Information and the History of Philosophy (Rewriting the History of Philosophy) [1 ed.] 0815355009, 9780815355007

In recent years the philosophy of information has emerged as an important area of research in philosophy. However, until

481 41 52MB

English Pages 392 [393] Year 2021

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Information and the History of Philosophy (Rewriting the History of Philosophy) [1 ed.]
 0815355009, 9780815355007

Table of contents :
Cover
Half Title
Series Information
Title Page
Copyright Page
Table of Contents
Figures and Tables
Contributors
Acknowledgements
Introduction
Natures
Access
Control
(Dangerous) Systems
Insurgencies
Notes
Part I Information Before 500 CE: Natures
1 Yinyang Information: Order, Know-How, and a Know-How,-Relation-Based Paradigm
A Short Description of Yinyang
Yinyang Information As Reflecting of the Order of Things
Yinyang Information As Know-How
Relations-based Yinyang Information Model
Notes
2 Plato On the Act of Informing: Meaningful Speech and Education
Introduction
Meaningful Speech
A Lógos As a Unit of Information
Names As Descriptions
Education
Different Sorts and Sources of Information
Accessing Information
Conclusion
Notes
References
3 On Information in Aristotle: Nature, Perception, Knowledge
Introduction
Environmental Instructional Information in Aristotle?
Reception of Perceptible Forms in Perception
Reception of Information and Acquisition of Knowledge
Notes
References
4 Information and History of Psychiatry: The Case of the Disease Phrenitis
The Label Phrenitis and Greek Traditional Discourses About the Mind
The Body of the Phrenitic Patient
Head Or Heart? Phrenitis at the Crossroad Between encephalocentrism and Cardiocentrism
Quasi Phreniticus: the Disease As Metaphor
Conclusions
Notes
References
Part II Information 500–1500: Access
5 Vacaspati On Aboutness and Decomposition
Sautrantika Decompositionalism
A Partial Argument for Sautrantika Decompositionalism
How Decompositionalism Leads to Idealism
The Argument Against Idealism
The Nature of Aboutness
A. Aboutness As a Natural Relation
B. Aboutness As a Self-Linking Relation
C. Aboutness As An Irreflexive Relation
Taking Stock
Notes
Bibliography
6 Seeing and Recognition in the Arabian Nights and Islamic Alexander Legends
Shahrazad, Benjaminian Storyteller of Experience
The Epistemology of Recognition Narrativized
Alexander, the Seeing Traveler
To See Is Not to See
Conclusions
Notes
7 Avicenna On Information Processing and Abstraction
Introduction
The Cognitive Content of Sense-Perception
Abstraction and Emanation
Avicennian Issues in Contemporary Epistemology
Notes
References
Further Reading
8 Thomas Aquinas On Cognition As Information
Cognizant Beings
“Receiving Forms Without the Matter”
Spiritual/intentional Vs Natural Existence
Intentional Species in the Medium
Intelligible Species
The Role of Species in Cognition
Notes
References
Part III Information 1500–1800: Control
9 Leibniz As a Precursor to Chaitin’s Algorithmic Information Theory
Introduction
The Rudiments of AIT: Complexity, and Irreducibility
Leibniz’s Anticipations
Determinism, Contingency and Computability
Physical Laws and Information
Conclusion
Notes
References
10 Information Visualisation in the Philosophical Transactions
Introduction
Journal Visualisations
Diagrams
Maps
Graphs
Visualise for User Performance
Conclusion
Notes
References
11 ‘Dwindled Into Confusion and Nonsense’: Information in a Copyright Perspective from the Statute of Anne to Google Books
Early Copyright: ‘An Act for the Encouragement of Learning’
Kant’s Theory of Copyright
Daniel Defoe and the Freedom of the Press
The Right to the Integrity of the Work
Google Books
Conclusion
Notes
References
12 Information in the Pursuit of Social Reform
Introduction: the Need for Adequate Information in the Pursuit of social and Democratic Reform
Requisites for Advances in Social Science
Major Sources On Methodology
The 17th Century, Emergence of Scientific Method
18th-century Enlightenment, French and British
19th-century Social Physics and Social Science
Mme De Staël and the Pursuit of “Geometric” Political Science
Abolition of the Slave Trade
Other Correspondence With Thomas Jefferson
Florence Nightingale: Information for Social Reform
A Cautionary Conclusion
Bibliography
Further Reading
Part IV Information in the nineteenth Century: (Dangerous) Systems
13 The Nineteenth-Century Information Revolution and World Peace
An All-Encompassing Revolution
The Post Office
After Hill
The Cultural Effects of Imperial Communication
Conclusion
Notes
14 Charles Babbage’s Economy of Knowledge
Introduction
Standardizing and Economizing On Knowledge
The Division of Labour and the Babbage Principle
Innovation
Transactions Costs
Search Costs and Middlemen
Knowledge and Progress
Notes
References
15 Mendel On Developmental Information
Introduction
Mendel and Gärtner On Development (entwicklung)
Mendel’s “developmental Series (entwicklungsreihe)”
Mendel’s Novel Conceptualisation: The Laws of Developmental Series
Mendel and the Study of Heredity
Conclusion
Notes
References
16 Information and Eugenics: Francis Galton
Introduction: The Scramble for Information
Data Gathering: Methods and Objects
Data Sharing
Inheritance
Ethics
Aftermath
Notes
Bibliography
Further Reading
Part V Information After 1900: Insurgencies
17 The Racialization of Information: W.E.B. Du Bois, Early Intersectionality, and Social Information
Du Bois and the Inauguration of Intersectional Social Information
The Distinctiveness of Du Bois’s Contributions to Social Information
Du Bois, Rural Information, and Urban Information
Du Bois and Racial Information
Du Bois and Gendered Information
Du Bois and Religious Information
Du Bois and Educational Information
Du Bois and Criminological Information
Du Bois, Intersectional Sociology, and Social Information
Bibliography
Further Reading
18 The Many Faces of shannon Information
Introduction
Recalling the Formalism of Shannon’s Theory
The Roots of Shannon’s Theory
About the Concept of Shannon Information
About the Interpretation of Shannon Information
From Shannon Theory to Quantum Information Theory
A Formal View of Shannon Information
The Many Faces of Shannon Theory
References
19 Computers and System(s) Science—the Kingpins of Modern Technology: Lotfi Zadeh’s Glimpses Into the Future of the Information Revolution
Introduction
The Information Revolution
A Biographical Sketch
A View of the Future (from the 1950s)
Let’s Go Digital!
The First Generalization of System(s) Theory
In the Wake of a Sabbatical at the Ias
Calling for An Institute
Fuzzy Sets
Computing With Words
Second and Third Generalization of System Theory
Education in Electrical Engineering and Computer Science
Two Further Views of the Future (from the 1970s)
Conclusions
Notes
Bibliography
Index

Citation preview

i

I N F O R M AT I O N A N D T H E H I S T O RY O F P H I L O S O P H Y

In recent years the philosophy of information has emerged as an important area of research in philosophy. However, until now information’s philosophical history has been largely overlooked. Information and the History of Philosophy is the first comprehensive investigation of the history of philosophical questions around information, including work from before the Common Era to the twenty-​first century. It covers scientific and technology-​centred notions of information, views of human information processing, as well as socio-​political topics such as the control and use of information in societies. Organised into five parts, 19 chapters by an international team of contributors cover the following topics and more: • • • • •

Information before 500 CE, including ancient Chinese, Greek and Roman approaches to information; Early theories of information processing, sources of information and cognition; Information and computation in Leibniz, visualised scientific information, copyright and social reform; The nineteenth century, including biological information, knowledge economies and information’s role in empire and eugenics; Recent and contemporary philosophy of information, including racialised information, Shannon information and the very idea of an information revolution.

Information and the History of Philosophy is a landmark publication in this emerging field. As such, it is essential reading for students and researchers in the history of philosophy, philosophy of science and technology, and library and information studies. It is also a valuable resource for those working in subjects such as the history of science, media and communication studies and intellectual history. Chris Meyns is a poet, developer and architectural conservationist based in Uppsala, Sweden. They have published on the history of data, on Anton Wilhelm Amo’s philosophy of mind, and their book The Philosophers’ Library: Books that Shaped the World (with Adam Ferner) will appear in 2021. Their current research focuses on vulnerability in information sharing ecosystems.

ii

Rewriting the History of Philosophy Series editors: Aaron Garrett and Pauliina Remes

The history of philosophy has undergone remarkable growth in the English language philosophical world. In addition to more and better quality translations of canonical texts there has been a parallel expansion in the study and research of sources, thinkers and subjects hitherto largely neglected in the discipline. These range from women philosophers and late ancient thinkers to new Western and non-​Western sources alike. Simultaneously, there has been a methodological shift to far greater intradisciplinary and interdisciplinary perspectives in the history of philosophy, cutting across the humanities and social sciences. Rewriting the History of Philosophy is an exciting new series that reflects these important changes in philosophy. Each volume presents a major, high quality scholarly assessment and interpretation of an important topic in the history of philosophy, from ancient times to the present day, by an outstanding team of international contributors. The Senses and the History of Philosophy Edited by Brian Glenney and José Filipe Silva Molyneux’s Question and the History of Philosophy Edited by Gabriele Ferretti and Brian Glenney Information and the History of Philosophy Edited by Chris Meyns Thought: A Philosophical History Edited by Panayiota Vassilopoulou and Daniel Whistler For more information about this series, please visit: https://​www.routledge.com/​Routledge-​ Handbooks-​in-​Applied-​Ethics/​book-​series/​RWHP

iii

I N F O R M AT I O N A N D T H E H I S T O RY O F PHILOSOPHY

Edited by Chris Meyns

iv

First published 2021 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 605 Third Avenue, New York, NY 10158 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2021 selection and editorial matter Chris Meyns; individual chapters, the contributors The right of Chris Meyns to be identified as the author of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-​in-​Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-​in-​Publication Data Names: Meyns, Chris, editor. Title: Information and the history of philosophy / edited by Chris Meyns. Description: Milton Park, Abingdon, Oxon ; New York, NY : Routledge, 2021. | Series: Rewriting the history of philosophy | Includes bibliographical references and index. Identifiers: LCCN 2020048364 (print) | LCCN 2020048365 (ebook) | ISBN 9780815355007 (hardback) | ISBN 9781351130752 (ebook) Subjects: LCSH: Information science–Philosophy–History. Classification: LCC Z665 .I5777 2021 (print) | LCC Z665 (ebook) | DDC 020.1–dc23 LC record available at https://lccn.loc.gov/2020048364 LC ebook record available at https://lccn.loc.gov/2020048365 ISBN: 978-​0-​8153-​5500-​7 (hbk) ISBN: 978-​0-​3677-​5564-​5 (pbk) ISBN: 978-​1-​351-​13075-​2 (ebk) Typeset in Bembo by Newgen Publishing UK

v

CONTENTS

List of figures and tables  Notes on contributors  Acknowledgements 

viii x xiv

Introduction  Chris Meyns

1

PART I

Information before 500 CE: Natures 

11

1 Yinyang information: Order, know-​how, and a relation-​based paradigm  13 Robin R. Wang 2 Plato on the act of informing: Meaningful speech and education  Tamsin de Waal

27

3 On information in Aristotle: Nature, perception, knowledge  Miira Tuominen

41

4 Information and history of psychiatry: The case of the disease phrenitis  Chiara Thumiger

59

PART II

Information 500–​1500: Access 

77

5 Vācaspati on aboutness and decomposition  Nilanjan Das

79

v

vi

Contents

6 Seeing and recognition in the Arabian Nights and Islamic Alexander legends  Anna Ayse Akasoy

107

7 Avicenna on information processing and abstraction  Luis Xavier López-​Farjeat

123

8 Thomas Aquinas on cognition as information  Cecilia Trifogli

137

PART III

Information 1500–​1800: Control 

151

9 Leibniz as a precursor to Chaitin’s Algorithmic Information Theory  Richard T. W. Arthur

153

10 Information visualisation in the Philosophical Transactions  Chris Meyns

177

11 ‘Dwindled into confusion and nonsense’: Information in a copyright perspective from the Statute of Anne to Google Books  Stina Teilmann-​Lock 12 Information in the pursuit of social reform  Lynn McDonald PART IV

198 211

Information in the nineteenth century: (Dangerous) systems 

227

13 The nineteenth-​century information revolution and world peace  Edward Beasley

229

14 Charles Babbage’s economy of knowledge  Renee Prendergast

245

15 Mendel on developmental information  Yafeng Shan

262

16 Information and eugenics: Francis Galton  Debbie Challis and Subhadra Das

281

vi

vii

Contents PART V

Information after 1900: Insurgencies 

299

17 The racialization of information: W.E.B. Du Bois, early intersectionality, and social information  Reiland Rabaka

301

18 The many faces of Shannon information  Olimpia Lombardi and Cristian López

324

19 Computers and system(s) science—​the kingpins of modern technology: Lotfi Zadeh’s glimpses into the future of the information revolution  Rudolf Seising

341

Index 

367

vii

viii

FIGURES AND TABLES

Figures 0.1 N-​g ram for “information age” and “information revolution”, 1900–​2000  1.1 A guqin (1634)  3.1 A classification of information  9.1 A replica of Leibniz’s calculating machine  9.2 A diagram of I Ching hexagrams owned by Leibniz  10.1 Number of documents published in the Philosophical Transactions, 1670–​1879  10.2 Percentage of documents published in the Philosophical Transactions with one or more diagram, map, or graph, 1670–​1879  10.3 Diagram of a setup for experiments with gunpowder  10.4 Diagram of the leg of a turkey  10.5 Map of trigonometrical operations  10.6 Meteorological map of winds and monsoons  10.7 Histogram of atmospheric pressure  10.8 Two graphs with “Life-​Tables” for “Healthy Districts” of England  12.1 Diagrams showing the causes of mortality in the Crimean War  15.1 Illustration of the behaviour of a dominating parental trait (denoted by A) and a dominating hybrid trait (denoted by Aa)  16.1 Weather map printing plates  16.2 Galton’s counting gloves  16.3 Galton’s sextant  16.4 A hand-​held dynamometer from the Galton Collection  16.5 A headspanner from the Galton Collection  16.6 Galton’s quincunx  18.1 Parts of a general communication system 

viii

10 18 42 157 159 178 179 181 182 184 185 187 188 216 272 284 285 285 287 287 289 325

ix

Figures and tables

18.2 Relationship between the entropies of the source H(S) and of the destination H(D) of a message in a general communication system  19.1 Lotfi A. Zadeh, undated photo, approximately in the 1940s  19.2 Headline of Zadeh (1950) in the Columbia Engineering Quarterly  19.3 Headline of Zadeh (1954) in the Columbia Engineering Quarterly  19.4 Zadeh’s (1960) call for institutions in IRE Transactions  19.5 Lotfi A. Zadeh, undated photo, October 1968  19.6 Zadeh’s hierarchical stack of methodologies  19.7 Participants in the Conference on Graduate Academic and Related Research Programs in Computing Science, June 1967  19.8 Elements of the fuzzy set “computer science” and their grades of membership 

325 344 345 345 351 353 354 356 358

Tables 15.1 Scholars cited in Mendel (1866)  15.2 Result of the first generation hybrid (Mendel 1866) 

ix

264 271

x

CONTRIBUTORS

Anna Ayse Akasoy is Professor of Islamic Intellectual History at the Graduate Center of the City University of Islam. Her areas of expertise include medieval Islamic philosophy, the relationship between the Islamic world and other cultures, and the history of falconry. Richard T. W. Arthur is a philosopher based in Toronto and Professor Emeritus at McMaster University, Hamilton (Ontario). He specialises in early modern natural philosophy and mathematics, the theory of time, and the infinite. He is the author of G. W. Leibniz: The Labyrinth of the Continuum (Yale UP, 2001), Leibniz (Polity Press, 2014), Introduction to Logic (Broadview, 2017), Monads, Composition and Force (Oxford UP, 2018), and The Reality of Time Flow (Springer 2019) and has several works in preparation. Edward Beasley is Professor of History at San Diego State University. He is the author of several books on imperial and racial categories in nineteenth-​century England, including The Victorian Reinvention of Race: New Racisms and the Problem of Grouping in the Human Sciences (2010) and The Chartist General: Charles James Napier, The Conquest of Sind, and Imperial Liberalism (2017). His latest book, to be published in 2021, is on the social construction of diabetes mellitus. Dr Debbie Challis is Education and Outreach Officer at LSE Library. She has curated a number of exhibitions including Typecast: Flinders Petrie and Francis Galton in 2011 and written several articles on racial science and archaeology as well as the book The Archaeology of Race: The Eugenic Ideas of Francis Galton and Flinders Petrie (Bloomsbury, 2013). Nilanjan Das is Lecturer in Philosophy at University College London. He works on epistemology and Indian philosophy in Sanskrit. His work has been published in Noûs, Philosophy and Phenomenological Research, British Journal for the Philosophy of Science, Review of Symbolic Logic, and Journal of Indian Philosophy. Subhadra Das is Curator of the Science Collections at University College London. In 2017 she curated Bricks + Mortals, an exhibition and podcast walking tour examining UCL’s pivotal

x

xi

Contributors

role in establishing eugenics as a science. Her forthcoming book is about science, power, and race and why we need to talk about Francis Galton. Olimpia Lombardi is an electronic engineer and has a PhD in Philosophy from the University of Buenos Aires. She is Superior Researcher of the National Scientific and Technical Research Council, Argentina, Director of the Argentine Group of Philosophy of Sciences at the University of Buenos Aires, Member of the Academie Internationale de Philosophie des Sciences, in Brussels, and of the Foundational Questions Institute, Charter Honorary Fellow at the John Bell Institute for the Foundations of Physics, in New York, and Research Associate of the Centre for Philosophy of Natural and Social Sciences (LSE). Cristian López has a PhD in Philosophy from the University of Buenos Aires and the University of Lausanne. From April to September 2020, he was Assistant Researcher at the University of Lausanne, and from October on, FNRS Postdoctoral fellow based at the CEFISES, University of Louvain. In 2017 and 2018, he was Visiting Researcher at the Munich Center for Mathematical Philosophy (LMU) and at the University of Oxford. He specialises in the philosophy of physics, in particular, in the philosophy of time, quantum mechanics, and space-​time symmetries. Luis Xavier López-​Farjeat is Associate Professor at the Faculty of Philosophy at Universidad Panamericana, Mexico City. He has written on Classical Islamic philosophy and is co-​editor of Philosophical Psychology in Arabic Thought and the Latin Aristotelianism of the 13th Century (2013) and of The Routledge Companion to Islamic Philosophy (2016). In 2018 he published Razones, argumentos y creencias. He is Associate Director of the Aquinas and ‘The Arabs’ International Working Group and Editor of Tópicos: Journal of Philosophy. Lynn McDonald, a Canadian, has a PhD from the London School of Economics, is a former member of Parliament and is now a Professor Emerita. She is the author or editor of 25 books, including the 16-​volume Collected Works of Florence Nightingale and three books on theorists, including the women contributors. Her latest paper on the subject, “Sociological Theory: The Last Bastion of Sexism in Sociology,” was given originally as a paper at meetings of the American Sociological Association. Chris Meyns is a poet, developer, and architectural conservationist based in Uppsala, Sweden. They have published on the history of data, on Anton Wilhelm Amo’s philosophy of mind, and their book The Philosophers’ Library: Books that Changed the World (with Adam Ferner) will appear in 2021. Their current research focuses on vulnerability in information sharing ecosystems. Renee Prendergast is Reader in Economics in the Management School at Queen’s University Belfast, Northern Ireland. Her main research interests are in technology and development and in the history of economic thought. Her publications include Choice of Technique in Iron Founding (with B. A. Bhat), Machine Tools in Developing Countries (with M. M. Huq), Development Perspectives for the 1990s (with H. W. Singer), Market Forces and World Development (with F. Stewart), Contributions to the History of Economic Thought (with A. Murphy), and A History of Irish Economic Thought (with T. Boylan and J. D. Turner).

xi

xii

Contributors

Reiland Rabaka is Professor of African, African American, and Caribbean Studies in the Department of Ethnic Studies at the University of Colorado, Boulder. He is also Research Fellow at the College of Human Sciences at the University of South Africa (UNISA). Rabaka is the author of numerous scholarly articles, book chapters, and essays, as well as more than a dozen books, including most recently Du Bois: A Critical Introduction (Polity, 2021). Rudolf Seising obtained his PhD in Philosophy of Science and the German habilitation in History of Science from the Ludwig Maximilian University in Munich after studying Mathematics, Physics, and Philosophy at the Ruhr-​University of Bochum. He is College Lecturer at the Faculty of History and Arts at the Ludwig Maximilian University of Munich and he is now leading the three-​year project “IGGI” regarding a history of artificial intelligence in West Germany, funded by the German Federal Ministry of Education and Research. Yafeng Shan is Research Associate in Philosophy at University of Kent, UK. He is the author of Doing Integrated History and Philosophy of Science: A Case Study of the Origin of Genetics (Springer, 2020) and has published research articles in Philosophy of Science, Synthese, and Philosophy Compass. He is primarily interested in the philosophy of science, epistemology, and metaphysics. He completed his PhD at University College London. Stina Teilmann-​Lock is Associate Professor at the Department of Management, Politics and Philosophy at Copenhagen Business School. Her research focuses on intersections of law, design, and technology and she has published widely on intellectual property law, design, and digital transformations. Chiara Thumiger obtained her Dr. PD (2004) from Kiel University and is a classicist and historian of science. She has worked on a variety of medical themes and authors from the Hippocratic Corpus (her monograph A History of the Mind and Mental Health in Classical Greek Medical Thought, published in 2017 by CUP) to the late-​antique world and beyond (Mental Illness in Ancient Medicine. From Celsus to Paul of Aegina, Brill 2018, co-​edited with P. N. Singer). Most recently she has researched the history of the ancient disease phrenitis; the results of this work are now under submission as a monographic volume. She has also published on tragedy, ancient animals, and the history of emotions. Cecilia Trifogli is Professor of Medieval Philosophy at the University of Oxford. She has published extensively on the medieval tradition of Aristotle's natural philosophy and psychology. She is the author of a book on the English commentaries, Physics (Oxford Physics in the Thirteenth Century). She also works on the critical editions of medieval philosophical texts. She has recently edited (together with Silvia Donati and Jennifer Ashworth) the Quaestiones on the Physics by the Oxford master, Geoffrey of Aspall. Miira Tuominen is Senior Lecturer of Theoretical Philosophy at the University of Stockholm and before that was Professor of Philosophy at the Department of Social Sciences and Philosophy at the University of Jyväskylä. Following on from her PhD (University of Helsinki, 2002), Tuominen has mainly published on ancient philosophy and on themes related to knowledge, philosophical psychology, and argumentation. More recently, she has been working on Porphyry’s On Abstinence from Killing Animals on which she has written several articles and a monograph that has not been published yet.

xii

xiii

Contributors

Tamsin de Waal is currently Visiting Scholar in Philosophy at Columbia University. She has previously taught at Kings College London and University College London. Her primary research interests are in the areas of Ancient Philosophy and the Philosophy of Mathematics. Robin R. Wang is Professor of Philosophy at Loyola Marymount University, LA. Her teaching and research centre on Chinese and Comparative Philosophy, particularly on Daoist Philosophy, women and gender in Chinese culture and tradition. She is the author of Yinyang: The Way of Heaven and Earth in Chinese Thought and Culture (Cambridge University Press, 2012) and editor of Chinese Philosophy in an Era of Globalization (SUNY Press, 2004) and Images of Women in Chinese Thought and Culture: Writings from the Pre-​Qin Period to the Song Dynasty (Hackett, 2003).

xiii

xiv

newgenprepdf

ACKNOWLEDGEMENTS

The editor of this volume would like to acknowledge a debt—​both moral and practical—​to all librarians, to book swappers, to indexers, editors, and typesetters, to data analysts, to system administrators, to file-​sharers, to anyone and everyone who sends and receives email, to indie web developers, to map-makers, to those who install and maintain road signage, to navigators, to language learners and instructors, to copyists, to open access advocates and science liberators, to painters, draughtspeople and photographers, to minute takers, to sculptors and sound engineers, to data centre operators, to adblockers, to privacy advocates, to all the information freedom fighters out there. Plus, a massive thank you to the rebels securing our (and the earth’s) informational futures.

xiv

1

INTRODUCTION Chris Meyns

At the start of the pandemic of 2020, facial recognition systems were momentarily thrown into disarray. How to identify faces, if everyone venturing outside covers mouths and noses with virus-​repellent cloth?1 That these machine learning systems could soon make do with simply people’s eyes and facial shape is, to me, telling, not just because of the nauseating extent of continued on-​and offline surveillance it reveals.2 It is curious, equally, because it illuminates how even the tiniest segments of our bodies can function as living, walking repositories of information. Not just our bodies: information extends from fossils to footprints, from DNA to demographics, from barometer readings to behavioural profiles, from statistical regularities to true randomness. Information is many things, information is all around us, information reaches into any and all aspects of people’s lives. Furthermore, once extracted, information can be stored, manipulated, transformed, analysed and transmitted to whomever or whatever stands in the right chain of communication (as well as being able to be lost or destroyed). Information’s omnipresence and adaptability signal its significance: ethically, individually, socially, economically and politically. Yet, perhaps hiding behind covers of novel terminology (“information” only introduced itself to the English language in the fifteenth century), within philosophical histories it still craves sustained study. This volume seeks to offer its readers such philosophico-​historical reflection, in the guise of a set of individual, self-​contained “confrontations” with information throughout history. Bundled into five sections, organised broadly chronologically, it contains 19 individual studies which grab a text, idea, or movement in the history of (thinking about) information and lay out its ethical, epistemic, socio-​political and environmental significance—​holding them up, drawing closer to the light for maximum scrutiny, even where that occasionally gets uncomfortable.

Natures Already well before the Common Era, we find philosophers theorise about the extent to which information is pervasive, manifest throughout the natural and social world. In ‘Yinyang information: Order, know-​how, and a relation-​based paradigm”, Robin R. Wang focuses on information in Classical Chinese philosophy, more specifically in yinyang thinking: that is, thinking attuned to yinyang. 1

2

Chris Meyns

In yinyang thinking, yin (陰) and yang (陽) are the ultimate opposing, interdependent, interacting, complementary principles that structure an ever-​changing dynamic reality. This much is evident in the change of seasons, operations of natural elements, as well as in the relationships among all entities. It also governs human affairs, such as social relations, correct governance and how to maintain a healthy body. Order, in yinyang thinking, is emergent rather than predetermined, may constantly change, leaving an uncertain, not fully predictable, future. Drawing on classical texts including the Shijing (first millennium BCE), and works from the Warring States period (476–​221 BCE) such as the Zhuangzi, the Mohist canon, the Mengzi and the Daodejing, Wang demonstrates that information in yinyang thinking is understood to capture the yinyang dynamic order. In contrast with practices of divination or magic, such information-​guided thinking rationalises: it attempts to manage uncertainty informationally, by a search for patterns and harmony. An expansive vision of nature being informationally structured as found in ancient China resonates, moreover, with a number of philosophical movements in the Mediterranean in the final centuries BCE. Plato of Athens, for example, turns to information in the individual, or as shared between people, with their focus on the basic act of informing, as Tamsin de Waal shows in “Plato on the act of informing: Meaningful speech and education”. If informing concerns the communication of a message from a sender to a receiver, what sources (if any) can information spring from? What sorts of information are there? Can information be stored, remembered? De Waal demonstrates that precisely Plato’s focus on informing as joint inquiry reverberates throughout his wider views of information. Interestingly, De Waal points out that Plato was particularly critical of one form of information storage: writing. When in his Phaedrus dialogue, the figure of Socrates reports an African philosophical story of the Egyptian god Theuth boasting about their invention of “letters”, a.k.a. writing, their interlocutor Thamus counters that writing merely “produce[s]‌forgetfulness” in people: Their trust in writing, produced by external characters which are no part of themselves, will discourage the use of their own memory within them. You have invented an elixir not of memory, but of reminding. (Phaedrus, 275a) Practices of recording textual or numerical information had at this point, of course, already been around for millennia: from the Ishango bone calendars found in what today is the Democratic Republic of Congo dating back circa 20,000 years, to Sumerian cuneiform tablets (ca. 3200 BCE) to the Inca Quipu system of accounting found in the Andes region of South America around 1500 BCE. Yet during much of history, orality would have been the dominant mode of information transfer. Be that either exclusively oral (as with Native North American and Australian Aboriginal intellectual traditions) or by using oral transmission paired with script (as was the case for early Buddhist, Vedic and Jain works, ancient texts of Judaism and the Catholic tradition). (Moreover, literacy and numeracy long had only limited distribution, remaining the preserve of a ruling elite, often tied to socio-​economic and political power.) Where writing crafts a semi-​persistent record, detachable from its producer, oral information transfer requires sender and receiver to be near another and, if immediate perishing of the message is to be avoided, some form of mnemonic and vocal reproductive capacities in the receiver, facilitating a chain of hearer-​speakers in an oral tradition. 2

3

Introduction

Thamus (as reported by Socrates, as written by Plato) reminds us that one’s choice of information vehicle is not neutral, but can have wide repercussions for its users, both cognitively and socially. (Something to which many living in attention-​deprived, memory-​externalising societies today can attest.) Our very nature as cognitive and sensory information-​processing animals fascinated Aristotle of Stagira, as Miira Tuominen makes clear in “On information in Aristotle: Nature, perception, knowledge”. Tuominen shows that in Aristotle’s philosophy, information is everywhere. In nature, Aristotle sees a form of non-​semantic, teleological information, as they regard any being as structured (that is, as being determined and regulated) by immaterial principles called “forms”, which inform matter. A wren, then, is structured by the form of wren-​ness. Those forms, though, may equally be transmitted, so as to inform knowers, which is where Aristotle’s theory of perception comes in. If I perceive a wren, the form in nature that structures the wren’s matter can come to be transmitted from the wren to myself as semantic (because meaningful) but non-​linguistic information: I grasp the wren sensorily. Tuominen points out that, epistemically, while Aristotle hence finds perception foundational for obtaining information, its role nonetheless remains limited: to get a cognising being anywhere near certainty or universal principles—​in short, near real knowledge—​further reasoning and dialectical assessment would be required. Information, we will learn time and again, does not equate to knowledge. The view of information as operating in and through the natural world would, during these centuries, also begin to characterise nascent medical professions, as becomes clear from Chiara Thumiger’s longue durée study of the presumed psychiatric condition of phrenitis (roughly, “inflammation of the mind”). In “Information and history of psychiatry: The case of the disease phrenitis”, Thumiger shows how medical information in the guise of diagnostic criteria, symptoms, labels, definitions and classifications shape phrenitis as an “object of information” across centuries of psychiatric cultural transmission. The backdrop of Thumiger’s study is formed by broader attempts to organise information, for instance in the gathering of huge library collections (think of the Assyrian clay tablet library of Ashurbanipal in the seventh century BCE, or the library of Alexandria in Egypt circa four centuries later), as well as the development of various medical corpora, such as the body of works associated with physician Hippocrates from the island of Kos in the Aegean, and in traditional Chinese medicine, the Han Dynasty text The Yellow Emperor’s Inner Canon, both dating from around the latest centuries BCE. Collections such as these aided not just the persistence and transfer of (medical) information but also, given their availability, its access.

Access Does how one accesses information matter, epistemically? Hindu philosophical traditions during the first millennium and onward distinguish between texts that were “heard” (śruti in Sanskrit) and those “remembered” (smṛti). Significantly, the body of “heard” texts (originally transmitted orally), which often were oldest and without a known author, would count as the more authoritative, while the authored, written-down ones descended somewhat in the hierarchy of trust and reliability. A key scholar in the Nyāya school of Hindu philosophy who did contribute authored texts was the great ninth-​century philosopher Vācaspati Miśra. In “Vācaspati on aboutness and 3

4

Chris Meyns

decomposition”, Nilanjan Das confronts Vācaspati’s work on information channels, in particular their stance on how cognitive information relates to the worldly objects that it is of. Das shows that Vācaspati argues against “decompositionalism”, that is, against the idea that what my cognitive (and in particular: perceptual) experience is about can “decompose” into several components (such as, say, a picture in my head, and a resemblance between that picture and a worldly object). Instead, Vācaspati holds that information I gain through perception is naturally and automatically (without further conditions) about mind-​independent particulars. The informativeness of perception is not due to further inference, but already lies within a successful act of perceiving itself. That Vācaspati, in addition to producing self-​originated work, also wrote abundant commentaries on key texts of Hindu philosophy, helped solidify this philosophical tradition around the turn of the first millennium. Roughly in this era, another monumental undertaking in the trans-​linguistic and cross-​ cultural safeguarding of information was taking place in and around Baghdad at the height of the ʿAbbāsīd Caliphate (750–​1258). A massive-​scale initiative of translating (mostly scientific) works from languages including Sanskrit, Syriac, and Greek into Arabic helped make these sources accessible to readers of that language. (Translation by itself, too, brings with it a host of philosophical questions: What is it to accurately carry information across to another language? Can translation ever be lossless, exact? What about the ethics of (not) translating, (not) making accessible?) When Anna Ayse Akasoy investigates the circulation of information around the figure of Alexander of Macedon in the stunningly rich narrative records of the Arabian Nights stories, it is against this backdrop of such freshly accessible Greek sources in the Arabic language. More precisely, Akasoy’s “Seeing and recognition in the Arabian Nights and Islamic Alexander legends” zooms in on the reliability of information channels represented in this work of philosophical literature. Is oral information more trustworthy than written words? How about sensory modalities, can I rely on hearing over sight? Akasoy demonstrates that in these stories, the primary and most reliable, trustworthy, mode of informing is deemed to be oral (or verbal), rather than visual. Moreover, the literature foregrounds how information, when transferred, is never simply “absorbed” as is, but that its successful transmission always requires contributions from both sender and recipient in a context of cooperation. (Incidentally, Akasoy also shows how memes—​from the Greek ēmimēma, “that is imitated”—​ or the spread of cultural information through copying was not a 1970s invention, nor a product of GIFs on the Internet, but very much already alive in the figure of “Alexander-​as-​meme” in the Arabian Nights stories.) Not a meme, but nonetheless similarly concerned with sensory-​cognitive information processing, is the work of Abū-​ʿAlī al-​Ḥusayn ibn-​ʿAbdallāh Ibn Sīnā. As Luis Xavier López-​Farjeat demonstrates in “Avicenna on information processing and abstraction”, Ibn Sīnā (or “Avicenna” to his Latin-​speaking friends) was particularly concerned with the acquisition of mental content (ma’nā in Arabic), or the cognitive access of information. In agreement with Aristotle in the latter’s work On the Soul, to which they had access courtesy of the translation-​into-​Arabic enterprise, Ibn Sīnā holds that the human mind can process information from the external world by apprehending forms. The question is: How? When I spot a wren, can I myself obtain its form for cognitive grasping? Or do I need external assistance to achieve such cognitive information? A key notion for Ibn Sīnā here, as López-​Farjeat shows, is that forms might be extracted from perceptible objects by abstraction. Precisely this insight links Ibn Sīnā’s theory of perception 4

5

Introduction

directly to contemporary discussions of perceptual epistemology, about intentional states and the exact informational contents we can gain from the senses, and which require further (cognitive or extra-​cognitive) processing. Compared to Ibn Sīnā, Tommaso d’Aquino (or “Thomas Aquinas”), writing in the thirteenth century, estimates the human capacity to process worldly information rather optimistically. As Cecilia Trifogli shows in “Thomas Aquinas on cognition as information”, this philosopher too had been inspired by Aristotle’s work on information extraction and processing in cognition, understood as the taking in or receiving of the form of a thing. Cognising, on this line, is taking on a form in addition to one’s own. Yet what is it to take on such additional forms? What makes it that the form of the wren I grasp through perception is in fact of that little bird on the branch? Key for d’Aquino here, as Trifogli flags, is the notion of intentional information, or the idea that the form one might grasp is “like”, or “agrees in nature” with the object in the world that it is of. Hence d’Aquino does not take it to be basic or inexplicable that the information gained through perception is about a particular mind-​independent object, but points to conditions that must obtain. (What Vācaspati might call out as “compositionalism”.)

Control Moving beyond 1500, various tendencies that had already been bubbling below the surface began to gain shape more clearly. For instance, where information-​storing manuscripts had enjoyed healthy circulation and copying in previous eras, the development of methods of mechanical reproduction—​initially with woodblock printing (developed in Han Dynasty China before 220 CE, with uptake in subsequent centuries in India, North Africa, the Middle east, and adopted in Europe only from around 1300), next with the more flexible process of printing with movable type, developed around 1040 by Bì Shēng (990–​1051) (echoed by Johannes Gutenberg’s work some centuries later)—​facilitated growing mass production of information-​ holding documents. Moreover, further tools for processing and organising information had taken shape, such as in encyclopedias, bibliographies and indexes to list summarised information. Tenth-​century bibliographer Abū al-​Faraj Muḥammad ibn Isḥāq al-​Nadīm offered a compendium of all current knowledge in their Kitāb al-​Fihrist (Book Catalogue); the Yǒnglè Encyclopedia (1403–​ 1408) commissioned by the Yǒnglè Emperor (Zhu Di) in Ming Dynasty China gathered all available book knowledge into over 20,000 manuscript rolls; and in Conrad Gessner took the inventive step to add a thematic subject index to their “universal” list of all books, Bibliotheca universalis (1545–​1549). In 1751 Denis Diderot and Jean-​Baptiste le Rond d’Alembert began to publish their Encyclopedia, “to collect knowledge disseminated around the globe”. Analogue computers were another appealing tool for information processing. Emerging as early as the first century BCE in the Aegean in the hand-​powered Antikythera mechanism (used to predict astronomical positions), computing devices began to proliferate in the seventeenth century. We have, for example, John Napier of Merchiston’s calculator (called “Napier’s bones”, published 1617) and Blaise Pascal’s arithmetic machine, designed 1642. A figure not to be missed here is Gottfried Wilhelm Leibniz, who—​inspired by hexagrams of the classical I Ching (Yìjīng) text—​not only studied binary arithmetic, which would later form the foundation of electronic computing, but also invented a “stepped reckoner” (1673) for mechanical calculation. In the trans-​historical paper ‘Leibniz as a precursor to Chaitin’s Algorithmic Information Theory’, Richard T. W. Arthur considers Leibniz’s work on information. Arthur shows some 5

6

Chris Meyns

of the surprising ways in which Leibniz anticipated key components of the idea of digital information associated with twentieth-​century Algorithmic Information Theory, as developed by Gregory Chaitin, including the idea of a computer program and an information-​theoretic notion of complexity. Significantly, Arthur also demonstrates that for Leibniz, information cannot ever be understood as disembodied, free-​floating but—​in an anti-​idealist gesture—​demands a physical basis. Information transfer requires energy; information must be materially embodied. Vehicles and embodiment of information are equally at stake in my own chapter, “Information visualisation in the Philosophical Transactions”, which focuses on approaches to providing information in visualised (not primarily text-​or number-​based) form, in one of the first scientific journals in Europe, the Royal Society of London’s Philosophical Transactions, between 1665 and 1886. Distinguishing how visualisations in the broad categories of diagrams, maps and graphs function epistemically, I show that in this platform, visualisation functions not to offer a distinct kind of information that could not have been communicated in other form. Rather, visualised information functions to optimise user performance: cumbersome and costly to produce, but it assists readers in efficient information uptake. Attention to the (social) uptake of information is one way to attempt to regulate or control information flows within societies. Other methods, including censorship or the vetting of information quality before documents get out into the wild, are equally gaining clearer shape in this period. For example, in 1538 Henry VIII requires that any printing of books in England gain pre-​publication approval, while from the mid-​sixteenth century onward the Tribunal of the Holy Office of the Inquisition—an arm of the Catholic Church—publishes successive indices of banned books to halt the spread of information it deemed ‘heretical’. Stina Teilmann-​Lock, in their “ ‘Dwindled into Confusion and Nonsense’: Information in a copyright perspective from the Statute of Anne to Google Books”, investigates the role of copyright in the distribution of information. Teilmann-​Lock argues that while some emphasise that intellectual property rights obstruct access to information, the motivation behind the introduction of these rights was to stimulate the distribution of accurate, authenticated information. Teilmann-​Lock also offers a caution: as the Internet lowers barriers to information sharing without guaranteeing quality or accuracy, we risk collapsing into “confusion and nonsense”. The social life of information is also at the forefront in Lynn McDonald’s chapter on “Information in the pursuit of social reform”. Focusing on British and French-​speaking authors in the seventeenth to early twentieth century, McDonald shows how a number of them strategically employed emerging empiricist scientific methods in gathering real-​world data, and put that information to use—​extracted, analysing and disseminating—​to effect social and political reform. McDonald pays specific attention to the work of two key authors: political scientist Anne Louise Germaine de Staël-​Holstein (1766–​1817), who had advocated precise, “geometric” science in support of political and human rights, and statistician Florence Nightingale (1820–​ 1910). Having received significant spurs from Abu Yūsuf Yaʻqūb ibn ʼIsḥāq al-​Ṣabbāḥ al-​Kindī’s (801–​873) work on frequency analysis, the development of mathematical probability theory in the seventeenth century, and the eighteenth-​century state-​driven collection of demographic and economic data (hence the “stat-​“ prefix), by the nineteenth century, work in statistics and statistical information was taking a new shape. Nightingale, in addition to pioneering new forms of information visualisation through statistical graphing, was foundational in modelling how statistical information could be used to great effect in health care advocacy.

6

7

Introduction

In meta-​level assessment, McDonald also points out the risk of misinformation when towering scientists such as de Staël and Nightingale are ignored (as they have been) in sexist information regimes and efforts of canon formation. A risk which, as will become clear, only looms larger in successive periods.

(Dangerous) Systems The philosophical life of information in the nineteenth century is marked by a pattern of increased systematisation, automation and organisation of information at scale. In few places is such systematisation of information processes more evident than in the information needs and channels used to shape and maintain “empire”, as Edward Beasley shows for the case of Victorian Britain in their chapter “The nineteenth-​century information revolution and world peace”. Early Victorian culture saw two critical (technological) changes which sped up information flows: an efficient postal service, and near-​instant telecommunication. An optimism only paralleled by the early days of the Internet took hold. Rapid information flows hold a promise of improving people’s lives and livelihoods, of letting societies flourish; a process culminating in world peace. Yet, as Beasley shows for Britain, such information orders did not only assist British-​ controlled imperialism, but also (to their alarm) empowered those who contrariwise sought to rise up against the coloniser, as was the case for people in India. A like enthusiasm about the potency of information can be found in the work of philosopher Charles Babbage—​known also for developing a “Difference Engine” (a calculator in the spirit of the Antikythera mechanism and seventeenth-​century calculating machines, started in 1821), and for conceiving a more general-​purpose computer called the “Analytical Engine” (conceived 1834) in collaboration with mathematician and programmer Augusta Ada King, Countess of Lovelace (1815–​1852). In “Charles Babbage’s economy of knowledge”, Renee Prendergast shows how Babbage’s enthusiasm for information machines was sparked at least in part by their potential economic benefits, combined with the view that information and knowledge should bear fruit for transforming the conditions of people’s lives. Prendergast makes clear that—​in a Baconian, empiricist spirit—​Babbage saw good use for the gathering of accurate, experimental data and for inductively building upon such information. Operating with absent or imperfect (or worse, erroneous) information could risk rising costs in production and transactions, Babbage noted. Making calculation mechanical, with the appropriate investments in capital and machinery, could economise on the information and skill needed to perform a task, and so help reduce costs and eliminate error. Yet it could also result in replacing people altogether. Information quality and its operation can have direct, real-​world economic and social consequences. The nineteenth-​century drive to systematise information made itself felt not just socially, but also in the move to organise, name and control nature. With the Swedish physician Carl Linnaeus’ System of Nature (1735), which had introduced its author’s system of biological classification, hovering in the background, a crucial contribution on this front came from the work of Gregor Mendel (1822–​1884). Mendel’s experiments with pea plants in the mid-​nineteenth century subsequently placed them as a founding figure of the discipline of genetics. In “Mendel on developmental information”, Yafeng Shan investigates Mendel’s contributions to ideas about biological information, in particular concerning development. What, exactly, did Mendel understand by “development” in relation to biological organisms,

7

8

Chris Meyns

and how does information figure in his assessments? Particularly and provocatively, Shan shows that the famous Mendelian laws of developmental information are not, in fact, about heredity; that is, about the genetically passing on of physical or mental traits from one generation to another. Heredity was not what Mendel studied or was concerned with. Rather it was the development of hybrids in progeny. Where Mendel did not seek to tread, the Victorian-​era statistician Francis Galton went. Like Nightingale, Galton made significant strides in mathematical statistics in analysis, developing the basis of the contemporary weather chart (in Meteorographica, 1863) and conceiving of the notion of a standard deviation to quantify variation in a normal distribution. Incidentally, as Debbie Challis and Subhadra Das discuss in “Information and eugenics: Francis Galton”, Galton was also an advocate of the racist wing of pseudo-​science known as “eugenics”. These two points are not coincidental. As Das and Challis show, information gathering was key to Galton’s development of eugenicist ideas. During a period that involved significant imperial self-​positioning, Galton jumped on that part of biology that Mendel did not concern themselves with, namely: heredity. Galton refined available methods for gathering and analysing information about people through statistical modelling, and did so at least in part in the process of seeking to mathematise biology and study how “intelligence” might be inherited. From their historical vantage point of describing Galton’s ventures into eugenics, Challis and Das are moreover able to issue another caution. They argue that, instead of assuming that information collection and analysis can and will always be neutral, even mathematical operations can be discriminatory. (A point reinforced in recent publications such as Cathy O’Neil’s Weapons of Math Destruction of 2016, and Safiya Noble’s Algorithms of Oppression from 2018.) More strongly, Das and Challis show that not only does information gathering for eugenic purposes still occur, it is even prevalent today.

Insurgencies In the year 1900, few could have foretold the exact manner in which the mathematised and increasingly automated use of information in science, industry, and societies at large would impact individuals, their communities, and the planet. Though the contours, we should say, were already becoming gradually clearer. One person who helped shape the scientific gathering, use and analysis of information for the social domain—​co-​building the emerging field of sociology—​was William Edward Burghardt Du Bois (1868–​1963) in the United States. In “The racialization of information: W.E.B. Du Bois, early intersectionality, and social information”, Reiland Rabaka pulls into focus two core points. One, the groundbreaking work carried out by Du Bois for sociological information. Du Bois was at the forefront of studying the “uniquely and unequivocally” American issues of race and anti-​Black racism, segregation, colonisation and racial oppression, being one of the first scholars to investigate how these manifested themselves at the intersection with social class, gender and sexuality, religion, education and economics. Methodologically, Du Bois extended existing inductive approaches, including the use of local community surveys and interdisciplinary studies, and enriched sociology with political economy, history, anthropology and “archaeologies” of social phenomena (an innovation later often credited to a prominent theorist in France). Second—​resonating with McDonald’s attention to the exclusion of authors from political science and statistics—​Rabaka also, historiographically, flags the artificial exclusion of Du Bois from the history of social information. Instead of being acknowledged as a groundbreaking 8

9

Introduction

sociologist, Du Bois is all too often ignored altogether, or only gets credit in one or more of sociology’s sub-​disciplines. Such continued exclusion, such “epistemic apartheid” in United States sociology, Rabaka makes clear, raises fundamental questions about what results if information ecosystems get artificially and detrimentally separated. The twentieth century was also the point at which information itself—​its nature and characteristics—​became the subject of sustained philosophical moves and reflections, as stands out in the final two chapters of the volume. In “The many faces of Shannon information”, Olimpia Lombardi and Cristian López investigate contributions to the study of information by mathematician and engineer Claude Shannon (1916–2001), whose work on mathematical communication has reshaped the field of communication technology. Placing itself amid a hoard of different interpretations and connotations of what “information” really is, López and Lombardi focus specifically on how “Shannon information” must be understood. Information could, among other things, be conceived as something epistemic, to do with human (or animal) cognitive abilities; it could be a physical magnitude; or, as within the biological sciences, concern whatever is contained in a DNA sequence. (Moreover, they note that Shannon information has relevant links to the concept of algorithmic information which, via Chaitin, we can trace back to Leibniz.) Perhaps surprisingly, Lombardi and López demonstrate that we would do well to set aside the assumption that information as Shannon saw it is a single thing. Instead, they show that an information pluralism must be correct here. There are, in different fields, various different, equally valid but mutually exclusive, interpretations of this single formal concept: information. In the final chapter, “Computers and system(s) science—​the kingpins of modern technology: Lotfi Zadeh’s glimpses into the future of the information revolution”, Rudolf Seising lays out one of recent history’s visions of information futures as embodied in the work of Lotfi Zadeh (1921–2017). A proponent of system theory—​that is, the study of both natural and artificial (living, intelligent) systems—​Zadeh emphasised the need for such study to be able to mathematically accommodate imprecise, uncertain or even vague information. To this end, Seising shows, one of Zadeh’s key contributions lay in developing a theory of “fuzzy” sets, where membership of elements in a set may be gradual (rather than assessed in binary terms of either belonging or not belonging, as would be the case in classical set theory). Only such more fuzzy mathematics would suit to understand complex systems. In addition to their technical work, already from the 1960s onward Zadeh championed the use of digital computers for use in education, research and libraries. Zadeh’s continued advocacy shows the need for information visions to continually remain alert to how the ever-​ growing role of information all around us can (technologically) be charted. During the march of twentieth and nascent twenty-​ first centuries, a sense grew that humanity was approaching a new era; an “information age”, accompanied by its own “information revolution” (see Figure 0.1).3 Over recent decades, the world’s total capacity to store information increased more than hundred-​fold, from 2.6 optimally compressed exabytes (an exabyte equalling 1018 bytes) in 1986, to 295 optimally compressed exabytes in 2007. The energy used by data centres running 24/​7 to facilitate such lucrative tasks as hosting mailboxes, mining cryptocurrency and keeping everyone’s personal pics stored securely in “The Cloud”, is having a drastic environmental impact.4 Some physicists have even speculated that the fundamental “building blocks” (so to say) of the universe might itself be information? (Something poet-​physicist John Wheeler coined as “It from bit”.) 9

10

Chris Meyns

Figure 0.1  Results for queries “information age” and “information revolution” in the Google Books corpus English (2019), period 1900–​2000, case insensitive, with a smoothing of 0.

Information, or data in its raw form, has (in a not-​so-​environmentally friendly analogy) been called the “new oil”, signalling its status as a commodity around which to spawn profitable capitalist industries.5 Everything—​from location, page clicks, spending patterns, busyness, transport, desires, fears, experiences—​is monitored, tracked and monetised (as Shoshana Zuboff details in The Age of Surveillance Capitalism, 2019). All this happens with a degree of centralisation by a few monopolising players, such that a diagnosis of “information imperialism” would not be amiss. The upside, if there is one, is that in recent years there has been a growing critical awareness of the extent to which we are impacted by this explosion of information generation and use. A growing attentiveness, not just to the risks of misinformation, of epistemic exclusion, of echo chambers, of speech bubbles, of locking scientific information behind paywalls, but also, with Egyptian philosopher-​god Thamus (via Plato), to the extent to which any of these informational operations are not neutral, but can actually impact on our psyches, on who we are, on our societies, lands, and by extension, our earth. We offer this volume to stimulate and solidify that rise in thinking about information both philosophically and historically.

