language. For instance: (1) we see universals and then merely give them a name established by convention (i.e., universal realism); (2) after sensation occurs, one can immediately (i.e., non-­inferentially) abstract such content into universal content, which can then be given a name by convention (i.e., traditional empiricism); or (3) our intellect naturally comes endowed with an understanding of universals, which are merely given names via convention (i.e., rationalism). In the above accounts of how language is learned, our capacity to understand the logical relations between different types of universals and our capacity to make comparisons by seeing how things resemble one another are taken as non-­ problematic and inherent functions of the mind, i.e., such abilities are not learned. Hence, according to the above theories, the differences among languages are due to the differences established by the ways universals and resemblances are named according to longstanding conventions. Universal realism, rationalism, and British empiricism additionally overlook the fact that the meaning of words cannot be reduced to reference, upon which a sentence is merely a list of names. For instance, the standard model of language presumes that it primarily invokes names, so that, when a subject is named, the predicate spoken language relation to it, in effect, “names” a relation between subject and predicate. Sellars argues that the above linguistic model oversimplifies the activity of spoken and written language: Sorting expressions by their meanings is sorting them by functional classifications. Just as we might sort kitchen utensils by their functions—by the ways in which they are used—so we can sort words by their functions or roles in language. A functionalist account of meaning states that two words (whether in the same or in different languages) have the same meaning iff [i.e., a necessary and sufficient condition] they have roughly the same linguistic function. . . . The relevant role played by an expression [or a word’s function] is multidimensional, for, as he [i.e., Sellars] makes clear in “Some Reflections on Language Games” (1954), the use of the expression in language-­entry transitions (e.g., verbal response to perceptual stimuli), language-­departure transitions (e.g., actions fulfilling statements of intention), and language-­language transitions (e.g., inferences, conversational responses, and implicatures) all count in determining the expression’s functional role in a linguistic economy. deVries and Triplett 2000, 65

According to Sellars, the ability to use a language is not acquiring conventional names for things already known, but rather, it is the acquired ability to understand

Introduction

5

and employ the functional roles words play in a particular language.4 The roles of words and expressions have multiple functions as suggested above (i.e., language-­entry, language-­departure, and language-­language transitions), and thus their meaning stems from their use and application. According to Sellars, the successful employment and understanding of terms and expressions require one’s ability to navigate a virtual sea of logical possibilities that are not present to the linguistically uninitiated. Sellars illuminates the failings of traditional views on language by arguing that one’s understanding of the distinct roles of different kinds of categorical terms requires a “logical space”: What does the notion of a “logical space” here amount to? We can think of logical space as determined by the categorical structure that we use to carve up the world conceptually. There are categorical distinctions amongst objects, events, and properties; and within those categories, there are distinctions between physical objects and abstract objects, first-­order properties (properties of objects) and higher-­order properties (properties of properties). deVries and Triplett 2000, 60

The ability to competently use categories and concepts (terms Sellars prefers over that of “universals”) requires the ability to understand why something is considered an event, object, or property (though this ability need not be overtly manifested during their use). Likewise, the ability to understand why something is a mental entity or physical object requires an understanding of the relationships or differences that these two classifications delineate. Moreover, qualities of qualities require logical distinctions. For instance, “the blue tint of the motor oil’s sheen,” designates a “quality” of the oil’s “sheen.” In this example, both the “blue tint” (i.e., the higher-­order property) and “sheen” designate qualities one can associate with the oil, but the oil itself is neither “blue” nor a “sheen.” These differences among the functional use of universals belie that their applications, insofar as they can be understood, presuppose an understanding of their logical relations (i.e., a logical space of reasons).

Knowing that, know how, and conceptual holism Sellars views the ability to use the logical space of reasons as not merely a matter of “knowing that” but more fundamentally “knowing how.”“Know how” concerns our ability to successfully perform some function, while “knowing that” requires

6

Introduction

conceptual content in the propositional form of “that is a tree,” “that is a bad idea,” “that is sad,” etc. While it is true one may need to understand that “this is a tree” and the corresponding reasons why something is a tree, in order to “know that” “this is a tree,” such abilities are only half the story. One also needs the more fundamental “know how” of being able to (1) appropriately generate such thoughts in the presence of trees and (2) form and interpret such statements as “this is a tree.” Sellars shares important similarities with Gilbert Ryle, who argues that “Knowing how to apply [reasons or] maxims cannot be reduced to, or derived from, the acceptance of those or any other maxims” (2000, 32). Sellars applies this Rylean concept to language. The use of language and logic cannot rest on reasons concerning “knowing that,” which in turn rests on additional acts of reasoning (i.e.,“knowing that”) ad infinitum.5 If it did, one would have to undergo an infinite string of reasons to communicate one’s observations, which of course exceeds human mental capacities. Thus, there must be some stop to “knowing that” and some start to “knowing how,” i.e., certain types of know how must operate as the precondition for certain types of knowing that. How then can one learn a language, if we as children are not already in possession of universals and the logical relations they share? Moreover, how does one acquire the know how that enables the multi-­dimensional application of universals to particulars and their variable functions in numerous settings? According to Sellars, one acquires concepts and the ability to “know how” to apply concepts. Concepts, once acquired, enable the capacity to recognize common traits and much more: “psychological nominalism according to which all awareness of sorts, resemblances, facts, etc., in short, all awareness of abstract entities—indeed, all awareness even of particulars—is a linguistic affair” (EMP, 60).6 In this regard, psychological nominalism directly opposes universal realism, British empiricism, rationalism, and all their numerous variants. In rejecting these views and their offshoots, Sellars is not merely making a case for a particular form of nominalism or arguing that the logical space that permits a functional command of universals is learned, but he is also making an important epistemic claim. To see the epistemic issue at stake here, consider the following points. What if one or more of the following were true: (1) people directly see universals or (2) people non-­inferentially create universals from sensation? Moreover, what if, in cases 1 and 2, people possessed an inherent command of the logic of universals? The universal realist, rationalist, British empiricist, and all those who are influenced by their basic perceptual claims, argue (or assume) that some

Introduction

7

combination of the above creates a foundation for observational claims. In short, 1 or 2 could enable “primitive perceptual” experiences that are distinguishable from the potentially false “judgments” or learned responses that occur when one attempts to infer what is being perceived. Put differently, either empirical experience or experience of one’s own mental states could provide “given” content that immediately and non-­inferentially supports, justifies, or irrefutably shows one’s sincere observational assertions to be true.7 According to Sellars, this view expresses the “Myth of the Given.” To understand how language can provide us with concepts, a logical space, and by extension the ability to reason about the world and ourselves, which in turn undermines the notion of the given, we need to address why concepts are interrelated and interdependent. According to Sellars, a person cannot know one or just a few concepts, because the functional operation of any one concept is dependent on a “whole battery” of concepts and know how. For instance, Sellars argues that “the ability to recognize that x looks green presupposes the concept of being green” (EPM, 44). This ability will also require, in addition, knowing in what circumstance to view an object to ascertain its color (EPM, 44). In turn, this additional ability requires us to recognize that certain objects have perceptual characteristics, one of which is being colored. Hence, the deceptively simple observation “this is green” presupposes the ability to know under which circumstance one can make color designations, what colors are, how to discern colors, as well as an awareness that individual objects can possess properties such as being a specific color. After arguing that an observation requires many different kinds of abilities, Sellars radicalizes his claim that concepts are interrelated and interdependent: [O]ne can have the concept of green only by having a whole battery of concepts of which it is one element. It implies that while the process of acquiring the concept green may—indeed does—involve a long history of acquiring piecemeal habits of response to various objects in various circumstances, there is an important sense in which one has no concept pertaining to the observable properties of physical objects in Space and Time unless one has them all—and, indeed, as we shall see, a great deal more besides. EPM, 45

Sellars believes that concepts have meaning in light of how they relate or fail to relate to other concepts.8 Thus, characterizations, which are judgments that assert/attribute conceptual content to some entity, are meaningful only insofar as one understands how they relate to a whole web of compatible and

8

Introduction

incompatible concepts. The “great deal more” that is needed over and above concepts is (1) a logical space of reasoning, which permits one to functionally navigate a “battery of concepts” and (2) the “know how” of knowing when and in what manner one should employ them. One begins to learn a language via acquiring the right kind of habits that enable one to functionally respond to the observational conditions one is exposed to, as well as the kind of observational claims they permit. Take, for instance, the kinds of obstructions that may occur when making an empirical observation. These obstructions, whether they are a physical object in one’s line of sight, a colored light, dim lighting conditions, atmospheric conditions, etc., change when and how we ought to make observational claims. One can quickly see why Sellars wants to emphasize that learning a language, and by extension learning how to make observations, requires a longstanding acquisition of habits.9 In review, Sellars is a “conceptual holist,” meaning that he believes concepts are interrelated and cannot be employed in isolation from one another, because their employment requires a shared “logical space.” He therefore rejects the notion that we obtain concepts “piecemeal” (i.e., one at a time) when we first begin to speak. Instead, he argues that the use of many concepts needs to be habitually acquired until a gestalt-­like understanding dawns on the whole. In other words, many individual habits are acquired piecemeal, but they do not afford conceptual understanding until the appropriate linguistic habits form a complex network which contains interrelated habituated responses to both stimuli and groupings of acquired habits. The resulting gestalt-­like occurrence does not require one to obtain an instantaneous, overt, and reflective understanding of each concept that one can use. Rather, Sellars’ claim is that one obtains the ability to play the language game, i.e., one can appropriately respond to the norms by which one can functionally employ concepts, and secondly assesses (if the need arises) whether or not they are properly employed. If one cannot be aware of particulars qua particulars or of universal content, have a logical space, or know how to make observations that contain the logic of propositional and conceptual content before learning a language, and all these abilities require “know how” that stems from learning numerous interconnected habits, then there are no “givens” of intrinsic epistemic value in experience. In Sellars’ alternative to given perceptual content, one employs a complex pattern of habituated skills that culminate in an interrelated command of concepts that permit one to employ conceptual content seamlessly within one’s experiences. If the latter is true, the epistemic merit of one’s observational accounts rests on

Introduction

9

acquired abilities rather than the type of inborn capacities argued by the empiricist, rationalist, universal realist, and all those who unwittingly follow in their wake. Psychological nominalism thus provides an account of the manner in which one obtains conceptual understanding and reason and hence the capacity to know in its most robust form. It claims that the process of learning a language provides the necessary habits, through which one ultimately obtains an interrelated framework to dispositionally respond to various stimuli. This process permits one to develop what can transcend pure dispositional states, i.e., one can acquire the ability to spontaneously or freely employ concepts as is required of rationality. Since concepts and reasons require an interconnected web of relations that determine their functional use (i.e., how we employ them), psychological nominalism is also an account of how mental states relate to one another and prompt behavior. In this regard, it can be considered a form of “functionalism.” Sellars’ functionalism results in the view that language is the only means by which humans can learn the types of socially shared practices that permit rationality. Before proceeding, it is important to emphasize that psychological nominalism should not be confused with the view that there are no mental states before language acquisition or that our inborn pre-­linguistic abilities and pre-­conceptual states fail to play an essential role in our capacity to form epistemic claims. Rather, it claims that one’s mental states before the acquisition of a language are non-­conceptual and non-­rational. Similarly, one should not confuse psychological nominalism with the view that all thoughts occur as a form of linguistic imagery (i.e., imagining that one is speaking). Instead, it is the claim that learning a language provides the normative framework or form through which reasoning and understanding transpire.

Normative functionalism and psychological nominalism If one cannot observe particulars or universal content, have a logical space, or know how to make observations that contain the logic of propositional and conceptual content before learning a language, and all these abilities require “know how” that stems from learning numerous interconnected habits, then what establishes or influences the habits we learn? Well, the people in our learning environment may seem like a potential answer, but what guides their choices concerning the habits they intentionally and unintentionally instill in the members of their linguistic community? At last we come to norms.

10

Introduction

For Sellars (and the rest of the Pittsburgh School, i.e., John McDowell and Robert Brandom), psychological nominalism accounts for how humans come to be rational and form conceptual observations. This account concerns a process, which begins and is mediated by a slow and persistent acquisition of habits shaped by the norms of one’s linguistic community. For instance, Chauncey Maher writes: Crucially, according to philosophers in the Pittsburgh School, the role of assertions . . . must be understood in normative terms. To make an assertion is to adopt a normative status, a status defined in the terms of what the assertor rationally may (or may not) and should (or should not) say or do. That is a manifestation of the normative character of epistemic states or episodes expressed in Sellars’s “placing” remark. Assertions must be understood in terms of what they rationally permit or license or warrant and what they prohibit or forbid. In this way, there are norms or rules that govern the connections between assertions, perceptions, and non-­verbal behavior. While someone capable of making an assertion can violate these rules from time to time, one can be an “assertor” only if one is generally in conformity with these rules. If one never conforms with these rules, none of one’s behavior could count as asserting. 2012, 30

Maher calls the above view “normative functionalism”: “It is functionalist because it emphasizes the function of assertions. It is normative because it stresses the place of normative statuses in characterizing these functions” (Maher 2012, 30). In short, the manner in which one’s assertions or judgments function, as they concern what something is, what it appears as, or what ought to be done are shaped by the normative principles that render such assertions or judgments possible. Norms touch nearly every significant aspect of human thought. First, norms shape the kind of concepts one possesses by establishing the practices that relate conceptual content to other conceptual content. This greatly affects one’s concepts, because the manner in which concepts relate to one another determines what they mean when used, e.g. conceptual holism. Second, norms shape when and how concepts are employed, by setting overt and implicit standards under which it is proper (or improper) to offer a “verbal response to perceptual stimuli,” initiate “actions fulfilling statements of intention,” and guide the formation of “inferences, conversational responses, and implicatures” (deVries and Triplett 200, 65). This latter point leads to a third role norms play: norms function as the

Introduction

11

principles that underlie rationality in the sense that they permit, license, warrant, prohibit, and forbid. Put another way, since norms influence when and how concepts are employed, they also shape how we form judgments and link conceptual content, as it concerns our ability to reason.

The relevancy of psychological nominalism No philosophical question or approach can be truly free of competing claims concerning the manner in which human beings think, communicate, act, and know. As a result, psychological nominalism is relevant to many philosophical disciplines, because it opposes many of the traditional views concerning the manner in which the human intellect comes into being and operates. In this regard, psychological nominalism touches upon many important issues facing analytic thought. First, psychological nominalism offers a linguistically centered account of the advent of the rational mind and how it functions. As noted above, this account significantly differs from traditional views concerning the philosophy of mind. Second, since psychological nominalism accounts for how humans come to be rational, it indirectly offers an account of agency. In other words, it explains how human beings can be responsible for their actions by explaining the manner in which a person comes to be rational and conceptually aware of him/herself, and as a consequence, endowed with the ability to deliberate about the correctness or incorrectness of one’s own actions. This is obviously of interest to moral philosophy. Similarly, if rationality is required for personhood in the fullest sense, the ramifications of language learning as a prerequisite for rationality and agency have radical moral implications, e.g., paralinguistic people and children do not enjoy personhood in its fullest sense. Third, psychological nominalism accounts for how we perceive and conceive of existence. It thus plays an important role in epistemology. For instance, knowledge is not reducible to non-­ conceptual states such as sensing or an inborn understanding of universals and particulars. Instead, it indicates that knowledge requires the capacity to employ concepts and that concepts are acquired via learning a language. Learning a language is largely a process whereby one forms concepts by being habituated into the normative (i.e., accepted) use and practices that render observations communicable. This view has significant repercussions concerning not only how knowledge is obtained (e.g., as a social achievement), but also the type of metaphysical accounts one should attribute to cognitive/rational/intentional

12

Introduction

states. In this vein, Sellars believes all true accounts of mental events, and even intentionality, are commensurate with an ideal physical science. This is also true of the manner in which he views his account of psychological nominalism.10 In conclusion, Sellars’ psychological nominalism is at odds with much of the Western philosophical tradition with respect to how one thinks, possesses knowledge, and learns a language. It is also a theory that offers far reaching effects in all the major fields dearest to analytic tradition, namely philosophy of mind, epistemology, metaphysics, ethics, the philosophy of language, and (perhaps less dear) the philosophy of personhood.

Psychological nominalism and German idealism Kant’s Critiques imply that an accurate model of the human mind should be established before epistemic and metaphysical claims concerning outer existence. To this end, Kant undertook a transcendental approach in which he sought to disclose the necessary conditions for experience. A natural extension of this view is the position that a defensible philosophical approach hinges upon an accurate account of the manner in which we reason, perceive, communicate via a language, and judge. As a result, an account of such content can be seen not only as an attempt to establish the groundwork for what we can claim or hope to know, but also as a necessary step to bring such projects to fruition. It is precisely this kind of Kantian view that is implied, though nonetheless, at the heart of Sellars’ psychological nominalism. In this regard, psychological nominalism can be interpreted as a viable alternative to Kant’s transcendental account of experience. Insofar as Sellars’ psychological nominalism hinges upon a normative account of cognitive experience, it draws into question the kind of access human beings possess in regard to mind-­independent existence. For instance, due to the thick layering of normative and conceptual requirements for observational knowledge, there are renewed challenges facing the analytic tradition—challenges which were thought to be resolved. One of these more pressing problems concerns how naturalism can overcome the epistemic limits argued by German idealists, if the following are true: (1) no logical relations are given; (2) no sensory content possesses intrinsic epistemic value; and (3) all knowledge (i.e., “knowing that”) requires concepts, which are mental constructs that enable self-­awareness and “knowledge” of so called outer reality. Here I do not wish to insinuate that it is a foregone conclusion that such doubts are insurmountable, but rather that they

Introduction

13

require serious engagement. In fact, these are issues that Sellars himself grappled with throughout his career. For instance, Science and Metaphysics: Variations on Kantian Themes—Sellars’ self-­proclaimed sequel to “Empiricism and the Philosophy of Mind”— is an extended argument that seeks to show that his robust scientific realism is in fact defensible: Kant’s account implies indeed that certain counterparts of our intuitive representations, namely God’s intellectual intuitions, are literally true; but these literal truths can only be indirectly and abstractly represented by finite minds, and there is an impassible gulf between our Erkenntnisse and Divine Truth. If, however, as I shall propose . . ., we replace the static concept of Divine Truth with a Peircean conception of truth as the ‘ideal outcome of scientific inquiry’, the gulf between appearances and things-­in-themselves, though a genuine one, can in principle be bridged. SM, 50

Given the norm-­driven account of Sellars’ views on meaning and the foreknowledge that Sellars is a realist, one might expect (and believe he needs) a view of perception that un-­problematically supports realism. Instead, Sellars surprisingly makes numerous Kantian commitments that are central to his perceptual model—commitments which are historically interpreted as counter to realism. For example, take the following five Kantian commitments that Sellars adheres to in Science and Metaphysics (and throughout all of his later career): (1) neither sensibility (in itself) nor conceptuality (in itself) are capable of providing knowledge of the thing-­in-itself; (2) without a pre-­existing conceptual framework, no knowledge of empirical content is possible; (3) we are not directly aware of sensations; (4) in order to recognize empirical content, as facts or states of affairs, a judgment is required; and (5) in order to relate concepts to sensory content, via a judgment, the mind must first synthesize sensory content into coherent units of time and space.11 The above views are surprising for a realist to embrace, because they play a central role in Kant’s claim that the thing-­in-itself (an entity’s true existence independent of how a person’s mind may contingently experience, believe, or feel about it) is unknowable. Despite Sellars’ Kantian commitments, he believes, as noted above, that the gap “between appearances and things-­in-themselves, though a genuine one, can in principle be bridged” (SM, 50). Sellars also makes a series of claims that are peculiarly close to important features of Hegel’s philosophy. Perhaps the strongest parallel is Sellars’ rejection of the Myth of the Given and the similarities it bears to Hegel’s rejection of

14

Introduction

“sense certainty” in the Phenomenology of Spirit. There are also looser connections to draw between Sellars and Hegel. For instance, Hegel’s account of the “Also” and “Determinate Negation” appear to be an early predecessor to conceptual holism (regardless if these sections concern the actual metaphysical properties of objects or our conceptual understanding of them). Another noteworthy connection between the two philosophers is that they both believed it very important to pay attention to the history of philosophy, to reconcile our everyday experiences of the manifest world with our philosophical views, while honoring both. Sellars’ “Philosophy and the Scientific Image of Man” and Hegel’s account of nature in the encyclopedia of philosophy both strive, though in incompatible ways, to achieve these ends. Likewise, there are important connections to be drawn between Sellars’ view that we become aware of our mental lives by learning a language, as seen in the “Myth of Jones” in “Empiricism and the Philosophy of Mind,” and Hegel’s account of the “Master-Slave” dialectic, in which Hegel argues one is first aware of other people before one becomes intimately aware of oneself. In each of these very different accounts, one finds both authors arguing in their own ways that self-­awareness is not innate, as well as implying that the same can be said of personhood, because in both instances, something acquired via interactions with one’s immediate community is required for its achievement. These are just a few connections that can be drawn between Sellars and German idealism. Since Sellars’ texts are thoroughly engaged in a dialogue with Kantian and Hegelian views, his work almost begs to be addressed, defended, and criticized by these Juggernauts of Western thought. His ability to use the best of their views and thoroughly integrate them into his own philosophical views, which are robustly committed to a contemporary view of naturalism, is a testimony to his brilliance. This text seeks to renew such rich fruits by bringing the German idealist commitments embedded in Sellars’ view of mind, experience, and knowledge to bear on various realist topics. In doing so, one should ask: in what way should idealism or realism be defended in light of Sellars’ psychological nominalism? The original chapters (i.e., unpublished elsewhere) in this volume address this problem as they explore numerous issues relevant to both idealism and realism.12

Notes 1 The term “realist” denotes several different and incongruent philosophical views. The aforementioned use is only one possible formulation.

Introduction

15

2 Beginning most notably with Peter Abelard (1079–1142), philosophers have agonized over the debate as to whether universals exist independent of the human mind (i.e., realism) or, as Abelard argued, are constructs formulated by the human mind (i.e., nominalism). 3 “Universal realism” is a fabricated term (i.e., a stipulative definition). Its sole purpose is to designate the type of realism to which I am referring, and as such, keep my readers from confusing it with a variety of other views called “realism.” 4 Wittgenstein’s illustration of this point made him famous in Philosophical Investigations. 5 For a good account of why this claim is also generally true of logic see: “What the Tortoise Said to Achilles” by Lewis Carroll, Mind, 104 (416), 1995. 6 “This view is nominalist because it denies that cognitive phenomena involve any direct relation to abstracta [or what was above defined as ‘universals’] . . .; instead, it seeks to substitute a relation to linguistic entities. It is psychological because it is not the ontological doctrine that abstacta do not exist (full stop), but the psychological doctrine that abstacta are not involved directly in psychological, especially cognitive, phenomena as such” (deVries and Triplett, 2000, 193). 7 “The general framework of the givenness consists of the assumption that there are epistemic primitives—beliefs or other mental states that have some positive epistemic status but that are noninferential, conceptually simple, and epistemically independent and efficacious” (deVries and Triplett, 2000, 7). 8 This Sellarsian view holds many parallels to what Hegel calls “determinate negations.” 9 This however does not mean that a proficient language user is merely a creature of habit; rather, it means that one obtains the initial ability to successfully employ language via habits. 10 Here, it is important to note that Sellars rejects many views commonly associated with scientific realism, e.g., Sellars does not appeal to traditional formulations of scientific laws, correspondence theory, or representationalism. 11 I use Sellars’ technical term “sensory content” to denote the following: content that sensations contribute to perception, while denying that we are aware of sensations or “raw feels” in a preconceptual field of awareness. 12 My thanks to Willem deVries for helpful comments on an earlier draft of this paper.

Bibliography Carroll, Lewis (1995), “What the Tortoise Said to Achilles,” Mind, 104 (416). DeVries, Willem, and Triplett, Timm (2000), Knowledge, Mind, and the Given, Cambridge MA: Hackett Publishing Company. Locke, John (1995), An Essay Concerning Human Understanding, Delray Beach: Powerline Publishing Group. Loux, Michael (2010), Metaphysics: A Contemporary Introduction, New York: Routledge.

16

Introduction

Maher, Chauncey (2012), The Pittsburgh School of Philosophy: Sellars, McDowell, Brandom. New York, NY: Routledge. Ryle, Gilbert (2000), The Concept of Mind, New York: Penguin Classics. Sellars, Wilfrid (1997), “Empiricism and the Philosophy of Mind,” Cambridge MA: Harvard University Press. —— (1992), Science and Metaphysics: Variations on Kantian Themes, Atascadero: Ridgeview Publishing Company.

Part One

Psychological Nominalism and Realism

1

“Psychological Nominalism” and the Given, from Abstract Entities to Animal Minds James R. O’Shea, University College Dublin (IE)

In what sense was Sellars’ thesis of psychological nominalism, as he called it, a “nominalism”? And in what sense was that nominalism a psychological thesis? As we shall see, it turns out that Sellars formulated his thesis of psychological nominalism in two very different ways: (1) first, and most famously, as the thesis that all awareness of sorts and particulars is a linguistic affair (a thesis that he later carefully qualified to take account of animal cognition, aesthetic sensibility, and so on1); but also (2) secondly, and less well-­known, as a thesis about the psychology of the higher cognitive processes. The latter psychological thesis both denies the standard philosophical view that relations to (platonic) abstract entities are required in order to explain human thought and intentionality, and asserts to the contrary that all such mental phenomena can in principle be accounted for causally, as we shall see Sellars put it, without any use of normative or semantic terms in the explanation. Recent “Hegelian Sellarsian” philosophers such as Rorty, McDowell, and Brandom have argued that the holistic, normative themes that follow from thesis (1), above, ultimately support what Brandom and McDowell characterize as various (German) idealist but simultaneously common-­sense realist philosophical outlooks. By contrast, Sellars’ own defenses of thesis (2), above, show that he himself conceived his psychological nominalism as a naturalistic empiricism that is supposed to sustain the normative-­holistic themes following from thesis (1), but within an exhaustively scientific naturalist conception of reality. The examination of these two connected themes in Sellars’ work will thus cast his psychological nominalism in a very different light than it is usually seen. Sellars once wrote that “Philosophers have a peculiar form of the Midas touch. Everything they touch becomes a puzzle, and eventually a problem” (NAO II, §1).2 Of particular interest to Sellars was the philosophical problem of abstract

20

Wilfrid Sellars, Idealism, and Realism

entities, a problem that of course stretches from Plato to the most technical reaches of contemporary philosophy of language and mathematics. Put crudely, the classic Platonic (or “platonic”) problem concerns the nature of such universal abstractions as justice or courage, for example, or of such abstract mathematical entities as the number five or triangularity, as opposed to all of the concrete particular instances of those abstract entities, qualities, or relations. Philosophers with the Midas touch have drawn a host of careful distinctions in this connection, but for present purposes I will follow Sellars’ general practice and distinguish broadly between platonic realism and nominalism, as two of the most important perennial positions on the problem of abstract entities. Sellars was a nominalist, and for him, as we shall see, the problem of abstract entities was essential to that complex philosophical position that he called psychological nominalism. It will be worth backtracking a bit right at the start to see how this might be so. The platonic realist about universals and other abstract entities holds that such entities exist independently of our thought and language, and that they are not reducible to the concrete individuals that provide the many instances of them or “participate in” them. Such abstract entities, being by their very nature shareable by or instantiated in all of their many concrete instances, have consequently classically been held to exist “outside” of space and time, so to speak, and to be unaffected by the domain of physical causality in general. The nominalist, by contrast, holds either that abstract entities do not themselves exist at all—only concrete individuals exist—or, if they do exist, as Sellars held, then they are in some sense to be identified with or otherwise explained entirely in terms of our own acts of thought and our linguistic practices, or in terms of the functioning of representational or symbolic systems in general.3 A nominalist of Sellars’ stripe attempts to spell out such a view consistently with the “naturalistic” philosophical hypothesis that all such cognitive achievements can in principle themselves be fully explained as goings-­on that are restricted entirely within the spatio-­temporal framework of nature and its physical laws. In particular, the naturalistic nominalist of Sellars’ sort seeks to explain the rich contents of our mental lives (broadly, our “psychology”), and in fact all of the phenomena of meaning or ‘intentionality’ generally, without any appeal to or “positing’ ” of any of the sorts of real platonic abstract entities that have been thought to be required to explain how such mental contents or “intentional relations” are possible in the first place. We shall see that Sellars’ psychological nominalism, put negatively, turns out to consist precisely in the denial that there exist any such alleged psychological relations obtaining between minds and abstract entities (cf. Sellars’ EAE, passim). By contrast, it is fair to say

“Psychological Nominalism” and the Given

21

that the vast majority of analytic philosophers of mind and language today believe that the positing of such abstract entities and relations to them are required to explain a multitude of facts and puzzles about minds, meanings, morals, modals, and mathematics. Most contemporary discussions of Sellars’ psychological nominalism do not, however, treat that topic primarily in connection with the traditional problem of abstract entities. This is understandable, particularly in the wake of Richard Rorty’s widely read Philosophy and the Mirror of Nature in 1978, due to the discussions that have subsequently grown out of the most famous use to which Sellars himself put his psychological nominalism: namely, in his groundbreaking attack on the Myth of the Given in his classic 1956 article, “Empiricism and the Philosophy of Mind” (EPM). Let us turn first, then, to a brief examination of Sellars’ psychological nominalism as it arose in its most familiar context, in EPM.

Psychological nominalism and the Myth of the Given in EPM The most frequently quoted passage in connection with Sellars’ psychological nominalism is no doubt the following from EPM part VI, entitled “Impressions and Ideas: A Historical Point,” in which Sellars discusses: a view of the general type which I will call psychological nominalism, according to which all awareness of sorts, resemblances, facts, etc., in short, all awareness of abstract entities—indeed, all awareness even of particulars—is a linguistic affair. According to it, not even the awareness of such sorts, resemblances, and facts as pertain to so-­called immediate experience is presupposed by the process of acquiring the use of a language. EPM §29

Here psychological nominalism is explicitly stated as the thesis that all awareness of abstract entities, and even of particulars, is a linguistic affair. And while we shall see that this particular formulation requires some important qualifications that were later introduced elsewhere by Sellars, this can indeed be taken as the heart of the thesis of “psychological nominalism.” The classically nominalist aspect of the thesis is clear: nominalists, as the French and Latin roots of the word suggest, have traditionally attempted to explain so-­called abstract entities in terms of the representational functions or roles played by their corresponding “names.” For example, a nominalist might attempt to explain the universal kind

22

Wilfrid Sellars, Idealism, and Realism

or abstract entity lionhood in terms of our classificatory practices involving the word “lion” in its regular application to concrete individual lions, without appealing to any awareness or implicit grasp on the part of the language user of any non-­concrete, non-­spatial, non-­temporal objects, properties, or kinds of the sorts traditionally posited by (platonic) “realists” about abstract entities. The immediate context for the above definition is Sellars’ contention in EPM VI that Locke, Berkeley, and Hume all share the basic “presupposition that we have an unacquired ability to be aware of determinate repeatables” (EPM §29). This is despite their famously differing views about what they called “abstract ideas,” that is, about how any particular “idea” or image in the mind could represent the “determinable” idea of color in general, for instance. An example of the idea of a determinate sense-­repeatable would be one’s “immediate awareness” or conscious “sense impression” or “idea” of red, which is “repeatable” in the sense that many items can be red. As Sellars puts it, “however much Locke, Berkeley, and Hume differ on the problem of abstract ideas, they all take for granted that the human mind has an innate ability to be aware of certain determinate sorts— indeed, that we are aware of them simply by virtue of having sensations and images” (EPM §28, italics in original). However, Sellars suggests that “it takes but a small twist of Hume’s position to get a radically different view” (EPM §29). For such empiricist thinkers “often treat impressions or ideas of red as though they were red particulars”; and this “would become the view that all consciousness of sorts or repeatables [including all awareness of determinate repeatables] rests on an association of words (e.g. ‘red’) with classes of resembling particulars” (EPM §29). When Sellars then adds the further twist that such a ‘Hume’ might seek to explain such associations between words and particulars without (perhaps unlike Hume) implicitly importing or relying upon any other mediating awarenesses of abstract entities, such as awarenesses “of facts of the form x resembles y,” or of “the repeatable resemblance” (EPM §29), it is clear that Sellars is here describing a position that is getting closer to his own nominalist outlook on the nature of thought, language, and reality. For what Sellars is doing is rejecting a form of what he calls the “Myth of the Given,” in this case in the form of the idea, as he put it just above, that “we have an unacquired ability to be aware of determinate sorts” of experiences, for example of red, “simply by virtue of having” such sensations (EPM §29). By contrast, what Sellars’ “twist of Hume” enables is a view according to which “basic word-­world associations hold, for example, between ‘red’ and red physical objects, rather than between ‘red’ and a supposed class of private red particulars,” for

“Psychological Nominalism” and the Given

23

example one’s own sense impressions of red (EPM §29). According to such a view our having ideas of or beliefs about red objects would not be mediated by the “givenness” of any awareness of sorts, whether determinable or determinate— or indeed by the alleged givenness of “private” particulars either. Our awareness of red physical objects would of course be “causally mediated by sensations of red,” but without our awareness being “immediately” of those mediating sensations (EPM §29). It is the physical red apple itself that I directly perceive, on this view; I do not perceive the causally mediating sensations of red that are nonetheless essential to having a perceptual experience of that kind. In perception, I am not normally aware of my sensations as such, i.e. as sensations, but rather of the object of the perception (the apple). Of course, as developed merely in terms of simple word-­object associations, as in the above “twist on Hume,” such a view would be “impossibly crude and inadequate as an account of the simplest concept” (EPM §29). Accordingly, in the two crucial sections §30–1 that immediately follow in part VII of EPM, entitled “The Logic of ‘Means’,” we find a brief sketch of what Sellars thinks it really would take for the nominalist’s “word-­object associations” to give us concepts or meanings proper. In particular, such “words”4 must function normatively as playing a rule-­governed “role in a certain linguistic economy” (EPM §31). Only as functionally embedded, as I shall put it, within such a holistic framework of norm-­governed word-­object (perceptual and volitional) and word-­ word (inferential) associations do such linguistic patterns of behavior constitute what Sellars here calls “a structured logical space” (EPM §30). To take a simplified example, the word or concept “red” cannot function in such a way as to mean what it does mean unless speakers in a given linguistic community normally utter that word as it ought to be uttered in response to objects (e.g., in response to apples but not bananas), in their standing inferences (for example, from “x is red” to “x is colored”), and so on. Put crudely, words and concepts come to have the meanings that they do in virtue of how they are used, or rather how they ought to be used, in accordance with the wider package of implicit rules or social “ought-­to-be” norms that thus practically serve to maintain a given community’s various patterns of linguistic behavior, where the latter is construed as above to include our perceptual responses, inferences, and volitions. By the close of EPM part VII we are thus supposed to have a better view of what it really takes to be a psychological nominalist, on Sellars’ own way of thinking. This has turned out to require embracing various views concerning the essentially holistic, constitutively normative, implicitly “ought”-governed nature of conceptual thinking (intentionality), meaning, and knowledge. Such views, it

24

Wilfrid Sellars, Idealism, and Realism

is important to note, are very different from the sorts of positions that have classically been held by nominalists of a more naturalistic and empiricist bent: [T]he role of the word “red” by virtue of which it can correctly be said to have the meaning it does is a complicated one indeed, and . . . one cannot understand the meaning of the word “red”—“know what redness is”—unless one has a great deal of knowledge which classical empiricism would have held to have a purely contingent relationship with the possession of fundamental empirical concepts. EPM §31

So Sellars’ holistic psychological nominalism will not be of the classical empiricist variety. As Sellars sees the matter, those holistic requirements on meaning and knowledge are essential both to any plausible psychological nominalism, and to the rejection of the Myth of the Given. For his contention is that to know or mean anything at all, however basic or seemingly simply “given,” one must always already possess or be systematically coming to acquire (both naturally and normatively) appropriate competences within a wider functioning framework or “logical space” of norm-­governed inferential connections and other linguistic behaviors. As Sellars puts it in this context, “I wish to emphasize, therefore, that as I am using the term, the primary connotation of ‘psychological nominalism’ is the denial that there is any awareness of logical space prior to, or independent of, the acquisition of a language” (EPM §31). There is a clear link between psychological nominalism as thus introduced in EPM and a principle that Sellars later in 1981 described as “perhaps, the most basic form of what I have castigated as ‘The Myth of the Given’ ” (FMPP I, §44), and which has come to be known as the myth of the categorial given (cf. O’Shea 2007, 115–16). As Sellars put the myth in this “most basic form”: “If a person is directly aware of an item which has categorial status C, then the person is aware of it as having categorial status C” (FMPP I, §44). The upshot, as I have explained more fully elsewhere (e.g., 2007, chs. 3–5) is that on Sellars’ view not even the very nature of any object or item of awareness as, for example, mental or physical, or “outer” or “inner,” or even as a particular or a quality or a state, is simply given in the awareness per se; rather, it is given to us only insofar as that object or item is conceived or represented in the awareness (correctly or incorrectly) as being an item of that fundamental kind or sort. And such a representation or meaning is possible only because something in the awareness—for example, a word (or on Sellars’ “Myth of Genius Jones,” something functioning like a word in a “language of thought” in the brain)—is such that, whether thanks to nature or nurture, it has thus come to function normatively and holistically as a

“Psychological Nominalism” and the Given

25

representation or signifier of items of that kind or sort.5 In this way the connections between Sellars’ normative-­functionalist “psychological nominalism” and his rejection of the Myth of the Given run very deep indeed, and did so throughout his career. With the above essential connections between Sellars’ psychological nominalism and his more famous holism-­inspired rejection of the Myth of the Given now more clearly in view, we are in a position to appreciate, the impact that Sellars’ psychological nominalism has had on some of the most well-­known philosophers who have been influenced by his views on these particular matters. Following this, I will then turn in the final section to Sellars’ “Empiricism and Abstract Entities” (EAE) and “Mental Events” (MEV), to tease out some of the less well-­known aspects of Sellars’ psychological nominalism, and this will ultimately bear centrally on problems pertaining to realism and idealism.

Rorty, Brandom, and McDowell on psychological nominalism, realism, and idealism Perhaps the most famous uses to which Sellars’ psychological nominalism has explicitly been put have been in the influential works of Richard Rorty, John McDowell, and Robert Brandom, the latter two having been strongly influenced by both Sellars and Rorty, as well as by Kant and Hegel. In Philosophy and the Mirror of Nature (1978), particularly in chapter four, section three on “Pre-Linguistic Awareness,” Rorty quoted and defended in some detail Sellars’ basic psychological nominalist thesis that all awareness of abstract entities and particulars is a linguistic affair. Rorty presents Sellars’ psychological nominalism as an “epistemological behaviourism” that is “indistinguishable from epistemological holism” (1978, 188), and “which might be called simply ‘pragmatism’ ” as Rorty understands the latter (1978, 176). Rorty rightly focuses on the essential link between psychological nominalism and the rejection of the given, and thus in relation to the latter on Sellars’ famous “logical space of reasons” conception of knowledge without reliance on any mythic given: The essential point is that in characterizing an episode or a state as that of knowing, we are not giving an empirical description of that episode or state; we are placing it in the logical space of reasons, of justifying and being able to justify what one says. EPM §37

26

Wilfrid Sellars, Idealism, and Realism

This is how Rorty sums it up: To sum up, Sellars’s psychological nominalism is not a theory of how the mind works, nor of how knowledge is born in the infant breast, nor of the “nature of concepts”, nor of any other matter of fact. It is a remark about the difference between facts and rules, a remark to the effect that we can only come under epistemic rules when we have entered the community where the game governed by these rules is played. We may balk at the claim that knowledge, awareness, concepts, language, inference, justification and the logical space of reasons all descend on the shoulders of the bright child somewhere around the age of four, without having existed in even the most primitive form hitherto. But we do not balk at the thought that a cluster of rights and responsibilities will descend on him on his eighteenth birthday . . . [I]n both cases what has happened is a shift in a person’s relations with others, not a shift inside the person which now suits him to enter such new relationships. Rorty 1978, 187

That is, knowledge and meaning on this Sellarsian view—in this respect similar to the right to vote—do not consist in and are not reducible to any detectable empirical properties or reliable causal relations per se (though such phenomena will of course require or presuppose that various appropriate empirical properties and causal relations are in place). Rather, such cognitive states are by their very nature normative statuses, for example, of justification or of correct inference or of proper immediate response. They have to do not with we are made of but with what can rightfully be done or said, within a rule-­governed “logical space” of giving and asking for reasons. At a later stage, however, some contrasting light will be thrown on Rorty’s opening remark above that “Sellars’ psychological nominalism is not a theory of how the mind works.” For we shall see that Sellars does offer such definitions of “psychological nominalism” as the following (in “Empiricism and Abstract Entitites,” written roughly contemporaneously with EPM):6 Platonism, therefore, is, in essence, a thesis in the psychology of the higher process; and to reject it—which by no means involves a rejection of the linguistic framework of abstract entities7—is to be what I shall call a “psychological nominalist”. EAE 442, III. 248

For Sellars, but not Rorty, the constitutively norm-­governed or rule-­conforming conception of our knowledge in terms of the “logical space” of giving and asking for reasons, which Rorty has correctly highlighted above, is itself inseparable

“Psychological Nominalism” and the Given

27

from what is precisely (pace Rorty) “a theory of how the mind works”: in this case Sellars’ normative-­functionalist theory of conceptual content or intentionality in general, which stands in opposition to the currently thriving industry of accounting for the “propositional contents” that are grasped by minds in terms of basic semantic relations to abstract entities of various kinds (or perhaps in terms of “possible worlds” as concreta or abstracta). Whether this difference in emphasis between Rorty and Sellars on “psychological nominalism” is philosophically significant or not is something that should emerge before we are done. What Rorty primarily takes from Sellars’ psychological nominalism is the view that knowledge is a matter of conversational justification amongst one’s social peers, rather than a matter of our minds, or our language, or the language of science, allegedly functioning as cognitive “mirrors” or adequate “pictures” of the ultimate intrinsic nature of mind-­independent reality. For Rorty, if not for Sellars, a “holistic approach to knowledge is not a matter of antifoundationalist polemic, but a distrust of the whole epistemological enterprise” (1978, 181). And similarly, for Rorty, to adopt Sellars’ psychological nominalism or a pragmatist, epistemological behaviorism is “to adopt the view that philosophy (and, specifically, ‘philosophy of mind’) cannot, by supplying a loftier critical point of view, reinforce or diminish the confidence in our own assertions which the approval of our peers gives us” (1978, 188). In a later paper on “John McDowell’s version of Empiricism,” Rorty puts the broader upshot this way: I take the linguistic turn in philosophy, the turn that made it possible for Sellars to envisage his doctrine of psychological nominalism, to be a turn away from the very idea of answerability to the world. . . . I regard the need for world-­ directedness as a relic of the need for authoritative guidance, the need against which Nietzsche and his fellow pragmatists revolted. Rorty 1998, 142–3

Here Rorty is criticizing McDowell’s particular way of seeking to attain what McDowell in Mind and World had described as “something Rorty himself aspires to, a frame of mind in which we would no longer seem to be faced with problems that call on philosophy to bring subject and object back together again” (McDowell 1996, 86). What Rorty disagrees with is McDowell’s view that there is a genuine insight lying at the root of the “distinctive anxieties of modern philosophy” (McDowell 1996, xvi). McDowell diagnoses these anxieties as the result of “a tension between two forces” or ideas. One of those ideas, which Rorty also endorses, is Sellars’ doctrine of psychological nominalism, that is, that “all

28

Wilfrid Sellars, Idealism, and Realism

awareness . . . is a linguistic affair” (EPM §29), including the essentially related idea that all knowledge or justification is a normative standing within the “logical space of reasons.” In his most recent work McDowell (2016) has expressed Sellars’ psychological nominalism this way: Consider a visual experience that enables its subject to know there is something red and triangular in front of her. It is a Sellarsian thought that, like all experiences, an experience of which that is true would be at least partly constituted by acts of capacities that belong to the subject’s power of discourse. That is a way of saying it would come within the scope of Sellars’s “psychological nominalism”. (See EPM §29.) [In ‘The Structure of Knowledge’] Sellars expresses the idea by saying “perceiving essentially is or involves a thinking”. Sellars SK I.30, McDowell 2016, §2

The other idea or “force” that contributes to generating the tension in modern philosophy is articulated by McDowell in terms of the attractiveness of a minimal empiricism, which makes out that the very idea of thought’s directedness at the empirical world is intelligible only terms of answerability to the tribunal of experience, conceived in terms of the world impressing itself on perceiving subjects. McDowell 1996, xvi

The tension for McDowell, as is well-­known, arises in large part from the fact that the raw “sense impressions” that affect our sensibility have tended to be conceived by modern philosophers in terms of natural scientific descriptions that do not by themselves make it intelligible how such brutally causal impacts by their very nature are able to function normatively within Sellars’ logical space of reason-­giving justification, and thus serve as a “tribunal of experience.” For a sense experience can provide justification for a belief as to how things are, and thus be able to serve as a “tribunal” or test of that belief, only if the sensory experience by itself can reveal the presence of a recognisable object or the obtaining of a stateable fact in the world—either of which requires concepts. What we need to recover, according to McDowell, is the idea that nature is not exhausted by such lawful scientific descriptions, and in particular that the sense impressions by which nature reveals itself to human knowers are not bare unconceptualized (and thus Mythic) Givens, but are rather, as Kant saw, both passively sensorily receptive and subject to our conceptual-­linguistic discursive capacities from the ground up: “Experiences are impressions made by the world on our senses, as products of receptivity; but those impressions themselves already have conceptual content” (McDowell 1996, 46).

“Psychological Nominalism” and the Given

29

The overall result is a view of the relationship between mind and world that McDowell sees as consistent with the central German idealist insights of Kant, as corrected by Hegel (for example, on sensibility, the “thing in itself,” and history); but it is a view that thereby, he argues, also retrieves empirical realism as a genuine normative answerability to nature as a mind-­independent world of fact, a world that by its very nature is open to our possible experience of it. For on this view what our sense experiences by themselves directly reveal to us, when all goes well, is the presence of a knowable object or an apprehension of how matters stand in the world. Toward the end of his second lecture in Mind and World, “The Unboundedness of the Conceptual,” McDowell framed it this way: It is central to Absolute Idealism to reject the idea that the conceptual realm has an outer boundary, and we have arrived at a point from which we could start to domesticate the rhetoric of that philosophy. Consider, for instance, this remark of Hegel’s: “In thinking, I am free, because I am not in an other.” This expresses exactly the image I have been using, in which the conceptual is unbounded; there is nothing outside it. The point is the same as the point of that remark of Wittgenstein’s [that] “We—and our meaning—do not stop anywhere short of the fact”. [Philosophical Investigations, §95], McDowell 1996, 44

In a more recent article entitled “Conceptual Capacities in Perception,” McDowell describes his view as “an idealism that does not diverge from common-­sense realism,” a view according to which “thought and the world must be understood together” (McDowell 2009, 143; for a full discussion see O’Shea 2009). As McDowell had argued in Mind and World (1996, 28), it is on his view sufficient for realism—that is, sufficient to avoid “a phobia of idealism” and to assert “the independence of reality”—to remind ourselves that the external “constraint” that is impressed on our thinking by the world “comes from outside [our acts of] thinking, but not from outside what is thinkable.” One is perhaps reminded, by McDowell’s idealistic defense of realism, of the following well-­known remark of C.I. Lewis’s in his classic 1929 book, Mind and the World-Order. Lewis, having defended the broadly Kantian idea that the object of our knowledge consists in a lawful “if–then” prediction of the future possible experiences that are appropriate to that type of object as conceived, argues furthermore that this is sufficient to account for the mind-­independence of the object of our knowledge. This is because the mind cannot dictate what will in fact be given in those future possible experiences. So both Lewis and McDowell in a sense argue that while all objects and facts are essentially conceptually thinkable, such objects are independent of our acts of thinking. And Lewis concludes by remarking:

30

Wilfrid Sellars, Idealism, and Realism If the idealist should find that there is nothing in such ‘independence’ which is incompatible with his thesis, then it may be that between a sufficiently critical idealism and a sufficiently critical realism, there are no issues save false issues which arise from the insidious fallacies of the copy-­theory of knowledge. Lewis 1929, 194

Thus both C.I. Lewis and McDowell are intending to defend an idealism, as it were, that is at one and the same time a common-­sense realism. I shall return to McDowell’s views at a later stage. Let me first close this section, however, with a brief examination of how a third “Hegelian Sellarsian” (cf. O’Shea 2002), Robert Brandom, has conceived the best way to make constructive use of Sellars’ psychological nominalism. In his ‘Study Guide’ to Sellars’ EPM, Brandom describes it this way: . . . Sellars will argue for what he calls “psychological nominalism” (not the best imaginable name), according to which all awareness of repeatables (whether determinate or determinable) is a linguistic affair, and hence may not be presupposed in one’s account of the acquisition and functioning of language. Sellars is proposing a linguistic, social theory of . . . classificatory awareness, awareness of something as something. . . . Such awareness, specifically conceptual awareness, requires something beyond being awake and classifying by [mere] differential response. Brandom 1997, 150–1

Brandom, unlike McDowell, thus stresses the constitutively social and specifically inferential origins of conceptual content, and hence of the type of awarenesses that fall under Sellars’ psychological nominalism. Like McDowell, however, Brandom has articulated the overall outlook that emerges as a view that is supposed to be consistent with the leading themes of German idealism in Kant as corrected by Hegel, while at the same time preserving the mind-­independence of empirical reality. (In particular, Brandom seeks to preserve a “structural objectivity” such that the truth of a claim does not depend on the expressive capacities of any particular linguistic community.) One way in which Brandom articulates his own version of an idealistic realism (if I may use that term to cover the views of both McDowell and Brandom, despite their many differences), for example in his recent book on Sellars (2015), is in terms of Kant’s transcendental idealism and empirical realism, and in particular in relation to what Brandom calls the “Kant–Sellars thesis” about modality (i.e., about possibility and necessity).