Notes 1 Ngan, Mei, Patrick Grother, and Kayee Hanaoka. 2020. “Ongoing Face Recognition Vendor Test (FRVT) Part 6A: Face Recognition Accuracy with Masks Using Pre-​COVID-​19 Algorithms”. NIST IR 8311. Gaithersburg, MD: National Institute of Standards and Technology. https://​doi.org/​10.6028/​ NIST.IR.8311. 2 Wang, Zhongyuan, Guangcheng Wang, Baojin Huang, Zhangyang Xiong, Qi Hong, Hao Wu, Peng Yi, et al. 2020. “Masked Face Recognition Dataset and Application”. ArXiv:2003.09093 [Cs], March. http://​arxiv.org/​abs/​2003.09093. 3 Michel, Jean-​Baptiste, Yuan Kui Shen, Aviva Presser Aiden, Adrian Veres, Matthew K. Gray, The Google Books Team, Joseph P. Pickett, Dale Hoiberg, Dan Clancy, Peter Norvig, Jon Orwant, Steven Pinker, Martin A. Nowak, and Erez Lieberman Aiden. 2010. “Quantitative Analysis of Culture Using Millions of Digitized Books”. Science. doi.org/​10.1126/​science.1199644. 4 Whitehead, Beth, Deborah Andrews, Amip Shah, and Graeme Maidment. 2014. “Assessing the Environmental Impact of Data Centres Part 1: Background, Energy Use and Metrics”. Building and Environment 82 (December): 151–​9. https://​doi.org/​10.1016/​j.buildenv.2014.08.021. 5 The Economist. 2017. “The World’s Most Valuable Resource Is No Longer Oil, but Data”. 6 May 2017. www.economist.com/​leaders/​2017/​05/​06/​the-​worlds-​most-​valuable-​resource-​is-​no-​longer-​oil-​ but-​data.

10

11

PART I

Information before 500 CE Natures

12

13

1 YINYANG INFORMATION Order, know-​how, and a relation-​based paradigm Robin R. Wang

The notion of “information” in the West became an important philosophical concept only in the twentieth century. According to Pieter Adriaans, the cluster of information contains two central properties: information is extensive and information reduces uncertainty.1 We are compelled to ask whether this philosophy of information theory can be applied to other philosophical traditions, such as classic Chinese philosophy. A meaningful interaction and discourse between different philosophical perspectives on the philosophy of information will enhance our scope of knowledge and our depth of understanding. This chapter will explore the role of yinyang information in Chinese thought; how yinyang information is the order of things and the building block of human knowledge; why yinyang knowing after all is to know how; what a yinyang information model can contribute to the philosophy of information at a broader and deeper level.

A short description of yinyang The terms yin 陰 and yang 陽 are the most commonly known concepts from Chinese philosophy, they have practically become English words themselves. They have been used spanning four centuries, and applied at all levels, from knowing the world to governing the state, to maintaining the health of the body. As Joseph Needham says, yinyang ideas “were the most ultimate principles of which the ancient Chinese could conceive.”2 The usage of yinyang signified a rationalizing tendency in relation to natural processes. It stands for the human intellectual power to explicate natural events, where human knowledge does not depend on divination or magical operations. It marked a transitional time when yinyang began to be used as a conceptual tool to clarify natural phenomena and when it provided a break from treating those events as auspicious signs determining human events. The first written record of using the two characters together is found in the classic text Shijing: “Viewing the scenery at hill, looking for yinyang [the light and the shade].”3 This line indicates that yin and yang in their earliest usage were separate terms that were connected to capture the result of the sun reflecting on the hill. The sunny side of the hill is yang, and the shady side is yin. This original usage already displays some primary aspects of yinyang information, suggesting the intersection of one thing (the sun) with a complex informational context,

13

14

Robin R. Wang

so that yinyang depicts the interplay between the sun, a hill, and the light. It implies a concrete reference and an observable fact. At the same time, yinyang is not a particular object but rather an informational phenomenon: the features of the sun as they relate to other things. Yinyang is an early attempt to reconcile human life with the sun. It gradually developed from ways of naming relationships with the sun into a complex way of thinking, which serves an indispensable role in making an ontological link between a unitary source and the diversity of the myriad things, wanwu 萬物 (literally, ten thousands of things). As Zhuangzi 莊子 (third century bc e) connects yinyang directly to life itself, referring to living things as “all the creatures taking shape between heaven and earth and receiving vital energy from the yin and yang.”4 Yinyang is not only a simple and particular term, but it also represents an underlying structure in an enduring tradition. It functioned as the warp and woof of the fabric of reality. Early Chinese texts tackled structural questions about the universe through yinyang as a way of highlighting the self-​generative and self-​organizing information of complex phenomena. Any given existence is a complex system such that parts are arranged based on a yinyang relationship.5 Rather than taking yinyang as a primitive form of thought, Graham takes it as a correlative thinking, a fundamental element of all reasoning. The focal point is “stimulus and response” (ganying 感應) among things and events. An event or action happening or performed in one domain affects corresponding information in another domain. This cosmology is not based on linear causality between distinct entities, but rather on making a connection between entities and phenomena. For example, yinyang is not just an important tool for grasping the cosmic body, namely the universe; it is also applied to the micro body: “human flesh.”6 The human body bears the same information, rhythm, and properties as the greater cosmic body. Yinyang presents a justification for this association and a conceptual tool for human understanding and embodies a wide range of linked information and meanings, many of which are in play simultaneously.7 There is an ambiguity and complexity in saying that two things “are like yin and yang.” Everything is bound up in a plurality of relationships at the same time, related both to multiple things and to the same thing in multiple ways. These relations are not distinct but reflect the actual complexity of life and nature. Yinyang claims must be taken as a point of reference and their information defined by location (wei 位) and time (shi 時). The yinyang information accentuates these complex, multidimensional frameworks to explore the wide array of practices that constitute human understanding and action. The defense of this pluralistic picture of yinyang information illuminates the diversity and variety within the paradigm itself, a diversity that has enabled yinyang to serve so many different functions throughout various aspects of Chinese culture.

Yinyang information as reflecting of the order of things The origin of the myriad things, the structure of the universe, the source of change, and the beginning of life are all fundamental problems requiring explanation. Yinyang information weaves a coherent understanding that integrates these issues with what is imminent in one’s own situation. Everything and every event can be seen either as yin or as yang, and embodied information related with other things on this basis. If the legacy of Greek logic is the Aristotelian deductive pattern of reasoning and the Euclidean vision of geometry, then the legacy of Chinese thought is this kind of yinyang information. It functions by linking phenomena across various levels and forms. This kind of categorizing through yinyang information is a deep-​seated way of thinking that is found in political systems, ethical orientations, art, and 14

15

Yinyang information

city planning. Yinyang, thus, supplies a rational and coherent basis for all structures and human behaviors. In short, yinyang information is the order of things. One of the most important functions of this yinyang order can be defined as a matrix to describe, guide, and structure concrete phenomena. Historically, in the third and second centuries b ce, yinyang information of spatial orientations and temporal cycles emerged and developed as an attempt to give a rational description of the world that went beyond shamanistic and magical operations. Yinyang information arranges knowledge into a simple, integrated, and flexible pattern, which can be applied to an extremely wide range of phenomena. As the Huangdi Neijing states: As for yinyang, it has a name but no form, thus if you count it, it can be ten; if you separate it, it can be a hundred; if you disperse it, it can be a thousand; and if you extrapolate from it, it can be ten-​thousand.8 且夫陰陽者, 有名而無形, 故數之可十, 離之可百, 散之可千, 推之可萬, 此之謂也 The yinyang matrix is a logical structure and method that classifies all things and reveals yinyang as something like a vital information-​bearing hologram, in the sense that a hologram “is a multidimensional entity where even the smallest part of the entity contains, in condensed form, all of the information necessary for a detailed and complete expression.”9 Many aspects of Chinese thought and culture are encrypted in this yinyang holographic matrix: an unfolding continuum, a net of relationships, and a paradigm. The direct way to approach the yinyang matrix is through the concept of lei 類 (category or kind; or as a verb, to place into categories or kinds). The notion and method of lei is the underlying logical structure of the yinyang information. In contemporary usage, lei literally refers to a group of things with shared features or qualities and, thus, placed under the same category. The He Guanzi (Pheasant Cape Master), a text most likely from the Warring States Period (475–​221 b ce), makes the same point: “There is nothing which does not belong to a lei” 物無非類者.10 All things under heaven are unified or differentiated through lei. We see the important role of lei as logic in the “Xiaoqu” chapter of the Mohist Canon, which says: “use kinds [lei] to accept, use kinds (lei) to propose [以類取, 以類予].”11 The lei method rests on the reasoning that everything can be categorized and classified into a kind, or a group based on its information, function, and movement. The origin of the term lei can be traced back to early ritual sacrifices in the Yin and Zhou dynasties. Lei was the one of five procedures used to select materials for sacrifice in rituals.12 Bilei 比類 (comparing categories) was used to arrange the offerings by dividing them into different groups for different kinds of ceremonies. To see how lei categorizing works, we can take an example: the sun and ginger belong to the same lei of yang, because they both embody properties or information of being hot and warm. Snow and watermelon belong to the lei of yin because they have properties or information of being cool. They are also different lei, however, because sun and snow belong in a group in heaven, and ginger and watermelon belong to the group in earth. This reveals the multiplicity of yin and yang information and relationships. The fundamental importance of the use of lei appears across a variety of texts from the late Warring States and Han. The greatest learning is to learn these lei. The Liji 禮記 (Record of Rituals) says that after seven years of schooling, one only has a small accomplishment. In contrast, “Learning the lei is the highest accomplishment after nine years of learning. It is the highest accomplishment (da cheng 大成).”13 To learn is to learn about lei, as lei is bound up with 15

16

Robin R. Wang

culture (wen) itself. To be cultured is to master the application of lei. The idea of information is inherent in an ability to categorize all things based on similarities, with yinyang as its foothold. Dividing phenomena according to yin and yang order constitutes lei at the broadest level, ultimately explaining all things. We find such claims across a range of texts: “Due to the interaction of yinyang, the lei of things resonate; this is the reason of the ten thousand events” (夫陰陽之感, 物類相應, 萬事盡然).14 This yinyang order is built on structures of relationships and dynamic tendencies rather than on individual characteristics of things. It functions primarily through analogies, information and correlations. All things categorized as yang share common information in dynamic relations to other elements, just as phenomena that are more yin will also share common information.15 The lei possesses an explanatory capacity within which all things are linked in a consistent whole. Although yin and yang are pervasive categories extending to any phenomena, a classificatory system based on only two classes would lack the specificity needed to make much sense of concrete information. Greater specificity comes through the concept of xiang, as in the phrase juxiang guilei 据象歸類 (using images to define categories).16 The term xiang 象 is of special philosophical interest, for it affords us an appreciation of the distinctive features of yinyang information.17 The concept of xiang came from early divination procedures in the Shang Dynasty. People carved lines on turtle bones and then burned them. The bone would crack and form different images, patterns, or figures called xiang. We read from the Zuozhuan (Xigong 15): “Turtles are xiang (images) while yarrow stalks are shu (numbers). When things are generated, they will have xiang; when xiang grow, they will have shu (numbers).”18 Another key aspect of the use of yinyang as lei is that the categorization of phenomena allows the derivation of information and inferences: bilei xiangtui 比類相推, which translates literally as “comparing kinds to infer to each other.” Bi refers to putting together things that are similar and analogous. By examining things of the same kind or having the same function, one can derive certain properties and information, and by placing new information into lei, one knows how to approach them. All things in the world can be organized into yang lei or yin lei, and from these, inferences or knowledge can be derived.19 The Yijing is the best example for this reasoning mode. It suggests that lei thinking takes a small thing (a concrete thing) and extends its application to a big thing (general thing) through a process that is called qu lei 取類 (getting the lei). One takes natural phenomena or an arrangement of hexagrams and then infers the characteristics of other things or events, “taking the image to elucidate the meaning, the small can refer to the big.”20 There are intrinsic informational links among all things; yinyang is a rhythmic pathway through these links. Once an entity or event is defined in yinyang terms, its order of managing things can be grown from it. Thus yinyang becomes the building block of our knowledge. Beyond classifying and grounding inferences, the important element of the yinyang lei is in explaining influence and causality. We have seen that categories primarily name dynamic situations, tendencies, orders, and patterns rather than static things. They are, thus, oriented toward change and specifically toward how to act effectively in relation to the world. The patterned information exemplifies a deeply rooted orientation toward growth and transformation. The fact that the tendencies and movements of things are located in complex systems of information suggests that things of the same kind can influence each other. This influence proceeds because there are mutual intrinsic informational links between things; like entities are mutually interconnected. The unity of heaven, earth, and human beings lies in their sharing structures, information, orders, patterns, and movements. Moreover, heavenly movements have implications for the human world. As Liezi puts it, “One body’s being full and empty, 16

17

Yinyang information

rising and falling, all of these interact with heaven and earth and respond to the lei of things” (體之盈虛消息, 皆通于天地, 應于物類).21 Let’s take an example of an early Chinese calendar, known as the “joined yinyang calendar” (yinyang heli 陰陽合歷). It was based on the movement of both the sun and the moon. The moon’s cyclical movement from full to crescent is called the yin calendar; the movement of the sun, which sets days and seasonal changes, is called the yang calendar. The sun and moon, joined together, provide an informational guide for farming. The classical Chinese “day” (ri 日) is the same word as “sun,” just as “month” (yue 月) is the same word as “moon.” The moon has an indispensable role to play, because individual days (ri, representing the sun) are situated within months (yue, representing the moon). When you decide what you want to do this particular day, you must also consult with the moon, the background information. This understanding is what allows one to act effectively. The month that a day falls within is much more important in the traditional calendar than the year. The sun and moon are the most vital information for farmers; however, they cannot be viewed as isolated bubbles of elements, but rather as related to each other to set the particular context for any decision. Situating oneself in any configuration of forces requires attention to both yin and yang information and order, although the yin commonly is the context and background outside of one’s main focus. Yinyang information is the hub for human understanding of all things. It contains at least three variables: (1) the rhythm of yinyang (jiezou 節奏): either yang or yin is too fast or too slow; (2) the balance of yinyang (pingheng 平衡): too much or too little yang or yin; and (3) the transformation of yinyang (bianhua 變化): yang or yin changing too much or too little. Yinyang information penetrates all aspects of human life and crafts. Let’s take music, including the selection of materials and design of the music instruments themselves, as an example. The guqin, 古琴, a seven-​stringed zither-​like musical instrument, is among the oldest of musical instruments, according to legend originating at the time of the sage kings Yao and Shun. The design of the guqin embodies yinyang information and order. It is made of yang-​type wood and yin-​type wood. Because of the seasons and other natural influences, some parts of a tree will be hard and strong, and some parts of tree will be soft and weak. Therefore, there is an art to picking the parts of the tree that make an instrument. One should select the yin portions that are soft and yielding to build the yin sections of the guqin, and one should select the hard and strong portions to construct the yang sections. The surface board is round and represents heaven, or yang. The bottom board is flat to represent the earth, or yin. The entire length of the guqin (in Chinese measurement) is 3 chi, 6 cun, and 5 fen, representing the 365 days of the year. Each part of the guqin contains information, some more obvious, like the “dragon pool” as the yang pool and the “phoenix pond” as yin pool. This guqin forms a circulation of qi movement such that, when played, the guqin transmits sound from yin to yang: from the supporter (yin), then to the peg pool (yang), and finally to the forehead (yin). This path of sound migration in the guqin reflects the fact that yinyang information is the pathway and order to all movements (see Figure 1.1). Cai Yong 蔡邕 (132~192), Eastern Han artist, used yinyang information to formulate nine positions (Jiushi 九勢) in calligraphy. These positions derive from the movement in calligraphy of up and down, left and right, in and out, vertical and horizontal, head and tail, and beginning and ending. Cai Yong writes: “My calligraphy is rooted in nature. Yinyang is generated in nature. The positions come from the generation of yinyang.”22 Yinyang is a contingent framework in the sense that it always exists within a background that contains a wide range of presuppositions and information. There is a yin mode, perhaps seen as a pre-​reflective mode, before awareness has occurred. There is a yang mode, a reflective 17

18

Robin R. Wang

Figure 1.1  A guqin made of wood, silk, jade, lacquer, and mother-​of-​pearl, dated 1634.

mode, in which one must observe one’s self and attend to the inner processes, becoming aware of what is taking place. Yinyang information presupposes knowledge of yinyang patterns. If we see a yang, we cannot be ignorant of yin-​ish things, because they have always been there, in time and space. It is not so much that we build knowledge based on what we objectively perceive, but that we perceive according to what we know. All representations, whether in language, thought, or experience, only succeed in representing the order of yinyang information. This understanding brings out an epistemic assumption underlying yinyang information: any given point of knowing is only a small knot in a giant and consistent web. Any knowing contains infinite unknowing because the known reveals only a part of the unknown. On the one hand, we might take this as leading to what Graham calls a “great man metaphysic”: “The Great Man, by identifying himself with whole, widens his perspective to a full view of everything, with the result that he sees finite things in proportion, as only relatively great or small, good or bad.”23 On the other hand, however, this yinyang information conceives of our surroundings as the immediately accessible portions of a to-​be-​known world, which extends in space and time beyond the limits of our immediate access. This directly accessible portion is like the stem of a plant: as the Chinese expression says, “following the runner is the best way to find a hidden melon” (Shunteng mogua 順藤摸瓜). Ordinary experience yields patterns, which lend themselves to immediate and spontaneous assimilation. The yinyang knower locates herself within this fixed and largely predictable perceptual world. This is not a mystical awareness of the whole, but rather a concrete way of interacting with particulars by tracing out their information and order in those broader networks.

Yinyang information as know-​how Everything we know about the world is based on yinyang information we received or gathered, therefore every human action should be aligned with this knowing. For example, yinyang was inspired by human experience of the sun, which provides a clock for daily routines. Farmers depend on it for light, which in turn dictates the daily rhythm of human life. Overall yinyang information is manifest in these aspects: heavenly images (tianxiang 天象), weather and climate (qihou 氣候), the developing stages of things (wuhou 物候), and human affairs (renshi人事). The images of heaven (tianxiang) provide the timing for all kinds of actions, including the ruler’s policies for scheduling punishments and rewards. The condition of the weather (qihou) guides farmers to plant or harvest. The stage of things (wuhou) indicates the proper timing for ordinary people. All of this information makes up the integral parts of human affairs (renshi).24 18

19

Yinyang information

This whole structure of information was conceived through the flowing of yinyang order, thus making yinyang the basis for human activities. To master this yinyang information in action is seen as a shu 術, a strategy or technique that enables one to function effectively in any given circumstance. The word shu in oracle bones refers originally to a road, thus connected to the way, dao. The “Zun Deyi” (“Respecting Virtue and Rightness”) text recently excavated at Guodian says: Yu the Great’s moving the waters was by following the Dao of water. Zao Fu’s riding horses was by following the Dao of horses. Hou Ji’s planting the earth was by following the Dao of the earth. There is nothing that does not have its Dao, but the Dao of human beings is nearest. Thus, gentlemen first select the Dao of human beings.25 禹之行水, 水之道也. 造父之禦馬, 馬也之道也. 後稷之藝地, 地之道也. 莫不7有道焉, 人道為近. 是以君子人道之取先 These three popular heroes exemplify advantageous human action, which comes from following along with the information of things, whether of water, horses, or human beings. Water is a dominant metaphor relating to terrain. Yu the Great (大禹 Dayu) was the legendary founder of the Xia Dynasty (2070–​1600 bc e). He mastered flood control techniques to tame rivers and lakes. In Mengzi, there is an account of how bad these floods were. Originally, Yu’s father Gun (鯀) was assigned by King Yao (堯) to tame the raging flood waters. Over nine years, Gun built strong dikes all over the land in the hope of containing the waters. During a period of heavy flooding, however, all of these dikes collapsed and the project failed miserably. Gun was executed by King Shun (舜). Yu learned from his father’s mistakes and took a different approach to manage the water. Instead of using force to combat the floods, such as dikes to stop up the water, he used the way of shudao 疏導 (redirecting) and shunni 順逆 (following along or going against). He dredged new river channels to direct the flow of the water, going with rather than resisting the tendencies of the water. These channels served both as outlets for the torrential waters and as irrigation conduits to distant farmlands. He, thus, successfully controlled the floods. His method serves as the metaphor of flowing along in attunement with the order of a terrain to get things done with excellence, ease, and sustainability. This narrative shows that through the embodiment of yinyang information, one will gain a natural power from heaven and earth, be granted the ability to manage one’s internal and external world, and finally, attain a kind of magical charisma. Human understanding is a matter of integrating the information that confronts us into our broader network of life. More importantly, all knowing is situated within knowing how. The Huainanzi takes the core of shu as “Looking at the root and knowing the branches, observing the finger and seeing the return [path], holding to the One and responding to the many, grasping the essentials and ordering the details. These are called ‘techniques’ ” [shu].26 This shu is best manifested in the metaphor of charioteering in the Huainanzi: Therefore, the Great Man calmly has no worries and placidly has no anxieties. He takes Heaven as his canopy; Earth as his carriage; the four seasons as his steeds, and yin and yang as his charioteers. He rides the clouds and soars through the sky to become a companion of the power that fashions and transforms us. Letting his imagination soar and relaxing his grip, he gallops through the vast vault [of heavens] … Thus, with Heaven as your canopy, nothing will be uncovered; With Earth as your carriage, 19

20

Robin R. Wang

nothing will be unsupported; With the four seasons as your steeds, nothing will be unemployed. With yin and yang as your charioteers, nothing will be incomplete.27 The four seasons are the horses, and yinyang is the driver. In this way, the Great Man can ride through the clouds and beyond the sky. He will go with transformation and change, doing as he pleases and unfolding with rhythm. He gallops across an infinitely vast land. The horse was a very significant image in many early Chinese texts. Managing horses effectively was a necessary condition for success, and horse-​driven chariots were crucial for early military battles. Of course, this can be extended to navigating any path, from one’s personal life to political organizations. Horse riding is the one of the six classical arts.28 It is about training someone to become a superb horse rider by cultivating a yinyang know-​how by which one can easily and artfully locate oneself in relation to one’s milieu. In the Hanfeizi’s commentary on the Daodejing, there is a classic story about King Zhao of Jin learning the art, shu, of charioteering. After learning the art, King Zhao was eager to defeat his master. He requested three races with three different horses, however, he lost all three times. He was angry at the master and thought that he had not taught him a complete skill. The master told him: I have given you all the techniques you need to ride a horse. However, there is a deficiency in your usage of these skills. The most important thing for the art of charioteering is to have the horse peacefully reside with the chariot and to have the rider’s heart/​mind come together with horse. However, you only care about who is in front and who is behind. If you were behind, you worried about catching me; but if you were ahead, you worried about being caught by me. There is always a rider either ahead or behind. If all of your attention focuses on me, how can you come together with the horse? This is the reason why you lost the race.29 Clearly, a good horse rider must be able to peacefully work with a controlled flow that responds to unrestrained forces and variations, and not focus on one single, specific external fact such as who will win the race. Skill at charioteering is not a case of courage (yong 勇), but rather a demonstration of a kind of intelligence (zhi 智), a know-​how for becoming an embodied navigator. Based on the Huainanzi’s metaphor, this technique, or shu, can be analyzed from two distinct points of view. First, yinyang know-​how is rooted in a view of the universe as an organic self-​generating system. Self-​organization and self-​stabilization presuppose interaction between system and environment. In the case of the shu of horse riding, effective interaction occurs through movement. The immediate interfaces of navigating a horse-​drawn carriage include the horses and their power, the terrain, the weather, and one’s purpose. The horse rider is linked to the many external factors that may disturb his or her inner state and draw out different kinds of responses. It is a kind of open system that deals with environmental disturbances and processes within it. The Huainanzi depicts that you feel with your hand, however, you respond through your heart/​mind. This is a common saying: dexin yingshou 得心應手, “getting it through your heart/​ mind and responding with your hands.” In this aspect, the world to a rider is not observed but felt. The whole nexus of senses (including vision) is a felt response. Like the sting of the sunlight or rush of the wind, the act of seeing has a similar feeling. It involves the mind and body working together. The Liezi presents another description of the art of charioteering:

20

21

Yinyang information

Internally, one focuses the center of the heart/​mind; externally, one unites with the will of the horse. One is able to go forward and backward but there is center and one goes around it as if with a compass. One can take the road on a long journey yet still have strength to spare.30 In riding horses, one can distinguish the internal (focusing the heart) and the external (the horse itself), however, the crucial point is being able to reach the center. Liezi clarifies further that one receives (de 得) the bit and responds with the bridle; one receives in the bridle and responds with the hand; one receives in the hand and responds with the heart/​mind. This way one sees without eyes and urges without a goad; relaxed in the heart/​mind and straight in posture, holding six bridles and pacing 24 hooves to advance, withdraw, and swing around with perfect precision. The heart/​mind plays a key role in adjusting the situation. One’s heart/​ mind is synchronized and functions with the natural flow of the horse and chariot. Here, the rider is in a condition lacking in deliberateness or discursive thought, but has a great awareness that allows overall optimal performance. This yinyang know-​how is an attunement and embodiment. The second aspect of yinyang know-​how is found in adaptivity. Horse riding works with external forces and internal constraints that lead to adaptive self-​organization. In this facet, adaptation is not synonymous with stability or harmony (he) but is closer to functional efficiency in coping with actual environmental disturbances. It is more about efficacy than about harmony. For example, what if the horse goes slowly, while the rider needs to travel faster? The rider needs to initiate a way to make the horse go as fast as it can. The rider must engage in yielding and pulling movements, a dynamic yinyang play: giving and taking, pushing and pulling with the powers of the horse. The Liezi explains: “equalizing the give and the pull is the ultimate principle of dealing with the world.”31 What is the “equalizing” (jun 均)? It is the center of the wheel that can turn to face any direction. The rider can only reach his or her goal by working with it, negotiating all variables to attain the desired result. The rider incorporates his surroundings into his perception-​response loops to maintain an efficacy. Horse riding also requires human adaptation, affecting the reorganization of inherited behavior patterns to fit the existing environmental situations. This yinyang know-​how as an adaptation is an indispensable instrument for the interaction between oneself and the world. It is a configuration of forces: fundamental forces that can be exerted only by certain types of configurations and the information that emerge from such configurations. Only sages know how to use or obtain dao. Dao is not just getting things, but also knowing how to use power, like navigating a boat in moving water, or using the wind to sail at sea.

Relations-​based yinyang information model Yinyang information invites us to see world in a different light and lifts our vision to a broader horizon, which implies a paradigm for thinking about change and effective action. The ultimate aspiration is to search for order, pattern, and harmony in the universe and direct human action within. This yinyang information model is rooted in various relationships that can simplify, exemplify, and expand as the underpinnings for all entities. The constructed argument here is that the yinyang information model fashions a configuration of relations in reality. This effort is to illuminate that yinyang is a multilevel structure that cannot be reduced to a single string of complementarity or balance. Yinyang is constituted by an all-​encompassing web of various relationships.

21

22

Robin R. Wang

A range of classical texts demonstrates that the yinyang information model is a pervasive spectrum of differences within relatedness, connection, and mutual influence. From the yinyang epistemic perspective, this yinyang spectrum is situated in the rhythm of interactions and mutual integration. There are six aspects of any given yinyang relation-​based information model. (1) Maodun 矛盾: Contradiction and opposition. Although yinyang thinking may prompt us to think of harmony, interconnection, and wholeness, the basis of any yinyang distinction is difference, opposition, and contradiction. Any given two sides are connected and related, but they are also opposed in some way, like light and dark, male and female, forceful and yielding. It is the tension and difference between the two sides that allows for the dynamic energy that comes through their interactions. This aspect of yinyang is often described in terms of maodun, which literally means “shield-​spear” and originates from a story in the Hanfeizi 韓非子 (280–​233 bc e). A person who sells shields and spears promotes his shields by saying they are so strong that nothing can penetrate them, whereas he promotes his spears by saying they are so sharp they can penetrate anything. Someone then asks him: what happens if one tries to use your spear to penetrate your shield (Gu 1996, 204)? (2) Xiangyi 相依: Interdependence. One side of the opposition of yinyang cannot exist without the other. This interdependence can be seen on several different levels. On one level, it points out the interdependence of opposites as relative concepts. In labeling something as “high,” one must implicitly label something else as “low.” One cannot have a concept of “good” without there existing a concept of “bad” (Daodejing, ­chapter 2). According to yinyang thinking, however, the interdependence of opposites does not simply refer to the relativity of our concepts, but also to how things themselves exist, grow, and function. One way that this interdependence appears most clearly is through the alternation of yin and yang. The sun is the best example of yang—​bright, warming, stimulating growth, and giving a rhythm—​but when the power of that yang is developed to the extreme, it is necessary for it to be anchored, regenerated, and sustained by the force of yin. The sun must set. Although yang is obvious, it cannot thrive without attention to yin. This interdependence appears in traditional Chinese medical texts, where the surge of yangqi 陽氣 depends on the regeneration of yinqi 陰氣, of the five internal organs. Without that basis, the yinqi of the organs, there will be no a surge of yangqi or its extension outward. (3) Huhan 互含: Mutual inclusion. Interdependence is linked closely to mutual inclusion. If yin depends on yang, then yang is always implicated in yin; in other words, yin cannot be adequately characterized without also taking account of yang. The same is true of yang—​ it necessarily involves yin. Regarding things themselves, even something that is strongly yang can be considered yin in some relations, as we have seen. One cannot understand or characterize one without the other (i.e. about knowledge and explanation). The constant alternation between yin and yang also entails that yang always holds some yin and yin holds some yang. In the cycle of four seasons, summer is the most yang of the seasons, yet it contains a yin force, which will begin to emerge in the summer, extend through the fall, and reach its culmination in the winter. Winter is the highest stage of yin, yet it unfolds a yang force that will attain its own full swing from spring through to summer. This mutual inclusion is best captured in the famous yinyang symbol which includes a small circle of yang within the fullest yin and a small circle of yin within the fullest yang. (4) Jiaogan 交感: Interaction or resonance. Each element influences and shapes the other. If yin and yang are interdependent and mutually inclusive, then a change in one will necessarily produce a change in the other. Thus, as yang ebbs in the autumn, yin strengthens,

22

23

Yinyang information

and as yin declines in the spring, yang grows. For example, in Chinese traditional medical diagnoses, too much yin in the body is a sickness of yang, and too much yang in the body is a sickness of yin. Changes in yin will affect yang, and vice versa. This mutual resonance is crucial to yinyang as a strategy because it entails that one can influence any element by addressing its opposite, which in practice most often takes the form of responding to yang through yin.32 (5) Hubu 互補: Complementary or mutual support. Each side of yinyang supplies what the other lacks. Given that yin and yang are different but interdependent, properly dealing with a situation often requires supplementing the one with the other, which is a way of achieving the appropriate balance between the two. For example, the Zhouli 周禮 (The Rites of Zhou, 5–​221 bc e) describes the craft of making a wheel: The way of making the hub of wheel must be measured according to yinyang. Yang is densely grained and thus is strong; yin is loosely grained and thus is soft. Therefore, one uses fire to nourish its yin, making it even with its yang. Thus, even if the wheel is worn, it will not lose its round form. (Wang, 1972, 424) This passage addresses the difficulty of creating a wheel that is firm but made of materials that are soft enough to bend into a circle. Here, softness and hardness complement and support each other. This complementarity is different from the submission of one to the other, because both sides stand on equal ground in performing different roles. Such rational control and receptivity should be in a pair of mutual support. (6) Zhuanhua 轉化: Change and transformation. One side of yinyang becomes the other in an endless cycle. Yinyang thinking is fundamentally dynamic and centers on change. In nature, there is decline, deficiency, decrease, and demise, as well as flourishing, surplus, increase, and reproduction. In the human world, life is filled with trouble, failure, exhaustion, and insufficiency, as well as fullness, fruition, mastery, and success. Considering these various states of being, one can derive that change is perpetual, never-​ending. The above six yinyang relationships reveal several propositions: (1) levels of relationship defined through degrees of integration; (2) emergent order as opposed to a predetermined order; (3) constant change; and (4) a not fully predictable future. These propositions can be captured in the model of a flowing circular or spiral movement (huanliu 環流) as a metaphor for the processes of generation, integration, and emergence. The term for circular or spiral (環 huan) refers to the movement of the flow. The character liu 流 refers most literally to the flowing of water, and the character itself contains the image of water on the left. We might thus also translate the phrase as “flowing circulation.” This flowing circularity has three properties. The first is that it revolves around an empty center. The Daodejing discusses images of the hub, the cart, and doors and windows, all of which identify an empty space in its enclosing frame. The thing depends on this empty space for its form and its “usefulness.” This “structural blueprint” hinges on the concept of an enclosure that surrounds the emptiness. Thus, concrete things themselves always exist through an element of emptiness that is called fuyin 負陰 (embodying yin). However, non-​presence is embedded not only in presence but also in motion. This will lead us towards the concept of the rotation of the empty center: the form revolves both

23

24

Robin R. Wang

metaphorically and literally around the empty space that rotates. The wheel spoke rotates around an “empty” hub; the pot spun on a potter’s wheel and the walls of the room “rotate” around the empty space and toward a spiral movement. We can also glimpse this spiral progression in natural phenomena like the whirlpool, vortex, sinkhole, and coil. The second characteristic of the flowing circularity is that the cycles have no end and no beginning, continuing on without limit or exhaustion. Change is perpetual, never-​ending. One side becomes the other again and again. This view appears across a wide range of texts. The Huangdi Neijing makes the same point more explicitly in terms of yinyang: Yin and yang are mutually connected, like a cycle without beginning. Thus, one knows that attack and defense always follow each other. Another passage says: “Yinyang are interlocking like a cycle without limit, yinyang follow each other and internal and external interlock each other like a cycle without limit.”33 The third aspect of the flowing circularity is reversal (fan 反),34 which is a constant theme in the Daodejing. Reversal invokes the image of a circle in motion, or more precisely, a nonlinear and infinite-​dimensional spiral movement. Reversal is a constant theme in Chinese thought, especially in the Daodejing. It invokes the image of a circle, or more precisely, a spiral movement that forever continues in a ring formation. The He Guanzi (Pheasant Cape Master) gives an influential characterization of this movement: “Beautiful and ugly adorn each other: this is called returning to the full cycle.” This is the case of wuji zhefa (物極則反), “a thing will reverse after developing to its extremes,” a popular idiom in contemporary China. It is somewhat similar to the term in contemporary Western science called “self-​organized criticality,” which refers to the tendency of large dissipative systems to drive themselves to a critical state, with a wide range of lengths and timescales. The idea provides a unifying concept for large-​scale behavior in systems with many degrees of freedom. It has been looked for in such diverse areas as earthquake structures, economics, and biological evolution. It is also seen as “regression toward means.”35 This flowing circularity intends to convey the idea that the occurrence of events is not linear, the circular flowing moves forward to a critical point of transition and then reverses away from it. Things reposition from improbable order to inevitable disorder. Destruction and loss turn out to be the very rhythm by which nature sustains its beauty and regeneration. The ramifications of this view disclose a paradoxical feature of reality. The yinyang relational information model originates in the desire of managing and dealing with uncertainty, which is one of the primary concerns in ancient China. It holds that unpredictability and change are unavoidable and actually dominate all things under heaven. Uncertainty is not a “thing” or “entity” that is created by human failure, but rather a continuous activity that is constituted and shaped by change and emerges from change. Human knowledge about the future understood as a degree of relationships is measured in terms of probability, which in its turn is explained in terms of the number of yinyang configurations in the grand scheme of things. It allows for differences, emergence, and transformation. The yinyang relational-​informational model seeks out meaning in experience that is not immediate, but begins in the formless and meaningless, as one aspect of the intricate web of life. Practically speaking, this indicates that when one faces a task, one needs to make an intentional shift from the immediate to the wider spectrum, paying attention to the background and implicit patterns. This model epitomizes spontaneous cognition of environmental relations. The cognitive power resides in its capacity to elucidate the yinyang relationships in a complex network and environment.

24

25

Yinyang information

Notes 1 Adriaans, Pieter, “Information,” The Stanford Encyclopedia of Philosophy (Winter 2018 Edition), Edward N. Zalta (ed.), https://​plato.stanford.edu/​archives/​win2018/​entries/​information/​. 2 J. Needham, Science and Civilization in China, 7 vols. (Cambridge: Cambridge University Press, 1956), vol. II, p. 232. 3 Z. Zhou 周振甫, Commentary on Shijing 詩經 (The Book of Odes) (Beijing: Chinese Press, 2002), p. 437. James Legge translated the word yinyang as “the light and the shade.” See J. Legge, The Chinese Classics, 5 vols. (Taipei: SMC Publishing Inc., 1994), vol. IV, p. 488. 4 B. Ziporyan (trans.), Zhuangzi: The Essential Writings with Selections from Traditional Commentaries (Indianapolis/​Cambridge: Hackett Publishing Company Inc., 2009), p. 69. 5 Robin R. Wang, Yinyang: The Way of Heaven and Earth in Chinese Thought and Culture (Cambridge: Cambridge University Press, 2012). 6 A. C. Graham, Yinyang and the Nature of Correlative Thinking (Singapore: The Institute of East Asian Philosophies, 1986), p. 1. 7 Robin R. Wang: Yinyang: The Way of Heaven and Earth in Chinese Thought and Culture (Cambridge: Cambridge University Press, 2012). 8 Niu Bingzhan 牛兵占, (ed.), Huangdi Neijing, 黃帝內經 (Shijiazhuang: Hebei Science and Technology Press 1993), p. 105. 9 D. Parrish M. D., Nothing I See Means Anything: Quantum Questions, Quantum Answers (Boulder: Sentient Publications, 2006), p. 15. 10 Wang Xinzhan 王心湛 (ed.). Collection and Interpretation of He Guanzi 鹖冠子集解 (Shanghai: Guangyi Press, 1936), p. 26. 11 I here follow the translation in Dan Robins, “The Later Mohists and Logic,” in History and Philosophy of Logic, Volume 31, Issue 3, 2010, pp. 247‒85. 12 Wang Mengou, 王夢鷗 (ed.) Liji, Record of Rituals, 禮記今註今譯 (Taipei: Taiwan Commercial Press, 台灣商務印書館, 1981), p. 289. 13 Wang (ed.), Liji, p. 595. 14 Ban Gu. 班固,(Han Dyansty) Hanshu 漢書 (The Book of Han), Commentary by Yan Shigu 颜师古 (Tang Dynasty), (Beijing: Chinese Press, 1962), C ­ hapter 24, p. 1140. 15 Here, we should recall the point that yin and yang themselves are relational terms, so that the same thing could simultaneously be yang in one relationship and yin in another. 16 Although xiang can be taken as a form of lei, they were sometimes distinguished. For example, the “Tiangua Shu” chapter of the Shiji associates xiang with heaven, and indeed one of the paradigms for xiang is in connection with the changes of constellations, calendars, and cyclic movement of seasons and planets. It then associates lei with earth (di), taking yinyang as the main form of lei. Si, Maqian 司馬遷, Shiji 史記 (Records of the Historian) (Beijing: Chinese Press, 2003), ­Chapter 27, p. 1342. 17 Hu Shi has given the explanation for the etymological origin of the word. See S. Liu, “The Use of Analogy and Symbolism in Traditional Chinese Philosophy,” Journal of Chinese Philosophy, 1 (1974), 316. 18 Li Mengsheng, 李夢生, (ed.) Zuozhuan Yizhu 左傳譯注 (Shanghai: Shanghai Guji Press 上海古籍出版社, 1998), p. 238. 19 One example would be the interpretation of dreams, as Zhouli and Liezi interpret six different kinds of dreams. There are yang dreams and yin dreams according to the intensity and the function of yinyang. B. Yang (ed.), Liezi Jishi, pp. 102–​3. 20 Gao, Commentary on Zhouyi, p. 417. 21 Yang Bojun, 杨伯峻(ed.) Liezi Jishi, 列子集釋, Commentaries on Master Lie (Beijing: Chinese Press, 1979), p. 102. 22 Ibid. 23 Graham (trans.), The Book of Lieh-​tzu, p. 144. 24 G. Ban (ed.), Hanshu 漢書 (The Book of Han) (Beijing: Chinese Press), p. 1079. 25 Liu, Zhao (ed.), Guodian Chujian Jiaoyi 郭店楚簡校釋 (Fuzhou: Fujian People’s Press, 2003), p. 122. 26 Roth, The Huainanzi, p. 720. 27 Roth, The Huainanzi, p. 52. 28 Information on six arts.

25

26

Robin R. Wang 29 Gu (ed.), Hanfeizi, p. 93. 30 Graham (trans.), The Book of Lieh-​tzu, p. 183. 31 Ibid., p. 104. 32 For more discussion on Chinese medicine see Ted Kaptchuck, The Web That Has No Weaver: Understanding Chinese Medicine (Chicago: Contemporary Books, 2000). 33 Niu Bingzai 牛兵占 1993, 59. 34 For more specific discussion on the term of fan, See Wang Zhongjiang “Abnormalities and Return: An Exploration of the Concept Fan 反 in the Laozi,” Religions 2019, 10, 32; doi:10.3390/​rel10010032. 35 For more information see, Kauffman, Stuart A., The Origins of Order: Self-​Organization and Selection in Evolution (New York: Oxford University Press, 1993).

26

27

2 PLATO ON THE ACT OF INFORMING Meaningful speech and education* Tamsin de Waal

Introduction Socrates, or, at any rate, Plato’s character, Socrates, would insist that we begin a discussion of information by trying to define it. The concept of information, however, is a very slippery one. As Floridi observes:1 Information is notorious for coming in many forms and having many meanings. It can be associated with several explanations, depending on the perspective adopted, and the requirements and desiderata one has in mind. Indeed, some would argue that there isn’t such a thing as “the concept of information” at all, and Floridi himself speaks of “the family of concepts of information”.2 For the purposes of this chapter, I proceed from a very general characterization of information, one which I take it broadly underpins both its colloquial and technical applications. On this characterization, information is: a message (or what is contained in a message) communicated from a sender to a receiver, which (ideally) informs the receiver (i.e. is informative). My aim here is to demonstrate that Plato was consciously and explicitly concerned with questions and issues relating to information, so understood. Questions as to the nature of information—​what counts as informative, what constitutes a unit of information, how information relates to truth and knowledge, whether there are different sorts and sources of information. And then questions relating to the access and storage of information, and transmission and understanding of information. These are questions that remain central to philosophical thinking about information today, so, to this extent, my aim here is to show that Plato shares contemporary concerns about information, and, to a degree, paves the way for contemporary thought on information. At the same time, however, I aim to bring out what is distinct about Plato’s conception of information, and how it diverges from the characterization I set out with. The approach I take is to focus on what Plato has to say about the act of informing where I take this to encompass, first, the act of speaking meaningfully, and, second, more broadly construed, the act of educating. Plato has a great deal to say about both, and, as we shall see, the 27

28

Tamsin de Waal

two are inextricably related. I set out here to highlight key features of his characterization of each, in order to highlight his concern with and conception of information. I start with Plato on meaningful speech, and then focus on education in Plato.

Meaningful speech A lógos as a unit of information In one of Plato’s later dialogues, the Sophist, Plato offers an analysis of lógos (261d ff.). The Greek word lógos, a noun derived from the verb légein (to say, speak, tell), has a wide range of meanings (including account, argument, reasoning or reason, reckoning, principle).3 Lógos in this context has the sense of “sentence” or “proposition”, and is generally rendered “sentence” or “statement”. The Sophist tells us that some words “fit together” and some don’t. That is to say, some words, when combined, signify something; other words, when combined, mean nothing (261d–​e). Words that fit together make a statement (lógos), where this can be true or false. The most basic (“simplest and briefest”, 262c) statement consists of a name or noun (ónoma) (i.e. an expression for things that perform actions), plus a verb (rhêma) (i.e. an expression for actions). For example: “Man learns” (262c). A statement “gives an indication about what is, or comes to be, or has come to be, or is going to be … [it] accomplishes something, by weaving together verbs with names” (262d). By contrast, a string of names, such as “lion, stag, horse”, or a string of verbs, such as “walks, runs, sleeps”, doesn’t “indicate either an action or an inaction or the nature of something that exists or of something that does not exist” (262b–​c). So, to say “Man learns”, or “Theaetetus sits”, is to state something, as opposed to merely naming something (262d). The name fixes a reference, and then the verb(s) describes or tells us something of or about the referent. A statement is thus about something, and indicates something, and so is informative, in a way that merely naming things isn’t. So on Plato’s account, a statement, where this minimally consists of a name and a verb, is, by its very nature, meaningful. To make a statement is to speak meaningfully. And since a statement can be either true or false, it is possible to speak falsely and meaningfully, as well as truly and meaningfully. The Sophist argues that false statement, hence false belief,4 is possible, against the Parmenidean thesis that what is not cannot be (237a). It does so by demonstrating that what is not, in some way is. Take, for instance, the true statement, “Theaetetus sits”, and the false statement, “Theaetetus flies”. For a statement to be a statement at all, on the Sophist account, it must be about something (262e–​263a). Both the true and false statement here are about Theaetetus. The true statement states of what is, that it is, about Theaetetus. The false statement states of what is not, that it is, about Theaetetus, when it is not. The false statement, that is, attributes to Theaetetus something—​flying—​that is not, with respect to Theaetetus.5 Flying nonetheless constitutes something that is, in the sense that there is such a thing as flying, and in the sense that flying is different from or other than whatever is with respect to Theaetetus. One point on which Plato is in agreement with Parmenides is that a statement cannot be about nothing. So whilst it is possible to speak falsely and meaningfully, it is not possible to speak meaningfully about nothing.6 In identifying statements, lógoi, as (complex) linguistic entities or units distinct from names in this way, Plato, in essence, discovers that statements are units of information (correct or incorrect). The term lógos is central to Greek philosophy prior to Plato. Heraclitus, for instance, makes prominent use of the term—​in places, he seems to be using it to denote a general cosmic principle, at other points, to denote an account or argument. And Parmenides likewise uses 28

29

Plato on the act of informing

lógos to denote an account or argument.7 Already in these contexts, lógos signifies some sort of information. Arguably, a lógos is also already conceived of as a unit of information. But Plato is breaking new ground in the Sophist in attributing this informational sense to lógos where this has the sense of sentence or proposition, such that the sort of information in question is information that can be true or false.