“Psychological Nominalism” and the Given

31

Brandom takes his working out of the modal Kant–Sellars thesis to constitute a pragmatic modal expressivism, in that what we are doing in applying modal vocabulary (for example, in asserting that “necessarily, copper conducts electricity”) is expressing or making explicit the conceptual connections and commitments that are already implicit in our ordinary uses of empirical concepts (such as “copper” and “conducts electricity”). For Brandom, however, the objective correlate of this modal expressivism is a semantic modal realism, in that what we are saying in applying modal vocabulary is that various modal properties, relations, and facts (for example, necessary causal connections) are instantiated or obtain in the objectively real empirical world. In fact, Brandom sees his view, like Kant’s and Sellars’, as one that “puts modal expressivism and modal realism together again” precisely because it is “recognizably a development and a descendant, for this special but central case, of Kant’s claim that one should be a transcendental idealist, but an empirical realist” (Brandom 2015, 215). In a nutshell, the “idealist” aspect of Brandom’s “modal Kant–Sellars thesis” is the idea that what is meant by our claims about what objective facts and laws there are in the world is dependent for its very sense on aspects of our norm-­ governed practices of asserting claims and making inferences. The modal realist semantics is thus “sense-­dependent,” as Brandom puts it, on the modal expressivist pragmatics. Put in Kant’s terms, this entails the transcendental idealist thesis that, for example, any causally lawful world must be one in which certain corresponding normative inferential practices and commitments would also be in place. But this sense-­dependence does not entail, according to Brandom, the absurd “reference-­dependence” thesis such that there could not exist in objective reality any facts and laws without the existence of corresponding practices of assertion or inference (Brandom 2015, 207–15). There were objective facts and laws concerning the melting point of copper, for instance, before any human beings evolved with the capacity to make the corresponding assertions and inferences.9 We have now before us John McDowell’s and Robert Brandom’s different ways of grounding in Sellars’ psychological nominalism what are supposed to be two idealist yet also common sense realist outlooks, both of them indebted to Rorty’s similarly grounded critique of the idea of the mind or of knowledge as a “mirror of nature.” Let us turn finally, then, to Sellars’ own further articulation of the significance of psychologism nominalism for his own philosophical views, and in particular how the nominalist aspects of that view are supposed to facilitate a certain kind of naturalistic empiricism with respect to the mind and its place in nature.

32

Wilfrid Sellars, Idealism, and Realism

Sellar’ psychological nominalism as a naturalistic empiricism Sellars wrote “Empiricism and Abstract Entities” in the mid-1950s for the Schilpp volume on Carnap (1963), and his focus was on Carnap’s account of abstract entities in “Empiricism, Semantics and Ontology.” For present purposes, however, I want to focus on Sellars’ various characterizations of psychological nominalism and its consequences. The crucial guiding theme behind Sellars’ analysis, as indicated earlier, concerns “the persistent (if currently repressed) notion that relations between minds and abstract entities must be invoked by an adequate psychological theory of the ‘higher processes’ ” (EAE 436, II.12). As also noted earlier, there is no doubt that this continues to be the standard view among contemporary analytic philosophers of mind and language. By contrast, Sellars sought a view of mind and language that “safeguards psychological nominalism” and does not “leave the door open to Platonistic metaphysics” (EAE 442, III.25), thereby making possible the ultimate conclusion that, in Sellars’ own eyes, psychological nominalism is intended to support, and which he thinks Carnap’s analysis partly aided, but partly hindered: namely, that “Today, for the first time, the naturalistic-­empiricist tradition has the fundamentals of an adequate philosophy of mind” (EAE 468, VIII.81, italics added). Sellars took this to be “a truly revolutionary situation, which is just beginning to make itself felt” (EAE 468, VIII.81). We have already seen in §1 some of the main normative, holistic, and anti-­ givennist reasons why for Sellars this “truly revolutionary” naturalistic empiricism, as he characterizes it, cannot be an empiricism of any of the classical varieties. Furthermore, Sellars’ empiricism must somehow be a naturalistic (and not an idealistic) empirical realism that nonetheless accommodates the sorts of holistic, Kantian-­conceptualist insights that are embodied in psychological nominalism and which were developed in various ways by Rorty, McDowell, and Brandom, as we saw in §2. Let us consider a further aspect of psychological nominalism as discussed in EAE: I shall use the term “Psychological Nominalism” to stand for the denial of the claim, characteristic of the realistic [i.e. platonist] tradition, that a “perception” or “awareness” of abstract entities is the root mental ingredient of mental acts and dispositions. In other words, the psychological nominalist argues that it is in principle possible to describe and causally account for the episodes and dispositions singled out by such sentences as “John believes that it is raining”, without positing a “perception” or “awareness” of abstract entities. EAE 445, III.31

“Psychological Nominalism” and the Given

33

This is not a characterization of Sellars’ “psychological nominalism” that one would find in Rorty, Brandom, or McDowell, or usually anywhere else either, apart from in Sellars. On the one hand, for Sellars in EAE just as much as for the Hegelian Sellarsians, “characteristically semantical words” such as “means”—and we could say the same about “John believes” in the passage above—“have a conceptual role which is no more reducible to non-­semantical roles than the role of prescriptive terms is reducible to non-­prescriptive roles” (EAE 459, VII.60). Nonetheless, for Sellars, “psychological nominalism [is] the thesis that linguistic phenomena can, in principle, be described and causally accounted for without using semantical or prescriptive expressions” (EAE 465,VII.73). What is going on here? What we thus run up against in Sellars’ own definitions of the thesis of “psychological nominalism” is what I have elsewhere characterized as Sellars’ naturalism with a normative turn (O’Shea 2007; 2009). The “program of psychological nominalism” (EAE 448, III.38) is an explicit statement of Sellars’ ambitious aspiration to provide a “conceptual role” theory of mind, meaning, and knowledge that would at one and the same time preserve the conceptual or normative-­pragmatic irreducibility of our rationality and intentionality—and hence, the irreducibility of persons themselves—while simultaneously, at least “in principle,” enabling the complete and exhaustive natural-­scientific causal explanation of those very same phenomena. I cannot pretend to be able to provide an adequate defense or analysis of the structure of this program, and it is doubtful whether Sellars was able to work out fully or with sufficient clarity what he had in mind. But I have argued elsewhere that Sellars’ overarching program remains worthy of our further consideration, and here what I have been concerned to point out is how central that program was to what Sellars himself conceived of as psychologism nominalism—as well as to highlight how utterly forgotten that conception has become in subsequent appropriations of that particular doctrine. In what ways, then, is Sellars’ psychological nominalism supposed to support or integrate with his comprehensively naturalist realism with a normative turn? The psychological nominalist “thesis that linguistic phenomena can, in principle, be described and causally accounted for without using semantical or prescriptive expressions” (quoted above) is puzzling in that on certain interpretations the thesis might be true but trivial, and certainly not what Sellars had in mind (cf. O’Shea 2009); while on more substantive interpretations the thesis threatens to run afoul of Sellars’ own anti-­reductionist conceptions of the constitutively normative “logical space of reasons” rightly highlighted by Rorty,

34

Wilfrid Sellars, Idealism, and Realism

McDowell, and Brandom. Consider this further statement of the thesis of psychological nominalism (where Sellars’ “—–” refers to a linguistic item): These questions bring us to the heart of the matter. The expression “the role of ‘—’ ” is ambiguous. If it is being used in a context of interest in which expressions are predicates, . . . logical constants, etc. etc. then of course the role of “—” cannot be specified without using the categories of syntax and semantics . . . . But “the role of ‘—’ ” can also be understood in another sense. In this sense, to ask What is the role of “—”? is not to ask about the role of an expression. It is to ask about the causes and effects of a certain empirically definable stimulus configurations [sic]. Here the word “role” is used as in What is the role of HCL in the electrolysis of H2O? And it is the thesis of psychological nominalism that the questions as to the role of “—” thus understood requires [sic] no use of semantical or syntactical terms in the answer. EAE 461, VII.64; the grammatical agreement issues are in the original text

Since Sellars holds that semantical terms such as “means,” for example, are essentially normative terms that serve the pragmatic function of functionally classifying the implicitly ought-­governed roles of linguistic and other representational items, he is here defining psychological nominalism as the thesis that non-­normative “cause and effect” answers can be given to questions concerning the roles of those same linguistic or other representational items. Psychological nominalism, for Sellars, is thus the thesis that the same conceptual roles that are described (indeed, constituted) normatively using semantic and mentalistic vocabulary can also be (sufficiently?) “described and causally accounted for” using non-­normative, scientific naturalist vocabulary. His basic idea, I believe, was that having argued for the replacement of the usual conception of mentality as involving an ostensibly non-­natural relation to abstract entities (realistically construed) with his own functional role semantics, which involves no relations between language and reality other than natural causal relations, then no bar remained in principle to what he describes as “a gapless description and explanation of the Borneo [island] social scene in behavioristic terms, and therefore in which no prescriptive term occurs” (EAE 453, IV.48). We can indeed say that the “logical space of reasons” has the shape that it does, and thus things mean what they do, only in virtue of persons using normative vocabulary in the ways that they do—or more accurately, in virtue of such persons “normally” using such terms in the ways that they “ought to do.” But at the same time we could in principle, Sellars held, give a sufficient causal-­ explanatory account of how normative-­classificatory terms themselves function in our lives in the ways that they do, without that explanation itself having to be

“Psychological Nominalism” and the Given

35

couched in normative terms.10 (For further clarification and a partial defense of exactly how Sellars thinks he can successfully defend both the causal-­explanatory reducibility and the normative-­pragmatic irreducibility aspects of his naturalism with a normative turn, in one coherent picture, see O’Shea 2007 and 2009.) Presumably the Hegelian or “left-­wing” Sellarsians, as they have also been called,11 such as Rorty, McDowell, and Brandom, would find Sellars’“psychological nominalism” implausible as Sellars explains it in EAE. McDowell, for example, has argued that Sellars had a “blind spot” in connection with his view that meaning and intentionality more generally are not “relations to the world” (McDowell 2009, chs. 11 and 12). McDowell argues that a (Tarski–Davidson) truth-­conditional semantics can respect all of Sellars’ correct normative-­holistic points connected with psychological nominalism without falling prey to either the Platonism or the (semantic or epistemological) atomism that Sellars thought inevitably must accompany any view that embraces basic semantic relations to the world. For example, Sellars argued that Tarski’s semantics as employed by Carnap rested ultimately upon lists of isolated (hence, atomistic) “name–object” pairings that provide no plausible account of how names actually function in any living language. But McDowell argues that Tarskian truth-­conditional semantics as developed by Davidson is both normative and holistic with respect to meaning and reference in ways that successfully avoid that style of criticism from Sellars, while preserving his key insights regarding the given and the space of reasons. In a very different way, on the other hand, the left-­wing Sellarsian Michael Williams (2016) has argued that those same Sellarsian insights can be preserved within an unreservedly “deflationist” outlook on truth and meaning, thus again presumably skirting the need for anything like Sellars’ more reductive-­ sounding claims in EAE on behalf of psychological nominalism as a way of “safeguarding” scientific naturalism. I regard it as a still open question whether a Davidsonian or a deflationist semantics can successfully preserve the normative-­ holistic insights of Sellars’ psychological nominalism without carrying in its train the sort of stronger naturalistic claims that Sellars took his own psychological nominalism to entail, as explained above. That said, however, I will close by pointing out a possible virtue of Sellars’ own way of understanding his conceptual role semantics and his psychological nominalism, and this will serve to bring out another relatively neglected aspect of Sellars’ own naturalistically conceived “program of psychological nominalism.” For Sellars not only regarded his psychological nominalism as providing a via media between Platonism (with its realistically construed abstract entities) and classical empiricism (with its semantic and epistemological atomism) in the

36

Wilfrid Sellars, Idealism, and Realism

philosophy of mind and language, thus avoiding both the rationalist (or Platonist) and empiricist versions of the Myth of the Given. He also—and this is the part that is usually overlooked—regarded his functional role accounts of meaning and intentionality as opening up the space for a seamless integration of non-­ linguistic (and pre-­linguistic) animal cognition within the purview of psychological nominalism. Sellars was most explicit about this in two of his last articles,“Mental Events” (MEV, 1981) and “Behaviorism, Language and Meaning” (BLM, 1980) Sellars’ basic idea12 was that something analogous to the sort of normative-­holistic “logical space of reasons” that we have seen to be required for thought and meaning according to psychological nominalism can also be attributed to non-­linguistic animal-­cognitive “representational systems” thanks to their broadly adaptive evolutionary origins: §56. Indeed, I propose to argue that to be a representational state, a state of an organism must be the manifestation of a system of dispositions and propensities by virtue of which the organism constructs maps of itself in its environment, and locates itself and its behavior on the map. §57. Such representational systems (RS) or cognitive map-­makers, can be brought about by natural selection and transmitted genetically, as in the case of bees. Undoubtedly a primitive RS is also an innate endowment of human beings. The concept of innate abilities to be aware of something as something, and hence of pre-­linguistic [non-­conceptual] awarenesses is perfectly intelligible. MEV 336, III; cf. BLM VI

Here biological proper functioning stands in for the normativity of a logical space of reasons, and the relevant holism in this case refers to a systematic embedding within an evolutionary framework instead of within a logico-­linguistic structure. But the core of Sellars’ naturalistically conceived psychological nominalism is preserved in both cases, with all cognitive awareness of a world respecting the requirement that any such state be embedded within a wider normative framework—or in this case, within a wider evolutionary “biological space” (not a logical space) of “proper functioning”—with the result that no mysterious intentional or semantic “relations to the world’ ” are posited in either case. One thing that is clear, as we have seen, is that Sellars himself regarded his psychological nominalism as supporting a robustly naturalistic realism (as opposed to idealism), which he also conceived as a kind of Kant-­corrected empiricism. What is also clear is that the Hegelian Sellarsians, not without good reason, have found a more idealistic empirical realism to be supported by the normative-­holistic dimensions of Sellars’ psychological nominalism. What is not

“Psychological Nominalism” and the Given

37

so clear, and a matter for further inquiry, is whether Sellars’ own naturalistic aspirations for his psychological nominalism, or rather the adaptations of that outlook adopted by his more idealistic interpreters, will be found in the end to have provided the more coherent synoptic vision of our thinking animal nature.13

Notes 1 Cf. Sellars “Mental Event’ ” and “The Structure of Knowledge” (Lecture I). 2 References to Sellars’ works are to the standard abbreviations listed in the References, and references such as “II.1” are to the relevant chapter (or part) and section (or paragraph) of the given work. 3 Following Sellars, I will include various aspects of “conceptualism” about universals under the umbrella of “nominalism” when, as for present purposes, it is not important to distinguish them. 4 Or also, for Sellars, the “inner” analogues of such words—“mental words,” as it were, in a “Mentalese” or “language of thought,” to anticipate Sellars’ famous “Myth of Genius Jones” account of inner, silent thoughts, modeled on outer linguistic behavior, in the second half of EPM. 5 Of course, whether this is an adequate nominalism, and does not rather import real “sorts” or universals of a specifically linguistic kind, would require further discussion. See for example O’Shea 2007, chapter 3, and Brandom 2015. 6 Sellars referred to “Empiricism and Abstract Entities” (EAE) in endnote twenty of EPM as “available in mimeograph form from the author.” It was not actually published until the Schilpp volume on Carnap appeared in 1963, but the views on psychological nominalism expressed in it were held roughly at the same time as those in EPM. 7 For an excellent defense of Sellars’ own theory of abstract entities, as a codification of our normative practises that saves the practise of metaphysics in a way that should not be objectionable to contemporary pragmatists, see Kraut 2016. 8 References to EAE are to the page number in the Schilpp volume (1963), followed by part and paragraph numbers (e.g., III.2). 9 The preceding two paragraphs have been adapted with minor revisions from O’Shea 2015, 211–12. 10 Of course, for Sellars any use of language, whether in the “manifest image” or in the ideal “scientific image,” will require that such sayings, thinkings, or representings are governed by the normative “ought-­to-be”s that make any language possible in the first place. Sellars’ point here is that such norm-­governed causal explainings will not themselves involve the use of normative vocabulary to characterize what is being explained (which is in this case the origin and nature of our use of normative

38

Wilfrid Sellars, Idealism, and Realism

vocabulary itself). Failure to make such clearly intended “act-­object” distinctions in this connection has led to serious confusion in interpreters’ criticisms of what Sellars means when he writes of the “non-­normative” character of scientific characterizations of reality. (His views on the latter may have other problems, but it is not guilty of that confusion.) 11 For further discussion of the problematic but sometimes useful distinction between so-­called “left-­wing” and “right-­wing” Sellarsians, see the introduction to O’Shea 2016. 12 I have discussed the importance of Sellars’ views in “Mental Events” in relation to animal and human cognition in several other locations, including O’Shea 2007, ch. 5 and 2010, part III. 13 I would like to thank Patrick Reider for his very helpful comments and suggestions, and also my colleagues in the School of Philosophy at UCD for their insightful comments in our work in progress seminar.

Bibliography Brandom, Robert B. (1997), “Study Guide,” in Wilfrid Sellars, Empiricism and the Philosophy of Mind, Cambridge, MA: Harvard University Press, 119–81. —— (2015), From Empiricism to Expressivism: Brandom Reads Sellars, Cambridge, MA: Harvard University Press. Carnap, Rudolf (1950), Revue Internationale de Philosophie 4 (1950): 20–40. Reprinted in the Supplement to Meaning and Necessity: A Study in Semantics and Modal Logic, enlarged edition, Chicago: University of Chicago Press, 1956. Kraut, Robert (2016), “Norm and Object: How Sellars Saves Metaphysics from the Pragmatist Onslaught,” in O’Shea (ed.) 2016. Lewis, Clarence Irving (1929), Mind and the World-Order: Outline of a Theory of Knowledge, New York: Dover Publications, Inc. McDowell, John (1996), Mind and World, Cambridge, MA: Harvard University Press. —— (2009), Having the World in View: Essays on Kant, Hegel, and Sellars, Cambridge, UK: Polity Press. —— (2016), “A Sellarsian Blind Spot,” in O’Shea (ed) 2016. O’Shea, James, R. (2002), “Revisiting Sellars on the Myth of the Given,” in the International Journal of Philosophical Studies, 10.4: 490–503. —— (2007), Wilfrid Sellars: Naturalism with a Normative Turn, Cambridge, UK: Polity Press. —— (2009), “On the Structure of Sellars’s Naturalism with a Normative Turn,” in Willem A. deVries (ed.), Empiricism, Perceptual Knowledge, Normativity and Realism, Oxford: Oxford University Press, 187–210. —— (2010), “Having a Sensible World in View: McDowell and Sellars on Perceptual Experience,” Philosophical Books, 51.2: 63–82.

“Psychological Nominalism” and the Given

39

—— (2015), “Concepts of Objects as Involving Laws: A Kantian and Pragmatist Line of Thought,” in Robert Stern and Gabriele Gava (eds.), Pragmatism, Kant, and Transcendental Philosophy, London: Routledge, 196–216. —— (2016), Sellars and His Legacy, Oxford: Oxford University Press. Rorty, Richard (1978), Philosophy and the Mirror of Nature, Princeton: Princeton University Press. —— (1998), Truth and Progress: Philosophical Papers, Cambridge: Cambridge University Press. Sellars, Wilfrid   BLM —— (1980), “Behaviorism, Language and Meaning,” Pacific Philosophical   Quarterly 61: 3–30.   EAE —— (1963), “Empiricism and Abstract Entities,” in Paul A. Schilpp (ed.), The  Philosophy of Rudolf Carnap (The Library of Living Philosophers), La Salle: Open Court, 431–68.   EPM —— (1956), “Empiricism and the Philosophy of Mind,” in Herbert Feigl and  Michael Scriven (eds.), Minnesota Studies in the Philosophy of Science, vol. I, Minneapolis: University of Minnesota Press, 253–329. References are to the section numbers that are also in all reprintings of EPM.   FMPP —— (1981), “Foundations for a Metaphysics of Pure Process,” (The Carus   Lectures), The Monist 64: 3–90.   MEV —— (1981), “Mental Events,” Philosophical Studies 39: 325–45.   NAO —— (1980), Naturalism and Ontology, Atascadero: Ridgeview Publishing   Company.   SK —— (1975), “The Structure of Knowledge: (1) Perception; (2) Minds; (3)  Epistemic Principles” (The Matchette Foundation Lectures for 1971 at the University of Texas), in Hector-Neri Castañeda (ed.), Action, Knowledge and Reality: Studies in Honor of Wilfrid Sellars, Indianapolis: BobbsMerrill, 295–347. Williams, Michael (2016), “Pragmatism, Sellars, and Truth,” in O’Shea (ed.) 2016. Wittgenstein, Ludwig ([1953] 2009), Philosophical Investigations, G.E.M. Anscombe trans., revised 4th edn. P.M.S. Hacker and Joachim Schulte (eds.), Oxford: WileyBlackwell.

2

Hegel and Sellars’ “Myth of Jones”: Can Sellars Have More in Common with Hegel than Rorty and Brandom Suggest? Paul Redding, University of Sydney (AU)

One characteristic shared by G.W.F. Hegel and Wilfrid Sellars is that both left behind follower-­interpreters portrayed as aligned along a “right-­to-left” continuum, but one might be skeptical that there could be anything more substantive in common between them. Is not the German’s absolute idealism the antithesis of the American’s uncompromisingly scientific realism? Nevertheless, Sellars hinted at connections between his work and the philosophy of Hegel and a number of his followers have taken that hint seriously. From the Sellarsian side, the feature of his realism that leans him towards idealism is the combination of his critique of Cartesian and empiricist conceptions of the mind and the irreducible role given to social norms, especially those of linguistic communication, within an otherwise scientific realist ontology.1 From the Hegelian side, recent interpreters who take Hegel’s idealism as free of any commitment to cosmic minds and the like, and who stress Hegel’s rejection of the very idea of some “Platonic realm” that transcends the concrete spatio-­ temporal world, nudge Hegel in the direction of Sellarsian naturalism and realism. In this chapter I explore some features shared by Sellars and Hegel in an area where philosophies of mind and language intersect, arguing for a contrary account to that put forward in their names by Richard Rorty and Robert Brandom—two Sellarsians perhaps most associated with the Sellars–Hegel parallel. In particular, I examine Sellars’ thesis of psychological nominalism, which informs his account of human mindedness found in the closing sections of Empiricism and the Philosophy of Mind, the myth concerning the radical linguistic innovator “Jones.” According to psychological nominalism, “all awareness of sorts, resemblances, facts, etc., in short, all awareness of abstract

42

Wilfrid Sellars, Idealism, and Realism

entities—indeed, all awareness even of particulars—is a linguistic affair” (EPM, §29, 63). While Brandom, following Rorty and Dennett, interprets this as a form of psychological anti-­realism, I argue that Sellars’ own account is compatible with the kind of psychological realism found in Hegel. Psychological nominalism plays an important role in Sellars’ critique of the “Myth of the Given,” which is commonly compared to Hegel’s critique of the stance of “sense-­certainty” at the outset of The Phenomenology of Spirit (Hegel 1977, ch. 1). As with Hegel’s critique, the “givens” targeted by Sellars are purported particular entities (traditional “sense impressions” or Russellian “sense-­data,” for example) thought by empiricists to be immediately “given” in experience.2 For empiricists, the mere experience of such entities is thought to be sufficient for the subject’s knowledge of them, and this knowledge is then thought of as the basis upon which all further knowledge is to be built. But any attempt to account for how knowledge as a whole can be derived from these particular instances of knowledge must assume that such givens are repeatable—that is, within the stream of experience a subject is able to have and recognize “givens” of the same kind and differentiate them from givens of a different kind. For example, repeatedly having impressions or sense-­data will lead to sorting these ones as red, those as blue, and so on. Against this common approach, psychological nominalism holds that the mental capacities assumed by empiricists are not primitive but presuppose the capacity for language. The learning of language could not then presuppose such psychological capacities, and the “Myth of Jones” is introduced to give us a sense of how a type of language could be learned in which “inner episodes,” including impressions, come to be ascribed to speakers without those speakers having a prior knowledge of such episodes. As we will see, Sellars’ mythical Jones had extended the linguistic resources of his community—a community of “neoRyleans.” As the name suggests, these neo-Ryleans had basically behavioristic attitudes toward others and themselves. More specifically, they had the linguistic resources to describe themselves in behavioral terms, but had no way of talking about such behavior as expressive of or resulting from “inner episodes” such as those we might think of as thoughts about or awarenesses of worldly things or states of affairs. Jones’ linguistic innovations are described as resulting in individuals subsequently coming to describe others and, as a consequence of this, coming to describe themselves, as having such mental lives with such “inner episodes.” But this then raises difficult and controversial questions, both interpretative and substantive. Are we to think that with the capacity to describe themselves and others as having mental lives, the neo-Ryleans thereby come to actually have such lives?

Hegel and Sellars’ “Myth of Jones”

43

To answer “yes” here is to endorse a type of psychological realism, albeit a type that is compatible with Sellars’ psychological nominalism. Hegel, I suggest, with his “recognitive” account of consciousness and self-­consciousness (Redding 2008), should be understood a psychological realist of this kind, and I would argue, so should Sellars.3 In contrast, Rorty was, we will see, a clear critic of psychological realism, and had argued early in his career against the idea that there was anything such as the “intentional content” or “phenomenal character” of mental states that played the role of a distinctive “mark of the mental.” For Rorty, I suggest, there could be no further point to be made about the consequences of Jones’ linguistic revolution: we remain “neo-Ryleans” with the added capacity to talk about others and ourselves as having “inner lives.” While Brandom is not as explicit here, I suggest the same holds for him as well. For example Rorty endorses Daniel Dennett’s theory of the “intentional stance” (Rorty 1979, 236n), according to which the attribution of intentional states to others, such as beliefs and desires, is merely a strategy adopted to make their behavior more predictable. We treat others as if they had such states to simplify our dealings with them. Brandom also endorses Dennett’s “stance stance” toward intentionality (Brandom 1994, 55–62), arguing that “the community members’ practical attitudes institute normative statuses and confer intentional content on them” (Brandom 1994, 61). It would seem that there can be really nothing more to the having of inner lives than being treated by others as having them—being the bearers of ‘normative statuses’ conferred by others. But Sellars seems to have been reluctant to abandon many of the traditional marks of mindedness, including the phenomenal dimensions of consciousness, that are happily shed in the psychologically antirealist “stance stance.” In relation to the mind’s phenomenal states, for example, such a reluctance is neatly captured in a claim about qualia that Dennett attributes to Sellars and includes as an epigram in his Consciousness Explained: “But Dan, qualia are what make life worth living!” (Dennett 1993, 383). Here I want to explore an interpretation of Sellars’ psychological nominalism that is compatible with a psychological realist stance toward the intentional and phenomenal characteristics of consciousness: one stating that we properly come to have inner, psychological lives as a consequence of having adopted the linguistically enabled intentional stance—mental lives that have a degree of independence from what others ascribe. As Rorty provides the clearest expression of a psychological anti-­realist interpretation, I will start with him, and then turn to the Rortian aspects of Brandom’s way of construing Sellars. After this, I offer a reading of Sellars’ Myth of Jones that is compatible with psychological realism, drawing on parallels between Sellars and Hegel.

44

Wilfrid Sellars, Idealism, and Realism

A Rortian world of augmented neo-Ryleans Rorty’s initial professional training had not been as an analytic philosopher: rather, as a graduate student in the 1950s his main influence seems to have been Alfred North Whitehead’s speculative philosophy. By the early 1960s, however, he was taking the “linguistic turn” of analytic philosophy seriously after discovering the work of Sellars in particular.4 Rorty was soon to become identified as the proponent of a radical “eliminative materialist” philosophy of mind, that he had seen as grounded in Sellars’ work.5 His starting point was behavioristic: a subject’s “mental states” were to be understood in terms of their dispositions to act, including dispositions to utter sentences. While the meaningfulness of such sentences had to be accounted for, this was to be conceived not because they expressed inner “thoughts,” but because they were subject to further linguistic acts, those of metalinguistic interpretation conceived as an interpreter’s translation of the subject’s utterance into their own language. Rorty’s radically eliminative materialist consequences for the mental were then drawn from this metalinguistic analysis of verbal behavior. While typically materialists had been challenged to explain various features that were taken to be fundamental to the mind—the “what it is like” of phenomenal consciousness, or the “aboutness” of intentional states, for example—Rorty argued that there was no distinctive “mark of the mental” that had to be accounted for. By the time of his 1979 Philosophy and the Mirror of Nature, the mind had simply become an “invention” that we could do without.6 Echoes of Rorty’s radical psychological antirealism can be discerned, I suggest, in Robert Brandom’s so-­called two-­ply account of perception, which he attributes to Sellars (Brandom 2002). The two “plies” to which Brandom refers concern broadly the linguistic behavior in which we talk about the world, and the particular bodily dispositions understood as underlying and recruited by such behavior. In relation to the “upper ply,” Brandom applies his Sellars-­based “inferentialist” approach to the semantic contents of assertions. Like Rorty, Brandom argues that we shouldn’t think of the meaning of an asserted sentence as contributed to by the intentional contents of the speaker’s mental state: the sentence’s meaning is wholly determined by the inferential relations that it bears to the contents of other utterances, for example, those that would be given as reasons when the assertion is challenged by others. What remains of the traditional empiricist contribution of experience can be accounted for at the level of the “lower ply” in terms of “reliable differential responsive dispositions”

Hegel and Sellars’ “Myth of Jones”

45

underlying our linguistic responses to the environment and that are “characterizable in a naturalistic physicalistic vocabulary” (Brandom 2002, 350). These reliable differential responsive dispositions (RDRDs) that underlie empirical judgments are not restricted to cognitive, or even animate, beings—an appropriately trained parrot, for example, might reliably distinguish red from blue things by appropriate squawks, and even an iron bolt can be said to “discriminate” between wet and dry environments by rusting or not rusting (Brandom 2002, 349–50). What distinguishes sapient beings is the response that unfolds on the upper normatively conceived ply, and that supplies the other part of the relevant justification. Rather than simply make noises reliably associated with features of the environment, we produce assertions with those noises, that is, place utterances into a normative “space” of inferences or reasons by entitling others to ask for reasons why they should accept and act on them, should they have doubts. Thus given the proper working of the RDRDs at the lower ply, all contributions to the properly semantic content of assertions will now derive from the normative inferential relations within which the utterances stand within this logical space. On this account, inferential relations are not only necessary for the semantic contentfulness of our utterances, they are also sufficient, and so what has thus dropped out from consideration is any idea of a contribution to the semantic contents of our words from any independently conceived cognitive states of the speaker—mental states that can be said to have some sort of “content” that is non-­ reductively derived from what others properly attribute to the speaker on the basis of what he or she says. There is no place here for the subject’s reporting on “givens” of sensory experience, and no need for it, the separation of the two plies thus marks a radical dualism between the sentience of experience and the sapience of cognition (i.e., conceptual understanding). Brandom disavows the use of the term “experience”: it is not, he says, one of his words (Brandom 2000, 205, n. 7). This is a feature of Brandom’s account warmly applauded by Rorty (2007, 9–11). With this Brandom opposes a number of commonplace assumptions such as the assumption that when we speak we express thoughts and experiences that, in some sense, exist independently of their particular expressions. Here I want to focus on one of Brandom’s counter-­intuitive claims that pertains to perceptual experience—his own example of the particle physicist who, reporting on the events happening within a cloud chamber, is said to see or “observe” sub-­atomic particles such as mu-­mesons.7 When reporting on the presence of mu-­mesons in the process of observing events within an experimental cloud chamber, the physicist is, Brandom insists,

46

Wilfrid Sellars, Idealism, and Realism

“directly observing mu-­mesons . . . rather than indirectly, inferentially coming to a conclusion about mu-­mesons on the basis of an inference . . . from the presence of a vapor trail with a certain shape” (Brandom 1994, 223–4). For Brandom, mu-­ mesons are observable because to be observable something has simply to be the subject of non-­inferential reports (Brandom 2002, 363). Many find Brandom’s account here counter-­intuitive (e.g., Apel et al, 2008), denying that sub-­atomic particles can be observed. Indeed, Sellars himself had expressed this latter view with an example involving the vapor trails of an overhead jet. There Sellars argued that in cases where an observer saw the vapor trails only, one could ascribe to that observer a type of de dicto “seeing that”— here, “seeing that the plane is overhead,” while it would not be correct to ascribe to them the de re seeing of the plane itself (Sellars ME, 26).8 Among the intuitions to which critics of Brandom’s position might appeal are those that avert to the experience of perceiving. For example, one might hold that it makes sense to ask of someone who had reported seeing or observing a jet overhead, “What did it look like?” This is, surely, a question that can’t be answered in the case of Sellars’ unobserved jet or Brandom’s purportedly observed mu-­meson. It can be answered, however, with regard to the vapor trails in both cases. To sharpen what is at stake here we might contrast Brandom’s approach to perception with the seemingly more intuitive approach of Tyler Burge in his discussion of “de re” intentional contents (Burge 2007, 44–81).9 Quine (1956) had argued that de dicto attitudes are basic, and that de re attitudes are somehow derived from them. A similar view is found in Rorty (2014, 99), and Brandom (1994, 502). This analysis fits with Brandom’s claim that the physicist sees the mu-­ meson, because “S sees the mu-­meson” will be understood as derived from “S sees that a mu-­meson is present.” Burge rejects the Quinean analysis, however, arguing that for perceptual beliefs, “de re belief is in important ways more fundamental than the de dicto variety” (Burge 2007, 44). And given that Sellars believes that one can see (de dicto) that a jet is overhead without seeing the jet, it would seem that he too rejected the Quinean analysis.

Hegel on perceptual and reflective judgments How, then, do things stand with Hegel? Here Hegel is, I suggest, on the side of Burge and Sellars. In his account of judgment in volume two of The Science of Logic, Hegel distinguishes judgments of determinate being (Urteile des Dasein) from judgments of reflection (Urteile der Reflexion, Hegel 2010, 557 and 570).

Hegel and Sellars’ “Myth of Jones”

47

Hegel characterizes the distinction in terms of the different relation between subject and predicate in each: in judgments of determinate being, the predicate is said to inhere in the subject, while in judgments of reflection the subject is said to be subsumed by the predicate. Clearly the former are meant to be taken as judgments about specific objects immediately available to perception, and Hegel takes the predicate involved to refer to the specific property-­instance in the way a name refers to a single object.10 In the example “the rose is red,” we are clearly meant to think of the subject term as referring to some specific rose, and the predicate to its particular way of being red.11 In contrast, the predicate “red” in the context of judgments of reflection is clearly an “abstract” classifying one. In the reflective case, the predicate “subsumes” the particular rose into a class of objects that would include post-­boxes and London buses, of which “red” can be truly said. The relation between the two forms of predication clearly maps onto the subject matters of chapters 2 and 3 of Hegel’s Phenomenology of Spirit. Chapter 1, “Sense-­certainty” had shown Hegel as a proto-Sellarsian critic of the modern Myth of the Given, the theory of epistemic acquaintance with bare thesis of sense-­certainty is seen to collapse under its own internal contradictions and gives rise to the model of perception, in which the original qualitative simples have become reconceived as contingent properties of publically perceptible individual instances of kinds—“this suches” (deVries 2008). The new ontological relation here is that of the inherence of property instances in substances modelled by judgments of determinate being, but this new model of perception will in turn similarly collapse, the process giving rise to a new cognitive attitude—“the understanding,” explored in chapter 3—an attitude that posits, rather than directly perceives, underlying forces involved in the production of any perceptual episode (Redding 2010–11). Clearly it is the subsuming, reflective judgment that models this ontology. In both phenomenological and logical treatments, the succeeding position will be said to “aufheben”—negate but preserve—the position it succeeds, but this “Aufhebung” will be reversed in the subsequent move, and this means that neither position can be simply reduced to or explained in terms of the other. For example, in Hegel’s Logic the de dicto judgments of reflection that succeed de re judgments of determinate being will themselves be aufgehoben by the succeeding form of judgment, the judgment of necessity. However, the first form of the judgment of necessity repeats the de re character of the earlier judgments of determinate being, with this de re character now manifests at a higher level. The “res” of the judgment of necessity is no longer an individual thing, such as an

48

Wilfrid Sellars, Idealism, and Realism

individual rose, but the genus or secondary substance, such as the kind “rose” or the “rose as such” (Hegel 2010, 575–6). In sum, while Hegel clearly is a critic of the modern Myth of the Given, in his account “the given” undergoes a different fate to the one it undergoes in both Rorty and Brandom. It is not the case that the ascribed de re object of perception is unilaterally explained in terms of de dicto ascription. I have suggested that by his example of the jet and its vapor trail, Sellars also resists Brandom’s reductive account of object perception. Moreover, other aspects of Sellars’ work also seem to appeal to a mediated dualism between de re and de dicto accounts of intentionality. For example, in distinguishing the manifest and scientific pictures of the world (PSIM), Sellars seems to construe the former as a view of the world made up of the type of objects typical of perception as commonly understood. As O’Shea puts it, the manifest world is a “world as conceived by common sense in terms of manifest sense-­perceptive properties—the colors and shapes . . . of ordinary persisting physical objects . . . as opposed to the often strange and colorless scientifically postulated world of swarming microphysical atoms and subatomic particles that is imperceptible to our unaided senses” (O’Shea 2007, 13).12 Such a non-­reductive attitude to object or “de re” perception will become apparent in what Sellars presents as the perceptual theory of Jones in the Myth of Jones.

Sellars on thoughts, impressions, and their modeling Brandom’s example of the “observing” of mu-­mesons seems to make the experiential dimension of observation redundant, but part of the motivation of the introduction of Sellars’ Myth of Jones in the concluding parts of Empiricism and the Philosophy of Mind seems precisely to address questions of the sorts of ways that similarities among experiences can lead to incorrect perceptual judgments. Thus, in section X, entitled “Private Episodes: the Problem,” the problem in question is described as that of explaining the similarity among three types of experience that can be ascribed to a subject: E1, the experiences “of seeing that an object over there is red”, E2, “its looking to one that an object over there is red (when in point of fact it is not red)”, and E3, “its looking to one as though there were a red object over there (when in fact there is nothing over there at all)”. EPM § 45, 85

Hegel and Sellars’ “Myth of Jones”

49

This was the sort of problem that the now discredited “given” had been meant to address. For example, the sense-­datum theorist would answer that the similarity was explained by the fact that in the three cases similar sense-­data were perceived. Sellars’ critique of the given, found in the first part of EPM, had been meant to show why that was an inadequate explanation. For his part Brandom will treat attributions like E2 and E3 merely negatively, as the type of expressions that simply indicate the withholding of the endorsement by the attributor of the content endorsed in the expression of E1. Such sentences can be thought of as descriptive of how things “look” only in the sense of their functioning as non-­ inferential reports without significant experiential dimensions, as do reports about mu-­mesons.13 But for Sellars, this alone does not seem to completely remove the philosophical problem. The problem of explaining the similarity among E1, E2, and E3 remains, and part of the Myth of Jones is meant to provide an alternative explanation rather than show that there is nothing further in need of explanation. Sellars thus now revisits possible explanatory alternatives to the idea of the given as a discovered “component” contained within these three forms of experience. Earlier, he had broached the alternative possibility of treating “impressions or immediate experiences” not as components found in experience as conceived in the Myth of the Given, but as “theoretical entities”—the kinds of imperceivables belonging to the explanations employed in the “scientific image.” He had provisionally dismissed this alternative, but now returns to it, and the idea of impressions qua theoretical entities will be broached in the light of a related account of treating thoughts as theoretical entities. These will be parts of the story of Jones’ two-­staged linguistic innovations, but introducing the story of Jones first requires two excursuses—first, a discussion of thought according to what Sellars calls the “classical tradition” as well as its modern distortion, and next, his account of the role of models in theoretical explanation. According to Sellars, in the “classical tradition” stemming from Aristotle thoughts are conceived as a family of mental or “inner” episodes that are different to both overt verbal behavior and what he calls “verbal imagery,” a type of internalized talking to oneself. Moreover, both overt verbal behavior and “verbal imagery” owe their meaningfulness to the fact that they express thoughts. According to this tradition, as he describes it, while thoughts are introspectable in the sense that a thinker has some sort of privileged access to them, this is not to be thought of as some perceptual scrutiny of one’s own mental states. That type of picture is characteristic of the early modern period and results from a distortion of the classical account, in which the thoughts of the classical account

50

Wilfrid Sellars, Idealism, and Realism

are conflated with other types of “inner episodes” such as sensations and impressions. To understand Sellars’ approach here we need to understand his account of the role of models in theoretical explanation more generally. Sellars presents his distinctive view of the role of models in theoretical explanation as contrasting with positivistic accounts of the relations of theoretical and observational languages. His alternative, which he believes better captures the actual practices of science, involves the use of models that “describe a domain of familiar objects behaving in familiar ways such that we can now see how the phenomena to be explained would arise if they consisted of this sort of thing. The essential thing about a model is that it is accompanied, so to speak, by a commentary which qualifies or limits—but not precisely, nor in all respects—the analogy between the familiar objects and the entities which are being introduced by the theory” (EPM § 51, 96). Focusing on the role of models in theory, he thinks, in turn allows us to understand the continuity between scientific theory construction and everyday commonsensical forms of explanation, and this will be crucial for his telling of the Myth of Jones. Once more I want to appeal to the work of Tyler Burge to help bring into focus an important feature of what, on Sellars’ account, is the position of the “classical tradition.” Burge, too, discusses a classical or “traditional” view of concepts and thoughts when discussing thoughts and their conceptual parts, a view that he traces back to Aristotle (Burge 2007, 291–4). According to Burge, thoughts on this account are taken to be “intrinsically representational” in that (1) their representational properties are used to explain the representational properties of the sentences used to express them, and (2) thoughts differ from the derivatively representational sentences that express them in that while linguistic entities have properties other than representational ones, thought has only representational properties. Thoughts so conceived would have no features allowing them to be identified and individuated independently of what they are about. However, such “transparent” thoughts are just what we would expect were we, following Sellars, first, to model thoughts on sentences, and next, to provide a commentary stating that only representational features of the sentences are to be counted as relevant to the thoughts being modeled by those sentences. Burge’s account clarifies why conflating thoughts and sensations produces a distortion of the classical view, as impressions, sensations and so forth, clearly do have non-­representational properties, that is, properties that can be appealed to when discussing such episodes in abstraction from their representational function. Actual words, of course, have non-­representational properties—the words “goat” and “boat” being phonetically similar in ways not reflected by

Hegel and Sellars’ “Myth of Jones”

51

similarities between goats and boats. Might not conscious mental contents be thought of as related in an analogous way? With this thought, we might now see how phenomenal similarities of this sort could be at issue when considering what is in common between the experiences involved in “seeing that,”“its looking that” and “its looking as though.” This is because what is being alluded to here is something common to the three experiences despite the differences in the semantic contents represented in those experiences. This in turn sheds light on the “de re”/“de dicto” distinction. If we think of holding a general belief as an attitude towards some entirely de dicto content, as in Hegel’s judgments of reflection, then we will think of this propositional content as itself possessing no intrinsic non-­representational properties. Thought of in this way, a proposition is modelled on a sentence, but with the non-­representational features of the sentences discounted. For example, the belief that all London buses are red abstracts away from the particular way any actual London bus exemplifies redness. It would indeed be a belief of which a red-­green color-­blind person would be perfectly capable. But a different type of modelling would be involved in explaining what was similar and different among the experiences E1, E2, and E3. When I see that that particular bus over there is red, my experience is such that there is something similar in it to the experience in which it looks to me that that (actually pink) bus over there is red (because of a trick of the light) or the experience of its looking to me as though there is a red bus over there (because I’m hallucinating). Only a mental state with a de re content, and thus possessing phenomenal features, could account for our ability to make this set of comparisons.