Names as descriptions Whilst the Sophist tells us that a statement must be about something, and that it is the job of names to secure a reference, Plato’s Cratylus tells us that the way a name secures a reference is by means of description.8 A name is the vocal equivalent of a painting. It describes or imitates the thing named. In so doing, it encodes information, and this informational content allows a name to pick out the being (ousíā) of the thing named. This, on the Cratylus account, is the function of names. Or rather, it is one of two functions: “A name is an instrument of teaching and of distinguishing natures (ousíās)” (388b–​c).9 The first of these functions relies on the second: names function as an instrument of teaching by distinguishing the being of one thing from other things in the world. A name constitutes an instrument of teaching in the sense that, in picking out the being of a thing (through imitation), it indicates the referent of a statement, and so facilitates meaningful speech. Names, at any rate, have this potential. For names to fulfil their function(s), clearly, they must successfully pick something out. Extending the painting analogy, the Cratylus asserts that just as some paintings are better imitations than others, so some names are better imitations than others (no name will perfectly capture the thing it imitates—​for an image “if expressing in every point the entire reality, would no longer be an image”, 432b). Whilst paintings are made up of pigments that imitate those in nature, names are made up of sounds or letters “which bear some degree of resemblance to the objects of which the names are the imitation” (434a ff.). So long as a name has sufficient appropriate elements (appropriate to the object picked out) to preserve the general character or outline (týpos) of a thing’s nature, it constitutes an adequate description—​“still the thing is signified” (433a).10 Plato allows that, in some cases, convention acts as an additional source of information. For instance, in a case like sklērótēs (hardness). The “r” sound expresses hardness, while the “l” sound expresses softness. Yet, as Socrates says to Cratylus, “the word is intelligible to both of us … you know what I mean” (434e). We don’t mistakenly take sklērótēs to mean softness. Therefore, “custom must be supposed to contribute to the indication of our thoughts” (435b). So although names, as descriptions, have a natural connection to the things they name, on the Cratylus account,11 and are not simply arbitrarily assigned, as an extreme conventionalist would claim, convention does play some part, in some cases. Names, then, on Plato’s account, whilst not in and of themselves true or false,12 can be more or less correctly, as well as incorrectly, applied to an object, depending on their informational content (with some input, in some cases, from convention). Plato allows that different names, hence different information, can pick out the same thing; therefore, more than one name can be correctly applied to the same object. On the Cratylus account, it is the job of the name-​ maker, the legislator, “who of all skilled artisans in the world is the rarest of all” (389a), to determine informational content, and assign names. But only the expertise and oversight of the dialectician—​“he who knows how to ask and answer questions” (390c)—​can ensure the correct application of names. This expertise, which allows the dialectician to correctly determine what information should be encoded in each name, comes from being a user of names, and knowledge of the objects that names describe. 29

30

Tamsin de Waal

Where the Sophist indicates that the objects signified must exist, the Cratylus highlights the fact that the capacity for names to describe objects relies on the stability of the objects signified. “Can we rightly speak of a beauty which is always passing away, and is first this and then that?” asks Socrates (473e). This is a point also emphasized in the Theaetetus: Socrates: But since not even this abides, that what flows flows white; but rather it is in process of change, so that there is flux of this very thing also, the whiteness, and it is passing over into another colour, lest it be convicted of standing still in this respect—​ since that is so, is it possible to give any name to a colour which will properly apply to it? Theodorus: I don’t see how one could, Socrates, nor yet surely to anything else of that kind, if, being in flux, it is always quietly slipping away as you speak? (182d) If everything is in constant flux, language has no stability, and language needs stability to have meaning. This, in turn, suggests that information itself, the informational content of a name, requires stability, on Plato’s view—​at least in as much as one of the vehicles of information, speech, requires stability. Arguably, the Cratylus tells us that this stability is provided by transcendent forms; that names describe forms such as beauty itself, or good itself, where these are eternal and unchanging (439b). In this case, on the Cratylus account, the capacity for names to successfully pick something out, and therefore the possibility of meaningful speech, relies both on the existence of forms and on knowledge of forms.13

Education Plato’s most famous depiction of education comes in the Republic. Here, in the cave image, he depicts the process of enlightenment as a turning around and ascent of the soul—​an ascent away from the sensible realm and a dependence on sense-​perception, towards the intelligible realm, accessible only through reason (514a–​518b). It is a process that culminates in knowledge of forms. The act of informing in Plato, in the broader sense of educating, is thus ultimately a process of facilitating progress towards knowledge of forms. But how can humans, who are part of the sensible world, possibly come to know such entities as forms? It is a problem that Plato raises in the Parmenides. And it is a problem that remains central to contemporary philosophy of mathematics. In this context, Paul Benacerraf, in his paper “Mathematical Truth”,14 famously argues that no reasonable epistemology can mesh with the (mathematical) Platonist’s account of mathematical truth. Platonism, says Benacerraf, “casts a shroud of mystery over how knowledge can be obtained at all”.15 It fails to provide a coherent account of how a human knower can come to know the non-​spatio-​temporal mathematical entities the Platonist is committed to. In this same paper, in an aside, Benacerraf notes: It might not be unreasonable to suppose that Plato had recourse to the concept of anamnesis at least in part to explain how, given the nature of the forms as he depicted them, one could ever have knowledge of them.16 Plato discusses recollection (anámnēsis), his thesis that all learning is somehow really remembering, in three dialogues: the Meno, Phaedo, and Phaedrus.17 Whilst the recollection thesis is most obviously concerned with the accessing (and storage) of information, which I will come to shortly, it also distinguishes between different sorts and sources of information. 30

31

Plato on the act of informing

Different sorts and sources of information In the Meno, which will be my main focus here, recollection is introduced to explain how knowledge of virtue is possible and, ostensibly, to explain the possibility of inquiry. Meno, a young aristocrat, asks Socrates if virtue (aretḗ) can be taught, or if it is acquired some other way. Prompted by Socrates to first offer a definition of virtue, Meno attempts and fails to do so, and then asks (80d): How will you look for something you don’t know at all? Which of the things you don’t know will be the object of your search? And if you find it, how will you know it is the thing you didn’t know? Socrates reformulates this as a dilemma, with an emphasis on inquiry (80e): A man cannot inquire either about what he knows, or about what he does not know—​he cannot inquire into what he knows, because he already knows it; and he cannot inquire into what he doesn’t know, because he doesn’t know what to inquire about. At this point, Socrates introduces the recollection thesis: The soul is immortal, has been born many times, and has seen all things. There is therefore nothing it has not learned, “so it is no wonder that it can recall all that it knew before about virtue and other things”. What we call learning is actually recollection of things the soul knew before (81c‒d). Asked for some proof of this thesis, Socrates agrees to give a demonstration with the help of one of Meno’s household servants (82b–​85d). The boy, who has never been taught geometry (85e), is questioned on a geometrical problem: given that a square with sides of two feet has an area of four feet, what would be the length of the side of a square with double the area of the original? The boy initially replies that the length of the side will be double the length of the original side. Prompted by Socrates to see that a four-​foot side is too long, and that the length of the side must therefore be between two and four feet, the boy confidently replies that the length of the side must be three feet. On realizing that this, too, is too long, the boy exclaims that he does not know the answer (84a). Eventually, after further questioning, the boy arrives at the correct answer. According to Socrates, he has not taught the boy. Rather, through his questioning, he has drawn truths out of the boy that were already in him—​that have been in his soul throughout all time (86a). That the process of recollection is one that is not confined to mathematics, is underlined by Socrates’ assertion that the boy “can do the same as this with all geometry and every branch of knowledge” (85e). Dominic Scott has argued that recollection in the Meno is not in fact intended to explain the possibility of inquiry—​it is neither necessary nor sufficient to address the eristic dilemma (the Meno’s distinction between knowledge and true belief can solve this)—​but, rather, to solve the problem of discovery, a problem raised in Meno’s challenge.18 Scott explains: Someone might launch a definitional inquiry with true beliefs about the object in question; it is also possible that they might find the right answer. Nevertheless, it might be said, all they can claim is that this answer matches their beliefs about virtue. But why should this amount to knowing that the definition is true of virtue?19 The problem requires that one know that the initial specification of the object of inquiry is correct, so that one knows one has arrived at the right answer. It requires that one already know the object sought. So it cannot, like the problem of inquiry, be solved by allowing that we start from true beliefs. “What is distinctive about the problem”, says Scott, “is its underlying assumption that discovery or learning is a process of realizing that one thing matches something that one already knows”.20 Scott calls this assumption, that knowledge must derive from pre-​ existent knowledge, the “foreknowledge principle”, and argues that it is an assumption Socrates 31

32

Tamsin de Waal

shares. “Ultimately, Socrates concludes that any act of learning must be explained by the existence of conscious knowledge in a previous life.”21 Recollection addresses the discovery problem by allowing for the possibility that we can discover something that we already know—​we can do so, because we have forgotten it.22 On this reading, the Meno claims that we must already be in possession of a certain sort of information in order to acquire any further information. The Meno is not explicit as to the nature of this information. That we must already be in possession of knowledge, as opposed to true belief, is indicated at Phaedo 73a–​b, which refers back to the Meno recollection account.23 As to what it is that is known, however, Socrates says only that “the truth of all things that are” is in the soul (86a), and that the soul has learned “about virtue and other things” (81c), and “all things” (81d). This raises questions as to the scope of recollection: Does it encompass all learning, and what counts as learning? Does the reference to “all inquiry and learning” (zēteîn kai manthánein) (81d) suggest that it is restricted to philosophical inquiry?24 Is the knowledge recovered by recollection restricted to things such as virtue? What counts as things such as virtue?25 Is recollection restricted to certain disciplines? Geometry, certainly, would be included among these. But then what should we make of the reference to “geometry and every branch of knowledge” at 85e (above)? Socrates’ insistence on the priority of definition (PD) in the main discussion with Meno suggests that the knowledge in question is, at any rate, definitional knowledge, where this is knowledge from which further knowledge is derived.26 Socrates argues that it is necessary to know what virtue is, to have a definition of virtue, in order to know anything about its attributes:27 “If I don’t know what a thing is, how can I know what sort of thing it is?” (71b).28 It is this insistence on PD that makes the discovery problem a serious one, calling for recollection as a solution. PD rules out the possibility that one might start from anything less than knowledge of the definition of the object of inquiry. What is required of a definition, emerges from Socrates’ response to Meno’s attempts at a definition of virtue, and from the two examples he gives Meno to help him along (74b ff.). A definition must pick out all and only the things defined, and must be explanatory. In the case of shape (or surface, schêma) (one of Socrates’ examples), it must identify one feature in virtue of which all shapes are shapes. It must, in other words, pick out a unitary property of F-​ness. It must do so in terms understood. And it must do so in terms not themselves under discussion (i.e. it should not be circular). The Meno itself makes no reference to forms. We have to look to the Phaedo recollection account for this. Allowing that what is recollected on the Meno account is definitional knowledge, the Phaedo (and Phaedrus) adds to this account, that the objects of definition are forms, and that it is knowledge of definitions of forms that is recollected—​knowledge of “all those qualities … which we designate in our question-​and-​answer discussions by the term ‘itself ’ ” (75c–​d). The Phaedo posits forms as fitting objects of knowledge, because they, unlike sensible particulars, do not suffer compresence of opposites, and are therefore stable—​a form is unqualifiedly F (the form of equal is only and always equal, it does not admit of its opposite). Since they do not suffer compresence of opposites, forms can also explain how sensible particulars come to have the properties they have—​forms can be proper causes (aitíāi) (the form of beauty is the cause of beauty in all sensible particulars) (100a ff.). Taken together, then, PD and the foreknowledge principle suggest that we must already be in possession of knowledge of what x is, knowledge of the form, in order to acquire knowledge of anything else about x. In distinguishing between sorts of things known, in this way, the Meno distinguishes between different sorts of information: on the one hand, definitions of things, and, on the other, 32

33

Plato on the act of informing

attributes of things, propositions arrived at through logical inference. The former, it claims, is information we must already be in possession of, information that is recollected. Why, outside of the discovery problem, should Plato insist on this? One answer would be that we cannot treat all information the same—​there must be some basic concepts or principles not requiring justification, otherwise there will be an obviously vicious regress, as demonstrated by Lewis Carroll’s paradox of Achilles and the Tortoise.29 Another answer would be that experience cannot account for the formation of concepts. The Phaedo arrives at the claim that we must already be in possession of knowledge of forms on the basis that where, for instance, we perceive sticks and stones to be equal, we are reminded of “the equal itself ”, and prompted to compare the equal itself and equal sticks and stones. In so doing, we find that the equal sticks and stones are distinct and “fall short” of the equal itself (74d).30 Evidently, we don’t experience the equal itself through sense-​perception, so we must already have the notion of absolute equality independently of sense-​perception: Before we began to see and hear and otherwise perceive equals, we must somewhere have acquired the knowledge of equality as it really is; otherwise we could never have realized, by using it as a standard of comparison, that all equal objects of sense are desirous of being like it, but only imperfect copies. (75b) Since we use our senses from birth, our knowledge of the equal itself must have been obtained prior to birth. Recollection, here, most obviously explains the fact that we can compare forms and sensible particulars. If we take it that it is also intended to explain concept formation, then this brings us back to questions regarding the scope of recollection—​is it intended to explain concept formation in general, or only the formation of certain concepts? If the latter, how does Plato account for the formation of other concepts? Does sense-​experience play a role here?31 The Phaedo does, at any rate, explicitly point to sense-​experience as a crucial source of information in the process of recollection. Whilst it emphasizes the fact that forms are only accessible through reason, and that the soul reasons best when it is most by itself,32 the Phaedo also tells us that it is the perceived inadequacy of sensible particulars that prompts recollection: This notion of deficiency did not and could not have occurred to us except by sight or touch or one of the other senses … So it must be as a result of the senses that we obtained the notion that all sensible equals are striving to realize actual equality but falling short of it. (75a–​b) The Meno recollection account makes no such explicit reference to sense-​experience. The emphasis is rather on the boy’s mental reflection, the thought process that allows him to grasp the geometrical theorem, and thus information accessed through reason. However, the geometrical demonstration does make conspicuous use of diagrams,33 and it is clear that the perception of these diagrams helps the boy to follow the reasoning.34 For instance, the diagrams help the boy understand why his initial answers are wrong, by demonstrating that a square with a four-foot long side, and a square with a three-foot long side, produce squares that are too big. The diagrams also help to establish what is under discussion. At the outset of the demonstration, Socrates asks the boy, “Do you know that a square figure is like this?” (toioûton estín) (82b). In this case, the name “square” is clearly also playing a role. In the case of “diagonal”, Socrates 33

34

Tamsin de Waal

first introduces the concept by means of the diagram—​“Does this line, drawn from corner to corner …?”—​and only subsequently (and conspicuously) gives it a name: “The sophists call it the diagonal; so if the diagonal is its name …” (85a–​b). Both diagrams and names, here, in imitating, encode information. Both, in this way, serve as tools for instruction, facilitating the process of inquiry and discovery. However, as Plato makes clear in the Cratylus, both diagrams and names can encode incorrect information (436c–​d). So knowledge is not to be derived from such imitations, but from those realities, those truths, they imitate. These themselves should be studied (439b).

Accessing information Since definitional knowledge is knowledge that we are already in possession of, and that we have always had, it is not, strictly speaking, knowledge that is acquired. With respect to definitions of things, we are, in a sense, informed for all time. Still, as latent information, it is information that needs to be recovered, or accessed. So how exactly is it accessed? The first thing to note is that on the Phaedo account, the process of recollection cannot be completed in the embodied state. Only discarnate souls can fully recover knowledge of forms. So on the Phaedo account, there is a certain sort of information that is unavailable to us in the embodied state. On the Meno account, recollection is a process that can be completed in the embodied state. However, if the participation of a household servant in Socrates’ demonstration suggests that anyone can complete this process, the Republic makes it clear that attaining knowledge is something that is achievable only by an elite. This is underlined by the Phaedrus:35 Every human soul by its very nature has beheld true being … but not every soul finds it easy to use its present experience to recollect the world of reality. (249e–​250a) The recollection process presented in the Meno geometrical demonstration is incomplete. At the end of the demonstration, Socrates says of the boy: At this moment, those opinions have just been stirred up in him, like a dream; but if someone questions him about the same matters on many occasions and in many ways, you know he will have in the end as exact an understanding as anyone. (85c–​d)36 The boy has true beliefs, but has not yet achieved (explicit) knowledge. Knowledge and true belief, on the Meno account, are differentiated on the basis of stability. The hallmark of knowledge is that it is stable, where true belief is not. What makes knowledge stable is causal or explanatory reasoning (aitíās logismós). Thus a true belief can become knowledge, where we tie it down with such reasoning.37 The Meno does not elaborate on the nature of this reasoning, but it does explicitly identify it with recollection. Recollection, in turn, is emphatically characterized as a process of questioning. We see this above, at 85c–​d; also at 84c–​d, below; and at 86a: “[The boy] has had true opinions in him which have only to be awakened by questioning to become knowledge.” This process of questioning is, above all, set against the transmission of information through testimony. In the geometrical demonstration, Socrates repeatedly contrasts “questioning” and “teaching”. For instance, at 84c–​d: 34

35

Plato on the act of informing

As a result of this perplexity, he will go on and discover something by joint inquiry with me, while I merely ask questions (erōtôntas) and do not teach him (didáskontas); and be on the watch to see if at any point you find me teaching him or expounding to him, instead of questioning him on his opinions.38 The contrast is between a Socratic dialectical process that aims to draw out information that is already within the interlocutor, encouraging thinking and understanding for oneself, and a Sophistic style of education that aims to instill information in a student. In the Protagoras, Socrates complains that orators (amongst others, he has sophists such as Gorgias in his sights) are like books: they expound at great length, and are “incapable of either answering you or putting a question of their own” (329a). On Plato’s view, knowledge is not something that can simply be expounded and handed over from one person to another, in the way the sophists of fifth-​century Athens envisage. As Socrates asserts in the Republic: We must reject the conception of education professed by those who say they can put into the mind knowledge that was not there before … Our argument indicates that this is a capacity that is innate in each man’s mind. (518b–​c) Socrates’ disavowal of knowledge—​his insistence in the Meno, for instance, that he does not know what virtue is—​underlines the point that knowledge is to come from within the interlocutor. In the Theaetetus, Socrates is famously characterized as a midwife (149a ff.)—​one who helps to give birth to other people’s intellectual offspring: At first some of them may give the impression of being ignorant and stupid; but as time goes on and our association continues, all whom God permits are seen to make progress—​a progress which is amazing both to other people and to themselves. And yet it is clear that this is not due to anything they have learnt from me; it is that they discover within themselves a multitude of beautiful things, which they bring forth into the light. But it is I, with God’s help, who deliver them of this offspring. (150d) There is no (explicit) suggestion in the Theaetetus that this knowledge is already within us. The emphasis here is on the student conceiving ideas (those students who do not appear to be pregnant with any ideas are sent to the sophist, Prodicus, 151b). Nonetheless, the characterization of the dialectical process as a maieutic one is one that is applicable to recollection.39 A common objection to the way the Meno geometrical demonstration proceeds is that Socrates is asking leading questions. That is to say, his questions, for the most part, supply the answer, such that the boy only has to confirm the answer. (For instance: “Is not an area of eight feet double the size of this one, and half the size of the other? Yes”, 83c.) Thus, for all that Socrates claims to draw true beliefs from within the boy, he is effectively giving him the information, and so the process of questioning is hardly distinct from the process of teaching he is rejecting. One key difference is that the initial aim of the questioning process is to get the interlocutor to the point where he understands—​or, at least, to some extent grasps—​that beliefs he held before are false beliefs, and that he does not in fact know what he thought he knew. We see this, first, in relation to Socrates’ questioning of Meno, which leads Meno to a state of puzzlement (aporíā), unable to say what virtue is, where previously he was confident he knew what it was (80a–​b). 35

36

Tamsin de Waal

And we see it again, in relation to his questioning of the boy in the geometrical demonstration. In response to the boy’s realization that he does not, after all, know the answer, Socrates says: There now, Meno, do you observe what progress he has already made in his recollection? At first he did not know what is the line that forms the figure of eight feet, and he does not know even now; but at any rate, he thought he knew then, and confidently answered as though he knew, and was aware of no difficulty (aporeîn); whereas now he feels the difficulty he is in, and besides not knowing does not think he knows. (84a) The importance of this stripping away of false beliefs is that it clears the way for discovery of truths, and provides the impetus to embark on the process of discovery—​“now he will push on in the search gladly, as lacking knowledge” (84b). Arriving at a point of aporía is thus a positive and necessary step in the recovery of knowledge. A sophist, by contrast, convinces students that they know things they don’t know (Sophist 232b–​233c), encouraging idleness (see Meno 81d–​e). Meno is happy to unreflectively adopt and reiterate Gorgias’ views on virtue wholesale (71d, 76a). The geometrical demonstration brings out a further important difference, too. In order to respond to Socrates’ questions, the boy must assess the information contained in the question—​ particularly since he twice arrives at the wrong answer. This, in turn, requires him to follow the reasoning, and to grasp logical relations at every step. Socrates can thus maintain, in spite of asking leading questions, that the boy has been thinking for himself—​that the answers are his own (85b). By the end of the demonstration, in addition to recognizing what the answer is (how to double the square), the boy has some understanding of why it is the answer. In the case of the straightforward transmission of information, on the other hand (Gorgias expounding to Meno, say), the same information—​the same set of propositions—​may be conveyed, but the understanding as to how to order that information will not be conveyed. This is something that can only be discovered for oneself. A consequence of this is that someone receiving information through testimony has a much weaker epistemic commitment to any one proposition than someone who has grasped for themselves how the different propositions in a set of propositions relate to each other. Their beliefs are less stable.40 These differences between the Socratic dialectical approach and the Sophistic approach, highlight an important aspect of Plato’s thinking on information and its relation to knowledge. Knowledge is not achieved through the simple accumulation of pieces of information, the accumulation of lógoi, albeit that this information is correct. The Theaetetus indicates as much, when it rejects the idea that being able to state one’s beliefs is enough to constitute knowledge: “Isn’t that a thing that everyone is able to do more or less readily—​I mean, indicate what he thinks about a thing …?” (206d). Attaining knowledge requires an understanding of how to order information, where this is something that cannot be transmitted. That grasping relations and interconnections, and ordering information, is key to attaining knowledge, on Plato’s view, is highlighted in the Republic. Here, in setting out the prospective philosopher-​rulers’ mathematical education, Socrates observes: If the investigation of all these studies goes far enough to bring out their community and kinship with one another, and to infer their affinities, then to busy ourselves with them contributes to our desired end, and the labour taken is not lost; but otherwise it is in vain. (531c–​d)41

36

37

Plato on the act of informing

A little later, Socrates asserts that the capacity to view such relations and interconnections is “the chief test of the dialectical nature and its opposite. For he who can view things in their connection (synoptikós) is a dialectician” (537c). This capacity to view things in their connection, in turn, enables the dialectician to give a lógos, in the sense of account, of the being or essence (ousíā) of each thing (534b). That is, it enables the dialectician to define each thing, where this, the Theaetetus suggests, is a matter of “being able to tell some mark by which the object you are asked about differs from all other things” (208c). The Meno boy, of course, whilst he has achieved some understanding, is a long way from the synoptic view of the dialectician. But with further questioning—​about the same matters on many occasions and in many ways—​the boy can make further progress towards this. Whatever the exact nature of this questioning,42 the crucial point here is that the boy’s (and Meno’s) progress depends on further questioning. Likewise, whatever the exact sense of aitíās logismós—​ whether we interpret it broadly to mean the very process of dialectical reasoning itself, or, more narrowly, to mean the sort of reasoning encouraged by this dialectical process—​it is clear that the process of questioning is key to attaining knowledge, on Plato’s view.43 Whilst sophists such as Gorgias expound, some sophists do engage in questioning—​as, for instance, the (eristic) sophists Euthydemus and Dionysodorus do in Plato’s Euthydemus. So it is not straightforwardly the process of questioning that sets the dialectical approach apart from that of the sophists. Nor is it the subject of the questioning. Euthydemus and Dionysodorus set out to demonstrate that virtue can be taught, and that they are the right ones to teach it. It is, more particularly, the manner of the questioning that distinguishes the two approaches, where this, in turn, reflects distinct aims. Socrates, in the Meno, contrasts the eristic and dialectical approaches to definition (75d). In the context of an eristic exchange, which is a combative affair, it is enough for one side to state a true definition; if the interlocutor deems it wrong (no matter his grasp of the definition), he must examine and refute it. In the context of a “milder” dialectical exchange, on the other hand, it is not enough simply to state a true definition; that definition, as we have seen, must also be expressed in terms that the interlocutor (that specific interlocutor) understands.44 This dialectical approach is what is characterized at Phaedo 73a as the right way of questioning. And this brings us back to the Cratylus. As we saw, the dialectician, here, is identified as a user of names, and, implicitly, the ultimate user of names. Since names are tools for instruction, the dialectician is also an instructor (388e), implicitly, the ultimate instructor. That this instruction proceeds by means of Socratic style dialectical discussion is made clear in the description of the dialectician as “he who knows how to ask and answer questions”. And it is striking that the dialectician is initially described as “he who knows how to ask questions” (390c), putting the emphasis on the dialectician’s role as questioner. This is a point highlighted by David Sedley, who argues that whilst the mundane function of names is to pick out a referent, the ultimate function of names is to facilitate the questioning process.45 In the hands of the dialectician, a name, by picking out the thing that is under discussion, allows for a process of definition that ultimately leads to an understanding of the essence that is the subject of inquiry.46 This suggests that truly speaking meaningfully (and truly), on Plato’s view, is a matter of using language correctly and effectively to facilitate philosophical inquiry into definitions (of forms).

Conclusion Plato is clearly preoccupied with the act of informing, and this sees him focus on a range of questions relating to information that remain relevant today. In so doing, Plato makes an

37

38

Tamsin de Waal

important contribution to philosophical thinking about information—not least, in his analysis of lógos. His contribution is as much in the asking and articulating of these questions, as in the discussion of how to answer them. But this focus on information should not lead us to think that Plato’s philosophy easily lends itself to what one might call an informational interpretation. I set out with a characterization of information on which information is a message (or what is contained in a message) communicated from a sender to a receiver, which (ideally) informs the receiver. Plato’s account of meaningful speech accommodates such a characterization, indicating what, minimally, could constitute such a message, and highlighting the fact that information is something that can be both true and false. Plato’s account of education, however, challenges the notion that information is something that is communicated from a sender to a receiver, and thus that the act of informing is the communication of information. For Plato, education is, first and foremost, a process of recovering information that we are already in possession of—a process of discovery for oneself. It is, of course, a process of communication, a process of question-and-answer, but it is not a straightforward process of communicating, or transmitting, information. Plato expressly rejects such a model of education. That Plato does not conceive of education as an informational enterprise, in this sense, is underlined, first, by Plato’s misgivings regarding writing as a means of communication, as famously set out in the Phaedrus (274 ff.), and, second, by the very nature of a Platonic dialogue.

Notes * I am very grateful to Dom Bailey for helpful comments on an earlier draft of this chapter. 1 Floridi (2010), 1. 2 Floridi (2010), 2. 3 The Theaetetus, in setting out three possible senses of lógos (statement, enumeration of the elements of a thing, account) at 201c ff., highlights both the range of senses that can attach to lógos, and the indeterminacy of the word (“Come then, what are we intended to understand by a lógos? I think it must be one of three meanings”, 206c). Here, as in the Sophist, lógos is characterized in terms of a weaving together (symplokḗ ) of names (where names here encompasses nouns and verbs) (202b). 4 “What we call thought is statement that occurs without the voice, inside the soul in conversation with itself ” (Sophist, 263e). This points to the possibility of meaningful speech within oneself—​see also the Theaetetus characterization of thinking as an internal discourse (lógos) of the soul (189e ff.). And see Sedley (2004), 169, n. 28. 5 Plato hereby distinguishes the sense in which a statement, lógos, is not true, and the sense in which a string of names or verbs, as mere lists, cannot be true. 6 See McCabe (1994) 197 on the Cratylus: “All speech, to get off the ground, must start with a something (or some somethings). Reference, that is, is fundamental to successful speech; speech that fails of reference is failed speech.” 7 On some interpretations, Parmenides also uses lógos to denote “reason” at (DK 28B7). 8 The Cratylus, a likely earlier dialogue, focuses on what makes a name the correct name for any given object, where “name”, here, includes nouns, adjectives, proper names. See esp. Sedley (2003). 9 See also 428e and 435d–​e. 10 See also Cratylus 393d and 394c. 11 “Names belong to things by nature” (Cratylus 390d). 12 Certainly, on the Sophist account—​only a statement can be true or false. And see Fine (1977) on falsity in the Cratylus. 13 And see Parmenides 135b–​c: “If someone, having an eye on all the difficulties we have just brought up and others of the same sort, won’t allow that there are forms for things and won’t mark off a form for each one, he won’t have anywhere to turn his thought, since he doesn’t allow that for each thing there is a character that is always the same. In this way, he will totally destroy the ability to engage in discussion/​dialectic.” 14 Benacerraf (1983) 403–​20. 15 Benacerraf (1983) 416.

38

39

Plato on the act of informing 16 Benacerraf (1983) 416. See also Maddy (1992) 21. 17 The Parmenides, notably, does not reference recollection. Nor does the Republic. But the general characterization of education here, is reminiscent of Plato’s characterization of recollection—​see esp. 518b–​c. 18 See esp. Scott (2006) 83 ff. 19 Scott (2006) 83–4. 20 Scott (2006) 84. 21 See Scott (2006) 84–​5. He points to the passage at 85d, towards the end of the geometrical demonstration, as the best support for this. 22 Against this view, Fine (1992), for instance, argues that the theory of recollection is not a theory of innate knowledge. On Fine’s view, recollection explains our tendency, in inquiring, to favour true over false beliefs: “We can all inquire and tend toward the truth in doing so, because, although we now lack the relevant knowledge, we once had it, in a prior life” (213). 23 In the Meno itself, Socrates refers both to the knowledge in the boy (85d), and the true beliefs in the boy (85b–​c, 86a). 24 See Scott (1995), esp. chs 1–​2, for the view that recollection is concerned with higher learning, and that mundane concept formation is accounted for empirically. See also Fine (1992), n. 26, 221, for the view that recollection is restricted. 25 See n. 35. 26 Thus in the mathematical context, one proceeds by logical inferences from definitions such as “odd number” or “even number”, to prove a proposition(s). 27 For discussion of whether or not Socrates/​Plato is in fact committed to the priority of definition, see Benson (1990), for example, for the view that definitional knowledge has to be prior; against this, see Vlastos (1985). 28 Similarly, in the Laches, Socrates says: “Then isn’t it necessary to start out knowing what virtue is? Because if we are not absolutely certain what it is, how are we going to advise anyone as to the best method of obtaining it?” (190b). 29 The tortoise does not accept that we treat logical principles such as modus ponens as rules of inference—​he wants to insist that we treat them (the information they encode) the same as the premises of the argument, in which case they stand in need of a prior principle to explain them, and then a further principle to explain that principle, and so on ad infinitum. 30 One sense, at least, in which sensible particulars “fall short” of forms is that they can be qualified by opposites (e.g. be at the same time both equal and unequal). Cf. Republic 523a ff. 31 Again, how generous a view of sense-​experience we attribute to Plato, depends on how restricted a view of recollection we take. A more restricted view, allows for a more generous interpretation of sense-​experience as a source of information. 32 The body is a hindrance, on the Phaedo account (65a–​c). 33 Whilst there is no explicit reference to the use of diagrams (which is notable in itself), Socrates consistently uses demonstratives, and there is explicit talk of drawing (e.g. at 83b). See, for instance, Bedu-​ Addo (1983) on this. 34 See esp. Patterson (2007) on the role of diagrams here. 35 Plato does, however, suggest that some forms (the equal itself, for instance, and forms of artefacts), are more accessible than others (moral and aesthetic forms, or value forms, such as the form of good, or beauty) (see, for example, Phaedrus 250a ff.). 36 I follow Scott (2006: 106) in the italics here. 37 “These (true beliefs), so long as they stay with us, are a fine possession, and effect all that is good; but they do not care to stay for long, and run away out of the human soul, and thus are of no great value until one makes them fast with aitíās logismós. And this process, friend Meno, is recollection” (97a–​b). 38 See also 85d: “Without anyone having taught him, and only through questions put to him, he will understand recovering the knowledge out of himself ”, and 85e. 39 See Burnyeat (2012) for a comparison of the Socratic method in the Meno and Theaetetus. 40 See Scott (2006) 100–​3. 41 See also Republic 537b–​c. 42 Perhaps, in the first place, this would be a matter of working through the same problem a number of times; perhaps, too, it would involve working through other sized squares, to confirm the theorem—​to understand that it generalizes. Ultimately, though, it seems that the questioning must prompt progress

39

40

Tamsin de Waal towards an understanding of the whole system of geometrical propositions, and then an understanding of how geometry relates to other branches of mathematics. 43 See Gonzalez (1998) 349–​50, n.83, on aitíās logismós in the Meno as reasoning about forms, as opposed to systematic, synoptic reasoning. 44 In the Sophist, Plato emphasizes the importance of achieving a common understanding, and, at the same time, highlights the distinction between defining and naming. Seeking to establish the nature of the sophist, the Stranger says to his interlocutor, Theaetetus: “At present, you see, all that you and I possess in common is the name. The thing to which each of us gives that name we may perhaps have privately before our minds, but it is also desirable to have reached an agreement about the thing itself by means of explicit statements, rather than be content to use the same word without formulating what it means” (218c). 45 Sedley (2003) 62ff. 46 Where, moreover, a name constitutes a sufficiently accurate description of the thing it names, it can offer guidance in the investigation of its being, its essence. See Sedley (2003) 151.

References Bedu-​Addo, J.T., “Sense-​experience and Recollection in Plato’s Meno”, The American Journal of Philology, 104(3) (1983), 228–​248. Benacerraf, P., “Mathematical Truth”, Journal of Philosophy, 70 (1973), 661–​679, reprinted in P. Benacerraf and H. Putnam, eds., Philosophy of Mathematics: Selected Readings, Cambridge: Cambridge University Press, 1983, 403–​420. Benson, H.H., “The Priority of Definition and the Socratic Elenchus”, Oxford Studies in Ancient Philosophy, 8 (1990), 19–​65. Burnyeat, M.F., “Socratic Midwifery, Platonic Inspiration”, in Explorations in Ancient and Modern Philosophy, Vol. 2. Cambridge: Cambridge University Press, 2012, 21–​35. Diels, H. and W. Kranz. Die Fragmente der Vorsokratiker. Zurich: Weidmann, 1985. Fine, G., “Plato On Naming”, The Philosophical Quarterly, 27(109) (October 1977), 289–​301. Fine, G., “Inquiry in the Meno”, in The Cambridge Companion to Plato, Cambridge: Cambridge University Press, 1992, 200–​226. Floridi, L., Information: A Very Short Introduction, Oxford: Oxford University Press, 2010. Gonzalez, F.J., Dialectic and Dialogue: Plato’s Practice of Philosophical Inquiry, Evanston, IL: Northwestern University Press, 1998. Maddy, P., Realism in Mathematics, Oxford: Oxford University Press, 1992. McCabe, M.M., Plato’s Individuals, Princeton, NJ: Princeton University Press, 1994. Patterson, R., “Diagrams, Dialectic, and Mathematical Foundations in Plato”, Apeiron, 40, 1–33, 2007. Scott, D., Recollection and Experience: Plato’s Theory of Learning and its Successors, Cambridge: Cambridge University Press, 1995. Scott, D., Plato’s Meno, Cambridge: Cambridge University Press, 2006. Sedley, D., Plato’s Cratylus, Cambridge: Cambridge University Press, 2003. Sedley, D., The Midwife of Platonism: Text and Subtext in Plato’s Theaetetus, Oxford: Clarendon Press, 2004. Vlastos, G., “Socrates’ Disavowal of Knowledge”, The Philosophical Quarterly, 35(138) (1985).

40

41

3 ON INFORMATION IN ARISTOTLE Nature, perception, knowledge Miira Tuominen

Introduction General introductions to information in the history of philosophy sometimes claim that information was a technical notion in classical ancient philosophy (e.g., Adriaans 2018, 2.1). This declaration risks taking us into an interpretive loop of some 2,400 years. It is true that both Plato and Aristotle talk about what in English are translated as “forms” (Greek eidos, idea; cf. “in-​form-​ation”).1 However, to say that Plato and Aristotle use “information” as a technical notion is a claim that is deceptively easy, and its meaning and implications should be articulated before proceeding to make such a statement. One reason why conceptual clarification is needed is that, also in Greek philosophy, one easily finds information everywhere, in discussions about metaphysics (forms), theory of causes, logic, theory of knowledge, perception, even rhetorics,2 and it is not immediately clear what exactly is meant by “information” in the different cases. Perhaps this is not necessarily a bad thing. In today’s discussion, the notion of information also pops up in various contexts and in Luciano Floridi’s words that might not always be a conceptual confusion but “the result of intentional bridging” (Floridi 2015, untitled introductory section). However, as in today’s discussions, we need at least a working definition or conception of what we are looking for. Otherwise we are in danger of chasing a creature like the sophist in Plato’s eponymous dialogue that is showing up in various branches of the divisions, always in a different guise. In fact, as we shall see soon below, in Floridi’s own classification information to an extent is such a definitional chameleon that can cross boundaries of divisions. It can be said that, in reality, as Aristotle conceives it, forms inform matter and that forms inform knowers.3 On the one hand, this means that things are structured and their way of being is determined and regulated by immaterial principles called “forms”. Aristotle’s view that such principles are intelligible entails that creatures with reason are capable of understanding, knowing or grasping them. However, he also maintains that there are perceptible forms that sentient creatures (human beings and non-​human animals) are aware of in perception. They can see colours, hear sounds, smell odours and taste flavours as well as become aware of tactile qualities such as roughness or smoothness through the sense of touch, while size, shape and movement, for instance, are not specific to any particular sense but can be perceived by many senses, and the perception of movement seems to require memory. Not all animals have all the 41

42

Miira Tuominen data (structured)

environmental

semantic (content)

instructional

factual

untrue

true

Figure 3.1  An illustration of Floridi’s (2015) classification of information.

senses, however, but even those that only have the sense of touch have some knowledge (gnôsis) about what the world is like, insofar as it is perceptible. Moreover, the process through which percipient creatures gain perceptual knowledge about the world (or cognise it) is explained by reference to forms “informing” the material medium between perceptible objects and percipients as well as their sense-​organs, so that the animal’s capacity to perceive is activated and it becomes aware of the perceptible quality. In each of these cases, the question of what it means for forms to inform something (matter, the medium, the sense-​organs or the intellectual capacity of the knowers) are among the hardest interpretive issues in the scholarship on Aristotle. It falls outside the scope of this chapter to tackle in detail the difficult questions and controversies concerning the question of how to understand Aristotle’s claims about forms in these different contexts. Instead, I shall focus on the question of how these discussions can be mapped on to some of today’s central debates about information. I shall take Floridi’s classification and working definition (the general definition of information GDI) as my starting point, although I shall also pay attention to some of Floridi’s critics. The following figure illustrates Floridi’s classification, and I shall refer to it in my discussion (see Figure 3.1). One of the puzzling aspects of Floridi’s classification is that, on the one hand, information is defined as being semantic and characterised by three conditions: (i) consisting of one or more data that is (ii) well-​formed and (iii) meaningful. On the other hand, while data is thus divided into semantic and environmental (e.g., Floridi 2010, 20), environmental data is also called “information” (and not only “data”) (ibid., 73ff.). Moreover, although Floridi is well known for his controversial claim that semantic information is true and thus distinguished from misinformation and disinformation, he allows instructional environmental information as well. This is noteworthy, since his distinction between instructions and factual semantic information is based exactly on the claim that while the former cannot properly be called true or false, the latter can –​and Floridi argues that information in this sense must be true. In the illustrations of the divisions, instructional environmental data is also identified as semantic information, thus showing that the division between environmental and semantic information is not exclusive for Floridi. It is this feature that makes information look like the sophist in Plato’s Sophist. Although there are problems in Floridi’s classification, it is useful because it is rather well known and because its divisions raise important questions for this chapter as well. The first problematic feature just mentioned also makes Floridi’s notion of information analogous to some of Aristotle’s technical terms: it is used both in a broad sense to cover various different kinds of information and, technically, only of one specific kind of information (semantic, 42

43

On information in Aristotle

factual and true or correct).4 For the present purposes, the important implication of this is that Floridi’s work also concerns information in its broader sense and is not only about the narrower technical sense he reserves for it. I shall leave the second problematic aspect, that is, Floridi’s treatment of environmental information as instructional aside in this context, since it is not directly relevant to my discussion of Aristotle. More generally, the question of whether environmental information in biology is semantic is a debated issue, and I shall not enter that debate either. I shall follow the general practice in today’s discussions of using the notion of information in a rather general sense, also to refer to items that on some conceptions of information are not information, properly speaking, but data. As just noted, Floridi himself does so, while also defining one of the subcategories (semantic factual data that is truthful or correct) as information, properly speaking. In this chapter, I shall consider Aristotle’s view according to which the forms that structure matter can also be realised in knowers and the question of how the forms are transmitted from the objects to the knowers. When referring to knowers, I use the term to refer both to human knowledge but also to “knowledge” in animals to which Aristotle refers by the Greek term gnôsis that can also be translated as “cognition”. Therefore, we should not take the reference to “knowers” to imply that all of them have propositional beliefs, beliefs that are also true and warranted. Rather, Aristotle clearly allows forms of veridical cognition that do not satisfy that kind of a definition. Instead of such knowers we could talk about “cognisers”, but that would seem unnecessarily awkward. Therefore, I use the term “to know” here to cover also animal cognition that can be veridical but is neither conceptual nor propositional. I shall argue that it is in fact helpful to take Aristotle to be concerned with a question about the transmission of information or data –​a discussion that must be distinguished from more directly epistemological concerns about the nature of knowledge.5 With respect to the transmission of information from objects to knowers in perception, I shall argue that it is important to grant that the transmission includes a material change in the knower, although the change does not need to be literal (i.e., the percipient’s sense-​organs do not need to take on the perceived quality in the process of perception). In the language of information this can be expressed as the point that it is not merely the semantic content that is taken on by the percipient but also some material realisation of it.6

Environmental instructional information in Aristotle? In today’s discussions concerning the use of informational language in biology, one central question is to what extent the analogies of coding and programming should be taken as analogies or whether there is a more literal meaning to them. The question is, for instance, whether and in what sense genes should be taken to “carry a message”. If we assume that they should, this implies that the notion of information in biology is a semantic one (or a teleosemantic one),7 whereas if the expressions are taken metaphorically (as, e.g., Levy 2011 does), a causal notion of information is sufficient to explain the same phenomena. One way of expressing the difference between a causal and a semantic notion of information is that while the latter is taken to have content or to be about something, the former (i.e., causal information) is a difference-​maker and having content is not necessary for it. In order to illustrate what it means to be a difference-​maker, let us imagine a device that is designed to detect red balls moving across a table and to count them. Its state after crossing the distance is different depending on whether or not it has encountered red balls; it shows a different number depending on how many red balls there were on the table. Therefore, whether or not there are red balls on the table (and how many) makes a difference and is informative to the device without semantic content in the device.8 Whether the language of carrying a message in the semantic notion of 43

44

Miira Tuominen

information is taken metaphorically or more literally, it is generally agreed that the messages that are typically compared to computer programs have instructional, prescriptive or imperative rather than descriptive content. The crucial point of this distinction is that instructional, prescriptive or imperative information is compared to commands or orders (“do this”, “do that”) and must be distinguished from factual information resembling indicative statements about facts (“it is raining at time t1”, “Paris is the capital of France”). Commands or orders do not have truth-​value, properly speaking, but are rather instructions of what to do in a particular situation, while indicative statements express matters of fact and can be correct or incorrect and thus true or false. In Aristotle’s biology, the model of explanation is certainly teleological: the cause for the sake of which (to hou heneka) occupies a central place in the explanations (e.g., De an. 2.4, 415b15–​ 17) as do the principle(s) that nature does nothing in vain and acts for the best (Progression of Animals 2, 704b12–​18).9 When discussing the works of nature, Aristotle also makes use of several metaphors that liken natural processes to various forms of human action. However, although Aristotle talks about nature as if she was deliberating on how to use certain materials for the good, this talk must be taken metaphorically. He specifically denies that nature is or has been organised by a divine craftsman and argues against Plato on this point.10 Therefore, although for instance Aristotle says that nature is devising (Parts of Animals 2.7, 652a31) things so that if there are contrary parts in animals, they counteract each other –​in this case, the opposition is between the coldness of the brain and the heat of the spinal marrow that is continuous with it –​this does not entail that he takes animal bodies to have been designed. Neither should we take the talk about nature being, for example, as it were, a good housekeeper (Generation of Animals 2.6, 744b16–​27)11 or a carpenter (Generation of Animals 730b19–​23; 740b25–​741a3) to mean that nature is actually managing things around us. Rather, while Aristotle talks about nature in this way, he does not think that nature should really be compared to the practitioners of the art but to the art or craft itself (Physics 2.3, 195b21–​25).12 Therefore, although Aristotle’s metaphors pull in the direction of design, we should resist this pull and take it seriously that nature is not a devising agent but rather something like a principle: like art and not the practitioner of the art. The reason why crafts provide a useful analogy probably is that they are better known to us than the workings of nature that are “better known in nature”. This is why the former can be used as analogues to the latter, while in reality it is art or craft that imitates nature; not the other way around.13 However, as scholars have also stressed, the principle(s) that nature does nothing in vain and works for the best does not mean that Aristotle would take natural organisms to be ideal. There are several kinds of constraints to the natures of things due to which the best arrangement of an animal body might not possible.14 The constraints can be due to the material constitution of the animal but also its essence. An example of the former kind is Aristotle’s discussion of why not all animals give birth to living young, although that would be best (Generation of Animals 2.1, 732a25–​733b16; 2.4, 737b15–​27). The explanation of why snakes have no legs is due to a constraint of the latter kind (Progression of Animals 8, 708a9–​20). There are also other kinds of constraints, for example, ones related to the arrangement of inner organs –​Henry calls them “architectural” (2013, 239) –​such as the respective locations of the windpipe and the oesophagus in animals with lungs that necessitate either an epiglottis or a collapsible larynx (Parts of Animals 3.3). An ideal arrangement would be to have the mouth directly connected to the stomach and not to have an entrance to the windpipe thus risking suffocation by food. Therefore, while the constraints prevent an ideal arrangement, there is some good achieved by it. It is difficult to explain in detail how we should read Aristotle’s metaphors in the case of natural teleology and how exactly to understand the workings of causes for the sake of an end 44

45

On information in Aristotle

in nature.15 In fact, considering the central role of teleology (the cause for the sake of which) in Aristotle’s biology and philosophy in general, one would have expected him to explain the matter in more detail and less metaphorically than he does. However, the point I would like to make with respect to the topic of this collection is relatively uncontroversial. Insofar as we articulate Aristotelian biological explanations in terms of the notion of information, we should not conclude that such information is semantic. Although the organisation of animal organisms is explained by reference to what Devin Henry calls “optimality reasoning” (nature does nothing in vain and works for the best), this does not mean that nature would literally be sending messages or programming natural organisms to function in certain ways. As has been pointed out above, for Aristotle nature is rather a principle like art, not a divine craftsman or administrator. Some of the constraints that prevent an ideal outcome and entail that nature “devises” another solution for some good might also seem to imply that there is some adaptation during the development of individual organisms.16 However, it seems to me that these considerations in Aristotle are quite far from the references in modern biology to organisms’ adaptability to circumstances and the effects of environment in the natural development of embryos, for instance.17 If nature engaged in the development of individual organisms and, for instance, removed legs from snakes because as blooded animals they could only have four and could not efficiently move about on four legs, there should also be snakes that do have those four legs and were not adapted in that way. However, since the features adapted in this way are uniform in the members of their species, there seems to be no space for adaptation of individuals during individual embryonic development.18

Reception of perceptible forms in perception Aristotle analyses perception within his general theory of causes in which a cause producing an effect requires interaction between the productive and the receptive party of the process. With respect to perception, he distinguishes between the effects perceptible qualities have on different kinds of (animate) receptors. Both plants and animals are affected by heat for instance, that is, heat is something that makes a difference for plants, to borrow a contemporary description of causal information or data.19 Heat, for instance, can cause plants to wither, while animals also perceive the heat and, thanks to their capacity of self-​movement (in those animals that have it; see 413b2–​4; 415a6–​7), can move away from it if it is excessive. One way in which the difference is articulated is that plants and inanimate creatures are affected with their matter (De an. 2.12, 424a30–​b3) while the senses of animals are affected “as being of such a quality and in accordance with its logos” (424a24), a description that is not self-​explanatory to say the least.20 In his discussion on perception in the De anima, Aristotle considers many questions that from the perspective of the modern division of philosophical topics seem odd. For instance, why is the capacity of all living organisms to absorb nutrition relevant for perception? Would one rather not want to hear about whether our perceptions are reliable and how we can distinguish between correct and incorrect perceptual appearances? Such features are partly related to the fact that Aristotle’s discussion about the soul concerns all living beings (e.g., 414b32–​33), and the soul is the principle of life in all of them (plants, non-​human animals, and human beings). Moreover, it has also been argued that the discussion of nutrition exemplifies Aristotle’s program in the De anima of explaining how the soul explains the activities and functions of the body: nutrition differs from mere augmentation, since while the latter goes on infinitely if there is no lack of material (e.g., fire burns as long as there is some combustible material around), nutrition has a natural goal of not only increasing the size of a plant or an animal but taking 45

46

Miira Tuominen

it to and keeping it in its natural size (Menn 2002, 117–​118). Therefore, both nutrition and perception are soul’s activities (Gr. energeia) that it performs through the body, for example, the creatures are nourished through their metabolism and perceive through their sense-​organs.21 In general, Aristotle distinguishes between different kinds of souls on the basis of their functions or capacities. Creatures with nutritive souls are capable of taking in nourishment, of growth or development and reproduction. By virtue of their sensitive soul, animals also have the capacities of perception and self-​motion. In addition to these, human beings have reason that enables them to attain generally valid knowledge, also knowledge concerning the causes or explanations and the natures of things. In brief, Aristotle distinguishes between different kinds of knowledge and while awareness or cognition of perceptible qualities, colours, sounds, smells and so on, as well as common sensibles such as shape and size and recognition of individuals, is common to many non-​human animals and human beings, human forms of cognition and knowledge differ from that of the animals because human beings can operate with genuinely universal notions that concern all their instances. Animals can form experiential generalisations on the basis of what they have perceived but this, in Aristotle’s view, falls short of universal generalisation that concerns all individuals of a certain type. Forming universal notions from experience is necessary for such generalisation that, in its turn, is crucial for acquiring the kind of knowledge that we need to engage in scientific inquiry. Through such inquiry, we can come to know essences and explanations of kinds and phenomena, and it is such knowledge about causes and essences that is vital in scientific knowledge (epistêmê) as Aristotle conceives it. However, all knowledge-​ acquisition even in human beings, he maintains, starts from perceptions in the sense that we acquire universal notions from experience and they can then be used to actively inquire into the natures and explanations of things. As regards perception, animals including human beings are also, as bodily beings, affected with their matter, for example, they are heated by heat. However, it is not such affection that causes them to perceive the heat. Rather, animals are sensitive or capable of perception because they can be affected in another way. Aristotle’s description of this affection has been the subject of intense controversy. His key description is that sentient or perceptive beings must be capable of receiving the perceptible form without its matter (De an. 2.12, 424a17–​19).22 Such reception must be distinguished, on the one hand, from receiving, for example, a red object with its matter. Getting a red colour patch on one’s skin from a paint jar could be an example of such affection. However, seeing red is not like this and even if some scholars have argued that there is a sense in which something in the eye becomes red in some sense of the word, this is different from getting red paint in one’s eye. Moreover, it is not clear whether we should accept that the eye becomes red in any sense of the word when we see red. I shall return to this question below. However, Aristotle does not merely explain perception by reception of forms without matter. If he did, he would perhaps risk a vicious circle: defining perception or sensation in terms of the reception of forms without matter and then saying that animals differ from plants precisely because they can receive forms without matter.23 Rather, he says that the reason why plants cannot perceive is that they do not have a mean (mesotês) or a principle (archê) receptive of perceptible forms without matter but are, as mentioned, affected with the matter (De an. 2.12, 424b3). It is the mean that explains our capacity to perceive and allows us to distinguish or discern (krinein) the perceptible qualities and/​or objects (2.11, 424a4–​6). The mean is supposed to be neutral with respect to the perceptible qualities that are proportions (logoi) between two extremes, for example, black and white for colours, hot and cold for tangible temperatures –​ and there can be several such pairs of contraries for each sense (ibid., 422b23–​31).

46

47

On information in Aristotle

It is also crucial that sentient beings can retain the mean. If not, they would, for example, be heated by heat, in which case temperatures that are colder than this newly acquired warmer one would be perceived as cool or cold, although they are rather towards the warm end of the range between the two extremes between which sensible hot or cold temperatures are located.24 Sentient beings can discern sensible or perceptible qualities in a reliable way, since the effect (in some special sense) the object has on the sense-​organ causes us to perceive the quality as it is in its perceptibility, and our perception of it is also the full actualisation of that quality. For example, a colour is a potentiality to be seen and that potentiality is actualised when the colour is seen (e.g., De an. 3.2, 426a15–​19). The surface of the body that has the colour (as such potentiality) affects the transparent medium25 between the object and the percipient and the transparent medium then affects the organ that, being an organ of an ensouled and sentient body, activates the creature’s capacity to see. Aristotle insists (in De an. 2.5) that the soul is not moved or altered by the object or the body (or the sense-​organ) but, rather, changed or perfected in a similar way as people who have knowledge but are not using it are “changed” when they start using their knowledge (ibid., 417b5–​7; 18–​19).26 Therefore, what it means for animals to be capable of perception and to have senses is that they have a mean that enables them to receive perceptible forms without the matter and to discern the perceptible qualities that fall within the range of extremes between which the mean is a neutral proportion.27 It is the perceptible qualities that are proper to each sense (colours for sight, sounds for hearing, olfactory qualities for the sense of smell and gustatory qualities for the sense of taste) that are such proportions. In addition, there are qualities such as shape and size that can be perceived by several senses, and the so-​called accidental sensibles that are not strictly speaking perceived (things or substances such as this human being or this horse) but can be recognised by the perceptual capacity. Aristotle’s discussion of the various functions and capacities of the sensitive soul allows a wide range of cognitive tasks28 from sensations of touch in some simple organisms to the human capacity to have experience from which we can start to achieve scientific knowledge thanks to our reason’s capacity to form general concepts.29 The simplest organisms discern the qualities that enable them to find their food, to feel pleasure and pain and to desire what is pleasant (414b1–​2; 4–​5), as well as to launch some kind of protective measures in the presence of danger. In more complex animals and human beings, perceptual cognition can amount to rather sophisticated acts of discernment. In many animals, perceptions are retained and that explains memory; some animals can also imagine things that are not present thanks to the capacity of phantasia and to accumulate experience about what they have perceived. In human beings, these capacities also lead to the formation of general concepts that enables us to process information in a different way, using general concepts and making inferences. Aristotle maintains that, as opposed to other animals, human beings are capable of making universal generalisations that require being able to subsume individuals under general concepts, and thus proceeding to engage in an inquiry into the causes and natures of things. While some animals are capable of rather demanding cognitive tasks, according to Aristotle, they lack the capacity of proper universal generalisation. One notorious element in Aristotle’s explanation of perception is his comparison of the reception of the perceptible form without the matter to the way in which a piece of wax receives the sign (sêmeion) of a signet ring without the bronze or gold (De an. 2.12, 424a19–​21). As the ring affects the wax, not as brazen or golden but with respect to the sign, similarly the senses or sense-​ organs are affected by the objects that have the colour, flavour or sound, not because of what they are (e.g., a ball, a piece of cake or a flute) but because they are of such and such a quality and in accordance with their logos (ibid., 424a22–​24). As mentioned, a lot has been written about what, if anything, happens in the sense-​organ when the perceptible form is received without the matter, 47

48

Miira Tuominen

the main debate being between a “literalist” interpretation according to which the eye turns red in some sense when we see, say, a red ball, and a “spiritualist” one maintaining that material changes are mere necessary conditions while what it means to receive the form without matter is to become aware of the quality, for example, red. In the following, I shall argue that a material change in the organ is necessary for defining perception, which means that Aristotle’s view should not be read in the spiritualist way. However, there is no need to understand the change in a literalist way either. Although I shall not argue for a particular reading of the material change, I shall conclude that the material change in the organ is important also from the perspective of the notion of information. Despite its name, however, the literalist reading does not necessarily mean that the sense-​ organ should take the quality on in the same way that the object has it, that is, that it would be a change with the matter as, for example, being heated by heat. Instead, the quality is taken on by the organ in a different way or in a different sense in which it is present in the object.30 To illustrate this point, let us imagine me looking at my red handbag. According to Aristotle, the cause of me seeing red is that the surface of the bag affects the transparent medium (illuminated air) between me and the handbag, and the medium affects my eye. If the medium took the quality on in the same way and in the same sense as my handbag has it, an observer that is looking at me and the handbag from the side should see some redness in the air, perhaps in the style that the light swords are coloured in the Star Wars films.31 However, since this is not the case, the medium cannot be affected by the colour in the way in which the bag has it. The literalist can thus claim, accordingly, that our sense-​organ, the eye-​jelly, takes on the quality in a similar way as the medium does, that is, without turning red in exactly the same sense as the object is red. The spiritualists maintain, by contrast, that nothing happens in the sense-​organ except that the creature having that sense-​organ becomes aware of the quality.32 While the opposition between these two views has dominated the discussion, I agree with Victor Caston that these two readings do not exhaust the logical space of options open to Aristotle, but we should rather understand Aristotle’s view in a way that falls between the two readings. Caston explains (2020, 30 n41) the reception by reference to Zenon Pylyshyn’s notion of transduction (1984, 151), referring to the way in which information is transmitted in different kinds of materials by a device that “receives patterns of energy and retransmits them usually in some altered form”. In Pylyshyn’s example (ibid., 151–​152), a computer can be described as a device that transduces the electrical (and computationally irrelevant) events within the machine into computationally relevant ones. Another example could be a device that receives vibrations of air and transduces them into pitch-​specific symbols, for example., a device that when receiving 440 Hz frequency produces the name “a1” on a screen.33 Whether or not one wants to use the notion of transduction to explain the way in which forms are received without matter, I want to stress that the literalist and spiritualist accounts do not exhaust the way in which the reception of forms without matter can be understood in Aristotle. I use Caston’s analysis to show how it is possible to adapt an intermediate view and explain the process without resorting to either spiritualism or literalism and thus to allow a non-​literal material change in the sense-​organ. If we consider Aristotle’s theory of perception with reference to the notion of information, at least the following relevant senses can be distinguished. First, the quality of the object (say, its colour) can be analysed as information and from this perspective Aristotle focuses on explaining how such information can be transmitted to percipients. The question is especially pressing in the case of seeing in which, as we just saw, the transmission through the air does not take place by the air taking on the quality in the fundamentalist literalist way. By contrast, smells are transmitted in that way, since the air between the percipient and the object becomes smelly, that is, takes on the same odour as the object. Second, the percipient’s awareness of the quality 48

49

On information in Aristotle

can also be analysed by using the notion of information: when the percipient has received that information, she is capable of processing it. Third, there is the question of how the information is present in the sense-​organ giving rise to the debate between spiritualist and literalist and the intermediate positions between them. Before considering the intermediate account according to which Aristotle takes there to be non-​literal material change in the sense-​organs in some more detail, one note needs to be added about the notion of information in Aristotle’s biology or theory of nature. As noted, there is an important parallel between Aristotle’s analysis of soul in living creatures and nature as a general causal principle, namely that both are likened to the art and not to the artisan. The lack of an agent corresponding to the artisan in Aristotle’s use of the analogy in nature entails that his biology should not be taken to include a notion of semantic information: nature is not a messenger, an engineer or an administrator. However, there are reasons to allow a notion of semantic information in Aristotle’s theory of perception. In the comparison of perception to the wax receiving the sign of a signet ring, the operative Greek term is sêmeion from which “semantic” is derived. In the De sensu (1, 436b18–​437a15) Aristotle talks about the senses that work at distance (sight, hearing and smell) as reporting (eisangelein) many differences, for example, in colour and shape thus entailing that their function lies in carrying a message from the object to the percipient (Caston, 2020, 26). It is important, however, that although the information that is transmitted in perception can be taken to semantic, it is not linguistic, not at least in the sense of human language. The difference between being significant (sêmantikos 420b32) and being linguistic is related to Aristotle’s view according to which animals with voice are capable of making significant sounds, as opposed to producing mere insignificant noises (De an. 2.8, 420b31–​33). This requires that the sound they make is accompanied with an appearance (phantasia) in their soul. However, animal significant utterances differ from human languages that operate on utterances that correspond to the affections of the soul by convention (De int. 1, 16a3–​4; 2, 16a19). Therefore, to the extent that the reporting is understood as carrying a linguistic message, that talk must be metaphorical. Perception is common to other animals and human beings, and in neither case is the message something like a sentence in a natural human language. In the animal case, there is significant communication based on perceptions without linguistic articulations, while in the human case Aristotle allows propositional articulations of what one is perceiving, but he does not identify perceptions with such articulations. Therefore, we can say that in Aristotle’s theory of perception, there is reception of semantic information that should be distinguished from linguistic semantic information. The point about signification is also emphasised by Victor Caston.34 He analyses the transmission with reference to the notion of transduction, stressing that while the information is received in a transduced form, that is, not literally let alone in a fundamentalist literal sense, it is received with matter as well. In his analysis, this means that the sense-​organ does not take on the quality in the same sense as it appears in the object, in which case he says it would turn into a replica of the quality in the object. Rather, in perceiving an object that has the perceptible quality F (e.g., crimson), it does not become F. However, it must receive another quality, say G, that is F in a transduced form and to receive G with the matter. By becoming G, the sense-​organ transmits the information of F. Caston explains G (that is F in a transduced form) as the proportion (logos) of the two extremes that determine the range of the qualities in which F belongs. This means that when we, for instance, see a pepper that is crimson in colour, which is, say, ratio r 12 between black and white, our eyes are contracted in ratio r 12. Such a contraction is a physical or material change, but it does not mean turning red, that is, the change is neither spiritual nor literal. Although I find Caston’s analysis helpful, it is not my task to consider it here in more detail. Rather, I would like to note the following with respect to an analysis of receiving perceptible 49

50

Miira Tuominen

forms without matter in terms of transmission of information. It is important that there is a material change in the sense-​organ in addition to the mere signal or the semantic information35 for the following reason. Aristotle argues in De anima 3.4 that intellectual understanding (noein)36 is analogous to perception because, while not literally affected by the object, it must be (a) receptive of the form without matter and (b) in the same relation to the intelligibles as perception is to perceptibles (429a13–​18). However, intellection also differs from sense perception because while (429a29–​b5) the senses are incapable of perceiving anything after being stimulated by an object that is highly perceptible, for example, after hearing a very loud noise we are unable to hear anything,37 we are even more capable of understanding something intelligible after grasping something highly intelligible. The argument seems to show that, with respect to information, the reception of semantic information is not sufficient to account for perception. If it were, the argument would have to assume that the signal itself is destroying a transmission device. While a device can of course be destroyed by something that happens in it while it is transmitting information, it is not the signal as semantic information that does so but rather the material realisation of that signal (such as an electric current).38 To add another analogy, to say that the semantic information makes the device unable to function would be like saying that it is the lyrics (such as “Ah, je ris …” in the Jewel Song by Margarita in Gounod’s Faust) of an aria that break a glass when a soprano is singing it, and this does not seem plausible. Therefore, when analysing Aristotle’s theory of perception, we need an account of the material side of what goes on in the sense-​organ39 and cannot simply refer to transmission of semantic information. We saw above that, similarly to how nature is compared to art, so is soul. However, there is an important difference between an art using its instruments and the soul using the body to perceive. While, say, a flute-​player can start playing the flute any time she wishes –​given that the flute is present –​the soul cannot start using the eyes to see, for example, white, on its own accord. Animals that have eye-​lids can open and close them of course but, even in that case, an animal can only see what is in front of its eyes (De an. 2.5, 417a6–​8; b19–​27).40 Because of the view that perception crucially involves reception of forms without matter, there is a tendency in some critics to see Aristotle’s theory of perception as a passive one and to ignore the fact that perception requires active cognitive processing, attention, and so on. However, it is important to note that Aristotle by no means denies that animals and human beings process the information they receive. Rather, the claim about reception is related to the point that in order for us to perceive something, the object needs to be present and we need to receive information from it. Therefore, while, on the one hand, the details of Aristotle’s analysis are controversial, and some authors rather identify perception with propositional judgements about what is being perceived on the other,41 the point that information is received in perception is not too controversial.42 It is quite another question how the information thus received is being processed, how attention influences what we are aware of, what the role of perception in justifying knowledge claims is, and so on. As to the last point, while Aristotle insists on the reliability of the perception of the proper sensibles, he of course allows mistaken perceptual beliefs (e.g., De an. 3.3, 427b20–​21). Therefore, from Aristotle’s assertion concerning the high reliability of the perception of the proper sensibles, there is no direct conclusion to epistemological claims about the reliability of all perceptual beliefs.

Reception of information and acquisition of knowledge In this chapter, I have discussed how we can approach some central themes in Aristotle’s theory of nature or biology and perception from the perspective of the notion of information. I have 50

51

On information in Aristotle

argued that we can articulate a sense in which there is information or data in both cases: non-​ semantic but teleological data in biology and semantic but non-​linguistic information in the theory of perception. I have also suggested that while the notion of transmission of semantic non-​linguistic information is important in Aristotle’s theory of perception, it is by no means an exhaustive account of perceptual cognition. I have suggested that the reception of forms can be taken to cover precisely that part of the theory that is concerned with what can be called the reception of information. This means that Aristotle is not focusing on explaining in what ways the information is processed through general concepts and attention, for example, and how it leads to perceptual beliefs and/​or justifies further knowledge claims. Moreover, discernment is articulated as being at the core of simple perception: perceptible qualities are discerned from others within the range of a sense, perceptibles are discerned from those falling under another sense (such as being yellow from being bitter in De an. 3.1, 425b1–​3), and things can be discerned from other things. In fact, discernment is also understood as crucial for intellection (De an. 3.3, 427a20–​21). This probably refers to the acquired ability to distinguish between members of different species and their different features. While these capacities are crucial for the acquisition of more advanced types of knowledge, they do not directly lead to the knowledge of essences and explanations without systematic inquiry. Therefore, although Aristotle especially underlines the reliability the perceptions of those qualities that fall under one sense (colours perceived by sight, sounds by hearing and so on), their epistemological role is rather modest. It is of utmost importance for the survival of animals and constitutes the starting point for the cognitive development of human beings but produces a very limited kind of knowledge or cognition. Aristotle dedicates no single treatise to explaining how he thinks animals and human beings process the information they receive from the world and especially how they achieve other kinds of knowledge. In the De anima, he talks rather generally about the kinds of cognitive activities that animals and human beings are capable of, but this discussion is not explicitly linked to his views on the acquisition of knowledge. Moreover, the treatise that focuses on the question of what knowledge is like, that is, the Posterior Analytics, concerns a very specific type of knowledge, knowledge without qualification (epistêmê haplôs). It is also notoriously difficult to decipher how exactly Aristotle explains the acquisition of knowledge in that treatise. Knowledge without qualification requires knowing that the cause or explanation is the cause or explanation of the thing and that the thing cannot be otherwise (An. Post. 1.2, 71b10–​12) and in the second book, some guidelines are given about how to arrive at definitions of things and how to select between competing explanations. However, it is not explained how exactly we arrive at perceptual knowledge of things or phenomena ‒ in Aristotle’s terms “that something is the case (to hoti)” or “whether something is” (ei esti) (An. Post., 2.1, 89b24) ‒ that is, to the knowledge from which we can start scientific inquiry from the perception of proper, common and accidental sensibles. Aristotle seems to assume, however, that the ability of our reason to acquire general or universal concepts or notions is vital for us to come to have such knowledge from which we can start more systematic scientific inquiry.43 Finally, the last chapter of Posterior Analytics in which Aristotle promises to explain how the principles of demonstrations become known does not seem to fill its promise. Instead of explaining how exactly we come to know the explanations and definitions of natural things and phenomena, it provides a rather obscure explanation of what seems like an account of how we acquire universal notions from experience. Those notions are then also identified as the starting point of knowledge without qualification (An. Post. 2.19, 100a3–​9). It has recently been argued that the chapter should not even be expected to offer the whole answer to the question of how we come to know the principles of demonstrations but as an explanation of how we begin to 51

52

Miira Tuominen

acquire such knowledge or what its origin is (Bronstein 2016, 229–​230). While I agree with Bronstein that the chapter is largely concerned with the question of how we come to have the kind of knowledge that enables us to begin inquiry, I argue against Bronstein that an account of how we acquire universal notions must precede inductive inference (Tuominen 2019). If universal notions are assumed to be formed through inductive inference, this seems to render the account circular, since inductive inference requires subsuming individuals under general notions.44 Yet even if we take Aristotle to describe a process through which we acquire universal notions, there is a lot of unclarity about the details of that process. At first, the cognitive capacities are named that are responsible for our capacity to acquire knowledge: perception, retention, supposedly phantasia that is not mentioned but probably accounts for experience, and reason (An. Post. 2.19, 99b35–​100a2). These capacities explain our capacity to have perceptions, memories, experience and universals (ibid., 100a3–​6). I have recently argued for a view according to which the capacity to acquire45 universal notions makes a difference in our ability to process information (translating my claim to the language of information here). When we have universal notions at rest in our soul (100a6–​7), we can combine them to (i) claims in which individuals are subsumed under them (“Socrates is of a metabolic type T”; “Socrates suffers from condition C,” and “Socrates is helped by the application of leeches L,” and so on) as well as (ii) claims about general connections between types (e.g., “all patients of metabolic type T have a feature F”, “all patients of type T are helped by the application of leeches”) and (iii) arriving at claims about cause and/​or explanation (“all patients of type T are helped by leeches (L) because of F”). Although the capacity of acquiring universal notions from experience can explain generalisations that enable scientific inquiry into the causes and/​or explanations and essences, it probably already functions in relatively simple perceptual claims such as identifying a person as Diares’ or Cleon’s son (Aristotle’s examples of accidental perception in De an. 2.6 418a20–​23 and 3.1, 425a25–​27).46 It seems that the person in a sense can be identified without general notions (non-​human animals recognise their masters and other familiar people), but forming the claim (i.e., taking it to be the case) that the approaching pale young man is Diares’ son probably requires the notion of being a son, which perhaps is a universal in the category of relative.47 Although the human capacity to form beliefs and combine universal notions into generalisations brings with it many possibilities of error, it also entails a capacity to acquire knowledge that is in Aristotle’s view of a higher kind than the cognition possible for animals. This is because it is about generally valid explanations of things and in Aristotle’s view also about what is necessary, and these matters are a topic of scientific and philosophical inquiry. Although such inquiry in a sense starts from rather simple perceptions, it is not based on perceptions of proper sensibles in any straightforward manner. Therefore, there is no easy overlap between any of Aristotle’s discussions and contemporary epistemology. I have argued here that in contemporary terms the discussion of perception (and to an extent even intellection) concerns reception of information, while the way in which such information is processed and leads to higher kinds of knowledge requires dialectical testing of the credibility of various claims and active scientific inquiry.

Notes 1 Augustine uses the term informatio for the effects of objects in the senses; see De trin. 11.2.3, 336.60–​62 (Mountain): “but that informing of the sense, which is called sight, is imprinted solely by the body that is seen” (illa tamen informatio sensus quae uisio dicitur a solo imprimatur corpore quod uidetur) quoted by Caston (2020, 24 n25). Cicero also uses informatio of Epicurus’ theory of preconceptions thus not confining the term to explanations of theories that assume that there are forms (De nat. deor. 1.43,11–​14). 2 As is the case in Rosenfield’s rather odd approach (1971).

52

53

On information in Aristotle 3 I am using “knowledge” here in a very broad sense; see also Bronstein (2016, 17 n26) who argues against the view that Aristotle accepts false gnôsis and hence “knowledge” is not an appropriate translation for it, see Fine (2010, 148–​52). I agree with Bronstein that the cases Fine refers to do not unquestionably show that Aristotle accepts false gnôsis. I shall return to the question about what “knowledge” in the broad sense entails in the body text shortly below. 4 For criticism of Floridi’s claim, see, for example, Fetcher (2004) and Lundgren (2019). Floridi himself (2011b) defines truth in terms of correctness but many others would rather restrict the notion of truth to propositions or propositional entities and use the notion of correctness more generally. 5 Victor Caston also notes the distinction, and I have considered similar issues from the perspective of descriptive and normative epistemological concerns in Tuominen (2014a). 6 One account of how this can be analysed is found in Caston (2020) that also uses the language of transmission of information. As can be seen below in my discussion on this question in the third section, my perspective is slightly different from his. 7 For early versions of this claim, see, for example, Millikan (1984); Sterelny et al. (1996); Maclaurin (1998); Maynard Smith (2000). For a more recent teleosemantic account of genes, see Shea (2013). 8 For a causal notion of information in genetics, see, for example, Griffiths and Stotz (2013), who say that they ground the notion of information in an interventionist account of causation. 9 For Aristotle on teleology, see Johnson (2005). With respect to nature doing nothing in vain, there is a debate about whether there is one principle (nature does nothing in vain and acts for the best,) or two at work here ((i) nature does nothing in vain and (ii) nature always acts for the best). Although I am more inclined to think that one principle is used in different ways, my intention is not to argue about this point and my discussion does not depend on which reading we adopt. For the claim that there is only one principle, see, for example, Henry (2013, 230) and that there are two (NP and NP*), see Lennox (1997; repr. 2001). See also Leunissen (2010, 130–​131, 154, 171); Huby (1991). Gottlieb and Sober (2017) have also recently argued against Lennox’s analysis of the principle(s). 10 Thomas Johansen has argued, however, that in Plato the demiurge should be compared to the craftsmanship rather than the craftsman (2004, 83–​6). 11 Leunissen (2010, 81–​5); Henry (2013, 230). For arguments against taking nature as an Aristotelian equivalent to Plato’s demiurge, that is, divine craftsman, see, for example, Lennox (1995; repr. 2001, ­chapter 8). 12 This is also noted by Sedley (2010, 16), who otherwise locates Aristotelian teleology controversially close to Plato’s. Devin Henry also articulates nature as an agent (2013, e.g., 251) in an article on the principle that nature does nothing in vain but works for the best. Although I find the article otherwise very helpful, I would resist talking about nature as an agent in Aristotle. This is why I talk about it as a principle, that is, as a source or basis that determines for things how they function or operate. Henry stresses, however, that we should distinguish between nature and rational powers (with reference to Metaphysics Θ‎ 2, 1046a36–​b2). While rational powers involve deliberation, the talk about nature’s deliberation should not be taken in a similar way. Nature produces its effect in a uniform way rather than being a rational power that deliberates about how to act; for references, see Henry (ibid., 260). For a similar point about the craft analogy in the case of soul, see Menn (2002): in Aristotle’s analogy, soul is not the artisan but the art that is using the natural organic (i.e., instrumental) body as an instrument. 13 See also Johnson (2005, 81); Sedley (2010, 13) with reference to Physics 2.8, 199a15–​20. 14 Here I follow Devin Henry’s discussion (2013, 237–​) of the constraints. 15 For recent general or comprehensive accounts, see, for example, Johnson (2005); Leunissen (2010); for more limited accounts, see Leunissen and Gotthelf (2010); Quarantotto (2014). Recently, Jessica Gelber has argued (2018) that we should not understand Aristotle’s distinction between two kinds of teleological cause in terms of aims and beneficiaries as scholars have previously done but rather in terms of directive and instrumental (that she calls “huperetic”) cause for the sake of which. 16 Leunissen (2013). 17 On this point I agree with Devin Henry (2018) who argues against Leunissen (2013). 18 For a similar discussion about the deformed kinds, see Witt (2012) who argues, convincingly, against earlier interpretations of how to understand the so-​called deformities. 19 In this sense, data can be likened to a causal notion of information. The environment makes a difference to plants in another way as well: sunflowers turn towards the sun, and plants react to chemicals and gravity, for instance; today such phenomena are discussed in plant gnosophysiology.

53

54

Miira Tuominen 20 Menn notes, for example, that “it has always been obscure what Aristotle means by this” (2002, 134). Other scholars take the logos here to refer to that proportion (called logos as well) that determines the perceptible quality between two extremes; see Corcilius (2014, 42); Caston (2020, 34–35). I shall return to this question later in the chapter. 21 In Menn’s analysis (2002), the soul uses the body as an instrument but differs from the Platonic analysis precisely because the soul is like the art that uses tools, not the artisan, which would make the soul an immaterial entity and like a homunculus. The analysis of soul using the body as instrument would seem to fit with Gelber’s (2018) reading of the second type of final cause that she calls “huperetic”. It could even perhaps be called “instrumental” rather than “huperetic”, that is, subservient. 22 Aristotle says that aisthêsis is that which is capable of receiving the sensible/​perceptible form without its matter. One of the debated issues is how to understand aisthêsis here. While the term can mean sensation or perception, it can also refer to the senses and even to sense-​organs, although Aristotle also has a separate term (to aisthêtêrion) for them. What is relevant here is that the capacity to perceive means a capacity of such reception and even if we took aisthêsis as sensation or perception as reception, we should not reduce Aristotle’s account of perception to such reception. I shall return to this point in the body text. 23 For a risk of such circularity, see Corcilius (2014, 35); Johansen (2004, 171–​2). 24 See also Corcilius (2014, 41). 25 The medium being transparent means that it must be activated by light. If it is dark, the medium is not transparent and thus cannot transmit the effects of visible bodies. 26 This, however, is compatible with there being a change or affection in the sense-​organ. 27 I take the discernment to mean that an animal is capable of distinguishing between the qualities that fall within the range of one sense modality, for example, sweet from other tastes among gustatory qualities (see also Menn 2002, 133). By this, I do not mean that the object of perception is the difference between the qualities, as Ebert seems to do, but rather that different tastes make a phenomenal difference to the perceiver. Klaus Corcilius argues, to my view, convincingly, against Ebert’s suggestion (2014, 38–​40). I am not sure whether the difference between sub-​personal and personal discernment that Corcilius makes is significant for my discussion. 28 For this claim, see also Corcilius (2014, 38–​40). 29 I have recently argued for this kind of reading of Posterior Analytics 2.19; see Tuominen (2019). Although I disagree with some aspects of David Bronstein’s reading of the chapter (2016), I agree with him that much of the chapter is concerned with explaining how we begin to have knowledge, not so much about how our quest for the principles of scientific demonstrations can be carried out or completed. 30 The literalist reading was first suggested by Richard Sorabji in a footnote (1974, 49 n22). He has also defended the view more recently (1995 [originally published 1992]; 2001); see also Everson (1997). 31 Similar considerations against fundamentalist (Caston 2005, 250) literalism in the medium are found in Alexander of Aphrodisias, De anima (62.5–​10). 32 See Burnyeat (1995, 421) who is the main proponent of spiritualism (Burnyeat 1992; 1995; 2001; 2002), and Thomas Johansen has also defended it (1998). For a discussion of the debate between the literalists and the spiritualists, see Caston (2005). As Caston notes, the contemporary spiritualist claim goes beyond the medieval scholastic readings (ibid., 262). 33 Caston (2020, 30) elaborates on the acoustic example in more detail. 34 Caston notes that there is an analogy to how forms are transmitted in procreation, from the male to the female (2020, 33 n49). While the traditional reading of Aristotle’s theory of procreation maintained that the male provides the form, this has recently been challenged and now the dominating view is that Aristotle rather opts for a theory of epigenesis (Kosman 2010). Kosman notes that (ibid., p. 155) if the form of the animal were provided by the male, cross-​fertilisation would be inconceivable in terms of Aristotle’s theory, and we know that he was well aware of it and often talks about mules, for example. Concerning the cases in which Aristotle says that the male parent provides the form and the female the matter, Kosman suggests that it is the form of the fertilised egg, not the human being that is born (ibid., p. 159). For further discussion about how we should understand the Aristotelian epigenesis, see Henry (2018), who has plausible arguments against Leunissen (2013). 35 For an account that differs from Caston’s but still assumes something to happen in the organ, see also Corcilius (2014, 47 n42). 36 The term noein is notoriously difficult to translate. While Aristotle sometimes uses it more generally to refer to thinking (e.g., De an. 3.4, 429a22–​23), there are narrower uses of the term for which this

54

55

On information in Aristotle translation is not appropriate. Aristotle is certainly not claiming that all thinking is reception of some sort. Rather, there is a form of action in the rational soul that is veridical (i.e., a kind of knowing) that is described as reception and thus as analogous to perception (429a13–​18). Although a kind of knowing, it must be distinguished from such knowledge (epistêmê) that can be called “scientific” and, as also noted above, is about things explained or defined by reference to their real causes and essences. I use “intellectual understanding” (and “understanding” in short) and “intellection” here to refer to the veridicality of such activity and to indicate that it is an activity of nous traditionally (although not unproblematically) translated as “intellect”. 37 Aristotle also adds “or after strong colours”, which means misplaced, since it rather seems that it is a bright light that prevents us from seeing rather than a strong colour as such. For the excessive effect in the organs, see also De an. 3.2, 426a30–​b3. Towards the end of the De anima, Aristotle also argues that in the case of excess in tangible qualities such as heat cold or hardness, the excess not only destroys the mean but in fact the animal itself (3.12, 435b13–​14; 17–​19). 38 Perhaps one could say that it is only in the excessive case that a material effect is taking place in the organ and it is precisely that that destroys the mean, while when the mean is retained, nothing material happens in the organ. While this reading is possible, it does not entail that nothing happens in the organ when it is functioning properly. 39 See also De an. 434b27–​29; 422b3; 426b31 and Phys. 244b10–​11 discussed by Corcilius (2014, 47). On the tension between Aristotle’s denial of alteration in the De anima and the Physics passage as a chronological development, see Menn (2002, 87–​8), who takes Aristotle to deny the Physics account in the De anima (ibid., p. 129). However, the point for Menn is that the soul is not moved, which he clarifies does not exclude the sense-​organ being moved (ibid., 135). For an attempt at a reconciliation of the passages, see Johansen (2012, 160–​6). Corcilius (2014, 43–​7) proposes to solve the tension by reference to sub-​personal discrimination, which means an opposition of a perceptual quality and the mean in the soul that is not the subject of perception. As I said, I am not sure if this distinction is relevant for my discussion. With respect to Caston (2020), let me stress that he underlines the material side of the transmission of information and by no means assimilates the process to the mere transmission of semantic information. 40 Insofar as the animal has memory and phantasia, it can become aware of a perceptible quality in another way, by remembering or imagining it, but if the object is not there, that is of course not perception. 41 In contemporary theory of perception, one central controversy is to what extent perception should be taken as conceptual and/​or propositional as, for example, McDowell (1994, Lecture III and Afterword Part II) or whether there is non-​conceptual perception, for example, Crane (1992), Peacocke (1992, ­Chapter 3); Dretske (1995). 42 “New spiritualists”, as Victor Caston calls Burnyeat’s position (2005, 262), for instance, would not seem to accept this, since, according to them, “when Aristotle sees a colour or hears a sound, nothing happens save that he sees the colour or hears the sound” (Burnyeat 1995, 421) and we, as it were, see through our eyes. Note, however, that even sceptics proposing a brain-​in-​the-​vat scenario allow stimulation of the brain by the manipulative scientist, or those subject to the Cartesian doubt allow the influence of a bad demon. Some late ancient Platonist commentators on the De anima, while identifying perception with a judgement of reason take over an Aristotelian account of the reception of forms (Tuominen 2014b). 43 Although earlier in scholarship reason and perception –​or empiricism and rationalism –​were understood as opposing poles (e.g., Barnes (1993, 259) who detects Janus-​faced deception in the last chapter of the Posterior Analytics when identifying both perception and reason or intellect as crucial in the acquisition of the knowledge of the principles of demonstrations), more recently scholars have rather emphasised how they work together; see Gregoric and Grgic (2006); Tuominen (2019). 44 I also base my disagreement with Bronstein on this on the way in which Aristotle formulates his point in the relevant passage (esp. in 100a3–​9); see Tuominen (2019). Although I have to an extent changed my earlier view (Tuominen 2010b), I do not think my thought is spoiled by including the possibility of forming general notions later (cf. Bronstein 2016, 229 n7). This is because, in my view, Aristotle moves rather easily between the reading that principles are concepts or notions and that they are propositional. My new view (Tuominen 2019) also allows that general notions can be formed in all stages of inquiry. However, as opposed to the early stages of the process in which the notions are formed through perceptual experience alone, later they can be formed in a process of inquiry including inference and through posing scientific questions about explanation and about whether some properties are essential or not, that is, through rational processing of information partly received in experience.

55

56

Miira Tuominen 45 There is a sense in which Aristotle indicates our intellect is receptive of universal notions (intelligible objects; see De an. 3.4, 429a15–​18). The claim has caused much puzzlement and is often connected to assumptions that the productive nous somehow takes part in their reception. I have analysed the assumption in Alexander’s De anima (Tuominen 2010a), and Börje Bydén discusses the assumption with respect to the light analogy in 3.5 and in Alexander in a forthcoming article. See also Robert Roreitner (diss.). For incidental perception in Aristotle, see Mika Perälä (forthcoming). 46 For an account of Aristotle on accidental perception, see Perälä (forthcoming). 47 For the notion of father as a universal, see another difficult passage in Physics 1.1, 184a24–​b14.

References Adriaans, P. (2018) “Information”, The Stanford Encyclopedia of Philosophy (Spring 2019 Edition), E.N. Zalta (ed.). https://​plato.stanford.edu/​archives/​spr2019/​entries/​information/​. Barnes, J. (1993) Aristotle: Posterior Analytics, Oxford: Oxford University Press. Bronstein, D. (2016) Aristotle on Knowledge and Learning: The Posterior Analytics (Oxford Aristotle Studies), Oxford: Oxford University Press. Burgin, M. (2010) Theory of Information: Fundamentality, Diversity and Unification (World Scientific Series in Information Studies, 1), New Jersey/​London: World Scientific. Burnyeat, M. F. (1995; originally published in 1992) “Is an Aristotelian Philosophy of Mind Still Credible?”, in M. C. Nussbaum and A. O. Rorty (eds.), Essays on Aristotle's De anima (Clarendon Aristotle Series), Oxford: Oxford University Press, 15–​26. Burnyeat, M. F. (1995) “How Much Happens when Aristotle Sees Red and Hears Middle C? Remarks on De anima 2.7–​8”, in M. C. Nussbaum and A. O. Rorty (eds.), Essays on Aristotle's De anima (Clarendon Aristotle Series), Oxford: Oxford University Press, 421–​434. Burnyeat, M. F. (2001) “Aquinas on ‘Spiritual Change’ in Perception”, in D. Perler (ed.), Ancient and Medieval Theories of Intentionality, Leiden: Brill, 129–​153. Burnyeat, M. F. (2002) “De anima II 5”, Phronesis 47, 28–​90. Bydén, B. (forthcoming) ‘Aristotle’s Light Analogy and the Greek Tradition’. Caston, V. (2005) “The Spirit and the Letter: Aristotle on Perception”, in R. Salles (ed.), Metaphysics, Soul, and Ethics: Themes from the Work of Richard Sorabji, Oxford: Oxford University Press, 245–​320. Caston, V. (2020). “Aristotle on the Transmission of Information: Receiving Form without the Matter”, in Bennett and Toivanen (eds.), Philosophical Problems in Sense Perception: Testing the Limits of Aristotelianism, Cham: Springer, 15–55. Charles, D. (ed.) (2000) Definition in Greek Philosophy, Oxford/​New York: Oxford University Press. Corcilius, K. (2014) “Activity, Passivity and Perceptual Discrimination in Aristotle”, in J. F. Silva and M. Yrjönsuuri (eds.), Active Perception in the History of Philosophy (Studies in the History of Philosophy of Mind 14), Dordrecht: Springer, 31–​53. Crane, T. (1992) “The Nonconceptual Content of Experience”, in Crane (ed.), The Content of Experience: Essays on Perception, New York: Cambridge University Press, 136–​157. Dretske, F. (1995) Naturalizing the Mind, Cambridge MA: MIT Press. Everson, S. (1997) Aristotle on Perception, Oxford: Clarendon Press. Fetzer, J. H. (2004) “Information: Does it Have to Be True?” Minds and Machines 14(2), 223–​229. Fine, G. (2010) “Aristotle on Knowledge”, Elenchos 14, 121–​156. Floridi, L. (2010) Information: A Very Short Introduction, Oxford/​New York: Oxford University Press. Floridi, L. (2011a) The Philosophy of Information, Oxford/​New York: Oxford University Press. Floridi, L. (2011b) “Semantic Information and the Correctness Theory of Truth”, Erkenntnis 74(2), 147–​175. Floridi, L. (2015) “Semantic Conceptions of Information”, The Stanford Encyclopedia of Philosophy (Spring 2017 Edition), E.N. Zalta (ed.), https://​plato.stanford.edu/​archives/​spr2017/​entries/​information-​ semantic/​. Gelber, J. (2018) “Two Ways of Being for the Sake of an End”, Phronesis 63(1), 64–​86. Gotthelf, A. (2010) Teleology, First Principles and Scientific Method in Aristotle’s Biology (Oxford Aristotle Studies), Oxford: Oxford University Press. Gottlieb, P. and E. Sober (2017) “Aristotle on ‘Nature Does Nothing in Vain’ ”, HOPOS: The Journal of the International Society for the History of Philosophy of Science 7, 246–​271.

56

57

On information in Aristotle Gregoric, P. and Grgic, F. (2006) “Aristotle’s Notion of Experience”, Archiv für Geschichte der Philosophie 88, 1–​30. Griffiths, P. and K. Stotz (2013) Genetics and Philosophy: An Introduction, Cambridge: Cambridge University Press. Günther, H.-​C. and Rengakos, A. (eds.) (1997) Beiträge zur Antiken Philosophie: Festschrift für Wolfgang Kullman. Stuttgart: Franz Steiner Verlag. Henry, D. (2007) “How Sexist is Aristotle’s Developmental Biology?”, Phronesis 52, 251–​269. Henry, D. (2013) “Optimality Reasoning in Aristotle’s Natural Teleology”, Oxford Studies in Ancient Philosophy 45, 225–​263. Henry, D. (2018) “Aristotle on EpigenesisTwo Senses of Epigenesis”, in A. Falcon and D. Lefebvre (eds.), Aristotle's Generation of Animals: A Critical Guide, Cambridge: Cambridge University Press, 83–​107. Huby, P. (1991) “What Did Aristotle Mean by ‘Nature Does Nothing in Vain’?”, in I. Mahalingam and B. Carr (eds.), Logical Foundations: Essays in Honor of D. J. O’Connor, London: Palgrave MacMillan, 158–​166. Johansen, T. (1998) Aristotle on the Sense-​Organs, Cambridge: Cambridge University Press. Johansen, T. (2004) Plato’s Natural Philosophy: A Study of the Timaeus-​Critias, Cambridge: Cambridge University Press. Johnson, M. R. (2005) Aristotle on Teleology, Oxford: Clarendon Press. Kosman, A. (2010) “Male and Female in Aristotle’s Generation of Animals”, in J. Lennox and R. Bolton (eds.), Being, Nature and Life: Essays in Honor of Allan Gotthelf, Cambridge: Cambridge University Press, 147–​167. Lennox, J. (1996) “Material and Formal Natures in Aristotle’s De Partibus Animalium”, in J. Cleary and W. Wians (eds.) Proceedings of the Boston Area Colloquium 11(1), Leiden: Brill, 217–​240. Lennox, J. (1997) “Nature Does Nothing in Vain …”, in H.-​C. Günther and A. Rengakos (eds.) (1997) Beiträge zur Antiken Philosophie: Festschrift für Wolfgang Kullman. Stuttgart: Franz Steiner Verlag, 199–​214. Lennox, J. (2001) Aristotle’s Philosophy of Biology: Studies in the Origins of Life Sciences, Cambridge: Cambridge University Press. Lennox, J. and R. Bolton (eds.) (2010) Being, Nature and Life: Essays in Honor of Allan Gotthelf, Cambridge: Cambridge University Press. Leunissen, M. (2010) Explanation and Teleology in Aristotle’s Nature of Science, Cambridge: Cambridge University Press. Leunissen, M. (2013) “ ‘Crafting Natures’: Aristotle on Animal Design”, Philosophic Exchange 41(1), 28–​51. Leunissen, M. (2015) “Introduction”, in M. Leunissen (ed.), Aristotle’s Physics: A Critical Guide (Cambridge Critical Guides) Cambridge: Cambridge University Press, 1–​9. Leunissen, M. and A. Gotthelf (2010) “What’s Teleology Got to Do with It? A Reinterpretation of Generation of Animals V”, Phronesis 55, 325–​356. Lindén, J-​I. (ed.) (2019) Aristotle on Logic and Nature, Leuven: Peeters. Lundgren, B. (2019) “Does Semantic Information Need to Be Truthful?” Synthese, 196/​7, 2885–​2906. Maynard Smith, J. (2000) “The Concept of Information in Biology”, Philosophy of Science 67(2): 177–​194. McDowell, J. (1994) Mind and World, Cambridge, MA: Harvard University Press. Maclaurin, J. (1998) “Reinventing Molecular Weissmanism”, Biology and Philosophy 13(1): 37–​59. Menn, S. (2002). “Aristotle’s Definition of the Soul and the Programme of the De Anima”, Oxford Studies in Ancient Philosophy XXII, 83–​139. Millikan, R. G. (1984) Language, Thought and Other Biological Categories, Cambridge, MA: MIT Press. Nussbaum, M. C. and Amelie Oksenberg Rorty (eds.) (1992) Essays on Aristotle's De anima (Clarendon Aristotle Series), Oxford: Oxford University Press. Peacocke, C. (1992) A Study of Concepts, Cambridge, MA: MIT Press. Perler, D. (ed.) (2001) Ancient and Medieval Theories of Intentionality, Leiden: Brill. Politis, V. (2010) “Explanation and Essence in Plato’s Phaedo”, in D. Charles (ed.), Definition in Greek Philosophy, Oxford/​New York: Oxford University Press, 62–​114. Pylyshyn, Z. (1984) Computation and Cognition: Toward a Foundation for Cognitive Science, Cambridge, MA: MIT Press. Quarantotto, D. (2014) “Aristotle’s Natural Teleology Seen from Above: A ‘Cosmogony’ of the Means-​ Goal Relation”, in M. Tuominen, S. Heinämaa and V. Mäkinen (eds.) (2014) New Perspectives on Aristotelianism and Its Critics (Brill’s Studies in Intellectual History 233), Leiden: Brill. 31–​49. Roreitner, R. (diss.) Passive Intellect in Aristotle and Commentators, Humboldt-​Universität-​zu Berlin.

57

58

Miira Tuominen Rosenfield, L. W. (1971) Aristotle and Information Theory: A Comparison of the Influence of Causal Assumptions on Two Theories of Communication, The Hague: Mouton. Salles, R. (ed.) (2005) Metaphysics, Soul, and Ethics: Themes from the Work of Richard Sorabji, Oxford: Oxford University Press. Sedley, D. N. (2007) Creationism and its Critics in Antiquity, Berkeley: University of California Press. Sedley, D. N. (2010) “Teleology: Aristotelian and Platonic”, in J. Lennox and R. Bolton (eds.), Being, Nature and Life: Essays in Honor of Allan Gotthelf, Cambridge: Cambridge University Press, 5–​29. Shea, N. (2013) “Inherited Representations are Read in Development”, British Journal for the Philosophy of Science 64(1): 1–​31. Silva, J. F. and M. Yrjönsuuri (eds.) (2014) Active Perception in the History of Philosophy (Studies in the History of Philosophy of Mind 14), Dordrecht: Springer. Sorabji, R. (1974) “Body and Soul in Aristotle”, Philosophy 49, 63–​89. Sorabji, R. (1995; originally published in 1992) “Intentionality and Physiological Processes: Aristotle’s Theory of Sense-​Perception”, in M. C. Nussbaum and A. O. Rorty (eds.), Essays on Aristotle's De anima (Clarendon Aristotle Series), Oxford: Oxford University Press, 195–​225. Sorabji, R. (2001) “Aristotle on Sensory Processes and Intentionality: A Reply to Myles Burnyeat”, in D. Perler (ed.), Ancient and Medieval Theories of Intentionality, Leiden: Brill, 49–​61. Sterelny, K., K. C. Smith and M. Dickison (1996) “The Extended Replicator”, Biology and Philosophy, 11: 377–​403. Tuominen, M. (2010a) “Receptive Reason: Alexander of Aphrodisias on Material Intellect”, Phronesis 55, 170–​190. Tuominen, M. (2010b) “Back to Posterior Analytics II 19: Aristotle on the Knowledge of Principles”, in J. H. Lesher (ed.), From Inquiry to Demonstrative Knowledge: New Essays on Aristotle’s Posterior Analytics, Apeiron 43/​2–​3, 115–​143. Tuominen, M. (2014a) “Naturalised versus Normative Epistemology: An Aristotelian Alternative”, in M. Tuominen, S. Heinämaa, and V. Mäkinen (eds.), New Perspectives on Aristotelianism and Its Critics (Brill’s Studies in Intellectual History 233), Leiden: Brill, 66–​91. Tuominen, M. (2014b) “On Activity and Passivity in Perception: Aristotle, Philoponus, and Pseudo-​ Simplicius”, in J. F. Silva and M. Yrjönsuuri (eds.), Active Perception in the History of Philosophy (Studies in the History of Philosophy of Mind 14), Dordrecht: Springer, 55–​78. Tuominen, M. (2019) “Reason, Experience and the Knowledge of the Principles in Aristotle’s Posterior Analytics 2.19”, in J.-​I. Lindén (ed.), Aristotle on Logic and Nature, Leuven: Peeters, 103–​138. Tuominen, M., S. Heinämaa and V. Mäkinen (eds.) (2014) New Perspectives on Aristotelianism and Its Critics (Brill’s Studies in Intellectual History 233), Leiden: Brill. Witt, C. (2012) “Aristotle on Deformed Kinds”, Oxford Studies in Ancient Philosophy 43, 83–​106.

58

59

4 INFORMATION AND HISTORY OF PSYCHIATRY The case of the disease phrenitis Chiara Thumiger Medicine, and in particular the history of medical labels and taxonomies, is an intriguing area to explore for an inquiry into the nature of scientific information. Organization of data and transmission of knowledge in the field of medicine, in fact, rest importantly on the issues of definition and classification, central in the case on which I propose to focus: the construction and transmission of the ancient mental disease phrenitis, from ancient Greek testimonies through to the modern history of medicine and psychiatry. The history of nosological classifications pose well-​known conundrums. The topic touches on central discussions in history of medicine and psychiatry,1 not least because of the hotly debated classifications of mental disorders regularly updated by the American Psychiatric Association (the DSM, now in its fifth edition), or those of its competing system, the ICD produced by the World Health Organization.2 Disagreement hits at various levels: what is a disease, a nosological entity? What is ‘mental’ in a mental illness? And, in combination: what is a mental disease, if such a thing exists? These are problematic issues for the clinical operations of doctors and psychologists in contemporary, highly institutionalized Western contexts; the problems are only multiplied when we consider psychiatric classification from a philosophical perspective,3 and from a historical-​anthropological one. To the epistemological problems of classification per se we then have to add the shifting parameters of different scientific cultures, definitions of health, notions of mind and body and technical vs. lay culture vis-​à-​vis medicine and psychiatry in particular.4 In this chapter I propose to look at an example from ancient Greco-​Roman medicine, the disease label phrenitis, whose life stretches from the fifth century BCE to the nineteenth century of our era. This example is historical—​the development, adaptation and survival of a medical concept; at the same time, I intend to present it as epistemological and mediatic case study. As a rare case of a mental disease which will continue being described, diagnosed and cured by (variously defined) professionals over twenty-four centuries, in fact, phrenitis begs the question of its own persistence in terms of communication, as object of information. As such, I am going to consider phrenitis in this chapter: leaving aside questions of treatment and clinics, of patient perspective, of physiology and retrospective diagnosis, I shall tackle instead the following question: which elements, in the presentations of this disease by physicians, functioned as vehicle for its durability through different epistemologies and models of the body, making it,

59

60

Chiara Thumiger

to use Berrios’ label, ‘ontologically robust’ enough to make this possible?5 How did the body of knowledge about phrenitis acquired through the centuries succeed in preserving and sharing such continuity across various communities and audiences? My answer is composite, and involves different aspects of the history of the disease phrenitis. A first, fundamental point is its having a quality of conspicuousness and traditionalism, to the ears of its audience for the first thousand years of its existence at least, already in the letter of its name: phren-​itis, or, the ‘disease/​affection/​inflammation of the phrenes’—​a very common and loaded ancient bodily and psychological term. Second, there is its being strongly, but flexibly embodied—​I shall explain this in a moment. Third, there is the fact that phrenitis incarnates a core ambiguity and rivalry in the history of Western philosophy, namely that between encephalocentrism and cardiocentrism—​the attribution of the centre of the mental faculties to the brain or to the heart respectively (with various qualifications). Finally, its becoming, at some point of its cultural history, a metaphor for a certain kind of human flaw, of ethical and cognitive deficiency, and identifying a sort of ‘human folly’, a condition between sin, ignorance and damnation that will be regularly referred to in prudential and theological literatures from the late antique period to the high middle age.6 Its status consolidated by these four lines of information, each with its mediatic reach (lay, professional, scientific-​historical, religious and pastoral), phrenitis will enter the domain of modern science, fitting comfortably within modern discussions of the anatomy of the brain and its pathology and will be identified, most univocally from the seventeenth century onwards, with forms of encephalitis and meningitis. I shall now look in more detail at each of these four itineraries, before drawing some broader conclusions about the life and death of diseases within medical traditions—​that is, not epidemiologically or biomedically, but as labels—​in terms of shared and transmitted information. With a broad paradigm for cultural historical changes in mind, Ineke Sluiter has offered the formula ‘anchoring innovation’ with reference to classical antiquity: the process through which, by nesting the old and traditional within a changing present, innovation is made acceptable and integrated within the personal and social context of its receivers.7 Ancient medicine, unfortunately, has not lent content to these discussions, which is understandable: history of science, in fact, appears to follow its own rules vis-​à-​vis innovation and conservation, and to pose its own set of obstacles, which have long been theoretically discussed.8 Mindful of the specific problems the evolution of medical concepts imply, I propose an account of phrenitis which can be also seen as a story of successive anchoring moves—​or rather, if the anchoring image retains perhaps too much of a suggestion of intentional agency9 or political strategy for pre-modern medical contexts, as a story of successful anchorage of a disease concept at the conjuncture of history of science, popular and religious paradigms and clinical medicine in Western history.

The label phrenitis and Greek traditional discourses about the mind In a traditional, oral context such as early Greek culture a meaningfulness of the name is key to understand the mediatic appeal of a disease concept. This is different in our contemporary (first-​world, official) medical culture in which defamiliarizing, often hermetic terms are rather the marker of technical language. In this respect phrenitis is particularly interesting. As a denominative from the root phr-​, its meaning is immediately understood (like, like hepa-​titis from hepar, nephr-​itis from nephros, and so on) as an affection, or inflammation, or a damage to the phren-​ phrenes. This term designates in Greek both a bodily location and, in an abstract/​metaphorical sense, one’s ‘mind’ or ‘self ’.10 Anatomically, phren-​phrenes indicate a location or bodily part in the torso; which one in particular, is a detail in continuous evolution in the earlier centuries

60

61

Information and history of psychiatry

of attested Greek literature (from the Homeric poems to the fourth century BCE), on which scholarly opinions have differed: in some sources, the diaphragm, the sheet of muscle between the thorax and the abdomen, is clearly meant; in some others, a general position in the chest, or the upper cavity of the torso seem to be indicated. This range of meanings is evident when one looks at the many occurrences from as early as the Homeric poems or tragedy, where the significance of the term is entirely psychological: ‘mind’, ‘character’, ‘soul’, ‘self ’.11 This range of poetic sources suggests that any audience would immediately interpret the disease label phren-​itis as localized disease; as a disease with a mental quality; and with a connection with the chest area—​to a modern ear, sounding like something in between ‘mind-​itis’ and ‘chest-​itis’. This lay appeal of the technical nosological construct at the time of its first attestation in the Hippocratic Corpus, in the fifth century BCE, already nails a semantic characteristic of phrenitis that is unusual for modern medical labels, but that obviously constitutes a first element of its capability to persist through time: its immediacy.12 Immediacy, however, does not mean that phrenitis was, for the coeval audiences, a vernacular, everyday term (such as, e.g., ‘hay-​fever’ is to English speakers). Interestingly, the early medical authors seem to avoid direct reference to the traditional connection between phrenes and mind in the explicit discussions of the disease. Consider Diseases 3.9 (Potter 76.20–​29 = 7.128.5 L.), where the medical author is discussing the possible onset of forms of the disease. The mental component is important but reference to phrenes is made only in the sense of a bodily localization, not of the mental organ: phrenitis … Patients suffer as follows: they experience such pain in the phrenes that they will not allow themselves to be touched, there is fever, they are deranged (ekphrones), they stare fixedly, and for the rest they resemble patients with peripleumonia who are deranged (ekphrones). Likewise, reference to a pain in the chest area where phrenes reside is also present in the discussion of phrenitis in Affections 10, but using alternative terminology, hypochondria: In phrenitis, at first there are mild fever and pain over the hypochondria, more on the right towards the liver. When the fourth or fifth day arrives, the fever becomes more intense, as do the pains, the colour becomes somewhat bilious, and the patient’s mind becomes deranged. Later in the same passage the phrenes are mentioned in a way that appear synonymous to the locations just indicated: at Affections 10 (Potter 19 = L.6.218.9) phrenitis, it is added, is caused by an overflow of bile ‘into the internal organs and phrenes’. My suggestion, based on these and other similar passages, is that in discussions of phrenitis the medical authors entertained a double relationship with the lay traditions and cultural contexts they belonged to. They exploited the etymological burden of the disease name, but were in turn constrained and conditioned by it. On the one hand, they rooted their accounts of the disease in the chest as general locus, following a traditional representation, and gave the disease a name of clear significance in this sense. But, on the other, they avoided explicit use of the phrenes as mental organ, or as indicating ‘mind’, as if to distance themselves from poetic and popular representations, even in the face of the evident etymology.13 This mechanism is a perspicuous case of anchoring: you have the diaphragmatic location; the phren-​ etymology; the

61

62

Chiara Thumiger

mental character, but these elements are shuffled and recombined, so that the identification of the mental phren with a place in the body is excided and localization of the mind is left undiscussed. In this way, the language used is familiar, the building blocks are the same, but the edifice raised is an entirely new one.