Jones’ theory of thoughts and perceptual thoughts We will remember that Jones belongs to a “neo-Rylean” linguistic community that has the linguistic resources for talking about the behavior of individuals but not about any states of mind that could be conceived as leading to or given expression by that behavior. They can say things like “Dick is making a cup of coffee” but not things like “Dick wants a cup of coffee.” The neo-Ryleans thus lack the capacity for thought as given in the classical account, and the first phase of Jones’ cultural revolution is going to enable them to acquire that capacity. In the myth, the linguistic resources of the neo-Ryleans are quite extensive and include the capacity for giving postulational accounts of worldly phenomena, and what Jones does is to extend such a postulational account to human behavior itself,

52

Wilfrid Sellars, Idealism, and Realism

and he does so by inventing a model. Precisely, human speech itself will be treated as a model for some “inner episodes”—thoughts—that will be adduced as that which explains the behavior of others, including their verbal behavior. Jones thus introduces Dennett’s “stance, stance.” In the myth, Jones’ theory catches on, and members of the community start talking about each other in this way, ascribing thoughts to their fellows in attempts to explain their actions. Moreover, in having learned to ascribe thoughts to others they soon learn to ascribe thoughts to themselves, and so come to not only think of themselves as having these inner episodes, but actually recognize them in themselves. Jones, for example, can move from uttering a sentence such as “Dick is thinking that p” to uttering one such as “I am thinking that p.” We might now see how, at this new stage of cultural development, the neoRyleans have come to have something approximating the “classical” view. As we will remember, in its undistorted form, the classical view has not been sullied with the conflation of thoughts with non-­representational properties such as is found in impressions and so on. We might now think of this classical view as a consequence of Jones’ use of the interpretative penumbra that must surround the use of a model. Jones can specify that thoughts are like verbal utterances in as much as they have representational properties,14 but that they are unlike verbal utterances in having no non-representational properties: “Thus, while his theory talks of ‘inner speech’, the commentary hastens to add that, of course, the episodes in question are not the wagging of a hidden tongue, nor are any sounds produced by this ‘inner speech’ ” (EPM §57, 104). Jones’ theoretical–linguistic innovations do not stop here, however, and in the second phase of the revolution Jones now employs a second model with which to attribute to people another class of “inner episodes”—the impressions, sensations and so on, which, on the unsullied classical view, had been kept distinct from thoughts. Whereas Jones had modeled thoughts on verbal occurrences, “this time the model is the idea of a domain of ‘inner replicas’ which, when brought about in standard conditions, share the perceptible characteristics of their physical source.” But Sellars insists on two things: what the theory introduces are states of the perceiver, not particulars; and “that the model is the occurrence ‘in’ perceivers of replicas, not of perceivings of replicas” (EPM §61, 110). Sellars’ account of this phase of the Jonesian revolution is complex, and his discussion here of “replicas” as models is, admittedly, confusing. Indeed, he seems to prevaricate over exactly what plays the role of “model” in this case. Models consist of “familiar objects behaving in familiar ways” (EPM §51, 96), but this

Hegel and Sellars’ “Myth of Jones”

53

hardly characterizes replicas conceived of as in perceivers. This general sense of “model” is in line with what, in the context of the present discussion, Sellars goes on to call “model entities,” for which he gives as an example red and triangular wafers—clearly the sort of ‘familiar objects’ that make up models in the general sense. Nevertheless, Sellars explicitly calls the replicas themselves “entities of the model.” One might think that the “replicas” qua models are themselves modeled on perceptible “familiar objects” like wafers of determinate shape and color. The picture involved, then, seems to suggest a two-­tiered relation of modeling: red triangular wafers providing models for their replicas, and the replicas in turn providing models for the theoretically posited “impressions.”15 Sellars’ insistence that the model is the “occurrence ‘in’ perceivers of replicas, not of perceivings of replicas” (EPM §61, 110) suggests that “replica” here not be thought of as having the type of representational significance that is carried by a notion like “picture” or “image,” which are often thought of as replicas.16 For example, he notes that taking the model to be “seeing a red and triangular replica . . . smuggles into the language of impressions the logic of the language of thoughts” (EPM §61,100), suggesting the tendency to import the representational significance of the latter logic into that of the former, which is properly nonrepresentational. In the case of thoughts, I earlier suggested that Sellars had been concerned with using the “commentary” to discount all non-representational properties of the model entities—uttered sentences. Now, I suggest that the inverse is the case for the modeling of impressions. In the modeling of impressions, only the non-representational properties of the model—what Sellars refers to as intrinsic properties—are relevant. In contrast to Brandom’s reading, this will leave a place for phenomenal experience in Sellars’ account of impressions. It will play a role analogous to that played by phenomenal experience in the perception of particularly qualified things.17 In perceptual thought, we should be left with something analogous to what had been excluded in Brandom’s and Rorty’s account of thoughts—some analogue to the feel of the wagging of the tongue and the sound of the “inner words.” This type of inverse symmetry between thoughts and impressions suggested here is, of course, familiar: it is analogous to the structure of Kant’s original account of the different logics of concepts and intuitions. Of course Kant’s idea of an empirical intuition might itself be taken as exemplifying the Myth of the Given,18 but with his insistence on the different logics of impressions and thoughts, Sellars seems to want to hold onto more of Kant’s original distinction than is found in Brandom. As deVries (2005, ch. 8) and O’Shea (2008, chs. 4 and 5) stress, Sellars had clearly been committed to finding a non-­particularizing

54

Wilfrid Sellars, Idealism, and Realism

successor notion to the mythical givens, and we have seen something of Hegel’s way of addressing the same issue. In Hegel, Kant’s intuition–concept distinction had been transformed into a distinction between judgment forms characterized by two different conceptions of predication. One of these, we might say, provides for de re judgments rich in determinate phenomenal content, while the other provides for de dicto judgments rich in abstractly mediated classificatory relations. As in Kant’s intuition–concept distinction, Hegel’s different judgment forms are complementary and reciprocally dependent, but unlike Kant’s distinction, Hegel does not freeze this distinction into a fixed dichotomy, but treats each judgment form as somehow able to be transformed or translated into the other in a series of “Aufhebungen.” I suggest that we bring this feature of Hegel’s account to bear on Sellars’ theory concerning the relation of one’s attribution of perceptual thoughts to others on the one hand, and to oneself, on the other. Brandom and Rorty seem to see no real difference here; Sellars and Hegel, I suggest, see a crucial difference. As we have seen, attributing thoughts to others’ works on the basis of a model in which thoughts are conceived as analogous to utterances. Here the de dicto mode of attribution captures this directly: what one attributes to others has the shape of the type of sentence that they would be likely to utter in making assertions. But this has to work in tandem with the attribution of distinctly perceptual thoughts in which impressions, with their different logic, are seen to have a role. For Alice to attribute to Brett perceptual thoughts about things that they can both perceive, she theorises a role for “impressions” in a causal story leading to Brett’s assertions. Theoretically, these impressions are to be organized according to isomorphisms that Alice bases on the properties of everyday objects of which she is directly aware. “The essential feature of the analogy is that visual impressions stand to one another in a system of ways of resembling and differing which is structurally similar to the ways in which the colours and shapes of visible objects resemble and differ” (EPM §61, 112). But when she comes to extend the attribution of “impressions” to herself it seems odd to think that she would self-­attribute them in just the way as she attributes them to Brett. This is because she has had to draw upon her own perception of everyday objects to model the type of isomophically conceived impressions that she attributes to Brett. Why wouldn’t she then attribute to herself impressions that “share the perceptible characteristics of their physical source” in a more direct way that the isomorphism story—a way that conceived the state involved more on the model of a replica of the physical source than as a sentence about it? It is here that Hegel’s idea of the de re judgment of determinate existence looks helpful, as it has both features. It has the logic of a type of sentence, but it is a type

Hegel and Sellars’ “Myth of Jones”

55

of sentence whose subject-­predicate structure is modeled on the familiar objects of the manifest world, objects understood as instances of kinds in which various non-­essential properties “inhere.” So while Alice will attribute perceptual thoughts to Brett within the framework of an overarching de dicto mode of attribution, in “extending” the ascription if impressions to herself, might this not be done within a framework for which the de re mode of expression was taken as primary? But, on the Hegelian model this difference should not be considered rigid and fixed, because Alice has to be able to conceive of herself and Brett as having perceptual thoughts with the same content, and this is what Hegel’s idea of the mutual translatability of de re and de dicto forms is meant to capture. We have learnt from Brandom how Alice can translate the de dicto content of Brett’s perceptual thoughts into de re ones by some type of Quinean derivation, but on the reading suggested here, the translation of her own states should be seen as moving in the inverse direction. For her the relevant transformation will be a matter of going from an expression that captures her seeing of the wafer—say, that it is red and triangular—to one that expresses a judgment that the wafer is red and triangular. On this reading, we cannot discount the existence of genuinely psychological de re intentional states to which the de re attributions of others are, in some sense, descriptive—states, moreover, that have a distinctive phenomenology. But this is a form of psychological realism, I suggest, that is at variance with the “stance stance” approach of Rorty and Brandom. In Brandom’s Rortarian telling of Myth of Jones, Alice and Brett, like the rest of us who live downstream of Jones’ revolution, are creatures who have learned to talk about themselves as if they have inner mental lives. On this more “Hegelian” reading of Sellars offered here, that we have become creatures who talk about ourselves as having inner lives has actually brought it about that we have come to have such lives.19

Notes 1 According to Robert Brandom, it was Richard Rorty who distinguished right-­wing from left-­wing Sellarsians on the basis of their leaning towards the normative or the naturalist dimensions of Sellars’ position (Brandom 2015, 31). 2 For a concise account see deVries 2005, 98–103. 3 I leave out a third alternative here: that having the inner lives we have is a characteristic possessed independent of the development of language. This is the thesis of original intentionality as conceived by Roderick Chisholm. On Sellars’ dispute with Chisholm see O’Shea 2008, 87–8.

56

Wilfrid Sellars, Idealism, and Realism

4 For us, a crucial paper is “The Subjective Principle and the Linguistic Turn” from 1963, contrasting Whitehead and Sellars. Rorty 2014, ch. 4. 5 See especially the essays making up chapters 6, 8 and 10 of Rorty 2014. 6 Thus the opening chapters are entitled “The Invention of the Mind” and “Persons without Minds.” 7 Mu-­mesons (or ‘muons’) are short-­lived sub-­atomic particles belonging to the genus “lepton.” 8 A so-­called “de re” content is a content of an intentional state thought of as directed to a particular, specific object or res. In contrast, a “de dicto” content is the content of a mental state thought of as directed to a proposition, conceived on the model of a saying or “dictum.” For example, one often talks about perception in “de re” or “objectual” ways, as when one says one saw Alice’s new car, and talks about belief in “de dicto” ways, as when one says that one believes that the earth is heating dangerously. 9 In Burge’s usage it is important that the “res” is specific or definite and not simply particular. For example, to have a thought about “the shortest spy” is not to have a properly de re thought if one doesn’t know who the shortest spy is. 10 Thus the predicate here is said to instantiate the category of singularity, and the subject universality, Hegel here effectively reversing the subject–predicate relation. 11 This is congruent with the singularity of the predicate. 12 For Brandom’s unhappiness with this distinction see Brandom 2015, ch. 1, pt. II. In contrast, some non-­reductive “fusing” of the two images, more in line with Hegel’s “Aufhebung” is urged by “centrist” Sellarsians such as O’Shea (2008). 13 This indeed leads Brandom to talk of the observability of thoughts and sense-­ impressions (Brandom 2015, 60). 14 “Thus, just as Jones has, like his fellows, been speaking of overt utterances as meaning this or that, or being about this or that, so he now speaks of these inner episodes as meaning this or that, or being about this or that” (EPM §57, 103). 15 In the earlier case of thoughts, it is “inner speech” that plays a role analogous to “replicas” here. 16 In fact, this is just how Brandom seems to interpret Sellars in the corresponding sections of his “Study Guide” to EPM (Brandom 1997). 17 Perhaps Sellars’ deliberate two-­step account of the model, from model entity to replica and from replica to impression is meant to capture the idea that there are two distinct sets of conditions, “commentaries,” defining which features of the model are included or excluded as relevant. Sellars mentions using the commentary to discount that impressions are particular objects, like the replicas (and the model objects), but there is the other implicit commentary noted above, that any representational properties are also discounted. 18 Thus Russell had thought of his sense-­data as similar to Kant’s intuitions.

Hegel and Sellars’ “Myth of Jones”

57

19 I am indebted to helpful comments and criticisms on an earlier draft from Bill deVries and Patrick Reider. This work was carried out with the aid of a Discovery Grant from the Australian Research Council, DP130102346.

Bibliography Apel, J., S. Bahrenberg, C. Köhne, B. Prien, and C. Suhm (2008), “Of μ–Mesons and Oranges: Scrutinizing Brandom’s Concept of Observability,” in B. Prien and D.P. Schweikard (eds.), Robert Brandom: Analytic Pragmatist, Frankfurt: Ontos Verlag. Brandom, Robert B. (1994), Making It Explicit: Reasoning, Representing, and Discursive Commitment, Cambridge, MA: Harvard University Press. —— (1997), “Study guide to ‘Empiricism and the Philosophy of Mind’,” in W. Sellars, Empiricism and the Philosophy of Mind: with an introduction by R. Rorty and a study guide by R. Brandom, Cambridge MA: Harvard University Press. —— (2000), Articulating Reasons, Cambridge, MA: Harvard University Press. —— (2002), “The Centrality of Sellars’ Two-Ply Account of Observation to the Arguments of ‘Empiricism and the Philosophy of Mind’,” in Tales of the Mighty Dead: Historical Essays in the Metaphysics of Intentionality (Cambridge, MA: Harvard University Press), reprinted in Brandom 2015. —— (2015), From Empiricism to Expressivism: Brandom Reads Sellars, Cambridge MA: Harvard University Press. Burge, T. (2007), Foundations of Mind: Philosophical Essays, vol. 2, Oxford: Clarendon Press. Dennett, D.C. (1993), Consciousness Explained, London: Penguin. deVries, W.A. (2005), Wilfrid Sellars, Chesham: Acumen Publishing. —— (2008), “Sense-Certainty and the ‘This-Such’,” in D. Moyar and M. Quante (eds.), Hegel’s “Phenomenology of Spirit”: A Critical Guide, Cambridge: Cambridge University Press. Hegel, G.W.F. (1977), Phenomenology of Spirit, trans. A.V. Miller, Oxford: Oxford University Press. —— (2010), The Science of Logic, ed. and trans. G. di Giovanni, Cambridge: Cambridge University Press. O’Shea, J.R. (2008), Wilfrid Sellars: Naturalism with a Normative Turn, Cambridge: Polity Press. Quine, W.V. (1956), “Quantifiers and Propositional Attitudes,” The Journal of Philosophy, vol. 53, 5: 177–87. Redding, P. (2008), “The Independence and Dependence of Self-Consciousness: The Dialectic of Lord and Bondsman in Hegel’s Phenomenology of Spirit,” in F. Beiser (ed.), The New Cambridge Companion to Hegel and Nineteenth Century Philosophy, Cambridge: Cambridge University Press, 2008, 94–110.

58

Wilfrid Sellars, Idealism, and Realism

—— (2010–11), “Hegel’s Anticipation of the Early History of Analytic Philosophy,” The Owl of Minerva, vol. 42:1–2: 18–40. Rorty, R. (1979), Philosophy and the Mirror of Nature, Princeton: Princeton University Press. —— (2007), Philosophy as Cultural Politics: Philosophical Papers, Volume 4, Cambridge: Cambridge University Press. —— (2014), Mind, Language and Metaphilosophy: Early Philosophical Papers, Cambridge: Cambridge University Press. Sellars, W.   EPM —— (1997), Empiricism and the Philosophy of Mind, with an introduction by  R. Rorty and a study guide by R. Brandom, Cambridge, MA: Harvard University Press.   ME —— (1989), The Metaphysics of Epistemology: Lectures by Wilfrid Sellars, ed.   P.V. Amaral, Atascadero: Ridgeview Publishing.   PSIM  —— (1963), “Philosophy and the Scientific Image of Man,” in Science,   Perception and Reality, Atascadero: Hackett, 8–42.

3

The Metaphysics of Sensation: Psychological Nominalism and the Reality of Consciousness Ray Brassier, American University of Beirut (LE)

For philosophers of a realist stamp, psychological nominalism, understood as the claim that all awareness is a linguistic affair (EPM 160), is of a piece with philosophical behaviorism, which allegedly denies the reality of consciousness. If psychological nominalism is complicit with this denial—which strikes many philosophers as absurd—some would say it deserves to be relegated to the dustbin of philosophical history along with its behaviorist sibling. Wilfrid Sellars first coined the expression “psychological nominalism” and defended the doctrine throughout his long philosophical career. But he also distinguished between two aspects of the mind, thinking and sensing, and laid claim to realism about both. This distinction between thinking and sensing (which can be traced back to Kant) essentially prefigures David Chalmers’s more recent distinction between the functional-­psychological and the phenomenal-­ experiential aspects of mind (see Chalmers 1995 and 1996). The crucial difference, however, is that while Chalmers endows phenomenal-­experiential states with a cognitive authority equal to if not greater than that of functional-­ psychological states, Sellars reserves cognition for functionally characterized thinking alone and characterizes sensation as a non-­cognitive state that plays a causal but not justificatory role in empirical knowledge. Yet Sellars cannot be accused of downplaying the significance of the phenomenal-­experiential aspect of mind. Indeed, his account of sensory consciousness leads him to make the controversial claim that sensation has a metaphysical purchase insofar as it (indirectly) reveals a fundamental aspect of physical reality: its intrinsic, qualitative aspects, which, as he sees it, cannot be reduced to the extrinsic or dispositional properties of things. Sellars’ Rortyan heirs have made much of his psychological nominalism, but tend to dismiss his suggestion that sensation has a metaphysical purchase as an

60

Wilfrid Sellars, Idealism, and Realism

aberrant relapse into dogmatic rationalism. I want to argue that Sellars’ account of the metaphysics of sensation should not be dismissed as a dogmatic regression; rather, it offers a way of resolving the deadlock between realists whose equation of consciousness with knowledge leads them to embrace dualism and anti-­ realists whose reduction of consciousness to conceptual awareness threatens to sever the causal link between mind and nature.

The transcendental difference and the rejection of givenness Thinking and sensing are both modes of awareness. How can Sellars insist that all awareness is a linguistic affair and still claim to be a realist about both thinking and sensing? To answer this question we need to understand what “realism” means for Sellars. Sellars is a transcendental philosopher and the transcendental perspective alters the conditions in terms of which the issue of realism is framed. Transcendental philosophy can be contrasted with dogmatic rationalism on the one hand and skeptical empiricism on the other. The dogmatic rationalist assumes that the mind enjoys a priori cognitive access to mind-­independent reality and believes that reason can deduce the fundamental features of that reality. The skeptical empiricist, for her part, insists that all knowledge is rooted in but limited by sensory experience and denies that reason can deliver a priori knowledge of mind-­independent reality. The transcendental philosopher rejects both dogmatism and skepticism. She rejects the rationalist’s assumption that the mind enjoys a priori cognitive access to reality just as she rejects the empiricist’s claim that all knowledge derives from sensory experience. Both dogmatism and empiricist skepticism remain beholden to what Sellars calls “the framework of givenness”1: dogmatism because it takes the correlation between thinking and being as given; skepticism, because it takes the intelligibility of sensory experience as given. At the heart of the framework of givenness is the assumption, common to empiricism and rationalism, that mental states are self-­intimating. Rejecting the framework of givenness, Sellars refuses the assumption that the mental is self-­ intimating. This means that minds do not necessarily know themselves. There is a fundamental difference between thinking and knowing what is thought. By the same token, there is a fundamental difference between sensing and knowing what is sensed. The awareness of something is not the awareness of something as something. This difference—between thinking and thought, or sensing and sensed—follows from the rejection of givenness. The core feature of the

The Metaphysics of Sensation

61

framework of givenness is the premise that the fusion of thinking and thought or of sensing and sensed is guaranteed, either by intellectual intuition in the case of rationalism or sensory intuition in the case of empiricism.

Awareness and thought Sellars’ insistence that all awareness is a linguistic affair is a transcendental claim, not an empirical or metaphysical one. The awareness in question is cognitive awareness, i.e. thinking. It is thinking as cognitive awareness that is a linguistic affair. Sellars defines psychological nomimalism as follows: “[A]ll awareness of sorts, resemblances, facts, etc., in short all awareness of abstract entities—indeed all awareness even of particulars—is a linguistic affair” (EPM, 160). Cognitive awareness is a linguistic affair because it involves being aware of something as something. Such awareness requires conceptualization and concepts are linguistically instantiated rules. We would not be aware of ourselves as thinking were it not for language. Once one rejects the premise that thoughts are essentially self-­intimating, or the assumption that to think is to know that one is thinking, it becomes possible to understand the concept of thinking as modeled on the concept of speaking. Sellars’ “Myth of Jones” is a philosophical reconstruction of the process through which minded beings come to recognize themselves as minded.2 Thoughts are understood as inner episodes modeled upon overt public utterances. These inner episodes are theoretical entities postulated as structurally analogous to linguistic utterances, yet devoid of the physical characteristics of utterances: they are silent, invisible, intangible, etc. The claim is not that thoughts are the causes of utterances or that utterances are the expressions of thought. Although thoughts are (implicitly) understood to be categorially different from linguistic utterances—principally insofar as they do not involve any of their physical characteristics, i.e. vocalizations, gestures, inscriptions, etc.—they are not confined within some withdrawn inner recess inaccessible to publicly perceptible space. They are not located anywhere at all; certainly not in speakers’ heads. They are introduced as provisionally unobservable theoretical entities, not “immaterial” entities. Jones’ theoretical innovation proceeds by identifying overt speech with thinking in the same way in which gases are identified with populations of molecules. Thus, thoughts “are ‘in’ language using animals as molecular impacts are ‘in’ gases, not as ‘ghosts’ are in ‘machines’ ” (EPM, 187). Empirical observables

62

Wilfrid Sellars, Idealism, and Realism

are identified with theoretical unobservables, without the former being the overt manifestation of the latter’s covert existence. Linguistic utterance is not the exteriorization of thought and thought is not the interiorization of utterance; rather, utterance itself is a “thinking-­out-loud” and thinking is a “silent saying.” But once our language has been enriched with these theoretical terms, they can acquire an observational role through which they can be used in perceptual reports. Thus we learn to perceive each other and ourselves as thinking (believing, hoping, fearing, wishing, etc.).

Meaning The meaningfulness of overt public utterance provides the model for understanding the meaningfulness of the thought-­episodes with which they are theoretically identified. The intentionality of thought is modeled upon the intentionality of language, where this is understood to depend upon the development of metalinguistic resources that allow us to speak about speaking. Speaking about speaking is the condition for thinking about thinking. Semantic vocabulary such as “means,” “designates,” “refers to,” “expresses,” provides us with the conceptual resources we need to think about meaning as a phenomenon exemplified by language. But the question then is whether the meaning of words derives from the meaning of thoughts or vice versa. Sellars’ claim is not that we learned how to think only after we had learned how to speak. Sellars allows for a proto-­conceptual language-­of-thought (“mentalese”) through which animals represent their environment. Mental representations are not linguistic or conceptual in Sellars’ sense, although they are endowed with propositional form, i.e. a referring and characterizing function.3 Nor is Sellars claiming that thinking depends upon language in an ontological sense.4 This is a metaphysical misinterpretation of psychological nominalism. The Ryleans in Sellars’ mythic parable certainly think but they do not think of themselves as thinking until Jones provides them with a concept of what it is to think. By embracing Jones’s theory of thoughts as inner-­episodes of saying, the Ryleans acquire the concept of thinking as what underlies meaningful utterance. They become aware of themselves as thinking. The concept of “meaning to say,” and of “intention” in the psychological and semantic sense, is ushered in with the shift from conscious to self-­conscious thought. This is precisely what Jones’s theory of thoughts as inner-­episodes is supposed to explain. But it entails that

The Metaphysics of Sensation

63

the intentionality of public language is the condition for the intentionality of thought: [T]he concept of thought essentially involves that of intentionality in the following sense. To say of a piece of verbal behavior that it is a thinking-­out-loud is to commit oneself to say of it that it means something, while to say of it specifically that it is a thinking-­out-loud that-­p, is to commit oneself to say of it that it is a piece of verbal behavior which means p. Thus, at the primary level, instead of analyzing the intentionality or aboutness of verbal behavior in terms of its expressing or being used to express classically conceived thoughts or beliefs, we should recognize that this verbal behavior is already thinking in its own right, and its intentionality or aboutness is simply the appropriateness of classifying it in terms which relate to the linguistic behavior of the group to which one belongs. LTC, 527, my emphasis–RB

The intentionality of public discourse is a matter of appropriately classifying linguistic behavior according to standards encapsulated by collectively accepted norms. These norms involve the rules of criticism, or “ought-­to-bes,” to which speakers are encouraged to conform, and the rules of action, or “ought-­to-dos,” which speakers are expected to realize. Since ought-­to-bes imply ought-­to-dos, conforming to the former is indissociable from accepting the latter. Linguistic action and linguistic behavior are both norm governed; they cannot be opposed to one another since the capacity to conform is the precondition for the ability to act. If psychological nominalism identifies thinking with verbal behavior, this is not in the same sense in which psychological behaviorism (allegedly) identifies thinking with behavioral dispositions. Verbal behavior expresses thought only insofar as thinking is understood as a kind of doing cognitively modeled on linguistic utterance. Thinking is a kind of doing because it involves inferring, which in turn involves representing—although thinking is not representing and concepts are not representations. Concepts are inferential roles and these roles are defined by networks of implication across webs of linguistic assertion. To say that intentionality is the appropriateness of classifying a linguistic expression relative to the behavior of a linguistic community is to say that intentionality is a functional characteristic. Sellars’ dot-­quoting device is the exhibition of this linguistic function. To dot-­quote a linguistic expression is to show what it does in the language in which the dot-­quoting occurs. This of course presupposes familiarity with the dot-­quoted expression’s proper function.

64

Wilfrid Sellars, Idealism, and Realism

This familiarity is the background of tacit understanding against which the exhibition of the meaning of particular expressions can stand out as an instance of explicit understanding. Exhibiting the meaning of a specific expression is only possible against the backdrop of a language and a form of life within which the use of words is necessarily embedded. But this exhibition, rather than stating, of complex rule-­governed functioning, which presupposes familiarity with the language in which the dot-­quoting device is deployed, re-­embeds language within a world of practical involvements. Thus Sellars’ psychological nominalism, and his claim that cognitive functioning is linguistically instantiated, envelops awareness within language only insofar as language itself is enveloped in a world comprising a variety of (human) practices and purposes. That this is a world whose structure is constellated around human interests, both practical and cognitive, is the necessary corollary of Sellars’ transcendental approach, which situates itself within the correlation between representing and represented and seeks to draw out its underlying conditions. Drawing out those conditions exposes the chain of nested mediations from awareness to awareness-­as, from awareness-­as to conceptual role, from conceptual role to linguistic practice, and from linguistic practice to a world of practical involvements. What is exposed through this chain of mediations is the way in which the content of any single state of awareness is bound up with a context of significance that includes a whole practical life-­world. Psychological nominalism is usually reproached for denying the reality of mental states. But if anti-­realism in this context means denying that mental states belong in the world, then this reproach is misguided, for on Sellars’ account, mental states are as real as anything else in the world of which they are a part. The fact that cognitive awareness of mental states is conceptual through and through in no way compromises their reality, since on Sellars’ transcendental account this is as true of tables, zebras, and galaxies as it is of thoughts and sensations. If the issue is one of realism, the more fitting reproach to psychological nominalism, and to the transcendental stance from which it ensues, would be that its realism, whether about mental states or non-­mental phenomena, comes at the price of an exorbitant holism which is obliged to take as given a total context of significance; i.e. a form of life in which the transcendental perspective is necessarily embedded. Why should this be so? On Sellars’ account, the “ought-­to-be”s and “ought-­todo”s which condition all our thinking and acting constitute the linguistic framework within which we live. This framework furnishes the resources for all justification and evaluation; although claims made within the framework can be

The Metaphysics of Sensation

65

assessed, criticized, or rejected, there is no higher authority or court of appeal with regard to which the framework as a whole could be judged and found wanting. As Wittgenstein put it: “You must bear in mind that the language-­ game . . . is not based on grounds. It is not reasonable (or unreasonable). It is there—like our life.” (Wittgenstein 1969 No. 554). If our linguistic framework is simply “there, like our life,” it is not susceptible of rational revision; it is just something to be accepted as it is. Sellars’ emphasis on holism—both semantic and epistemic—would seem to point towards a quietist acceptance of the form of life within which we find ourselves as the ineluctable horizon of all thinking and doing. If the rejection of the empiricist and rationalist versions of givenness ultimately entails accepting the givenness of a form of life as the ultimate horizon circumscribing reflection, this may prove too high a price. Transcendental reflection would find itself constitutively incapable of critically interrogating the legitimacy of the practices—and specifically of the cognitive practices—within which psychological states and conceptual roles are necessarily embedded. Sellars’ conception of philosophy as a theoretical endeavor requiring conceptual construction as well as conceptual revision stands opposed to such quietism. The root of this opposition lies in Sellars’ bifurcated holism. Our understanding of the world is split into two competing conceptual structures, the manifest and scientific images of man-­in-the-­world.5 Both are autonomous categorial schemes, neither of which can be wholly subsumed within the other. Moreover, Sellars accords to the scientific image the peculiar privilege of describing and explaining why the world appears to us as it does. This is something the manifest image cannot do: transcendental reflection within the bounds of the manifest image reveals normative practice as the fundamental horizon of understanding, but cannot explain why the world manifests itself to us as it does or why we engage in these particular practices but not others – indeed, such questions make no sense so long as transcendental investigation is confined to the manifest image. However, they are perfectly legitimate if one understands the task of transcendental reflection to be that of articulating the two images. Ultimately, “the common sense picture [manifest image] of the world, in spite of its delicate coherence, poses problems which it lacks the resources to resolve” (SK, 310). The topic of sensation is crucial in this regard because it lies at the interface between the two images: it is the juncture between the causal nexus investigated by science and the normative space of reasons within which we conduct our daily lives. Thus, “sensations are essential to the explanation of how we come to construct the appearance which is the manifest world” (PSIM, 36).

66

Wilfrid Sellars, Idealism, and Realism

In response to the charge that holism entails quietism, it is worth pointing out that Sellars’ bifurcated holism makes room for conceptual revision because it allows for the possibility of diagnosing categorial dysfunction, not only within each image but also across the two images. This is perhaps the decisive difference between Sellars’ critical holism and the quietist varieties of holism embraced by Wittgenstein or Rorty. Sellars’ controversial proposal for a metaphysics of pure processes exemplifies his critical-­constructive conception of philosophy because it is motivated by concern over the malfunctioning of the category of sensation, rather than by dogmatic deference towards the epistemic authority of qualia.6 In order to understand this we have to clarify why the awareness involved in sensation is not the same as the awareness involved in intuition, or more simply, why sensation is not intuition.

Intuition and sensation Sensing is not thinking because it does not involve conceptual classification: it is not awareness of sorts, resemblances, facts, or any other sort of abstract entity. This much seems relatively uncontroversial. But where Sellars breaks with empiricist orthodoxy is with his claim that sensory awareness does not qualify as awareness of particulars. To be aware of a particular understood as a “this such,” e.g. “this red brick facing me edgewise,” is a conceptual affair, requiring the intervention of conceptual intuition. What is conceptual intuition? On the traditional account, intuitions deliver particulars which are perceived as being thus and so. Consider the following perceptual report: I see this as a red rectangular brick The traditional view maintains that my beliefs about the brick are distinct from my seeing the brick: I see this as a red rectangular brick and I believe that this a brick with a red and rectangular facing surface Here we have a distinction between intuitive perceptual taking, or “seeing as,” and belief proper, which has propositional form. My seeing this as being thus and so and my belief that it is (or is not) thus and so are distinct, with the latter presupposing the former. This traditional account falls prey to the Myth of the Categorial Given: the assumption that to be aware of X is to be aware of it as X. In this version, the myth

The Metaphysics of Sensation

67

fuses thinking and sensing: it assumes that things present themselves to sensory consciousness already endowed with categorial form. We sense something as something before superimposing onto it our belief that it is thus and so: I see this as a red rectangular brick and I believe it is too big for the job at hand. What is wrong here is the assumption that objects cause us to be in certain sensory states, and these sensory states are already endowed with the categorial form that allows them to play a justificatory role in empirical knowledge. The object causes me to see it as what it is and this justifies my subsequent beliefs about it and its relations to other objects. Causation and justification are illegitimately fused. By separating them, we distinguish between the sensory states which objects cause perceivers to be in, and the perceptual states in terms of which perceivers respond to their sensory states. For these perceptual responses to play a justificatory role in empirical knowledge they must already be endowed with categorial form: they must be seeings-­as (or hearings-­as, tastings-­as, touchings-­as, smellings-­as). This is to say that they must involve categorially-­formed conceptual intuitions of sensible particulars. Once we acknowledge the fundamental role of conceptual intuition in empirical perception, we can subsequently distinguish between what we see of objects, and what we see objects as. What we see of an empirical object is a function of our embodied, perspectival relation to it. But this perspectival relation already presupposes the conceptual intuition of the object as something thus and so. What we can see things as is not limited by what we see of them: we see the red exterior of the apple as containing white flesh, and the white flesh is imagined as co-­present along with the red exterior, yet we do not see of the apple its white flesh.7 What is intuited is never a bare particular; rather, it is a condition of our ability to intuit particulars that they be conceptually intuited as internally complex. They are endowed with conceptual content and grammatical form. In his reconstruction of Kant, Sellars explains why conceptual intuition is an epistemically irreducible type of representation: Consider the statements: This is a pyramid This pyramid is made of stone The first has the explicit grammatical form of a sentence. So does the second. But notice that the grammatical form of a sentence is lurking in the subject of the second sentence. From the standpoint of transformational grammar we would think of it as derived from the deep structure: This is a pyramid and it is made of stone

68

Wilfrid Sellars, Idealism, and Realism

One might be tempted to think of “this” as a pure demonstrative having no other conceptual content than that involved in being a demonstrative. Kant does think of an act of intuition as a demonstrative thought, a mentalese “this.” However, he does not think of this mentalese demonstrative as a bare mentalese “this.” An example of an act of intuition would be the mentalese counterpart of: This cube facing me edgewise where this is not to be understood as, so to speak, a mentalese paraphrase of: This is a cube which faces me edgewise The role of an intuition is a basic and important one. It is the role of bringing a particular object before the mind for its consideration. Thus, though there is a close relationship between: This cube facing me edgewise . . . and This is a cube which faces me edgewise the former is an irreducible kind of representation. It is a demonstrative representation which has conceptual content and grammatical form. As noted above it contains the form and content of the judgment “This is a cube.” Thus for Kant, intuitions are complex demonstrative thoughts which have implicit grammatical (and hence categorial) form (IKTE, 428–9). Intuitions are conceptually formed “this-­suches”: “this cube facing me edgewise,” “this stone pyramid,” etc. To say that conceptual intuitions are irreducible representations is to say that they provide the fundamental data for perceptual experience and that they deliver the ultimate subjects of predication for empirical judgments.8 It is because intuitions are representations endowed with conceptual content and grammatical form that they can play this fundamental role in empirical knowledge. Thus perception cannot be decomposed into the sensing of bare particulars coupled with propositionally structured beliefs about those bare particulars. What is intuited is categorially determined and thus already available for propositionally structured belief. How then do intuitions relate to sensations or what Sellars calls “sense-­ impressions”? Our perceptual reports are primarily about physical objects; they are not about the sensory states caused by those objects and responded to by our reports. We perceive physical objects as thus and so and we deliver perceptual reports about those objects, although those reports are in part responses to the

The Metaphysics of Sensation

69

sensory states caused by those physical objects. But according to the Jonesean theory, just as our psychological vocabulary is modeled on candid public speech, so our sensory vocabulary is modeled on the publicly perceptible properties of physical objects.9 We must have mastered that vocabulary before we could subsequently learn to deploy it to make perceptual reports about our own sensory states. Thus the ability to perceive our own sensations presupposes the ability to perceive publically accessible objects and the properties of sense-­ impressions are the postulated counterparts of the properties of physical objects. Whereas physical objects are particulars, sense-­impressions are postulated as states of sentient organisms. They are postulated to explain the experiential content common to veridical perceptions and the kinds of “ostensible perceiving” exemplified by perceptual illusions and hallucinations. Sense-­impressions are what perceivers are responding to when they ostensibly perceive things that are not physically present. As states of sentient organisms, sensations cannot be decomposed into a relation between sensing and sensed. The notion that the sensation of a red rectangle involves a relation to a red and rectangular sense datum follows from the philosophical reification of sensory states into hybrid particulars comprising physical and mental characteristics. Qualia and sense data are reifications of sensory states; a reification that results from conflating the analogical properties of theoretical postulates with the empirical properties of physical entities. Sellars rejects reification and opts for a non-­relational account of sensation. Rather than postulate sense data as immaterial relata for acts of sensing, Sellars adopts an adverbial account whereby the sense impression of a red rectangle is not a relation between a sensing and a red rectangular sensum but a sensing red rectangularly. This adverbial characterization of sensation allows sensings to be described as ways in which perceivers sense, rather than as relations between perceivers and non-­physical particulars. Since sense impressions as states of perceivers are the analogical counterparts of the physical particulars that serve as their models, the model features a commentary specifying the disanalogies between model and copy. The sensation of a red rectangle can be characterized as a “sensing red rectangularly” even though it is understood that strictly speaking there is no red and rectangular state of the nervous system in the organism having the sensation. In what sense then can sensations be adverbially characterized as sensing red rectangularly? This is to ask whether it can ever be literally true within the manifest image to say of someone that they are having a sensation of a red rectangle. Since sensations play a causal role in explaining our perception of the world, they must be accounted for. Characterizing sensations as states of perceivers

70

Wilfrid Sellars, Idealism, and Realism

explains why we seem to perceive or believe we are perceiving something in cases where the physical objects standardly responsible for those perceptions are not present. If sensations play a causal role in explaining what perceivers say or believe within the manifest image, it must be possible to make true claims about them within the confines of that image. Once we reject the Myth of the Given, together with the philosophical reification of sensory states and the dualism that ensues from it, the problem is to explain what the counterparts of redness and rectangularity are when we sense “red rectangularly.” Two options present themselves: either identify the properties of sensations with physical properties of the central nervous system (which means accepting that neurophysiology tells us what sensations really are); or deny that such physical identification is possible and embrace mind-­ body dualism. It is important to note that thoughts do not present the same obstacles to reductive identification as sensations. Since thoughts can be characterized in terms of their functional roles, Sellars insists that “there is no barrier in principle to the identification of conceptual thinking with neurophysiological process” (PSIM, 34). Thus the conceptual role of “red rectangular” is identified through dot-­quoting as the metalinguistic sortal •red rectangular•, and this linguistic function is (at least in principle) identifiable with the set of physical states in which its particular tokenings are realized. But although a sensation can be functionally characterized as a “sensing red triangularly,” it cannot be identified with its functional role.

Perceptual form and perceptual content To understand why sensation cannot be identified with its functional role we must distinguish between the formal and content properties of objects. Formal properties include all of an object’s causal, dispositional, and structural properties: they are conceptual. Content properties comprise the object’s sensible properties in the strict sense: whatever can only be directly seen, touched, heard, tasted, or smelled of an object. While the former are categorial features, the latter are purely occurrent qualities: [W]e do not perceive of the object its causal properties. What we see of it are its occurrent sensible features. This can now be generalized as follows. We do not perceive of the object its character as a substance having attributes, its character

The Metaphysics of Sensation

71

as belonging with other substances in a system of interacting substances, its character as conforming to laws of nature. In short, we do not perceive of the object what might be called “categorial” features. IKTE, 427

What we see of an object are its sensible, content properties; what we see an object as are its conceptual or categorial features. The latter are intuited, not sensed. While formal properties can be functionally characterized and physically identified, content properties possess intrinsic qualitative features that resist functional characterization and physical identification.10 Thus sensation is endowed with a metaphysical significance because it is keyed in to what Sellars calls the “ ‘occurrent content” of physical objects: It should be noted that if physical objects are genuine individuals, they can scarcely have only iffy properties (powers, propensities, dispositional properties, and the like). They must have some non-­dispositional or occurrent attributes. Nor, as Whitehead reminds us, will it do to limit their occurrent attributes to such “primary” qualities as shape and size; for, to use an Aristotelian turn of phrase, geometrical qualities are “formal” qualities and presuppose a “content” or “matter”, thus color. Things which had “primary” qualities without “content” qualities would have “vacuous actuality.” That Whitehead construed the occurrent content of physical objects in terms of feeling, rather than color, is a symptom of the revisionary character of his metaphysics. SK, 302

Whitehead construes occurrent content in terms of “feeling” because he is a pan-­ experientialist for whom actuality divorced from subjectivity is necessarily vacuous. A vacuous actuality in Whitehead’s sense is a non-­subjective actuality, i.e., an un-­perceived actuality wholly independent of subjective experience. At the root of Whitehead’s pan-­experientialism lies a conflation of sensing with perceiving, awareness with awareness-­as. Sellars is too much of a Kantian to accept Whitehead’s revisionary re-­ categorization of occurrent physical content as an instance of feeling. He endorses the traditional characterization of the metaphysical task as the identification of ultimate categories understood as summa genera. “Categories in general,” says Sellars, “are classifications of conceptual roles” (FMPP, 19). Like Kant, Sellars views categories as ways of classifying conceptual representings rather than attributes of unrepresented things. Thus he insists that metaphysics must accurately describe our extant categorial framework, i.e. our extant ways of

72

Wilfrid Sellars, Idealism, and Realism

representing things, before engaging in the revisionary task of postulating new categories or re-­categorizing entities. From the viewpoint of descriptive metaphysics, the category of actuality is not subordinate to the category of subjectivity as species to genus; they are distinct genera. Occurrent actuality manifests itself to us through sensation, and sensations are occurrent states of perceiving subjects, but they need to be conceptually apperceived, i.e. represented, to be known by the subject whose states they are. Although occurrent contents are revealed through sensation, sensation itself is not the cognitive awareness of occurrent contents. This is what prevents Sellars from embracing Whitehead’s pan-­experientialism, as well as the kind of property dualism championed by Chalmers. It is also why Sellars’ “grain argument” for the irreducibility of content qualities is not phenomenological; at least not as that term is used by analytic philosophers of mind.11

The grain argument The grain argument can be summarized as follows. According to Sellars’ “principle of reducibility”: If an object is in a strict sense a system of objects, then every property of the object must consist in the fact that its constituents have such and such qualities and stand in such and such relations or, roughly, every property of a system of objects consists of properties of, and relations between, its constituents. PSIM, 27

If perceivers are ultimately identical with systems of microphysical particles, then every property of perceivers must be explicable in terms of the properties and relations of their microphysical constituents. Now, perceptions are caused by sensations, which are states (i.e. properties) of perceivers. But sensations are occurrent content properties. (Since the content properties of manifest physical objects can occur without being perceived, so can sensory content properties, which is why sensory awareness is not awareness that one is sensing or of what one is sensing. Actuality is not bound to subjectivity—this is Sellars’ rejoinder to Whitehead.) Consequently, the properties of sensation cannot be explained in terms of the properties and relations of the microphysical constituents of the nervous system since (ex hypothesi) these are all formal properties. Occurrent content properties feature an “ultimate homogeneity” which is not explicable in terms of the formal

The Metaphysics of Sensation

73

properties and relations of the constituents of the nervous system. A manifest pink ice cube, writes Sellars, “presents itself to us as something which is pink through and through, as a pink continuum, all the regions of which, however small, are pink. It presents itself to us as ultimately homogeneous” (PSIM, 26). Since the homogeneity of occurrent pinkness is not explicable in terms of the formal properties and relations of the ice cube’s microphysical constituents, it is irreducible. Moreover, since Sellars appears to hold that ultimate homogeneity is a characteristic of all occurrent properties, and since he insists that the properties of sensation are occurrent properties, he concludes that sensations are irreducible to physical states. Stated in this way, the argument is as perplexing as it is straightforward. The claim that occurrent pinkness “presents itself ” as ultimately homogeneous seems to be a particularly egregious regression to the Myth of the Given. The argument also appears to hinge on the presumptive identification of properties that are only supposed to be analogically correlated: the properties of sense-­impressions were supposed to be modeled upon the content properties of manifest physical objects, not identified with them. Last but not least, the argument seems to require that imperceptible entities can only have formal properties, not content properties. Since Sellars rejects the (Whiteheadian) claim that occurrent content properties must be perceived, it is hard to see why he should think imperceptible entities lack occurrent properties. To see why he does not, and why the “grain argument” is neither regressive nor careless, we must draw upon Richardson and Muilenberg’s patient and illuminating reconstruction of it. In their reconstruction, Richardson and Muilenberg distinguish formal analogies from material analogies.12 Whereas material analogies correlate properties of individuals, formal analogies correlate properties of properties, i.e. second-­order properties. Clearly, the Jonesean theory of sensation proposes a formal analogy between manifest physical particulars and sense-­ impressions. This is to say that it establishes a correlation between the second-­order properties of perceptible physical objects and sensations. Thus the respect in which a sense-­impression is like a physical particular is not that of first-­order object properties, such as redness or rectangularity, but the properties of these properties. If the analogy between manifest physical particulars and sense-­impressions is formal, rather than material, then the homogeneity that Sellars ascribes to the ice cube’s occurrent pinkness is a second-­order property of the colored pink expanse; it is not itself an occurrent property somehow present in the pink; a property that could be perceived in the same way in which one perceives the pinkness of the ice cube. The ice cube, says Sellars,

74

Wilfrid Sellars, Idealism, and Realism presents itself to us as ultimately homogeneous; and an ice cube variegated in color is, though not homogeneous in its specific color, “ultimately homogeneous”, in the sense to which I am calling attention, with respect to the generic trait of being colored. PSIM, 26, my emphasis–RB

Homogeneity pertains to the generic, i.e. categorial, trait of being colored, not to occurrent pinkness as such. The pink is seen as homogeneous insofar as it is a color, but homogeneity is not something one sees of the ice cube. More precisely, it is the coloredness of the ice cube that is seen as homogeneous, rather than its occurrent pinkness. Homogeneity is not an occurrent content property but a formal-­generic property of pink as a species of color. And as we saw earlier, Sellars insists that we do not perceive the categorial features of objects. Crucially, although content properties are categorially different from formal properties, the properties of content properties need not be content properties. Moreover, while content properties can have formal properties, the reverse does not hold because content properties are necessarily first-­order properties. This is why Sellars believes they provide a clue to the fundamental material or “stuff ” of the world, even if they cannot be straightforwardly identified with it. The occurrent sensory qualities present in perception (what we see of things) provide the epistemic link between the empirical properties of perceptible things and the content properties of their imperceptible constituents. If there is a sense in which the grain argument is phenomenological, it is the classical Husserlian sense in which phenomenological description is a transcendental exercise rendering visible the invisible categorial structure latent in everyday perceptual experience. Sellars’ claim about homogeneity does not rely upon an empirical description of how things seem to the subject of experience. What motivates the grain argument is not antipathy to reductionism but a worry about categorial dysfunction both within the manifest image and across the two images, i.e. manifest and scientific. The occurrentness of sensations must be preserved because it is what allows them to exercise a causal function and bring about perception and action. Within the manifest image, sense-­impressions are invoked to explain otherwise puzzling features of our behavior. It is important to account for them without resorting to mind-­body dualism; not only because such dualism is the result of a philosophical misprision of the categorial structure of the manifest image, but also because it is the biggest obstacle to the stereoscopic integration of the manifest and scientific images. Sensation is destined to play a key role in

The Metaphysics of Sensation

75

the synoptic fusion of the manifest and scientific images: as the link between the mental and the physical (or reasons and causes), it can help science explain why the world appears to us as it does. Sellars’ argument is not anti-­reductionist per se; rather, it is directed against a peremptory reduction which, because it elides the connection between sensible qualities and content properties, threatens to jeopardize the cognitive integration of the two images.

Homogeneity and process The elimination of homogeneity as a categorial feature of sensory content vitiates our understanding of the occurrentness of sensation and hence of its causal efficacy. Having identified what he thinks is omitted in the attempt to reduce sensory qualities to the properties of microphysical particulars, Sellars believes he has a rationale for a revisionary metaphysical proposal. This proposal is intended to correct the transcendental dysfunction afflicting the category of sensation. As Sellars sees it, what is required is a theoretical counterpart for occurrent sensory content among the properties and relations of the imperceptible constituents of organisms. Thus Sellars postulates “sense” as the imperceptible counterparts of occurrent sensations. It is important to note that this is a postulation within the manifest image; albeit one that Sellars hopes will be eventually corroborated within the scientific image. Nevertheless, the introduction of sensa requires amending one fundamental component of the scientific image: its commitment to a particulate paradigm of fundamental physical entities. If sensa are to be understood as properties of fundamental physical entities, the latter cannot be elementary particles as currently conceived within the Standard Model since these are devoid of homogenous content properties. Sellars fastens upon the manifest category of “process” as a model for the physical entities in which sensa will feature as “ingredients.” “Pure” or “absolute” processes are distinguished from “object-­bound” processes in that they cannot be attributed to objects. While running has an attributive status in expressions like “Socrates runs,” thundering is not attributed to anything when we say “It thunders.” Absolute processes are said to be “subjectless” insofar as they cannot be attributed to an underlying substratum. Their other significant characteristic for Sellars’ purpose is that they are characterized in terms of an intrinsic qualitative aspect that, although associated with their typical causes, cannot be identified with those causes. Thus while

76

Wilfrid Sellars, Idealism, and Realism

buzzing is commonly associated with bees and can be attributed to their activity, the qualitative aspect proper to buzzing is independent of the activity that generates it. Buzzing can be generated by many different causes, but the qualitative aspect that makes of it a buzzing (rather than a fizzing or a roaring) remains the same: [W]hat is primary in the various senses of the verb “to buzz” is the concept of the intrinsic character of a certain kind of process which can be identified in terms of its typical causes. The verb “to buzz”, then, would have a sense in which processes of that intrinsic kind would be buzzings, even when they were not being brought about by one of these typical causes. Thus, in this sense of the verb “to buzz” we could say that a buzzing is going on without implying that some object, e.g., a bee, is buzzing. FMPP, 60

This is to say that buzzing as an occurrent content quality is irreducible to the formal properties of its constituents (e.g. the activity of bees). The concept of an absolute process is the concept of an intrinsic quality occurring independently of the properties of objects or relations among objects. This occurrent aspect is decisive not only because it allows us to conceptualize content qualities independently of any objects that might bear them, but also because it provides us with a paradigm of an absolute occurring; an occurring that is not attributable to something that begins and ceases to occur: In the case of absolute processes we can speak of absolute coming to be and ceasing to be, because when a sounding, e.g., a C#ing, begins, there is nothing which begins—in the relevant sense—to sound. FMPP, 51

Absolute processes combine three basic characteristics: intrinsicality, occurrentness, and homogeneity. Unsurprisingly, these are precisely the characteristics Sellars also ascribes to sensa. Yet the two are distinct: while sensa occur as components of absolute processes traversing the nervous systems of sentient organisms, not all absolute processes involve sensa; moreover, the characteristics of processes are defined without any reference to organisms. But can sensa be stipulated to occur only “in” the nervous systems of sentient organisms without lapsing into dualism? Only if this localization is explicable in physical terms. Sellars distinguishes between a generic and a specific sense of the adjective “physical,” for which he reserves the terms “physical1” and “physical2” respectively. An entity is said to be physical1 if it exists in the framework of space and time; an entity is said to be physical2 if it is “definable in terms of theoretical

The Metaphysics of Sensation

77

primitives adequate to describe completely the actual states though not necessarily the potentialities of the universe before the appearance of life” (CE, 252). The sense in which entities are described as physical2 is obviously a restriction of the broader generic sense in which any entity existing in space-­ time is physical1. Thus physical2 predicates are not sufficient to characterize all the properties of living things. Sensa and absolute processes are obviously physical1; but if the distinction between these two senses of physical is not to turn into an alibi for vitalism or pan-­experientialism it is necessary to insist that physical2 entities are also composed of absolute processes. Sellars calls the latter ϕ2 processes and jokingly proposes “electronings” and “quarkings” as possible examples. In a similar vein, the absolute processes constituent of sensation, such as reddings, buzzings, and C#ings, can be called σ-ings. With these distinctions in place, it becomes possible to sketch the relation between sensa and physical1 processes in terms of the ingredients of σ-ings within physical2 objects: [W]hereas the objects of contemporary neuro-­physiological theory are taken to consist of neurons, which consist of molecules, which consist of quarks – all physical2 objects—an ideal successor theory formulated in terms of absolute processes (both ϕ2-ings and σ-ings) might so constitute certain of its “objects” (e.g., neurons in the visual cortex) that they had σ-ings as ingredients, differing in this respect from purely physical2 structures. FMPP, 124

Reddings, buzzings, and C#ings would be construed as constituents of nervous systems alongside neurons, and the particulate entities of which neurons are composed—i.e. molecules and quarks—would be understood as objects abstracted from underlying physical2 processes such as quarkings and electronings. In the final analysis however, both quarkings and reddings would be physical1 processes.