The body of the phrenitic patient A mental disease is, first and foremost, a story. The way this story is told through a period as long as the one we are examining—​two millennia and a half—​can vary in consistency and ‘robustness’, as we anticipated. In the case of a mental disease, the nailing down of a bodily, physiological core in its description is a key part of the disease’s ability to survive consistently and recognizably despite the changing medical premises and scientific-​philosophical cultures. As Plutarch suggestively put it, offering an early instance of the ‘anchoring’ image, the human body itself is guarantee to the persistence of diseases: How then should the body be subject to new diseases, since it has not, like the soul, the principle of its own alteration in itself, but by common causes is joined to Nature, and receives a temperament (krasis) whose infinite variety of alterations is confined to certain bounds, like a ship rolling and tossing in a circle about its anchor?14 The discussion of phrenitis is from its early beginnings in classical medicine strongly embodied, which means that it displays a fixed range of symptoms: fever, first of all. The presence of hot fever is fundamental to its categorization as acute (i.e., swift, serious and critical disease, according to the ancient sub-​division in acute and chronic),15 and is directly related to much of its symptomatology and therapeutics. The emphasis on this aspect is especially evident in the earlier Hippocratic sources where the disease is grouped together with peripleumonia, pleuritis and other so-​called ‘ardent fevers’ (kausoi).16 In addition, there are the many signs brought in by overheating: the dryness and roughness of the tongue these authors repeatedly mention in connection with phrenitis (as, for example, at Coan Prenotions 229 (Potter 159 = L. 5.634.17), ‘rough, very dry [tongues] indicate phrenitis’) which directly depends on the rise in temperature; a pathological thirst; shivers, chills, and sweating. The importance of fever is such that phrenitis could be also seen, in the early stages of its history, to be a possible complication, or a development of ardent fever (kausos), as well as exemplary among clear-​cut cases of fevers in general (puretoi). For instance, within a class of individuals suffering from eye symptoms a moderate heat is a phrenitic sign: ‘patients who are not burning hot to the touch develop phrenitis’ (Coac. 223, Potter 158.1–​2 = L. 5.633). At Morb. 1.30 (Wittern 86–​7=L. 6.201) the entire description of the disease is constructed around the heating of the patient’s blood, and his or her body as a consequence, causing a formidable fever that is responsible for the mental derangement. There are then the visible elements of the external appearance of these patients, especially their eyes and glance—​the phrenitic have fixed and staring eyes. They may suffer from persistent headaches and painful or stiff neck. Their sleep is troubled or there is sleeplessness, or more generally a comatose state, as in Epidemics 3.11 (88.1–​5 Jouanna = 3.90–​2 L.): coma attended mostly the cases of phrenitis and ardent fevers, without excluding, however, all the other diseases of the most severe sort that were accompanied by fever. Most patients throughout either were sunk in heavy coma or slept only in fitful snatches.17 62

63

Information and history of psychiatry

Most striking for a modern reader is a behavioural sign that is systematically reported in these patients: the so-​called crocydism or floccillation. This is a well-​known characteristic of our disease, and a visually intriguing detail: a kind of hand compulsion well known in medicine as a neurological fever-​related (‘thyphoid’) symptom, as described in this exemplary case: the sister of Hippias, in the winter, was phrenitic; regimen bad; very busy with her hands, lacerating herself on the fifth day. On the sixth day, towards night, speechless, comatose, puffing into her jaws and lips like sleeping people. She died towards the seventh day. (Epidemics 7.53, Jouanna 84.21–​25 = 5.422 L.) This sign18 is described in multiple cases of high fever by Hippocratic medicine, and remains as important part of the embodied repertoire of our disease as paradigmatic fever. Its reshaping through the tradition offers us a good example of the evolution in communication about the disease: while in the earlier sources floccillation is described but left without explanation, as a fever-​related disturbance and a marker of severity or impending death it will later develop to fit new explanatory frames. This is clearly seen in this Galenic passage: Likewise, [hallucination] happens also in acute fevers and in inflammation of the lungs, when the humours in the body rise as vapours to the head, that the clear fluid around the pupil shares in their exhalation. And wherever and in whatever way it is made turbid, the aforesaid images are generated. But in violent headaches, just as also in cases of phrenitis, because the head becomes full, and some part of the humours reaches the eyes, this causes the same symptoms. And ‘picking at loose flocks’ and ‘picking at things’, verbs that are habitually used by all doctors, especially for those suffering from phrenitis, have acquired their meaning from the following. Some people have described the image of flocks (krokudon) and of chaff, both while this was actually happening and after it, recalling it later. The difference between these patients is twofold: for some became delirious whilst having these very sensory images while their critical capacity remained vigorous, but others also, through exceptionally strong intellectual activity, although they had been affected for a short while, nevertheless held on to their critical capacity and recovered so as to overcome the delirium and fight it off, and understand what had happened. (Galen, Commentary to the Hippocratic Prognosticon I.23, 237.8—​Heeg (= XVIIIB 74 K.) Transl. by C. Salazar) Galen takes a traditional sign related to fever and phrenitis, and inserts this embodied datum into his own explanatory frame—​encephalocentric and humoural, with a cognitive theory attached: vapors rise to the head and obfuscate the eyes, hence the hallucinatory experience which, for him, is behind floccillation. Two more embodied features recur constantly with phrenitis: the quality of urine and the pulse. The urine of phrenitic patients is described as pale, and with a suspension in it,19 by the Hippocratics as well as later authors. The second element will instead only emerge after the Hellenistic era, in concomitance with the rise of concepts of fluid movement through vessels: namely, the idea of the existence of a specific pulse for phrenitic patients. ‛Pulse doubled, small, thick’ (Anonymus Parisinus 1.2.1 (3 Garofalo); in greater detail: the pulse of the phrenitic is low/​small; however most rarely it may appear big, and it has moderate tone and it is hard, nervy, and too thick and fast. But it has also 63

64

Chiara Thumiger

something wavy; sometimes it will be felt by you as trembling; but at other times to be spasmodically intermittent. (Galen, On the Causes of Pulses IV, 9.184, 8 K.) Again, ‘the pulse of the phrenitic is short and vigorous because of the continuous motion of the breath due to the lack of sleep’ (Rufus, Syn. de Pulsibus, 6.2.1–​3.1, 1–​2 CE); and so on. That of the pulse is also a telling example of the embodied sign being a vehicle for change as well as continuity in a disease’s history: as a new important system or medical idea enters the horizon of ancient medicine (e.g., the pulse), a set of strong observations are made to fit the older disease that now gains its own profiling vis-​à-​vis, in this case, pulsation. Both a continuous adaptation to evolving models of the body and ideas about pathology, then, and a resistant core of embodied signs that substantiate this disease, combining traditional traits and innovation with a strategic result. If we take a leap of a thousand years after the Hippocratic texts and read the summarizing description of our disease offered by Caelius Aurelianus, a medical author from the fifth century CE, the same key points emerge: we recognise phrenitis from the combination of all the signs. For any single sign, e.g. mental derangement or fever, does not indicate phrenitis; but the case is otherwise if many signs concur which together can indicate only this disease. In this case an indication is obtained, as we have said, from many circumstances, and constitutes a single sign indicative of the situation. We therefore recognise phrenitis, as I said, from the combination of acute fever, mental derangement, small and rapid pulse, and the plucking of straws and hairs: for it is on the basis of these that the kind of disease is recognised.20 Of course, one may note, per definitionem, a disease that persists through the ages must display a consistent, if not an entirely stable set of characteristics. This is the case, for example, for mania, for hysterical suffocation, for melancholy, to offer three famous instances. It would be very hard for an historian, however, to mention for any of these diseases a repertoire of physical signs as hardwired and as consistent as those we have seen above. The names of these diseases survive but their physiologies and causations are disparate, their descriptions so different from one era to the next that only aurally do prior versions survive under the umbrella of the identical name. Phrenitis, instead, is both very codified in its corporeal and physiological features, and open to inclusion of new scientific paradigms, discoveries and modes, its status both continuously updated and rooted deep in tradition.21

Head or heart? Phrenitis at the crossroad between encephalocentrism and cardiocentrism From the point of view of intellectual and philosophical history, phrenitis also held a unique advantage that made it of enduring appeal in history of science. This is its positioning vis-​à-​vis a key bifurcation in ancient biology, medicine and philosophy but also lay perceptions about the human body: that between positing the centre of human mental, animated and spiritual life in the head or brain (as in the Hippocratic Sacred Disease, in Plato’s Timaeus, and in a series of medical thinkers, notably from Galen onwards) and centring it in the chest, especially in or around the heart (the Aristotelian and Stoic views22). Complications and refinements to these two major alternatives were of course offered in the course of Western intellectual history;23 what interests us here is the open advertising of the issue in the case of phrenitis, where 64

65

Information and history of psychiatry

the oscillation between chest and head, inner organs and brain is nested in the name label for starts—​the phren which is both abstract word for self and mind, and concrete body part—​and returns again and again in the medical discussions of the disease, as well as being possibly present in its lay understanding.24 In the earlier testimonies, those from the Hippocratic Corpus, the explicit references to localization made are to the chest and diaphragm, or to a hypochondriac region: for instance, a discussion of phrenitis in On the Diet in Acute Diseases (23, 46.3–​5 Joly =2.274 L.) is the occasion to say that ‘in cases when pain fixates beneath the phrenes, but does not appear in the shoulder, it is necessary to sooth the intestines, with black hellebore or purslane’; at Coan Prenotions 405 (5.676.7 L.) we read that ‘if persons with pain in the side, who do not have pleurisy, evacuate favourable thin stools, they will turn out to be phrenitic’. The emphasis on expectoration with reference to mental disturbance, and in phrenitic cases in particular, is significant in this sense, as it associates the upper cavity of the body with mental life: in Prorrheticon 1.6 (75.11–​12= 5.512 L.) we read that ‘frequent expectoration, if some other sign is present as well, indicates phrenitis’. Examples of this kind are many. In parallel to these remarks about the chest, it is interesting for us to notice how the same sources include, at the same time, also reference to a pathology located in the head as far as phrenitis is concerned. For example, in a patient case described in the Epidemics we find that ‘when the pain in his head did not stop, a sternutatory was applied on the tenth day. Afterward he had a severe pain into the neck … his mind was unsound in a phrenitic manner’.25 Or, ‘her urine was shaggy for a long time and she had headaches. She became phrenitic and died with powerful convulsions’;26 ‘the phrenitic … painful heaviness of the head and neck’.27 The two locations, head and chest, are easily combined and with no sense of contradiction, as evident in the following: ‘headache in acute fevers, tension in the hypochondrion, unless there is a hemorrhage through the nostrils, develops into phrenitis’.28 These references to an involvement of the head as location, in fact, or as position in the body—​not yet to the brain as affected part—​does not result in an etiological statement of an encephalocentric kind before the Hellenistic era: in the classical sources a twofold discourse is maintained as unproblematic. In the Hellenistic period and after, with the crucial development of the anatomical inquiries about the brain and nerves, as well as of an anatomical approach to the body more generally, a clear interpretation of the disease as inflammation of the brain will appear to compete with the chest-​based account. The great Alexandrine physician Erasistratus (third century BCE) is among the first said to claim in so many words that the cause of phrenitis is encephalic: [phrenitis occurs when] ‛the activities of the [cerebral] membrane are affected; for at the place where, according to him, thinking is reasoning, there disturbance of thinking is likely to be disturbance of reasoning’.29 At the same time, voices continue to be heard from the opposite, cardiocentric camp: as Galen reports, the fourth-​century doctor Diocles ‘says that phrenitis is an inflammation of the diaphragm (phlegmonē tou diaphragmatos)’; but, Galen stresses, Diocles says so not because he thinks such a disease could be an ‘affection of the phronein (“reasoning”)’, that is, of the reasoning faculties (which he posits in the heart), but because it is an affection of the diaphragm, the phrenes (fr. 72 van der Eijk). In this passage the emphasis is on phrenes as anatomical (the diaphragm), and on phrenitis as a bodily disease, as the agenda of the discussion in Galen is to emphasise Diocles’ cardiocentrism. The same disease and label are flexible enough, however, to be manipulated into fitting different levels of discourse and different localizations. Many examples could be offered for this shifting of localization, etymology and pathology around one another in our disease. I shall bring two examples which are especially illustrative: first, in the Hippocratic Sacred Disease 17 (Jouanna 30, 4–​17=L. 6.392.6–​19) we find an

65

66

Chiara Thumiger

early medical discussion in which the diaphragm (phrenes) but also the kardiē (a location later identified with the heart) are openly demystified as location of cognitive faculties: the phrenes (= the diaphragm) have obtained their name (phr-​enes, from phronein, ‘to reason’) from accident and usage, and not from reality or nature, for I know no power which it possesses either as to sense or understanding, except that when the man is affected with unexpected joy or sorrow, it throbs and produces palpitations, owing to its thinness, and as having no belly to receive anything good or bad that may present themselves to it, but it is thrown into commotion by both these, from its natural weakness. It then perceives beforehand none of those things which occur in the body, but has received its name vaguely and without any proper reason … Some people say that the heart (kardiē) is the organ with which we think, and that it feels pain and anxiety. But it is not so; it merely is convulsed, as are the phrenes. This fifth-​century treatise, itself a famous manifesto of encephalocentrism in its detailed attribution of cognition, emotional responses, judgement and will to the working of the brain indulges for a long passage in an articulate refutation of the phrenes as organ of mental life, turning its traditional psychological meaning into nothing more than a false etymology and a sloppy habit of speech; and reading down its—​subjectively felt—​responsive nature in states of psychological affection (we may suppose, the feeling of ‘flutter in the stomach’, the breathlessness, the racing heart, the reflux, and other similar embodied experiences of heightened emotion we might identify with) to its passive ability to suffer from mental events, rather than ability to originate any.30 Another source from around a hundred years later returns to the problem of location and of the phrenes upon discussing the name of diseases, and uses precisely the example of phrenitis—​ but in an opposite spirit. At Anonymus Londinensis IV (Manetti 7. 7–​9) it is said that diseases are named from the ‘attendant affection’ or from the ‘affected place’. The author says, as an example, that ‘from the place affected phrenitis receives its name. For the affection makes its seat around the phrenes’. Then, he moves on to clarify: ‘not the diaphragm, but that which is the rational part of the soul ... (τὸ λόγιστικὸν μέρος τῆς [ψ]υ[χ]ῆς)’. The thought is the same we have found in Galen’s report on Diocles’ views on phrenitis; but here phrenes returns to be the abstract phronein, reasoning, as opposed to the insignificant diaphragm. The rivalry, or tension between (broadly) cardiocentric and encephalocentric interpretation in the stories, descriptions and definitions of this disease will linger on for as long as this binary in Western intellectual history holds. This is strikingly illustrated by the reception of the Greek medical concept phrenitis in the Arabic sources of the end of the first millennium CE, where our label is translated as bir-​sam and/​or sir-​sam, ‘inflammation of the chest’ and ‘inflammation of the brain’ at different stages of its history, renewing the debate, and the ambiguity we have illustrated.31 Bir-sam and sir-sam are Persian terms which were adopted by Arabic translators to render the Greek phrenitis. Most interestingly for us the first, birsam, literally indicates an inflammation of the chest (bar, ‘chest’, and sam, ‘inflammation’ in Persian) and the second term, inflammation of the brain (sir, ‘head’, in Persian). The two have been said to never have been sharply distinguished, for various reasons, and indifferently applied to phrenitis as synonymous,32 but in a recent study Carpentieri illustrates the existence of a development in the usage from the ninth–​tenth century CE and in later stages. Earlier, birsam defined two illness with identical symptoms. It referred to an inflammation of either the meninges (brain fever) or the diaphragm (diaphragmatic fever). Both inflammations would 66

67

Information and history of psychiatry

cause delirium, high fever, and were, typically, fatal. In the second and later stage, the usage of birsam became more restricted, designating only diaphragmatic fever. Brain fever, on the other hand, came to be referred to only as sirsam.33 In short, through changing scientific and cultural environments—​classical medicine, Galen, the Arabic reception of Greek medicine, ​which we could here only mention and briefly sample—it appears that very different authors engage, in various ways, with the problem of the double location of phrenitis, adapting the disagreement about its causes and pathology to the changing terms of medical and scientific debates. The tying-in of the definition of our disease with such a fundamental point of philosophical and medical discussion in our tradition sustained the durability of phrenitis through the centuries and protected it from anachronism as it passed from one medical culture and bodily paradigm to the other.

Quasi phreniticus: the disease as metaphor I now turn to the last of the four aspects I mentioned: phrenitis is a disease label endowed with a highly technical status, as an easy search through the surviving texts of Greek antiquity reveal beyond doubt. Precisely because of its heightened technicalism it grew to acquire a metaphorical, ethical meaning in the first centuries of our era, incarnating flaw of character, spiritual shortcoming and a general portrayal of ‘human folly’, and functioning as rhetorical catalyzer in religious-​moral invectives.34 This last aspect of ‘information’ is the more pervasive for our current discussion, concerning as it does the social outreach of a disease notion, and a shifting of its cultural, and social positioning from the elitist status of a highly technical term, only known and used by physicians, to a popular, widely understood token for common human weaknesses in the context of theological, protreptic or prudential literatures—​which have, by definition, a far wider reach than medical treatises. Phrenitis is a special case in this sense, too. While other terms for mental pathology used by the medical authors, like mania and melancholy are well-​known presences in various literary genres, to convey emphasis in serious contexts or with comical overtones,35 phrenitis and its cognates are not only rare, but basically absent until the imperial age. The only occurrence of the term before Cicero is the proverbial exception that confirms the rule: a reference in a third century BCE’s fragmentary play, Menander’s Aspis in which a faux doctor is evoked, and a parodistic diagnosis is offered. The scene is rich in randomly used medical jargon in the service of creating a parody—​that of a ‘certain kind of philosophizing doctor’.36 At the turn of the millennium and in the first centuries of our era phrenitis begins to appear outside the medical contexts, but in a clearly defined constellation of authors and contexts: some prudential and ethical reflections (Seneca, Plutarch, Cicero), satyre (Petronius, Martial, Juvenal) and the work of Christian authors (patristic and hagiographical works of various kinds). These divergent trajectories constitute only apparently a contradiction. Both comedy and moral exhortation, in fact, stem from the same origin: the moralized, popularized appropriation of a term strongly connoted ontologically and lexically—​that is, concrete in its embodied manifestations and highly technical. Thus, the technical term for the mental disease ‘par excellence’, phrenitis, becomes a metaphor for stupidity, delusion (in comic genres) or, in serious contexts, for the foolishness and irrationality of man who refuse the ‘real medicine’ and rejects the ‘real doctor’—​God. I survey this rich material elsewhere;37 this is a long story we can only tackle here by offering some illustrative pointers. Phrenitis appears, first of all, to be used antonomastically to convey weakness of body and soul, and a generalized ‘fury’ as hyperbole for intellectual deficiency already in the first century CE, by Seneca and Plutarch: 67

68

Chiara Thumiger

if you practice philosophy, it is good. To be in good health is ultimately this. Without this your soul is sick. Also the body, even if it has great strength, it is no more strong that it is in manic or phrenitic people (non aliter quam furiosi aut phrenetici validum est).38 More detailed are the references found in Plutarch, an author versed in medicine and whose texts display a great deal of technical knowledge. In De Latenter Vivendo (128 D 4) the phrenitic gains an ethical profile as the proverbial ill who ignore his own illness: if you are talking to a fool, a wretched or a senseless person, then you are no different from someone who is saying ‘hide the fact that you have a fever’, and hide the fact that you suffer from phrenitis (λάθε φρενιτίζων), so that the doctor might not find you; go and throw yourself somewhere dark, unrecognized among the other diseased … Turning to Christian literature, key in the rhetorical creation of a metaphorical phrenitis is above all Augustine of Hippo; he is also the most articulate in his use of our disease as ethical symbol. Further aided by his exceptional weight as author within Christianity, his images and allegories will remain influential through Christian literature.39 For example, in his De Genesi ad litteram libri duodecim phrenitis means irrationality and nonsensical talk (‘he was feverish and was saying these things [in the same way as] through phrenesis’40 (tamquam in phrenesi ista dicebat).41 Along similar lines, a spiritual and existential evil that befalls humanity is identified with phrenitis by Peter Chrysologus (fourth–​fifth century CE), with the image of a community ‘lying in the darkness of its own depravity, oppressed under the weight of his sins, feverish with perversion to the point of frenesis’ (uitiis usque ad frenesem febrientem, Collectio sermonum);42 or, again, ‘Brothers, when the disease of sins, vices, wickedness, impiety, frenesis pervades the human minds, and extinguishes with its insane fury whatever it has of wisdom, sense and reason, it causes people throughout the earth to flee away from God’.43 In a similar way, phrenesis becomes image for the irrational mob in the words of Peter Lombard (eleventh–​twelfth century CE): ‘for a disordered turmoil arose in the crowd, as if coming from someone suffering from phrenesis’ (fit enim tumultus quidam inconditus multitudinis, quasi phrenesim patientisi), while William of Auvergne (Guillelmus Alvernus, twelfth–​thirteenth century CE) equates it to the human sins of greed and arrogance:44 ‘likewise about greed and arrogance and the like, which are like a form of continuous sleep, like frenesis and the like’ (idem de auaricia et superbia et huiusmodi, que sunt quasi fixus somnus, sicut frenesis et huiusmodi). The sources one could mention are innumerable, and reach as far as the modern era; perhaps most illustrative is the work of the ninth–​tenth century CE bishop of Verona, Rather, who composed an invective against his enemies in the form of a pamphlet entitled Phrenesis, where the disease as allegorical instrument of polemic is used in every page. When it comes to the flaws this metaphorical phrenitis indicates, the ethical and the cognitive are inseparable levels of critique. Phrenitic are the sinners, the fools, the presumptuous, the unfaithful; those who lack philosophical rigour and training, or magnanimity, or awareness of their own limits. Already in Seneca the phrenitic is that particular type of incapacitated patient whom a good doctor should not engage with on equal terms: ‘for which doctor angers himself against a phrenitic patient? Which would take badly the bad words coming from a feverish person, impeded by fever?’ (Quis enim phrenetico medicus irascitur, De Const.Sapientis 13.1.3); or again, at De Ira 3.26.1, 4, ‘why do you take badly the fury of an ill person or the words of a phrenitic, or the insolent gestures of children?’ (quare fers aegri rabiem et phrenitici verba, puerorum protervas manus?).45 The idea of the phrenitic as quintessentially incapacitated individual persists to the end of the medieval era, when the Czech theologian Jan Hus (Defensio articulorum Wyclif, 68

69

Information and history of psychiatry

lectio: 2, 204, 790) will revive the famous Platonic anecdote of the madman who should not be left with a sward, even if it belongs to him, precisely to qualify the status of the phrenitic: Thus it is an action of greater compassion to take away the sword from a phrenitic who wants to kill himself than to give a sword to a persecuted person to defend himself from someone who wants to kill him, because it would be worse if a man were to die at his own hand in this way, than if one were killed by another; the first case would be condemnable, while the second deserved and right. (Sicut enim est opus maioris misericordie aufferre gladium a frenetico volente se ipsum occidere quam dare gladium persecuto ad defendendum se a volente eum occidere, quia peius foret hominem occidi sic a seipso quam occidi ab alio, quia primum est dampnabile et secundum meritorium sive iustum.) As phrenitis becomes the ethical-​existential ill par excellence, a specific allegorical pattern on the basis of the elements we have seen already emerges—​the phrenitic’s lack of awareness, his furious behavior, his being disabled and irresponsible for his own actions, culminating in the narrative of the sinner as patient facing a good doctor who tries to help—​ultimately, the one doctor who only can offer man the true medicine, God. Jerome (fourth–​fifth century CE) already casts the phrenitic as the ungrateful beneficiary of his advice in his Epistulae: ‘Why do you try to insult others while neglecting your own flaw? Why do you assault me with your bite, like a phrenitic, while I have always advised you well and with great care?’;46 most clearly Augustine, with reference to God and humanity as ungrateful patients: ‘for if they had been sick in a milder way, they would not have killed their doctor like phrenitics’ (nam si uel mitius aegrotassent, non quasi phrenetici medicum occidissent).47 The furious bite against the doctor will become a centre piece of the repertoire, as testimonies from a much later time show. Consider the words of Bernard de Clairvaux (eleventh–​ twelfth century CE): so that the one who goes with the deaf becomes deaf himself, sinking into wicked doctrines to find excuses for their sins and, in the way of the phrenitics, not only rejecting but even trying to bite the hand of the doctor. (ut qui in sordibus erat, sordescat adhuc, declinans in verba malitiae ad excusandas excusationes in peccatis ac, more phrenetici, non solum repellens, sed et mordere tentans medici manum.)48 Or Rupert of Deutz (eleventh–​twelfth century CE), who in his De sancta trinitate et operibus eius writes: he will see how diseased is that hand of his and will recognise his own disease, while until now it appeared to him to be healthy and strong, just as if his hand, made insane because of the violence of a disease, like a phrenitic one, should begin to beat the true doctor as it is said in the Psalm: ‘and they will rush against me with violence’. (Videbit inquam quod manus illa infirmata sit et infirmitatem suam agnouerit quae hactenus sibi uidebatur sana et fortis ita ut prae infirmitatis ui insaniens ut phrenetica medicum uerum percuteret quemadmodum dicit in psalmo: et irruerunt in me fortes.)49 If the patient misrecognizes doctor and medicine, that of tough love and bitter medicine also become key vehicles within the normative allegory that sees the phrenitic as the quintessential 69

70

Chiara Thumiger

sinner-​patient: a good friend or relative is the one who can impart ‘tough love’ to the weak and misguided. See Augustine again: ‘and the one who ties up the phrenitic and tries to urge the lethargic into action, by being aggravating to both is actually loving both’ (et qui phreneticum ligat et qui lethargicum excitat, ambobus molestus ambos amat).50 The motif is picked up by other authors, as we see in Philip of Harveng (twelfth century CE), in Sermones de tempore: Likewise, who would say that God is not at peace with the one whose insults, blows, blasphemies and rejection he patiently suffers; just like the pious mother of a phrenitic son ties him up so that he cannot rage against himself or others with his ill acts; and in this way, when he returns to his senses and repents and makes peace with him, he can in addition gain paradise? (Item, quis dicat Dominum non habere pacem cum illo cuius obprobria et uerbera et blasphemias et exactiones pacienter sustinet, sicut pia mater filii frenetici uel eum ligat ne seuire | possit in se uel in alios infirmitatibus et huiusmodi, quando enim ad cor reuertitur et penitet et pacem eius habet et lucratus est insuper paradisum?)51 Or, again, in De silentio, ‘but pitying and embracing the phrenitic he just refused to love the phrenesis in him’ (sed miserans et amplectens phreneticum solam illius phrenesim non dilexit.)52 On the psychological and psychiatric side, we can single out two more themes which reinforce the enduring success of phrenitis as pathological concept: the topic of unawareness, directly affiliated with the role played by hallucinations in the profile for this disease, and that of perverted joy, also rooted in distorted judgement and perceptions. With his hallucinations and pathologically heightened bodily strengths, the more the phrenitic suffers, the more severely he is unaware of his illness. Peter Chrysologus (fourth–​fifth century CE) topically evokes the Pharisee, who ‘wounded with the disease of depravity, feverish with the flame of arrogance, through his frenesis did not know he was insane’ (Pharisaeus, perfidiae morbo saucius, superbiae flamma febriens, per frenesim se nesciebat insanum);53 or, again, ‘a brother remains such when his brother suffers from fever, your neighbour remains such even when the neighbour commits crimes through frenesis, unaware of himself, exiled from humanity’ (fratri frater est in febre cum laedit, est in frenesi proximo proximus cum delinquit, est sibi nescius, est humanitatis extraneus).54 Thomas Aquinas (thirteenth century CE) also returns to phrenitis to qualify the topos of the arrogance and formalism of the Pharisees: ‘in which matter it is to be noted, that as the Pharisee is convinced of his own opinion, he is carrying his own rope with which he should be tied up, like a phrenitic’ (Qua in re notandum est, quod dum sua sententia Pharisaeus convincitur, quasi phreneticus funem portat ex quo ligetur).55 Finally, an eminently psychological feature of these allegorical phrenitics is the pathological joy that belongs to their misplaced strength. Augustine is again seminal: at In Iohannis euangelium tractatus he compares misdirected happiness to phrenitis: ‘for their joy is like the joy of the phrenitic’ (gaudium enim ipsorum quasi phreneticorum est);56 at In Iohannis euangelium tractatus, more elaborately, he writes: ‘in the way in which the phrenitic greatly rejoices in his insanity, and laughs, while those who are sane mourn him; in the same way we too, my beloved, if we accept the medicine that comes from heaven, since we too were once phrenitic, we will also be saved, ... and we would weep to God for those who are still insane’ (quomodo autem phreneticus gaudet in insania plerumque, et ridet; et plangit illum qui sanus est; sic et nos, carissimi, si recepimus medicinam de caelo uenientem, quia et nos omnes phrenetici eramus tamquam salui facti, ... gemamus ad deum de iis qui adhuc insaniunt).57 The tradition will preserve the image, as in Sedulius Scottus (ninth century CE), Collectaneum miscellaneum diuisio: ‘theirs is like the joy of the phenetics’ (Gaudium itaque ipsorum quasi freneticorum est),58 or Philip of Harveng (twelfth century CE), De dignitate 70

71

Information and history of psychiatry

clericorum: ‘the dissolute path of life of which, like phrenitics, they congratulate themselves’ (dissolutum uiuendi tramitem quo quasi frenetici gratulantur); and so on.59

Conclusions At the beginning of this chapter I resorted to the accepted image and critical concept of ‘anchoring’ which describes the manner in which the survival of cultural concepts is made possible through changing scientific and cultural environments and communities. In the light of the four paths of knowledge transfer we have explored in the case of the disease phrenitis—​its traditional appeal, its hardwired embodiment, its allegiance with a key theme in Western philosophical history, and its lexical positioning between technicality and allegorical exploitation—​perhaps a different, more collective image appears to be needed for an entity with such an important medical-​scientific component; an image placing less emphasis on intentional agency from the side of a propaganda-​minded entity as origin of the anchoring move, and less commitment to ‘new’ and ‘old’ as stable concepts,60 a more dangerous pitfall in history of science as in other spheres of human production. In introducing his monumental discussion of the history of evolutionary theory, Stephen Gould recalls the architectural image used in 1863 by the naturalist Hugh Falconer in appreciation of Darwin’s newly formulated model: [Darwin] has laid the foundations of a great edifice; but he need not be surprised if, in the progress of erection, the superstructure is altered by his successors, like the Duomo of Milan, from the Roman to a different style of architecture. To which acknowledgement, and suggestive image Darwin will reply, in seeming assent, ‘I expect and hope that the framework [scil: at least] will stand’. As Gould subtly comments, the two scientists are here getting at two different models of survival, whose differences are slight but greatly important for the historian of science or cultural historian. While the first, in fact, claims that the fundamentals of the theory, the ‘evolutionary principle of descent with modification’ (the basement), will stand, although future theories might describe differently ‘the tree of life’ (the façade of the Duomo, with its syncretism of different styles), Darwin for his part envisages that that general shape of his theory (the general design of the cathedral) will persist, despite the possibility of radical corrections in any of its fundamental parts.61 This exchange best illustrates the choral and complex quality of the existence, and survival of scientific creations. If phrenitis and evolutionary theory are only partly comparable items, they share an important aspect against other kinds of cultural products, which maybe can be described as undergoing forms of ‘anchoring’ in the service of integrating innovation within political, economical, or religious communities. Both are inextricably cultural and scientific at the same time, that is, with a stake in Truth with the capital T (the life and death of human beings, the working of their bodies, the ultimate alleviation of their suffering; the origin and development of the human species respectively) as well as being narratives about ourselves, who we are and in which way we live, function and may die, also being strongly couched in cultural elaborations. The image of the architectural complexity of a cathedral whose completion spanned across two hundred years, with its hybrid stylistic elements, discordant designs and yet final coherence, constitutes a fair representation, to return to our case of the disease phrenitis, of continuity in a piece of scientific information, which must have a structure—​the foundations—​ and a form—​its design, shape, general look. Both can change, be radically altered, even be replaced; but the item can survive with some continuity if the relationship between foundation 71

72

Chiara Thumiger

and form is maintained to an extent sufficient to keep the building identifiable through its evolutions. The disease phrenitis underwent changes in the foundations (in its etiology and physiology, from a broadly cardiocentric frame to an encephalocentric qualification) while maintaining the original integrated structure in which both localizations were present; it maintained visible elements of ‘design’ (the consistent signs and symptoms described: fever, crocydism, hallucinations) while integrating new ones, keeping pace with the changing acquisitions of medicine (e.g., the nerves, the pulse); it also developed a varied but consistent set of ‘decorations’ (the metaphorical and allegorical uses of the disease) which, somehow aurally, contribute to offering coherence and identity to the edifice. Gould notices that a third option, unmentioned by either Falconer and Darwin in their hopeful appreciation of evolutionary theory, is always possible in the history of scientific ideas: as it is ‘often the fate of cathedrals … destruction, either total or partial, followed by a new building of contrary or oppositional form, erected over a different foundation’. With the rise of scientific anatomy in the seventeenth century, and within an increasingly segregated view of the matters of the soul-​mind from those of the body, the disease entity known as phrenitis will meet a bifurcated destiny: as bodily disease, it will be increasingly assimilated to inflammation of the brain and meninges, and assimilated to encephalitis and meningitis; as a phenomenon of psychiatric interest, it will develop into various parallel outcomes, among which a syndrome described as ‘delirium’, and no longer be diagnosed after the end of the nineteenth century.62 Once the delicate balance, and modulation of changes between foundations, details and general design ceased to hold, the edifice too was demolished. The case of phrenitis, and in general of the life of diseases, incarnates to perfection the mediatic and epistemological channels through which human stories can be told but, most concretely, through which human lives can be lived—​that fundamental sense in which information, far from being a simple matter of delivered content, deeply informs human life and health; and human death.

Notes 1 An excellent survey is Murphy (2015). 2 Cf. www.psychiatry.org/​psychiatrists/​practice/​dsm and www.who.int/​classifications/​icd/​en/​respectively. 3 See the essays in Hauswald, Keil and Keuck (2017). For a representative discussion; se also Murphy (2017); Busfield (2011) 7–​39. 4 See Thumiger (2017) 44–​50; Thumiger and Singer (2018) 1–​15; now the wide-ranging chapters in Steinert (2020). 5 Berrios (1996) 11. 6 See Wright (2016) 259–​318, on late-​antique theological uses of phrenitis; (2020) on Augustinus. 7 See Sluiter (2017). I find this metaphor, and the cultural phenomenon it indicates, most useful as items in intellectual history, rather than social strategy behind which the agency of individuals can be traced. 8 Illustrative is the hugely influential publication of Kühn’s The Structure of Scientific Revolutions (1962) and its immediate critique by Popper, Feyerabend and Lakatos among others (Criticism and the Growth of Knowledge, 1970). 9 A pitfall of which Sluiter is well aware (2017) 11. 10 See Thumiger (2007) 72–​3. 11 Thumiger (2013) on medicine, (2007) 60–​86 on literary sources; Sullivan (1979) on Homer; Clarke (1999) 74–​9; 83–​6; 106–​10; Stefanelli (2010) 19–​24, 44–​51; Onians (1951) 13, 23–​30, 39–​40. 12 On ancient technical languages see Schironi (2010), van der Eijk (1997). 13 On this definition of territory in classical medicine against poetic and popular traditions, see Thumiger (2017) 419–​22.

72

73

Information and history of psychiatry 14 Plut. qu. conv. 8.9, 731d4. On the topic of ‘new diseases’ in ancient medicine and its importance in sectarian debate, see Harris (forthcoming, 2021). 15 The only exception to this is Celsus, in many respects a unique discussion—​see Thumiger (2020), Thumiger and Singer (2018) 7–​15. 16 For instance, at Prognosticon 4 (Jouanna 13.3–​5 = 2.122 L.), ‘in those who suffer under acute fevers, or peripleumonia, or phrenitis …’, or Epidemics 1.6 (Jouanna 10.12–​13=L. 2.620.6) ‘I know of no ardent fever proving fatal at this time, nor of any phrenitis’ (οὐδὲ ἀποθανόντα οὐδένα οἶδα τότε καύσῳ, οὐδὲ φρενιτικὰ τότε γενόμενα); cf. also Epidemics 3.6 (Jouanna 83.11–​12= 3.80.3), Epidemics 3.14 (Jouanna 90.16= 3.98.2). At Epid. 1.6 (Jouanna 10.12–13 =L. 2.620.6), Epid. 1.22 (Jouanna 32.3–5 = L. 6.666.11), Epid. 3.5 (Jouanna 83.11–​12 = 3.80.3), and Epid. 3.14 (Jouanna 90.16 = 3.98.2) καῦσος and τὰ φρενιτικά are associated and implied to be categorically related. 17 See Epidemics 3.17, case 4 (Jouanna 98.1–​11= L. 3.116.15–​118.5) for an example combining all these. 18 Galen reflects at length on the nature of pathological signs as indicator of a disease, especially in his Commentary to Prorrhetikon I; see Thumiger (forthcoming). 19 Cf. Coan Prenotions 571 (L. 5.716.16), Aphorisms 4.72 (Magd. 426.7–​8= 4.528 L.). 20 Caelius Aurelianus, Acute Diseases 1.3, 40.15–​22 Bendz. 21 One could perhaps compare the sacred disease as a similar case (on which see Temkin 1945). 22 See Manuli (1977) on this topic. 23 As well as attempts to combine them into a compromise, as in the Platonic tripartite account of the soul most clearly offered in the Timaeus and adopted by Galen in PHP; see Rocca (2003) 18–​47 on the ‘development of the hegemonic concept’ of a ruling seat of the soul in history of medicine. 24 As we are led to think by the one classical testimony we have about this, a passage from Menander’s Aspis—​see below. 25 Epidemics 7.112 (Jouanna 112.3–​9 = L. 5.460). 26 Epidemics 7.112 (Jouanna 112.12–​15 = L. 5.460). 27 Epidemics 3.17, case 4 (Jouanna 98.3–​4 = 3.117.16). 28 Coan Prenotions 116 (L. 5.608). Hypochondrion or hypochondria indicate the epigastric regions of the upper chest, locate on the two sides. 29 φρενίτιδος αἰτία. Ἐρασίστρατος μὲν ἐξ ἀκολουθίας τῶν ἑαυτοῦ δογμάτων φησὶ τὴν φρενῖτιν κατά τι πάθος τῶν κατὰ τὴν μήνιγγα ἐνεργειῶν· οὗ γὰρ τόπου κατ’ αὐτὸν ἡ νόησις φρόνησις, ἐπὶ τούτου ἡ παρανόησις παραφρόνησις ἂν εἴη)(fr. 176 Garofalo). 30 Similar points are touched on by Aristotle in his discussion of the phrenes in Parts of Animals 3.10 (672 b 24–​673a28), however less critically from a cardiocentric perspective. 31 See Carpentieri (2017); Carpentieri and Mimura (2017). 32 See Carpentieri (2017) 1; Jacquart (1992) 184, who says Razi employs all three terms indifferently; cf. also Bornemann (1988) for a survey on phrenitis in Arabic sources. 33 Carpentieri (2017) 2. 34 In classical Greek medicine ethical evaluation is entirely absent from discourses on mental health, while the question of value and human flourishing began to be visible as object of medical interest in the imperial era—​Galen, most notably, engages with ideals of ethical improvement as therepeutical process. Phrenitis however is not discussed by him in this sense, but remains, qua medical disease, a topic for purely physiological investigation, not even mentioned once in Galen’s ethical writings. 35 See Thumiger (2013) 62–​70 for discussion of these two diseases. 36 The two relevant passages: Aspis (339–42): ἔπειτα παραληφθήσεται ἐνταῦθ’ ἰατρός τις φιλοσοφῶν καὶ λέγων πλευρῖτιν εἶναι τὸ κακὸν ἢ φρενῖτιν ἢ τούτων τι τῶν ταχέως ἀναιρούντων.

Afterwards we will call a doctor, a wise philosopher who will say that it is a case of pleuritis, or phrenitis, or some other acute disease such as these Med. ‘It is the phrenes, I think. … we usually call this phrenitis’. Smicr. ‘I understand. And then? Is here no chance to save him?’ Med. ‘These are serious diseases, if you don’t want me to comfort you with vain hopes …’ Smicr. ‘Don’t deceive me, but tell me the truth.’

73

74

Chiara Thumiger Med. ‘It is impossible to him to survive. He is throwing out bile, he is darkened in his vision … has foam at the mouth…his glance shows death.’ And Aspis (444–50): {(Ια)} [α]ὐτὰς τὰς φρένας δή μοι δοκῶ []. ὀνυμάζειν μὲν ὦν εἰώθαμες [φ]ρενῖτιν τοῦτο. {} μανθάνω. τί οὖν; {(Ια)} οὐκ ἔστ]ιν ἐλπὶς οὐδεμία σωτηρίας. καίρια] γάρ, αἰ μὴ δεῖ σε θάλπεν διὰ κενᾶς, τὰ τοια]ῦτα. {(Σμ)} μὴ θάλπ’, ἀλλὰ τἀληθῆ λέγε. {(Ια)} οὐ πάμπαν οὗτός ἐστί τοι βιώσιμος. ἀνερεύγεταί τι τᾶς χολᾶς· ἐπισκοτεῖ []εντ̣.[..] καὶ τοῖς ὄμμασι []υ̣κνον ἀναφρίζει τε καὶ []. ας ἐκφορὰν βλέπει. See Thumiger (forthcoming) on this passage. 37 Thumiger (forthcoming). 38 Seneca, Epistulae Morales ad Lucilium 15.1, p. 44. 39 See Wright (2016) 86‒127; Wright (2020). 40 The Greek phrenitis is transliterated into Latin as phrenesis/​frenesis, later phrenesia/​frenesia. 41 CPL 0266 lib.: 12, par.: 17, pag.: 403, linea: 25. 42 CPL 0227 + (M) SL 24, sermo: 18, linea: 67. 43 Collectio sermonum (CPL 0227 + (M) SL 24, sermo: 38, linea: 55. 44 Sermones de communi sanctorum et de occasionibus, sermo: 80 (de uno confessore), pag.: 276, linea: 18. 45 Plutarch makes the same point at Biogr., fr.136.4. 46 CPL 0620, Epist.: 147, vol.: 56, par.: 9, pag.: 324, linea: 10. 47 Enarrationes in Psalmos (CPL 0283), SL 39, psalmus: 65, par.: 4, linea: 67). Cf. Also Contra Cresconium (CPL 0335), lib.: 4, cap.: 51, par.: 61, pag.: 558, linea: 18 In Iohannis euangelium tractatus (CPL 0278), tract.: 17, par.: 15, linea: 27, Enarrationes in Psalmos (CPL 0283), SL 39, psalmus: 58, sermo: 2, par.: 5, linea: 69, Sermones (CPL 0284) sermo: 80, ed.: PL 38, col.: 496, linea: 12, Sermones (CPL 0284), sermo: 80, ed.: PL 38, col.: 496, linea: 20, Contra litteras cuiusdam presbyterorum coniugatorum causam defendentis, pag.: 249, linea: 14. 48 Sermones in die paschae, sermo: 2, par.: 9, vol.: 5, pag.: 99, linea: 15. 49 CM 22, lib.: 19, In Deuteronomium II, pag.: 1091, linea: 1108. 50 CPL 0262, Epist.: 93, vol.: 34.2, par.: 2, pag.: 449, linea: 1. 51 Sermo: 34 (dominica IV aduentus), CM 230, pag.: 139, linea: 27. 52 De silentio, cap.: 64, col.: 1077. 53 SL 24A, sermo: 94, linea: 35. 54 SL 24B, sermo: 139, linea: 38. 55 Catena aurea in Lucam, cap.: 7, lectio: 6, linea: 141 (pag.: 105, linea: 65. 56 CPL 0278, tract.: 7, par.: 2, linea: 3. 57 CPL 0278, tract.: 7, par.: 2, linea: 4. 58 24, linea: 24. 59 cap.: 122 (olim: De continentia clericorum), col.: 831, linea: 58. 60 On the topic of tradition and retrospective ‘traditioning’ as ways to protect a form of knowledge or practice, see the seminal points made by Bouillon (1977); on the ‘precursor’ fallacy in history of science, Canguilhem (1968/1988) 49–51. For a multifaced revision of the category ‘new’ (and ‘old’) in Greek culture see D’Angour (2011). 61 Gould (2002) 3, also on the complicated differences between the two scientists’ claims; 2–​7. 62 On the history of this concept, see Berrios (1981).

References D’Angour, A. The Greeks and the New: Novelty in Ancient Greek Imagination and Experience. Cambridge; New York: Cambridge University Press, 2011.

74

75

Information and history of psychiatry Berrios, G.E. ‘Delirium and Confusion in the 19th Century: A Conceptual History’. British Journal of Psychiatry 139 (1981) 439–49. Berrios, G.E. The History of Mental Symptoms: Descriptive Psychopathology since the Nineteenth Century. Cambridge: Cambridge University Press, 1996. Bornemann, A. Das Krankheitsbild der Phrenitis in der Medizin des arabischen Mittelalters unter besonderer Berücksightigeung von Ibn Sînâ (980–​1037). Diss. Bonn (1988). Busfield, J. Mental Illness. Oxford: Polity Press, 2011. Canguilhem, G. ‘The History of the History of Sciences’. In A Vital Rationalist. Selected Writings from Georges Canguilhem, trans. Arthur Goldhammer. Cambridge: MIT Press (1994/1968) 49–63 (in part originally published in Etudes d’histoire et de philosophie des sciences, Paris 1968). Carpentieri, N. ‘On the Meaning of Birsām and Sirsām: A Survey of the Arabic Commentaries on the Hippocratic Aphorisms’. MIDEO 32 (2017) 81–92. Carpentieri, N., and Mimura, T. ‘Arabic Commentaries on the Hippocratic Aphorisms, vi.11: A Medieval Medical Debate on Phrenitis’. ORIENS 45 (2017) 176–202. Clarke, M. Flesh and Spirit in the Songs of Homer. Oxford: Oxford University Press, 1999. Gould, S.J. The Structure of Evolutionary Theory. Cambridge, MA: Harvard University Press, 2002. Harris, A. ‘New Diseases and Sectarian Debate in Hellenistic and Roman Medicine’. Apeiron (2021, forthcoming). Harris, W.V. (ed.). Mental Disorders in the Classical World. Leiden: Brill, 2013. Hauswald, R., Keil, G. and Keuck, L. (eds.). Vagueness in Psychiatry. Oxford: Oxford University Press, 2017. Jacquart, D. ‘Les avatars de la phrénitis chez Avicenne et Rhazès’ in Maladie et Maladies: Histoire et conceptualisation. Mélanges en l’honneur de Mirko Grmek, ed. D. Gourevitch. Geneva: Droz, 1992, 181–​92. Keil, G. and Stoecker, R. ‘Disease as a vague and thick cluster concept’, in Vagueness in Psychiatry, ed. R. Hauswald, G. Keil and L. Keuck. Oxford: Oxford University Press, 2017, 46–​74. Keuck, L.K. ‘In quest of “good” medical classification systems’, Medicine Studies 3(1) (2011): 53–​70. doi: 10.1007/​s12376-​011-​0065-​5 (accessed 5/​1/​2021). Kühn, T.S. The Structure of Scientific Revolutions. Chicago: Chicago Press, 1962. Lakatos, I. and Musgrave, A. (eds.) Criticism and the Growth of Knowledge. Cambridge: Cambridge University Press, 1970. Lewis, O., Thumiger, C. and van der Eijk, P. ‘Mind, body and the concept of gradual health in ancient medicine’, in Vagueness in Psychiatry, ed. R. Hauswald, G. Keil and L. Keuck. Oxford: Oxford University Press, 2017, 27–​45. Manuli, P. ‘La techne medica nella tradizione encefalocentrica e cardio emocentrica’, in Corpus Hippocraticum, ed. J. Robert. Mons: Editions universitaires de Mons, 1977, 182–​95. Murphy, D. ‘Concepts of Disease and Health’, The Stanford Encyclopedia of Philosophy (Spring 2015 Edition), Edward N. Zalta (ed.), https://​plato.stanford.edu/​archives/​spr2015/​entries/​health-​disease/​ (accessed 25/​1/​2020). Murphy, D. ‘Philosophy of Psychiatry’, The Stanford Encyclopedia of Philosophy (Spring 2017 Edition), Edward N. Zalta (ed.), https://​plato.stanford.edu/​archives/​spr2017/​entries/​psychiatry/​ (accessed 25/​ 1/​2020). Onians, R.B. The Origins of European Thought: About the Body, the Mind, the Soul, the World, Time and Fate. Cambridge: Cambridge University Press, 1951. Pouillon, J. ‘Plus c’est la même chose, plus ça change’. Nouvelle Revue de Psychanalyse. “Mémoires”. 15 (1977) 203–11. Rocca, J. Galen on the Brain: Anatomical Knowledge and Physiological Speculation in the Second Century AD. Leiden: Brill, 2003. Schironi, F. ‘Technical languages: Science and medicine’, in A Companion to the Ancient Greek Language, ed. E. Bakker. Oxford: Wiley-​Blackwell, 2010, 338–​54. Sluiter, I. ‘Anchoring innovation: A classical research agenda’, European Review 25(1) (February 2017): 20–​38. Stefanelli, R. La Temperatura dell’Anima: Parole Omeriche per l’Interiorità. Padova: Unipress, 2010. Steinert, U. (ed.) Systems of Classification in Premodern Medical Cultures. Sickness, Health, and Local Epistemologies. London: Routledge, 2020. Sullivan, S.D. ‘A person’s relation to φρήν in Homer, Hesiod and the Greek lyric poets’, Glotta 57 (1979): 159–​73. Temkin, O. The Falling Sickness. Baltimore: Johns Hopkins University Press, 1945.

75

76

Chiara Thumiger Thumiger, C. ‘Hidden paths: Self and characterization in Greek tragedy: Euripides’ “Bacchae”’, Bulletin of the Institute of Classical Studies. Supplement 99 (2007). Thumiger, C. ‘The early Greek medical vocabulary of insanity’, in Mental Disorders in the Classical World, ed. W.V. Harris. Leiden: Brill, 2013, 61–​95. Thumiger, C. A History of the Mind and Mental Health in Classical Greek Medical Thought. Cambridge: Cambridge University Press, 2017a. Thumiger. C. ‘Mental disability? Galen on mental health’, in Disability in Antiquity, ed. Ch. Laes. Oxford: Routledge, 2017b, 275–​7. Thumiger, C. and Singer, P.N. ‘Introduction’, in Mental Illness in Ancient Medicine: From Celsus to Paul of Aegina (edited with P. Singer). Leiden: Brill, 2018, 1–34. Thumiger, C. ‘Therapy of the word and other psychotherapeutic approaches in Ancient Greek medicine’. Transcultural Psychiatry 57(6), 2020, 741–752. Thumiger, C. Phrenitis. Mental Pathology and Bodily Localisation. Forthcoming. Van der Eijk, P. ‘Towards a Rhetoric of Ancient Scientific Discourse Some Formal Characteristics of Greek Medical and Philosophical Texts’, in Grammar as Interpretation: Greek Literature in its Linguistic Contexts, ed. E.J. Bakker. Leiden: Brill, 1997, 77–​129. Wright, J. ‘Preaching phrenitis: The medicalization of religious difference in Augustine of Hippo’, Journal of Early Christian Studies (forthcoming). Wright, J. Brain and Soul in Late Antiquity. PhD Diss. Princeton, 2016. Wright, J. ‘Preaching Phrenitis: Augustine’s Medicalization of Religious Difference’, Journal of Early Christian Studies 28(4) (2020): 525‒53.