Concluding remarks While vitalism insists that living things are fundamentally different in kind from and irreducible to non-­living things, Sellars’ monism of absolute processes proposes a common basis for living and non-­living things. It also suggests that both living and non-­living things are composed out of this fundamental element. The ingredience of physical1 processes within physical2 objects, and of σ-ings within physical1 processes, is supposed to stave off vitalism and

78

Wilfrid Sellars, Idealism, and Realism

pan-­experientalism. However, certain questions remain as to the definition of absolute processes, particularly the relation between homogeneity, intrinsicality, and occurentness. As Richardson and Muilenberg point out, there is a distinction to be made between the reducibility of structural properties, e.g. being a ladder, and the reducibility of content properties, e.g. being colored.13 The property of being a ladder is reducible because it consists in the fact that the parts of a ladder—the rungs, the frame, etc.—have certain properties and are related to each other in certain ways. Obviously, not every part of a ladder is a ladder. By way of contrast, the property of being a red brick wall consists in the fact that each part of the wall, i.e. each brick, is itself red. In this instance, being a red brick wall is a reducible property only if each of its parts, i.e. each brick, is also red. Content properties require homogeneity from whole to part in order to be reducible. However, in Sellars’ account, while occurrent pinkness must be taken as a content property, its homogeneity seems to be a structural property (structural features are sufficiently akin to formal properties to allow this equivalence). According to the grain argument, the reason why we cannot apply the principle of reducibility to occurrent pinkness considered as a content property is because it cannot be attributed to each physical constituent of the organism, since we know that the physical particles composing organisms cannot be colored. But if homogeneity is in fact a structural-­categorial feature of occurrent sensation, then it need not be attributable to the organism’s component parts to be explicable in terms of their properties and relations. Thus while the reducibility of occurrent pinkness qua content property requires its attribution to the constituent parts of the object bearing the property, the reducibility of the homogeneity of occurrent pink does not. There seems to be an equivocation between occurrentness and homogeneity lurking in Sellars’ account of sensation: while homogeneity is a property of occurrent properties, such as sensations, it is not itself an occurrent property, i.e. an intrinsic feature of sensation. To qualify a property as occurrent is also to qualify it as homogeneous, since the property of homogeneity is entailed by the property of occurrentness; but the converse does not hold: to qualify a property as homogeneous is not to say that it is occurrent. The intrinsic qualitative particularity that is central to Sellars’ characterization of absolute processes is clearly tied to the property of occurrentness; but its connection to homogeneity is less clear. Thus while the formal nature of homogeneity can be used to rebuke those critics of Sellars who mistake it for a perceptible property, it is also what prevents it from being inherent in occurrentness in the way that Sellars’ argument seems to require.

The Metaphysics of Sensation

79

Notes 1 2 3 4

EPM, 128. See EPM, sections X–XVI, 175–96. See ME, 336. The postulate of a language of thought—to which Sellars is not unsympathetic, and with which his own verbal behaviorism is surprisingly compatible—claims that thinking has a linguistic structure, but that this structure is not that of a natural language. Verbal behaviorism in Sellars’ sense does not rule out the possibility that the capacity for overt public speech depends upon the existence of an inner language of thought. What it does claim is that our initial conceptualization of thinking is modeled upon overt public speech, even if subsequent investigation of the nature of thinking should reveal that competence in a natural language depends upon the operation of an inner language of thought. Familiarity with the structure of public language will still be prior to familiarity with the language of thought in the order of knowing, even if the language of thought (or mentalese) is prior to public language in the order of being. 5 Note that the bifurcation is in our conceptual grasp of the world, not in the world as such. Sellars’ transcendental dualism of images should not be taken to be incompatible with ontological monism. 6 Dennett’s rejection of Sellars’ case for the irreducibility of sensation seems to assume that the argument is empirical-­phenomenological. See Dennett 1981. I side with Seibt, as well as Richardson and Muilenberg, in taking Sellars’ argument to be transcendental-­categorial. See Seibt 1990, 233–70; and Richardson and Muilenberg 1982. The issue is not about what is or is not empirically obvious but about whether and how occurrent sensory qualities can have causal powers. 7 See Sellars IKTE §12–24, 421–3. 8 This is not to say that conceptual intuitions deliver ultimate subjects of predication in the metaphysical sense. Sellars is careful to distinguish between complex particulars, exemplified by perceptual “this suches,” and the simple or ultimate particulars sought for in metaphysical discourse. See Sellars LCP. 9 See EPM §60–2, 190–5. 10 This is essentially how Chalmers characterizes the basic difference between states of awareness and states of consciousness: see Chalmers 1995 and chapter 1 of Chalmers 1996. 11 See Sellars PSIM, V–VII, 25–38. 12 Richardson and Muilenberg 1982, 189–90. 13 See Richardson and Muilenberg 1982, 177–8.

80

Wilfrid Sellars, Idealism, and Realism

Bibliography Chalmers, David (1995), “Facing up to the Problem of Consciousness,” Journal of Consciousness Studies vol. 2 no. 3: 200–19. —— (1996), The Conscious Mind: In Search of a Fundamental Theory, New York and Oxford: Oxford University Press. Dennett, Daniel (1981), “Wondering Where the Yellow Went,” The Monist 64: 102–8. Richardson, R. C. and G. Muilenberg (1982), “Sellars and Sense-Impressions,” Erkenntnis vol. 17 no. 2: 171–211. Seibt, Johanna (1990), Properties as Processes: A Synoptic Study of Wilfrid Sellars’ Nominalism, Atascadaro: Ridgeview Publishing Company. Sellars, Wilfrid   BBK —— (1991), “Being and Being Known,” in Science, Perception and Reality.,   Atascadero: Ridgeview Publishing, 41–59.   CE —— and P.E. Meehl (1956), “The Concept of Emergence” in H. Feigl and  M. Scriven (eds.), Minnesota Studies in the Philosophy of Science: Vol. I: The Foundations of Science and the Concepts of Psychology and Psychoanalysis, Minneapolis: University of Minnesota Press. 239–52.   EPM —— (1991), “Empiricism and the Philosophy of Mind,” in Science,  Perception and Reality, Atascadero: Ridgeview Publishing Company, 127–96.   FMPP —— (1981), “Foundations for a Metaphysics of Pure Process,” The Monist   64: 3–90.   IKTE —— (2002), “The Role of Imagination in Kant’s Theory of Experience,” in  J.F. Sicha (ed.), Kant’s Transcendental Metaphysics, Atascadero: Ridgeview Publishing Company, 419–30.   LCP —— (2005), “On The Logic of Complex Particulars” in J. Sicha (ed.) Pure  Pragmatics and Possible Worlds: the Early Essays of Wilfrid Sellars, Atascadero: Ridgeview Publishing Company, 140–67.   LTC —— (1969), “Language as Thought and as Communication,” Philosophy and   Phenomenological Research, vol. 29, no. 4: 506–27.   ME —— (1981), “Mental Events,” Philosophical Studies, vol. 39, no. 4:   325–45.   PSIM —— (1991), “Philosophy and the Scientific Image of Man,” Science,  Perception and Reality, Atascadero: Ridgeview Publishing Company, 1–40.   SK —— (1975), “The Structure of Knowledge,” in Hector-Neri Castañeda (ed.),  Action, Knowledge, and Reality: Critical Studies in Honor of Wilfrid Sellars, Indianapolis: Bobs Merrill Company, 295–346.   SM —— (1968), Science and Metaphysics: Variations on Kantian Themes,   London: Routledge & Kegan Paul.

The Metaphysics of Sensation

81

  SRLG —— (1991), “Some Reflections on Language Games,” in Science, Perception,   and Reality, Atascadero: Ridgeview Publishing Company, 321–58.   TTC —— (2002), “Towards a Theory of the Categories,” in J.F. Sicha (ed.),  Kant’s Transcendental Metaphysics, Atascadero: Ridgeview Publishing Company, 321–39. Talhouk, Omar (2014), “Natura Naturans: The Concept of a Nature for Representation,” MA thesis, American University of Beirut. Wittgenstein, Ludwig (1969), On Certainty, G.E. Anscombe and G. H. von Wright (eds.), Oxford: Basil Blackwell.

4

Language, Norms, and Linguistic Norms Willem deVries, University of New Hampshire (USA)

In this essay, I argue that Sellars did not, but should have recognized a form of natural normativity that derives from the functionality of biological processes and structures. This form of natural normativity is not reductive, but it is naturalistic. It is also a precondition of intentionality, and recognizing how it conditions intentionality helps to disarm the all-­too-common notion that intentionality forces either dualism or even idealism upon us. Thus, the view I defend supports Sellars’ psychological nominalism, which was built on the premise that abstracta cannot play a direct explanatory role in human behavior. Interestingly, we can draw on Hegelian arguments that have recently been echoed in the enactivist literature to motivate my emendation of Sellars’ views.

Normativity and modality The norms to be discussed here are prescriptive or evaluative, not reportive or descriptive statistical generalizations. A descriptive norm generalizes over a set of data using some statistical rule, often that of an average or a mean. In reports of a recent meeting of the Seismological Society of America, for instance, we read, concerning the recent upsurge of earthquakes in Oklahoma, that “Scientists believe wastewater disposal [in the process of fracking] could have a cumulative effect and that as more wastewater is shot underground, more intense earthquakes could become the norm.”1 Prescriptive norms, however, are tied to the notions of obligation and permission. They encourage action of some kind—even if it is sometimes inaction. Such norms are usually expressed using some deontic modality, one of the set of linguistic modifiers, such as “ought” or “may,” by which we mark whether something is required, permitted, or proscribed.

84

Wilfrid Sellars, Idealism, and Realism

A full consideration of prescriptive norms has to have some story about the fundamental nature of the modalities, some story about just what words like “ought” and “may” express. In the rationalist/idealist tradition, modalities are usually taken to refer to inference-­like real connections among the entities involved.2 I say “inference-­like” because the entities connected are not always themselves construed as directly propositional or judgmental.3 Connections referred to are real in that they are identity-­constituting for those things they relate. But in being identity-­constituting, the relations do not have to be definitional. Dogs may be four-­legged creatures, but three-­legged dogs are surely possible; matches typically light, if struck. In this tradition, accepting the reality of concepts or judgments entails accepting as equally real the modally qualified connections among them. Thus, idealism and modal realism are usually closely tied, because modal relations are tied to the very constitution of concepts, judgments, and inferences. The opposing empiricist tradition, however, is much more atomistic in its view of ideas and concepts. The relations among concepts, judgements, and other idealities are thought to be either trivial (as in “brother = male sibling”) or merely contingent. Thus empiricism tends to be skeptical about modalities altogether: at very best they express a subjective association or feeling that indicates nothing in particular about “reality.” The poster child for this view is Hume’s treatment of the causal modalities, according to which causal necessity is not a real connection “in the objects” but only something in the mind There is a middle path between these two traditions that gives each of them a piece of the prize: the modalities indeed play an expressive role, but what they express is not a merely subjective feeling or association. Rather, they are used to express our commitment to certain inferential structures embodied in our shared conceptual framework or language. They are therefore subject to rational assessment, for we can (and should) worry about whether the inferential commitments of our shared conceptual framework are the ones we want, given our overall aims. This path is an attempt to recognize the structural insights of the idealist tradition—in particular their recognition that the modalities are intrinsically related to normative questions about the propriety of certain conceptual connections—without committing to metaphysical excess. It is on this middle path that we find Wilfrid Sellars, whose thought on these matters I want to exploit and explore. Thus, in Sellars’ view, when I say something like “Brothers are necessarily male,” I am expressing our commitment (not, I remind you, my commitment, but our commitment) to the validity of the inference from

Language, Norms, and Linguistic Norms

85

David is my brother to David is male. We call this a conceptual necessity, because anyone who doesn’t recognize its validity is normally not considered to be in full command of the concepts involved. A claim such as If the light is on, then, necessarily, the circuit must be closed does not expresses a conceptual necessity because there’s a perfectly good sense in which one can command the concept of a light’s being on without commanding the concept of a circuit being closed. But empirical investigations lead us to a theory in physics according to which there is a law-­like connection between electric lights being on and the circuit involved being closed. Accepting this theory amounts to endorsing adding to our language the standing rule of material inference The (electric) light is on. So, the circuit involved is closed. According to Sellars, the story concerning “ought” is related in kind, but translated to the practical sphere. A recommendation such as If you’re going to mow the lawn, you ought to wait until the dew is gone endorses, under the circumstances, an inference something like the following Shall be (X mows the lawn). [Mowing the lawn requires (or is optimized under) conditions C, which include dryness.] Shall be (X waits until the dew is gone to begin mowing).4 Oughts (and their partners, permissions and proscriptions) express our recognition and endorsement of such practical inferences. Such practical reasoning is reasoning about values, or better, it is value-­ directed or value-­oriented reasoning, since engaging in it doesn’t entail commanding any explicit conception of value per se. We could also say that it is goal-­directed reasoning. Goals aimed at have presumptive value, though they may not have value “all things considered.” What is valued or aimed at can be idiosyncratic, subjective, transient, and entirely particular. Those qualities do not put values or goals outside the reach of reasoning. But norms, I take it, have to

86

Wilfrid Sellars, Idealism, and Realism

have some generality. This is true for “norm” in both of its principal senses, as a reportive regularity and as principle of action or rule. In both cases, norms are general in a given population, even if the limiting cases (where the population is either a single individual or the entire domain) deserve special treatment. So, given what I’ve said so far about value-­directed or goal-­directed reasoning, norms are not yet in explicit view for us, nor is anything like morality.

Norms, purposes, and rules The kind of norms, prescriptive norms, that are the target here, are, as far as I can see, always connected to contexts of goal-­directed or purposive structure or behavior. This is one of the ways they differ from merely reportive norms. Normally, iron left in water (or high humidity) rusts. This states a reportive norm. Of course, iron does not rust “on purpose” nor is the iron itself criticizable for doing this (though to users of iron implements it may seem that those implements are hell-­bent on rusting and require a great deal of preventive maintenance to avoid it). Descriptive norms are ubiquitous. There are general patterns of properties and behaviors that are “normal”; this is why it makes sense to talk about “the natural course of events.” Prescriptive norms, in contrast, have a much smaller range of application. It is tempting to think that prescriptive norms apply only to organisms or related biological phenomena. But that would be a bit too quick. Inanimate, non-­biological objects and phenomena can easily have a tremendous effect on organisms and become deeply entwined with them. Consider the claim that it ought to be the case that the earth’s air and water contain, at most, low levels of the chemicals we consider pollutants. “Oughts” can apply to inanimate objects and phenomena, but it looks like they can do so only in a broader context of their interactions with and effects upon organisms, purposefully organized bodies. Sellars accommodates these points by distinguishing between “rules of action” (which he also calls “ought-­to-do”s) and “rules of criticism” (aka “ought-­ to-be”s). Rules of action presuppose cognitively and conatively sophisticated beings to whom they are addressed, beings that are capable of recognizing and understanding rules or action prescriptions and applying them in light of current circumstances. Rules of criticism, in contrast, apply without particular regard for the agency that might come to realize them. They can apply to inanimate objects: Clock chimes ought to strike on the quarter hour. But, according to Sellars, such ought-­to-bes

Language, Norms, and Linguistic Norms

87

point beyond themselves in two ways. In the first place they imply (in some sense of this protean term) a reason, a because clause . . . In the second place, though ought-­to-be’s are carefully to be distinguished from ought-­to-do’s they have an essential connection with them. The connection is, roughly, that ought-­ to-be’s imply ought-­to-do’s. LTC §10

So, if it is the case that clock chimes ought to strike on the quarter hour, then clock makers ought to design and build their clocks to chime on the quarter hour. I think there is also a sense in which ought-­to-dos imply ought-­to-bes: any time one ought to do some particular thing (e.g., tell the truth), must it not also be the case that successful fulfillment of the ought-­to-do also realizes some state of affairs that ought to be the case? Sellars is committed to the idea that rules of criticism imply rules of action. We need to worry: does that doctrine have the converse implication that if, in some particular context, there are no rules of action, we must conclude that there are also no ought-­to-bes either? He claims only that the “implication” involved is implication “in some sense of that protean word,” so we cannot assume that he thinks it operates just like the formalized notions of material implication or strict implication. Suppose we consider a moment in evolutionary history before the development of cognitively sophisticated beings. Rules of action (ought-­to-dos) would not apply to anything in such a situation, because nothing could recognize and act upon them. Still, we would normally be inclined to think that there are salient ought-­to-bes that pertain to the structure and function of the organisms that do exist. It would still be the case that such organisms could be deformed, disabled, or diseased, and being in such a state seems possible only in contrast to what the organism ought to be. My argument will be that we do need to recognize such ought-­to-bes. They are part of our understanding of biological functionality and do not violate the principles of a sound naturalism. In such a situation, rules of action apply nowhere, because there are no agents, nothing capable of undertaking a conscientiously-­applied program of action aimed at realizing what ought to be. We need not, however, think that we must abandon all conceptions of rulishness and normativity in such circumstances. We can still make sense of counterfactual rules of action. The ought-­to-bes that define health, for instance, would still imply that were there any veterinarians or horticulturalists, they ought to do X, Y, or Z to restore afflicted organisms to health.5 We will see, however, that Sellars can’t be fully satisfied with such counterfactual oughts, for his model requires that rules of action are real only to the extent that

88

Wilfrid Sellars, Idealism, and Realism

(1) they are explicitly represented by agents, and (2) such representations are involved in the causal process generating the required actions. This follows from his psychological nominalism, whether in the stringent version given in EPM or the more relaxed version found in “Mental Events.” There may be ideal rules of action that no one ever follows, but they will not contribute to an understanding of human behavior. To make some contribution to understanding human behavior, a rule of action must have explanatory traction, and to gain such traction, the rule must be represented by some agents in a way that enables it to play a role in the deliberative process that generates their behavior. Counterfactual rules of action are abstractions of a fairly high order, and an explanation that appeals to abstractions is immediately suspect for Sellars, unless it is clear how the relevant abstraction gains causal traction in the material world. This is the real motivation behind Sellars’ psychological nominalism: abstracta are, on their own, causally impotent. Any abstractions that seem, as such, to play an explanatory role must be embodied in order to acquire any causal efficacy, and the primary way in which abstractions can be embodied is in language-­like items. We can see here how Sellars’ commitment to psychological nominalism affects his views on normativity. Sellars’ basic objection to standard forms of Platonism is that it leaves unexplained both our knowledge of abstracta, and the explanatory role of abstracta more generally, given that abstracta do not have any causal properties. Sellars famously asserted that “in the dimension of describing and explaining the world, science is the measure of all things,” (EPM §41; in SPR 172) and he always held that description and explanation go hand in hand (that’s why one article is titled “Concepts as Involving Laws and Inconceivable without Them” (CIL), but he also has to hold that prescriptions have explanatory force as well, if they are to provide explanations of human behavior. To have such force, however, they must also be describable as items within the causal nexus. So it seems that norms can have prescriptive reality only to the extent that they also possess descriptive reality, and that means being expressed in a representation that plays a causal role in the generation of action. Thus, one of my concerns is that Sellars does not seem to be able to account for the use of normative vocabulary in biology, a vocabulary that includes the notions of deformity, disability, and disease as well as that of adaptation and reproductive advantage or success. At best, given the way Sellars has arranged things, normative vocabulary can be used metaphorically in such contexts. Ultimately, though, this makes the job of accounting for the full-­fledged normativity of rule-­obedience exhibited by humans all the more difficult.

Language, Norms, and Linguistic Norms

89

Norms, language, and explanations This problem appears a bit more clearly from another angle when we consider Sellars’ account of language acquisition. He thinks of a language as a structure of interlacing, rule-­governed behaviors that both enable and facilitate further complex behavioral organizations in social animals. In particular, it enables a certain form of reasoning in which principles of inference (of various sorts) are explicitly represented and acted upon—truly logical reasoning. How does one acquire the mastery of such a structure of rules? If we equate mastering such a structure of rules with coming to obey the rules, Sellars realizes, there is a big problem. What is a rule but a sentence that prescribes some action? A rule governing linguistic behavior is therefore a sentence in a metalanguage of the base language. So it might look as if learning one’s first language presupposes having learned its metalanguage first. That’s an idea that goes nowhere fast, because it quickly generates a vicious regress (Cf §1–10 in SRLG). Sellars rightly infers that there must be a mode of acting in accordance with a rule that is not full-­fledged rule-­following or obeying, but is also not a matter of mere consistency or conformity with the rule, i.e., something that could happen “by accident.” This mode of behavior is one where it is correct to say that the behavior elicited occurs because of the rule, even though the rule, as a linguistic representation, does not play a specific proximate causal role in the organism’s production of the behavior. I complained earlier that the connection Sellars draws between ought-­to-bes (rules of criticism) and ought-­to-dos (rules of action) makes it difficult to apply oughts or normative conceptions unproblematically in biological contexts, even though we seem to do that all the time. But Sellars is quite aware of the ways in which biological contexts beg us to mobilize such conceptions. Two of his examples in SRLG when he introduces the notion of pattern-­governed behavior as behavior that is neither full-­fledged rule-­obeying behavior nor mere rule-­ conforming behavior, come from biology, and, in particular, evolutionary theory. It is instructive to give what he says about those examples closer scrutiny, particularly when issues concerning idealism are in view. The problem is that if purposive or functional vocabulary always implies the presence of a mind that can follow rules, then the very existence of biology seems to imply the existence of a mind somehow implicated in the existence of the biological. Sellars seeks to convince us that we should not accept a stark dichotomy between two kinds of behavior that contributes to the realization of a complex pattern (such as organisms exhibit): (1) behavior that merely (and “accidentally”)

90

Wilfrid Sellars, Idealism, and Realism

conforms to certain rules (and yet contributes to the realization of a complex pattern), and (2) full-­fledged rule-­obeying behavior. In order to point to a third alternative, Sellars distinguishes two senses of “accidental.” In one sense, “accidental” means unintended; in this sense a great deal of purposeful behavior (such as the beating of my heart) is “accidental.” But in another sense “accidental” is the opposite of “necessary”, and there can surely be an unintended relation of an act to a system of acts, which is nevertheless a necessary relation—a relation of such a kind that it is appropriate to say that the act occurred because of the place of that kind of act in the system. SRLG §12: in SPR 325; in ISR 326

This latter sense opens us to the possibility that functional characterizations of events could be rule-­governed without being instances of explicit rule-­following. To clarify this point, Sellars introduces two examples from the biological realm. The first is this: it is quite proper to say that the sequence of species living in the various environments on the earth’s surface took the form it did because this sequence maintained and improved a biological rapport between species and environment. SRLG §13: in SPR 325; in ISR 32–3

I want to emphasize how he then follows this claim up: It is quite clear, however, that saying this does not commit us to the idea that some mind or other envisaged this biological rapport and intended its realization. It is equally clear that to deny that the steps in the process were intended to maintain and improve a biological rapport is not to commit oneself to the rejection of the idea that these steps occurred because of the system of biological relations which they made possible. SRLG §13: in SPR 326; in ISR 33

The idea that certain events happen because of the systems of objects and events they make possible can be held on to without thinking that there must be some intentional agency at work designing and operating to realize those latter systems. Why does Sellars move immediately to deny the involvement of an intending mind? He is not specifically looking to refute the doctrine of intelligent design. But he is worried about the apparent commitments of the causal claim being made. Given the occurrence of mutations and the facts of heredity, we can translate the statement that evolutionary phenomena occur because of the biological rapport they make possible—a statement which appears to attribute a causal force to an

Language, Norms, and Linguistic Norms

91

abstraction, and consequently tempts us to introduce a mind or minds to envisage the abstraction and be the vehicle of its causality—into a statement concerning the consequences to particular organisms and hence to their hereditary lines, of standing or not standing in relations of these kinds to their environments. SRLG §13: in SPR 326; in ISR 33

The train of thought that Sellars wants to ward off is now clear: the occurrence of certain events is explained by reference to an abstraction—in this case a “rapport between organism and environment.” Abstractions have no causal powers of their own, so the explanation is easily misinterpreted as an invocation of some mind that can entertain the abstraction and act upon it. That pushes us at least towards a dualistic view of nature, and opens us up to a full-­blown idealism. There is an historical side to Sellars’ analysis too: arguably, primitive thinking often moves from “non-­accidental” (i.e., functionally goal-­directed) to “intentional”—isn’t that how myths “explain”? To the extent that this move seems natural or taken for granted, necessities and other modal properties are still treated as fundamentally grounded in minds and their actions. As an advocate of the “middle way,” Sellars thinks there’s something to this thought, but it is far too easy to reify the necessities and other modal properties, giving them a spurious independence of the actual, embodied minds that are persons and loses sight of their grounding in our modes of thinking. This keeps the prospect of a general idealism alive longer than Sellars thinks called for. A proper account of thinking as the activity of material, embodied persons, a psychological nominalism, which Sellars believes he can give us, will prevent a rush to idealism. That is, material reality will retain ontological priority, because thinking activity is taken to be dependent on materially embodied persons, and not prior to or even fully independent of materiality. Sellars’ second example is similar, but even more closely related to the problem concerning linguistic rules, for it concerns the bee dance in which bees inform their hivemates about the location of nectar or pollen. Sellars wants to preserve the idea that the individual wiggles in the dance “occurred because of the complex dance to which it belongs” (SRLG §14: in SPR 326; in ISR 33). But such an explanation, he again worries, appears “to attribute causal force to an abstraction, and hence tempts us to draw upon the mentalistic language of intention and purpose” (SRLG §14: in SPR 326; in ISR: 33). What Sellars wants is an explanation in which it remains proper “to say that the dance pattern as a whole is involved in the occurrence of each wiggle and turn” (SRLG §14: in SPR

92

Wilfrid Sellars, Idealism, and Realism

326; in ISR: 33) without, however, invoking any minds in which “the dance pattern as a whole” shows up as an explanatory posit. How can we account for the efficacy of a whole (the dance pattern) without reducing it to the efficacy of its parts, the way mechanistic explanations usually work? A long-­standing form of answer is to put the idea of the whole in some mind which then acts on the world. The idea of the whole is abstract, however, and both dualism and idealism leave its causal efficacy unexplained. Sellars thinks evolutionary accounts can fill this bill, for in such an account “the dance pattern comes in not as an abstraction, but as exemplified by the behaviour of particular bees” (SRLG §14: in SPR 326; in ISR 33). The behavior in question is first exemplified in a way that cannot be said to be “because of the pattern,” but, since the pattern itself enhances the survival value of the organisms involved, it comes to be wired into the population. Ruth Millikan has picked up and elaborated upon this kind of story to great effect. I see no problem with such accounts. Nonetheless I do think that a potential problem arises from the fact that one possible reading of the moves Sellars makes here is that it constitutes a rejection of any serious use of normative vocabulary in biological contexts. My own view is that we need to find an analysis according to which the normal use of normative vocabulary in biological contexts can be seen to be unproblematic. Sellars seems very concerned to show that there is no need to believe that simply because an abstract noun occurs in the “because” clause of an explanation, the explanation somehow commits one to the presence of a mind executing some conscious thought or plan. Sellars takes it, however, that normative discourse is always mind-­involving, for real normativity depends on the existence of full-­fledged agents who can represent and follow rules explicitly. But if normative discourse is always mind-­involving, then one can read the second section of SRLG (§11–17) as an argument for the dispensability of explicitly normative conceptions in biological contexts. Since the seventeenth century, there has been a strong tendency to view nature as “disenchanted,” devoid of normativity, and to reject teleology as an unscientific conception. These prejudices may have been important in separating the new science from its forebears, but there is now good reason to rethink them. It would be a mistake to think that, like Ruth Millikan, Sellars himself would try to drive the argument further to show that explicitly normative conceptions are equally dispensable in the contexts of thought and language. Sellars does not, as I read him, think of thought and language as further “biological categories,” à la Millikan. They are social categories, he thinks, that become explicit in our own reflective thought, capable of informing our action, and therefore incapable of

Language, Norms, and Linguistic Norms

93

being “explained away.” This is why Sellars believes that the scientific image must be “joined” to the language of individual and community intentions, which is itself indispensable to rational beings. (Again, intentional states figure in good explanations of human behavior. They cannot do this as mere abstractions or relations to abstracta. They must be embodied in ways that fit into the causal network of material nature.) It is not Sellars’ purpose to explain normativity away; only to explain what it is. To put the point simply and in terms we’ve already encountered, Sellars clearly believes that we often do things because of rules. Rules are abstractions, but in the cases of linguistic rules and moral rules, these abstractions, he believes, do bottom out in someone’s actually “envisaging” or “comprehending” the abstraction, in activities to which the “mentalistic language of intention and purpose” apply literally. Because he thinks he has an analysis of what it is to think of an abstraction that does not violate materialism, he is happy to admit that normativity is essentially mind-­involving. Normativity exists because we can represent norms, which, via those representations, gain causal traction in the world. Sellars is careful to argue that there must be rules of language that are not “ought-­to-do’s” requiring cognitively and conatively complex processing in order to obey. That would lead to the regress he seeks to escape at the beginning of SRLG. The linguistic rules lying at the bottom of language, the formation rules and basic, underived transformation rules, are ought-­to-bes that we come to act in accord with as we absorb our initial language in medias res, together with guidance from our adult mentors: A language is a many-­leveled structure. There are not only the ought-­to-be’s which connect linguistic responses to extra-­linguistic objects, but also the equally essential ought-­to-be’s which connect linguistic responses to linguistic objects. There could be no training of language users unless this were the case. Finally, there would be no language training unless there were the uniformities pertaining to the use of practical language, the language of action, intention, of “shall” and “ought”, which, as embodying epistemic norms and standards, is but one small (but essential) part of the conceptual structure of human agency. One isn’t a full-­fledged member of the linguistic community until one not only conforms to linguistic ought-­to-be’s (and may-­be’s) by exhibiting the required uniformities, but grasps these ought-­to-be’s and may-­be’s themselves (i.e., knows the rules of the language). One must, therefore, have the concept of oneself as an agent, as not only the subject-­matter subject of ought-­to-be’s but the agent-­subject of ought-­to-do’s. Thus, even though conceptual activity rests on a foundation of conforming to ought-­to-be’s and of uniformities in linguistic

94

Wilfrid Sellars, Idealism, and Realism behavior, these uniformities exists in an ambience of action, epistemic or otherwise. To be a language user is to conceive of oneself as an agent subject to rules. My point has been that one can grant this without holding that all meaningful linguistic episodes are actions in the conduct sense, and all linguistic rules, rules for doing. LTC 513; in ISR 64

All this I take to be essentially correct. The rub comes, as one might suspect, not in the account of the ontogenic transmission of language from practiced language-­user to budding language-­user. Something like the story Sellars tells has to be right. But how do we get from non-­language-­user (even if possessed of a relatively sophisticated animal communication system) to full-­fledged language-user, given Sellars’ constraints? In his view, before the development of language proper, it is strictly inappropriate (and at best metaphorical) to apply normative concepts to the behavior—whether communicative or otherwise—of the proto-­humans still shy of full language. We are once again driven, Sellars holds, to believe that there is no normativity, even ought-­to-bes, until there are agents subject to rules. Rule-­talk has no explanatory value unless rules can be materially represented and thus inject themselves into the causal network. Psychological nominalism is the condition of taking rule-­talk seriously, because psychological nominalism insists that abstractions such as rules must have a material embodiment in something like linguistic entities in order to be able to have explanatory force. Since the justificatory structure of norms is in part historical, Rebecca Kukla has suggested that Sellars has to believe that we systematically “misremember” and “misrecognize” the origins of the norms we live by, treating them, essentially, as “always already” having been in place.7 There’s no non-­question-­begging argument that gets one from a purely causal story about human behavior that would be absent any normativity to the kind of richly normative story we use (and use effectively) to understand ourselves. That is to say, in effect, that in the rich language of the manifest image, redolent with normativity as it is, a true, coherent story about the origins of normativity itself cannot be formulated. Precisely because the manifest image is so deeply imbued with normativity, any attempt to formulate a theory of the origin of normativity in manifest image terms will look question-begging. What science can provide is a coherent story about how a population comes to exhibit the uniformities of behavior that reflect the norms espoused by the community. Then, using Sellars’ principle, “Espousal of principles is reflected in uniformities of performance,” (TC: in SPR 216) a hermeneutic investigation can

Language, Norms, and Linguistic Norms

95

hope to reveal the principles those uniformities of performance reflect. But it is yet a different endeavor to attempt to justify for the community the norms it espouses. The fact that the community does act in accord with a norm is not itself a justification of that norm. We need a different vocabulary, the vocabulary of individual and community intentions, to engage in that justificatory project, because norms can be justified only in terms of other norms and purposes.

A partial grounding of social norms I now want to suggest that someone who, like Sellars, seeks a coherent but non-­ reductive naturalistic account of the norms that constitute human societies and their languages, could take a page from German idealism, without falling into idealism itself. Such an account would respect the logical peculiarities of normative language without locating norms ontologically outside the natural realm. What I have in mind is the dialectic of teleology that we can trace from Kant’s Critique of Judgment to the teleology chapters in Hegel’s Logics. After reviewing this dialectic, I will argue that Sellars is still too Kantian in his conception of teleology and therefore also intentionality, and needs to make the move Hegel argued is necessary: the recognition of a form of natural teleology that is prior to and the basis of any form of intentionality.8 Kant was wedded to the idea that teleological concepts were necessarily both mind-­involving and subjective. He used what I call the “intentional model” of teleology. According to this model, when we say that S did A in order to G, we are attributing to the subject, S, a complex intentional state (complex in that it involves both beliefs and desires, cognitive and evaluative elements) which is causally sufficient for S’s doing A. The goal, end, or telos enters into this account only in the description of the intentional state, and in particular the desiderative component, as its content, so we are neither committed to the existence of the state aimed at nor to its possessing a (backwards) causal efficacy, i.e., the goals we seek to achieve do not reach backwards in time to make themselves actual. This intentional model of teleology has been around for millennia, and it easily accommodates attributions of purposiveness to human and even animal action and to artifacts as well, but it runs into trouble in explaining attributions of purposiveness to things that have no intentional states and are not artifacts. The model has difficulty accounting for the fact that hearts beat in order to pump blood, that flowers are colorful in order to attract pollinators, etc. Furthermore,

96

Wilfrid Sellars, Idealism, and Realism

while it assumes that intentional states can be causally efficacious, it does not tell us how that is possible. Rather, it seems to presume dualism; intentional states operate on different principles from mechanical causes, even though the explanation retains much of the form of a mechanical explanation (think of Ryle’s complaints about Descartes’s “mental mechanism” picture of mind). Hegel recognizes both the strengths and the weaknesses of the intentional model of teleology. He recognizes, that is, that the conception of teleology he calls the “subjective end” does work well when applied to intentional action and to artifacts, though even there it is clearly subject to strict limitations. But Hegel clearly recognizes as well the weaknesses of the intentional model of teleology: it cannot account for the teleology in biology, and it divorces mental causes too sharply from the material world in which they are to have effect. Most important, Hegel thinks the “subjective end” or intentional model of teleology cannot capture the very heart of teleology, which is that the end or final cause is productive of itself. Final causes are self-­realizing causes. Hegel, that is, thinks that all teleology is fundamentally self-­realization. Organs like hearts and kidneys have particular functions or goals, but these all ultimately play a role in the self-­realization of the organism. The phenomena that escape the grasp of the intentional model of teleology that Kant employed are those to which the notion of inner design applies. Hegel takes there to be two cases of such inner design: natural organisms constitute one such case; the purposiveness of the world constitutes the other. I’m not going to worry about the latter, though Hegel constructs some interesting arguments to show that the purposiveness of the world cannot depend upon an extramundane divinity. The analysis of inner design Hegel gives us in the middle section of the teleology chapter of the greater logic is devoted to showing us that the intentional model of teleology must itself be an abstraction from a more complex, more fundamental form of teleology, an abstraction which is useful under certain conditions. The analysis of the concept of a means mediating between the subjective end and the objective world brings us to realize that there are certain special objects in the world, without which the intentional model of teleology could not have application: animated bodies, organisms that embody natural purposiveness. If there were no animated bodies, the concept of a subjective end, an intention, would be a useless dangler without traction in the world. We use the same form of words to attribute purposiveness to actions and to organisms: S does A in order to G. In the former case, it is the intentional model of teleology that spells out what would make the attribution true. In the latter

Language, Norms, and Linguistic Norms

97

case, what I call the “functional model of teleology” spells out the truth conditions of the attribution. In functional teleology, however, it cannot be the case that doing A in order to G is a one-­time event. The heart cannot beat just this one time if it beats in order to circulate the blood. In functional teleology, event-­ tokens inherit their purpose from their respective event-­types. The other obvious difference between functional and intentional teleology is that functional teleology does not attribute any intentional states, beliefs or desires, to the subject of the attribution. The explanatory power of a functional explanation arises not from the ill-­understood causal efficacy of inner, subjective states, but from an orchestration of individually normal interactions directed at some ultimate goal. That is, a certain pattern of events or interactions exists because prior instances of that pattern have contributed to the later occurrences of that pattern. Hearts that beat now beat for the purpose of circulating blood because prior instances of hearts beating in ways that made blood circulate were significant contributors to the existence of those hearts that beat now, and thus contributed to the self-­ replicating and self-­maintaining pattern of which the current heart beats are instances. Such explanations presuppose some sort of inherent goal-­structure for the organism or object we apply them to.9 Hegel thinks that self-­realization is a universal and inherent goal of all being. Natural objects—and organisms are the primary models—have the inherent, unconscious goal of realizing themselves as the kinds of things they are, though they do so with varying levels of perfection. More recently, we have come to believe that evolution by natural selection explains why organisms naturally or inherently strive to survive and reproduce. The self-­sustaining pattern adumbrated by evolutionary theory explains why the individual organisms seek survival for reproduction, and, thus, why we can count survival for reproduction as an intrinsic good for organisms, universally pursued. Any failure to pursue this goal we count as a pathology of the organism. Hegel did not yet have the idea of speciation by natural selection. But we have an argument in his work to the effect that our ability to employ the intentional model of teleology to explain particular actions and artifacts essentially presupposes that there are already embodied beings (animated bodies) endowed with a kind of natural teleology, a built-­in, natural purposiveness directed at goals in terms of which the very identity of the being is to be defined. Hegel’s argument is echoed today in arguments that minds can be made sense of only as embodied in a naturally active organism. Seeing intentionality as a characteristic of purposively organized natural systems interacting with an environment that

98

Wilfrid Sellars, Idealism, and Realism

simultaneously threatens and provides resources to the organism is a powerful motivation to interpret the intentional states of such organisms in a thoroughly realistic fashion. We can make sense of intentional states only by viewing them in their practical context, modulating one’s interaction with a real, surrounding world. The goal-­directedness of the structure and function of organisms induces a form of normativity that Sellars ought to have paid greater respect. It is, in my view, not normativity “so- called” or a merely metaphorical normativity. It is not identical to the social normativity available to cognitively sophisticated, language-­using, rule-­following, self-­conscious organisms, but it isn’t merely an “as-­if ” normativity either. Perhaps we can distinguish assessment normativity from prescriptive normativity. Assessment normativity requires only ought-­tobes or rules of criticism, whereas prescriptive normativity engages rules of action as well. Evolution provides a context in which we can see the applicability of rules of criticism, even though there need be no beings capable of inferring from them any rules of action, much less acting on them. Sellars himself noted that abstractions seem to play an explanatory role in evolutionary contexts, but in a way that need not bother the naturalist. He was on the cusp of recognizing the sui generis being of natural normativity, but did not make the final step. Such natural normativity is the precondition for all other forms of normativity; prescription presumes an ability to assess. Recognizing natural normativity is a step we should be happy to make. It does better justice to the structure of our thinking about biological matters, for one thing. But it also helps fill in the gap between the supposedly full-­fledged normativity of rule-­obeying consciousnesses and the merely mechanical world in which they operate and from which they somehow emerge. It seems utterly unintelligible how a collection of material objects could be or become a rule-­ obeying person, without some further mediating structures. It is still difficult to understand how an organism fraught with ought-­to-bes can become and then be a person who recognizes and acts upon both ought-­to-bes and ought-­to-dos, and I do not pretend to understand just how that story may go. But I at least think that it is a story we can hope to be able to understand.

Notes 1 http://america.aljazeera.com/articles/2014/6/26/oklahoma-­new-earthquakecapital oflower48.html [accessed September 7, 2015].

Language, Norms, and Linguistic Norms

99

2 As I’ve discussed elsewhere, there are really two separable forms of idealism and thus two idealist traditions in modern Western philosophy. One of these, associated with empiricist thinkers such as Berkeley or Hume, is fundamentally epistemological, motivated primarily by the conviction that we have direct epistemic access only to our own mental states and worried that we have no right to infer the existence of anything different in kind from them. The idealist tradition I am concerned with is more directly ontological, motivated by the conviction that primary being must have the (active) structure of a mind. The structure of mind is then understood in terms of the fundamental cognitive/conative activities: reasoning (both theoretical and practical) and judging, both of which employ the fundamental elements of mental activity: concepts and sensations. Even apparently non-­mental things exhibit structural features and relations that can be assimilated to such mentalistic structures. As Hegel might put it, “Substance is subject.” My earlier discussion is in deVries (2009). 3 Before the rise of modern logic, though, not everyone thought of inference as a relation between propositions. Despite their otherwise immense differences, both Hume and Hegel think of inference as primarily a relation among ideas or concepts. 4 I follow Sellars here in using “shall” and “shall be” as operators that signal an expression of intention, which differs from both a description of intention and an attribution of intention. Practical reasoning combines one’s antecedent intentions with one’s understanding of the factual situation to generate new intentions and, eventually, volitions. 5 We could also get into complex metaphysical arguments about the status of possible modal statuses. But that is not a direction I want to explore here. 6 There is a standard set of abbreviations for Sellars’ works that I use, many of which have been reprinted in several places (see http://www.ditext.com/sellars/bib-­s.html). The bibliography spells out the relevant abbreviation for the works cited here. I give section or paragraph numbers when possible, as well as page location in specific texts. 7 Rebecca Kukla (2000). 8 I have worked out my interpretation of Hegel’s response to Kant on teleology in deVries (1991). It is available as well on my website: http://pubpages.unh.edu/~wad/ Archive/archivefrontpage.htm 9 Mark Okrent asks “How can teleology and agency make naturalistic sense if it isn’t reducible to beliefs and desires, and intentions treated as causes?” He points out that there are several different ways to try to answer this. My own view is that philosophers have gotten overly narrow in their interpretation of causation, thinking that all causal explanation must be in terms of mechanical causation, whereas the manifest image notion of causation is not so narrowly constrained. Not all good explanations are mechanical. There is some indication that Sellars himself thought that restricting causation to the mechanical was overly narrow. See FMPP, III (“Is Consciousness Physical?”), section VI: 83–4.

100

Wilfrid Sellars, Idealism, and Realism

Bibliography deVries, Willem A. (1991), “The Dialectic of Teleology,” Philosophical Topics, 19 (2): 51–70. —— (2009), “Getting Beyond Idealisms,” in W.A. deVries (ed.), Empiricism, Perceptual Knowledge, Normativity, and Realism, Mind Association Occasional Series, Oxford: Oxford University Press, 211–45. Hegel, G.W.F. (1934), Wissenschaft der Logik, ed. Lasson, Hamburg: F. Meiner, —— (1969), Hegel’s Science of Logic, trans. A.V. Miller, London: George Allen & Unwin. Kukla, Rebecca (2000), “Myth, Memory and Misrecognition in Sellars’ ‘Empiricism and the Philosophy of Mind’.” Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition, 101 (2/3): 161–211. Sellars, Wilfrid   CIL —— (1948), “Concepts as Involving Laws and Inconceivable without Them,”   Philosophy of Science, 15: 287–315.   EPM —— (1963), “Empiricism and the Philosophy of Mind,” in Science,  Perception and Reality. (1991) Atascadero: Ridgeview Publishing Company.   FMPP —— (1981), “Foundations for a Metaphysics of Pure Process” (The Carus  Lectures), The Monist 64: 3–90.   ISR —— (2007), In the Space of Reasons: Selected Essays of Wilfrid Sellars, eds.  Kevin Sharp and Robert B. Brandom, Cambridge MA: Harvard University Press.   LTC —— (1969), “Language as Thought and as Communication,” Philosophy and  Phenomenological Research, 29: 506–27.   SPR —— (1963), Science, Perception and Reality, London: Routledge &   Kegan Paul. [Reissued 1991 Atascadero: Ridgeview Publishing Company.]   SRLG —— (1951), “Some Reflections on Language Games,” Philosophy of Science,  21: 204–28. [Reprinted in ISR; reprinted with significant additions in SPR.]   TC —— (1962), “Truth and Correspondence,” The Journal of Philosophy 59:  29–56. Reprinted in Sellars (1963), Science, Perception and Reality, London: Routledge & Kegan Paul [Reissued 1991, Atascadero: Ridgeview Publishing Company], 197–224.

Part Two

Psychological Nominalism and Idealism

5

On the Pittsburgh School, Kant, Hegel, and Realism Tom Rockmore, Peking University (CN)

In “Epistemology and the Philosophy of Mind,” Sellars defends a representational form of Kantianism in part through psychological nominalism. The Pittsburgh School further develops this doctrine in their turn to Hegel. Though they are interested in Hegel, the members of the Pittsburgh School are strong realists, hence defend a form of realism incompatible with Kantian idealism as well as his German idealist successors. With this in mind, I will be arguing three points: First, this way of reading the critical philosophy, which is restated and developed by the Pittsburgh School, is incompatible with the constructivist reading, which persists in Hegel. Second, this way of reading Kant fails to solve the epistemological problem in leading to a blind alley, which Kant properly rejects and which is incompatible with Hegel’s position. And, third, correctly understood, Hegel, in building on Kant, suggests another and more interesting epistemological approach. One consequence of this analysis is to reject strong or metaphysical realism in favor of empirical realism.

Sellars’ Kantianism and psychological nominalism Wilfrid Sellars is an unusual early analytic Kantian, committed to a reformulation and defense of selected Kantian doctrines. “Psychological nominalism,” a theory apparently first put forward by Sellars, suggests language is an indispensable means through which humans learn the types of socially shared practices that permit rationality. This is a norm-­governed theory in that norms influence and shape the practices that render language and thought cognitively meaningful. Psychological nominalism incorporates views traditionally associated with idealism, but its main proponents—Sellars, Robert Brandom,

104

Wilfrid Sellars, Idealism, and Realism

and John McDowell—are all strong realists. But observers sympathetic to German idealism find their appropriation of German idealism problematic. In one sense, the cognitive problem, which interests Sellars, is a form of the difficulty Kant faces after Hume in reestablishing the claim for knowledge. In another sense, the cognitive problem is more general, and concerns the unclear relation of idealism and realism. This relation is unclear since there is no agreed upon view of “idealism” or “German idealism” and “realism” is used in many different ways. In reacting against Berkeley, this problem is raised by Kant and later by a number of observers. They include Fichte, who simply rejects the Kantian conception of the thing in itself, or mind-­independent reality, as well as F.H. Jacobi. The latter famously objects that the view of the thing in itself is unintelligible but central to the critical philosophy. He asks: “How is it possible to reconcile the presupposition of objects that produce impressions on our senses, and in this way arouse representations, with a hypothesis intent on abolishing all the grounds by which the presupposition could be supported” (Jacobi 1994, 337)? A version of this post Kantian query remains important in Sellars’ view. Sellars, who was interested in Kant at a time when this was unusual among analytically-­trained philosophers, later described his main philosophical concern as “to formulate a scientifically oriented, naturalistic realism which would ‘save the appearances’ ” (Sellars and Castañeda 1975, 289). This is a recognizable restatement of the Kantian concern with knowledge as based on an interaction between subject and object, or in McDowell’s language between Mind and World. In the important Herz letter, Kant famously states the question that later occupies him in writing the Critique of Pure Reason as a question: “I asked myself . . . on what grounds rests the reference of what in us is called representation [Vorstellung] to the object [Gegenstand]” (1977, 177)? Kant’s question, which is ambiguous, can be read in two main ways: as calling for a representational theory of perception of the mind-­independent real, or reality, as his formulation in the Herz letter suggests, or, on the contrary, as suggesting a constructivist view of cognition, in short some version of the claim that a necessary condition of knowledge is to construct the cognitive object, which runs counter to a representational theory of perception. I come back to this view below. The question of how best to interpret the critical philosophy remains undecided. In part because Kant appears to have difficulty in making up his mind, different interpretations find support in the texts. Numerous Kant scholars prefer a representational interpretation. Thus Allison’s widely known defense of Kantian idealism argues in favor of a so-­called two aspects reading of the critical philosophy in which appearance and reality are two sides of the same object so to

On the Pittsburgh School

105

speak (2004). Sellars’ statement also points toward a representational reading of the critical philosophy, more specifically toward an approach to knowledge in which there is a distinction between appearances and reality, and under appropriate conditions we can claim that the former correctly identifies the latter. In “Empiricism and the Philosophy of Mind,” where Sellars introduces the logical space of reasons, he can be understood to be pursuing a successor version of Kant’s critical philosophy, understood as a representational approach to cognition. Sellars, who follows Kant in rejecting claims for direct or immediate cognition, rejects the standard empiricist view that knowledge can be analyzed into non-­epistemic facts.1 Yet he does not reject but rather takes up a version of the Kantian task. He continues down the Kantian road in striving to realize a recognizable version of the critical philosophy through his suggestion that the space of reasons allows us to justify linguistic claims about the world as given in experience.2 Sellars supports the latter point in introducing psychological nominalism (EPM §29, 31), or the cognitive thesis that “all awareness of sorts, resemblances, facts, etc., in short all awareness of abstract entities—indeed, all awareness even of particulars—is a linguistic affair” (§29). He can be understood as denying “that there is any awareness of logical space prior to, or independent of, the acquisition of a language” (§31). Language and logical space are related concepts. For only an individual who knows how to use natural language can function in the logical space. In EPM, Sellars applies his view of psychological nominalism in treating thoughts—here following McDowell’s unpublished paper—as “inner perceptual episodes with conceptual content” (§6) in countering the so-­called Myth of the Given with his account of the mythical Jones. McDowell understands this to mean that through the appropriate use of language individuals refer to the way things are. According to McDowell, “ ‘Inner’ episodes with perceptual content are to be understood on the model of overt performances in which people, for instance, say that things are thus and so. The directedness at reality of overt verbal behavior affords the model on which we are to understand the directedness at reality of non-­overt conceptual episodes.”3

Psychological nominalism, the given, and the Pittsburgh School The Pittsburgh School aka the Pittsburgh Hegelians or as the Pittsburgh neoHegelians, is often associated with Sellars, McDowell, and Brandom, but oddly

106

Wilfrid Sellars, Idealism, and Realism

not with Rescher. The latter is arguably closer to idealism, closer as well to German idealism, and, hence, since Hegel is a German idealist, closer to Hegel.4 The Pittsburgh School features a series of readings of the conception of the given by Sellars and others in related efforts to work out an acceptable approach to cognition after the given, for instance in relying on psychological nominalism. Sellars professes to abandon the idea of the given in his important text on “Empiricism and the Philosophy of Mind” (EPM). If, to use current jargon, the given is empirical, or exogenous, and not nativist, or endogenous, then to abandon the given is part, perhaps even the central part, of abandoning any recognizable form of the empiricist approach to knowledge. Following many others, I take this to mean some form of the view, routinely identified with British empiricism, that knowledge comes only or at least primarily from experience. Conversely, to contest giving up the given is to manifest a concern to hold on to in whole or in part a recognizably empiricist approach to knowledge. The idea of the given, which is now firmly identified with Sellars, was anticipated earlier by others, for instance C.I. Lewis. The latter famously distinguishes between sensations, through which objects are given to us, and categories lodged in the understanding through which they are thought (Kant 1998, 193). Though he was strongly influenced by Kant, Lewis rejects Kant’s famous claim that “intuitions without concepts are blind” (1998, 193–4). In Mind and the World Order, his most important work, he draws attention to a quasiKantian distinction between the given or immediate sense data about which one cannot be mistaken, the interpretation of the given, and the concept through which we interpret it.5 In following Kant, Sellars in turn rejects any form of the given. The given is the hallmark of empiricism. Kant rejects the given in turning from empiricism to a categorial approach to experience and knowledge. In rejecting the given, Sellars distantly follows Kant down the epistemological path. Unlike Kant, who relies on categories, or concepts of the understanding, Sellars relies on linguistic competence. I believe no one can say with certainty precisely how Sellars understands the given, though it is at plausible that he intends to counter Lewis and others committed to empiricism. Moore and Russell, the founders of analytic philosophy in England, were broadly empirical thinkers. Sellars’ attack on the given loosely belongs to the analytic attack on empiricism underway roughly since the later Wittgenstein. The analytic turn against empiricism featured in recent analytic philosophy was anticipated in Kant’s critique of Humean naturalism. For naturalism, he substitutes a transcendental approach to knowledge. In relying on

On the Pittsburgh School

107

the supposed parallelism between judgments and categories, he proposes a transcendental deduction of the latter. Kant relies, in going beyond the so-­called Aristotelian rhapsody, on an a priori deduction of the categories. If it were successful, the transcendental deduction would provide Kant with a finite and exhaustive set of non-­normative categories. Yet it is difficult to understand how a deduction of the categories would count as an answer to the justification of claims to know, if knowledge requires a grasp of mind-­independent reality, whose possibility Kant denies. A possible answer is available in Kantian constructivism. Constructivism is the view, which arises in ancient Greek geometry, and comes into modern philosophy through Hobbes, Vico, and Kant, that we know only what we in some sense “construct.” Kant formulates a version of this view in a famous passage in distinguishing between the traditional view that cognition must conform to objects and the new view that objects must conform to cognition. Kant, who thinks that the assumption “all our cognition must conform to the objects . . . [has] come to nothing” proposes as an experiment the opposite assumption, the centerpiece of his Copernican revolution, “that the objects must conform to our cognition” (Kant 1998, 110). Kant’s Copernican view is constructivist. He is not claiming to know the mind-­independent world as it is. He is rather claiming to know the mind-­ dependent world, which we construct and therefore can know. Kant’s Copernican view is sometimes given a semantic interpretation. According to Hanna: Kant’s Copernican Revolution of 1781–7 is in this way an all-­things considered answer to the fundamental semantic question he raised in 1772: how can mental representations—and more specifically necessary a priori mental representations— refer to their objects. And the answer is that mental representations refer to their objects because “objects must conform to our cognitions”; hence our true a priori judgments are necessarily true independently of all sense experience because they express just those cognitive forms or structures to which all the proper objects of human cognition automatically conform. Hanna 2001, 22

Hanna’s interpretation points to part of the answer, but not to the whole answer. Kant is correct that we can only experience content corresponding to our mental capacities, for instance colors belonging to the visible spectrum. Yet it does not, therefore, follow that mental objects correctly represent mind-­independent reality. Hence it seems that Kant after all does not answer the so-­called semantic question about how a priori mental representations refer to their objects.

108

Wilfrid Sellars, Idealism, and Realism

Kant, Sellars, and the given Kant tackles the question of knowledge after the given, hence knowledge after empiricism, on a priori grounds. Sellars and analytic thinkers influenced by him tackle this question on a posteriori grounds. The Pittsburgh neo-Hegelians offer different but related takes on the problem of knowledge after the given. Sellars’ view of the given can be read in different ways. According to Reider, in “Empiricism and the Philosophy of Mind” Sellars is concerned with, among other possibilities, at least three possibilities: the views of realists who claim that we “see” universals and their logical relations; the views of rationalists who, on the contrary claim, that we do not cognize universals or their logical relations but are naturally endowed with an understanding of both; and traditional empiricists who claim that the mind can immediately (and inherently) transform sensory content into universal content and their logical relations (2013, 4). Sellars appears to be primarily concerned with closing off the possibility of traditional empiricism. DeVries and Triplett describe Sellars’ view of the given as follows: “The general framework of the givenness consists of the assumption that there are epistemic primitives—beliefs or other mental states that have some positive epistemic status but that are noninferential, conceptually simple, and epistemically independent and efficacious” (deVries and Triplett 2000, 7). Sellars’ suggestion to turn from the given toward the space of reasons amounts to giving up empiricism, a main approach to cognition, in favor of the space of reasons as a replacement concept. The given can be understood in different ways, not all of which are empirical. Plato can be read as suggesting that on grounds of nature and nurture a selected part of the population has intellectual intuition of the forms, or mind-­independent reality, which is, hence, in selected circumstances, given so to speak since according to Plato philosophers can intuit reality. Reliance in different ways on the given is presumably a central feature of modern philosophy. Descartes, for instance, apparently relies on a version of the given since in appropriate circumstances we can infer from ideas to the world. Locke, perhaps the paradigmatic empiricist, holds that simple ideas match up one to one with the world. Yet as part of his distinction between sensation and perception, Kant abandons the effort to base claims to know on sensation, hence on what is given. In the critical philosophy, claims to know cannot be justified through the relation of experience and knowledge to sensations. Unlike Locke and perhaps Descartes, Kant appears to anticipate what Husserl calls psychologism. In the critical philosophy, the logical cannot be reduced to and must be sharply separated from the psychological. In other words, Kant’s anti-­psychologism, to

On the Pittsburgh School

109

employ Husserl’s term, commits him to turning away from the given. Whether that is the case for Husserl as well falls outside the scope of this paper.

Sellars’ view of knowledge after the given Sellars, Brandom, and McDowell all participate in the analytic rejection of empiricism. I take them to be making three very different claims for knowledge on the basis of their shared agreement about Sellars’ rejection of the given. Sellars’ positive approach to knowledge is broadly Kantian in sharply distinguishing receptivity and spontaneity. In an important passage in EPM he writes: “The essential point is that in characterizing an episode or a state as that of knowing, we are not giving an empirical description of that episode or state; we are placing it in the logical space of reasons, of justifying and being able to justify what one says” (EPM §5, 131). Sellars, who is a scientific realist, is, hence, not a cognitive skeptic. Yet he rejects traditional empiricist claims for knowledge. According to Sellars, any claim for knowledge of reality, or the way the world is, say through epistemic intuition, is problematic, and must be rejected. I take Sellars to be abandoning the given but not as abandoning the view that to know is to know reality. I further take him to be claiming that since there is no given, the given is a myth, and in its place we must rely on the very briefly evoked so called space of (scientific) reasons to cognize reality, or in informal language the way the world is. One of Sellars’ central goals is apparently to unite the conceptual behavior of the “space of reasons” with subjective sense experience. Some forms of empiricism (e.g. H.H. Price) suggest that there are certain identifiable, non-­impeachable facts. The rejection of the idea of the given obviously does not mean there is no subjective sense experience, but rather that, unlike the empirical subject, or perhaps better the empiricist’s subject, the post-­empirical subject, which turns away from the given, cannot rely on it for cognitive purposes, hence needs other cognitive sources. In virtue of his reliance on the so-­called space of reasons, Sellars gives up empiricism as ordinarily understood since he thinks that the traditional empiricist analysis of epistemic facts into non-­epistemic facts cannot succeed. “Now the idea that epistemic facts can be analyzed without remainder— even ‘in principle’—into non-­epistemic facts, whether phenomenal or behavioral, public or private, with no matter how lavish a sprinkling of subjunctives and hypotheticals is, I believe, a radical mistake—a mistake of a piece with the so-­ called ‘naturalistic fallacy’ in ethics” (EPM §5, 131).

110

Wilfrid Sellars, Idealism, and Realism

Sellars further rejects both foundationalism and Hegelianism, which, on his reading, is allegedly committed to givenness, but which he does not further characterize. “One seems forced to choose between the picture of an elephant which rests on a tortoise (What supports the tortoise?) and the picture of a great Hegelian serpent of knowledge with its tail in its mouth (Where does it begin?). Neither will do. For empirical knowledge, like its sophisticated extension, science, is rational, not because it has a foundation but because it is a self-­correcting enterprise which can put any claim in jeopardy, though not all at once” (EPM §38, 170). Sellars rejects traditional empiricism, traditional foundationalism, and traditional coherentism in favor of what he calls the logical space of reasons. The term “psychological nominalism” suggests one of the dimensions of the “logical space of reasons” is linguistic competence. Sellars can be taken as suggesting that to use language correctly in referring to reality we go beyond simply describing the contents of consciousness in being able to justify what one says in claiming that reality is thus and so. In short, Sellars is not claiming that mere observation is sufficient since in rejecting the view that observational knowledge “stands on its own feet” (EPM §36), he rejects traditional empiricism. He is rather pointing to the way that being able to give inductive reasons today “is built on a long history of acquiring and manifesting verbal habits in perceptual situations . . .” (EPM §37). Thus according to Sellars part, indeed an important part of knowledge of reality without falling into circularity, is to use language correctly in cognitive situations. How does the correct use of language, even allowing coherence, justify cognitive claims about reality? McDowell points out the difficulty linked to coherence. The so briefly limned view of the space of reasons relies on the interrelation of concepts in a conceptual framework. Yet since the coherence in question cannot rely in any way at all for its justification on the given,6 it is an instance of what McDowell, who is largely sympathetic to Sellars, calls “unconstrained coherentism” (1994, 143). Sellars, who abandons the given, relies on linguistic competence and coherence to justify claims to know. Yet a theory can be coherent but false. Many individuals in mental institutions have coherent worldviews. Others go into politics. If we cannot rely, as McDowell suggests, on mere coherence, can we rely, as psychological nominalism suggests, on linguistic competence? A clever Sellarsian could argue, rightly so, that science, which differs from the folk model, in probing experience relies on the stringent application of techniques and technologies elaborated over centuries to support is cognitive claims. That is all to the good, and bolsters scientific claims to cognition. Yet it seems clear that with or without

On the Pittsburgh School

111

reliance on rigorous science, a correct use of words is necessary but not sufficient for cognitive purposes. Linguistic competence or even, if there is a difference, using words correctly does not permit a justified inference from what one thinks is the case to what is the case, nor a justified inference from appearance to reality. This suggests that though one gives up the given in one sense as sufficient in itself to justify epistemic claims, there is no alternative to retaining it in another sense as a verifiable limit on our cognitive claims. In short, if “reality” means that the world is thus and so, then neither coherence nor linguistic competence either separately or together seems sufficient to make out claims to know reality.

Brandom, McDowell, and knowledge after the given Brandom and McDowell, the main members of the Pittsburgh School, read Sellars differently, and, on that basis make very different cognitive claims about knowledge of reality. Brandom asserts that the same reality makes our concepts true or false with respect to normative concepts. McDowell suggests that cognitive claims can be falsified through external constraints, as in a fallibilist approach. I distinguish the weaker view that a cognitive assertion about reality can be falsified from the stronger view that a cognitive assertion about reality can be shown to be true. It is possible that no cognitive claim can be shown to be true but at least some cognitive claims can be shown to be false. In the logical space of reasons Sellars, as noted, relies on coherence and psychological nominalism. Brandom goes beyond either coherence or psychological nominalism in pointing to an inferentialist relation between concepts and the world. Inferentialism is often understood as inferential role semantics, or as an approach to theory of meaning espoused by Brandom, Gilbert Harman, Paul Horwich, Ned Block and others. In inferential role semantics, the meaning of an expression is related to the meaning of other expressions, hence suggesting some form of holism. Brandom’s form of inferentialism apparently centers on two main claims: first, concepts are normative in determining what counts as reasons for particular beliefs; and, second, the mind-­independent real “makes” the conceptual structure true or false by telling us how it is with the world. The latter claim suggests that under appropriate conditions we can and do cognize reality or the mind-­ independent world, not merely as it appears but as it is. Here a reference to Peirce may be helpful. Peirce introduced the idea of abduction, which is often understood as a form of logical inference from

112

Wilfrid Sellars, Idealism, and Realism

so-­called data description to a hypothesis accounting for the data. This suggests that theories, which aim at, but cannot lay claim to truth, are hypotheses based on experience, which are subject to later correction. Hegel, who influenced Peirce, defends an analogous view. Brandom seems to make a stronger, more problematic claim than either Peirce or Hegel, a claim which points to some kind of traditional analytic verificationism. For instance in Making It Explicit, he writes: A semantically adequate notion of correct inference must generate an acceptable notion of conceptual content. But such a notion must fund the idea of objective truth conditions and so of objectively correct inferences. Such proprieties of judgment and inference outrun actual attitudes of taking or treating judgements and inferences as correct. They are determined by how things are, independently of how they are taken to be. Our cognitive attitudes must ultimately answer to these attitude-­transcendent facts. 2001, 137

Brandom seems to be suggesting that concepts must fund their own verification as either true or false. Verification, which was earlier popular, is now regarded as problematic, even as incoherent by such former enthusiasts as Hilary Putnam.7 It is difficult, perhaps not possible, to formulate a coherent verificationist theory and there is no reason to hold verificationism can be successful. Verificationism which arose in the 1920s in relation to led to a spirited debate that led to the final abandonment of the project. A.J. Ayer’s early formulation of verificationism presupposed three elements: the analytic/synthetic distinction, the obervation/ theory distinction, and the empirical criterion of meaning. Quine’s later attack on the analytic/synthetic distinction is often regarded as undermining this version of verificationism. There is a further a philosophical difficulty, which apparently undermines any form of verificationism. In Brandom’s position, verification consists in ascertaining which concepts are true through confronting them to reality. This is no more than a promissory note, which cannot be redeemed. The insuperable difficulty lies in going from fact-­dependence to objectively correct inferences through relevant conceptions of normativity. That an inference is objectively correct according to prevailing normative standards is unrelated to what is really correct about the real. In short, the suggestion that inferences respect the facts according to the best current standards is insufficient to show they succeed in grasping how things are. In other words, as Kant points out in his Copernican turn, no one has ever been able to show how one can correctly infer from appearance to reality.

On the Pittsburgh School

113

This crucial distinction is sometimes overlooked, by Brandom as well. Consider further the following passage from Articulating Reasons, where he distinguishes physical, chemical, and biological things that supposedly have natures, and presumably are the stuff of science, from those, like English romantic poetry that have histories, Brandom holds that our concepts about such objects can be known to be either true or false: For the properties governing the application of those concepts [that is, electrons and aromatic compound] depend on what inferences involving them are correct, that is, on what really follows from what. And that depend on how things are with electrons and aromatic compounds, not just on what judgments and inferences we endorse. 2001, 27

This passage apparently conflates the crucial difference between concepts that are normatively correct with how things stand in reality. Yet it does not follow that if inferences are normatively correct, that is that they are acceptable according to current standards, which may later change, that they are correct about the world. At most, certain inferences about the world are plausible in that they are based on the information we currently possess and they respect our current standards. No one is suggesting that there is not a mind-­independent reality, which science seeks to know. Yet there seems not to be any persuasive way to claim to know it as it is. The suggestion, say, that natural science is “correct” about the world is perhaps a necessary hypothesis with respect to natural science but obviously indemonstrable. Natural science includes a collection of theories that, since they are merely theories, can presumably be falsified. It makes commonsense to invoke experience to point out that a theory is incorrect. It does not make commonsense to invoke experience to claim a theory is correct if that means it in fact tells us the way the mind-­independent world really is. McDowell is critical of both Sellars and Brandom, but more sympathetic to the former and increasingly more critical of the latter. He invokes in passing what he calls “a rational constraint on empirical theories” (McDowell 1994, 8–9) to signal a tactical retreat in retaining just enough of the given to avoid abandoning all cognitive claims about experience as well, but not enough to base cognitive claims on the given, for instance as that was traditionally done in empiricism. According to McDowell, the idea of the given is that the space of reasons is wider than the conceptual space (1994, 7). McDowell, who opposes theories that leave rational constraint outside the space of thought (1994, 15), wants a constraint that is outside spontaneity but not outside the space of

114

Wilfrid Sellars, Idealism, and Realism

concepts. According to McDowell, “a belief or judgement to the effect that things are thus and so—a belief or judgement whose content (as we say) is that things are thus and so—must be a posture or stance that is correctly or incorrectly adopted according to whether or not things are indeed thus and so . . . This relation between mind and world is normative, then, in this sense: thinking that aims at judgment, or at the fixation of belief, is answerable to the world—to how things are—for whether or not it is correctly executed” (1994, xi–xii). Unlike Sellars (and Davidson), he apparently thinks that the “belief that an object has an observable property can be grounded in an impression itself: the fact’s impressing itself on the subject” (McDowell 1994, 145). In short, he seems to think that causal relations in fact disclose the world, not, to be sure in the sense they reveal what is is, but in the more minimal sense of revealing what it is not (McDowell 1994, 145). McDowell insists that the empirical constraint must be inside the space of reasons. Yet it must paradoxically also be outside the space of reasons to function as a constraint. In German idealism, Fichte anticipates both this problem as well as the solution through the concept of Anstoss. According to Fichte, the world, which impinges on us, is paradoxically both dependent on as well as independent of the subject.

On a Hegelian approach to experience and knowledge Sellars, who understands “givenness” as equivalent to the Hegelian term “immediacy” (EPM §1), further thinks that a commitment to the given affects “dogmatic rationalism,” “skeptical empiricism,” and, without argument, even Hegel (EPM §2). I say without argument since he does not explore the latter’s position, which in the Phenomenology of Spirit begins through an analysis and rejection of immediacy under the heading of sense-­certainty. Sellars and Hegel differ with respect to immediacy, which Sellars apparently rejects in favor of the logical space of reasons but on which Hegel builds as the initial, but insufficient step in a naïve approach to cognition. From a Hegelian perspective, the problem is not to give up the given in simply discarding the empirical dimension of experience but rather to understand the relation of judgments, hence concepts, to it. Kant, for instance, recognizes that a theory of knowledge must contain both a subject pole, that is, what the subject contributes in the form of mental activity, as well as an object pole, or what the object contributes through a causal relation. Neither is sufficient, and the

On the Pittsburgh School

115

difficulty, which Kant is never able to resolve, lies in bringing them together in a single coherent theory. I am sympathetic to the idea that, under appropriate conditions, causal relations serve as reasons supporting conceptual frameworks, hence have epistemic force in disclosing what we take to be the world. Modern science depends on the assumption that we disclose what through hypothesis we take to be the world through an appropriate analysis of causal laws. That does not mean that causal relations in fact disclose the world. That would only be true if we could reliably represent reality, which simply cannot be shown. McDowell criticizes Davidson in arriving at his view. According to Davidson, the world outside our thinking exerts a rational causal influence on it, an influence through which we can triangulate a common world as it were. For McDowell, the world exerts a rational influence on our thinking since it is inside the conceptual framework (1994, 34–5). Kant correctly tells us that in a sense the world is both inside and outside our conceptual framework, since it is both represented as well as constructed as it were. This claim allows us both to make sense of knowledge while avoiding what McDowell mistakenly takes to be the idealist view of slighting the independence of reality (1994, 34). The solution lies in adopting a different view of the difference between impressions and appearings, or causes and effect. Unless we can claim to know reality, we cannot know it appears, nor know that our views of reality correspond to it. It follows that the suggestion that our views correspond to reality is regulative but cannot be constitutive. We can do no better than to compare our views of the real with what is given in experience in continually adjusting the former in the light of the latter. On this view, which I take to be Hegelian, concepts or theories arise within the effort to come to grips with the contents of experience, and are either refuted or temporarily confirmed by further items of experience. This approach has the advantage of not reducing concepts to experience, and not giving up the conceptual value of experience, in bringing together both within the cognitive process.

On the disjunction between analytic neo-Hegelianism and Hegel This paper has so far examined the cognitive claims of Sellars, McDowell, and Brandom, who each lays claim to Hegelian or neo-Hegelian status. The former describes his attack on the given as méditations hégeliennes; McDowell indicates

116

Wilfrid Sellars, Idealism, and Realism

Mind and World is a prolegomenon to a reading of the Phenomenology; Brandom refers to the Pittsburgh neo-Hegelians. I infer that the Pittsburgh School as currently constituted thinks of itself as Hegelian. This claim is not unimportant. If the members of the Pittsburgh School were Hegelian or neo-Hegelian in a still Hegelian sense, that is in adopting views Hegel recommends or might recommend, then, since Hegel is a German idealist, German idealism would continue and perhaps even reach a new peak in the Pittsburgh School. Some observers, such as Paul Redding, clearly think this is so (2007). On the contrary, I believe the Pittsburgh School is neither Hegelian nor neo-Hegelian nor even non-Hegelian. On even a sympathetic interpretation it is anti-Hegelian, since its representatives insist on approaches to knowledge Hegel as well as the other German idealists reject. In fact were it to be Hegelian, it would be in conflict with Sellars’ reported intentions. According to deVries, one of Sellars’ main aims in EPM is to counter the empiricist route to idealism (deVries 2009, 211–45). Perhaps this is what Sellars has in mind in reproaching Hegel’s commitment to the given. One way to bring out the veritable gulf between the Pittsburgh School and Hegel concerns the relation between what McDowell aptly calls mind and world. Here a reference to Kant will be useful. In the critical philosophy Kant presents an unresolved contradiction between incompatible representational and constructive approaches to cognition. Representationalism consists in claiming to represent the mind-­independent world, in short to know reality. Constructivism lays claim to construct (or perhaps better to reconstruct) not reality but rather a view of reality as given in experience through concepts. Kant’s so-­called Copernican revolution offers a constructivist approach to knowledge. German idealism turns away from representation and toward constructivism. What we call German idealism can be characterized by the constructivist claim initiated by Kant and further developed by the post-Kantian German idealists: we know only what we “construct.” Epistemological representationalism yields a one-­to-one relation between mind and world in which at least in principle the idea in the mind matches up with reality. Epistemological constructivism yields no more than a holistic theory of experienced reality but not reality in a theory based on and revisable in terms of the information available at a given moment. Now, I do not say that the thinkers in the Pittsburgh School are representationalists in the Kantian sense of the term. Perhaps they are not. Brandom apparently features inferentialism as an alternative to representationalism. Yet, unlike Hegel, they are certainly not constructivists in terms of the letter or even the spirit of their views. Rather each in his own way

On the Pittsburgh School

117

holds that through a version of the space of reasons they successfully recover what was earlier thought to be accessible through the given.

Conclusion: Psychological nominalism, conceptual frameworks, and the given Since Sellars does not write simply, his view of the given remains elusive. Yet this much seems clear: to give up the given is to give up empiricism as ordinarily understood, hence to abandon the possibility of cognitive representationalism, or a representational approach to grasping the way the mind-­independent world is. Kant distinguishes between spontaneity and receptivity. Sellars as well as Brandom and McDowell, the members of the Pittsburgh School, each rely on versions of Sellars’ view of the given in formulating cognitive views featuring a strong distinction between spontaneity and receptivity. Whatever his intentions, Sellars’ attack on the given turns away from a traditional representational approach, perhaps any representational approach,8 while adopting a cognitive approach based on the space of reasons. Hegel, on the contrary, follows constructivism in adopting a position incompatible with any version of the Sellarsian space of reasons understood as an alternative cognitive approach through the logical space of reasons after a rejection of the given. The point can be made in Kantian terms. According to Kant, all cognition necessarily begins in but is not limited to experience. In the critical philosophy, the categorial framework of cognition is deduced prior to and apart from experience, hence in independence of the given. For Hegel, on the contrary, categories, or concepts arise out the effort of the subject to come to grips with the given understood as no more than the contents of consciousness, to come to grips with immediate experience, hence on an a posteriori basis. Hegel, who rejects empiricism as ordinarily understood, is not an empiricist in, say, the classical British sense. Yet he retains an empirical component as the basis of the formulation of the categorial grasp of experience. In part, the difficulty can be situated relative to the Kantian thing in itself. McDowell indicates a desire to domesticate absolute idealism, which he apparently attributes to Hegel. The difficulty is to acknowledge a reality outside the conceptual sphere, in other, more Kantian language, the thing in itself or again the noumenon. Kant needs the distinction between noumena and phenomena. He needs to be able to say that what is given to consciousness is a clue to what lies outside it, and which, through, say, science as well as other forms

118

Wilfrid Sellars, Idealism, and Realism

of cognition, we believe exists but cannot know that we discover. The solution is, as Kant realizes, to claim that we “construct” what we know, where “to know” means at least temporarily to correspond to what is given in experience, and which, if refuted by further experience, as Hegel points out, needs to be reformulated. I believe the most satisfactory version of this constructivist approach, which lies at the center of the critical philosophy, and, since later German idealists react to Kant, hence of German idealism in general, is formulated by Hegel. Hegel restates Kant’s a priori approach to constructivism in terms of the practical relation to experience. Unlike the Pittsburgh “Hegelians,” Hegel does not claim to grasp the real within any form of the so-­called space of reasons. Indeed from his perspective that is not possible. He rather claims that knowledge emerges as a self-­correcting view of what we at any given time and on the basis of empirical constraints take the world to be.

Notes 1 “Now the idea that epistemic facts can be analyzed without remainder—even ‘in principle’—into non-­epistemic facts, whether phenomenal or behavioral, public or private, with no matter how lavish a sprinkling of subjunctives and hypotheticals is, I believe, a radical mistake—a mistake of a piece with the so-­called ‘naturalistic fallacy’ in ethics” (EPM, §5). 2 “The essential point is that in characterizing an episode or a state as that of knowing, we are not giving an empirical description of that episode or state; we are placing it in the logical space of reasons, of justifying and being able to justify what one says” (EPM §36, in SPR 169; in KMG 248). 3 John McDowell, “Sellars and the Space of Reasons,” (unpublished), 5. 4 Rescher has often written on idealism, but not, to the best of my knowledge, on Hegel. See, for his overall view, his trilogy, entitled A System of Pragmatic Idealism, including Human Knowledge in Idealistic Perspective, Princeton: Princeton University Press, 1991; The Validity of Values: Human Values in Pragmatic Perspective, Princeton: Princeton University Press, 1992; and Metaphilosophical Inquiries, Princeton: Princeton University Press, 1994. 5 See C.I. Lewis, Mind and the World Order, New York: Scribner’s, 1929, esp. ch. 2: “The Given Element in Experience,” 36–66. 6 See John McDowell, Mind and World, 14, 15. 7 For recent argument tending to disqualify verificationism, see Hilary Putnam, Philosophy in an Age of Science: Physics, Mathematics and Skepticism, Mario de Caro and David Macarthur (eds.), Cambridge MA: Harvard University Press, 2012.

On the Pittsburgh School

119

8 Sellars’ view evolves. In Science and Metaphysics he can be read, unlike his view in “Empiricism and the Philosophy of Mind,” as taking a representational approach. See Wilfrid Sellars, Science and Metaphysics: Variations on Kantian Themes, Atascadero: Ridgeview Publishing Company, 1992.

Bibliography Allison, Henry E. (2004), Kant’s Transcendental Idealism: An Interpretation and Defense, New Haven: Yale University Press. Brandom, Robert (2001), Articulating Reasons, Cambridge, MA: Harvard University Press. —— (2001), Making It Explicit: Reasoning, Representing and Discursive Commitment, Cambridge, MA: Harvard University Press. deVries, Willem A. (2009), “Getting Beyond Idealisms,” in Empiricism, Perceptual Knowledge, Normativity and Realism: Essays on Wilfrid Sellars, Willem A. deVries (ed.), New York: Oxford University Press. deVries, Willem A., and Timm Triplett (2000), Knowledge, Mind and the Given: Reading Wilfrid Sellars’ “Empiricism and the Philosophy of Mind,” Indianapolis: Hackett. Hanna, Robert (2001), Kant and the Foundations of Analytic Philosophy, Oxford: Clarendon Press. Jacobi, Friedrich Heinrich, and George Di Giovanni (1994), The Main Philosophical Writings and the Novel Allwill, vol. 18, Montréal: McGill-Queen’s University Press. Kant, Immanuel (1977), “Letter to Marcus Herz,” in Prolegomena to Any Future Metaphysics, ed. and trans. James W. Ellington, Indianapolis and Cambridge: Hackett. —— (1998), Critique of Pure Reason, trans. and ed. Paul Guyer and Allen W. Wood, New York: Cambridge University Press, 1998, B 75, 193. Lewis, C.I. (1929), Mind and the World Order, New York: Scribner’s. McDowell, John (1994), Mind and World, Cambridge, MA: Harvard University Press. Putnam, Hilary (2012), Philosophy in an Age of Science: Physics, Mathematics and Skepticism, eds. Mario de Caro and David Macarthur, Cambridge MA: Harvard University Press. Redding, Paul (2007), Analytic Philosophy and the Return of Hegelian Thought, New York: Cambridge University Press. Reider, Patrick J. (2013), “Normative Functionalism in the Pittsburgh School,” in Social Epistemology Review and Reply Collective, 1(12): 4. Rescher, Nicholas (1991), Human Knowledge in Idealistic Perspective, Princeton: Princeton University Press. —— (1992), The Validity of Values: Human Values in Pragmatic Perspective. Princeton: Princeton University Press. —— (1994), Metaphilosophical Inquiries, Princeton: Princeton University Press.

120

Wilfrid Sellars, Idealism, and Realism

Sellars, Wilfrid   EPM —— (1991), “Empiricism and the Philosophy of Mind,” in Science, Perception   and Reality, New York: The Humanities Press.   SM —— (1992), Science and Metaphysics: Variations on Kantian Themes,   Atascadero: Ridgeview Publishing Company. Sellars, Wilfrid and Hector-Neri Castañeda (1975), Action, knowledge, and reality: Critical studies in honor of Wilfrid Sellars, Indianapolis: Bobbs-Merrill.

6

Reading Sellars’ “Philosophy and the Scientific Image of Man,” with Robert Brandom at One’s Side Joseph Margolis, Temple University (USA)

I think of Wilfrid Sellars as a transparently enigmatic figure in contemporary philosophy. The reason—I suspect, more than confidently believe—has to do with Sellars’ being as passionately absorbed as he is with the question, How best to begin philosophically? His passion and argument confound his natural allies; isolate him in his personal limbo; and yet, now, more than fifty years since the opening thrust of his floruit, time has turned back to favor him. I’m aware that this may seem an all-­too-obvious judgment. But then, your reproach may soften a little if I persist, by adding that it remains Sellars’ abiding question—perhaps never satisfactorily resolved—even at the end of his career or among ourselves. Perhaps it’s not a question one should think of as having a determinate answer. But it is certainly not accurately read by either Richard Rorty or Robert Brandom. I mean, it’s the same question that haunts the initial energy of classic pragmatism after the American Civil War and classic positivism after World War I, and (that) Sellars was aware of the triple convergence in himself, when he turned back, early in his career, to consider his father’s congenial philosophical inquiries and, contemporaneously, the lessons of his actual and imagined exchanges with figures like Herbert Feigl, Rudolf Carnap, W.V. Quine, and, especially, C.I. Lewis (with whom he studied)—and, of course, in the real world of philosophical imagination, Immanuel Kant. I therefore hazard the proposition that Sellars’ question is close to being the only question that matters in philosophy, if you construe it in Sellars’ terms, as pointedly featured in the paper, “Philosophy and the Scientific Image of Man” (1962), where it appears as an explicit puzzle regarding the apparently dual (divided, opposed, paradoxically alienated though inseparably conjoined) presence of the “manifest” and “scientific” images of “man-­in-the-­world” (PSIM).

122

Wilfrid Sellars, Idealism, and Realism

In fact, the “images” (or conceptions) are obviously contrived, never quite the views of anyone in particular. Sellars finds the quarrel disjunctively engaged, irresistibly and invisibly (that is, to a degree, uncomprehendingly), by nearly every leading figure of his day, in spite of the fact that (as he sees the matter) the doctrinally opposed positions he himself abstracts are scandalously irreconcilable, formulated so that they cannot escape stalemate, and that the oblique influence of Sellars’ own picture of the contest that joins them confrontationally may actually block any genuine resolution of the apparent agon.1 (It keeps morphing into other unacceptable positions. There seems to be no end to the effort to capture the flux of experience systematically.) What strikes me here is the corollary question: Can we ever escape the contingency of being attracted to such options, if the best players in the field continually fall into the plausible trap of attempting to capture the conceptual ephemera of the philosophical tradition itself—and how would we know we had succeeded? I don’t wish, however, to set this particular puzzle before us as a question to be firmly and finally answered. I draw it from Sellars’ apparent attempt to resolve the puzzle of the “two images,” and the upshot of his unseen corrective penance—and the temptation on our part to provide a better answer to the disjunction itself. It’s a piece of philosophical theater intended to draw us in a new direction. I had met Sellars a number of times under very friendly circumstances and have counted a number of his ablest and most ardent students good friends of mine, perhaps most rewardingly, Hector Castañeda, with whom I came to understand Sellars better than I could possibly have done entirely alone. But I admit that my first readings (plural readings) of the “Scientific Image” paper were completely negative. It seemed to me to be the most inept confrontation between what was described as the so-­called “commonsense” (or “internalist”) metaphysics and epistemology that runs continuously (though not transparently) from the seventeenth- and eighteenth-­century rationalists and empiricists down through the phenomenalism of Rudolf Carnap and C.I. Lewis (the latter, at Harvard, where Sellars had been a graduate student) and the “scientific” (or “externalist”) alternative, distinguished largely by its effort to escape the epistemological primacy of the first “image” (the “manifest image”) by favoring (somehow) the relatively autonomous accomplishments of the formal and natural sciences (and accompanying metaphilosophical reflections as to how that autonomy could possibly be vouchsafed conceptually). The scientific image, I should add, appeared to run, in diverse, often obscure ways, from Kant again,

Sellars’ “Philosophy and the Scientific Image of Man”

123

through the would-­be reductionisms and eliminativisms of the positivists and those they influenced down, say (however tangentially or quarrelsomely), to the proxy Sellars of the “Scientific Image” piece and the would-­be corrective Brandom of our own day. (The would-­be “scientific image” has snookered many an unwary Fregean.) It was only when I grasped the genuine force of Sellars’ question (so I believe) that I was able to relent and read the essay “correctly”—that is, as not depending at all on the deliberately lame formulation (in the essay) of the opposed “images” themselves. That’s to say, once the question was rightly conceived—the question, whether the “internalist” (or “subjectivist”) and the “externalist” (or “objectivist”) analytic idioms might be “reconciled” (Sellars’ term) or could never be more than “joined” or “fused” (also Sellars’ terms) into one “stereoscopic vision” (PSIM, 4)?—whatever that might be—the deliberate crudity of the “two” visions fell away completely and I could see the point of Sellars’ insisting that if “there is one picture to be grasped reflective as a whole, the unity of the reflective vision will be seen as a task rather than an initial datum” (PSIM 4). (On my reading, this is as close as Sellars comes to exposing his deepest suspicions—or fears. He runs with the “scientific image,” but—as I read him—he already suspects that it cannot defeat the “manifest image” satisfactorily.) If you’re at all tempted by this alternative option, let me recommend the additional finger exercise of reading and rereading no more than the first five pages of the essay. Chances are, you’ll see the point of Sellars’ question much more quickly than I did. But then, I’ve come to see it, partly at least, by puzzling over another question, Whether Brandom’s reading of Sellars’ “realism”—for that it what the matter comes to—is entirely, or largely accurate, or misleading or uncertain, or just plain wrong?2 And why? I read Sellars, I may say, largely as a philosophical “collagist”: that’s to say, most of his central papers are “thought-­ experiments” regarding possibilities he’s not (yet) committed to, but wishes to run with, as faithfully as possible, in order to decide what his best options actually are. Temperamentally, Sellars is clearly drawn to an extreme scientism; but, scrupulously, as his career lengthens, he returns (I conjecture) to the need to test a hybrid (if not a mongrel) resolution of the agon of the two “images.” It’s a corollary of this reading that affirms that the “Scientific Image” paper and “Empiricism and the Philosophy of Mind” must be read together. In that sense, the decision of Rorty and Brandom to reissue the second paper as a separate volume is itself a straightforward confirmation of a deep misreading of Sellars’ corpus. I should also add here that the “internalist”/“externalist” disjunction is completely unworkable, incompatible with the Kantian intuitions Sellars favors,

124

Wilfrid Sellars, Idealism, and Realism

and not satisfactorily congruent with Sellars’ own presentation of the “manifest” and “scientific images” sketched in his essay, as if to capture a sense of the “framework” (a notion Sellars and Brandom and Carnap and Lewis and Nelson Goodman seem to share), within the terms of which the inferentialism associated with the resolution of Sellars’ question may be “placed.” And yet, one cannot doubt that Sellars takes the puzzle (he poses) to bear in an important way on the diverse “dualities” he finds problematically entrenched in the “phenomenalist” views of Carnap, Lewis, Quine, Feigl, and others. (I’ll come back to “inferentialism.”) The fact remains that Sellars was unable to settle the question beyond (what I regard as) the completely slapdash resolution appended at the very end of the essay (PSIM 40): the expression, finally, of a sort of good-­humored exasperation, seemingly defeated by a stubborn mistake Sellars could expose only by induction. There’s the point of my dissatisfaction with the entire piece—and with Sellars’ entire venture. Taken seriously, it cannot fail to color our reading of Rorty, McDowell, and Brandom, who, in different ways, also begin with Kant, but not with Sellars’ naturalistic patience. I do indeed believe that the “solution” Sellars provides—such as it is (that is, the lame resolution of the contest between the two images)—is completely unworkable and justifiably discarded (I’m sorry to say). But I also think Sellars deliberately (didactically or for theatrical reasons) left it in that form (congruently with the comic characterization of the two images themselves). He’s acknowledging (correctly, in my opinion) that it’s probably futile to try to demonstrate the validity of scientific realism (though he’s tempted by the “reconciliation” and disappointed that he cannot succeed). The clue I now see leading to the right reading and resolution of Sellars’ puzzle is indeed embedded in the essay itself, though Sellars was (apparently) unwilling to run with it in the cartoon context of the “Scientific Image” piece: that is, with the straightforward import of the discovery that, on the strength of Sellars’ own account, the scientific image was originally (according to the premises given)— had to be—a projection from the vantage of the manifest image, which, “therefore,” outflanks the other claim; hence, that the “duality” of the two images was and remains wrongly characterized as a pair of free-­standing, independent, opposed perspectives (as Sellars actually declares). One might as well say that the manifest image was itself grievously “victimized” by Sellars’ yielding—“unwittingly,” shall we say?—to the disjunctive two-­ness he found favored, and which he opposed (rightly enough) in early Carnap or Lewis (and might have found even more baldly formulated in the “comic” Sellarsian themes professed by literal-­minded readers like Paul Churchland and Daniel Dennett).