76

77

PART II

Information 500–​1500 Access

78

79

5 VĀCASPATI ON ABOUTNESS AND DECOMPOSITION 1 Nilanjan Das

I seem to see a white picket fence outside my window. Suppose my perceptual system is functioning normally. Am I aware of a mind-​independent particular, for example, the white picket fence? If you say “Yes,” I’ll call you a realist. If you say “No”, I'll call you an idealist. Realism is the view that perception can yield information about mind-​independent particulars; idealism denies this. Both realism and idealism come in different varieties. While all realists agree that perception can yield information about mind-​independent particulars by generating perceptual awareness or knowledge, they differ on how it does so. Indirect realists typically claim that such awareness or knowledge is epistemically indirect, that is, based on an inference from some body of evidence. For example, an indirect realist might say that I can only become aware of the white picket fence by inferring its existence on the basis of my non-​inferential awareness of a mind-​dependent entity, for example, a sense-​datum. In contrast, direct realists will deny this. For them, our perceptual awareness or knowledge of mind-​ independent particulars can be epistemically direct, that is, may not depend on any inference from any body of evidence. While all idealists agree that perception cannot yield knowledge or awareness about mind-​ independent particulars, they disagree on the ontological status of mind-​independent items. Some idealists claim that there are no mind-​independent items at all; call them ontological idealists. Others don’t rule out the existence of mind-​independent objects, but merely insist that we are never aware of or gain any knowledge about them. Call such idealists epistemological idealists. In the history of Indian philosophy, we find all these positions defended in one form or another. Some Buddhists, working within the Sautrāntika and Yogācāra traditions, defend indirect realism and idealism respectively. In contrast, philosophers who belong to the Buddhist tradition of Sarvāstivāda and non-​Buddhist traditions of Nyāya and Mīmāṃsā defend direct realism. The aim of this chapter is to focus on a 9th-​century Nyāya philosopher, Vācaspati Miśra, who defends direct realism against Buddhist arguments for indirect realism and idealism. Why is this debate worth caring about? As we shall see, Vācaspati opposes what I call decompositionalism about aboutness. Suppose I am aware of a white picket fence by means of perception. My experience, plausibly, is about the white picket fence. For decompositionalists,

79

80

Nilanjan Das

this aboutness relation between my perceptual experience and the picket fence obtains in virtue of two kinds of conditions: internal conditions like an internal picture in my head and external conditions like the resemblance between the internal picture and the relevant mind-​ independent particular.2 The Buddhist indirect realists—​the Sautrāntika philosophers—​defend decompositionalism. They claim that perceptual experiences, and all conscious thoughts and experiences more generally, possess what they call an objective phenomenal form (grāhyākāra, literally “the phenomenal form of what is apprehended”). At least on one interpretation, this is a mental image which represents a mind-​independent particular.3 A perceptual experience is about a certain mind-​independent particular insofar as this mental image resembles and is caused by that particular.4 In his sub-​ commentary Commentary on the Import (Tātparya-​ṭīkā) on the Aphorisms on Reasoning (Nyāyasūtra) 4.2.33–​4, Vācaspati wants to resist decompositionalism about aboutness.5 He motivates this view from the perspective of the Sautrāntika philosophers by drawing on some arguments given by Dharmakīrti (fl. c. 6th or 7th century CE) in his Ascertainment of Epistemology (Pramāṇa-​viniścaya). Then, he shows that this view leads to disastrous consequences. His argument is this. If decompositionalism about aboutness is true, then a perceptual awareness-​event cannot carry information about mind-​independent particulars and properties. This paves the way for epistemological idealism. But epistemological idealism is incoherent. So, decompositionalism about aboutness is false. Here’s how I shall proceed. In §1, I will briefly outline the Sautrāntika theory of perception, highlighting its commitment to decompositionalism about aboutness. Then, in §2, I will explain how Vācaspati motivates this brand of decompositionalism in light of Buddhist texts. In §3, I will show how, according to Vācaspati, decompositionalism leads to epistemological idealism. In §§4–​5, I will discuss how Vācaspati resists idealism and rejects decompositionalism about aboutness.

Sautrāntika decompositionalism The Sautrāntika theory of perception arose in opposition to a form of direct realism defended by Vaibhāṣika philosophers (who belonged to the Sarvāstivāda tradition of Abhidharma Buddhism). The Vaibhāṣikas thought that when we become aware of a mind-​independent particular by means of perception, our perceptual awareness-​event (jñāna) reveals to us a worldly item that exists at the time of our awareness. But Sautrāntika philosophers were uncomfortable with this idea.6 Implicitly, they thought that our best metaphysical theories should constrain our theories in epistemology and philosophy of mind, that is, whenever there is a conflict between one of our best metaphysical theories and a proposed theory in epistemology or philosophy of mind, we should reject the latter. So, contrary to how things appear, they argued that we have good metaphysical grounds for thinking that our perceptual awareness or knowledge about mind-​independent particulars cannot exist at the same time as those mind-​independent particulars. They were committed to: The Doctrine of Momentariness. There are no (ultimately real) temporally extended entities.7 The Thesis about the Object of Awareness. Any intentional object (ālambana) of perceptual awareness must both cause the awareness, and resemble what appears in it. The Thesis about Causation. If an object of perceptual awareness causes that awareness, then it must be present prior to the perceptual awareness. 80

81

Vācaspati on aboutness and decomposition

If these are correct and perceptual processing is temporally extended, then the mind-​independent particular that was in contact with the relevant sense-​faculty cannot exist at the time when the final conscious perceptual experience takes place. So, whatever item appears to the subject at this stage can only be a mental image left by the initial sense-​object contact. How, then, do we become aware of, or gain knowledge about, mind-​independent particulars? The Sautrāntika philosophers defend the following proposal. Any perceptual awareness is produced due to the causal interaction between a mind-​independent particular and a sense-​ faculty. An intrinsic property of the awareness-​event is an objective phenomenal form, a mental image that the subject is immediately aware of, which resembles the mind-​independent particular that produces the perceptual awareness. It is on the basis of this phenomenal form that the agent can infer the existence of a mind-​independent particular that produced the sensory awareness. This inferential judgement constitutes knowledge or awareness of the mind-​independent particular that produces the initial sensory awareness-​event. In a later Buddhist epistemology textbook, Mokṣākaragupta (fl. between 11th–​13th centuries CE) describes this view as follows. The view of the Sautrāntikas is this. All that appears in the form of blue, etc. is awareness itself, not an external object. For a non-​conscious object cannot be manifested. Thus it is said: “That which is detected by the senses is not observed, even though it gives rise an awareness that possesses its form.” [The opponent:] Well, if it is only awareness that is manifested, how do you know that there are external objects? [Reply:] The existence of external objects is demonstrated by a principle of absence. It is not the case that objective phenomenal forms, e.g., blue, are manifested always everywhere. They cannot also be produced just by dint of their own causal bases (upādāna) [i.e., the resources for conscious mental activity internal to the mind]. For, then, each specific awareness-​event wouldn’t be directed at a specific object. Therefore, we can conclude that there is some cause of these objective phenomenal forms, distinct from the immediately preceding condition [i.e., the mental event that serves as the immediate cause of perceptual experience], which makes them arise only at certain times at certain places. That, indeed, is the external object. (Krishnamacharya 1942, pp. 34–​35) I will return to this argument later. For now, just note this. On this view, perceptual experiences cannot make us non-​inferentially aware of mind-​independent particulars. We can become aware of them by means of some inference based on our awareness of the objective phenomenal forms of the experiences. Thus, it is tempting to draw a parallel between this theory and the kind of indirect realism that sense-​data theorists defend.8 Within the Sautrāntika framework, objective phenomenal forms of perceptual experiences, that is, the mental images that they involve, play an important role in explaining how a perceptual experience can be about or represent a mind-​independent particular. For the Sautrāntikas, a mind-​independent particular that a perceptual experience is about must satisfy two conditions: it must be a cause of the perceptual experience, and the perceptual experience must involve an objective phenomenal form, that is, a mental image, in virtue of which it resembles the mind-​ independent particular. This commits the Sautrāntika to: Sautrāntika Decompositionalism. For any subject S and a mind-​independent entity e, if S undergoes a perceptual experience about e, then the fact that S’s perceptual experience is about e holds solely in virtue of: 81

82

Nilanjan Das



(i) the fact that S undergoes experience that involves a certain objective phenomenal form, (ii) the fact that e causes the experience to have that objective phenomenal form, and (iii) the fact that the experience resembles e in virtue of its objective phenomenal form. This is an instance of a general decompositionalist view on which the fact that a perceptual experience is about a mind-​independent particular is necessarily decomposable into certain internal and external conditions. The internal condition has to do with an intrinsic property of the subject. In this case, it is a mental image in the subject’s head. The external conditions have to do with some representation-​grounding relation. In this case, it involves the relations of causation and resemblance between the experience and the mind-​independent particular.

A partial argument for Sautrāntika decompositionalism Sautrāntika decompositionalism is partially motivated by the idea that if a perceptual awareness-​ event didn’t have an objective phenomenal form, it couldn’t be about any mind-​independent particular. Vācaspati explains this idea in some detail.9 When an external object is apprehended by an awareness-​event that doesn't have any objective phenomenal form, is this merely due to the existence of the external object? Or, is this because the external object is one of the causal conditions that together bring about that awareness-​event? Or, is it because the external object depends on the same set of causal conditions as the awareness-​event? Or, is it because the awareness-​ event produces some result (phala) in the external object? ( Tātparya-​ṭīkā on Nyāyasūtra 4.2.33–​4 in Thakur 1996, p. 625, ll. 13–​14) There are four options here. Option 1. An awareness-​event is about an external object just in case that object exists. Option 2. An awareness-​event is about an external object just in case that object is one of the causal conditions that together bring about the awareness-​event. Option 3. An awareness-​event is about an external object just in case that object is caused by the same set of causal conditions as the awareness-​event itself. Option 4. An awareness-​event is about an external object just in case the awareness-​ event produces some result in that object. None of these options appeal to the objective phenomenal form of perceptual experiences. The Sautrāntika philosopher wants to argue that none of these options work. The first three options are non-​starters, so we can deal with them quickly. Consider Option 1: namely, that an awareness-​event can be about a mind-​independent item in virtue of its existence. This, as the Buddhist quickly points out, is a bad proposal. For existence is neither necessary nor sufficient for an object to be grasped by an awareness-​event. It's not sufficient, because, if it were, then every agent would be aware of every existent object. It's not necessary, since one can also be aware of non-​existent objects.10 Next, focus on Option 2: namely, that an awareness-​event can be about a mind-​independent item just in case it’s one of the causal conditions that together bring about the awareness-​event. This proposal, too, must fail. First, non-​existent objects—​past or future objects—​cannot be 82

83

Vācaspati on aboutness and decomposition

causal conditions but can still be objects of awareness. Second, there are some causal conditions of awareness-​events (e.g., the senses themselves) that are not grasped by the awareness-​events. Finally, if the Buddhist theory of momentariness is true, then this proposal rules out the possibility of there being any awareness of a currently existing object. For, by the time the awareness-​event arises, the previous time-​slice of the object which causes the awareness-​event will have gone out of existence.11 (This last remark is directed at Buddhist direct realists, i.e., the Vaibhāṣika philosophers from the Sarvāstivāda school, who think that the object of perception exists at the same time as the perceptual awareness.) In light of the last problem, Option 3 does look more promising. On this view, the object of awareness needn't be one of the causal conditions of the awareness-​event; rather, it may itself be produced by the same set of causal conditions that produce the awareness-​event. This makes room for the possibility that the awareness-​event may be about an object that exists at the same time as the awareness-​event itself. However, this proposal faces two serious worries. On the one hand, it cannot explain how we become aware of past or future objects. For those objects needn’t be produced by the causal conditions that produce an awareness-​event. On the other hand, it predicts that certain other effects brought about by the set of causal conditions that produce the awareness-​event may also be grasped by that awareness-​event.12 Option 4 is different: it’s based on some remarks that Kumārila Bhaṭṭa, the Mīmāṃsā philosopher, makes in a section called “The Debate About Emptiness” (Śūnyavāda) in his Detailed Commentary in Verse (Ślokavārttika). Unlike the Sautrāntika philosophers who construct their theory of intentionality on the basis of their metaphysical commitments, the Mīmāṃsā account of intentionality proceeds from a semantic theory (defended by the classical Indian grammarians and widely accepted by other classical Indian philosophers) known as kāraka theory. Take any declarative sentence of the following sort: (1) Mira kicks the football. Here, the verb “kicks” picks an act (kriyā) of kicking. Mira is the agent (kartṛ) who brings about that act, while the football is the patient (karman). (1) is true if and only if there exists an act of kicking, with respect to which Mira plays the agent-​role and the football plays the patient-​role. Now, consider: (2) Jo perceives the picket fence. Here, the verb “perceives” picks out the act of perceiving; Jo is the agent who brings about that act; the picket fence is the patient with respect to that act. (2) is true if and only if there exists an act of perceiving, with respect to which Jo plays the agent-​role and the picket fence plays the patient-​role. Since this way of analysing the truth-​conditions of the sentence can be generalised to any ascription of awareness-​events, some philosophers—​especially within the Mīmāṃsā and Nyāya traditions—​treat intentional objects of awareness-​events as things that play the patient-​ role with respect to such events. What does this patient-​role consist in? According to one popular definition, a patient is just an object in which the relevant act produces an effect, for example, some property. For instance, in the event of Mira’s kicking the football, the football counts as the patient, because the activity of kicking produces a certain property in the football, namely, motion. Similarly, we could say that in the event of Jo’s perceiving the picket fence, the picket fence becomes the object of Jo’s perceptual awareness, because that awareness-​event produces a certain property in the picket fence. This, in effect, is the proposal defended by the Bhāṭṭa Mīmāṃsakas, that is, the followers 83

84

Nilanjan Das

of Kumārila Bhaṭṭa: a perceptual awareness-​event is about a mind-​independent particular just in case it produces some effect in the mind-​independent particular.13 But this proposal seems mysterious: what is this property that an awareness-​event produces in its object? The Bhāṭṭas do have an answer to this question.14 Their proposal is described by Vācaspati as follows: It's manifestness (prākaṭya). This property isn't a property like whiteness, etc., which is epistemically accessible to everyone (sarva-​puruṣasādhāraṇa). For the property produced in the object by an awareness-​event is observed to be epistemically accessible solely to the subject of that awareness-​event. This is just as in the following case: the subject whose combinative awareness-​event (apekṣābuddhi) produces the property of two-​ness (dvitva) in certain objects is the person to whom that property is specifically accessible. (Thakur 1996, p. 626, l. 1–​4) The thought is roughly this. Some properties of objects are produced by mental states of this or that agent. These properties are different from other properties of objects such as its shape or size: while shape and size of an object may be epistemically accessible to several agents, these mind-​dependent properties are only epistemically accessible to that agent whose mental states produce them. Here's how the theory can be applied to numerical properties. On the view in question, when an agent encounters two objects, she may undergo a combinative awareness of the form, “Here is one thing, and here is one thing.” This in turn produces the property of two-​ness that resides in the two objects together. Then, the agent becomes aware of the two-​ness; she judges, “Here are two things.” Here, two-​ness is accessible only to the agent who undergoes the relevant combinative awareness. How does this help us with our original question? According to the Bhāṭṭas, when an agent undergoes an awareness-​event, that awareness-​event produces a property of manifestness in an entity, which thereby becomes the intentional object of the awareness-​event. However, this property of manifestness isn't epistemically accessible to everyone; it is only epistemically accessible to the agent whose awareness-​event produces the relevant manifestness property. Take the case where Jo is looking at the picket fence. When she is perceptually aware of the picket fence, she is aware of the picket fence, the white paint on it, the rectangular wooden panels that are part of it, and so on. She is also aware of certain perspectival features of the picket fence: the distance between herself and the picket fence, the angle from which she is looking at it, and so on. These perspectival features belong to the picket fence only insofar as Jo is aware of the picket fence under those circumstances. And they are epistemically accessible solely to Jo. These together may be thought of as the manifestness of the picket fence. This is important, because it helps us explain how we become aware of our own awareness-​events, and why we seem to have privileged access to our own awareness-​events. On the basis of these perspectival features, Jo can infer that she is aware of the picket fence. But no one else can. Thus, when an agent become aware of something, the property of manifestness that her awareness-​event produces in the object of awareness is also epistemically accessible to her, and her alone. That is why she is in a unique position to infer the presence of her own awareness-​event on the basis of that manifestness. This proposal might be plausible when it comes to perceptual awareness-​events: since the objects of perception (plausibly) exist at the time of the relevant perceptual experience, it is at least conceivable that a perceptual awareness-​event can cause its object to become manifest. As the Buddhist notices, this view is implausible when it comes to inferential or testimonial awareness-​events directed at past or future objects.15 For example, my testimony-​based 84

85

Vācaspati on aboutness and decomposition

awareness about dinosaurs cannot intelligibly be said to produce any property in dinosaurs themselves. While it is true that I become aware of dinosaurs when I read about them, there is no causal interaction between my awareness-​event and them.16 Vācaspati goes on to describe another proposal possibly defended by another group of Mīmāṃsakas who follow Prabhākara Miśra and therefore are called Prābhākaras. They explain the fact that a certain awareness-​event is only about certain objects and not others by appealing to the fact that each awareness-​event only has the capacity (śakti) to reveal itself and certain other objects.17 This capacity is produced by the same causal conditions that produce the awareness-​event itself. But the Buddhist points out that this proposal suffers from its own share of problems. Any capacity has its own distinctive effect. So, if awareness-​events have a capacity in virtue of which they reveal their objects, we need to explain what the distinctive effect of that capacity is. Now, the object of awareness cannot itself be the effect of the relevant capacity. For, in the case of perceptual awareness at least, the object produces the awareness-​event and therefore cannot be brought about by any capacity that the awareness-​event possesses. Moreover, we also cannot say that the relevant effect of the capacity is some property that the awareness-​event produces in the object. For this won’t be true of past or future objects that our awareness-​events can be about. The moral is clear. It’s extremely hard to explain how an awareness-​event can be only about a certain mind-​independent particular and not others without appealing to some intrinsic feature of that awareness-​event in virtue of which it can represent it. This leads the Buddhist to the following conclusion: Therefore, it's a mere fantasy to posir that an awareness-​event that lacks a phenomenal form has as its intentional object an external entity. So, let an awareness-​event that has an external entity as its intentional object have a phenomenal form! That is to say, an awareness-​event that has the phenomenal form of blue, etc. has as its intentional object something that is blue. For its own nature is blue. (Ibid., p. 626, ll. 11–​14) The claim is this. Appealing to a mental image helps us explain what an awareness-​event is about. If an experience includes a blue mental image (and conditions are normal), it represents a blue particular outside in the external world. This blue mental image, according to the Buddhist, is part of the nature of the awareness-​event; it isn’t distinct from the experience itself. Thus, aboutness is decomposable into internal and external conditions.

How decompositionalism leads to idealism Suppose we grant that we cannot explain how perceptual experiences can represent mind-​ independent particulars without appealing to their phenomenal forms. However, as Dharmakīrti notes in the chapter on perception (pratyakṣapariccheda) in his Detailed Commentary on Epistemology (Pramāṇavārttika), merely appealing to the phenomenal forms of perceptual experiences doesn’t make it easy to explain aboutness either.18 In fact, it leads to a form of epistemological idealism. Here, I describe how Vācaspati explains this argument. On a simple version of the Sautrāntika view, a perceptual experience can be about a certain mind-​independent particular if it resembles that particular in virtue of its phenomenal form, that is, the mental image it involves. In response to this proposal, we may ask whether this resemblance is complete or only partial. On the one hand, there cannot be complete resemblance between an experience and a mind-​independent item. For the former is a mental event, 85

86

Nilanjan Das

but the latter isn’t. On the other hand, partial resemblance alone cannot be enough to generate awareness of a specific mind-​independent particular; there are far too many things that the experience, in virtue of the mental image it involves, may resemble partially.19 In order to solve this problem, the Sautrāntika might point out that his brand of decompositionalism doesn’t explain aboutness merely in terms of resemblance but also by appeal to causal relations between experiences and mind-​independent particulars. So, partial resemblance between the phenomenal form of the experience and a mind-​independent item isn’t sufficient for that experience to be about that item; the experience must also be caused by that item. But this proposal will collapse in a scenario where an agent undergoes two successive perceptual awareness-​events, which involve the same mental image, such that the second awareness-​event is caused by the first. Here’s an example. Suppose you are looking a white picket fence. First, at t1, there is a contact between your visual sense-​modality and the picket fence. Then, at t2, you undergo a sensory awareness-​event which represents something that resembles the white picket fence. Finally, at t3, this sensory awareness-​event produces a further awareness-​event, which involves a similar mental image. For example, the sensory awareness-​event might make you attend to the object more carefully than at the first moment. This second awareness-​event, which has the same (or similar) phenomenal form as the sensory awareness-​event, is also caused by it. Thus, the conditions of causation and resemblance are fulfilled. But then, the preceding sensory awareness-​event (sometimes called immediately preceding homologous condition or samanantara-​pratyaya) could just as well be the object of the following awareness-​event.20 The lesson: mere resemblance or causation plus resemblance cannot explain how a perceptual experience, in virtue of having a phenomenal form, can be about a certain mind-​independent particular. Sautrāntika decompositionalism doesn’t quite work. Vācaspati knows that there are other kinds of decompositionalist proposals that are available to the Sautrāntika philosopher. For Yogācāra-​Sautrāntika philosophers like Dignāga (5th–​6th century CE) and Dharmakīrti, perceptual awareness-​events are essentially non-​conceptual. Even though we may perceptually apprehend an external particular without conceptually distinguishing it from any other objects, we can only act towards external particulars if we are able to conceptually distinguish it from others. This process of conceptually distinguishing the external particular from things of other kinds is what Dharmakīrti calls determination (adhyavasāya).21 Dharmakīrti’s commentator, Dharmottara (8th century CE), turns this into a theory of what perceptual awareness-​events represent. He writes: An source of knowledge (pramāṇa) has two kinds of intentional objects: the apprehended object (grāhya), i.e., the item in the form of which the relevant awareness-​ event arises, and the determined object (adhyavaseya), i.e., that which is to be attained. The determined object is distinct from the apprehended object. The apprehended object of perception is a momentary entity. The object, which is determined by a certain judgement produced by dint of the perceptual awareness, is a sequence of such momentary entities. The object to be attained on the basis of perception is this sequence, since the momentary entity cannot be attained. (Sastri 1994, p. 39) On this view, a perceptual awareness can be said to represent two sorts of things. The first kind of thing—​the apprehended object—​is a momentary mind-​independent item which causes the perceptual awareness and which resembles the perceptual awareness in virtue of the phenomenal form that the latter possesses. The second kind of thing—​the determined object—​is rather a temporally extended mind-​independent item (i.e., an ordinary material object composed of 86

87

Vācaspati on aboutness and decomposition

instantaneous temporal parts), which the subject judges to be what she perceives. This temporally extended entity is what an agent can act towards. On behalf of the Sautrāntika philosopher, Vācaspati entertains a proposal along these lines. Let the awareness-​ event be related by causation-​ plus-​ resemblance to its causal conditions, e.g., attention (manaskāra), sense-​faculties, etc. Still, whatever entity is determined by the awareness-​event is the thing that serves as its intentional object. Since it determines (adhyavasyati) the entity, the intentional object of that awareness-​ event is that entity, and not attention, etc. (Thakur 1996, p. 626, ll. 20–​22) Here’s a way of understanding the proposal. When an agent’s sense-​faculty causally interacts with an external particular, she at first undergoes an initial sensory awareness, which involves a mental image that the agent is immediately aware of. Then, on the basis of that initial sensory awareness, she judges that there’s an ordinary material object with such-​and-​such features before her. The temporally extended sequence of mind-​independent particulars, which is thus determined, becomes the intentional object of the agent’s perceptual awareness-​event. However, the notion of determination that lies at the heart of this proposal is underspecified. Well, what is this determination? It isn’t just apprehension. For two phenomenal forms are indeed not apprehended. The first-​hand awareness (anubhava) is of the form, “This is blue”, and not of the form “These two are blue.” (Ibid., p. 627, ll. 1–​2) If the determination of an object just consists in an apprehension of that object, then an agent’s perceptual awareness of a blue material object would involve two phenomenal forms: the agent would apprehend both the form of the blue mental image and the form of the blue mind-​ independent particular. But then this seems to be incongruent with the ordinary phenomenology of perceptual awareness. This just reinforces the distinction between the apprehended object and the determined object that Dharmottara mentions. So, following Dharmottara, the Sautrāntika philosopher might just claim that the determined object isn’t apprehended at the time of the perceptual awareness; rather, determination involves a projection (āropa) of features of the mental image onto a sequence of momentary mind-​independent particulars. Thus, when I perceive a blue sphere, my perceptual experience involves a blue, sphere-​shaped mental image, and so on. Then, I project the blue colour, the spherical shape and so on, on to a sequence of mind-​independent entities, and take myself to perceive a mind-​independent blue sphere. This yields a different form of decompositionalism.



Dharmottara’s Decompositionalism. For any subject S and a sequence M of mind-​ independent entities, if S undergoes a perceptual experience about M, then the fact that S’s perceptual experience is about M holds solely in virtue of: (i) the fact that S undergoes an experience that involves a certain phenomenal form, and (ii) the fact that S projects certain qualitative properties instantiated by the phenomenal form of that experience onto M.

But even this proposal doesn’t succeed. 87

88

Nilanjan Das

Vācaspati explains why. The Sautrāntika faces a dilemma with respect to the process of projection. Is the relevant mind-​independent entity apprehended before the projection occurs, or not? If the agent already apprehends the mind-​independent entity before she projects certain features onto it, then she is in fact aware of two things: the blue sphere-​shaped image and the mind-​independent blue sphere. This again conflicts with ordinary perceptual phenomenology. This, in turn, leaves us with the conclusion that, before the projection, we aren’t non-​ inferentially aware of any mind-​independent particulars.22 On this reformed Sautrāntika view, a perceptual awareness-​event doesn’t in itself carry any information about a mind-​independent object; it can come to do so only insofar as it produces an inferential judgment that there’s a mind-​independent object which resembles and causes the relevant awareness-​event. This view is vulnerable to several objections.23 The first objection proceeds from the premise that, ordinarily, projection works much like a cognitive illusion, where we judge a mind-​independent particular to be F when it’s not in fact F. If this is right, then it’s not clear that we can project anything onto an external object without being antecedently aware of that external object and its properties in the first place. Suppose, out for a walk on the beach, you perceptually judge a piece of seashell to be silver. Here, you project the property of being silver on to the seashell. But this projection is possible only insofar as you are first aware of the white and shiny seashell. If this is right, then we cannot project the features of our perceptual experiences on to mind-​independent particulars without being antecedently aware of those particulars themselves.24 The second objection is that this view cannot explain how specific actions can be directed towards specific external objects. For example, suppose I have a jumper that I really like, so I wear it as often as I can. But, if I really never apprehend an external object like my jumper, how can I tell that my practical undertaking, with respect to wearing that jumper, is directed at that jumper and not something else that produces the same mental imagery in me? Thus, it’s quite hard to explain how specific types of practical undertakings are directed at specific objects on this view. For our purposes, however, the most serious objection is this. If this Sautrāntika view is true, then an agent can only become aware of or gain knowledge about mind-​independent items by inference. For the judgements, by means of which the intentional objects of perceptual awareness are determined, are produced by an inferential process without any independent awareness of external entities. But the worry is that such an inference can never yield awareness or knowledge about external objects. This, in turn, means that we can never be aware or have knowledge of mind-​independent particulars and properties. Thus, this view leads us to epistemological idealism. As we may expect, the Sautrāntika philosopher might attempt to resist this conclusion. She might argue that the idealist, who thinks that we cannot gain knowledge of the external world by inference, faces an explanatory challenge. She must explain how certain perceptual experiences only arise at certain times but not at others. The standard Buddhist idealist explanation of how our perceptual experiences arise appeals to various memory impressions or traces (vāsanā) contained in storehouse awareness (ālaya-​vijñāna). Storehouse awareness serves as a repository of memory traces that can later be used for conscious mental activity. When these traces mature (due to karmic factors that depend on past good or bad deeds performed by the subject), they give rise to conscious mental activity, for example, conscious experiences and thoughts. However, the Sautrāntika philosopher can argue that, even within this framework, in order to explain how our ordinary experiences and thoughts arise, we need to appeal to the existence of mind-​independent items. The argument (a version of which we saw in Section 1) proceeds from the following general principle: in the presence of a condition C, 88

89

Vācaspati on aboutness and decomposition

if a phenomenon X arises only sometimes but not always, then X causally depends on something other than C. Suppose I undergo an awareness-​event where something that looks like a white picket fence appears. Such an awareness-​event only arises sometimes, even though the ingredients generating that experience (e.g., mental images of whiteness, picket fences, etc. derived from past experiences) may be present in storehouse consciousness. Then, we are forced to conclude that the relevant awareness-​event cannot be explained solely by appealing to the resources for conscious mental activity present in storehouse awareness. Something else—​ for example, a mind-​independent particular—​is required.25 So, the Sautrāntika philosopher can resist the charge that her view leads to epistemological idealism. However, it is far from obvious whether this argument succeeds. For the idealist could, in principle, explain the fact that our perceptual awareness-​events only arise at certain times and not others without appealing to external objects: “Awareness-​events arise only sometimes, because their causal condition—​namely, the maturation of impressions—​also obtains sometimes” (Ibid., p. 627, ll. 10–​1). Here is how Vācaspati imagines the realist could resist this response. [The realist:] Surely, an impression is just the power that different store-​ house awarenesses belonging to the same sequence have of producing a specific occurrent awareness-​event (pravṛtti-​vijñāna, i.e., a conscious experience or thought). The maturation of an impression consists in its state of being about to produce its effect. And the causal condition for that maturation is the previous moment of that storehouse awareness which occurs in the same sequence and causes the present moment of the store-​house awareness to exist. For you don’t accept the view that the maturation of the impressions can depend on a distinct sequence. But then, either all the mental events that belong to that sequence of storehouse awarenesses must serve as the cause of that maturation, or none of them. For they aren’t distinct from each other. [The idealist:] There are differences amongst effects due to differences amongst the moments [of storehouse awareness]. So, since certain moments occur only at certain times, the effects can also only arise at certain times. [The realist:] Surely, if that were the case, only one time-​slice of storehouse awareness would have the capacity to produce an awareness of blue, etc., as well as the capacity of activating the power to give rise such an awareness. Other moments [of storehouse awareness] couldn’t have it! For, if they had it, then how could there be differences amongst effects due to differences amongst moments [of storehouse awareness]? So, all the moments belonging to the same sequence of storehouse awarenesses would be capable of making the impression mature. And it is impossible for an effect not to arise when the causal conditions capable of producing it are present. (Ibid., p. 627, ll. 11–​18) The idealist story about how our awareness-​events arise only sometimes but not at others requires us to say that different time-​slices of the same mental stream have arbitrarily different causal capacities. This is because the idealist wishes to appeal to the maturation of impressions carried by storehouse awareness. An impression is just a dispositional state: it is the power of different time-​slices of storehouse awareness occurring within the same mental stream to produce experiences or thoughts of a certain kind. But, if the idealist is right, although all time-​ slices of storehouse awareness may carry the same impressions, only some of those time-​slices may be capable of producing a specific awareness-​event by activating a specific impression. But this seems difficult to accept, since ex hypothesi the time-​slices all have the power to generate the relevant awareness-​event. 89

90

Nilanjan Das

Prima facie, therefore, the Sautrāntika does have an argument for the existence of external objects: “Awareness-​events of blue, etc., are not caused by the mind-​sequence in which they occur, since they only arise sometimes and not always.” The argument seeks to show that perceptual awareness-​events have a certain property, that is, the property of not being caused by the stream of mental events in which they occur. This is the target property (sādhya) of the inference. The inference seeks to establish the target property by appealing to another property of those awareness-​events, that is, the property of arising sometimes and not always. This is the reason property (hetu) that provides evidence for the target property. In her refutation of the Sautrāntika realist, the idealist notes that, in order for this inference to be successful, the reason property has to be exclusive: whenever it is present, the target property has to be present. And the idealist wants to say that the reason property fails to be exclusive in this sense.26 Why? The idealist repeats herself: we don’t need to appeal to any external object—​that is, any object external to the stream of mental events in which an awareness-​event occurs—​in order to explain why that awareness-​event occurs at a certain time but not at others. So, the fact that an awareness-​event only occurs sometimes but not others is compatible with the claim that awareness-​events are caused solely by conditions that are internal to the stream of mental events in which they occur. Thus, the reason property can be present in an awareness-​event even when the target property is absent. This makes the reason property non-​exclusive (anaikāntika): its presence fails to exclude the absence of the target property. The idealist explains this thought in two steps. First, the crucial problem for the Sautrāntika realist, as the idealist notes, is that her position itself involves a tension. The realist is committed to the claim that ordinary material objects—​which are conceptually constructed streams of time-​slices—​can causally affect an agent’s perceptual awareness-​events in distinct ways. At the same time, she doesn’t want to say (as the idealist does) that the difference amongst time-​slices of storehouse awareness can causally explain why perceptual experiences only arise at certain times and not others. So, her position is somewhat inelegant. On the one hand, she thinks that these conceptual constructions can account for the difference in effects. On the other hand, she thinks that ultimately real differences amongst time-​slices of storehouse awareness can’t.27 By contrast, the idealist’s position is more coherent: she thinks that these differences amongst time-​slices can explain why different awareness-​events arise at different times and not always. In response, the realist could highlight the problematic consequences of the idealist’s position: if the idealist is right, then the different time-​slices of storehouse awareness can’t possibly carry the same impression, that is, the power to produce the same awareness-​event. This seems false.28 This is where the idealist can make her second point. She could point out that the realist faces an analogous challenge. Note that sometimes, two distinct material objects can produce phenomenally or introspectively indistinguishable perceptual experiences. Either the realist must say that these experiences are the same, or she must say that they are distinct. If they are the same, then there may be a sense in which different momentary time-​slices of storehouse awareness could produce the same experience or thought. If they are distinct, then the idealist could also say that different momentary time-​slices of storehouse consciousness produce distinct experiences and thoughts.29 So, this isn’t really a problem for the idealist. Thus, on balance, the realist’s position seems much worse than that of the idealist. So, the reason property of the realist’s inference hasn’t been shown to be exclusive with respect to the target property. This, in turn, clears room for the idealist to say: “Therefore, even though awareness-​events may have objective phenomenal forms, no [external] entity can be inferred, whether or not it is detected by the senses” (Ibid., p. 628, ll. 11–​2). So, we cannot non-​ inferentially or inferentially know anything about mind-​independent particulars. 90

91

Vācaspati on aboutness and decomposition

While the realist argument fails, the epistemological idealist can still run her own argument: “Intentional objects aren’t distinct from the mind, since they can be apprehended, just like feelings” (Ibid., p. 628, ll. 13–​14). Everyone agrees that there are intentional objects, for example, the blue thing that appears in a perceptual experience as of something blue. The debate between the realist and the idealist is about whether such objects are distinct from awareness (and its intrinsic features) or not. According to the idealist, apprehensibility (i.e.., the property of being something that can be apprehended) is invariably accompanied by the property of being non-​distinct from awareness and its features, for example, in the case of feelings (i.e., hedonic states like pain and pleasure) which aren’t distinct from the awareness-​events that apprehend them. So, since intentional objects can be apprehended, they must also be non-​ distinct from the awareness-​events that apprehend them. The upshot is this. If we accept the Sautrāntika realist’s claim that a perceptual awareness-​ event can be about a mind-​independent particular only in virtue of possessing a phenomenal form, we need to say how this happens. The idealist argues that no good answer to this question can be given. This undermines the claim that perceptual awareness-​events can carry information about mind-​independent particulars. This paves the way for epistemological idealism.

The argument against idealism Vācaspati agrees that the Sautrāntika brand of decompositionalism leads to epistemological idealism. However, he thinks that epistemological idealism is incoherent. According to the epistemological idealist, the intentional objects of our awareness-​events aren’t distinct from those awareness-​events (or their intrinsic features). But this view is untenable. Here, first, this is to be said. If apprehensibility were pervaded [i.e., invariably accompanied] by the property of not being distinct from awareness, then distinction would be refuted. But this is not pervaded by such non-​distinctness. That is to say, in a case where one undergoes a first-​hand awareness of a blue object that is spatially extended and separate, its extension and separateness isn’t the nature of the awareness-​event. Extension involves pervading multiple spatial regions. Since an awareness-​event is simple, non-​spatial, and impartite, it cannot pervade multiple spatial regions. This is just as has been said [by Dharmakīrti:] Therefore, the appearance of spatial extension belongs neither to the object nor to the awareness. Since the nature [of being extended] is refuted with respect to one [i.e., atoms], it is also not possible with respect to many.30 In the same manner, separateness from an object also not the nature of an awareness-​ event. If it were so, then an awareness-​event would ultimately be separate from an object. Therefore, since separateness is also not [part of the nature of] the awareness-​ event, it must be accepted as non-​existent. Thus, since that which exists cannot be identical to that which doesn’t exist, you—​even if you don’t like it—​would have to accept that awareness manifests that which is distinct from it. (Thakur 1996, p. 628, l. 19–​p. 629, l. 5)31 The argument is complicated, so let me unpack it. In any case of perception, we seem to be perceiving something that is spatially extended and located away from different objects. Spatial extension (sthaulya), for Vācaspati, is the property of pervading different regions of space (nānā-​dig-​deśa-​vyāpitā), while separateness (viccheda) is the property of occupying 91

92

Nilanjan Das

separate regions (vicchinna-​deśatā) from other things.32 These represented properties of spatial extension and separateness cannot belong to the nature of the perceptual awareness. After all, perceptual awareness-​events cannot be spatially extended, and, if epistemological idealism is true, they can’t be located away from any of their intentional objects. This yields the conclusion that what perceptual awareness-​events represent doesn’t exist. If that is right, then what is represented by perceptual awareness-​events must be distinct from awareness-​events themselves. That already contradicts the conclusion that the epistemological idealist arrived at. To make things clearer, we can formalise the argument as follows (by supplying a missing premise). The argument from spatial properties Idealism. If epistemological idealism is true, the intentional objects of our perceptual awareness-​events are not distinct from them or their intrinsic features. Spatial Content. Some intentional objects of our perceptual awareness-​events have spatial properties that cannot belong to those awareness-​events or their intrinsic features. The Missing Premise. For any two entities x and y, if x has certain properties that cannot belong to y, then x is distinct from y. Conclusion 1. Some intentional objects of our perceptual awareness-​events are distinct from them and their intrinsic features. Conclusion 2. Therefore, epistemological idealism is false. In response to this argument, the idealist might concede that the intentional objects of our perceptual awareness-​events are not identical to those awareness-​events, but are in fact non-​ existent. But, even then, the conclusion will remain intact. For that which exists cannot be identical to that which doesn’t exist. As long as the idealist concedes that awareness-​events can be about something other than themselves, they will be contradicting the conclusion of the earlier idealist argument. So epistemological idealism is incoherent. Suppose we grant that our perceptual awareness-​event can represent non-​existent objects that are distinct from them and their intrinsic features. Vācaspati now says that, if the idealist allows, they must also allow that these awareness-​events can represent existent particulars and properties that are distinct from them and their intrinsic features: “Then, even an existent object [which is distinct from awareness] could be manifested by it” (Thakur 1996, p. 629, l. 5). The rest of Vācaspati’s discussion is a defence of this claim. Call it The Analogy. Vācaspati starts out from the observation that the epistemological idealist must herself tell some story about how perceptual awareness-​events can represent non-​existent objects. Whatever that story about aboutness is, there will be an analogous—​if not better—​story available to the Nyāya realist about how perceptual awareness-​events can represent existent objects. So, the Analogy cannot be easily be rejected.

The nature of aboutness In this section, I will show how Vācaspati defends the Analogy by appealing to a non-​ decompositionalist account of aboutness.

A. Aboutness as a natural relation The idealist could try to reject the Analogy by offering the following argument.

92

93

Vācaspati on aboutness and decomposition

The idealist argument from causal inefficacy The Act Model of Awareness. If there is an intentional object of an awareness-​event that is distinct from the awareness-​event and its intrinsic features, then it must play the role of a patient in relation to the act of awareness. The Patient-​role. If something plays the role of a patient in relation to an act of awareness, then the act must produce a result in that object. Causal Inefficacy. If there is an existent intentional object of an awareness-​event that is distinct from the awareness-​event and its intrinsic features, then the relevant act of awareness needn’t necessarily produce a result in that object. Conclusion. Therefore, there is no existent intentional object of an awareness-​event that is distinct from the awareness-​event and its intrinsic features. The Act Model of Awareness is motivated by ascriptions of awareness-​events, like “Jo sees the picket fence.” The Patient-​role is motivated by the commitment of kāraka theory, which underlies the Bhāṭṭa theory of aboutness. Causal Inefficacy is plausible at least given the failure of the Bhāṭṭa view of aboutness. If this argument succeeds, then the realist is in trouble. Vācaspati responds to this worry in three stages. The first part of the response involves pointing out that the same challenge can be raised against the idealist, who (according to the reasoning presented in the last section) must also think that perceptual awareness-​events can represent distinct non-​existent objects. [The idealist:] How can an awareness-​event, which doesn’t causally affect an object that is distinct from it, manifest it? [Vācaspati:] How does an awareness-​event, which cannot causally affect the spatial extension, manifest this spatial extension? (Ibid., p. 629, ll. 6–​7) In response, the idealist could argue that there is something about the nature of awareness that allows it to represent such non-​existent objects. But the same idea could also be deployed by the Nyāya realist. [The idealist:] By its own nature. [Vācaspati:] There is nothing wrong with saying the same with respect to external objects. If something is related to another thing in virtue of its own nature, it doesn’t depend on anything else in order to be so related. This is just as in the case of an inferential mark and its relation to the target property. Since an awareness-​event is about a certain object in virtue of its own nature, then its being about the object doesn’t depend on anything else. (Ibid., p. 629, ll. 8–​9) The key idea here is that the relation of aboutness between an awareness and its object is a natural relation. The thought can be explained using the example that Vācaspati mentions. According to Vācaspati, it is possible for an agent to become certain that an object o is F on the basis of the inference, “The object o is F, because it is G,” only if the agent is able to ascertain beforehand that the reason property (or the inferential mark)—​that is, the property of G-​ness—​bears a natural connection (svābhāvika-​pratibandha) to the target property—​that is, the property of F-​ness. Vācaspati explains the idea as follows. 93

94

Nilanjan Das

Therefore, whatever the relation may be, it’s reasonable to say that if it is a natural and invariable relation that belongs to a certain entity, then that entity is the indicator (gamaka) [i.e., the reason property], and the other relatum is the indicated object (gamya) [i.e., the target property]. Thus, the relation of smoke, and so on with fire, and so on is natural, not the relation of fire, etc. with smoke, and so on. For those fire, and so on are indeed apprehended without smoke, and so on. In contrast, when they become connected to wet fuel, and so on, then they are related to smoke, and so on. Therefore, the relation of fire, and so on [with smoke], which obtains in virtue of an incidental condition (upādhi), i.e., wet fuel, and so on, is not natural, and therefore isn’t invariable. In contrast, the relation of smoke, and so on with fire and so on is natural, since no incidental condition for it is apprehended. For, since no deviation [i.e., a case where the reason property is present, but the target property is absent] is observed, it doesn’t make sense to postulate an unapprehended deviation. Therefore, an invariable relation is a constitutive element of an inference. And it is not appropriate to say that the naturalness of a relation is blocked by an unobserved incidental condition which is suspected to be observable. For there is no evidence that either proves or disproves its existence. (Ibid., p. 135, ll. 8–​16)33 The passage contains a number of important ideas, so let me explain them carefully. Compare two inferences: (3) The hill contains smoke. Therefore, the hill contains fire. (4) The hill contains fire. Therefore, the hill contains smoke. The inference in (3) can generate certainty in the conclusion that the hill contains fire only if the relevant agent is antecedently certain that fire always accompanies smoke. In other words, the agent must be sure that the relation of smoke with fire is an invariable one; fire never fails to obtain at a place where smoke exists. How can one become so sure? The answer is that one must ascertain that relation of smoke with fire is natural, that is, not an artefact of some other background incidental condition (upādhi). An incidental condition is a condition such that (i) in the presence of the condition, the intended target property accompanies the intended reason property, but (ii) in the absence of the condition, the intended target property may not accompany the intended reason property. So, if there were an incidental condition that induces the connection of smoke with fire, then, in its absence, smoke could be present at a place where there is no fire. If an agent isn’t able to find any such condition even after searching thoroughly for it, she can be certain that the relation of smoke with fire is natural and therefore invariable. Compare this inference to the inference in (4). The latter can generate certainty in the conclusion that the hill contains smoke, only if the relevant agent is antecedently certain that smoke always accompanies fire. In other words, the relevant agent has to be sure that the relation of fire with smoke is invariable. But Vācaspati points out that an agent (who undertakes a thorough investigation) cannot easily be sure of this. This is because even though the agent may perceive smoke and fire occurring together on numerous occasions, she will in fact find an incidental condition, that is, contact of fire with wet fuel, such that: (i) in the presence of that condition, fire invariably gives rise to smoke, and (ii) in the absence of that condition, fire doesn’t give rise to smoke. Given that the connection of fire with smoke depends on such an incidental condition, the relation between them cannot be natural. Since the agent cannot ascertain that the 94

95

Vācaspati on aboutness and decomposition

relation of fire with smoke is natural, she cannot take it to be invariable. So, she cannot take the relation for granted for the purposes of inferring smoke from fire. Let’s see how this applies to the case of awareness-​events and their intentional objects. If Sautrāntika decompositionalism about aboutness were true, then the relation between an awareness-​event and its intentional object couldn’t be natural. Why? According to Sautrāntika decompositionalism, the relation of aboutness that connects a perceptual awareness to its intentional object doesn’t belong to the nature of the perceptual awareness itself: it obtains in virtue of an internal condition, namely the phenomenal form of the awareness-​event, and some further external condition like causation and resemblance. If the relation of aboutness that an awareness-​event bears to its intentional object could be decomposed in this way into internal and external conditions, then the presence of the awareness-​event would be accompanied by the presence of the intentional object only when the background external conditions were in place. So, the external conditions would serve as the incidental condition due to which the relation of the awareness with its intentional object holds. This, in turn, would mean that the awareness-​event could sometimes obtain even without an intentional object. So, even though an agent may undergo a veridical perceptual experience about a yellow banana in one situation, it is in principle possible for an agent to undergo a hallucination that has the same phenomenal form, even though the relevant intentional objects—​the banana, its yellowness, and so on—​are absent. This is not how things in fact are. Whenever we undergo an awareness-​event, the awareness-​ event manifests something. So, given that we can’t find any case where there is an awareness-​ event but no intentional object, we must take the relation between awareness-​events and their intentional objects to be a natural one. More specifically, it means that the relation of aboutness that connects awareness-​events with their respective intentional objects isn’t dependent on any further incidental condition such that (i) the presence of that condition ensures that awareness-​ events are accompanied by their intentional objects, but (ii) in the absence of that condition, an awareness-​events could be present without the presence of any such intentional object. More generally, this means that when an agent is aware of an object, she undergoes a mental event or state of a certain kind, such that an agent couldn’t undergo a mental event or state of that kind unless that object were present. Why is this significant? The challenge that the idealist raised against the realist was this: how can a relation of aboutness obtain between an awareness-​event and its (existent and distinct) intentional object even though the awareness-​event cannot affect its intentional object in any way? The realist’s response was that the idealist could be asked the same question, since the idealist (as shown by The Argument from Spatial Properties) also should think that perceptual awareness-​events represent non-​existent objects that are distinct from them and their intrinsic features. Now, if the idealist’s claim is that perceptual awareness-​ events do this in virtue of their own nature, the realist could make the same appeal to natural relations in order to explain how perceptual awareness-​events represent existent objects that are distinct from them and their intrinsic features. The point is that, if an entity x bears a natural relation to an entity y, nothing over and above the existence of x is needed for the relation to obtain; x doesn’t have to causally impact y in order to be so related to it. The point is clear in cases where the natural relation is a causal one: smoke doesn’t have to causally impact fire in order to be naturally related to it. The point is also clear in the case of other non-​causal natural relations, for example, the relation between certain properties of precious stones (“rainbow-​like lustre”) and the property of being a specific kind of jewel like a ruby (Ibid., p. 136, l. 22).34 The same, Vācaspati claims, is true of the natural relation of aboutness. He explains the point as follows. 95

96

Nilanjan Das

Suppose you ask, “How can the awareness of a material object—​which is causally inefficacious with respect to it—​have the material object as its intentional object?” Is the expression “awareness of a visible entity” (rūpavijñāna) supposed to pick out a means of knowing, or its result? If it is a means of knowing, then it isn’t causally inefficacious with respect to the material object, since it causes acts of avoidance, grasping, and indifference to arise. If it is a result, then the question is unreasonable. What result of an act is causally efficacious with respect to the patient of that act? Caitra’s arrival at the city, which is caused by his movement, doesn’t causally affect the relation of inherence by which the contact between Caitra and the city resides in the city. In the same way, in order for there to be a natural relation of aboutness between an awareness and its object, the awareness doesn’t have to produce some further effect. (Ibid., p. 629, ll. 14–​20) The argument is this. On the idealist’s view, if an awareness of a material object were possible, then it couldn’t causally affect that entity, and if it couldn’t affect that entity, it couldn’t be about that object. The underlying idea is that an awareness, insofar as it is an act or event, can only have a distinct entity as its patient if it could produce some result in the entity. Vācaspati’s response is this. If we take the awareness of the material object to be a means of knowing whether the object is to be obtained, avoided, or neither, then it can indeed affect the relevant material object by generating the relevant actions. If we take the awareness of the material object merely to be an awareness in which the relevant causal process culminates, then it’s just a mistake to ask whether it is causally efficacious with respect to the material object. Results of acts or events don’t have to be causally efficacious. For example, consider Caitra’s act of going to the city. Here, the city plays the role of the patient with respect to the act of going. The result of that act is Caitra’s arrival at the city, which in turn consists just in the physical contact (saṃyoga) between Caitra and the city. But once this result obtains, it doesn’t produce any further effect in the city. Similarly, once the agent’s perceptual or cognitive processes has culminated in an awareness-​event directed at a certain object, the awareness-​event doesn’t have to produce any further effect in the object in order to be about it.