Sellars’ “Philosophy and the Scientific Image of Man”

125

Once you see the matter this way—perhaps a theatrical gesture, signaling Sellars’ awareness that his closing resolution won’t do at all—you cannot fail to see that his later pertinent papers draw us closer to a deeper concession to the manifest image. The seemingly reliable “metaphilosophy” associated with the assessment of the two images falls away completely: the heuristic loses its assured purpose, the agon appears to be abstractly and arbitrarily contrived. There is no explicit or reliable link between Sellars’ metaphilosophical “inferentialism” and the putative victory of the scientific image over the claims of the manifest image. There’s a profound gap in the argument, a gap and an error that seems to fill it— room enough for indefinitely many incompatible or incommensurable undertakings (hardly worth bothering with, once the gap is noticed)—that cannot be closed in the manner Sellars appears to favor (which, on my reading, he does not actually favor)! If so, then Brandom, who is drawn to reading Sellars as if Sellars reads himself as committed to the apparent adequacy of the argument of the “Scientific Image” piece, is justifiably uncertain (uneasy, perhaps) about Sellars’ seeming laxity in retreating from the reconciliation option. The point is, there is no assuredly realist solution to be drawn from Sellars’ scientistic vantage, regarding the metaphilosophical conditions under which Sellars’ original question might be satisfactorily answered—which, if it could be provided, would enable us to specify in some improved measure the inferentialist entailments of the right “framework” within which valid empirical inquiry proceeds. (This, I should add, is very close to Brandom’s Fregean thesis, seconded by some of Sellars’ own conjectures, in Science and Metaphysics.) I’m stumbling into terminological niceties here that require a bit of care—but they’re not my terms. You realize that Brandom wishes to retire what he regards as a mistake on Sellars’ part in pursuing a naturalized account of the prospects of scientific realism. He reads Sellars’ “mistake” (wrongly, I say) as somehow committing Sellars to treating the would-­be testing of an adequate such realism as rendering noumenal claims empirically accessible! The noumenal reading cannot be squared with Sellars’ “Kantian” intentions; and, in any case, it’s the comparison between Brandom’s and Sellars’ inferentialisms that counts at all. I oppose Brandom’s reading on the strength of what may be called rhetorical considerations in interpreting Sellars’ actual narrative and argument (especially if we concede that Sellars is, after all, a knowledgeable and respectful reader of Kant’s texts). But, beyond all that, it’s more than reasonable to suggest that if, say, the “metalinguistic inferentialism” that Sellars and Brandom appear to share— which Brandom is more than willing to affirm they share—gains in plausibility, if only to the extent to which the nonmonotonicity of its inferential framework

126

Wilfrid Sellars, Idealism, and Realism

can be manageably “diminished” for all practical purposes. Brandom’s inferentialism might be strengthened thereby, though not yet in terms of Sellars’ original account of “material inference,” which is not strictly logical or confined to “purely” formal or “rational” (as opposed to perceptual or experiential) linkages. Thus, if the doctrine of the “scientific image” were feasibly confirmed (which it can’t be), then Brandom’s own project—responsibly acknowledging the philosophical primacy of inferentialism (read along distinctly Fregean lines)— would be significantly strengthened by Sellars’ own reflections. As things now stand, however, the “failure” of Sellars’ mock agon (the “ Scientific Image” piece) appears to widen and deepen the logical informality of pragmatic inferentialism itself, since just about all reasonably construed “causal,” “normative,” “instrumental,” “interpretable,” “semantic,” “contextual,” “logical” and similar distinctions, applied to what, loosely, we have in mind in speaking of the things that belong to the “manifest image,” are, nomologically, prescriptively, historically, idiosyncratically, rationally, only weakly regular at best and maximally subject to unforeseen “defeasor” intrusions in particular circumstances—which yield in the direction of the manifest image (that is, along anti-Fregean lines). The mongrelizing of realism (if I may call it that)—the sheer multiplication of diverse conceptual (“descriptive”) options in interacting with whatever we encounter in the world—permits us, at one and the same time, to agree with Brandom’s blunderbuss inferentialism: viz. “that autonomous discursive practices essentially, and not just accidentally, involve the association of ranges of counterfactual robustness with at least some [sic] material inferences” (2015, 163), while at the same time insisting that the nonmonotonicity of any would-­be specialized “pragmatic rules” (answering to the concerns just mentioned: causality, normativity, rationality, and the like) are simply too readily defeasible (in the manifest-­image zone, so to say) to be said to have been captured meaningfully as determinate “rules” or “laws” or as rationally “necessary” (in any reliably “Fregean” way). There are at least two unanswered worries here: one, how, if Brandom’s “metalinguistic” inferences (in pragmatic settings) do not conform closely enough with Frege’s model (in the Begriffsschrift) and, in fact, are (as he concedes) decidedly nonmonotonic, can he possibly account for a suitable measure of inferentialist rigor? The other, how, if he applies his inferentialism in pragmatic contexts, can he avoid the extreme informality and guesswork of material inference itself in perceptually and experientially qualified transient contexts? An unpersuaded reader of Brandom’s response might not unfairly add: Well, yes, anything we “say” or “do” meaningfully is bound to be embedded in, or

Sellars’ “Philosophy and the Scientific Image of Man”

127

infected with, some “entailed” or “necessary” implications. But, how do we draw them out and how do we confirm that they hold true and what does it mean to say that they hold true? And, by the way, how tight or loose is the notion of causality or rule or necessity or even agency in any supporting philosophical brief? And if we cannot count on the firmness of our “noninferential,” descriptively responsible empirical or experiential “data,” then how can we count on the validity of the “material inferences” we advance in pragmatic contexts? (You remember, these were Sellars’ original worries about “material inference”: he never showed us how to pick them out validly; he did indeed help us to see how plausible it was to speak as if we might.3) Ultimately, we cannot address material inference regarding the perceptually encountered world in the same way we address the “logic” of mathematical arguments. We’re addressing two very different sorts of question here, though they’re obviously inseparable. I’d say the primary questions are the ones just mentioned, epistemological questions (bearing on the natural world and human agency), which cannot fail to rely on first-­order empirical practices of testing what we believe to be true. I don’t deny that, following Carnap and Frege, both Sellars and Brandom deem “metalinguistic” inferences to be technically restricted to fragments of language, not to first-­order experiences or psychological episodes of any kind. But to say only that is to remain with the rhetoric of the “linguistic turn” or “first philosophy” or the primacy of “inferentialism” (or “deflationism” or different kinds of “naturalism” or “nonnaturalism” or “anti-­descriptivism” and so on); it ignores the epistemological question just posed. Brandom’s sensible, even courageous admissions regarding the drastically loose nonmonotonicity of material inferences are epistemologically pertinent, in pragmatic contexts, not mere rhetorical concessions. Hence, it doesn’t matter (epistemologically) that Sellars asks us to consider the “metalinguistically”— second-­order—inferential expressivity of the conceptual or “natural” “framework” of valid empirical work; what matters is that there cannot be any way to decode such metalinguistic inferences (drawn from what we say—in context) without bringing the empirical content of our first-­order claims and assertions—and nonverbal actions—to bear on the second-­order question (in the context of pragmatic agency). Accordingly, when Sellars says, in agreement with Brandom, who endorses what Sellars says (in “Realism and the New Way of Words,” p.  61), namely: “It is with some hesitation that I speak of these metalanguages as pragmatic, for they have nothing to do with language as expressive or persuasive, or with such other concepts as empirical psychology as have come to be characterized as the subject-­matter of a science of pragmatics,”

128

Wilfrid Sellars, Idealism, and Realism

Brandom is himself caught in an awkward equivocation. For, here, we must ask ourselves whether the supposed “autonomy” of “metalinguistic” inference in pragmatic contexts (bearing on the “rational” or “empirical” structure of the “framework” of pertinent inference) is similar enough, say, to mathematical inference, à la Frege, to justify the claim that its inferential power escapes being “descriptivist,” that is, cast in perceptually or experientially freighted terms. What this suggests is that Brandom may well have put the cart before the horse. (Brandom apparently believes that this is precisely what his colleague, John McDowell, continually but mistakenly chides him about.) As I see the matter, it’s just at this point that Sellars, in the “Scientific Image” paper, effectively overrides the impression of the early “Realism and the New Way of Words” (from which I’ve drawn the comment just cited), which may indeed have been a prescient anticipation of what has come to be called (fatally) the “linguistic turn”—which, at its most extreme, is an undefended “Fregean turn,” whether Sellarsian or Brandomian.4 Both Brandom and Sellars are tempted by a “modular” account of epistemological (or inferential) powers—which, as I read the puzzle, is precisely what Fregean rationalism (applied in pragmatic contexts) comes to: a strictly boundaried disjunction between “rational,” metalinguistic inference and, say, perceptually or experientially qualified contexts. No one, to my knowledge, has made the case for the natural (or the natural and human) sciences—certainly not Noam Chomsky or Jerry Fodor for instance. I more than doubt that it can be done. Strict modularism makes no sense in Kantian terms or, for that matter, wherever cognition is treated holistically. Alternatively put: modularism is entirely heuristic. The semantic and syntactic structures of natural language cannot be treated as modularly disjunctive; and neither can pragmatic inference in metalinguistic and first-­order linguistic contexts. Worries of these sorts take precedence over Brandom’s would-­be emendations of Sellars’ various conjectures regarding different levels of inference in practical life. In fact, the “Scientific Image” paper may be convincingly read as signifying that Sellars is hardly assured that the scientific image can, satisfactorily, ever absorb or displace all of the important resources of the manifest image! I think it entirely fair to acknowledge Brandom’s skill in making the case for his sort of inferentialism, in the face of well-­nigh disabling counter-­intuitions. But it’s hardly a negligible option that the free-­wheeling use of terms like “rule,” “law,” material “necessity” and the like, in pragmatically complex contexts, may well signify that the proposed “regularities” (sometimes said to be “idealized”— as by Brandom himself) are really more like Weberian “ideal types” than robust approximations to any “actual” regularities. (“Necessity” itself comes to mean

Sellars’ “Philosophy and the Scientific Image of Man”

129

little more than “contingency” under pertinent circumstances.) Curiously, Sellars actually suggests that the terms “manifest” and “scientific image” are “designed to illuminate the inner dynamics of the development of philosophical ideas, as scientific idealizations illuminate the development of physical systems. From a somewhat different point of view [Sellars continues] they can be compared to the ‘ideal types’ of Max Weber’s sociology” (PSIM 4–3).5 In that case, it seems to me, inferentialism dwindles to an altogether different lesson drawn from an analysis of ad hoc consensual tolerance in the circumstances of practical life, which masquerade as defeasible linguistic or metalinguistic rules, as in keeping up the appearance of there being a genuinely determinate, rational, relatively (or entirely) autonomous rule-­governed order in the life of a particular society.6 But there is no adequate enabling argument (in either Sellars or Brandom) to confirm the supposed autonomy of “metalinguistic” (rationalist) inferentialism. Here, it needs to be said that Sellars and Brandom do indeed share an entire account of what Sellars calls a “pragmatic metalanguage”—even a “pure pragmatics” (an entirely “rational” inferentialism free, say, of perceptual or “descriptivist” encumbrances). Read that as shorthand for a naturalistic rendering of a “pure pragmatic naturalism” to replace Kant’s transcendental capture of truth-­bearing discourse, idealized or imagined in “Realism and the New Way of Words,” for an omniscient but not transcendent being—which, in Brandom’s summation, is (literally) “an expressive pragmatic (subject) naturalism that avoids scientism by rejecting its genus: descriptivism” (Brandom 2015, 95). I understand here that Brandom opposes Sellars’ conviction that a strict scientism (initially bruited in the “Scientific Image” paper) must be validated, if we are ever to justify the empirically descriptive and explanatory work of a “descriptivist” vocabulary (a vocabulary addressed to the “appearances” of the world). Brandom takes this to be a “Bad Idea” and turns instead to a “rational” or Fregean-­like treatment of the “discursive framework” of empirical description. I confess I find the enabling argument largely missing.7 Brandom takes this to be “Sellars’ best wisdom on the topic of naturalism [that is, a naturalized realism] . . . contained in his early ideas about pure pragmatics.” He rejects (in the same passage) what he (now) calls the “object naturalism expressed in the scientia mensura passage” (2015, 95): effectively, Sellars’ stab at what a scientific realism would look like, which, bear in mind, the “Scientific Image” piece fails to confirm. Here, Brandom confronts the adverse (the implied) lesson drawn from the manifest image8—what (as already remarked) he mistakenly treats as Sellars’ mistaken drift into noumenalism. (A heady business.)

130

Wilfrid Sellars, Idealism, and Realism

You realize that Brandom prefers several early papers of Sellars’ to the “Scientific Image” piece: notably, as we’ve just seen, “Realism and the New Way of Words” (which draws in its wake other papers included in Pure Pragmatics and Possible Worlds). Now Sellars was very careful in selecting the papers included in the Science, Perception, and Reality volume (1963). They constitute (still constitute) the first tier of papers that begin to define a proper picture of Sellars’ evolving theory: they cannot be simply overridden by earlier papers therefore; they form, dialectically, a kind of historied collage of deliberately fragmentary pieces moving toward a conjectured unity. (A loosely Hegelian unity, one might say, without a predetermined telos.) As it happens, “Realism and the New Way of Words” (1948) was itself selected (in a similar spirit) to be included earlier (1949), in a reasoned collection (from different hands) of current analytic philosophy, edited by Herbert Feigl and Sellars himself. So that Brandom’s preference must (as we’ve seen) count as a distinct departure from Sellars’ own account. In any case, Sellars never overtakes the “implied” lesson of the “Scientific Image” paper: he never reaches a conclusion on inferentialism that would vindicate Brandom’s would-­be replacement. What I make of the division, principally, is that Sellars is content to model (in a heuristic sense not far removed from Weber’s use of “ideal types”) the inferential complexities of what we say and do; whereas Brandom wishes to precise his model so that it may be used methodologically—to some extent, criterially. But then, it’s difficult to concede the force of Brandom’s argument if the risks of nonmonotonicity (vis-à-­vis inferentialism) are as high as Brandom insists they are: The . . . considerations advanced so far [he says; there are five considerations: pp.  163–5] together entail that epistemically responsible believers face a potentially intractable updating problem. Every change of belief, no matter how small, is potentially relevant to the justification of every prior belief. Acquiring a new belief means acquiring what, for any material inference the believer endorses and relies upon for justification, might possibly turn out to be a defeasor. And giving up any belief means giving up not only a premise that might previously have been relied upon in justification, but also a potential counter-­defeasor . . . 2015, 165

Just try to read Sellars’ evolving corpus of papers according to Brandom’s inferentialist formula: effectively, a sort of Fregean modularism. Disaster all around. I can’t see how the model can be applied criterially: it’s a statement of the idealized intention of inferentialism, viewed as a way of construing the practical

Sellars’ “Philosophy and the Scientific Image of Man”

131

complexity of human life itself; it’s not a formula that can be determinately applied. (Certainly, it’s not preordained to confirm anything like a “Fregean” reading.) Obliquely put: the “failure” of the “Scientific Image” piece effectively blocks any effort to bring “metalinguistic” inference in pragmatist contexts into accord with anything like the would-­be autonomy (meant to yield a “rational” framework for empirical inference) of deriving the “rules” of mathematical reasoning from mathematical practices, if it requires support from a Sellarsian scientism. Practical human life, it seems, defeats any such analogy—which, then, adversely affects inferentialism in the physical and human sciences as well. At the very least, then, considerations of these sorts count toward denying inferentialism the philosophical primacy Brandom favors. If you add to this, even modest admissions about the indeterminacy, vagueness, alterability, or contentiousness, of notions like “causality,” “rule,” “law,” “necessity,” “justification,” and the like, you cannot fail to wonder whether there’s not a better way of approaching Brandom’s problem than by featuring inferentialism (metalinguistic or “framework” inferentialism). Although, to be sure, the minimal admission of the implicative complexity of the “content” of whatever we say or do remains unchallenged. Inferentialism cannot be summarily defeated, but we may not be able to fashion an effective methodology for an inferentialism applied to pragmatic contexts—of the sort Brandom proposes. In that case, inferentialism would not be able to claim conceptual primacy in any sense matching Frege’s model of mathematical reasoning (in the Begriffsschrift). But the contrast the model provides illuminates the profound informality of the inferentialist norms of natural-­language reasoning.9 There must be some common ground, somewhere, by which to bridge our understanding the meaning of mathematical and natural-­language discourse. My sense is that, however attenuated, the bridge must rest finally with the conjunction of perceptual and inferential discourse in natural-­language contexts—not “top-­down,” as in anything like a Fregean model of mathematical thought. I see a parallel here between Sellars’ temptation to believe that the scientific image will ultimately displace the manifest image and the neo-Fregean’s conviction that the “exact sciences” (in principle, perhaps, the whole of science) will go completely “rational” (in the mathematical sense) as opposed to remaining (however tenuously it may seem) experientially grounded. My own intuition has it that, in the “empirical” sciences at least, the “rational” is always “hybrid”: the interpretation of a domain (shared by practical life and practical science) is grounded in perceptual and experiential terms. (This is what is often characterized, nowadays, as the need for “friction.”)

132

Wilfrid Sellars, Idealism, and Realism

It cannot be true that we improve our inferentialist position by ignoring the potential epistemological linkage between the second-­order linguistic analyses (of the framework of empirical work), and the “content” of empirical work itself. But, if so, the resolution of Sellars’ question, posed in the “Scientific Image” piece, remains open to any makeshift solution we might muster and usefully exploit. I find a telltale clue, discreetly insinuated toward the end of the “Scientific Image” piece, that signals the inadequacy of the final solution offered at the close of the essay: To say [Sellars famously explains] that a certain person desires to do A, though it is his duty to do B but was forced to do C, is not to describe him as one might describe a scientific specimen. One does, indeed, describe him, but one does something more. And it is this something more which is the irreducible core of the framework of persons. PSIM 39

Possibly, not quite “the irreducible core”: bear in mind the holism of persons; but certainly a significant part of it. Sentience and sapience (in Brandom’s vocabulary) are not as easily disjoined, among enlanguaged persons, as one might suppose: as far as inferentialism is concerned, the right contrast may lie with the disjunction between sensorily descriptive and nondescriptive (rational or implicative) attributions; sentience among humans normally has a sapient element, and sapience tends to be conceptually embedded in linkages selectively grounded in the sentient. Where Brandom yields an almost unconditional primacy to the sapient over the sentient—when he says, for instance, that, “ ‘experience’ is not one of my words,” he seems to be yielding to that Rorty that Rorty himself ultimately rejects (in the “Twenty-­five Years After” piece), the ideologue who (according to the later Rorty) champions the “linguistic turn” over “experience” but cannot say what it requires epistemologically. It’s entirely possible, acknowledging the fragility of inferentialism, that the pragmatic “rules” of inferential regularities involving the “framework” of “nondescriptive” (rational or logical) inferential linkages (say) may be more readily trusted when qualified, contextually, by salient bits of sentience (sensory perception, say) already memorably associated with (or embedded in) successful sapient guesses.10 Here, sentience (involving humans) cannot, Brandom holds, be disjoined from sapience; and sapience answers, minimally, to our discursive ability to explicate (“give and ask for reasons,” by which to make explicit) the inferential input of what we say and do—in effect, explain why we take what we believe-­true to be true. In this sense, human

Sellars’ “Philosophy and the Scientific Image of Man”

133

sentience is already “rational” and, indeed, well on its way to being “logical” as well. “Framework” is a term both Sellars and Brandom use, recalling Carnap’s innovation, to mark any reasoned choice of (the) “metalinguistic” space within which categorially enabling, rather than merely first-­order empirical, inferences may be rightly drawn. Hence, Sellars invites our attention, at one and the same time, to pertinently embedded inferential linkages that may be easily lost or placed at risk by favoring the hegemony of the scientific image over the manifest image (where reconciliation proves unworkable), or, more specifically, to whatever we may find we are not willing to lose in that way. To offer “fusion” or “addition,” here, signals such a loss or risk—which calls the entire “experiment” into question. I’d say Sellars was drawn to the need for a better solution in actual pragmatic contexts. What Sellars features, and what Brandom takes up from Sellars, is Carnap’s innovation about “metalinguistic vocabularies” that are not tethered (are said not to be tethered) to first-­order empirical description or to some merely parasitic or derivative second-­class role with respect to the first, but which play a putatively autonomous “nondescriptive expressive role” applied to the first by the use of otherwise normally accessible vocabularies (2015, 4–5). (It’s the functionality that counts, not the approximation to inviolable or autonomous rules.) The strongly “rationalist” claims are merely declared—and nowhere confirmed. Thus it is that Brandom cites part of a passage from Sellars’ (early) paper, “Realism and the New Way of Words,” which he associates with Michael Dummett’s slogan “Philosophy of language is first philosophy” and Richard Rorty’s “linguistic turn.” I cite a slightly different part of the same passage in order to strengthen the sense in which the new metalinguistic dimension is taken to be “robust”: . . . “verified,” “confirmed,” and “meaningful” are to be understood [Sellars holds] as predicates belonging in a type of meta-­language the central concept of which is that of a confirmed world-­story [or framework for description]. As a matter of fact, meta-­languages of this type alone are [he adds] meta-­languages in the complete sense of the term, for they alone deal with languages as languages, that is to say, as meaningful symbols. Syntactics and semantics as epistemological rather than empirical disciplines are abstractions from pure pragmatics, and are misunderstood in a way which leads directly to psychologism when their fragmentary character is overlooked. It is with some hesitation that I speak of these meta-­languages as pragmatic, for they have nothing to do with language as expressive or persuasive or with such other concepts of empirical psychology as

134

Wilfrid Sellars, Idealism, and Realism

have come to be characterized as the subject-­matter of a science of pragmatics. Pure pragmatics or, which is the same thing, epistemology, is [please note] a formal rather than a factual area. In addition to the concepts of pure syntactics and semantics, pure pragmatics is concerned with other concepts which are normative as opposed to the factual concepts of psychology as “true” is normative as opposed to “believed,” or “valid” is normative (. . . remember that the use of the term “normative” is tentative) as opposed to “inferred”. 2015, 8911

This captures Sellars’ strategy for naturalizing Kant’s transcendentalism. But it does not confirm the “epistemological” thesis advanced. This is Sellars at his most Fregean—at his most attractive, as far as Brandom is concerned. I’ve cited a longish passage—a little out of sync—first, because it provides a compendious sense of the coherent use of such terms as “pure pragmatics,” “metalinguistic framework,” “pragmatic inferentialism,” “epistemology,” “nondescriptive expressive role,” and the like within a single theory that addresses realism, normativity, the definition of a person, the agon of the manifest and scientific images, the prospects of inferentialism, a naturalistic reading of Kant, and the plausibility of extending our conception of pragmatism to include some of the work of the so-­called Pittsburgh School; and, second, because the running argument shows that, even here, Sellars’ “inferentialism” requires no additional warning regarding the pertinence of an empirical vocabulary already in use. Sellars’ “experiment” (in the “Scientific Image” piece) would make no sense if it did not draw some essential categorial distinctions, inaccessible to the supposed vocabulary of the scientific image, from what Sellars calls the “manifest image,” which, in time, should be completely displaced. So there’s something seriously amiss in Brandom’s summary of what Sellars’ proposal is expected to yield—as well as in Sellars’ own account, construed literally. There cannot be a way of disciplining the inferential use of metalinguistic terms (admitted to function in some nondescriptive way) without reference to some perceptual and experiential distinctions, even if epistemology may be sometimes said to be (on Sellars’ view) directly occupied with fragments of language (and not, ostensibly, with mental or psychological elements). Given the fragility of inferentialism, however, epistemology cannot convincingly be restricted to the use of pieces of language qua language alone, and language itself cannot fail to yield, even within the bounds of its “expressive” metalinguistic functions, some run of “representational” information keyed to first-­order description. Which is to say: the extreme, decidedly premature claims made in the name of the “linguistic turn”—effectively arrayed against admitting the

Sellars’ “Philosophy and the Scientific Image of Man”

135

evidentiary function of “experience” in inferential contexts involving the natural and human sciences and practical life, as in Rorty and Brandom and more equivocally in Sellars—cannot convincingly support any analogy with a Fregean-­ like treatment of meta-­mathematical inference. There is, however, no reason to deny that the inquiries of the Pittsburghers are sufficiently pragmatist in spirit to be so construed as passably pragmatist. Nevertheless, pragmatism is distinctly holist—not modular at all. Much of the formalism Sellars and Brandom favor (in their different ways) may be little more than rhetorical: certainly that would have to be acknowledged if epistemology were itself said to address the realism issue as well as normativity. Let me put the point in the plainest way: we’re coming down to the nerve of the matter. If Sellars’ effort at “naturalizing Kant” is to make any sense at all, then Sellars must, surely, address the question head on as to whether there are or are not any categorially necessary (“Kantian”) constraints on his ontology and epistemology to be drawn from the vocabulary of the manifest image, apparently inaccessible to the scientific image. He must, in short, say whether (for instance) “intentionality” assumes a decisive (but utterly mysterious) role in the closing pages of the “Scientific Image” paper, which may be fairly treated in the perfunctory way it is, though still in terms of the “framework” of pragmatic inference. (I’ll come back to the analysis of what Sellars actually says, but we don’t need all that for the moment.) And, if we are to suppose that when Brandom, in accord with Sellars’ general (“Kantian”) strategy, speaks of embedded, necessary inferential linkages of a categorial sort drawn from a “nondescriptivist” analysis of the pragmatic “framework” of empirical inquiry, isn’t Brandom obliged to explain whether (for instance), and how (where pertinent) his metalinguistic (rational, but not perceptual or experiential) analysis of the supposed inferential structure of “saying” and “doing” (or of “is” and “ought,” or of the “sapience” of persons, or of the practice of “giving/asking for reasons”) can be shown to yield normative necessities that we actually and rationally invoke in the inferential practices of ordinary life. If we adopt the dictum, “naturalizing Kant,” we may take Brandom to mean, both with regard to Sellars’ work and his own (however differently in each and however opposed one to the other), the inclusion of metalinguistic (“rational,” “nondescriptive,” “pure pragmatic”) inferential necessities that belong to the “framework” of empirical inference and truth-­claims. In this sense, Brandom favors Frege over Kant and replaces transcendental necessities with metalinguistic ones. I read this as a pragmatic promissory note, given Brandom’s admissions regarding the vagaries and deep nonmonotonicity of pragmatic inference.12

136

Wilfrid Sellars, Idealism, and Realism

There’s a profound discrepancy between Sellars’ and Brandom’s accounts— precisely in “Kantian” terms: Sellars, in his evident attraction to the rigors of scientism (apart from the vagaries of the “Scientific Image” piece) completely abandons the transcendental standing of persons and intentions (a fatal mistake if he is to succeed in naturalizing Kant—in salvaging normativity for instance); and Brandom, despite loosely conforming with Sellars’ general Kantian strategy, neglects to explain what, methodologically, he means by the categorial (“pure pragmatic”) normative necessities alleged to be discerned in the inferential structure of the “framework” of practical life. It’s not easy to sort out the differences between Sellars’ and Brandom’s uses of the same bits of philosophical vocabulary (due, of course, to Brandom’s acknowledging Sellars’ inventiveness). But, by and large, I’d say Brandom was, first of all, an inferentialist and an analytic rationalist who speaks like a pragmatist; and only then as a Sellarsian in part; and then, very loosely indeed as a “Kantian” or a “Fregean” or a “Wittgensteinian.” Whereas, Sellars is, first of all, a committed realist of a pragmatic and naturalizing bent, and then, as opportunistically as possible, a Kantian drawn to accommodating the persistently mongrel and inherently informal fixities of philosophical analysis; and thereupon drawn to acknowledge the sheer intractability of puzzles like that of “material inference” without actually invoking an all-­purpose inferentialism. These differences are temperamental of course, but also profoundly philosophical. Brandom believes he glimpses (relatively autonomous, rational) invariances in the flux of life and Sellars is convinced that inferential fixities are pragmatically useful idealizations of fluxive potentialities. Both, then, are “pragmatists” of a sort. The point of the harangue is that Sellar’s puzzle remains critical for the well-­ being of philosophy itself: how to “reconcile”—not merely “join” or “fuse”—what are said to be the essential or defining attributes of the reflective creatures who must answer (human beings, persons, ourselves), “admitted to exist,” according to the language of the manifest image, now conceptually trapped in an impossible limbo, unable, ontologically, to be recognized within the idiom of the scientific image—where their attributes are required. But what are these creatures to which “their” attributes (however defined) should be “assigned,” if they can no longer be assigned “descriptively,” to persons? (Intentions, for instance.) As far as I can see, Sellars does not deign to answer, unless he means, by what he says about persons, to have exited (for a clarificatory moment) from the agon involving the two images, in order to remind us of what we are unwilling to jettison, within the space of our home language. He must mean—at least for the sake of the essay’s thought-­experiment—that everything

Sellars’ “Philosophy and the Scientific Image of Man”

137

regarding the human animal can be descriptively and explanatorily salvaged within the terms of the scientific image, “except” what is essential to our functioning as persons; and that whatever else we may wish to preserve from the discarded world of the manifest image—“intentionality,” chiefly—can be successfully “joined” or “fused,” non-descriptively, that is, externally, rhetorically, fictively, with elements drawn from the scientific image, precisely because it is no longer entitled to autonomous ontological standing. (I don’t see how we can have it both ways.) We cannot hold both that intentionality must be secured (if secured at all) from the resources of the manifest image (thereupon “joined to” but not “reconciled with” the concepts of the scientific image) and that the scientific image itself can always displace important manifest-­image attributions, without significant conceptual disadvantage. In any case, Sellars does not venture any compelling reasons for his confidence. The idea cannot be coherently formulated: the attributional act would have to be intentional, but the attributes attributed would no longer have any proper meaning. If we succeeded, we would have already “reconciled” the two images: there would have been no need for “fusion”—which, of course, would be tantamount to adopting a fiction. In fact, I don’t believe the matter is as difficult as Sellars supposes—and I see no need for “going” Kantian here (unless, as in Peirce’s verdict, Kant is himself deemed to be no more than “a confused pragmatist”).13 But that’s an issue for another occasion. The formula under which Sellars advances his own lessons (not his solution) is more interesting than the mere advocacy of inferentialism, because it provides a substantive link between realism and normativity—mediated by the definition of what it is to be a person. (We’ve now seen that Sellars unaccountably scants his Kantian loyalties, in abandoning the transcendental standing of the personal Ich.) I take Sellars’ essay to afford the sparest possible glimpse of the stubborn, extremely complex distinction of the human subject and agent (and its world), which utterly baffles the would-­be champions of the scientific image (in effect, Sellars himself, in one of his guises). How could Brandom have failed to notice that and what is Brandom’s own replacing argument? In any event, here is what Sellars notoriously affirms: A person can almost be defined as a being that has intentions. Thus the conceptual framework of persons is not something that needs to be reconciled with the scientific image, but rather something to be joined to it. Thus, to complete the scientific image we need to enrich it not with more ways of saying what is the case, but with the language of community and individual intentions, so that by construing the actions we intend to do and the circumstances in which

138

Wilfrid Sellars, Idealism, and Realism

we intend to do them in scientific terms, we directly relate the world as conceived by scientific theory to our purposes, and make it our world and no longer an alien appendage to the world in which we do our living. We can, of course, as matters now stand, realize this direct incorporation of the scientific image into our way of life only in imagination. But to do so is, if only in imagination [my emphasis: that is, without any compelling argument], to transcend the dualism of the manifest and scientific images of man-­of-the-­world. PSIM 40

Surely, Sellars offers no more than a petitio here. What he confirms—theatrically or unwittingly—is the ubiquitous and irreplaceable executive importance of the functional role of the human person. (I find it impossible to believe that Sellars could have been unaware of that. Hence, I construe the entire essay as a piece of philosophical irony at its most arch.) As I read this, it is at least as plausible to think that the manifest image has absorbed the scientific image as it is to reverse the dependency; in fact, I think Sellars makes a better case for the hegemony of the manifest image (or a successor that could absorb the defeasible claims of the scientific image), though it appears that both fail along different lines. (Sellars is on both sides of the fence, at the same time.) I don’t see how the remark—“by construing the actions we intend to do and the circumstances in which we intend to do them in scientific terms, we directly relate the world as conceived by scientific theory [my emphasis] to our purposes”—could possibly support the primacy of the scientific image or acknowledge the unreality of the world of the manifest image; and I see no evidence that Sellars (or anyone else) has successfully demonstrated that (or what it means to say that) everything but the intentionality of persons can be successfully accounted for by the scientific image (which, then, entails the “placement” of norms)—perhaps within the terms of the notion of the “space of reasons” (now rendered inoperative by the stalemate, impossible to take seriously)—which directly affects the “placement of language, norms, and the ‘space of reasons’.”14 In a word, if “intentions” are allowed to “dangle” conceptually, without systematic linkage or replacement within the vision of the scientific image, then the same will be true respecting actions, norms, thoughts, language, and persons. Which is to say: scientism must fail, hands down. Sellars does not address the matter further in the essay, and I find no evidence that he’s solved the puzzle elsewhere. I take him to have demonstrated, obliquely, how difficult (better: impossible) the task is bound to be, confined within the terms of the scientific image. (As I say, the “solution” is no more than a theatrical gesture.) I believe we cannot make any philosophical progress here without a

Sellars’ “Philosophy and the Scientific Image of Man”

139

robust account of the nature of the human person, natural language, intentionality, and human agency; and none of these notions, I claim (but shall not argue here), which belong to the manifest image (within the terms of Sellars’ account), may be satisfactorily transferred to the space of the scientific image. I don’t think it can be solved in the form Sellars presents; but then I don’t see that we need follow either Sellars or Brandom in agreeing to the two-­image thesis. I myself have no doubt that a reasonable resolution must be hybrid but not dual—and that Sellars has come within a hair of the main premise of the solution. On my view, the following doctrines, preparatory to a full-­dress realist solution, are well-­nigh indisputable: (i) that, canonically, the mastery of language is exclusively assignable to persons; (ii) that the formation of persons (the transformation of human primates—usually, human infants) is the same process as that of mastering language; (iii) that, therefore both language and persons are cultural artifacts (biologically and culturally formed hybrids)—respectively, paleoanthropologically achieved transforms of primate forms of communication and of the actual primate members of the human species; and (iv) that normativity is itself a conceptual distinction and resource accessible only to enlanguaged persons (that is, discursively). I must add here, since it bears directly on the “objective” standing of norms, as well as on the continuity of animal and human intelligence (including inference); (v) that there must be a continuum between perceptual and experiential “concepts” and “judgment” and what, paradigmatically, are thought to be concepts and judgments in the specifically linguistic sense, if we are to explain animal evolution and survival and the human infant’s prelinguistic ability to learn human languages quickly and efficiently. I read items (i)–(v) as the core of a post-Darwinian improvement of classic pragmatism. Item (v) is particularly important, strategically, since all of the casually acknowledged members of the Pittsburgh School—to include at least Sellars, Brandom, McDowell, and Rorty—either oppose outright the functional aptitude of perceptual and other nonlinguistic concepts (hence, animal knowledge, learning, inference, and allied skills, without mastering normativity), or they tend to stonewall on the matter. (Animal cognition challenges the ubiquitous application of the linguistic turn.) Sellars shows more sympathy for the barest possibility than the others, but not as much as one might wish (or need); and, quite unexpectedly (perhaps unnecessarily), McDowell offers a remarkably unyielding Kantian-­like transcendental insistence to the effect that perceptual judgment must be discursive “all the way down.” Curiously, McDowell gives no argument for his thesis (except dialectically, against Sellars’ claims)—which must

140

Wilfrid Sellars, Idealism, and Realism

be empirically false, if indeed infants master their home languages by exercising prelinguistic cognitive skills. In any case, McDowell stands even to the right of Kant in his apriorist certitude.15 Frankly, I take the infant’s feat to threaten (or defeat) the Pittsburghers’s opposition to perceptual concepts, on whatever grounds they favor. It’s quite possible that McDowell fears that to yield on animal intelligence may be to subvert the necessity of norms: both the categorial necessity of normativity (the ineliminably accessible “space of reasons” that McDowell shares with Sellars), as well as the discernible necessity of contingently operative norms (whether causal, logical, instrumental, or analogously invoked)—which I call “agentive” and “enabling” norms, that is, norms governing the right or obligatory objectives of human persons qua persons and norms governing the rational instrumentalities by which contingent objectives are effectively (intelligently, reasonably) realized by persons. But if McDowell fears such challenges, then (I should say) his cause is lost, inasmuch as, given the entire tally comprising (i)–(v), normativity cannot be anything but an artifact of human culture as is also discursivity and discursive judgment. (These are concessions that profoundly affect the would-­be objective standing of agentive norms, whether with Aristotle or Hume or Kant or Hegel.) In any case, solely on empirical grounds, the human infant’s feat all but defeats the denial of animal “rationality” and nonlinguistic “concepts.” What Sellars “demonstrates” is that the disjunctive option (the two-­images model) leads to stalemate and paradox. The literal narrative does not acknowledge that there cannot be a disjunctive choice where it insists on one; the would-­be disjunction is itself no more than a faulty heuristic fiction generated within the terms of a holist picture of the human person. I’m guessing that that is what Sellars was obliged (or possibly had decided) to hide in plain view. I agree that, for the sake of the argument (and possibly in accord with Sellars’ own conviction), some scientistic solution is bound to dominate our conjectures—but not because the “scientific image” has actually defeated the “manifest image.” In effect, the two “images” are the main counters in a Weberian-­like fiction that has fruitlessly dominated Western philosophy down to our own day. My point is that if the agon was indeed a Weberian ploy, then Sellars must have intended its inaptness to be noticed. (He drops enough clues: well, the deliberately lame ending of the essay itself.) There’s the ground for making more of the “Scientific Image” essay than, say, Brandom does. Sellars, I find, is given to fragmentary thought-­experiments. He has satisfied himself that the two-­images solution can’t work. So he kills off the argument by

Sellars’ “Philosophy and the Scientific Image of Man”

141

conceding, impossibly, that all that’s needed is an external “fusion” of the two images. But that’s merely to admit the argument’s defeat. Surely, Sellars is aware of that. At the very least, he cannot claim both that the intentionality of actions can be managed within the idiom of the scientific image and that intentionality can only be “joined,” externally, to the scientific image. (I’ve cited the pertinent text from the essay’s closing pages.) Brandom has no intention of following Sellars here: hence his verdict that Sellars has gone over to the noumenalists. But that suggests, in turn, that Brandom’s treatment of the “Scientific Image” piece (in From Empiricism to Expressivism) may signify no more than that he has as yet no plausible reading of the running paper that suggests a fresh option, or the need for one. The principal source of the trouble rests with the fact that the “two-­images” thesis defeats itself. Speaking loosely, the presence of the enlanguaged person is required (at least provisionally, short of some successful elimination or displacement) among the standard concerns of the “scientific image” as much as among the concerns of the “manifest image.” The holist presence of persons must, in some sense, be common linguistic coin in both sorts of theory, since there cannot be an answer to Sellars’ question that is not effectively “mongrelized” (hence, unified in that way) to include whatever may be disjointly or incommensurably salient in either “image.” Hence, Sellars’ “Weberian” maneuver must have been flawed from the start, and may have been favored for that very reason, didactically and dialectically. We must begin with a single unified concept of the holist person ranging over every admittedly cognitive, intelligent, or rational concern, no matter how difficult it may prove to be to produce a coherent account of all that that provisionally comprises. A suitably “improved” scientific image (or successor) must already include a substantial contribution from the side of the manifest image, and vice versa. (I take that to be close to Sellars’ oblique instruction.) Our “image” must be inclusive from the start, holist (as I say), committed to premises at least as extensive and as reasonable as those tallied as items (i)–(v). Up to a point, this should pose no difficulties: we have only to invoke the vocabulary of any natural language. For, assuming universal bilingualism, the vocabulary of every distinct language is accessible, in principle, to the speakers of every other language. Sellars’ agon is already less comprehensive (conceptually), more restrictive and more troublesome than the mongrel possibilities of any natural language, because it pretends that the vocabularies of the manifest and scientific images are, in spite of being disjunctive, effectively comparable extensionally. But that’s

142

Wilfrid Sellars, Idealism, and Realism

not strictly true, since Sellars himself (please note) confirms that the scientific image lacks the vocabulary of “intentionality,” which (he also affirms) effectively captures the essential attribute of persons. But why should it matter if the scientific image lacks that particular vocabulary if persons have no substantive role in the scientific image itself? Sellars does not rightly say. The only pertinent reason is that the loss must be more important than Sellars concedes: he says we can simply add or fuse this part of the manifest image’s vocabulary to or with our scientific language (if we wish), since it doesn’t matter that it’s lost its descriptive role: we need it “only” to secure prescriptives of the normative sort, which are said to be nondescriptive (by both Sellars and Brandom). Read literally, Sellars’ solution is both preposterous and inapt: normative distinctions may not be descriptive, but their application, in living contexts, surely presupposes the use of descriptive vocabularies likely to affect, in consequential ways, the inferential linkages that we concede. (I suggest that complications of this sort can never be discounted. I find that they are not adequately addressed by Brandom or the neo-Fregeans.) Consider that, functionally, according to (my) premise (ii), a person is an apt speaker of a natural language: hence, champions of the idiom of the scientific image must be able to save the language-­speaking ability of persons, if they are to defend their own image successfully; and, if we agree to the holism of persons, these same champions are bound to lose the argument whether they save the language-­using ability of persons or not. For if they save it, they will have demonstrated that the two images cannot be disjoint after all; and if they fail, they will have conceded (by admitting one impossibility or another) the superiority of the manifest image. They cannot afford to lose this part of the argument. In conceding the holist thesis, intentionality and normativity and discursivity and autonomous agency (and more)—important Kantian themes—cannot be secured piecemeal: which is to say, normal persons must be sufficiently robust and functionally integrated so that, in exercising any of their distinctive abilities, they are able to exercise all or most of them. But, of course, in that case, scientific-­ image “persons” will be indistinguishable from the persons of the manifest image. I should add that I’ve deliberately avoided mentioning transcendental arguments, though a “Kantian” of Sellars’ stripe cannot possibly fail to discern the deep significance of the dangling, completely arbitrary resolution of his own puzzle. He’s breached Kant’s categorial constraints. After all, Kant requires the transcendental Ich in his theory of apperception; but Sellars puts the would-­be naturalistic self or person at mortal risk within the boundaries of the scientism he hopes will finally prevail!

Sellars’ “Philosophy and the Scientific Image of Man”

143

You see the difficulty. There’s no escape. But now, there’s a further question that begins to surface. For what is it to think and speak in normative terms? In my opinion, we invoke normativity in (at least) two very different ways. I’ve already suggested that norms are either agentive or enabling—though, whether they are of the grading or ranking sort, they must remain inescapably discursive. In that sense, valuing and preferring need not be normatively qualified (which allows attribution to honey bees and monkeys, though, doubtless, there will be incipiencies that will strain the determinate disjunctions usually favored). The conceptual importance of the normativity distinction (agentive or enabling norms) cannot be denied. Because a fair case can be made to the effect that enabling (but not agentive) norms can, standardly, be paraphrased in descriptive and non-­normative terms. For example, we readily allow the paraphrase of medical prescriptives regarding improved health, in terms of the describable results of determinate causal interventions of one kind or another. I see no reason why something similar should not hold of all instrumental processes relative to feasible objectives—causal, logical, inferential, interpretive, or otherwise cast in terms of means and ends. But agentive norms—especially the putatively “noblest” ones (moral or religious, for instance)—are usually regarded as terminal, categorical (rather than merely contingent or conditional), approaching ultimate concern, interpreted as integrative or inclusive, in such a way as not to be able to be directly aimed at in the enabling sense. The upshot is—though I shall not argue the thesis here—that, inasmuch as normativity is discursive, it is also artifactual. Hence, agentive norms are never more (cannot be shown to be more) than sittlich, that is, always already entrenched in the prevailing practices of a viable society of persons, within which new cohorts of infant primates successfully master their home language (and, thereby, the culture that that subtends); or, as apt agents, they propose or project would-­be normative improvements, as deliberate modifications of their current sittlich norms, which, in turn, may acquire a comparable sort of sittlich standing. There are then “anthropological facts” that, descriptively and in causal or historical ways, account for the sittlich presence of agentive norms, though such facts cannot be said to validate such norms, qua agentive norms, in any sense that exceeds the logically trivial but socially momentous importance of being sittlich. It is, then, one of the great ironies of philosophy that nearly all informed discussants of the normativity issue are captured by the seeming need to discover the true ground of objective morality, politics, Bildung, civilizational goals and

144

Wilfrid Sellars, Idealism, and Realism

the like (Hans Kelsen, for instance, in a famous dispute with Max Weber). To my mind, it is quite enough, conceptually, that human societies are actually able to generate modifications of their sittlich practices that they themselves are willing to treat as normatively endorsed in the sittlich way. Admittedly, the idea is a dangerous, even unruly one. But this is as much as one can elicit from Sellars’ “Scientific Image” paper—as well as from Brandom’s spotty discussion of it. My guess is that what more can be said about the linkage between “commonsense” and scientific realism and normativity requires, at the very least, a review of the interrelationship between the “Scientific Image” piece and both Sellars’ “Empiricism and the Philosophy of Mind” and his Science and Metaphysics book (1967)—and, accordingly, what John McDowell (as well as Brandom) has to say about the use of these further materials. But that requires a fresh beginning and a glimpse of the gathering unity of Sellars’ philosophical vision. Nevertheless, I venture to say—provocatively and much too briefly—that if the argument presented thus far is at all valid, then Sellars’ affirmation, in the “Empiricism” paper, of the doctrine of “psychological nominalism” (the linguistic standing of the self or person) will prove untenable and the standing of Sellars’ famous formula, “the space of reasons,” will betray its being seriously compromised: because language itself (and “linguistic nominalism” and the Pittsburghers’s insistence on the linguistic nature of concepts) has nowhere been accounted for in terms freed from manifest-­image terms. There’s the essential pons of any version of the “scientific image”—a fortiori, of the “space of reasons.” Effectively, the “Empiricism” paper cannot make sense without a scientistically favorable resolution of the “Scientific Image” piece. Accordingly, the lame conclusion of the “Scientific Image” paper puts the “Empiricism” piece at risk as well. Their arguments are finally inseparable, though I cannot afford to limn the full argument here. These conclusions confirm that the “Empiricism” piece cannot be treated as free-­standing. The rest of the story requires another sizable analysis.

Notes 1 The argument of the “Scientific Image” paper is effectively confined to the completely abstracted notion of the two images. No history or context or well-­known champions are mentioned. But several associated papers, perhaps most pointedly, “Phenomenalism,” which appears in the same volume as the “Scientific Image” piece, pp. 60–105, begin, though very sparely, to identify familiar contemporary philosophers who have favored variants of the two images doctrine or doctrines closely related to them. See, also, Robert B. Brandom, From Empiricism to

Sellars’ “Philosophy and the Scientific Image of Man”

2

3

4

5

6 7 8

145

Expressivism: Brandom Reads Sellars (Cambridge MA: Harvard University Press, 2015), passim, which fills in a great many details but does not endorse this part of Sellars’ philosophical program. Brandom seems to think Sellars is attempting to revive a form of scientific realism that would make noumena accessible to analysis; whereas it’s my contention that Sellars means to instruct us about the fatuousness of any such maneuver which, admittedly, he’s been drawn to, as are most of the leading philosophers of the day. We now have the benefit of a new publication, Brandom’s From Empiricism to Expressivism, essentially devoted (I would say) to a careful reading of Sellars’ “Empiricism and the Philosophy of Mind,” scanned against the weight of the intriguing fact that Brandom does not discuss the “Scientific Image” paper frontally, by name, though it is clearly implicated on every page and pointedly mentioned in the book’s Introduction (pp. 23–4), where Brandom confesses it took him a longish while to grasp Sellars’ question correctly. Chapter 1, however, is, essentially, a reading of the “Scientific Image” paper: it treats Sellars’ scientism (scientia mensura) as a naturalistic rendering of Kant’s noumenalism, which I take to be an interpretive mistake, however plausible. Sellars’ effort is a Kantian-­inspired attempt in favor of scientific realism, but it couldn’t possibly be a form of noumenalism. Realism regarding the empirical (so-­called “independent”) world is, if Kantian, not the noumenal world. In any case, on my reading, Sellars “demonstrates” his dissatisfaction with the “two images” conception. See Wilfrid Sellars, “Language, Rules, and Behavior” (1949), Pure Pragmatics and Possible Worlds: The Early Essays of Wilfrid Sellars, ed. Jeffrey F. Sicha (Atascadero: Ridgeview Publishing Company (1980, 2005), 117–34, especially 123, for the formulation “The mode of existence of a rule is as a generalization written in flesh and blood or nerve and sinew, rather than in pen and ink.” See, for example, the neglected paper by Richard Rorty, “Twenty-­five Years After,” added to the second edition of Rorty’s anthology The Linguistic Turn: Recent Essays in Philosophical Method (Chicago: University of Chicago Press, 1967, 1992), 371–4. I come back very shortly to the longer passage from which I’ve just excerpted the line cited from “Realism and the New Way of Words,” just above—and provide, there, the bibliographical reference needed. See, also, Max Weber, “ ‘Objectivity’ in Social Science and Social Policy,” in Edward A. Shils and Henry A. Finch (trans. and eds.), Methodology of the Social Sciences (Glencoe: Free Press, 1949), 49–112. See Brandom, From Empiricism to Expressivism, 163–6 especially, but, also, the whole of ch. 4. See, further, Brandom, From Empiricism to Expressivism, 93–8. The expressions (and options), “subject” and “object naturalism” are introduced by Huw Price, who may be read as a cousin to the Pittsburgh School. See his Naturalism without Mirrors (Oxford: Oxford University Press, 2011); and Expressivism,

146

Wilfrid Sellars, Idealism, and Realism

Pragmatism, and Representationalism (Cambridge: Cambridge University Press, 2013). As Brandom explains: “What Price calls ‘subject naturalism’ . . . is a pragmatic naturalism, rather than a representational semantic naturalism. The subject naturalist makes no assumptions about whether the target vocabulary admits of a properly representational semantics,” From Empiricism to Expressivism, pp. 91, 95. As far as I can see, Sellars sets no liens of any kind on the selection of viable descriptive vocabularies. His aim, rather, is to make provision for the specifically “nondescriptive” function of (pragmatic) metalinguistic inference—leading to collecting “Kantian [or alternative] categories” of the basic “framework” (the epistemology) within which ordinary or specialized truth-­bearing discourse obtains. Such use could, conceivably (Brandom concurs), employ the same “target vocabulary” that, in other circumstances services first-­order empirical description. (I regard all of this conceptual architecture as entirely heuristic: the main concern is the profound informality of the would-­be inferentialism both Sellars and Brandom have in mind.) We’re talking about a way of modeling the inferential complexity of whatever we “say” or “do”: every determinate claim in conformity with the model is so profoundly defeasible that we cannot rightly claim to rely on the model alone, “methodologically.” 9 For a scrupulous account of Frege’s model and its possible applicability in the “exact sciences” (or physics), see Danielle Macbeth, Realizing Reason: A Narrative of Truth and Knowing (Oxford: Oxford University Press, 2014), chs. 7–9. I see the tempting analogy, but admit (perhaps benightedly) that I cannot see the force of the full supporting argument required. I agree that physics is mathematized to a remarkable degree, but I find myself persuaded that, at bottom, it’s still an empirical science: the analogy with mathematical thinking may prove to be insuperably heuristic. 10 Compare Robert B. Brandom, Articulating Reasons: An Introduction to Inferentialism (Cambridge MA: Harvard University Press, 2000), 157–8, 205n7. 11 Wilfrid Sellars, “Realism and the New Way of Words,” Pure Pragmatics and Possible Worlds: The Early Essays of Wilfrid Sellars, ed. Jeffrey F. Sicha (Atascadero: Ridgeview Publishing, 1980, 2005), 61; cited by Brandom, From Empiricism to Expressibism, 89. 12 See Brandom, From Empiricism to Expressivism, 87–96. 13 I offer a sketch of a solution, complete with an account of agency and normativity, in a forthcoming book, Toward a Metaphysics of Culture (London: Pickering & Chatto, publication pending). 14 See Wilfrid Sellars, “Empiricism and the Philosophy of Mind,” Science, Perception and Reality (London: Routledge & Kegan Paul, 1963), 127–96; also, John McDowell, Mind and World (Cambridge MA: Harvard University Press, 1994, 1996), Lecture IV. 15 See John McDowell, Having the World in View: Essays on Kant, Hegel, and Sellars (Cambridge MA: Harvard University Press, 2009), chs. 1–3 (The Woodbridge Lectures), especially “Sellars on Perceptual Experience” (ch. 1), 3–22; and “The Logical Form of an Intuition,” (ch. 2), 23–43.

Sellars’ “Philosophy and the Scientific Image of Man”

147

Bibliography Brandom, Robert B. (2000), Articulating Reasons: An Introduction to Inferentialism, Cambridge MA: Harvard University Press. —— (2015), From Empiricism to Expressivism: Brandom Reads Sellars, Cambridge MA: Harvard University Press. Macbeth, Danielle (2014), Realizing Reason: A Narrative of Truth and Knowledge, Oxford: Oxford University Press. Margolis, Joseph (2016), Toward a Metaphysics of Culture, London: Pickering & Chatto. McDowell, John (1996), Mind and World, Cambridge MA: Harvard University Press. —— (2009), Having the World in View: Essays on Kant, Hegel, and Sellars (The Woodbridge Lectures), chs. 1–3, Cambridge MA: Harvard University Press. Price, Huw (2011), Naturalism without Mirrors, Oxford: Oxford University Press. —— (2013), Expressivism, Pragmatism, and Representationalism, Cambridge MA: Harvard University Press. Rorty, Richard (1992), “Twenty-­five Years After,” in The Linguistic Turn: Recent Essays in Philosophical Method, 2nd ed., Chicago: University of Chicago Press, 371–4. Sellars, Wilfrid (1963a), “Empiricism and the Philosophy of Mind” (1956), in Science, Perception, and Reality, London: Routledge & Kegan Paul, 127–96. —— (1963b), “Phenomenalism” (1959), in Science, Perception, and Reality, London: Routledge & Kegan Paul, 60–105.   PSIM —— (1963c), “Philosophy and the Scientific Image of Man” (1962), in   Science, Perception, and Reality, London: Routledge & Kegan Paul, 1–40. —— (2005a), “Language, Rules and Behavior,” in Jeffrey F. Sicha (ed.), Pure Pragmatics and Possible Worlds: The Early Essays of Wilfrid Sellars, Atascadero: Ridgeview Publishing Company, 117–34. —— (2005b), “Realism and the New Way of Words,” in Jeffrey F. Sicha (ed.), Pure Pragmatics and Possible Worlds: The Early Essays of Wilfrid Sellars, Atascadero: Ridgeview Publishing Company, 46–78. Weber, Max (1949), “ ‘Objectivity’ in Social and Social Policy,” in Edward A. Shils and Henry A. Finch (eds. and trans.), Methodology of the Social Sciences, Glencoe: Free Press, 49–112.

7

A Kantian Critique of Sellars’ Transcendental Realism Johannes Haag, Universität Potsdam (Germany)

Wilfrid Sellars developed his own philosophy in constant exchange with Kantian thought. In what follows I would like to show how some of the theses central to Sellars’ own systematic approach are most appropriately understood by acknowledging their methodological and argumentative foundation in Transcendental Philosophy. In particular, I will be concerned with the two most famous criticisms to which Sellars subjects Kant’s theory: his critique of Kantian arguments for the phenomenality or transcendental ideality of empirical reality and his rejection of Kantian agnosticism concerning the thing in-­itself. Both criticisms are closely connected to Sellars’ own doctrine of Scientific Realism. In what follows I will expound this doctrine as a piece of Transcendental Philosophy and criticize it on the grounds that it undermines the very transcendental presuppositions on which, according to my interpretation, it seeks to base itself.

Objects of experience In many of his writings, Sellars makes ample use not only of Kantian terminology but also of Kantian insights and methodology. And, as we shall see shortly, even where he criticizes Kant on these matters and replaces Kantian doctrines with his own, he is often working within the framework Kant first introduced for the philosophical discussion of these problems. One Kantian insight that found Sellars’ wholehearted support was the Kantian analysis of the function of the concept object of experience and the connected question of the phenomenality of empirical reality. This analysis is the subject of

150

Wilfrid Sellars, Idealism, and Realism

his late paper on “Kant’s Transcendental Idealism” (1976) and its results will prove of particular importance for my purpose. At the very end of this paper he writes by way of conclusion: Kant saw that the concept of an object of perception contains a reference to the perceptual takings which are the criteria for its actuality. He also saw that the concept of a perceptual taking as the taking of an object contains a reference to material things and events which, if actual, would imply its own actuality. The actuality of perceptual takings and the actuality of material things and processes are not logically independent. KTI §53

In this remark Sellars claims a mutual dependence of the concept of an object of perception or experience and the (possible) experiences or perceptual takings of this same object. In KTI he investigates both directions of the dependence by means of an exegesis of Kant’s concept of an object of experience or, more broadly, an empirical reality that exists as actual but not in-­itself.1 This distinction is also important for what follows in that it grounds the viability of the phenomenalism both Sellars and Kant envisage, i.e. a phenomenalism concerning empirical reality that is compatible with its claim to objectivity. What is the concept of actuality at work here? The concept of an actually existing object is, it turns out, the concept of an object of perception or experience that stays the same through actual or possible changes of perspective throughout a sequence of the minimal and basic acts of perception Sellars calls perceptual takings. Taking up an example that Kant uses at the beginning of the second Analogy2 of the Critique of Pure Reason (CPR) the content of the perceptual takings of any given house would be of the general form “house-­from such-­andsuch-­a-point-­of-view” (KTI §48), for instance “That house facing me edgewise.” The concept of an object of experience would then be the concept of what gives unity to a given string of successive perceptual takings of an object. Not, however, as something existing per se, but as that part of the content “which [the perceptual takings] share” (KTI §48), i.e. the content house that is common to each of the successive perceptual takings.3 The objectual content, of course, is not the only content these representations share, as Sellars is quick to point out. The perceiving subject is the other constant common factor in this varying flow of perceptual takings of an object. And it is in this sense that he can claim that “the core of the knowable self is the self as perceiver of material things and events” (KTI §52). Thus conceived, the concept of a perceiving, experiencing, and knowing self already incorporates a reference

A Kantian Critique of Sellars’ Transcendental Realism

151

to the actual existence of the objects thus perceived, experienced, and known— and vice versa. In short, they are mutually dependent. Let us treat this rough sketch of Sellars’ thought as adequate for the moment. It is important to note that Sellars does not claim that this line of reasoning demonstrates anything about the specific structure that the world or the self must have in order to generate experience of the required kind. In this context, he restricts his concerns to showing that the world (and, indeed, the self as part of that world) must objectively or actualiter exist in order to make experience possible—by and large regardless of the specific structure of world and self, required for their so existing.4

The “bifurcation” of nature rejected It is, I would like to argue, the acceptance of Kant’s doctrines of the mutual dependence of objective perception and object perceived that ultimately forces Sellars to accept some form of scientific realism—at least as a regulative ideal. For it is this mutual dependence between the possibility that the transcendental subject engage in conceptual takings and the conceptual necessity (Denknotwendigkeit) of an objective correlate, so central to Kant’s theory, that compels Sellars to frame a number of successor concepts that ultimately have to do with a science ideally conceived. This particular kind of conceptual necessity is characteristic of any philosophical thesis and corresponding line of argument that deserves the name transcendental in the original Kantian sense: a thesis is transcendental if it concerns the conditions of the possibility of the reference of any cognitive representations to its purported object.5 What makes this introduction of successor concepts necessary is Sellars’ most important critique of Kant, i.e. his critique of what he calls Kant’s “bifurcation” of nature. This bifurcation of nature is Kant’s asymmetrical treatment of spatial and temporal qualities on the one hand and objective sensations (e.g. colors) on the other. Sellars thinks that this asymmetrical treatment was a serious mistake on Kant’s part: [R]eflection on the nature of empirical Space and spatial attributes (if he had not . . . taken the subjectivity of colour for granted) would surely have convinced Kant that the objects of perception are as essentially coloured as they are extended; indeed, that their spatial characteristics essentially involve the contrast of colour with colour. An empirical line, for example, is a white streak on a black background, or the edge of a ruler. Thus Kant should have recognized

152

Wilfrid Sellars, Idealism, and Realism

that colour itself . . . is as essential a feature of the objects of outer intuition as is shape. SM 586

This criticism serves as the foundation for other criticisms, i.e. the rejection of Kant’s argument for the phenomenality of the empirical reality and his agnosticism concerning the in-­itself. This critique of Kant’s bifurcation of nature hence proves important because it is this critique that ultimately leads to Sellars’ endorsement of scientific realism.7 Sellars himself hints at a connection between these issues at one point in his paper “Scientific Realism or Irenic Instrumentalism” (SRI 1966).8 But what he has to say there can also be somewhat misleading, in my view, since it suggests a line of argument from a different treatment of colors to the phenomenality of empirical reality that presupposes scientific realism.9 While I think that Sellars does have a tendency to suggest this line of argument, I will try to make plausible, admittedly somewhat speculatively in some places, that it should not be conceived as the most interesting or most important argument in this context precisely because it presupposes scientific realism. The idea that Sellars is in effect presupposing scientific realism seems, for instance, to underlie McDowell’s critique of Sellars’ theory, put forward first solely with respect to EPM (in J. McDowell Having the World in View. Essays in Kant, Hegel, and Sellars (Cambridge, MA: Harvard University Press 2009), 15–16), and later extended to Sellars’ later “transcendental” conception (cf. ibd., 125). In rough outline it goes like this: there is no place in the scientific image for the concepts of macroscopic objects that permeate the manifest image. But the corresponding successor concepts introduced by means of analogical concept-­ formation cannot be colored in the sense in which macroscopic objects are conceived of as colored.10 Therefore the manifest image is phenomenal.11 As I said, I hope to show that this is not the decisive reason for deeming the manifest image phenomenal, although I do not want to deny that these considerations are present in Sellars’ work. But to the extent that they are connected with what I take to be the decisive argument for the phenomenality claim, they can, from Sellars’ perspective, only constitute additional evidence in support of the latter argument’s conclusion. The argument I would like to ascribe to Sellars is a transcendental one that emerges from one of the points where the manifest image generates an aporia that ultimately results in its replacement as a viable framework in which to approach the world. The phenomenality of empirical reality, therefore, is not the result of a presupposed dominance of the

A Kantian Critique of Sellars’ Transcendental Realism

153

scientific image but already threatens the manifest image from within. The phenomenality claim arises at the level (and from the resources) of the manifest image itself. The theorist of the manifest image responds to the challenge of explaining sensibilia—i.e. the sensible aspects of objects of experience such as their colors or sounds—by changing his conception of sensibilia within the manifest image: namely, by introducing sense-­impressions qua modifications of the perceiving subject.12 This modification of the manifest image to accommodate novel explanatory claims is, in the case of sensibilia, one of the most persistent aspects of Sellars’ philosophy. It results in the sense-­impression inference that Sellars first introduced in sections III and IV of EPM—and never tired of repeating. The name “sense-­impression-inference” itself serves two purposes: it marks an important difference from the notorious sense-­datum-inference while at the same time indicating some continuity.13 This inference was based on and licensed by what he took to be a basic phenomenological fact which no successful theory of perception should ignore: the intrinsic indistinguishability in content between the experience of a veridical perception, of the corresponding illusion, and of the matching hallucination, respectively. This phenomenological fact, on Sellars’ view, licenses an inference to the existence of entities he calls sense-­impressions or sensations.14

The sense-­impression-inference Since it is so important, let us take a step back and examine the reasoning behind the sense-­impression-inference. This inference takes as its point of departure two claims which Sellars basically took for granted—although at least the second one is by no means uncontroversial. If we take his famous example of the perception of a pink ice-­cube, we can put the first of these claims as follows: (1) Something, somehow, a cube of pink in physical space, is present in the perception other than as merely believed in.15 This is the descriptive core16 of our perceptions that accounts for the actual existence of something in our perceptions. No talk of merely intentional existence can serve as a substitute for this actual existence: The one thing we can say, with phenomenological assurance, is that whatever its “true” categorial status, the expanse of red involved in a . . . seeing17 of the very

154

Wilfrid Sellars, Idealism, and Realism

redness of an apple has actual existence as contrasted with the intentional in-­ existence of that which is believed in as believed in. FMPP 20/1

This is how far phenomenology takes us in the “analysis of the sense in which we see of the pink ice cube its very pinkness” (SSOP 89), as Sellars puts it. This does not, however, imply that this descriptive core itself is literally a part of the perception conceived of as a conceptual episode. It leaves open the possibility that this descriptive core is (part of) what is perceived, i.e. the object of perception, where the term “object” is not meant to have ontological implications concerning the nature of what is thus perceived—the object of perception might just as easily be a mental or bodily state or process, so long as the latter is distinct from the act of perceiving in question and is perceived in some sense of this multifaceted term. Yet even in this characterization of the seeing of the very pinkness of the ice cube—one of its sensible as opposed to its causal properties, the latter of which we do not see in the sense under scrutiny18—Sellars is presupposing another claim that he again takes to be grounded in phenomenological fact. For, shortly afterwards, he denies that in seeing the very pinkness and cubic shape of the pink ice cube we see a property of the cube of pink in physical space: Nor is its character as a cube of pink in physical space facing me edgewise a matter of its actually being a cube of pink in physical space. SSOP 8819

The very pinkness and cube-­shape of the object we see, in other words, need not belong to (i.e. be a property of) the physical object that we take to be responsible for our perceiving the object as a pink cube. Why, it surely will be asked, cannot this “character” be a property of the physical object itself or, for the sake of integrating the “facing me edgewise” of Sellars’ example, a relational property between the physical object over there in physical space and me as perceiving subject (another object in physical space), as for instance the Theory of Appearing20 has it? Why, in other words, should this somehow-­presence of the cube of pink in perception be conceived of as a presence of the sensible properties of the pink ice cube in perception? Is not this introduction of a further (qualitative) element in perception premature? What comes out in this analysis of the qualitative aspects of perception is another piece of Sellarsian phenomenology. Throughout his philosophical writings, Sellars insists that the sense-­datum-inference has a phenomenological

A Kantian Critique of Sellars’ Transcendental Realism

155

foundation that—while seriously misinterpreted by many sense-­datum theorists—should be taken seriously and that stands in need of interpretation by a philosophical account of perception: Now, the basic phenomenological fact from which I shall take my point of departure is that when an object looks red to S, and S is, so to speak, “taken in”—I make this stipulation only to put irrelevancies to a side—S has an experience which is intrinsically like that of seeing an object to be red. FMPP 1621

It is important to keep in mind here that the experiences Sellars is talking about in this quote, i.e. perception and seeing, are conceptual episodes according to Sellars’ Psychological Nominalism. (Non-­conceptual episodes have yet to be introduced at this point of the argument!) Sellars, in other words, is impressed by the following observation: (2) There is an intrinsic likeness in qualitative content between a veridical perception of a pink ice cube and the corresponding case of an illusion of a pink ice cube, i.e. between a case of real seeing and merely ostensible seeing.22 I will not discuss this claim in depth, although it definitely is in need of discussion if one is to defend Sellars’ theory of perception. A proper defense, of course, would have to take into account Austin’s critique, in Sense and Sensibilia, of similar reasoning in the context of sense-­datum-theories. To my knowledge, Sellars never attempted such a defense. He simply took this point for granted. Part of an explanation could be that he believed the main obstacle to the preservation of this piece of phenomenology had been dissolved by his own critique of sense-­datum-theories and their sense-­datum-inference. The real mistake that sense-­datum theorists make in reasoning from the indistinguishability of conceptual content to a conceptually structured sensible item immediately given in perceptual experience (sense-­data), on Sellars’ view, is to interpret the sense-­data or sensations—whose existence is claimed as the conclusion of the corresponding inference—as conceptually structured immediate givens. Exposing this as a mistake is, after all, one of the central lessons of Psychological Nominalism. Since his own sense-­impressions or sensations are in no way conceptual, Sellars is convinced that it is once again available to him to exploit what he therefore took to be a quite harmless phenomenological fact about experience as the basis for an inference to the existence of sensory items in perception.

156

Wilfrid Sellars, Idealism, and Realism

In contrast, what is under attack by the opponents of indistinguishability claims is often the assumed identity of conceptual content. But Sellars does not assume such an identity of conceptual content without qualification. He merely claims that something non-­propositional or descriptive (“other than merely believed in”) is present in the conceptual perception that is in need of explanation (1). And he maintains that this very descriptive element, which can never constitute the whole conceptual content of an experience, can be present in cases where the resulting conceptual state is due to an illusion (or, for that matter hallucination) just as it is in the felicitous case (2). Maybe this should not be so very controversial after all.23 But since my purpose here is only to unfold the transcendental reasoning behind Sellars’ scientific realism, it suffices to merely hint at this stumbling block for any would-­be Sellarsian. We may safely ignore any difficulties that may loom in its defense for the time being. It is on the basis of those two phenomenological facts—for that is what Sellars takes these two claims to be—that Sellars grounds his own sense-­impression inference. This inference is, as he puts it in Science and Metaphysics, an inference to an explanation.24 What it is “designed to explain” (SM 17) is the occurrence of certain minimal conceptual episodes—even in cases where nothing “out there” exists that has the properties (sometimes only provisionally) ascribed to it by the conceptual episodes in question. In other words, it is designed to explain the occurrence of ostensible seeings or perceptual takings of the form (a) “This pink ice-­cube facing me edgewise” as they occur as parts of full-­fledged perceptual judgement like (b) “This pink ice cube facing me edgewise has a nice color.” Conceptual episodes like (a) are characterized as minimal, because they are not complete judgments themselves, but only essential parts of such judgments. They are what is initially “taken” by the judging subject and serve a purpose similar to Kantian intuitions. They are conceptual responses25 that, as it were, take up the perceived object into the conceptual or intentional order, thus making it accessible for the judgment of a perceiving subject. According to Sellars’ rejection of the Myth of the Given, in order to be able to take the perceived object up into the conceptual order they have to be themselves conceptually structured. The second phenomenological fact—i.e. the intrinsic indistinguishability in the content of a veridical perception, of the corresponding illusion, and of the matching hallucination – is based on the occurrence of minimal conceptual

A Kantian Critique of Sellars’ Transcendental Realism

157

episodes of this kind with a common descriptive core. This common descriptive core is, as Sellars makes clear in Science and Metaphysics, what accounts in large part for the identity of the minimal conceptual episodes—independently of whether we are taken in or know better. Even when we know better the minimal conceptual episode Sellars has in mind will be the very same—although, it has to be emphasized, even these minimal conceptual episodes essentially consist in more than just this “descriptive core.” As Sellars puts it in his Locke Lectures: It is . . . essential to note that the correlation of the correct conceptual response with objects perceived in normal circumstances by normal perceivers is as much in need of explanation as the correlations of conceptual responses with abnormal perceptual situations. . . . For even in normal cases there is the genuine question, “Why does the perceiver conceptually represent a red (blue, etc.) rectangular (circular, etc.) object in the presence of an object having these qualities?” SM 18

Sellars is convinced that an adequate explanation of these facts has to transcend phenomenology26 and introduce new theoretical explanations. Phenomenology can describe the descriptive core in no other way than is done by our claim (1), i.e. as: something, somehow a cube of pink in physical space, is present in the perception other than as merely believed in. Yet this amounts only to a mere empty claim like “this is responsible!”—which in itself does not deserve to be called an explanation at all. Mere ostensive reference to something as a descriptive core does nothing to explain the occurrence of the minimal conceptual episodes in question. It serves to pin down a difficulty without solving it. Phenomenology thus generates explanatory pressure: what is it, we are inevitably led to ask, that accounts for the somehow presence of the descriptive core in cases of veridical perception and ostensible seeing alike? What are the colors and shapes we ordinarily take to be properties of the objects “out there”? As Sellars puts it: “What is at stake is their [i.e. the colors’] status and function in the scheme of things” (FMPP 73). The theory of sense-­impressions or sensations is explicitly designed to supply us with this explanation. For these sense-­impressions or sensations are a class of non-­conceptual entities introduced exclusively to explain the occurrence of ostensible seeings in the abnormal case just as much as in the normal case, where proper seeing takes place. To this end we re-­categorize27 what up to that point we took to be sensible properties of empirical objects in physical space as modifications of the perceiving subject that have the very same logical manifold of

158

Wilfrid Sellars, Idealism, and Realism

these properties as properties of empirical objects.28 It is, as Sellars often puts it, the very same properties transposed into another (categorial) key. The ontological status of these entities is that of episodes or states of the perceiving subject or person. It is important to note that this move subsumes sensations—as far as their basic ontological category is concerned—under the same category as thoughts or, more generally, conceptual episodes: states of thinking subjects or persons.29

An explanatory gap The result of this development within the manifest image is, as we will shortly see, dramatic. It leads to an explanatory gap that has to be overcome to guarantee the “survival” of our concept of an object of experience—a concept we cannot do without. To close this gap, we have to substantially alter our “world-­story.” It turns out that, while the reason for this alternation is one that occurs within the manifest image, its consequences force us to transcend it. How exactly does this gap come about? Sense-­impressions contain the sensibilia, albeit in different categorical form, that we (mistakenly) ascribe to objects of experience. As such they are properties of modifications of perceiving subjects. Such properties cannot serve as properties of objects in the empirical world: “[If] the cube of pink of which we are perceptually aware is a state of ourselves as perceivers, then neither it nor anything resembling it could be an object in physical space” (FFMP 76).30 Sellars is here not only excluding the possibility that we could infer the common-­sense properties of objects of experience from the properties of our sense-­impressions,31 but is also rejecting the apparent possibility that there could be completely different explanations for what is present to us having the properties in question in the normal and the abnormal case. If sense-­impressions are responsible for the intrinsic qualitative identity of the normal and the abnormal case, then they are so only by constituting this qualitative dimension of experience in both cases.32 The consequences for the picture of our perception of objects of experience are, as I said, quite dramatic: If one thing is clear, it is that in perception we do not take what in point of fact . . . are [sense-­impressions]33 to be such. But might we not take them to be, for example, the red and rectangular facing surface of a physical object . . . in the corner? Might we not, so to speak, mis-­categorize them as items in the physical environment? Of course, such a taking would be a mis-­taking. But, after all, we

A Kantian Critique of Sellars’ Transcendental Realism

159

were given our perceptual abilities not for the purpose of ontological insight, but to enable us to find our way around in a hostile environment—just as we were given pain to get our hands quickly off the stove. SSOP 10934

The fundamental re-­categorization that takes place in the aftermath of the sense-­ impression inference (which is, in turn, a reaction to the fundamental phenomenological fact I referred to above) constitutes Sellars’ decisive reason for the phenomenality or ideality of the empirical world as far as objects of experience are concerned.35 For after the sense-­impression inference we conceive of sensibilia only as properties of states of the empirical subject, no longer as properties of the empirical objects. The objects of experience, therefore, which we (mistakenly) took to be essentially characterized by these very properties prior to this step, turn out to be merely phenomenal objects. This means that the coherence of the concept of an object of experience, i.e. of an object that exists as actual, but not in itself (as Sellars puts it in KTI), is endangered. The concept of an object of experience was introduced above36 as the concept of what gives unity to a given string of successive perceptual takings of an object. This unity is given by thinking of these perceptions as brought about by something that exists as located in time and space, and hence as different from us (one of the main ideas of the Transcendental Deduction that Sellars whole-­heartedly embraces in KTI). But this concept of an object in space and time is now lost to us through the sense-­impression inference. We lose the concept of spatiotemporal objects because Sellars reintegrates Kant’s bifurcated world and accordingly places colors and similar “objective sensations” (Critique of the Power of Judgment, 5: 206) on the same level as properties like shape and spatiotemporal location: both secondary and primary qualities hence turn out to be only mistaken for being properties of objects of experience. In a world-­view that rejects bifurcation of nature, when the sense-­impression inference does away with the former (i.e. secondary qualities like colors), the latter (i.e. the primary qualities) are also lost. The concept of an object of experience as essentially spatiotemporal is consequently no longer applicable. Since this very concept (together with the concept of a person) lies at the very heart of the manifest image, this image itself is dissolved by an aporia it itself generated.37 It is, on this interpretation, Kant’s “bifurcation of nature” that, somewhat ironically, saved him from the consequence of transferring all sensibilia into the empirical subject. (At this point, by the way, the significance of the sense-­ impression inference being a development within the manifest image becomes obvious.)

160

Wilfrid Sellars, Idealism, and Realism

Sellars’ (Cartesian38) theoretician, on the other hand, having introduced the sense-­impression inference, is now in trouble. Since, as the Transcendental Deduction was designed to show (and as Sellars accepts!), we cannot do without thinking an object of experience, it is necessary to somehow save a concept of an object of experience which is not a concept of a Kantian (i.e. essentially spatiotemporal) object of empirical reality. In this context Sellars’ method of analogical concept-­formation will turn out to be of central importance. And this method itself will thereby prove to be a piece not only of scientific but of transcendental methodology as well.

Transcendental philosophy: The sense-­impressions Before we explore the subsequent moves of Sellarsian transcendentalism let us turn back one last time to the sense-­impression inference to try to clarify its transcendental dimension. As I see it, there are three distinguishable aspects of the sense-­impression-inferences’ transcendental dimension: Its first aspect is that without it we would, on Sellars’ view, lack a solid foundation for an argument against the view that empirical reality, i.e. the reality of the manifest image, is the true and only reality there is. (This view would be labeled “transcendental realism” with respect to empirical reality according to Kant’s usage of these terms. Kant, of course, would disagree; but Sellars is skeptical about the success of the Antinomies that Kant took to deliver the decisive blow against this kind of realism.39) But there is another aspect to this inference’s transcendental dimension apparent in the following remark: “The [manifold of intuitions; J.H.] has the interesting feature that its existence is postulated on general epistemological or, as Kant would say, transcendental grounds, after reflection on the concept of human knowledge as based on, though not constituted by, the impact of independent reality” (SM 9; my italics–J.H.).40 This impact of an independent reality corresponds to the “guidedness” of our perceptual content, which Sellars shows himself to be so impressed by in the opening pages of Science and Metaphysics. This guidedness, for Sellars, is a transcendental necessity that makes itself felt in the passivity of our experience. Guidance from without, in the manifest image, could, however, be provided by the physical objects naively constructed. A new substitute is not called for until sense-­impressions are introduced. It is only after this introduction that it becomes clear that, on pain of the impending idealistic consequences, Kantian

A Kantian Critique of Sellars’ Transcendental Realism

161

objects of experience—as conceptual constructs produced by the operation of the imagination upon sense-­impressions—are themselves essentially unfit for adopting this role of guidance. For Kantian objects of sense experience, thus conceived, presuppose something that fulfils this role “from without” and cannot—as constructs out of sense-­impressions—be responsible for this guidance themselves. The phenomenality of empirical reality implied by the sense-­impression inference is the second aspect of its transcendental dimension.41 Thus, the pressure for an explanation, even in the admittedly more ‘transcendentally oriented’ Science and Metaphysics, still comes from a comparative consideration of the so-­called “normal” and the “abnormal” case.42 As soon as this comparison is drawn, there is explanatory pressure to account for the qualitative identity of experience in the normal and the abnormal case. And the explanation demanded has to be one in which the normal case is explained just as much as the abnormal case. This, on my interpretation, is the meaning of the quote from Science and Metaphysics McDowell is invoking as evidence for the “transcendental turn” in Sellars’ treatment of the sense-­impression inference, that it “is . . . essential to note that the correlation of the correct conceptual response with objects perceived in normal circumstances by normal perceivers is as much in need of explanation as the correlations of conceptual responses with abnormal perceptual situations” (SM 18). As I have repeatedly observed, the sense-­impression inference in Sellars’ philosophy not only leads to acceptance of the phenomenality of the empirical world. Unlike Kant’s own arguments for the transcendental ideality of empirical reality, it is, furthermore, responsible for our loss of the concept of a Kantian object of experience.43 This, then, is the third aspect of the transcendental dimension of the sense-­impression inference. And it is this dimension which, in Sellars’ philosophical system, must finally lead to scientific realism.

Scientific realism In this summary of the aspects of the sense-­impression inference’s transcendental dimension, I have already referred to the aspect of “guidedness.” This concept is of central importance for assessing Kant’s and Sellars’ view of our relation to the world of which we are a part: this guidedness is the other great presupposition in Kant’s and Sellars’ reflections on this relation. And although Sellars took himself

162

Wilfrid Sellars, Idealism, and Realism

to disagree with Kant on this point, I think one can show that, for both authors, this guidance has to come “from without” the conceptual order.44 After the sense-­impression inference and the loss of the objects of experience, what could possibly play this role of guidance in Sellars’ system? There are only two candidates left at this point of the argument: either it is the sense-­impressions themselves, or the joint operation of the sense-­impressions with that which brings them about—whatever that might be. On the first option we would not have to turn to scientific realism. But this is no longer a genuine alternative, since it does not do justice to the fundamental result of the Transcendental Deduction: the mutual dependence of self-­knowledge and the thinking of an objectively existing reality. Merely subjective sense-­ impressions clearly cannot compensate for the transcendentally necessary measure of objectivity provided by spatiotemporal objects of experience. This option, therefore, is excluded for transcendental reasons. We thus have to think a reality which exists independently of us and which is responsible for bringing about the sense-­impressions from which we construct the phenomenal objects of the manifest image—thereby guiding our experience from without. The guidance is, consequently, the joint effect of independent reality and the sense-­impressions brought about by its impact, i.e. Sellars’ “sheer receptivity.”45 This independent reality is, of course, the Kantian thing in-­itself. According to Kant it is an unknowable condition of the possibility of our knowledge. Sellars, too, is convinced, for the reasons elucidated above, that this dependence is a necessary one. But this independent reality is guiding us from without via the sense-­impressions: we cannot do without the latter because they alone are immediately accessible to the workings of our spontaneity. Even this immediate contact with “inner” sense-­impressions is, however, a guidance “from without” in the sense that these impressions are for us not “given” as what they are in themselves, but always are synthesized, i.e. brought to a conceptual unity, in accordance with the conceptual resources of rational subjects of experience. The capacity responsible for this synthesis is called the imagination. (Here, again, Sellars’ Psychological Nominalism brings him into close attunement with Kant. For Sellars, however, unlike Kant, this disfiguring synthesis can eventually be overcome in the course of scientific development. This difference will shortly prove important for the argument.) Thus, sense-­impressions are, qua states of consciousness, entities we have immediate contact with, but they are at the same time not themselves changed by our spontaneity because this contact is not itself subject to our apperception.

A Kantian Critique of Sellars’ Transcendental Realism

163

That is the reason for Sellars’ insistence that they are states of consciousness that can never themselves become objects of consciousness: as objects of consciousness they would be changed by our spontaneity. For this reason, they can play an important role in the guidance of our experience “from without,” although they cannot, as I argued above, play this role all on their own: we still need the thought of an independent reality that brings about these impressions. It is from our conception of this independent reality that we have to reconstruct a new concept of objects existing independently from us. We have to reconstruct this concept because we cannot settle for a wholly negative conception of that which is responsible for bringing about our sense-­impressions. For otherwise the source of these impressions could, after all, turn out to be we ourselves qua transcendental subjects. But this, in turn, is something we cannot positively think since it would undermine the concept of experience itself as outlined in the Transcendental Deduction. Given that we accept this necessity how should we reconstruct this concept of objective reality? What means are available, on Sellars’ view, for us to do so? At this point there must be a serious difference to Kant’s system, since Kant was convinced that this would not be possible for principled reasons. And it is very important even on Sellars’ own terms that we do not simply give up the sensible features of objects of experience, but build our new concept with the help of the old concept it is used to overcome. To this end Sellars employs another element of the Kantian system and brings it together with his own method of analogical concept formation. This Kantian element is provided by the concepts of pure understanding (i.e. the Kantian categories) as they are employed in thinking about objects independently from our specific human forms of intuition. Unlike Kant, Sellars believes that we can determinately think of (though not intuit) alternative ways of intuiting the objects of our experience. What we need is the input of receptivity together with a method of analogical concept formation that gives this input a new conceptual form. We need “new ways of schematizing the pure concepts of the understanding.”46 The result of this intellectual process of a new schematization of the categories is a new analogical conceptual framework (i.e. successor concepts of the schematized categories of the manifest image) which can then serve as a framework for the development of new empirical concepts on the basis of the same receptive input (i.e. successor concepts of empirical concepts). Sellars repeatedly insists that in this process of “imaginatively” envisaging new world-­stories we cannot abstract completely from the individual terms

164

Wilfrid Sellars, Idealism, and Realism

constitutive of the manifest image. They serve as the coordinates that even the new mapping has to take seriously: According to the picture which I have been sketching, the concepts in terms of which the objects of the . . . manifest image are identified have “successor” concepts in the scientific image, and, correspondingly, the individual concepts of the manifest image have counterparts in the scientific image which, however different in . . . structure, can legitimately be regarded as their “successors”. In this sense, which is not available to Kant, save with a theological twist, the objects of the manifest image do really exist. SM 150

The paradigmatic individual terms are, on Sellars’ view, provided by intuitions: they fix the reference in one system of reference that persists over a change of world-­stories, albeit with a dramatic shift in connected descriptions. This allows for some continuity concerning the individuals, the objects, we are talking about in the different world-­stories: The ‘ “presence” of this unique [world-]story at each stage in the development of the language makes possible the referential framework of names, descriptions, and demonstratives and, by so doing, makes possible the exploratory activity which lead to the story’s enrichment and revision. NAO 11047

It is the concept of picturing and the related conceptual framework that thus allows Sellars to think of the successor framework to the Kantian objects of experience, which he conceives of as objects of the manifest image. Thus, the theory of picturing itself—along with the possibility of analogical concept formation—becomes a transcendental necessity. I therefore understand Sellars’ reference to science in the term “scientific realism” as a reference to the ongoing process of supplementing a new concept of objective reality for the empirical reality we lost in the wake of the sense-­impression inference: the idea of science stands for the idea of the quest for objective knowledge. That this process has not yet come to an end is obvious. But the important thing to keep in mind is that, on Sellars’ view, we have to think of it as completable. It will only be finished when we finally reach an adequate and hence stable picture of how things objectively are. From the point of view of transcendental philosophy, however, it is enough to envisage this end of inquiry as a regulative ideal: something we have to think of as a condition of the possibility of empirical knowledge. The regulative ideal of scientific realism stands for the newly gained

A Kantian Critique of Sellars’ Transcendental Realism

165

possibility of objective knowledge—and therefore guarantees the coherence of the concept of empirical knowledge.48

The “bifurcation” of nature regained? I have pointed to an important difference between Sellars’ and Kant’s approaches to the synthesis of apprehension and the consequences for their respective accounts of sense-­impressions. It has often gone unnoticed that Sellars implicitly rejects Kant’s conception of the synthesis of apprehension, i.e. the taking up of sensible material into the synthesizing activity of the (conceptual) imagination. For Kant, but not for Sellars (!), the synthesis of apprehension subjects the material of the sense-­impressions to a restructuring beyond recognition. For Kant, however, there is, even before that most fundamental synthesis, an unknowable, wholly passively, and non-­conceptually construed output of our forms of receptivity: a “synopsis of sense” (Kant CPR A97). The forms of receptivity have to be thought of, according to Kant, as a passive pre-­structuring or filtering of the input of affections by the thing in-­itself. Those forms of receptivity therefore have to be carefully distinguished from the forms of intuition characteristic of the synthesis of the imagination (a distinction that seems to me lacking in Sellars’ critique of Kant in the first chapter of Science and Metaphysics). The features of the result of this pre-­structuring are as unknowable to us as is the input of affection itself. In contrast, the most basic “material” we can know the features of, is the result of a basic synthesis operating upon this unknowable “synoptic” result—the result of a Kantian synthesis of apprehension. For Sellars, it is the sense-­impressions themselves that are the immediate objects of this fundamental synthesis of apprehension. They thus provide us with the material necessary as a basis for the transcendentally required re-­structuring in the “new schematization” of the scientific image. Kant, in this respect, turns out to be an even more avid enemy of the Myth of the Given than Sellars himself. On the other hand, Kant seems to cross a dangerous line. Was he not aware, one might ask, of the Sellarsian threat losing the object of experience altogether? On Sellars’ account, as I have tried to show, the sense-­impressions (together with their causal history) are the only candidates left for regaining a robust sense of a knowable object existing independently of us. At this point, however, it is important to remember that we observed above that Kant bifurcates nature in a way Sellars finds objectionable.49 The immediate

166

Wilfrid Sellars, Idealism, and Realism

consequence for our present inquiry is that Kant’s theory was never in danger of losing the object of experience (and hence objectivity in phenomenal reality) in the first place. In this respect, Kant consequently seems to be in a much better position, both with respect to the rejection of the given and with respect to conceptual securing of the concept of an object of experience. The reason for this is that Kant was confident that he had given a transcendental argument for the bifurcation that Sellars rejects on phenomenological grounds. The argument for the transcendental necessity of space, put very briefly, goes like this. To conceive of something distinct from ourselves, we need to conceive of it as being outside us. And to conceive of it as outside us, we need to conceive of it as located in space.50 The decisive difference between space and color, in this regard, is that there cannot be a comparable argument to the end that color is equally necessary for conceiving of something as distinct from us. Color (or optical contrast in general) is just one possible qualitative ingredient in the conception of something as distinct from us, as can be easily seen in the example of a person born blind: tactile or auditory sensory input might serve this purpose just as well. We need some quality or other to conceive of things as distinct from us, that much is clear a priori—but not necessarily color. Hence follows the difference between extension and color with respect to their apriority. From Sellars’ perspective, this means neglecting the phenomenological “seamlessness” of color and extension. But given his adherence to transcendental methodology we have to ask why this observation should trump the transcendental reasoning just sketched. It seems, after all, that the bifurcation of nature, so reviled by Sellars, is what enables Kant both to construe guidance through sensory receptivity as a guidance strictly from without and to construct a robust conception of an object of experience. Certainly these features would give us strong grounds for ultimately preferring a Kantian framework to Sellars’ transcendental realism.51

Notes 1 Cf. KTI §17. 2 Cf. CPR A 190/B 235. 3 That is the sense in which the concept of an object of experience “contains a reference to the perceptual takings which are the criteria for its actuality” (KTI §53). 4 This kind of consideration has features very similar to a class of arguments that John McDowell has characterized as a special kind of transcendental argument, cf. John

A Kantian Critique of Sellars’ Transcendental Realism

167

McDowell, The Engaged Intellect. Philosophical Essays (Cambridge, MA: Harvard University Press, 2009), 239. This type of argument aims to establish “how we must conceive the epistemic positions that are within our reach, if it is to be possible that our experience is as it is in having objective purport” (ibid.). 5 This is a very sketchy characterization but suffices for the task at hand. For a careful analysis cf. Eckart Förster, The 25 Years of Philosophy. A Systematic Reconstruction (Cambridge, MA: Harvard University Press, 2012), 1–5, 103–5. 6 This is the phenomenological reason for Sellars’ rejection of the Kantian bifurcation. But he has a further reason as well. On Sellars’ view, Kant failed to distinguish between ideal space as conceived by Newtonian physics and space as form of empirical intuition. Cf. SM 54. I have criticized this argument in another place. Cf. J. Haag, Erfahrung und Gegenstand. Das Verhältnis von Sinnlichkeit und Verstand (Frankfurt a.M.: Klostermann, 2007), ch. 4.2.3. 7 We will see below that this criticism need not be taken to be ultimately successful. And this, in turn, will lead to serious difficulties for Sellars’ own argument. 8 Cf. SRI 191. 9 The argument of SRI brings with it other problems. In connection with the critique of bifurcation (and, thus, with the argument I propose in the remainder of this paper), one might ask why Sellars is talking only about colors and neglecting spatial properties. I suggest that this is due to the special strategic purpose of this text. Only shortly before this passage, Sellars had tried to put considerations about “the physical geometry of micro-­physical entities” (SRI 176 n.9) to one side. His commitment to the ideality of space and time as conceived by the manifest image is, however, unwavering. 10 Cf. SM 171/2. Of course Sellars insists that they may be colored in a different sense in which colors are not treated as epiphenomena. Cf. SM 172 f., SSIS 410 and FMPP 85 ff. 11 Cf. SM 56 n. In this note Sellars seems to lend support to this line of argument. 12 I think it should ultimately be uncontroversial that the introduction of sense-­ impressions does not, on Sellars’ account, transcend the boundaries of the manifest image. This may initially appear to conflict with Sellars’ account of the scientific image as the postulational image. The postulation of sense-­impressions might therefore seem to be a scientific sophistication. But this appearance is easily removed once we recall that sense-­impressions (like thoughts) are modifications or states of persons, i.e. the basic entities of the manifest image. Without going into detail, we can pinpoint the difference between the two images as follows: the manifest image allows as its basic entities only macroscopic physical objects (things) and persons. Every theoretical development that leaves these two categories basically untouched is a development within the manifest image. Therefore, even limited revisionary metaphysics may take place within the manifest image. As soon as one of these basic categories is replaced (of course for theoretical reasons) we are transcending the

168

Wilfrid Sellars, Idealism, and Realism

manifest image and turning to the scientific image. (Bruce Aune was, to my knowledge, the first to forcefully make this point. Cf. Bruce Aune, “Sellars’ two Images of the World”, Journal of Philosophy 87 (1990), 542.) 13 For the continuity cf. for example EPM §21. 14 Cf. SM 16–17. 15 Cf. SK 310, SRPC 178, SSOP 89. 16 Cf. SK 310. The descriptive core of SK is what Sellars in EPM calls the descriptive content. Cf. EPM §22. 17 Sellars writes “ostensible seeing” to incorporate the second phenomenological fact. Cf. below. 18 Though in another sense we can directly see these properties as well. Cf. section 10 below. 19 Cf. SRPC §36. 20 Cf. for example W.P. Alston, “Back to the Theory of Appearing”, Noûs 33 (1999), 181–203. 21 A similar passage can be found, for example, at the beginning of “Phenomenalism,” PH 60. 22 I leave aside the case of hallucination, although Sellars would correctly include it in a statement of the phenomenology of sensations or sense-­impressions. However, since hallucination is not a proper case of perception, I hesitate to add it to my discussion of Sellars’ analysis of perception in the manifest image. 23 In one of his very last papers, “Sensa or Sensings” (SSOP, 1982), Sellars himself in one place hints at the controversial status of this conviction. Cf. SSOP 90. He does so, unfortunately, without providing any new argument for his case. 24 Cf. SM 17. 25 Cf. FMPP 89 n.11 26 Cf. FMPP 21. 27 Cf. FMPP 74. 28 Cf. FMPP 22. 29 With this explanation we reach the proper place of Adverbialism concerning the objects of sensation within Sellars’ philosophical system. As a postulational theory, Sellarsian Adverbial Theory of Sensation is basically a theory about the ontology of perception, not about its epistemology. It is significant that Sellars subtitles his late paper on “Sensa or Sensings” (SSOP) with “Reflections on the Ontology of Perception,” cf. FMPP 33 n. 19. 30 In a footnote to this remark Sellars encourages the reader “to ponder Berkeley’s categorial claim that ‘only an idea can be like an idea’.” (FMPP 89 n.14). 31 Cf. his critique of James Cornman in FMPP 75 f. 32 Thus, sense-­impressions are, on Sellars’ account, not “idle wheels” (McDowell, Having the World in View, p. 16). 33 Sellars writes here “sensa (or sensing)” (ibid.).

A Kantian Critique of Sellars’ Transcendental Realism

169

34 Sellars is touching here upon the evolutionary theme that recurs throughout his writings and gains new prominence in his late papers on “Behaviorism, Language, and Meaning” (BLM 1980) and “Mental Events” (ME 1981). 35 A story has to be told about the ideality of the empirical subject as well. Sellars relates some of the details in the third of his Carus-Lectures, FMPP. In this outline, however, I will concentrate on the side of the object. 36 Cf. section 1. 37 In SK Sellars describes as a “boundary condition” (SK §58) for the introduction of sense-­impressions in the manifest image that “the proper and common sensibles involved are to be construed as qualities of physical objects” (ibid.). The sense-­ impression inference, it ultimately turns out, cannot do justice to this condition and thus destroys the framework from within. 38 That is how he refers to this position in the Carus-Lectures. It led Jay Rosenberg to exclaim “Sellars, then, is (mirabile dictu!) in this sense a Cartesian,” (Jay Rosenberg, “The Place of Color in the Scheme of Things. A Roadmap to Sellars’ Carus Lectures,” The Monist 65 (1982), 329). 39 Cf. SM 53 40 Kant would not speak of “postulating”—and with good reason. And even on Sellars’ own interpretation of the sense-­datum inference as interpretive inference this is not a completely happy way of putting it, since it evokes an additional (FMPP 74) picture of sense-­impressions. 41 Cf. SM 16 42 McDowell argues towards a different interpretation of the transcendental turn in SM in his Woodbridge Lectures. Since his discussion is intimately connected with an interpretation of Sellars’ position in SM that he later explicitly rejects (in his “Sensory Consciousness in Kant and Sellars” and “Preface” in McDowell, Having the World in View), it seems plausible that he would not subscribe to this part of his interpretation anymore. (Further evidence can be found in the remarks in sec. 9 of “Sensory Consciousness” (McDowell, Having the World in View).) 43 That this aspect is separate from the issue of guidance can be seen by realizing that Kant’s objects of experience (at least on a Sellarsian interpretation) are essentially unfit to guide us ‘from without’ anyway. 44 Cf. Haag, Erfahrung und Gegenstand 2007, ch. 4. 45 Cf. for example SM 18. 46 SM 49. Cf. Sellars’ reference in Science and Metaphysics to the “purely structural features.” This is a deeply Kantian idea. (Kant would formulate a similar idea by distinguishing purely formal from material concepts.) 47 All this is, of course, completely alien to Kant’s view. (Here, again, the remarks on the different treatment of the synthesis of apprehension below are important for a clear grip on how different those two views really are with respect to the question of agnosticism concerning the in-­itself.)

170

Wilfrid Sellars, Idealism, and Realism

48 Cf. KTE 635. 49 Cf. section 2. 50 Dryer is, to my knowledge, the first to have pinpointed this argument in his discussion of the Transcendental Aesthetic of the Critique of Pure Reason. Cf. D.P. Dryer, Kant’s Solution for Verification in Metaphysics (Toronto: Toronto University Press, 1966). The argument for time is somewhat more complicated and rests on the mutual dependence of spatial and temporal determination. 51 A more detailed version of this argument can be found in Haag, Erfahrung und Gegenstand, ch. 10.

Bibliography Alston, William P. (1999), “Back to the Theory of Appearing,” Noûs 33: 181–203. Aune, Bruce (1990), “Sellars’ Two Images of the World,” Journal of Philosophy 87: 537–45. Dryer, D.P. (1966), Kant’s Solution for Verification in Metaphysics, Toronto: Toronto University Press. Förster, Eckart (2012), The 25 Years of Philosophy: A Systematic Reconstruction. Cambridge, MA: Harvard University Press. Haag, Johannes (2007), Erfahrung und Gegenstand. Das Verhältnis von Sinnlichkeit und Verstand, Frankfurt a.M.: Klostermann. McDowell, John (2009a), The Engaged Intellect. Philosophical Essays, Cambridge, MA: Harvard University Press. —— (2009b), Having the World in View. Essays in Kant, Hegel, and Sellars, Cambridge, MA: Harvard University Press. Rosenberg, Jay (1982), “The Place of Color in the Scheme of Things. A Roadmap to Sellars’ Carus Lectures,” The Monist 65: 315–35. Sellars, Wilfrid   EPM —— (1963), “Empiricism and the Philosophy of Mind,” Science Perception   and Reality, London: Routledge & Kegan Paul, 127–96.   FMPP —— (1981), “Foundations for a Metaphysics of Pure Process (The Carus   Lectures),” The Monist 64: 3–90.   KTI —— (1976), “Kant’s Transcendental Idealism,” Collections of Philosophy 6:   165–81.   KTE —— (1967), “Some Remarks on Kant’s Theory of Experience,” Journal of   Philosophy 64: 633–47.   NO —— (1980), Naturalism and Ontology, Reseda: Ridgeview Publishing Company.   PH —— (1963), “Phenomenalism,” Science, Perception and Reality, London:   Routledge & Kegan Paul, 60–105.   SK —— (1975), “The Structure of Knowledge,” in H.N. Castañeda (ed.), Action,  Knowledge, and Reality: Studies in Honor of Wilfrid Sellars, New York: Bobbs-Merrill, 295–347.