B. Aboutness as a self-​linking relation This response to the idealist might seem uninformative. For we can still intelligibly ask what sort of natural relation obtains between an awareness-​event and the material object that serves as its object. Is it a causal relation like the relation between smoke and fire? Or is it a non-​causal relation like the relation between rainbow-​like lustre and the property of being a ruby? If it is the former, then the proposal cannot work for awareness-​events that represent past or future objects. If it is the latter, then the intentional object must just be a feature of the awareness-​ event itself just as rainbow-​like lustre is a feature of the ruby. If the realist wants to say that the aboutness relation is neither a causal nor a non-​causal natural relation, then there can’t be any natural relation between the awareness-​event and its intentional object. Vācaspati’s first reaction is that, if the question is simply in virtue of what relation the natural relation between the awareness-​event and its intentional object holds, then the question may not have an informative answer. [The idealist:] If the awareness-​event isn’t related to its object, how can it be of the object? 96

97

Vācaspati on aboutness and decomposition

[Vācaspati:] How can a relation obtain between two relata? If one posits a further relation, there would be a regress. Therefore, a relation can obtain between two relata without any further relation. Similarly, an awareness-​event can be of an object without a further relation. There is no problem of overgeneration. A certain awareness-​event, produced by its causal conditions, can only be of a certain object, not another. One cannot subject the very nature of things to one’s commands or complaints. The same goes for your questions like “Let this not be the case! And why is it not so?” (Ibid., p. 629, ll. 9–​14) For any property that an entity a may possess, we can ask if there is a further relation in virtue of which a possesses that property. But if we keep positing new relations in response to such questions, we will be in trouble. For example, take any entity a such that a bears the relation R to b. We can ask: By which relation does the property of being R-​related to b reside in a? Suppose we say that there is a further relation R* by which this property resides in a. We can then ask: By what relation does the property of R*-​related to the property of being R-​related to b reside in a? Thus, positing new relations (or properties) to explain existing properties launches us on an infinite regress. At some point, we would have to appeal to the very nature of a, and say that the fact that a is R-​related to b is at least partly explained by a brute fact about the nature of a. This brute fact isn’t explainable in terms of some further relation or property. Similarly, according to Vācaspati, the fact that an awareness-​event is about an object should be treated as a brute fact about that awareness-​event. It isn’t explainable in terms of some further deeper fact about the awareness-​event. The idea can be explained once again using the example of Caitra’s going to the city. When Caitra goes to the city, he comes into contact (saṃyoga) with the city. We may ask, “In virtue of what relation is Caitra in contact with the city?” Since contact is a quality (guṇa) in the Nyāya-​Vaiśeṣika ontological scheme, such qualities always reside in substances by a relation of inherence (samavāya). But, once again, we can ask, “In virtue of what relation does the contact inhere in the city?” Given that there is a danger of an infinite regress here, the Nyāya-​Vaiśeṣika philosophers claim that there is something about the nature of one of the relata, for example, the inherence, in virtue of which it is related to the other, that is, the city. Such a relation, in later Nyāya-​Vaiśeṣika ontology, came to be called a self-​linking relation (svarūpa-​sambandha): if an entity x is related to y by a self-​linking relation R, then there is something about the nature of x that explains how it is related to y. On Vācaspati’s view, a feature of a self-​linking relation like this is that the nature of one of its relata cannot be determined without determining the nature of the other. For example, the nature of the inherence relation by which the contact with Caitra resides in the city cannot be determined without determining the nature of the city itself. An awareness-​event’s property of being about an object is just its being invariably connected to the object. We have said that an awareness, which is to be determined by an introspective awareness, cannot be so determined unless its object is determined. If you ask, “How can that which doesn’t inhere in the patient be of the patient?” then we can say, “A relation of inherence that connects the result of the action with its patient doesn’t inhere in the patient, so how can it be of the patient?” If you say, “It’s because it is dependent on the patient,” then we can ask, “What is this dependence on the patient?” If it is the property of being determined on the basis of the determination of the patient, then the same can be said here on the basis of different considerations. (Ibid., p. 629, ll. 21–​23) 97

98

Nilanjan Das

The claim then is that the relation of aboutness that connects the act of awareness with its intentional object is a self-​linking relation, such that the nature of the awareness cannot be determined without determining the nature of its intentional object. For example, when I am looking at the picket fence outside my window, I cannot undergo an awareness of the form, “I am seeing that picket fence”, without determining what it is I am seeing. In this sense, our epistemic access to the facts about what awareness-​events we are undergoing depends on our epistemic access to what objects we are aware of. Thus, Vācaspati seems to reduce the relation of aboutness to a self-​linking relation: in order to explain how an awareness-​event is related to its intentional object, we don’t have to appeal to a further relation, only to the nature of the awareness-​event itself. This doesn’t imply that the intentional object is identical to the awareness-​event, or is a part of its nature. It just means that there is a constitutive relation between the awareness-​event and its intentional object; an awareness-​event of that kind couldn’t obtain unless it was about the relevant intentional object. Moreover, self-​linking relations involve a kind of epistemic dependence, whereby the nature of one relatum cannot be determined without determining the nature of the other relatum. This is true of awareness-​events: in order to introspectively determine that an awareness-​event obtains, the agent must antecedently determine that its intentional object obtains. This, once again, conflicts with the Sautrāntika decompositionalist picture. For, on that picture, we can determine whether we are undergoing an awareness-​events simply by becoming aware of certain intrinsic properties of the awareness-​event, for example, its phenomenal form; we don’t have to be determine the intentional object of awareness in order to gain such self-​knowledge.

C. Aboutness as an irreflexive relation Vācaspati’s last move will be deemed problematic by Yogācāra idealists like Dharmakīrti. For they are committed to: Reflexivism. An awareness-​event constitutes a non-​inferential awareness of itself. In fact, Reflexivism is the basis for an idealist argument that Dharmakīrti offers in The Ascertainment of Epistemology. A blue thing and its awareness are not distinct because they are invariably apprehended together. Even though the two appear as different, the form of the blue thing isn’t distinct from the experience of the blue thing. For the two are invariably apprehended together, like the double-​moon seen by someone with eye-​disease (timira). There is no apprehension of one of the two when the form of the other is not apprehended. And this is not possible if they are distinct in nature because nothing else could cause them to have this connection. (Pramāṇaviniścaya 1.54ab in Steinkellner 2007, pp. 39–​40) If Reflexivism is true, then, whenever we are aware of any putative object of perceptual awareness, for example, a blue object, we are also aware of the awareness itself. But this fact cannot be explained if the blue object and its awareness are distinct. For example, a causal theory of representation cannot account for this, because an awareness cannot cause itself, and therefore cannot be about itself. Similarly, a resemblance-​based theory cannot account for this either; for, as we have seen, resemblance-​based accounts of aboutness suffer from problems of 98

99

Vācaspati on aboutness and decomposition

overgeneration. From this, this Dharmakīrti concludes that the blue object that seems to appear in perceptual awareness isn’t distinct from the awareness itself. So, epistemological idealism is true. Why might Reflexivism be true? Dharmakīrti notes that Naiyāyikas and Mīmāṃsakas could insist that an object and its awareness aren’t apprehended together: first, we become aware of the object, and then, by means of the manas or the inner sense, we become introspectively aware of that first awareness. Dharmakīrti offers an answer: The awareness of an object cannot be established for someone who isn't directly aware of his apprehension. The apprehension of an object is not due to the existence of the intentional object. What, then? It's due to the existence of the apprehension of that [intentional object]. And if the existence of the apprehension is not revealed by a means of knowing, then it does not attach itself to everyday practices that are based on its existence. Therefore, if the apprehension were unestablished, then the object would also be unestablished. If that were the case, the entire world would vanish! For even if something exists, we cannot act on the supposition that it exists unless it is established. Therefore, someone who is not directly aware of her awareness-​events is not aware of anything at all. (Pramāṇaviniścaya 1.54 cd in Ibid., pp. 40–​1) The argument is this. Dharmakīrti’s Argument The AA Principle. If an agent is able to act (i.e., make judgments and speech reports, or perform physical acts) on the basis of an awareness-​event, then the agent must be aware of that awareness-​event. The Principle About Action Guidance. If an agent undergoes an awareness-​event, then she is able to act on the basis of that awareness-​event. Conclusion. Therefore, if an agent undergoes an awareness-​event, she must be aware of that awareness-​event. Dharmakīrti’s reason for accepting the AA Principle is just that if an agent is to act towards an object on the basis of an awareness-​event, she must be aware of the awareness-​event that serves as a means of knowing with respect to that object. The Principle About Action Guidance is implicit, and nowhere defended. But it is plausible enough. Together, these two principles entail that if an agent undergoes an awareness-​event, she must be aware of that awareness-​event. The Naiyāyikas cannot accept this conclusion. If the conclusion of Dharmakīrti’s Argument is correct, and the Naiyāyikas are right in thinking that any awareness of an awareness is distinct from that awareness, then there will be a regress: without undergoing an infinite sequence of awareness-​events and higher-​order awareness-​events about them, it won’t be possible for an agent to act on the basis of any awareness-​event. If the awareness of the object is apprehended by another awareness-​ event, then the latter will itself be unestablished [i.e., won't be an object of awareness]. So, it won't establish [i.e., bring about an awareness of] the first-​order awareness. Thus, another awareness must arise. Then, if a person were to wait for the end of these 99

100

Nilanjan Das

awareness-​events, he wouldn't be aware of anything. For when one awareness-​event is unestablished [i.e., not an object of awareness], all of them are. And since there would be no end to the production of these awareness-​events, the world would be blind and mute. (Pramāṇaviniścaya 1.54 cd in Steinkeller 2007, p. 41) This gives us reason to accept Reflexivism. If Reflexivism is true, then the claim that the relation of aboutness consists in a self-​linking relation and therefore involves a kind of epistemic dependence between awareness-​events and their intentional objects will become hard to defend: if the apprehension of an object depends on the apprehension of the awareness (as Dharmakīrti seems to suggest), then the determination of the awareness cannot depend on an antecedent determination of the object. So, Vācaspati is wrong. Vācaspati resists this line of reasoning. Following his predecessor Uddyotakara (6th century CE), he thinks that Reflexivism is highly counterintuitive if we think of awareness-​events as cognitive acts of a certain kind. In his sub-​commentary on the Nyāyasūtra4.2.34, Uddyotakara writes: “An act cannot be the same as its patient” (Nyāyavārttika on Nyāyasūtra 4.2.34 in Thakur 1997, p. 490, l. 2). To see what this means, recall the view on which any episode of awareness is an act and its intentional object counts as its patient. Uddyotakara’s claim basically is that an awareness-​event cannot be its own object, since there is a general metaphysical truth about acts, namely that an act cannot be its own patient. So, the aboutness relation between an awareness-​ event and its intentional object is irreflexive. To explain the idea, Vācaspati quotes a verse from Kumārila Bhaṭṭa’s lost work The Great Commentary (Bṛhaṭṭīkā): “Just as the tip of a finger cannot be touched by itself, so also an awareness-​event cannot be apprehended by itself ” (Thakur 1996, p. 630, ll. 3–​4). This contradicts Reflexivism. Here, Vācaspati anticipates a response from defenders of Reflexivism. The Buddhist could argue as follows. If an awareness-​event is to manifest anything to an agent, the agent must become aware of those intrinsic properties in virtue of which it represents that thing. Thus, awareness of awareness is necessary for any awareness whatsoever.35 If this is right, then any awareness-​event must constitute an awareness of itself. However, Vācaspati finds this response implausible. This is the import here. What, then, is this property of being manifested in virtue of the manifestation of the awareness? It is not the case that, due to the manifestation of the awareness, a different manifestation with respect to an object is produced. Moreover, the manifestation of a target property takes place only when the manifestation of the inferential mark [i.e., the reason property] occurs. And the target property isn’t manifested when the inferential mark is manifested, but rather when the manifestation of the inferential mark is destroyed. Suppose you say, “The manifestation of the object or of the awareness isn’t anything distinct from the awareness. The awareness itself, insofar as it consists in reflexive awareness, is both of the object and itself.” If this is so, what follows? A thing cannot depend on itself. Neither is it established that an awareness can manifest both itself and something else. For the manifestation of awareness is due to the manas [i.e., the inner sense], while the manifestation of the object is due to the senses. Nor is the manifestation of the object dependent on the manifestation of the awareness that serves as a means of knowing the object; it depends only on the existence of the awareness that serves as a means of knowing the object. It has been made known that nothing here is dependent on any effect. Thus, it is established that the awareness, which is the topic of dispute, is 100

101

Vācaspati on aboutness and decomposition

distinct from its own manifestation. For it is manifested, just like the awareness of one being with respect to the awareness of another. (Ibid., p. 630, ll. 14–​22) The thrust of this argument is just that it’s unclear how the intentional object of an awareness-​ event could be manifested on the basis of an awareness of the awareness-​event. First, it cannot be the case that the agent first becomes aware of the awareness, and then the object is manifested. For that would mean that the initial awareness of the awareness produces a distinct awareness of the object. If that were true, then the process in question would be exactly analogous to an inferential process where one infers the presence of smoke on a mountain the basis of fire. In such a scenario, one first becomes aware of the reason property (e.g., the smoke), but then that initial awareness disappears, making way for the final awareness of the target property (e.g., the fire). But that’s not how our awareness of objects is generated: in ordinary cases, we don’t become aware of objects on the basis of an initial awareness of our awareness itself. Second, one also cannot say that an awareness-​event simultaneously manifests both itself and its object. On the one hand, this would lead to a problem of self-​dependence (ātmāśraya). Since, on this picture, the manifestation of an object of awareness is supposed to depend on the manifestation of the awareness, this would make the manifestation of the awareness depend on the manifestation of awareness. But dependence is essentially irreflexive: a thing cannot depend on itself. On the other hand, it is far from obvious that our awareness of our awareness-​events is identical to our awareness of the first-​order objects of awareness. It seems that we often become aware of our conscious mental occurrences only by turning our capacity for attention (i.e., the manas or the inner sense) on those mental occurrences, while we become aware of objects of perception by turning our senses towards them. Finally, Vācaspati turns to Dharmakīrti’s own argument for Reflexivism. Vācaspati denies the conclusion of the argument: he claims that the awareness of an object cannot require an awareness of the means of knowing by which the object is revealed. According to Vācaspati, the existence of an awareness is sufficient for the manifestation of its object: no further awareness of the awareness that serves as a means of knowing with respect to that object is necessary. But it remains unclear which premise of Dharmakīrti’s argument he rejects. On the one hand, he might be rejecting the AA Principle, which was motivated by the thought that in order to act towards an object on the basis of an awareness-​event, an agent must be aware of the awareness which serves as the means of knowing that discloses the object. On the other hand, he might be rejecting the Principle About Action Guidance, which just says that any awareness-​event is able to guide actions with respect to its object. Both these moves seem plausible in light of Vācaspati’s earlier criticism of Reflexivism. What’s important, for our purposes, is Vācaspati’s idea that the relation of aboutness cannot be reflexive: an awareness cannot be about itself. Once again, this idea is not only incompatible with idealism, but also with Dharmottara's decompositionalism. Dharmottara seeks to decompose the relation of aboutness into two components: the phenomenal form of the awareness and the act of determination that the agent makes on the basis of her immediate awareness of that phenomenal form. This proposal can only work if some version of Reflexivism is true. Since Vācaspati rejects Reflexivism, he also rules out this view.

Taking stock Let’s take stock. In this chapter, by focusing on the work of Vācaspati Miśra, I have attempted to explain a debate about aboutness in classical Indian epistemology and philosophy of mind. 101

102

Nilanjan Das

For the Sautrāntika Buddhists, the fact that an awareness-​event—​for example, a perceptual experience—​carries information about a mind-​independent entity is explainable in terms of certain further facts about that awareness-​event. The first of these facts is that the awareness-​ events involves a phenomenal form, that is, a mental image that the agent is immediately aware of. The second is that this mental image stands in some representation-​g rounding relation, like resemblance, to the mind-​independent entity. I went on to explain how the Yogācāra idealists took this view to undermine realism. In response, Vācaspati offered an argument against these idealists. To support this argument, he painted a picture of aboutness that is incompatible with the Sautrāntika philosopher’s decompositionalist approach. What’s significant in Vācaspati’s rejection of decompositionalism is his characterisation of aboutness. On his view, it is a natural, self-​ linking and irreflexive. It is natural in the sense that an awareness couldn’t occur without the relevant intentional object. It is self-​linking because we can’t explain why a certain awareness-​event about a certain object by appealing to any condition beyond the very nature of the awareness-​event. It is irreflexive in the sense that an awareness-​event cannot be about itself. The lesson that Vācaspati intends to impart is just this: our conscious experiences and thoughts can carry information about mind-​independent particulars only if the aboutness relation between these mental states and those particulars is prime, that is, non-​decomposable into internal and external conditions.

Notes 1 I am grateful to Davey Tomlinson, Chris Meyns and Cat Prueitt for helpful comments on earlier drafts of this paper. 2 Here, an internal condition is a condition depends only on the intrinsic properties of the agent, while an external condition is a condition that isn’t internal. 3 This line of interpretation has been accepted by many: for example, see Dunne (2004), Dhammajoti (2007a), and Moriyama (2008). Kellner (2014) distinguishes two different uses of the notion of ākāra in Abhidharma texts that seem to conflict with each other. On one view, the ākāra of a perceptual experience is just an intrinsic property of the perceptual experience; the perceptual experience just arises endowed with this feature. On the other hand, the ākāra is a mode of operation, i.e., the manner in which a certain sense-​modality apprehends the mind-​independent particular. 4 This idea is present in a number of Sautrāntika texts. For a representative instance, see Chapter 9 of Vasubandhu’s Commentary on the Treasury of Metaphysics (Abhidharma-​kośa-​bhāṣya) (Pradhan 1967, pp. 473–​4). 5 This discussion overlaps quite a bit with certain passages in Vācaspati’s The Grain of Reasoning (Nyāyakaṇikā), his commentary on Maṇḍana Miśra’s (c. 8th century CE) Discrimination of Injunctions (Vidhiviveka). See Goswami (1978, pp. 181–​92). 6 This theory is explained in Dhammajoti (2007a; 2007b, ch. 9). 7 In light of the Buddhist commitment to the idea that the basic constituents of reality are ontologically simple and impartite, Vasubandhu shows that the Doctrine of Momentariness just follows from two premises (Pradhan 1975, p. 193). The first premise is that all causally conditioned things, for example, material objects, eventually perish. The second premise is that we can only explain how an object could be destroyed after persisting for a while by appealing to some changes in it, e.g., a change in the configuration of its parts or its properties. The first premise is common ground amongst all Buddhists, and the second seems plausible. But if the basic constituents of reality are ontologically simple and impartite, they aren’t bearers of properties and they don’t have parts. So, their destruction couldn’t be explained in terms of changes in their properties or the configuration of their parts. Conclusion: no causally conditioned thing can perish after persisting for a while; they must perish immediately after being produced. 8 See Ayer (1956, 1963, 1967), Jackson (1977), and Robinson (1994). For discussion of the connection between the sense-​data and objective phenomenal forms, see Moriyama (2008).

102

103

Vācaspati on aboutness and decomposition 9 Vācaspati’s discussion here is possibly inspired by Dharmakīrti. See Verses k. 34–​37 and the commentary on them in Chapter 1 of Dharmakīrti’s Ascertainment of Epistemology in Steinkellner (2007, pp. 30–​32). 10 Ibid., p. 625, ll. 14–​17: “First, one thing isn’t the intentional object (viṣaya) of another merely in virtue of its existence. For, then, since existence is also present in other intentional objects, every entity would be the intentional object of every awareness-​event. So, there would be the undesirable consequence that every awareness-​event apprehends everything. And mere existence isn’t the factor that determines whether something is an intentional object. For even non-​existent things can be intentional objects.” 11 Ibid., p. 625, ll. 17–​20: “Therefore, an entity doesn’t have the status of being an intentional object in virtue of being a cause. Moreover, since our visual sense, etc. are causes of awareness-​events, there would be the undesirable consequence that they are intentional objects of visual awareness. Also, an awareness-​event would never manifest an object that currently exists. For, due to its momentariness, the momentary time-​slice of an object that produces the awareness-​event passes away at the time when the awareness-​event that is to be produced by it arises.” 12 Ibid., p. 625, ll. 20–​22: “By contrast, if a thing were to have the status of being an intentional object in virtue of being dependent on the same set of causal conditions as the awareness-​event, then the property of manifesting a currently existing object would make sense. But it doesn’t make sense. For it suffers from a problem of underextension with respect to the apprehension of past and future objects that don't currently exist, and from a problem of overextension insofar as it leads to the undesirable consequence that the momentary time-​slices of the senses, etc., which exist at the same time as the awareness, are intentional objects of the awareness since they depend on the same set of causal conditions as the awareness.” 13 Ibid., p. 625, l. 22–​p. 626, l. 1: “Suppose you say, ‘The intentional object is the locus in which the awareness produces a result.’ What, then, is produced by the awareness?” 14 For the discussion of this Bhāṭṭa view, see Umbeka’s and Pārthasārathi’s commentary on Verse 182 in “The Debate About Emptiness” in Kumārila’s Detailed Commentary in Verse at Sastri (1940, p. 183) and Śāstrī (1978, pp. 227–​8) respectively. 15 Ibid., p. 626, l. 4–​7: “That is false, since it suffers from a problem of underextension. For there are thousands of awareness-​events, produced by inference and testimony and belonging to omniscient beings, which have past and future entities as their intentional objects. And these awareness-​events aren’t able to produce any result in their intentional object. For their intentional object doesn't exist at that time. It is not possible that there is a property-​bearer [i.e., the object] that hasn’t been produced, but its property has the nature of being produced while remaining undestroyed.” 16 In The Grain of Reasoning (Nyāyakaṇikā), Vācaspati offers a different solution to the problem: he claims that manifestness isn’t a property of the intentional object, but rather a specific relation between the subject and the intentional object. This, on the face of it, avoids the problem that manifestness cannot be produced in past objects. See Goswami (1978, p. 190). On reflection, it’s far from obvious how successful this solution is: at least on an ontology in which relations are just dyadic properties, the same problem would arise again. 17 Ibid., p. 626, ll. 7–​11: “Suppose you say, ‘The awareness-​event of a certain kind is produced from its causes having been endowed with a capacity to manifest itself and other things, such that only a certain entity is its intentional object and not everything.’ Alas, good sir! Since a capacity is indexed to the target of that capacity (śakya), you must say what the target of this is. Suppose you say, ‘It’s the object.’ No. First, the object isn’t produced by this capacity, since the object is indeed what produces the awareness-​event. And it’s not reasonable to say that the awareness-​event produces a result that has the object as its locus. And when the target of a capacity is absent, then the capacity isn’t possible.” For a version of the Prābhākara view, see Sastri (1961, pp. 179ff). 18 See Verses 416–​446 in the Chapter on Perception (Pratyakṣa Pariccheda) of Detailed Commentary on Epistemology (Pramāṇavārttika) (Pandeya 1989, pp. 146–​51). For discussion, see Kellner (2017). 19 Thakur 1996, p. 626, ll. 14–​17: “Here, this is to be said. How does an awareness-​event come to have an intentional object—​by complete resemblance or partial resemblance? An awareness-​event, which is conscious in character, cannot resemble a non-​conscious object completely. If there were such resemblance, the awareness-​episode itself would be non-​conscious, so it would no longer have the property of being awareness. If the resemblance is partial, where is that absent? So, every awareness-​event would apprehend everything.”

103

104

Nilanjan Das 20 Ibid., p. 626, l. 17–​20: “Moreover, an awareness-​event, which is produced from an awareness of a blue object that serves as its immediately preceding homologous condition, more closely resembles the awareness of the blue object than the blue object. So, the immediately preceding homologous condition itself would be its intentional object in virtue of causation and resemblance. Therefore, the defining characteristic of the intentional object is also not causation plus resemblance.” 21 For discussion of this concept in Dharmakīrti and his successors, see Katsura (1986, 1991), Dunne (2004), and McCrea and Patil (2006). 22 Ibid., p. 627, ll. 2–​4: “Suppose you say that determination consists in the projection of the awareness-​ event's own phenomenal form on to an external object. Does that happen when the external object is apprehended, or in the absence of any such apprehension? It cannot be while the external object is apprehended. As we have already said, in an awareness-​event, two blue things are not manifested.” 23 Ibid., p. 627, ll. 4–​7: “It is not possible to project a projectible thing (āropya) onto the object of projection (āropa-​viṣaya) without the apprehending the object. Without undergoing a first-​ hand awareness of a white and shiny object before oneself at all, one cannot project silver onto it. Moreover, if no external object were apprehended, there wouldn’t be any regularity with regard to practical undertakings (pravṛtti). For, since other external objects are also not apprehended, there would be the undesirable consequence that there are practical undertakings directed at those as well. And, on the view that awareness-​events have a phenomenal form, there is no means of knowing that there are external objects, since they are unapprehended.” 24 A natural worry at this stage might be that we can sometimes project the objective phenomenal forms of awareness outward on to an external object even though we may never have apprehended it before, e.g., in dreams or hallucinations. The standard Nyāya response to this, starting with Uddyotakara’s sub-​commentary on Nyāyasūtra 4.2.34, is that, even in such cases, we project previously experienced properties on to previously experienced objects (Thakur 1997, pp. 489ff). For example, when I dream that I am in Delhi when I am not, I am ascribing to myself (an existent object) the property of being in Delhi (another property that I’ve encountered before). 25 Thakur 1996, p. 627, ll. 7–​8: “Well, the fact that the phenomenal forms of blue, etc. only occasionally arise counts as the means of knowing [the existence of an external world]. Now, if something happens only occasionally even in the presence of a certain condition, then that occurrence depends on something other than that condition.” 26 Ibid., p. 628, ll. 1–​2: “[The idealist:] This is unreasonable. Since the fact that awareness-​events of blue, etc. arise only sometimes can be explained even if they are caused solely by the stream [of mental events] in which they occur, the reason property put forward here is suspected to be compatible with the absence of the target property. As such, it is non-​exclusive (anaikāntika).” 27 Ibid., p. 628, ll. 3–​5: “[The realist:] Surely, we have said that a difference amongst moments [of material objects] cannot account for the difference in effects. [The idealist:] Who other than honourable sir can say that the constructed sequences [of momentary time-​slices of material objects], which is devoid of all causal capacities, can be a reason for the difference in effects, but the ultimately real difference amongst moments [of storehouse awareness] can’t?” 28 Ibid., p. 628, ll. 5–​8: “[The realist:] If the difference amongst moments of storehouse awareness could account for the difference in effects, then different moments of storehouse awareness couldn’t have the same power. If distinct moments don’t have the same causal capacity, once a certain moment of storehouse awareness produces a certain awareness of a blue object, a different moment [of storehouse awareness] couldn’t again produce that awareness of the blue object.” 29 Ibid., p. 628, pp. 8–​11: “[The idealist:] Well, can the defender of external objects account for how different streams of blue lotus moments produce streams of awareness-​events that have the same phenomenal form, i.e., of a blue lotus? If he says that the awareness of a blue lotus that is produced by a distinct stream of blue lotus moments is also distinct, then we can also say that the awareness of blue object that is produced by a distinct moment [of storehouse awareness] is also distinct. Therefore, from a particular set of causes, a specific particular (svalakṣaṇa) is produced, and that particular produces a certain effect and not any other effect.” 30 This is a variant of verse 211 in The chapter on perception (pratyakṣapariccheda) of Dharmakīrti’s Detailed Commentary on Epistemology (Pramāṇavārttika) (Pandeya 1989, p. 107). Unlike Thakur, I am reading the Sanskrit as, “tasmānnārthe na vijñāne sthūlābhāsas tadātmanaḥ/​ekatra pratiṣiddhatvād vahuṣv api na sambhava//​.” In the relevant context, Dharmakīrti is arguing that spatial extension that appears in perceptual awareness can belong neither to a material object nor to the awareness itself. It can’t

104

105

Vācaspati on aboutness and decomposition belong to the material object, because either it has to be one (i.e., an atom), or many (i.e., a plurality of atoms). An atom can’t have spatial extension; if one atom can’t have spatial extension, it’s hard to explain how a plurality of them could have it (given that it’s hard to explain a plurality of them compose a spatially extended object). Similarly, given that awareness-​events are themselves simple and impartite, they can’t be spatially extended either. 31 In his commentary The Grain of Reasoning (Nyāya-​kaṇikā), Vācaspati states the same argument; see Goswami (1978, pp. 187–​8). 32 For more detailed explanation of these notions, see The Grain of Reasoning (Nyāya-​kaṇikā), especially Goswami (1978, p. 187 ll. 10–​15). 33 For a discussion of this notion of a natural relation, see Oberhammer (1964, 1965), Vattanky (1984, pp. 76–​9), and Patil (2009, pp. 107ff). 34 For discussion, see Patil (2009, p. 107 and pp. 122ff). 35 Thakur 1996, p. 630, ll. 4–​10: “Well, an apprehension isn’t apprehended. For it doesn’t have the status of being a patient. A patient is that which comes to possess an effect produced by the act. An apprehension doesn’t produce any distinct effect in itself. In fact, an apprehension, once its inherent nature has emerged, is said to be apprehended when it is manifested independently of anything else. If it didn’t manifest itself, then it would also not manifest any objects. For the objects depend on its manifestation for their own manifestation. Therefore, anything, which is manifested on the basis of the manifestation of something else, is said to be manifested when that other thing is manifested. Just the manifestation of the stick-​bearer [as a stick-​bearer] depends on the manifestation of the stick, so also does the manifestation of intentional objects depend on the manifestation of the awareness itself …”

Bibliography Aiyaswami Sastri, N. (1990). Ālambanaparīkṣā and Vr̥tti by Diṅnāga. Madras: Adyar Library. Ayer, A. J. (1956). The Problem of Knowledge. London: Macmillan. Ayer, A. J. (1963). The Foundations of Empirical Knowledge. London: Macmillan. Ayer, A. J. (1967). “Has Austin Refuted the Sense-​Datum Theory?” Synthese, 17: 117–​140. Coseru, C. (2015). Taking the Intentionality of Perception Seriously: Why Phenomenology is Inescapable. Philosophy East and West 65 (1): 227–​248. Dhammajoti, B. (2007a). Ākāra and Direct Perception (Pratyakṣa). Pacific World 9: 245–​272. Dhammajoti, B. (2007b). Abhidharma Doctrines and Controversy on Perception. Hong Kong: Centre of Buddhist Studies, The University of Hong Kong. Dunne, J. D. (2004). Foundations of Dharmakīrti's Philosophy. Boston: Wisdom Publications. Ganeri, J. (1999). Self-​ intimation, Memory and Personal Identity. Journal of Indian Philosophy 27 (5): 469–​483. Ganeri, J. (2012). The Self: Naturalism, Consciousness, and the First-​Person Stance. Oxford: Oxford University Press. Ganeri, J. (2017). Attention, Not Self. Oxford: Oxford University Press. Goswami, M. L. (ed.) (1978). Vidhiviveka of Śrī Maṇḍana Miśra with the commentary Nyāyakaṇikā of Vācaspati Miśra. Varanasi: Tara Publications. Jackson, F. (1977). Perception: A Representative Theory, Cambridge: Cambridge University Press. Katsura, S. (1986). Jñānasrīmitra on Apoha. In B. K. Matilal and R. D. Evans (eds.), Buddhist Logic and Epistemology: Studies in the Buddhist Analysis of Inference and Language. Dordrecht: D. Reidel. Katsura, S. (1991). Dignāga and Dharmakīrti on Apoha. In E. Steinkellner (ed.), Studies in the Buddhist Epistemological Tradition. Vienna: Ôsterreeichischen Akademie der Wissenschaften. Kellner, B. (2014). Changing Frames in Buddhist Thought: The Concept of Ākāra in Abhidharma and in Buddhist Epistemological Analysis. Journal of Indian Philosophy 42 (2–​3): 275–​295. Kellner, B. (2017). Proofs of Idealism in Buddhist Epistemology: Dharmakīrti’s Refutation of External Objects. In J. Tuske (ed.), The Bloomsbury Research Handbook of Indian Epistemology and Metaphysics. London: Bloomsbury. Krishnamacharya, E. (1942). Tarkabhāṣā of Mokṣākara Gupta. Baroda: Oriental Institute. Levi, S. (ed.) (1925). Vijñaptimātratāsiddhi: deux traités de Vasubandhu: Viṃśatika et Triṃsikā. Paris: Librairie Ancienne Honore´ Champion. McCrea, L. J. and Patil, P. G. (2006). Traditionalism and Innovation: Philosophy, Exegesis, and Intellectual History in Jñānaśrīmitra’s Apohaprakaraṇa. Journal of Indian Philosophy, 34 (4): 303–​366.

105

106

Nilanjan Das Moriyama, S. (2008). Sense-​data and Ākāra. In M. K. Chakraborty, B. Lowe, M. Nath Mitra, S. Sarukkai (eds.), Logic, Navya-​Nyāya & Applications. Homage to Bimal Krishna Matilal. London: College Publications. Oberhammer, G. (1964). Der Svabhavika-​sambandha: Ein geschichtlicher Beitrag zur Nyaya-​Logik. Wiener Zeitschrift für die Kunde Südasiens, 8:131–​181. Oberhammer, G. (1965). Zum Problem des Gottesbeweises in der Indischen Philosophie. Numen 12 (1): 1–​34. Pandeya, R. (ed.) (1989). Ācāryadharmakīrteḥ Pramāṇavārttikam: Granthakartr̥viracitayā Svopajñavr̥ttyāĀcāryam anorathanandikr̥tayā Pramāṇavārttikavr̥ttyā ca Samupetam. Dillī: Motīlāla Banārasīdāsa. Patil, P. G. (2009). Against a Hindu God: Buddhist Philosophy of Religion in India. New York: Columbia University Press. Pradhan, P. (ed.) (1967). Abhidharmakośabhāṣya of Vasubandhu. Patna: K. P. Jayaswal Research Institute. (Tibetan Sanskrit Works Series VIII.) Robinson, H. (1994). Perception. London: Routledge. Sastri, S. K. Ramanatha (ed.) (1940). Ślokavārtikavyākhyā (Tātparyaṭīkā) of Bhaṭṭombeka. Madras: University of Madras. Sastri, S. (ed.) (1961). Prakaraṇa-​pañcikā of Śālikanātha Miśra with Nyāyasiddhi. Varanasi: Banaras Hindu University. Sastri, S. D. (ed.) (1978). Ślokavārttikam Kumārilabhaṭṭapādaviracitam. Varanasi: Tara Publications. Sastri, S. D. (ed.) (1994). Nyāyabindu of Ācārya Dharmakīrti with the Commentaries of Ārya Vinītadeva and Dharmotta and the Dharmottara-​ṭīkā-​tippanī. Varanasi: Bauddha Bharati. Steinkellner, E. (ed.) (2007). Dharmakīrti's Pramāṇaviniścaya: Chapters 1 and 2. Vienna: Austrian Academy of Sciences. Thakur, A. (ed.) (1996). Nyāyavārttikatātparyaṭīkā of Vācaspati Miśra. New Delhi: Indian Council of Philosophical Research. Thakur, A. (ed.) (1997). Nyāyabhāṣyavārttika of Bharadvāja Uddyotakara. New Delhi: Indian Council of Philosophical Research. Vattanky, J. (1984). Gaṅgeśa’s Philosophy of God: Analysis, Text, Translation, and Interpretation of Iśvaravāda Section of Gaṅgeśa’s Tattvacintāmaṇi with a Study of the Development of Nyāya Theism. Madras: Adyar Library and Research Centre.

106

107

6 SEEING AND RECOGNITION IN THE ARABIAN NIGHTS AND ISLAMIC ALEXANDER LEGENDS Anna Ayse Akasoy

Ofenröhre wird Katze. Beim Worte Ingwer ist an Stelle des Schreibtisches plötzlich eine Fruchtbude da, in der ich sofort darauf den Schreibtisch wiedererkenne. Ich erinnerte an 1001 Nacht.1 Walter Benjamin, Über Haschisch, December 18, 1927, between 3 and 4 in the morning Hārūn al-​Rashīd is bored.2 Kept awake by insomnia, the Abbasid caliph is looking for relief. Neither the beautiful gardens nor spying on his concubines appeals to him. Masrūr, his executioner, becomes desperate. ‘Summon the men of learning,’ he proposes to the ruler, ‘the philosophers and poets, so that they may conduct investigations, recite poetry and produce stories and histories.’ The suggestion does not appeal either, nor does Masrūr’s idea to invite Hārūn’s ‘boon companions and men of wit’ who might entertain him ‘with remarkable and witty tales’. Masrūr gives up. ‘Then cut off my head’, he says to the caliph. At this point, Shahrazad, most accomplished of all storytellers, interrupts her account. Her audience, murderous king Shahriyar, has to wait until the following night to hear how Hārūn responds. The caliph laughs. Kings often laugh in the Arabian Nights. It is an expression of amazement at the many strange events that transpire in the stories of the one thousand and one nights. Finally, ʿAlī ibn Manṣūr is announced, ‘the witty Damascene’. ʿAlī and the caliph engage in a revealing conversation. ‘Shall I tell you of something that I have seen myself or something that I have heard, Commander of the Faithful?’, asks ʿAlī. ‘If you have seen something remarkable, then tell me about it, for what has been said is not like what has been seen (Arabic: fa-​innahu laysa al-​khabar ka’l-​ʿiyān),’ the caliph replies. ‘Listen to me, then, and pay attention’, says his interlocutor. ‘I am listening to you with my ears, looking at you with my eyes and paying attention to you with my heart,’ the caliph assures him. The story ʿAlī has to offer is a humorous tale of two lovers. Jubayr and Budūr are separated by a disagreement or misunderstanding, ironically because Jubayr saw a slave girl planting an innocent kiss on his beloved Budūr’s cheek, having been moved by her beauty.3 ‘If my love is shared with someone else.’ Jubayr declares stubbornly, ‘I shall leave her and live alone.’ In the course of the story, told by Shahrazad over eight nights, ʿAlī reports how he orchestrated the reconciliation of the lovers. 107

108

Anna Ayse Akasoy

Here, as in so many other stories, the Arabian Nights show themselves to be a complex reflection of Islamic intellectual and cultural history –​rather than the collection of fairy tales for children they often became in global reception. Indeed, when the reader first encounters Shahrazad, we learn that she ‘had read the books of literature, philosophy, and medicine. She knew poetry by heart, had studied historical reports, and was acquainted with the sayings of men and the maxims of sages and kings.’4 Like a number of other characters in the Arabian Nights, Shahrazad is thus presented as a source of high-​prestige and canonized information, her stories implicitly an applied reflection of her learning. In these ‘embedded stories’, characters sometimes express views regarding different sources of information and different modes of conveying and receiving information. As Hārūn illustrates, they may value certain information or, indeed, they may find it boring. The stories offer us snapshots into categories the storytellers of the Arabian Nights and their audiences may have used to evaluate information, categories not limited to more or less reliable in its truth value, but also more or less amazing or entertaining. Hārūn exemplifies an ideal audience who receives information in a holistic manner, with several senses as well as an appropriate internal disposition, whereas Jubayr in the embedded tale illustrates what happens if information is received in a less than ideal manner. The present contribution is part of an effort to read literature of the medieval Islamic world in the context of intellectual and philosophical history. This approach is both historical and literary in nature –​historical because the stories reflect historical intellectual milieus, trends, views and concerns, and literary because with their plots, their characters and compositional aesthetic, the stories themselves embody epistemological, psychological and ethical principles. Similar exercises can be conducted for other examples of Arabic narrative literature. What makes the Arabian Nights an especially compelling case, however, is its structure of multiple layers of storytelling. Shahrazad is the storyteller in the frame tale, but within the embedded stories we find embedded storytellers such as ʿAlī who have heard or witnessed something remarkable. Many participants in the Arabian Nights phenomenon have thus a double role to play. They are storytellers and audiences, sources and recipients of information. And there are of course those storytellers too who presented the very story of Shahrazad to audiences. With this structure, the Arabian Nights lend themselves to a reader-​centric approach in which information is not simply absorbed from a source, but the product of the storyteller’s choice for a specific recipient under specific circumstances. In this manner, the stories offer a rich combination of principles articulated and principles narrativized, of information in the abstract and of information in context.

Shahrazad, Benjaminian storyteller of experience Hārūn’s preference for something that has been seen over something that has been heard is a significant comment on sources of information and modes of obtaining and transmitting information. Here, the difference stands for the preference of firsthand knowledge over secondhand knowledge, or hearsay, a common distinction in law, Islamic or otherwise. Elsewhere in the Arabian Nights too, the distinction is made, although not always with the same preference. The following story in the collection, for example, has a similar question posed to al-​Maʾmūn, but this ruler just wishes to hear whatever is stranger, whether heard or seen.5 His decisive criteria are the contents and effect of the information, rather than its source or mode of transmission. In another Arabian Nights story, The City of Brass, the Umayyad caliph ʿAbd al-​Malik hears about brass bottles in which Solomon imprisoned jinns and then he desires to see such a bottle himself, perhaps not so much to verify the account, but to enjoy amazement in a direct manner.6 108

109

Seeing and recognition

The caliph being the caliph, however, rather than traveling himself to the distant lands where these bottles can be found, sends out a delegation. It is they who see. Here, information can be purchased when personal pursuit is not feasible or convenient. Socio-​economic and political stratifications are an important dimension of many Arabian Nights stories, and they sometimes determine the access people have to information. What the group sees, is one and the same message inscribed in the remains of human civilizations they encounter throughout their journey. An ubi sunt regularly leads to a memento mori. Andras Hamori called the story the ‘gloomiest of travelogues’.7 As Bruce Fudge has shown, the landscape of The City of Brass is imbued with Qur’anic references.8 Scripture becomes literally inscribed in a Book of Nature. Seeing, however, is also treacherous. When the expedition reaches the City of Brass, they find it surrounded by a high wall. Those who climb it, are tempted to jump into a wonderful pool on the other side where beautiful women await them. The men jump to their death. The pool is an illusion, the women mirages animated by magic. Long before E.T.A. Hoffmann’s Sandmann of Freudian fame, the storytellers of the Arabian Nights evoked the dangers of the uncanny that rest in the beautiful humanoid. Only those in their right mind and pious spirits can truly see. The episode echoes Hārūn’s reference to the significance of the inner disposition. Once the delegation has made it into the city, they discover a dead princess protected by automated guardians. An inscription she left behind on a golden tablet describes the city’s fate. Once flourishing and ruled by just kings, it declined after years of droughts and famine. Their wealth was of no use and so they surrendered to their fate and to God. Literally at the heart of the City of Brass is thus further confirmation of the transience of life and the futility of material obsessions. At the end of the journey, the delegation finally come across a settlement at the limits of the known world where further wonders await and where they obtain the brass bottles. Then, they return to the caliph who, after all, gets to see the objects of his curiosity, albeit not in their amazing location and without having undergone the experience and various moral tests of the travelers. Information here stems from a physical and mobile source, but also invites questions about the relationship between that source and its original environment and about the significance of experience. We can zoom in and out of stories, but we can also consider information in isolation –​a moral lesson, a fact about the world –​and in context –​what does it mean for somebody and why does somebody else tell us a story about it? Listeners to the story may compare their own experience in hearing the account of the journey with that of the caliph who also primarily hears information, but is physically exposed both to those with firsthand knowledge and to an object which testifies to their experience and serves as a material vessel of information. The City of Brass is instructional, even homiletic. Time and time again, the story reminds the listener that there are realities more fundamental and significant to humans than others. Insights into these realities are not merely theoretical. They are meant to guide our actions. As much as the story illustrates the pursuit of knowledge by way of describing the delegation’s journey, both spatial and spiritual, it contains information as well for the static listeners to recognize and internalize. The information is presented factually as a travel account, but also aesthetically and imaginatively as a story. It records, evokes and creates experience. Insofar as The City of Brass is allegorical it invites diverse audiences in different places in life to contemplate our attitude to what we consider important in life. Put differently, the information it offers is wisdom which stretches across human particularities. As much as Arabian Nights scholarship has been critical of the text’s classification as a fairy tale, the story in fact exemplifies the virtues of the fairy tale Walter Benjamin identified in The Storyteller.9 In his 1936 essay, Benjamin outlined an evolution of literary forms in conjunction with changing modes of production and social life. He distinguished three such forms: story, novel and what he called information. Products of 109

110

Anna Ayse Akasoy

age-​old human cultures with their boredom of monotonous manual labor and the excitement of long-​distance trade, stories capture experience in the collective format of oral literature. The novel, by contrast, is a much more isolated and individual affair. Product of a single mind, read by individual and isolated readers, the novel tells a single and finite story. To Benjamin’s mind, the novel was incommensurable, unlike the story which can be remembered, adapted and repeated, generation after generation. Furthermore, because, as Benjamin argued, there was always room for curiosity how a story continued, stories are open-​ended. That curiosity is not always or necessarily plot-​driven, but might simply reflect a demand for expanding on a particular message and applying it in real life.10 The novel, on the other hand, offered all of a human life and asked the reader to assess this life from its end, if only the end of the novel. Benjamin considered the story superior on account of its greater ‘Schwingungsbreite’, that is, its breadth, flexibility or variability (54; 109). In his assessment, none other than Shahrazad became emblematic of this feature. It is memory that weaves the net all stories ultimately form together. Each is tied to the next, as the greatest storytellers, especially those from the Orient, were always pleased to demonstrate. In each of them lives a Scheherazade, for whom every episode in a story evokes a new story. (62; 117) How open-​ended The City of Brass is, is not just obvious from the fact that Shahrazad moves on to the next story, but also within the story itself. The caliph’s delegation moves from one stop to the next, discovering ever stranger realities. Inspirational in nature, the story’s potential is only ever and continuously actualized in what happens next. If we understand the story as a source of information, this information is shaped by the story’s openness and the audience’s curiosity. It is not information in the Benjaminian sense. As much as the novel seems to compare poorly to the story, Benjamin’s criticism in The Storyteller is directed towards the third literary format. Associated with journalism, ‘information’ provides immediate knowledge. Detailed and unambiguous, information lacks the malleability of stories. Capturing a specific moment in time, it is static and limited. The story, by contrast, could not be any more fundamental and encompassing of human experience. In part, it is mortality or death that lends the story its powerful significance. ‘Death is the sanction of everything the storyteller can relate. It is death that has lent him his authority’ (59; 114). Benjamin does not spell this out in his essay, but the threat of death, her own as well as that of the other women in the kingdom, is of course also what animates Shahrazad’s storytelling. Likewise, we can see this principle at work in embedded stories such as The City of Brass. Against the backdrop of Benjamin’s observations about the disappearance of death from the public sphere in modernity, we can understand Hamori’s description of the story as ‘gloomy’ as precisely such a modern response. To premodern audiences, the story was presumably on some level encouraging. In terms more widely used in the Islamic tradition, it is a promise as much as a warning. Characters in the story tend to weep whenever they come across messages of transience, a response which may have reflected as well as modeled reactions of the audience to the story. Medieval Muslim sermons too were meant to make listeners weep, but overall inspire a sense of urgency to heed the warning and obtain the promise.11 The plot unfolds as narrative hermeneutics. The specific experiences of characters, their insights and actions, resonate with a universal and profound human experience, here the awareness of mortality, although the characters in the story experience this awareness more acutely. As told by the returning travelers, the story of their discoveries has its desired effect on the caliph who has the treasure they brought distributed 110

111

Seeing and recognition

among Muslims. Musa, leader of the expedition, withdraws to Jerusalem to worship God. His companion Talib, on the other hand, had been executed by the automatons who guard the dead princess. He saw her jewels, but without recognizing them for the material trap that they are. The story thus illustrates a well-​known Qur’anic principle about signs for those who see (15:75). And yet, more is required to read these signs than the ability to see in the sense of pious cognition. One has to find them in a very literal, physical and visual sense. In The City of Brass, on one level revelation unfolds in a narrative mode. On another, it assumes a geographical mode. The quest for the true knowledge embedded in the curious strangeness of Solomon’s bottles requires traveling to distant regions, the end of the known world almost. Physical and spatial ubiquity is part of what makes the story’s moral message so powerful. At the same time, these strange phenomena were localized in specified areas. The Arabian Nights’ City of Brass is but one variant of a trope where an inaccessible and dangerous, but attractive city is set in a remote region. Islamic geographical literature of the medieval period describes a castle with similar features. Located in Tibet, it miraculously attracts foreigners. They attempt to climb its walls, but then die from laughter.12 As much as The City of Brass can thus be read in an allegorical way, assumptions seem to have been made about a physical, empirically verifiable reality as well. In fact, this story illustrates well how the Arabian Nights functioned as a vehicle for transmitting information about the world, not just within the cosmos of the story, but between real-​life storytellers and their audiences. This information was very much derived from specific experiences. Even though authors and compilers of medieval geographical literature relied on hearsay, the very genre is a testimony to the authority of visual, personal observation. Secondhand knowledge requires firsthand knowledge. The Abbasid caliph al-​Wāthiq (reg. 842‒847) provides us with another example where imaginary geography is physically pinned down. In 842, the ruler dreamed of the wall Alexander the Great had erected to keep the world safe from the wild tribes Gog and Magog. In the caliph’s dream, the wall was breached –​a sign of the last days. Al-​Wāthiq sent out a delegation headed by Sallām the Interpreter to inspect the state of the barrier. Sallām, who allegedly mastered 30 languages, traveled to what at the time seems to have been thought to be the or at least a possible location of the wall, which was in western China. Upon his return to Iraq Sallām dictated an account of his journey to a geographer. The account includes a detailed description of the appearance of the structure and the reassurance that the only damage was a small crack.13 The account thus offers a curious juxtaposition between dreams as sources of information, validated here by the authority of the dreamer, and firsthand physical seeing as a source of information in its own right which is required to evaluate the dream’s information. Evidence where mythological or legendary places are unambiguously located in such a way that they could be visited and seen coexists in medieval Arabic literature with the fact that such stories were told about different places in the world. This geographical adaptability suggests that we are dealing with an expression of experience, to put it in Benjamin’s terms, and a fundamental and replicable aspect of our relationship to the world. It is also a reflection of what Thomas Bauer describes as ‘culture of ambiguity’.14 Seeing is thus an ambiguous phenomenon, simultaneously particular and universal, figurative and literal. Accordingly, as a source of information, it offers both concrete and adaptable information.