A Kantian Critique of Sellars’ Transcendental Realism

171

  SM —— (1968), Science and Metaphysics. Variations on Kantian Themes,   London: Routledge & Kegan Paul.   SRI —— (1977), “Scientific Realism or Irenic Instrumentalism: A Critique of  Nagel and Feyerabend on Theoretical Explanation,” Philosophical Perspectives: Metaphysics and Epistemology, Reseda: Ridgeview Publishing Company, 157–89.   SRPC —— (1977), “Some Reflections on Perceptual Consciousness,” in R. Bruzina  and B. Wilshire (eds.), Selected Studies in Phenomenology and Existential Philosophy, The Hague: Martinus Nijhoff, 169–82.   SSIS —— (1971), “Science, Sense Impressions, and Sensa: A Reply to Cornman,”   Review of Metaphysics 25: 391–447.   SSOP —— (1982), “Sensa or Sensings: Reflections on the Ontology of Perception,”   Philosophical Studies 41: 83–111.

8

Psychological Nominalism and Conceptual Relativism: An Idealist’s Take Patrick J. Reider, University of Pittsburgh (USA)

“What is characteristic of philosophy,” Sellars claims,“is not a special subject-­matter, but the aim of knowing one’s way around with respect to the subject-­matters of all the special disciplines” (PSIM 2). The goal of philosophy is thus the achievement of a “unified vision” of “man-­in-the-­world,” i.e., as man perceives himself as a part of the world and the various ways he can intellectually engage it (PSIM 19). This task is extremely challenging, because human knowledge is not pre-­ established as a unified whole for the philosopher to swoop in and clarify. Instead, the philosopher must begin with distinctions such as “fact finding, the ethical, the aesthetic, the logical, the religious, and other” components of experience and then contrive some means to express them as aspects “of one complex picture which is to be grasped reflectively as a whole” (PSIM 4). Sellars’ objective, though not his method, is clearly reminiscent of ambitious German idealistic approaches that sought to bring under one roof the scope and limits of human experience and knowledge. For those familiar with Hegel, Sellars’ objective may illicit the concept of “absolute knowing.” Here, I have in mind a reading of Hegel’s absolute knowing, in which one seeks to understand the basic ways humans come to know and experience the empirical, ethical, aesthetic, logical, religious, etc., and in doing so, arrive at a holistic meta-­ awareness of the types and scope of human knowledge (see section 6). In this chapter, I argue that a loosely pluralistic Hegelian approach to securing a unified view of man in the world is more defensible than Sellars’“stereo-­scopic” view, which seeks to combine the “man-in-the-world image” (of everyday experience) with scientific realism. I argue for this claim by sketching the hitherto unanalyzed link between psychological nominalism and Hillary Putnam’s “conceptual relativism.” I close by outlining the manner in which this link undermines the type of scientific and metaphysical realism that Sellars supported.

174

Wilfrid Sellars, Idealism, and Realism

Sellars’ stereo-­scopic vision and his materialist reduction of man-­in-the-­world The “manifest image” and the “scientific image” are, according to Sellars, the two primary views comprising the “contemporary intellectual scene” (PSIM 6–7). According to him, the manifest image is the contemporary outcome of the conceptual, and hence, categorical investigation of the empirical “experience of ‘man-­in-the-­world’ ”: “It is not only disciplined and critical; it makes use of those aspects of the scientific method which use ‘correlational induction’ ” (PSIM 6–7). What distinguishes the “scientific image” from this manifest image is “stipulation” that “involves the postulation of imperceptible entities and the principles pertaining to them, to explain the behavior of perceptible things” (PSIM 7). Sellars’ dual account shares some common features with Arthur Eddington’s 1928 classic, The Nature of the Physical World. In the “Introduction” of this work, Eddington illustrates that one’s common conception of objects, such as tables, are radically reconceived in science. While some philosophers may talk of unseen or unknowable qualities of material objects, many philosophers and most people going about their daily lives believe that their everyday conception of tables and chairs is adequate for depicting reality. In other words, even though there are diverse ways to conceive of everyday objects, one’s so called “commonsensical” way of thinking about tangible objects and one’s manifest experiences of them are typically believed to be adequate for describing and representing their existence. Put more strongly, objects like tables, chairs, rocks, and bodies as a whole, living or otherwise, are core, non-­reducible features of the manifest image. Take, for instance, Eddington’s explanation of our everyday experience of a table: It has extension; it is comparatively permanent; it is colored; above all it is substantial. By substantial I do not merely mean that it does not collapse when I lean upon it; I mean that it is constituted of “substance” and by that world I am trying to convey to you some conception of its intrinsic nature. It is a thing: not like space, which is a mere negation; nor like time, which is—Heaven knows what! But that will not help you to my meaning because it is the distinctive characteristic of a “thing” to have this substantiality, and I do not think substantiality can be described better than by saying that it is the kind of nature exemplified by an ordinary table. 2012, XI

Psychological Nominalism and Conceptual Relativism

175

What then is the scientific table and how can it challenge the obvious reality of the manifest table described above? At first blush, this question may seem counterintuitive, because science is popularly seen as supporting our commonsense views of existence. As Sellars and Eddington point out, science does not defend the manifest image of substantiality, or in other words, one’s common conception of objects. Instead, it offers an entirely different picture. The table, according to the scientist, is made up of a variety of compounds that are constructed by molecules and atoms. The atoms of an object are not tightly packed together or inert like the manifest image of grains of sand on the beach. They are themselves composed of smaller particles, each of which is moving in a different fashion depending upon the type of particle it is and its relation to other atomic and sub-­atomic influences. Hence, there are not merely spaces between the atoms of so called “solid” objects; there are also spaces between the sub-­atomic parts of atoms. Now the whole of each atom, along with its independently moving sub-­particles, are moving. If one could directly see them, they may appear akin to an uncoordinated and flying swarm of insects. There is thus a sense in which objects are a seething storm of tiny particles. For example, Eddington notes that his table holds up his paper, because it bumps (with small imperceptible motions) against the opposing particles that constitute his paper: two tightly woven clouds of particles (i.e., the table and his paper) exerting force upon one another, which produces the result of being impenetrable at the macro level of sensibility (2012, XII). Moreover, such sub-­atomic particles may themselves, according to some physicists, be constituted by bandwidths of highly localized bits of energy. Egads! Not only is matter not solid, that is, it is no longer substantial as viewed in the manifest image, it may not even be matter—for it may be energy. According to some scientists, the manifested experience of physicality is a type of illusion constructed by our macro level senses that are incapable of perceiving its inner workings and causes. It is believed that the inner universe of micro particles and their interrelated causal chains are so numerous that they make the astrophysicist’s “astronomical” calculations seem like the local kiddie pool, rather than the worldly ocean we thought we knew. Now, unlike Eddington, Sellars does not believe that science is doomed to be theoretical or instrumental, in the sense that it merely provides symbolic ways of penetrating and predicating our manifest world (2012, 36). As a “scientific realist,” Sellars believes that science can offer the best description and explanation of existence. And though any rational scientist will fully admit that there is plenty we do not know, the scientific realist holds a type of faith in regards to the

176

Wilfrid Sellars, Idealism, and Realism

claim that there is nothing a complete science cannot (one day) comprehensively explain or properly describe. Despite Sellars’ scientific realist commitments, Sellars surprisingly rejects the view that the scientific image should dominate or slowly replace the manifest image. Why does he bother to take such a stance, if science can properly reveal and describe the way things are? Here is where Sellars makes things interesting for the scientific realist and idealist, who believe knowledge is largely a product of our shared human orientations and the manner in which the mind functions: [The “manifest image” is the] framework in terms of which man came to be aware of himself as man-­in-the-­world. It is the framework in terms of which, . . . man first encountered himself—which is, of course, when man came to be man. For it is no merely incidental feature of man that he has a conception of himself as man-­in-the-­world, just as it is obvious, on reflection, that “if man had a radically different conception of himself he would be a radically different kind of man”. PSIM 6

One cannot simply do away with the manifest image by replacing it with the scientific image, because human beings are not the type of beings they are without the manifest image. Consequently, if philosophy ever hopes to achieve a unified vision of man-­in-the-­world, it must not only account for the theoretical potency of science but also understand how the manifest image can legitimately co-­exist with the scientific image. Up to this point, I support the majority of Sellars’ claims. Namely that philosophy ought to seek a unified picture of man-in-the-world, the manifest image and the scientific image are distinct views of existence, and that one of these images should not replace the other. It is Sellars commitment to scientific realism that I reject. As a scientific realist, Sellars must accept the claim that even behaviors and thoughts of human beings can be reduced (without explanatory loss) to physical terms. He “provisionally” assumes “that although behavioristics and neurophysiology remain distinctive sciences, the correlational content of behavioristics points to a structure of postulated processes and principles which telescope together with those of neurophysiology theory, with all the consequences which this entails” (PSIM 25). In short, “man” can be seen as “a complex physical system” without any explanatory loss concerning our manifest experiences of oneself, others, and existence (PSIM 26). If the above were true, as Sellars believes must be the case, thoughts, intentions, awareness, and personality are all fully explainable in some properly advanced material science. I hold deep reservations concerning the viability of explanatory/

Psychological Nominalism and Conceptual Relativism

177

materialist models of such mental states; however, I will not address them here.1 Instead, I argue that there is a more primary way in which materialism/scientific realism, as a form of metaphysical realism, fails.

Conceptual holism and schemas Kant believed human reason requires systematic unity. He argued that a manifold of inborn capacities and the multifaceted interrelations that such capacities enabled made such unity possible. This conviction paved the way for his transcendental idealism, by which he sought to deduce the necessary conditions for experience. Kant believed this approach ensured timeless (or ahistorical) truths concerning the human mind. Sellars interprets Kant’s transcendental idealism as claiming that there is an insurmountable epistemic gap between the framework of human representation and existence itself. In other words, he reads Kant as claiming that knowledge of reality, as it is in-­itself, is unknowable (SM). Sellars’ Science and Metaphysics is largely a protracted argument for why he believes (in theory) that the thing in-­ itself is knowable. Like Kant, Sellars’ molding of concepts is predicated upon a manifold of pre-­ established conceptual networks that enable the possibility of reason by ensuring the possibility of coherency and the underlying unity of conceptual thought. This aspect of Sellars’ philosophy has been aptly called “conceptual holism.”2 Unlike Kant, Sellars rejects the claim that conceptual abilities are inborn and largely unchanged by historical development. Even though they are drastically opposed on this issue, they nonetheless agree that one can discern universal requirements or preconditions for human reason. They further agree that this premise makes certain a priori insights about the human mind possible. For instance, both philosophers hold the basic presumption that the human mind is knowable, in that all human minds share the same basic conditions for rational thought and experience. So, while Sellars is not a transcendental idealist by any stretch of the term, the underpinnings of his philosophical approach are very much indebted to Kant’s approach. In this regard, Sellars’ conceptual holism (and more broadly, his psychological nominalism) can be interpreted as a continuation of this a priori thread of Kant’s transcendental idealism (i.e., there are universal and necessary conditions for human reason and experience). It would, however, be untrue to claim that Sellars’ conceptual holism was solely a product of his familiarity with Kant. Sellars’ holism, like Quine’s “meaning

178

Wilfrid Sellars, Idealism, and Realism

holism,” was also a reaction to failed attempts in the early half of the twentieth century to offer a term-­by-term, and subsequently, a sentence-­by-sentence reduction of meaning.3 The Positivists, for instance, argued for a non-­holistic account of the meaning of words/concepts, because they assumed that cognitive meaning and content are ultimately modular, independent, and self-­sustaining. Similar to conceptual holisms and meaning holism, Robert Brandom argues that concepts are “modally robust,” in that they invoke or preclude other concepts depending on their use and the occasion in which they are employed. This seems to be a deeply Hegelian theme found in both Sellars and Brandom. Hegel, for example, emphasizes that “determinate negation” is a fundamental aspect of the phenomena of conceptual determination, i.e., what gives anything a determinate cognitive status stems from our typically unspoken ability to distinguish what something is, by understanding what the concepts associated with an entity are not. With the shared ability to understand which concepts are invoked and which concepts are excluded by a given concept, the meaning, use, and intention of our words, thoughts, and language are rendered reflectively intersubjective, i.e., one can intentionally return to a given body of content and express it to other members of her linguistic community. For example, the assertion that “this ball is red” obtains its meaning from the unspoken expectation or principle that its “redness” precludes it from being entirely a different color at the time of its utterances. In this manner, the designation “red” only has meaning in a larger framework in which one conceives the existence of many colors and understands that colors are ascribed to the appearance of objects. This is similarly true of its status as a “ball,” which precludes it from being a pyramid or a cube. Conversely, its status as an “object” calls one to associate with it concepts such as extension and mass, which function as an implicit linkage, even though a person may utter, or explicitly think, only one concept. And of course, one’s conception of “objects” and the “properties” that properly belong to them can encompass a large and sophisticated interlocking of commitments, beliefs, orientations, and expectations. In this fashion, even one’s thoughts or assertions concerning something as simple as a toy can quickly snowball into cascading relations that engage broad swaths of one’s conceptual framework. Now that these points have been made explicit, it should be obvious that one cannot start from a tabula rasa (i.e., blank slate) and then spontaneously generate the relevant linkages that define the holistic requirements of rational thought without some pre-­established guidelines. According to Sellars’ “psychological nominalism,” and those in the Pittsburgh School who have taken it up, such as

Psychological Nominalism and Conceptual Relativism

179

Robert Brandom and John McDowell, only through the slow accumulation of habituated behaviors via language acquisition can one manifest the gestalt effect of these necessary interrelations that enable rationality/conceptual understanding. The achievement of this gestalt effect requires what Brandom calls “policing.” The term “policing” refers to a given community’s collective efforts to sanction or repudiate proper and improper linguistic and assertive practices. The act of “policing” thus concerns the enforcement of the acceptable application of concepts as an unspoken obligation for each member of a linguistic community. Policing thus ensures pre-­established and navigable approaches of relating concepts—for those, like a child, who are in the process of entering into a linguistic community. This phenomenon creates shared practices and exceptions (i.e., “norms”) by which a shared public use of concepts transpires according to pre-­established conditions generally deemed appropriate (according to a particular linguistic community). Much of these normative expectations, principles, and practices are unspoken and may even be unarticulated—like those who learn to speak without ever being taught the formal rules of grammar or even those who are entirely ignorant of their existence. Though of course, normative practices, expectations, and principles are apt to change and grow as the conditions and desires of one’s linguistic community develop (Reider 2016a). Such developments are typically far more fluid than changes in formal grammar. Hence, the normative aspect of rationality via the acquisition of language, as implied in Sellars’ physiological nominalism, is a community centered enterprise that is subject to various historical influences and advancements. In this manner, Kant’s transcendental philosophy is replaced with a social/linguistic model that engenders the possibility of understanding and reason. As Sellars famously notes in “Empiricism and the Philosophy of Mind,” normative practices guide the manner in which concepts are employed. For instance, I should not claim that a ball is “gray” when I am looking at it under the moonlight or certain forms of artificial lighting, when I know that it will look green in the daylight.4 Conventions like using daylight as the appropriate condition for determining colors are often an unspoken and unreflective response, like one’s speech conforming to grammatical rules. Such conventions establish the preconditions for intersubjectivity by creating the conditions for common use.5 Do such Sellarsian inspired insights overemphasize the role of the linguistic community rather than the mental activity of individuals? Hilary Putnam provides a good response to this type of question:

180

Wilfrid Sellars, Idealism, and Realism

There is a linguistic division of labor. Language is a form of cooperative activity, not an essentially individualistic activity. Part of what is wrong with the Aristotelian picture is that it suggests that everything that is necessary for the use of language is stored in each individual mind; but no actual language works that way. . . . In sum, reference is socially fixed [i.e., normative] and not determined by conditions or objects in individual brains/minds. Looking inside the brain for the reference of our words is . . . [in most cases] just looking in the wrong place. 1988, 25

Putnam’s point is that one cannot possess all the relevant features that represent an object and all the ways it makes sense to conceptually think about that object through the spontaneous generation of one’s own mental processes.6 Such capacities are not merely inhuman in that they misrepresent humans at a psychological level; they also require god-­like powers, which are inconsistent with our finite character. I have been building the case that Kant is correct in assuming that our empirical and rational determinations require pre-­established structures, i.e., a manifold bundle of linkages that renders our judgments about empirical content and rational determinations coherent, relatable, and hence unifiable. The brilliance of Sellars is displayed in his ability to recognize that the schematic relations of concepts are not an inborn attribute of being human, as Kant (and more recently individuals such as John Searle) believed. Instead, he argued that such relations are linguistically acquired, historically developed, and socially distributed. This is the case, because linguistic use is derived from a community’s norms—for it is the normative practices of a community that underwrite the conditions in which one is deemed warranted or unwarranted in uttering certain types of assertions. As argued above, the conceptual linkages that render concepts understandable and open to public assent or dissent lie in the shared practices and commitments of one’s linguistic community and not the disparate activities of individuals. This facet of human rationality ought to have been the death knell of individualistic conceptions of epistemology and given rise to the popular realization that epistemology is irreducibly a social phenomenon. In short, all epistemology must be “social” epistemology, just as a tiger or bear must be a mammal to be a tiger or a bear.7

Conceptual bias and conceptual relativism Let us now return to the point at hand: Kant’s central claim in the Critique of Pure Reason is that human beings can only rationally understand concepts when they have some pre-­established conceptual system in place to coherently

Psychological Nominalism and Conceptual Relativism

181

relate concepts. This is perhaps Sellars’ greatest debt to Kant, as he reworks Kant’s transcendental idealism into psychological nominalism. An important outcome of both theories is that, if you change how concepts relate, you change the manner in which experiences are conceptually articulated, and hence, understood. This is one of the reasons why Sellars believes we cannot do away with the manifest image and entirely remove and replace it with a purely scientific one. The point here is not that such a feat is merely impossible, but rather that such an occurrence would change what it means to be human. It is also what I take to be the most salient feature of Hilary Putnam’s conceptual relativism: that, historically, many conceptual frameworks are in place for human beings, each of these conceptual frameworks permits different types of conceptual linkages, and in turn, such frameworks render different types of rational experience. And while some of these conceptual frameworks are surely false, it may nonetheless be possible that two or more opposing views can be simultaneously true. Let us begin by unraveling these claims. First, let’s readdress the notion that the way we view the world requires a large system of conceptual orientations that function according to normative principles. As Putnam notes: . . . Newton’s theory of Universal Gravitation (without any added statement specifying boundary conditions) is compatible with any orbits whatsoever. One could even reconcile square orbits with the theory of universal gravitation, by saying, “Well, that means there are non-­gravitational forces acting on the system.” It is only in the presence of a large body of statements that one derives all of its so-­called “Consequences” from a scientific theory. As Quine puts it, sentences meet the test of experience “as a cooperate body,” and not by one. 1998, 8–9

Here, Newton’s theory of orbits needs the supplemental principles concerning their elliptical natures. The reason why orbits are elliptical draws on other areas of science with each of their own conceptual domains, guiding norms, and principles. Hence, the manner in which the world appears to hang together, and the implicit bias of any such view, is relegated to how people understand the conceptual relations that are implied in the assertions they and others make. Conceptual bias thus has enormous metaphysical undertones, because it predetermines what type of entities are available to the linguistic user when making determinations. In other words, different conceptual schemas may offer alternative pictures of existence. Perhaps the best way to see what is at stake in this issue is to reduce the number of entities that exist, as Putnam does in The Many Faces of Realism. If we

182

Wilfrid Sellars, Idealism, and Realism

assert, for the sake of argument, that just three entities exist in a particular universe, how many entities exist as a whole? Putnam points out that many logicians, metaphysicians, and scientists perceive the relations that entities bear to one another as entities unto themselves. Since a relation can be seen as an entity, three entities could proliferate into as many as seven (or possibly eight!) entities (depending on how one works out the combinations in which they may be related) (Putnam 1987, 19). The question then becomes: in what sense do relations exist? By what means does one discern a real from false relation? The above questions cannot be revealed by science as an empirical endeavor, for there are innumerable ways entities can be viewed as bearing some relation. For instance, as Putnam points out, one can form a relation between his nose and the Eiffel tower. Perhaps our intuition is to ask: yes one could form such a relation, but is such a relation real or artificial? It seems unlikely that it is a natural relation; yet, there is nonetheless a fine line between which objects bear a natural relation from a constructed relation. For instance, what is the boundary of a cloud, a mountain range, or a solar system? There are many legitimate ways science can address these questions. This is the case, because it is unclear how one can be empirically informed about “natural” relations, especially if one does away with the direct seeing of essences as psychological nominalism requires. Here, I am not claiming that all relations are imperceptible. Rather, I am attempting to build the case that many asserted relations are conventional and that any line of demarcation between the two seems utterly dependent on the conceptual scheme one uses, which makes it unlikely that such issues are fully resolvable as having solely one legitimate answer. In short, the logical entailments of relations hinge upon the conceptual scheme one employs. These issues are extremely important, as they bring real challenges to Sellars’ scientific view of the world, e.g., when he claims that science can provide (at some unspecified future date) one true account of existence (SM). Consider, for instance, the following claims: despite objects being composed of atomic parts, they are the singular entities that we commonly perceive them to be; objects are mere aggregates of atomic parts; objects are a mere effect of their individual atoms; objects are the sum total of all their sub-­atomic relations; and objects are the relations they bear to each other as seen in the notion of an event. Which, if any, of these metaphysical claims is the truth? Well, according to Putnam’s conceptual relativism, there is a sense in which they are all correct, even though many of them can oppose one another at a certain metaphysical level. For Putnam, the existence of opposing claims that are simultaneously true, according to different conceptual schemes, is not the issue. Nor is he simply

Psychological Nominalism and Conceptual Relativism

183

claiming that there is more than one way to conceptually cut up the same substance, like cookie cutters that create different shapes but nonetheless cut the “same dough” (1987, 19). Rather, Putnam’s conceptual relativism is the stronger claim that more than one opposing claim, as it concerns real metaphysical differences, and not just conventions of linguistic characterizations, can be simultaneously correct: The sentence “Between any two points on a line there is a third point” and its “translation” into the version in which points are identified as the convergent sets of concentric spheres have the same truth conditions in the sense that they are mathematically equivalent. The answer to the question “Do the two sentences have the same meaning?” is that the ordinary notion of meaning simply crumbles in the face of such a question. It was never meant to do that job. Putman 1992, 119

These two ways of understanding space cannot be translated as having the same meaning, because the conceptual schemes and the principles that make them function as mathematical equivalents also constitute different metaphysical pictures of existence or at least incompatible ways of describing it. Our current topic has important implications as to whether or not bivalence is even a sustainable claim: Metaphysical realists to this day continue to argue about whether points (space-­ time points, nowadays, rather than points in the plane or in three-­dimensional space) are individual or properties, particular or mere limits, etc. My view is that God himself, if he consented to answer the question, “do points really exist or are they mere limits?”, would say “I don’t know”; not because his omniscience is limited, but because there is a limit to how far questions make sense. Putnam 1987, 19

Here, Putnam is suggesting that the assertions “points are properties,” “points are individuals,”“points are particulars,” or “points are mere limits” are neither true nor false in the sense that the true one represents a unique one-­to-one correspondence with a prestructured existence, and all the rest are false. Here, the various ways of conceiving points are tied to deep and sophisticated ways of understanding space, all of which have their own merits and faults. This suggests that existence may not be definable in one absolute or, more importantly, solely true and unique way. If it is the case that meaning stems from use in a conceptual scheme, then there is no extra-­human framework to endow the truth or falseness of our statements, because there is no meaning outside of human conceptual schemes. The above claim, however, is not a denial of facts. Putnam unequivocally claims that “[t]here are ‘external facts’, and we can say what they are. What we cannot

184

Wilfrid Sellars, Idealism, and Realism

say—because it makes no sense—is what the facts are independent of all conceptual choices” (1987, 33). The point then is that we should not confuse the way we speak about the world as the world itself, our human mode of understanding with existence itself, or that there must be one true way to describe existence. Instead, when using a particular conceptual scheme that does not have manifest flaws, one is correctly describing existing according to that scheme, and not existence qua existence. Since one cannot employ meaning and assertions outside of a language and conceptual schema, it is simply misguided to expect that the objects and relations one asserts are wholly the way things are independent of the human subject. This is a significant Kantian move. In The Many Faces of Realism, Putnam reminds the reader that in his Reason, Truth, and History, he “presented Kant as the first philosopher to reject the idea of truth as correspondence to a pre-­ structured Reality” (43). He goes on to say that, if Kant is claiming that “truth must not be thought of as correspondence to a pre-­structured or self-­structured Reality, if he was saying that our conceptual constitution cannot be factored out and that the ‘markers-­true’ and the ‘markers-­verified’ of our beliefs lie within and not outside the conceptual systems,” then Kant holds the same type of realism as himself (Putnam 1987, 43). The problem then is what to make of this type of “relativism” that Putnam upholds. On the one hand, it is not a radically subjective relativism in that the individual is the sole proprietor of right and wrong, true or false. Rather, his claim is that the only viable means of thinking about such designations occurs within a publicly shared conceptual scheme. The scheme itself predetermines, to some existent, what is permissible and impermissible to say about an entity. The other hand concerns experience itself as it plays out under a particular conceptual schema. Either certain assumptions will be “confirmed” or “supported” by experience or they will not. So in this fashion, not all conceptual schemes are of equal epistemic value, because some are inconsistent. Experience, even while operating under a deficient conceptual system, can be counter indicative of the system itself, because the system fails to achieve what it says it achieves, it fails at the practical level, it is overtly self-­contradictory, etc. Hence, not only can different conceptual schemes be epistemically unequal, some can be overtly false. In this regard, Putnam’s conceptual relativism, as well as my own view on this matter, is a rejection of the post-­modern turn, in which everything considered to be “truth” or “knowledge” is merely a matter of narrative, while at the same time embracing the fact that metaphysical realism is unattainable. Let us now consider the entailments of the above claims. Rationality understood as the proficient utilization of interrelated concepts requires the

Psychological Nominalism and Conceptual Relativism

185

skillful employment of normative practices. Such practices are functional for humans, if and only if such practices are finite and limited. In this regard, once one begins with one set of concepts over that of another, and once one links a limited number of concepts in a particular way that permits their public use, one correspondingly smuggles in, via their finite and limited number, use, and relation, a particular perspective that precludes or engenders various kinds of oriented experience towards oneself, others, and existence. This latter claim has further consequences. On the one hand, concepts are, in the above sense, schematized by linguistic practices of entire communities, which charge the meaning of words with specific metaphysical entailments by predisposing their users to form certain conceptual linkages. Yet, on the other hand, different schemas use different concepts, and more importantly, they form different relations among concepts that authorize certain linguistic practices and prohibit others. Both these components of human Spirit, i.e., the resulting product of the economy of many consciousnesses cooperating together in the communitive act over a historically significant span of time, form a reciprocal shaping of linguistic practices, schemas, and metaphysical orientations. These normative forces thus create conceptual bias by predisposing people to certain conceptual characterizations and by predisposing them not only to the concepts they tend to employ, but far more importantly, the conceptual relations they perceive and take for granted. In review, conceptual bias brings with it a unique metaphysical picture insofar as it affects the manner in which things are perceived to hang together, i.e., one’s conception of a particular state of affairs. This occurs, because our very mode of thought/language requires the use of a particular conceptual schema. Since each particular conceptual schema is limited, it starts with certain relations that could be drawn other than those currently expressed. This engenders certain underlying assumptions garnered from the contingent and particular conceptual scheme one enacts. In short, conceptual schemas hold within themselves implicit conceptual biases, which predispose one to use certain types of conceptual determinations.

Picturing, structural fallibilism, and the contingent life-­world Sellars writes, in Science and Metaphysics: “If, however, as I shall propose . . . we replace the static concept of Divine Truth with a Peircean conception of truth as the ‘ideal outcome of scientific inquiry’, the gulf between appearances and thing-­ in-themselves, though a genuine one, can in principle be bridged” (SM 50).

186

Wilfrid Sellars, Idealism, and Realism

Unlike Sellars, Putnam argues that conceptual relativism makes it impossible for human beings to obtain, even in theory, a final and singular account of existence. According to Putnam, metaphysical realism, as it concerns scientific realism, is actually a collection of claims concerning existence and knowledge—namely “Independence, Uniqueness, Bivalence, and Correspondence are regulative ideas that the final scientific image is expected to live up to, as well as the metaphysical assumptions that guarantee that . . . a final scientific resolution for all physical problems must be possible” (1998, 107). As we will soon see, Sellars does not accept many of the key views of knowledge, truth, and representation, as implied in Putnam’s above account of metaphysical/scientific realism. So the question then arises, whose epistemic stance is correct, as it concerns the possibility of a final and adequate account of existence—Sellars’ or Putnam’s? While Sellars’ views on knowledge, representation, and truth are complex and open to several types of interpretation, it should be relatively uncontroversial to note that his conception of “picturing” plays an important role in his understanding of the above topics. It also plays a dominant role in Science and Metaphysics (his self-­proclaimed sequel to EPM), wherein he concludes his section on “Picturing” as a “token payment” for his claim that it is “ ‘scientific objects’, rather than metaphysical unknowables, which are the true things-­in-themselves” (SM 79, 143). For Sellars, picturing “maps,” as a kind of representation, the “factual features” of physical existence. The ability of certain types of statements to reference these relations permits a conception of truth that Sellars believes is more primary than traditional conceptions of truth: “within this level essential distinctions must be drawn if we are to grasp the difference between the primary concept of factual truth (truth as correct picture), which makes intelligible all the other modes of factual truth, and the generic concept of truth” (SM 9, 119). DeVries notes two important views that follow from Sellars’ notion of picturing as a type of “mapping”: “First, there must be some kind of ‘projection relation’ that specifies how the configurations of objects in the map are related in the configuration of the objects they represent,” i.e., something that spells out the isomorphic relation between the map and the objects the map concerns. He then writes: Secondly, it takes a whole system to represent in any interesting sense. A dot in the middle of an otherwise blank page could be said to represent New York and its environs in some systematic way, but without some other structures present in the “map” relating to New York in some systematic way, the assertion seems empty. deVries 2005, 53

Psychological Nominalism and Conceptual Relativism

187

Sellars believes truth and knowledge are not a direct word to world correspondence or a simple imagining that “mirrors” existence, but rather knowledge and truth concern how one can systematically map/represent factual features of existence. Knowledge and truth must primarily concern the objects of material science for Sellars, because, insofar as one is concerned with “describing and explaining the world, science is the measure of all things, of what is that it is, and of what is not that it is not” (EPM 83). Sellars notes that there are different linguistic systems, each of which offers varying degrees of “adequate picture[s]” and presumably varying degrees of inadequate pictures (SM 68, 140). Sellars, however, is not troubled by this occurrence, for he believes an ideal language is possible: “Let us now go one step further and conceive of a language which enables its users to form ideally adequate pictures of objects, and let us call this language Peirceish” (SM 68, 140). While I have written about the problems of a Periceish approach elsewhere (Reider 2016a), the faults of this view, as it concerns Sellars, are perhaps more convincing coming from a strong supporter and key commenter of his work. Take for instance Willem deVries’ following commentary on Sellars’ Peircean view of mapping: [I]t is far from clear to me that there is a well-­defined notion of final and ultimate adequacy, even if the picturing relation is naturalistic and not semantic, for adequacy seems fundamentally interest-­relative. For instance, a thick tome of US Geological Survey maps would be quite a bad thing to use on a cross-­country drive, whereas a standard road atlas is geared to that purpose. There are many different maps of the Earth, depending on the projection used, and no one map is optimal for all purposes. . . . Could there be a map adequate for all purposes? Since purposes themselves can conflict, the presumption, it seems to me, is against such an ideally adequate map. deVries 2005, 54–5

It is unclear how a “ ‘well-­defined notion of final and ultimate adequacy” will ever be understood to be achieved. As deVries’ above excerpt alludes to, that which counts as a better answer or a more accurate account depends on human interests, i.e., a map’s structure is a derivative of the human purposes it serves. If this is true, it seems unreasonable to assume we can arrive at a standard that indicates an ideal scientific map that is free of human interests, because the presumed standard of the ideal will be largely indebted to the specific purposes the mapping serves within the multifaceted goals of the researchers developing it. It seems likely that all “maps,” in so far as they serve some function, no matter how much

188

Wilfrid Sellars, Idealism, and Realism

their authors strive to just “stick to the facts,” are necessarily marred by the human interest. In other words, structured mapping of phenomena possesses strengths and weaknesses dependent on its specific use and the purposes such use serves. My latter point calls into question how much explanatory cash Sellars can put into his claim that science can permit knowledge of the thing-­in-itself (at some unspecified future date). For instance, Sellars claims that picturing can provide an “Archimedean” point lacked by Peirce, and as a result, he implies that it kept Peirce from achieving a program that permitted knowledge of “what really exists” (SM 75, 142). Presumably, Sellars believes his model of picturing achieves such results. However, if it is the case that all mapping/picturing is necessarily intertwined with human interests, such circumstances call into question the reliability of the fulcrum (i.e., picturing) Sellars needs to get his metaphysical realism off the ground. Let me be clear by what I mean above. It is not the specifics of so-­called well-­ established scientific facts that I dispute. Nor do I wish to deny science its function (or purpose) as an explanatory tool or its power to allow us to knowingly influence our perceived environments. Instead, I have doubts that the manner in which facts are claimed to hang together, i.e., how they are pictured, can legitimately overcome the “gulf between appearances and things-­in-themselves.” The origin of my doubt lies in the fact that the understanding of epistemic claims always occurs within conceptual schemes, which in turn provide the claims concerning a specific entity their particular meaning and truth value. Most of Sellars’ commentators find such a view epistemically innocuous as it concerns his realism, but I do not. Once one accepts that concepts are our only access to understanding maps/pictures, we run the risk of conceptual relativism and, as a result, all the challenges it presents to metaphysical realism providing a singular account and epistemic view of existence. The problem of conceptual relativism begets other problems as well. To frame some of these problems, we need to first address the historical change of conceptual schemes. For instance, knowledge claims will change in epistemically relevant ways that place knowledge claims of one era at odds with subtle, yet significant, structural differences in how knowledge claims are conceived in another era. This creates conceptual bias, which is apparently a stable feature of historical change. The danger of conceptual bias is that its occurrences are rarely explicit. This is problematic when the biases of one’s scheme leads to false assertions about a variety of phenomena. It is doubly troublesome when it is perfectly adequate on

Psychological Nominalism and Conceptual Relativism

189

other matters, which only muddies the waters concerning the manner in which the scheme goes astray. These schematic shortcomings are tricky to spot for two reasons: (1) as already noted, conceptual bias is typically implicit, rather than overt, and (2) there are always unforeseeable ways that conceptual schemas can be organizationally and relationally changed. The latter is a huge epistemic thorn for a conceptual holist like Sellars, in that, while a scientist may weed out many structural problems in a given portion of a scheme, new ones tend to sprout up elsewhere due to the unforeseen structural changes that can occur to the entire scheme when one makes tiny changes to even relatively small portions of it. Hence, falsifying schematic changes can transpire, i.e., structural fallibilism, even when much of a given scheme’s core content remains intact. This phenomenon makes it difficult to be aware of such epistemically relevant occurrences, until one is afforded the luxury of historical hindsight. Conceptual relativism lends support to the view that matter-­of-factual maps (if such things are even possible in quite the way Sellars hopes) can be assessed to offer two or more “true” but opposing claims at the conceptual level. Not only is this a problem in and of itself for the metaphysical realist, but what if these two competing conceptual schemas tended to produce false claims in other areas that neither concerned the conceptual relative claims nor had a counterpart in its competitor? This possibility illustrates how difficult it is to discern the extent to which a conceptual schema is adequate, as it concerns some specific purpose, much less to argue that a conceptual schema is progressively marching forward as the proper medium to convey knowledge of things-­in-themselves. Even if one denies that the never ending historical process of conceptual schematic change offers a real threat to the metaphysical realist, there is also the contingent state of human existence with which to contend. In brief, the manner in which humans seem to be temporally and spatially oriented (primary aspects of any empirical investigation) to the world casts our depictions of existence according to certain human orientations, i.e., our life-­world. So, while we might guess at what happened billions of years ago or what transpires in the smallest sub-­atomic particles, we are still anchored to the conceptual life-­functions of our everyday life, which Sellars calls the “manifest world.” This affects not only how we think about temporal/spatial/causal factors but also how we map, model, and represent phenomena. I see no reason not to believe that, if we radically changed our dimesons, e.g., if humans were the size of quarks or galaxies, or if we consciously lived out our entire lives in milliseconds or billions of years, that the pictures in which we cast scientific facts would hang together in a much different fashion than they

190

Wilfrid Sellars, Idealism, and Realism

currently do. (Of course, this does not require that each of our facts would necessarily be disputed by such alternative versions of ourselves.) In short, if you change the manifest image, as Sellars argues, you change what it is to be human, and in doing so, you change how the world is conceived. More importantly, it may change the manner in which a physical science maps and pictures existence. It is not that the human concept of the world must be fundamentally flawed, or that it cannot be fundamentally correct, but as my teacher Thomas Rockmore often said, “how would one ever know?” In other words, how can we know to what extent our claims/representations are accurate, true, final, or adequate, independent of the contingent (i.e., could be otherwise) framework in which one casts the wide net of interrelated concepts, expectations, and norms that underpin all understanding and upon which all truth and falseness is judged? Faced with such concerns, I do not believe “picturing” is the Archimedean point Sellars believed he discovered.

Closing remarks We have uncovered a stance that opposes Sellars’ unified picture of “man-­in-the-­ world”: there are not just two images of existence that hold legitimate claim to our modern intellectual understanding but as many as there are distinct and effective conceptual schemas. As a result, I believe the scientific realist (of a Sellarsian bent) is forced to accept one of two alternatives, both of which he or she will find undesirable: either that psychological nominalism is fundamentally wrong (which seems to be very indefensible) or that the plausibility of science as a means of securing metaphysical truth and as an ultimate and final word on existence (i.e., metaphysical realism) is far more problematic than traditionally upheld under non-Sellarsian models of concepts. This is the case, not because there is something fundamentally flawed about science or modest forms of materialism as an approach or guide to the study of existence (as opposed to a final and complete account). Rather, neither is well suited to overcome the issues implied in psychological nominalism (e.g., conceptual bias and relativism) that challenge the metaphysical realist paradigm as an achievable end. What then ought we seek, if not an ultimate description of existence? Why don’t we pursue the Absolute defined as the limit and scope of human experience? What if we study this according to the different shapes that various historical, religious, aesthetic, and fields of science portray it as? Why not look at the influence of the family, government, and culture on our conception of ourselves

Psychological Nominalism and Conceptual Relativism

191

and existence? In seeking answers to these questions, let us become aware of the different enumerations and progressive developments of mankind’s epistemic standing towards existence. If we can achieve some moderate satisfaction of these tasks and reflectively obtain a meta-­awareness of each of these relevant features as they form, in very broad strokes, the unity or totality in human thought, then I suggest we have arrived at something very close to what Hegel calls “absolute knowing.” If the ultimate goal of philosophy is to arrive at a unified picture of “man-­inthe-­world,” Sellars’ approach fails. Why? Because scientific realism as the final arbitrator and description of the truth is a fool’s errand, due to the combined phenomena of psychological nominalism, conceptual holism, conceptual bias, structural fallibilism, and conceptual relativism. The better option then seems to be some loose approximation of the Hegelian program, which honors (or in the very least accommodates) the real historical presence of these phenomena. In closing, why should we accept Putnam’s position that his conceptual relativism is a type of “lower” case realism he calls “internal realism?” Why not simply call it “internal idealism?” Does the determining of facts or asserting that there is an external existence independent of us preclude one from idealism? Clearly, anyone who would proclaim “yes” misunderstands the idealist tradition, for few idealists ever made such claims. Perhaps the better question is simply to ask, “why introduce the term ‘idealism’ when ‘realism’ is all the rage?” If we wish to show that (1) we do not have a direct way to know existence that is free of judgment, (2) what makes something true or false is the way phenomena conform or fails to conform to a well constituted conceptual scheme, (3) if due to the latter we cannot know existence independent of conceptual schemes that may significantly differ, and (4) these combined points shed doubt on our general ability to understand mind-­independent reality, then what stance could we be holding other than idealism? If these themes do not strike you as stemming from the German idealist tradition and simultaneously undermining the main assumptions of traditional realism, then we are at an impasse, for your conception of history is irreconcilable with my own. And in this instance, I believe only one of us can be right.8

Notes 1 Here, I do not reject outright the strong possibility that consciousness must originate from the physiology of the brain. Instead, I doubt that science can offer an accurate

192

2 3 4 5 6 7

8

Wilfrid Sellars, Idealism, and Realism

explanation of how such structure permits experience, intentionality, and freedom. However, the scope of this claim is outside the topic of this chapter. DeVries and Triplett characterize Sellars’ model of concepts as requiring “conceptual holism.” See Knowledge, Mind, and the Given, 29–32. Putnam, see 8–9, and Quine. See Sellars’ handling of this issue in “Empiricism and the Philosophy of Mind,” 73–7. This insight that seems to be at the heart of much of Wittgenstein’s Philosophical Investigations. In a way, this can be seen as an early precursor to extend mind theory. For more on social epistemology and German idealism see chs. 9 and 10 from Social Epistemology and Epistemic Agency: Decentralizing the Epistemic Agent, Patrick J. Reider (ed.), London: Rowman & Littlefield, 2016. I wish to thank Johannes Haag for helpful comments on an early version of this chapter.

Bibliography Brandom, Robert (1994), Making it Explicit, Cambridge MA: Harvard University Press. deVries, Willem (2005), Wilfrid Sellars, Durham: Acumen. deVries, Willem and Timm Triplett (2000), Knowledge, Mind, and the Given, Indianapolis: Hackett. Eddington, Arthur (2012), The Nature of the Physical World, New York: Cambridge University Press. Hegel, G.W.F. (1997), Phenomenology of Spirt, New York: Oxford University Press. Kant, Immanuel (2005), Critique of Pure Reason, New York: Cambridge University Press. Putnam, Hilary (1987), The Many Faces of Realism, Chicago: Open Court. —— (1992), Renewing Philosophy, Cambridge MA: Harvard University Press. —— (1998), Representation and Reality, Cambridge MA: Massachusetts University of Technology. Reider, Patrick, J. (2016a), “Epistemic Agency as a Social Achievement: Rorty, Putnam, and Neo-German Idealism,” in Social Epistemology and Epistemic Agency: DeCentralizing Epistemic Agency, Patrick J. Reider (ed.), London: Rowman & Littlefield International. —— (2016b), “A ‘Dialectical Moment’: Desire and the Commodity of Knowledge,” in The Future of Social Epistemology: A Collective Vision, James Collier (ed.), London: Rowman & Littlefield International. —— (2015c), “The Internet and Existentialism: Kierkegaardian and Hegelian Insights,” Social Epistemology and Technology: Toward Public Self-Awareness Regarding Technological, Frank Scalambrino (ed.), London: Rowman & Littlefield International, 59–68.

Psychological Nominalism and Conceptual Relativism Rorty, Richard (1980), Philosophy and the Mirror of Nature, Princeton: Princeton University Press. Sellars, Wilfrid   EPM —— (1997), Empiricism and the Philosophy of Mind, Cambridge MA:   Harvard University Press.   PSIM —— (1991), “Philosophy and the Scientific Image of Man,” in Science,   Perception, and Reality, Atascadero: Ridgeview Publishing.   SM —— (1992), Science and Metaphysics, Atascadero: Ridgeview Publishing   Company.

193

Index abstract ideas 22 animal cognition 36, 139 Aristotle 49–50

judgment 7, 10–11, 45–8, 51, 54–5, 68, 84, 107, 112, 114, 139–40, 156, 180, 191 justification 26–8, 45, 64, 67, 95, 107, 130

Brandom, Robert 10, 19, 25, 30, 33–5, 41–9, 53–5, 105, 109, 111–16, 117, 121, 123, 125–37, 139–44, 146, 178, 179 Burge, Tyler 46, 50

Kant, Immanuel 12–14, 28–32, 36, 53–4, 59, 67–8, 71, 95–6, 103–9, 112, 114–18, 121–5, 128–9, 134, 137, 139–40, 142, 145 n.2, 146, 149–52, 156, 159–66, 177, 179–81, 184

categories 5, 34, 71–2, 106–7, 117, 146, 163 cognition 59, 104–8, 110, 114, 117, 128 coherency 110–11, 165, 177 concepts 5, 6–9, 10–11, 12–14, 23, 24–5, 28, 27–36, 50, 54, 61, 64, 66–8, 71, 84–8, 93–6, 111–15, 117, 133, 139, 151, 156–7, 161, 163, 177–80, 182–5, 188–90, 191 conceptual bias 180–5, 188 conceptual holism 10, 14, 177–8 conceptual relativism 182–9, 191 consciousness 22, 43–4, 59, 67, 117, 162–3 constructivism 107, 116–18 deontic modality 83 empiricism 2, 4, 6, 19, 24, 28, 31, 35, 60, 84, 106, 109–10 epistemology 11, 134–5, 146, 180 given, the 7, 14–15, 21, 22, 24–5, 35–6, 42, 47–9, 53, 65, 70, 73, 105–6, 108–17, 156, 165–6 Hegel, G. W. F. 13–14, 25, 29, 30, 35, 41–3, 46–59, 51, 54–5, 95–6, 103, 105, 106, 112, 114, 115–18, 140, 173, 178, 191 idealism 14, 29–30, 41, 84, 89, 91, 95, 99, 103–4, 106, 114, 116–18, 150, 177, 181, 191 intentionality 19, 20, 23, 27, 35–6, 43, 48, 62–3, 83, 95, 97, 135, 137–9, 141–2

manifest image 65, 70, 74–5, 94, 122–6, 128, 133–42, 152–3, 158–60, 163–4, 174–6, 181, 190 McDowell, John 10, 19, 25, 27, 25–35, 104–5, 109–11, 113–17, 124, 128, 139–40, 144, 152, 161, 179 metaphysical realism 103, 184, 186, 188 metaphysics 12, 32, 60, 66, 71–2, 122, 167 natural laws 31, 71, 115 naturalism 12, 14, 33, 35, 87, 106, 129, 146 normativity/norms 9–13, 19, 23–36, 43, 45, 63, 65, 83–9, perception 13, 23, 32, 44, 46–9, 52–4, 67–70, 72, 74, 104, 150–9 person 3, 11, 33, 91, 132, 134–59 picturing 164, 186–8, 190 Pittsburgh School 10, 103, 105–6, 111, 116–17, 134, 139, 178 Plato/Platonism 19, 26, 32, 34–6, 41, 88, 108 Platonic realism 20–2 psychological realism 42–3, 55 Putnam, Hilary 112, 173, 179–84, 186, 191 qualia 43, 66, 69 realism 1–2, 4, 6, 13–14, 29–33, 36, 41, 59, 60, 64, 84, 104, 125–6, 129, 135, 151–2, 156, 160, 161–4, 176, 184, 186, 191

196

Index

reference 4, 31, 35, 91, 134, 151, 157, 164, 180 representation 20, 24–5, 34, 36, 50, 52–3, 62–3, 67–8, 88–9, 103–5, 107, 116–17, 146, 150–1, 177, 186, 190 Rorty, Richard 19, 21, 25–7, 31–3, 35, 41–6, 53–5, 66, 121, 123–4, 132–3, 135, 139 scientific image 14, 49, 65, 74–5, 93, 121, 123–5, 128–9, 131, 135–8, 141, 152–3, 164–5, 167 n.12, 174, 176, 186 scientific realism 129, 151–2, 156, 161–2, 176, 186, 191 sensation 2–4, 22–3, 50, 52, 59, 65–6, 68–74, 77–8, 106, 108, 151, 153, 155, 157–9

sense-­impressions 68–9, 73–4, 153, 155, 157–8, 160–5, 167 n.12 space of reasons 5, 25–6, 28, 33–6, 65, 105, 108–14, 117–18, 138, 140, 144 spirit 185 structural fallibilism 189 teleology 92, 95–7 thought 1, 6, 9–10, 19–20, 24, 28–9, 36, 42, 49–55, 61–3, 70, 92, 103, 105, 113, 158, 167 n.12, 177–8, 185, 191 transcendental 2, 30–1, 60–1, 64–5, 74, 106–7, 129, 134–7, 139, 149–53, 156, 159–6, 177, 179, 181 universals 1–6, 108