The epistemology of recognition narrativized In the frame story of the Arabian Nights, seeing is often very literal and indeed critical to matters of life and death. The characters do not recognize any ambiguity in this source of information, nor does the story imply that audiences should recognize any ambiguity. At the very beginning 111

112

Anna Ayse Akasoy

of the frame story, just before setting out to visit his royal brother, king Shahzaman pops into his residence one more time, only to find his wife in the arms of the kitchen servant. He arrives at the court of his brother Shahriyar depressed, but his spirits are lifted when he watches Shahriyar’s wife betray her husband in an orgiastic open-​air encounter in the palace garden. Shahzaman is thus not alone in his predicament, an insight which cheers him up to such an extent that his brother wishes to understand the reasons for his recovery. He then sees the betrayal himself. Disappointed and disoriented, the two brothers leave. During their travels, they come across a jinn who holds a woman captive in a box. They hide in a tree. While the jinn is asleep, the woman discovers the two kings and forces them to sleep with her. She takes their rings, adding them to a chain with a large number of other rings, the precise figure varying from version to version. These trophies are the visible signs of previous abuse and betrayal. Strangely relieved that even the mighty jinn can be betrayed by a woman, the brothers return from their journey. Shahriyar now sets out on his murderous course, taking a new woman as wife every evening and executing her in the following morning. His own experience has taught him that women are treacherous. Among the numerous interpretations of this famous frame story, Daniel Beaumont has presented a psychoanalytical reading, applying Hegel’s master-​slave dialectics to the psychological state of Shahriyar and his brother.15 Beaumont emphasizes the moments of literal, visual seeing as foundational experiences of the self. It is in these moments that matters of life and death are decided as one character, the ‘slave’, decides that he’d rather submit than die, and the other, the ‘master’, that he’d rather die than submit. Seemingly despite his superiority, the ‘master’ needs the ‘slave’ for his sense of self. The meaning or significance of the information derived from sense perception is thus only complete in the response it elicits. Shahriyar’s seeing of his wife ultimately leads to the femicide, especially because she conducts her betrayal with a slave, this man a literal one. The ‘box woman’s’ discovery of the brothers forces them to decide whether they’d rather die than sacrifice their positions of power. Their submission and the conflicting insights the brothers obtain about who they are in relation to others enforce a crisis of the self and the profound disorientation that, in this reading, accounts for Shahriyar’s violence. From the course the story takes we can assume that the audience was not meant to agree with Shahriyar’s conclusion either. He inflicts pain and suffering on his innocent people and Shahrazad who ends this is a positive figure. As much as stories can illustrate how people can obtain information, they also tell us that this information is sometimes inaccurate or taken in an immoral way. This lesson in itself serves as information to audiences who observe the plot from the outside. Scenes of visual discovery are prominent in Arabic literature, as Philip Kennedy has argued in his recent study on recognition or anagnorisis.16 Analyzing recognition scenes from a variety of Arabic narratives, from the Qur’an to the picaresque Maqāmāt, Kennedy illustrates how such moments function as turning points in the plot to the extent where they become ‘signature features of fiction’ (2), but simultaneously serve as tools in the hermeneutics of narrative. Using Aristotle’s Poetics as his starting point, Kennedy points out that while anagnorisis may have initially been conceptualized as a recognition of the true identity of persons, it was already Aristotle who allowed for a broader category of objects of recognition. Recognition, however, gains in significance and potential meaning in a reader-​centric perspective. Readers see alongside characters, moving in a variety of ways from ignorance to knowledge, as recognition is often explained. Listeners of The City of Brass imaginatively accompany the traveling delegation. Unlike narrative characters, however, audiences are capable of multiple layers of recognition. They can revisit moments of recognition and sources of information, obtaining ever new insights into the knowledge acquired by characters, how and where they obtained that 112

113

Seeing and recognition

knowledge, but also into the knowledge they gain as readers by understanding these moments of recognition. Unlike narrative characters, they can thus experience multiple and expandable moments of recognition. It illustrates that what kind of information is conveyed from which sources, in which manner and to which effect depends very much on the level of storytelling within the Arabian Nights and whether we are finding ourselves within the world of the plot, the plot as an embedded story, as an embedded story told within an embedded story and so on. Accordingly, information gains increased levels of context and meaning. The narrative significance of visual recognition in Islamic cultural history extends well beyond Arabic literature, as Kennedy suggests in considering the Persian mystical text Conference of the Birds or a Turkish hagiographic collection in his exploration of Joseph narratives. It was in the Seljuk-​ruled Caucasus that Niẓāmī (1141‒1209), author of an extensive two-​part book on Alexander the Great, composed his influential story of Khosrow and Shirin in Persian. Many misunderstandings and external obstacles had to be removed for the princely lovers to be united. Khosrow is in surprising need of self-​disciplining given his deep love for Shirin which began when his grandfather appeared to him in a dream and promised him four things: the horse Shabdiz, the throne Taqdis, entertainment from the musician Barbad and ‘beyond all these, you shall have Shirin, your destined love, whose sweetness and beauty will sustain you all your days.’ The painter Shapur later tells Khosrow17 about Shirin, the beautiful niece of the queen of Armenia. Having heard about the prince’s dream, Shapur returns to Armenia and puts paintings of Khosrow up for Shirin to see. She promptly falls in love with the prince. Khosrow’s love in turn is sealed when he comes across the princess bathing. Just how critical especially this scene of recognition is for the extensive love story is obvious from the number of Ottoman and Safavid paintings which show Khosrow on horseback, approaching the pond in which beautiful Shirin is taking her bath. Khosrow typically places his index finger on his mouth, a gesture conventionally understood as an expression of amazement.18 In addition to recognizing Shirin the woman, Khosrow recognizes the realization of his grandfather’s promise. The idea of visual intimacy is more widely discussed in Middle Eastern literature. Much earlier and on the other end of the Islamic world, Ibn Ḥazm (d. 1064) devoted a lengthy passage in his Neckring of the Dove (Arabic: Ṭawq al-​ḥamāma) to the ability of lovers to communicate with their eyes. ‘You should know,’ the Andalusi polymath wrote, ‘that the eyes are sometimes messengers and we perceive with them what we desire. The four senses are open gateways to the heart and windows to the soul; but sight is the most reliable and insightful guide, and the one which acts with greatest clarity. It is the faithful leader of the soul, its guide and the well-​polished mirror in which it recognizes truths, discerns attributes and understands sensory perceptions. For it is said: “What is reported is not the same as what is seen.” (laysa al-​makhbar ka’l-​muʿāyan) Polemon, the author of The Physiognomy [or: the physiognomist], mentioned this and considered eyes foundational to judgment.’19 The parallel to the exchange between Hārūn and his companion ʿAlī al-​Manṣūr in the Arabian Nights is clear –​the Arabic phrases in which preference for something seen is expressed resemble each other closely, reflecting a more widely known principle.20 But Ibn Ḥazm embedded his observations in an entire scientific theory of vision. Later in this section, as well as elsewhere in his work, he elaborated his view that sight works by way of rays emanating from the eyes. The reference to Polemon too enhances the scientific character of his presentation. Alongside a pseudo-​Aristotelian book on the subject, Polemon’s treatise was the best-​known textual authority on physiognomy in medieval Arabic literature. The proposition underlying this tradition is that visually observable corporeal features of humans reflect their character. The principle of physiognomic analysis was sometimes framed more widely in hermeneutic terms as the ability to recognize the hidden (Arabic: bāṭin) in the apparent or obvious (Arabic: ẓāhir).21 113

114

Anna Ayse Akasoy

What these examples cumulatively reflect is the widespread appreciation of seeing as an immediate acquisition of accurate information, uncorrupted by human representation and communication, unmitigated by language and sound. It is a principle expressed in general and theoretical terms, but also narrativized in scenes of recognition. In narrative nutshells, these scenes enhance the nature of stories as sources of information, especially where dream vision is followed by literal vision. Experience corroborates information initially obtained in ways other than immediate vision. Information in the factual sense is complemented by information about its epistemological nature. It is worth reiterating, however, that the contrast between something seen and something heard reflects more a preference of first-​over secondhand knowledge than a hierarchy of the senses, even if Ibn Ḥazm exemplifies the latter. The counterexample in the Islamic tradition is the Qur’an which compels by virtue of its acoustic quality, although The City of Brass may be an example where the truth of the Qur’an is also being seen.22 The high status of hearing is obvious in Muslim education as well, where the Arabic word samāʿ denotes hearing as a direct acquisition of knowledge from a teacher which provides the student with the authority to pass on this knowledge.23 The appreciation for firsthand knowledge as represented by seeing and for traveling as the ideal mode of obtaining such knowledge was no mere literary or theoretical construct. As Houari Touati has illustrated in a study that combines social history and epistemology, this principle was a cornerstone of the structure of Islamic learning.24 A remarkable feature of premodern Islamic scholarly culture is just how common it was for learned individuals to travel. The pilgrimage to Mecca was and remains an obligation for any Muslim with the necessary means. But both students and established scholars traveled extensively beyond that. The phenomenon was known in Arabic as riḥla fī ṭalab al-​ʿilm, or journey in search of knowledge. During the first two centuries of Islamic history, oral transmission prevailed by reputation over the written word, but the skepticism against the latter waned. This shift had important repercussions for the authority of travelers, as James Montgomery has demonstrated for Ibn Faḍlān, emissary of the Abbasid caliph al-​Muqtadir (reg. 908‒932) to the Volga Bulghars.25 Their authority was distinctly that of eyewitnesses, but travelers had to contend with information circulating in written form. Even after the book established itself as a reliable source in the ninth century, students still traveled to hear their teachers directly. Instances of immediate visual recognition in which the deep and essential truth of a matter becomes obvious to a seer, however, were not universally and uncritically accepted in the medieval Islamic world. Al-​Ghazālī (1059‒1111), a scholar familiar with many branches of Islamic learning, offered a sort of thought experiment in what has variably been described as his autobiography or as a polemical treatise. In The Deliverance from Error, al-​Ghazālī described a spiritual and intellectual crisis in which he is no longer certain of the truth and how to recognize it. In a manner as reminiscent of Pyrrho as it is of Descartes, he considers the possibility that his sense perceptions are faulty.26 ‘The strongest of the senses is the sense of sight,’ he begins. Now this looks at a shadow and sees it standing still and motionless and judges that motion must be denied. Then, due to experience and observation an hour later it knows that the shadow is moving, and that it did not move in a sudden spurt, but so gradually and imperceptibly that it was never completely at rest. Sight also looks at a star and sees it as something small, the size of a dinar; then geometrical proofs demonstrate that it surpasses the earth in size. In the case of this and of similar instances of sense-​data the sense-​judge makes its judgements, but the reason-​judge refutes it and repeatedly gives the lie in an incontrovertible fashion.27 114

115

Seeing and recognition

Walter Benjamin’s observations while intoxicated illustrate al-​Ghazālī’s objections both poetically and epistemologically. As much as the truth of visual recognition was thus celebrated in Islamic literature, writers were also wary of the many ways our eyes can lead us astray for a whole variety of reasons. The theoretical observations of those who wrote in a philosophical or scientific mode can be supplemented with literary evidence where the modalities of providing and obtaining information are narrativized with their complex consequences. In stories, information is in part facts about the world, but also a function of how we interact with the world. What information is conveyed to audiences depends on the context we consider.

Alexander, the seeing traveler As indicated above, in The City of Brass, seeing is seeing in situ. It requires traveling. For Benjamin, traveling traders and seamen were protagonists in the enterprise of storytelling. He recognized their particular significance for the Arabian Nights, an observation recent scholarship vindicates. Notably Aboubakr Chraïbi described the Arabian Nights in parallel to mirrors for princes as a ‘manual of basic rules in manners and customs for young merchants’.28 In medieval Islamic literature, however, nobody traveled farther than Alexander the Great.29 Alexander, who became absorbed into the Islamic tradition through the Qur’an and adaptations of the Alexander Romance, a product of late antique Alexandria, was removed in multiple ways from medieval Muslim readers: historically, culturally and to some extent geographically as well. He functioned as a traveler, but as a traveling conqueror and city builder of the past he also shaped a landscape for later travelers to see. In addition to several stories which reflect the ‘sustained textual influence’ of Alexander narratives such as the travels of Bulūqiyā and indeed The City of Brass, the Arabian Nights include a variant of a widely-​told encounter between Alexander and Indian ascetics at the extreme East of his conquests.30 The anecdote typically throws the conqueror’s questionable violence and obsession with this-​worldly achievements into relief, but it also illustrates his wisdom and justice since he does not respond to the philosophical challenge the sages present in the punitive manner so common to him. The extent to which the account captures an actual historical reality is controversial. Like many other stories about Alexander, it is presumably to a large extent the product of later concerns, interests and imaginations. Stories about Alexander illustrate how narratives can serve as a vehicle for conveying a variety of information, which again confirms their adaptable nature as stories in the Benjaminian sense. As Yuriko Yamanaka has established, the version of Alexander’s encounter we can read in the Arabian Nights appears to have been adapted from a text by the above-​mentioned al-​Ghazālī.31 In this version, the connection to India has disappeared. Alexander merely comes across a community of very poor people and enquires about their habits in a conversation with their leader. The structural hallmark of the anecdote is thus always and characteristically discursive –​indeed, it is in the format of question and answer that some contemporary scholars have recognized Indian or Buddhist features.32 Remarkably, however, it is not the verbal arguments the king of these ascetics presents to Alexander that shake his confidence, but rather what the conqueror sees. First of all, it is already what he sees that inspires his curiosity: A story is told that Alexander the Great passed by a people who were so poor that they owned no worldly goods at all. They used to bury their dead in graves dug at the doors of their houses, which they would constantly visit to clean and to sweep away 115

116

Anna Ayse Akasoy

the dust, and where they would worship Almighty God. Their only food was grass, together with plants that they got from the earth. (325) Ironically, it is in fact what Alexander does not see that especially inspires his wonder –​he does not see any gold, silver or worldly goods in the locals’ possession. The episode is reminiscent of Alexander’s encounter with Diogenes. As recounted by Plutarch,33 one of several sources, the philosopher showed no interest in making an appearance in front of the king, which in turn inspired Alexander to visit Diogenes. In response to Alexander’s question what he could do for him, however, the philosopher famously requested that he step out of the sun. The encounter with the Indian Brahmins –​or the ascetics in the Arabian Nights –​is more complex. In the latter version, their king explains to Alexander that they surround themselves with graves because seeing them reminds them of the world to come. The exchange between the two men thus relies on paradoxical relationships between the visible and existence. In this ascetic semiotics, the absent riches and the signs of what has been here before, but no longer is, point to what is actually meaningful, although the world to come is not yet visible either. To absorb the reality of this, however, Alexander first needs to see something that is actually there, which is two skulls. One belongs to a just king who enjoys the pleasures of the hereafter, the other to a tyrant who is punished in hell. In order to make the lesson explicit, the king of the ascetics ‘laid his hand on Alexander’s head and said: “Which of these two, do you think, will be yours?” ’ Alexander despairs, presumably because he doubts the quality of his own leadership, but perhaps also because of this vision of death and the ascetic’s touch. This tactile experience enhances Alexander’s confrontation with the stark visual evidence which so much defines the caliphal delegation’s journey in The City of Brass. In both cases, the politically and militarily powerful obtain information derived from physical reality because they see it. Again paradoxically, while the physical world is thus granted a powerful sense of reality, the knowledge this material sphere grants is precisely how flimsy its reality is, if compared to the reality of the other, transcendent world. The version of the anecdote in the Alexander Romance ends with a similar lesson, but it is obtained differently. Here, the Indian sages ask Alexander for immortality, a request he acknowledges he cannot fulfil because he is only a mortal.34

To see is not to see Scholarship tends to consider the Alexander of classical literature separate from the Alexander of medieval literature. There are good reasons for this. Under the influence of the Romance, the biography of Alexander changed considerably. Supernatural elements became more common, but he was also claimed by a variety of peoples, notably Egyptians and Persians, who insisted that Alexander was really the son of one of their kings. Reading medieval alongside classical Alexanders, however, reveals interesting continuities. The friction between ambition, failure and achievement is a common context throughout this body of literature. Furthermore, a comparative analysis of material from across very different periods of history allows us to recognize more clearly the distinctive features of each. It is at the crossroads of literal and figurative seeing that Alexander recognizes the limitations and thus the essence of human nature. His self-​awareness exceeds that of Shahriyar who only sees, but does not not see. The transition from literal to figurative seeing involves the interesting paradox that precisely what cannot be seen in the former mode appears visible in the latter mode. The challenge posed to us is to transcend the information in front of us, to understand the implications of the information we have and to recognize what we are missing. And yet, 116

117

Seeing and recognition

we are not dealing with two entirely separate binaries here. If seeing in the figurative sense involves the acquisition of abstract knowledge, spiritual seeing involves some vision in the literal sense as well. This has a parallel in the paradox of dreams and other forms of divination where one obtains information about something that has not yet happened, or where the object of vision is not accessible to empirical, collective seeing, as in the case of Khosrow’s dream, or where it requires precious confirmation, as in the case of al-​Wāthiq’s dream of Alexander’s wall. The phenomenon of ambiguous seeing, however, exists in classical Alexander biographies as well. In Arrian and elsewhere, Alexander has his own diviner, Aristander, who interprets omens and dreams. For Plutarch, the fact that it is Alexander’s dreams that predict future events indicates his position near the divine. But even information derived from a predicative dream requires proper analysis, that is a combination of literal and figurative seeing. King offers the example of the dream of King Darius which should have been interpreted as a warning.35 In the evolution of Alexander biographies, Aristander becomes a middleman who is being removed. The Alexander of the Romance and medieval versions has himself hermeneutic expertise. Accordingly, Alexander as a source of information gains complexity and authority. The information he receives and offers is not that of the rash Alexander of the classical tradition, a man who killed a companion in drunken rage, but reflects superior insight, evaluation and thus validation. That Alexander gains such fundamental insights into his own and human nature in general is another theme which transcends literature produced in different cultural contexts. As far as we can tell from the classical biographies, traveling gave Alexander the opportunity to gather knowledge of a great variety of things, either because of the scholars who traveled with him, or because he saw them himself. The conqueror at the forefront of his wide-​traveling armies has fascinated readers since antiquity and provided them with information about the world. Large part of that fascination is of course also what Alexander does during his journeys. Emulating the heroic Hercules and seemingly not fearing death, for example, Alexander reportedly hurled himself over the walls of a fortress in the Punjab during his Mallian campaign. While he sustained a severe injury, his lung being pierced by an arrow, he survived again. Adventures such as this, and the sheer intransience of the landscapes he crossed, gave Alexander’s movements in quantity and quality a superhuman flavor. Although cutting through all those layers of legend is a virtual impossibility, it remains reasonable to speculate about the historical Alexander’s ambition. A successful founder of cities, he failed as a state builder. Rather than the model king he became in legendary reincarnation, he seems to have seen himself as another Achilles, taking revenge in Asia Minor against the enemies of Greece. Like Achilles, he was struggling with the ever so remote possibility of immortality.36 Rather than being the obsession of an exceptional individual, however, or a product of literary romance, Alexander’s concern with mortality appears to have been typically Macedonian –​Georgios Halkias lists among their distinctive trends the ‘construction of funeral monuments instead of temples, and the deification of their rulers’.37 Even before his legendary transformation, the Alexander of classical literature served as a source of information about larger themes by embodying and illustrating them, justice and kingship, virtues and vices, mental stability and power. In medieval Islamic adaptations of the Alexander legend, mortality assumes again a different significance. The rise of the subject predates these versions to some extent and can already be detected in the Alexander Romance.38 While classical biographers such as Plutarch or Curtius were read in medieval Western Europe in addition to this text, the Alexander Romance had a much more comprehensive influence in the Middle East. In a variant that gained particular prominence in medieval Persian Alexander legends, the conqueror enters the land of darkness in his search for the source of life, but only his companion drinks the water which makes 117

118

Anna Ayse Akasoy

him immortal. As Fudge suggests, the failure of the quest thus turns into a success insofar as Alexander gains knowledge highly valued in the Islamic tradition.39 It is precisely insight into his limitations that gave Alexander authority as a source of information. That Alexander’s quest for immortality would require a physical journey is significant. As indicated above, geography defines Alexander as an epistemic character in the Islamic tradition. The contrast with the sedentary Umayyad caliph ʿAbd al-​Malik in The City of Brass could hardly be greater. There is another important difference between the two men as well. While ʿAbd al-​Malik seems to have seen himself as God’s representative, but a leader of a specific polity, medieval Muslim authors presented Alexander as a prophetic figure.40 They took their cue from the Qur’an where Alexander –​or rather, the ‘man with the two horns’, as he is called in Scripture –​is directed by God. Curiously, the story of Alexander in the Qur’an is an embedded story. The frame story is Muhammad’s prophecy. Tested by the Jews of Medina, he recounts in elliptical style the story of Alexander who traverses the full range of humanly accessible space. The geography of Alexander’s actions is mythical and generic. As the Islamic empire expanded in the centuries subsequent to Muhammad’s movement, this myth was mapped on a new political and religious landscape. The edges of the known world moved further apart, as indeed the story of Sallām and Alexander’s wall illustrates. Following the logic of the trope of the world conqueror, it may not have been desirable to remain outside of his general realm of human history or civilization. As ever more diverse Muslims claimed their own space in Islamic salvation history, they stretched the edges of Alexander’s conquests to incorporate their own lands of origin. Where Adam might signify humankind in its pre-​history, Alexander did so for history. Alexander thus served as a prism which offered a spatial or geographical mode of acquiring and communicating knowledge. But because he was a figure of history and his past entangled with the past of an ever-​expanding Muslim oikumene, his story also functioned in a temporal or historical mode. As an epistemic figure, Alexander took on several roles. He obtained knowledge, traveling, discovering and enquiring. He was also a more passive recipient of information, addressee of pseudo-​Aristotelian letters of advice in matters of philosophy and state, but also divine guidance. And he appears in the historiography of knowledge. Persian historians held him responsible for the dispersion of what they claimed to be knowledge gathered at the Achaemenid court.41 We can thus recognize an instrumental and pedagogical function of Alexander as an epistemic figure as well. To communicate information through narrative, as attested widely in medieval Islamic literature, is another function with parallels in classical literature.42 The geographical expanse of Alexander’s biography and his wide-​ranging, almost existential ambition facilitated an equally broad perspective on the world. He presumably also lends the information he obtains an air of desirability and nobility because it is subject of Alexander’s knowledge. What the value of this information is, however, depends on what kind of Alexander obtains it, at what point in his journey we encounter him and whether he is one of us. In other words, as sources of information, literary characters, historical or otherwise, are not neutral. Even as figures of authority, the circumstances of the way they obtain and disseminate information require evaluation against the backdrop of how they act on that information.

Conclusions Alexander’s story is not unique in world literature. In the Buddhist tradition, Prince Gautama led a sheltered life. When he left the palace, he was for the first time exposed to the dire realities of human life: sickness, old age and death. His mode of obtaining this information was to see it. These are the most existential matters of the human experience, although love too ranks 118

119

Seeing and recognition

high. Versions of this story circulated in the Islamic world under the title Bilawhar wa-​Būdhāsaf, or Barlaam and Josaphat in Christian adaptation. From his own time onwards, perhaps even in his own mind, Alexander allowed audiences to contemplate these parameters of human existence. It was already in Roman times that ‘Alexander-​as-​meme’, as Diana Spencer puts it, was mobilized by political leaders in specific relationship to issues of life and death. ‘ ‘Picking up on Augustus’s visit to Alexander’s tomb and Caligula’s tomb-​robbing, we find that from Nero to Domitian, death and afterlife and even the perils of immortality are rarely far from Alexander.’ Still, in 1982, Andy Warhol presented his own43 Alexander, reflecting the artist’s own concerns with mortality.44 Paradoxically, these examples illustrate Alexander’s achievement of the heroic immortality of fame. In medieval narratives, the shift of the journey’s aim from material conquest and immortality to insight and knowledge turns the quest from failure to success. One aim becomes replaced by another. For Niẓāmī, the Alexander who had established his heroic credentials moved on to philosophical and spiritual wisdom in the second part of his biography. Unlike Shahriyar, Alexander approaches information with ethical self-​awareness. As stories in the Benjaminian sense, these accounts do more than describe Alexander’s vision and his insights into human nature. Despite the great value attributed to the seeing of eyewitnesses, the primary mode of conveying and receiving information in literature is not visual instances of recognition, but oral and verbal. These stories are also invitations to contemplation. Niẓāmī recounts an episode according to which the mirror was invented under Alexander’s rule. Because Alexander was the first to look into a mirror, whenever we look into a mirror, we see some of Alexander in ourselves. Alexander reveals himself here as an ancestor of the human recognition of the self with all out potential and limitations. The episode reinforces the suitability of Benjamin’s concept of the story as something malleable, flexible and widely applicable. It was the novel the philosopher described as concerned with a single individual, but Alexander seems larger than any single individual. Likewise, as a source of information, his appeal lies in the combination of the specific and the adaptable. Shahrazad herself understood the efficacy of stories. While she poses to kings and caliphs the question of which one they prefer, something heard or something seen, what she offers to king Shahriyar is not only something she has only heard, but Shahriyar himself does not see the stories and their locations himself. No object of amazement is brought to him. The narrative practice in the frame tale does not seem to conform to the epistemological principles in embedded stories. Shahriyar, however, sees the storyteller –​just like Hārūn who promises ʿAlī that ‘I am listening to you with my ears, looking at you with my eyes and paying attention to you with my heart.’ The same is true of the audience of the Arabian Nights who watched stories being told. There is more to a story than the text. Shahrazad, the storyteller whose authority rests on the threat of death, succeeds. Her success lies in deferring mortality –​her own, but that of her fellow women as well.

Notes 1 ‘Oven turns into cat. The word “ginger” is uttered and suddenly in place of the desk there is a fruit stand, in which I immediately recognize the desk. The Thousand and One Nights came to mind.’ Walter Benjamin, On Hashish (Cambridge 2006), 21. 2 For the following story see The Arabian Nights. Tales of 1001 Nights, translated by Malcolm C. Lyons, Robert Irwin and Ursula Lyons, 3 vols (London, 2010), vol. II, 68‒83. This translation is based on the Arabic text in the Calcutta II or Macnaghten edition which was published in 1839‒1842. Hārūn al-​ Rashīd is an actual historical figure and appears in several Arabian Nights stories. He ruled the Islamic empire from 786 to 809. The roots of the Arabian Nights go back to Persia and India, the Arabic version of this collection having been traced back to the ninth century. The preserved versions of

119

120

Anna Ayse Akasoy the collection date to later times, from about the fourteenth century onwards. The text gained global popularity with the French translation by Antoine Galland published in 1704 to 1717. For an introduction to the text and its history see Robert Irwin, The Arabian Nights: A Companion (London, 2004). 3 I read Jubayr and Budūr as a parody of a love story. The story contains jokes as well as parodies of conventions of love poetry, most notably in a food-​themed poem. 4 The Arabian Nights, Selected and edited by Daniel Heller-​Roazen, translated by Husain Haddawy (New York, 2010), 13. This version of the Nights dates to about the fourteenth century. The Arabic text edited by Muhsin Mahdi has been published as The Thousand and One Nights (Alf Layla wa-​Layla) from the Earliest Known Sources, 3 vols (Leiden, 1984‒94). 5 Al-​Maʾmūn, another Abbasid caliph, was a son of Hārūn’s and reigned from 813 to 833. The story al-​Maʾmūn hears is that of a Yemeni who moved to Baghdad and arranged for a poetic competition among six enslaved women of different appearances who are instructed to praise themselves and denigrate their opposite. 6 The Arabian Nights, trans. Lyons et al., vol. II, 520. ʿAbd al-​Malik ruled the caliphate from 685/​692 to 705. 7 Andras Hamori, ‘An Allegory from the Arabian Nights: The City of Brass’, Bulletin of the School of Oriental and African Studies 34 (1971), 9‒19, 9. 8 Bruce Fudge, ‘Signs of Scripture in the “City of Brass” ’, Journal of Qur’anic Studies 8 (2006), 88‒114. 9 ‘Der Erzähler’, in Walter Benjamin, Erzählen. Schriften zur Theorie der Narration und zur literarischen Prosa, Ausgewählt und mit einem Nachwort von Alexander Honold (Frankfurt, 2007), 103‒28. For an English translation see The Storyteller Essays, translation by Tess Lewis, introduction by Samuel Titan (New York, 2019), 48‒73. References are to the English version first, then to the German. 10 Thus, commenting on the function of wonders in this circle of Arabian Nights stories, Bruce Fudge states that ‘the strangeness of sights and situations is meant not to make us ask “what will happen next,” but to impress upon us the awe that God’s power demands.’ ‘Underworlds and Otherworlds in The Thousand and One Nights’, Middle Eastern Literatures 15/​3 (2012), 257‒72, 264. 11 Linda Jones, The Power of Oratory in the Medieval Muslim World (New York, 2012); ‘ “He Cried and He Made Others Cry”: Crying as a Sign of Pietistic Authenticity or Deception in Islamic Preaching’, in Elina Gertsman (ed.), Crying in the Middle Ages (London, 2012), 102‒35. 12 Anna Akasoy, ‘Tibet in Islamic Geography and Cartography: A Survey of Arabic and Persian Sources’, in Anna Akasoy, Charles Burnett and Ronit Yoeli-​Tlalim (eds), Islam and Tibet: Interactions along the Musk Routes (Farnham, 2010), 17‒41. 13 Emeri van Donzel and Andrea Schmidt, Gog and Magog in Early Syriac and Islamic Sources. Sallam’s Quest for Alexander’s Wall (Leiden, 2009), 133‒51 for the Arabic text and an English translation. 14 Thomas Bauer, Die Kultur der Ambiguität. Eine andere Geschichte des Islams (Berlin, 2011). 15 Daniel Beaumont, ‘King, Queen, Master, Slave: The Master/​Slave Dialectic and the Thousand and One Nights’, Neophilologus 82 (1998), 335‒56. For the episode involving ‘box woman’ see Samar Attar and Gerhard Fischer, ‘Promiscuity, Emancipation, Submission: The Civilizing Process and the Establishment of a Female Role Model in the Frame-​story of 1001 Nights’, Arab Studies Quarterly 13/​ 3‒4 (1991), 1‒18. 16 Philip Kennedy, Recognition in the Arabic Narrative Tradition: Discovery, Deliverance and Delusion (Edinburgh, 2016). 17 For a short summary and the quotation Peter J. Chelkowski, Mirror of the Invisible World. Tales from the Khamseh of Nizami (New York, 1975), 22. 18 It was this scene too which the Turkish writer Orhan Pamuk put at the center of his historical novel My Name is Red, which is set among Ottoman miniature painters. With Khosrow or Shirin seeing in the painting, especially Shirin seeing a painting, we find ourselves with a similar constellation of multiple layers of seeing. 19 José Miguel Puerta Vílchez, ‘Art and Aesthetics in the Work of Ibn Ḥazm of Cordoba’, in Camilla Adang, Maribel Fierro and Sabine Schmidtke (eds), Ibn Ḥazm of Cordoba. The Life and Works of a Controversial Thinker (Leiden, 2013), 253‒372, 305 (translation amended). 20 Houari Touati, Islam and Travel in the Middle Ages (Chicago, 2010), especially 101‒55. 21 See Simon Swain (ed.), Seeing the Face, Seeing the Soul: Polemon’s Physiognomy from Classical Antiquity to Medieval Islam (Oxford, 2007), especially the contributions by Antonella Ghersetti. For connections between the epistemological underpinnings of medieval physiognomy and Islamic theology see Anna Akasoy, ‘Arabic Physiognomy as a Link between Medicine and Astrology’, in Anna Akasoy, Charles

120

121

Seeing and recognition Burnett and Ronit Yoeli-​Tlalim (eds), Astro-​medicine: Astrology and Medicine, East and West (Florence, 2008), 119‒41. For Ibn Ḥazm see Puerta Vílchez, ‘Art and Aesthetics’, in 303‒11. 22 For the question whether the sense of hearing is indeed required in order to be a Muslim see Maribel Fierro, ‘Notes on Reason, Language and Conversion in the 13th Century in the Iberian Peninsula’, Quaderns de la Mediterrània 9 (2008) = Ramon Llull and Islam, the Beginning of Dialogue, 49‒58. 23 Touati, Islam and Travel, 9. 24 Touati, Islam and Travel. 25 Especially James Montgomery, ‘Travelling Autopsies: Ibn Faḍlān and the Bulghār’, Middle Eastern Literatures 7/​1 (2004), 3-​32 and ‘Spectral Armies, Snakes, and a Giant from Gog and Magog: Ibn Faḍlān as Eyewitness among the Volga Bulghārs’, The Medieval History Journal 9/​1 (2006), 63‒87. 26 Recent scholarship has advanced theories about a relationship between Pyrrho’s skepticism and Buddhist views of the world which developed as a result of Alexander’s conquests. See especially Christopher I. Beckwith, Greek Buddha. Pyrrho’s Encounter with Early Buddhism in Central Asia (Princeton, 2015). For al-​Ghazālī, see Frank Griffel, Al-​Ghazālī’s Philosophical Theology (New York, 2009) and Kenneth Garden, The First Islamic Reviver: Abu Hamid al-​Ghazali and his Revival of the Islamic Sciences (New York, 2013). 27 Al-​Ghazali, Deliverance from Error, An Annotated Translation of al-​Munqidh min al Ḍalāl and Other Relevant Works of al-​Ghazālī by Richard Joseph McCarthy (Louisville, 1980), 56. 28 Benjamin, ‘The Storyteller’, 49-​50/​104‒105, 66/​121 for the Arabian Nights. Aboubakr Chraïbi, ‘Situation, Motivation, and Action in the Arabian Nights’, in Ulrich Marzolph and Richard van Leeuwen (eds), The Arabian Nights Encyclopedia (Santa Barbara, 2004), vol. I, 5‒9, 6. 29 For Alexander the Great in the Islamic world see for Arabic Faustina Doufikar-​Aerts, Alexander Magnus Arabicus. A Survey of the Alexander Tradition through Seven Centuries: From Pseudo-​Callisthenes to Ṣūrī (Paris, 2010), and for Persian literature Haila Manteghi, Alexander the Great in the Persian Tradition: History, Myth and Legend in Medieval Iran (London, 2018). 30 Richard Stoneman, ‘Naked Philosophers: The Brahmans in the Alexander Historians and the Alexander Romance’, The Journal of Hellenic Studies 115 (1995), 99‒114. ‘Sustained textual influence’ is Kennedy’s phrase to describe the impact of Joseph stories. See Recognition in the Arabic Narrative Tradition, 163. For geographical journeys which reveal and reflect theological truth see also Fudge, ‘Underworlds and Otherworlds’. 31 Yuriko Yamanaka, ‘Alexander in the Thousand and One Nights and the Ghazālī Connection’, in Tetsuo Nishio and Yuriko Yamanaka (eds), The Arabian Nights and Orientalism: Perspectives from East and West (London, 2006), 93‒115. For a translation of the episode see The Arabian Nights: Tales of 1001 Nights, trans. Malcolm C. Lyons (London, 2008), ii, 325‒6. 32 The Questions of Milinda, a conversation between the Indo-​Greek king Menander (d. 130 BCE) and a Buddhist sage, are a key source for this subject. For this text and the dialectical method see David Sick, ‘When Socrates Met the Buddha. Greek and Indian Dialectic in Hellenistic Bactria and India’, Journal of the Royal Asiatic Society, third series 17/​3 (2007), 253‒78, and Richard Stoneman, The Greek Experience of India: From Alexander to the Indo-​Greeks (Princeton, 2019), 365‒74. 33 Plutarch, The Parallel Lives (Cambridge, 1919), VII, 259 (14). 34 The Greek Alexander Romance, translated with an introduction and notes by Richard Stoneman (London, 1991), 133. See also Stoneman, ‘Naked Philosophers’, 99. 35 Carol King, ‘Plutarch, Alexander and Dream Divination’, Illinois Classical Studies 38 (2013), 81‒111. 36 Fabian Horn, ‘The Death of Achilles in the Iliad’, Mnemosyne March 2020, 1‒28. 37 Georgios Halkias, ‘When the Greeks Converted the Buddha. Asymmetrical Transfers of Knowledge in Indo-​Greek Cultures’, in Peter Wick and Volker Rabens (eds), Religions and Trade: Religious Formation, Transformation and Cross-​Cultural Exchange between East and West (Leiden, 2013), 65‒115, 103. See also Paul Christesen and Sarah C. Murray, ‘Macedonian Religion’, in Joseph Roisman and Ian Worthington (eds), A Companion to Ancient Macedonia (Oxford, 2010), 428‒45, especially 431‒5 for death as passage to the afterlife and 436‒40 for tombs instead of temples. 38 Stoneman, ‘Naked Philosophers’. 39 Olga M. Davidson, ‘The Burden of Mortality: Alexander and the Dead in Persian Epic and Beyond’, in Epic and History, eds David Konstan and Kurt A. Raaflaub (Chichester, 2010), 212‒22. For the recognition of one’s mortality as a ‘common katabatic theme’ see Fudge, ‘Underworlds and Otherworlds’, 260. 40 For ʿAbd al-​Malik and the Umayyad conception of the caliphate, see Patricia Crone and Martin Hinds, God’s Caliph: Religious Authority in the First Centuries of Islam (Cambridge, 1986), 24‒42.

121

122

Anna Ayse Akasoy 41 Dimitri Gutas, Greek Thought, Arabic Culture: The Graeco-​Arabic Translation Movement in Baghdad and Early Abbasid Society (2nd‒4th/​8th‒10th Centuries) (London, 1998), 36‒41. 42 Spencer, ‘Roman Alexanders’, 272‒3 for ‘Alexander and the sum of all knowledge’, especially 273 about Pliny, Naturalis historia. 43 Diana Spencer, ‘Roman Alexanders. Epistemology and Identity’, in Waldemar Heckel and Lawrence A. Tritle (eds), Alexander the Great: A New History (Chichester, 2009), 251‒74, 265. For ‘Alexander-​ as-​meme’, see page 252. 44 Travis Nygard and Vincent Tomasso, ‘Andy Warhol’s Alexander the Great: An Ancient Portrait for Alexander Iolas in a Postmodern Frame’, Classical Receptions Journal 8/​2 (2016), 253‒75.

122

123

7 AVICENNA ON INFORMATION PROCESSING AND ABSTRACTION Luis Xavier López-​Farjeat

Introduction As one of the most iconic philosophers of the classical Islamic intellectual tradition, Avicenna’s (980–​1037) insights into the nature of human cognition still have parlance in contemporary theories of philosophy of mind, particularly his approach to sense-​perception, abstraction, intentional states, and self-​ awareness. In his philosophical works the notion of ‘information’ (taṣawwur) refers to the process by which the human mind acquires content provided by sense-​perception. Therefore, in this sense, Avicenna has a great deal to say about the topic of ‘information’, that is, the acquisition of cognitive or mental content (maʿnā). Maʿnā is a term commonly used among Islamic scholars and could refer the meaning of a word, properties of the external world, and also cognitive or mental content (forms, images, intentions, intelligible forms or concepts). Generally, Avicenna uses the term in this last sense. Inspired by Aristotle’s On the Soul and its Peripatetic and Neoplatonic commentators (D’Ancona 2010; Gutas 2016), Avicenna developed sophisticated explanations for the human mind’s ability to process information coming from the external world, as well as the way in which the mind (or ‘intellective soul’, to use Avicenna’s terminology) apprehends intelligible forms, that is, the highest form of cognitive content. Nevertheless, his explanation for human cognition is problematic: on the one hand, it can be read as conceiving of knowledge as an empirical process in which the human rational soul is able to abstract intelligible forms by itself; on the other hand, his conception of human cognition, at times, raises the need for an external agent for cognition and, in this sense, relies entirely on a specific cosmological-​metaphysical model. These two approaches, as we shall see, have led to divergent interpretations of Avicenna regarding the way in which cognitive content reaches the human mind. Avicenna deals with cognitive processes in several places. However, here I shall limit the discussion to the Book of the Healing (Kitāb al-​Shifāʾ), one of his most important philosophical works, where his most exhaustive discussion of the cognitive capacities of human beings is found. The Book of the Healing is an encyclopedic work divided into four main parts: logic, natural philosophy (which is itself divided into eight sections), mathematics, and metaphysics. Given that, for Avicenna, the cognitive acts are due to capacities of the intellective soul, he deals with these acts predictably in the Book of the Soul (Kitāb al-​Nafs),1 which is one of the eight 123

124

Luis Xavier López-Farjeat

sections devoted to natural philosophy. After studying several matters such as the heavens, the celestial bodies, and the characteristics of natural bodies, Avicenna deals with animate bodies or living beings in the sublunary realm, that is, in the sphere of changing nature. Living beings are characterized, as can be empirically verified, by their possession of a set of capacities, starting with the most basic, as displayed by plants (nourishment, growth, reproduction), and continuing, with increasingly complexity, to more developed capacities, as displayed by animals and human beings (sense-​perception, desire, locomotion, and, in the case of human beings, rational thought) (Avicenna 1959: 1.5, 39–​51). Avicenna was deeply concerned with the way by which the human mind processes sensible data in order to produce cognitive content, a process which requires the intervention of several mediating faculties, distinct from sensation and intellect, such as, for instance, imagination and the so-​called estimative faculty (a faculty that, as will be shown, receives non-​sensible content linked to sensible forms). Hence, he raised several questions in this regard: What is the process by which sense data or sensible forms can become concepts or, in Avicenna’s terminology, ‘intelligible forms’? Can these concepts remain in our minds or intellects? Is our mind able to store ‘intelligible forms’ by itself? Can human knowledge be explained by the personal efforts of our rational individual soul, or is some external agent needed to explain our capacity to process sense-​perceptions and attain ‘intelligible forms’? In order to provide a satisfactory response to the previous questions, it is essential to explain the most basic operation through which the human mind attains cognitive contents or information, namely, sense-​perception. As will be shown, for Avicenna, sense-​perception consists in extracting or separating the form of perceptible objects (Avicenna 1959: 2.2, 58). This means that our mind or intellect can take something from the external world and then process the information it has taken or extracted. This act of ‘extraction’ is what Avicenna means by ‘abstraction’ (tajrīd). Through this process, the human mind apprehends or separates the form from an external object, transforming it into what we commonly call ‘knowledge’. From this perspective, it seems that human knowledge can be explained as an empirical process, one which consists of a human capacity to abstract or separate, by itself, cognitive contents in different degrees: forms, images, intentions, and intelligible forms. In the last degree, sense-​perceptions are transformed into intelligible forms. Nevertheless—​hearkening back to a problem which originates in Aristotle’s On the Soul 3.5—​what is apprehended through sense-​perception is the form of the individuated, particular sensible thing (that is, the form of a particular object), and what characterizes human cognition—​in contrast to animal cognition—​is the capacity to transform individuated forms into non-​individuated intelligible forms. In other words, human beings are able to understand intellectually, that is, to ‘conceptualize’ and, in this sense, transcend the information provided by sense-​perception: human beings not only know ‘this horse’, but also the ‘horseness’ as such. The empirical approach, however, does not fully explain how sensible forms are transformed into intelligible forms and how these intelligible forms are stored in our particular minds. This opacity justifies, as some scholars have proposed, why Avicenna holds the need for an external agent that works both as the provider and the storage of intelligible forms. To sum up, it appears that Avicenna has two divergent models for human knowledge that, as will be shown, scholarly literature has intensely debated between, along with raising various ways to reconcile these divergent models. In what follows, I expand upon the first model by explaining the importance of sense-​perception and introducing Avicenna’s understanding of ‘abstraction’. The aim of this first section is to problematize the empirical approach and pave the way towards understanding the need for an external agent, which Avicenna calls the ‘active intellect’ (ʿaql al-​faʿʿāl). In the second section, I explore the second model by discussing the 124

125

Avicenna on information and abstraction

different interpretations as to why the ‘active intellect’ is necessary for the cognitive process. (Here, I shall present my own modest provisional conciliatory position.) However, the aim of this article is not to provide a resolution or a novel alternative interpretation to this problem, but to describe the way in which Avicenna understands human knowledge. While avoiding anachronism, I will ultimately evaluate if this way of explaining cognition has something to say to contemporary epistemology.

The cognitive content of sense-​perception Sense-​perception (idrāk) is a capacity shared by animals and human beings. As mentioned, Avicenna defines ‘sense-​perception’ as the extraction of the form (ṣūra) of a perceptible object (Avicenna 1959: 2.2, 58–​74). The sense-​perceptive faculties are divided into two parts, namely, the external and the internal. The external faculties are the external senses that make possible the most elementary mode of perception, namely, sensation, which consists in extracting the sensible forms or properties of the external objects. These sensible forms are impressed on the organ corresponding to each external sense. The external senses supply information that will be processed by the internal faculties or internal senses. Avicenna recognizes the following external senses: (1) sight, which, through the eyes, perceives the image of the forms of colored bodies; (2) hearing, which, through the eardrum, perceives sounds produced by the vibration of the air when two solid bodies strike each other; (3) smell, which, through the nasal passages, perceives the odor of the bodies transmitted through the air; (4) taste, which, through the nerves distributed in and around the tongue, perceives the sensible forms of tastes when they are mingled with the saliva; (5) and the sense of touch, which, through the nerves distributed over the skin, perceives four different kinds of sensations (hot and cold, dry and moist, hard and soft, rough and smooth) by means of four different faculties. Since Avicenna divides touch into four faculties, his account comprises a total number of eight external senses (Avicenna 1959: 1.5, 42–​43). Sensation is not the only kind of perception. There are other kinds of perception depending on the degree of abstraction or separation from matter: the forms, intentions (maʿānī), and intelligible forms or intelligibles (maʿqūlāt). As mentioned, the internal senses, which are located in the brain, are able to process the information provided by the external senses. So, they can grasp content and meaning that is not part of sensation, that is, not directly perceived by the external senses, even if this content is in conjunction with the perceptibility of objects. Avicenna calls this kind of content ‘intentions’ (maʿānī). These ‘intentions’ are ‘meanings of the external world’ that cannot be reduced to mere sensorial experience or to images, nor are they intellective content. When Avicenna explains these ‘meanings’, he mentions the well-​known example of the sheep perceiving danger in the wolf. When the sheep faces the wolf, this innocent animal sees nothing but a mass of dark fur and a pair of jaws and perceives its smell. The sheep does not ‘see’ or ‘smell’ hostility, unfriendliness, or danger. The senses of sight and smell limit themselves to particular sensations that must be processed by some internal faculty that endows the aforementioned sensations with meaning, so the sheep can ‘understand’ that the wolf represents ‘hostility’ and ‘danger’ (Avicenna 1959: 1.5, 43). In other words, intentions cannot be identified with the input data coming from the external senses. Intentions are properties that are recognized in the external world and, in this sense, as Hasse (2000: 132) holds, they are ontologically independent from the mind and are present in the perceptible object. They can only be perceived by living beings endowed with the internal faculties suitable for perceiving them. Through these faculties, human beings, and even some animals, are able to perceive and abstract the content of the external world thus receiving 125

126

Luis Xavier López-Farjeat

pre-​intellective or pre-​conceptual content, that is, information or, in other words, what Kaukua (2014b: 109–​110) calls ‘quasi or proto-​concepts’. This level of perception transcends the reception of mere sensible data and consists in ‘perceiving the world as something’. Note, however, that although intentions are already cognitive content, given that they are still particularized, they cannot be considered intellective content. The internal faculties that are needed to be able to perceive intentions, according to Avicenna, are the following: (1) the common sense (ḥiss al-​mushtarak), whose function is to receive and unify the sensible forms provided by the external senses; (2) the retentive imagination (al-​khayāl), devoted to the retention of images; (3) the estimative faculty (wahm), responsible for the reception of intentions that are in turn retained by (4) memory (dhikr); and (5) the compositive imagination (mutakhayyila), a permanently active faculty that composes and divides both imaginative forms and intentions. Avicenna adds, in the case of human beings, a faculty named mufakkira (cogitatio, as it was translated in the Latin world), which is the compositive imagination (mutakhayyila) under the control of the intellect, whose activity is crucial for preparing the intelligible forms: Then comes the faculty denominated imaginative in relation to the animal soul, and the cogitative faculty in relation to the human soul. This is a faculty found in the middle ventricle of the brain, in the cerebellar vermis, and its function is to combine and divide at its will any [form] in the images. Then there is the estimative faculty, found in the posterior area of the middle ventricle. This one perceives the intentions not perceptible by the senses but that only can be with respect to particular sensible objects, like the faculty with which a sheep judges that it must flee from the wolf and love the sheep. It could also act with respect to objects of the imagination by combining and dividing them. (Avicenna 1959: 1.5, 45) Note that these faculties provide forms, images, and intentions. These cognitive faculties play an active role in cognition and, as a consequence, they are providers of pre-​intellectual cognitive content, that is, non-​conceptual content common to animals and human beings. The estimative faculty is the highest in the case of animals which, as mentioned in the example of the sheep perceiving the hostility of the wolf, are able to receive non-​sensible content (intentions) linked to sensible forms, which is stored in the memory and constructed by the compositive imagination (mutakhayyila). Avicenna explains that the estimative faculty is able to control the compositive imagination, enabling animals to attain a representation of the external world and react and act according to what is beneficial for their survival (Avicenna 1959: 4.3, 184–​185; López-​Farjeat 2012; 2016). In sum, perception is not limited to the sensitive experience; rather, it includes the reception of non-​material properties conjointly with perceptible objects (Black 1993; 2000). Any living organism endowed with the appropriate faculties to perceive is capable of retaining cognitive information that allows it to interact with the external world. In the case of human beings, the operation of the internal faculties is somewhat more sophisticated. In rational beings the compositive imagination can sometimes act independently, while at other times it is subordinated to the estimative faculty (as happens in the case of animals) or to the intellect. When acting independently, the compositive imagination, as already mentioned, acts permanently, combining and separating images in the mind without any voluntary control, as happens, for instance, when dreaming. Note, then, that the compositive imagination is perpetually providing information to the estimative faculty and, hence, our mind (or intellective soul, in Avicenna’s terminology) is constantly producing images. But, as 126

127

Avicenna on information and abstraction

mentioned, the compositive imagination can be used by the intellect, thus acting as the cogitative faculty: it is natural for us to combine part of the perceptible objects with other parts, not according to the form that we have found in them externally nor even affirming that some of them exist or do not. Thus, in us there must be some faculty by which we do that. When the intellect is using it, this [faculty] is called the cogitative, while when the animal is using it, it is called compositive imagination. (Avicenna 1959: 4.1, 165–​166) The cogitative faculty, thus, is not a distinct faculty, but the compositive imagination assisting the intellect in its own operation, namely, the attainment of intelligible forms. The proper function of the cogitative faculty is to provide the intellect with images in which there are potentially intelligible forms which are not linked to any particular existing individual. The cogitative faculty separates the particularities or accidents of the specific perceptible objects and focuses on the intelligible form, for example, after meeting Zaid and Salma, one can imagine a human being without thinking about Zaid or Salma, in particular, even though the image is constructed from data gained through meeting Zaid and Salma. As can be seen, up to this point, Avicenna’s explanation of the cognitive process in the Book of the Soul emphasizes the role of the internal senses. In a passage that will be commented upon later (Avicenna 1959: 4.5, 234–​236), he explains that the process of intellection consists in bringing cognitive contents (maʿānī) from potency to act, after having been received by the internal faculties, that is, the intellect abstracts the content of the internal faculties, transforming this content into intelligible forms. According to this model, perception is the foundation for the formation of intelligible forms. Those things that have been extracted from perceptible objects and their associations (images and intentions) reach the intellect (ʿaql) completely separated (mujarrada) from matter and its accidents. But there are some difficulties when trying to clearly understand how content in the internal faculties can turn into intelligible forms. In the Book of the Soul, Avicenna seems to suggest that the individual rational soul, by its own powers, is able to separate any material residue from sensible forms, thus completing the cognitive process by attaining an intelligible form or, in other words, a concept. As Hasse (2001: 46–​ 58) has pointed out, through the cognitive process, starting with sensation, Avicenna describes four different kinds of abstraction:2 (1) the external senses abstract the sensible forms of perceptible objects; (2) the retentive imagination abstracts the image of the particular form provided by the common sense; (3) the estimative power abstracts intentions or non-​sensible properties and, in the case of human beings, the cogitative faculty abstracts the particularities of perceptible objects; (4) the intellective faculty removes any material residue from the potentially ‘intelligible form’ provided by the cogitative faculty completing the process of abstraction (intazaʿa). When the abstracted form comes to be in the intellect, it is impressed in what Avicenna calls the ‘material intellect’,3 that is, the capacity of the intellective soul to receive intelligible forms. (This process is analogous to the way a sensible image comes to be impressed in the eye.) But how is it that a cognitive process that, until now, has depended on organic faculties (the external and internal senses) to obtain mental contents, can, by itself, separate material residue from its content? Avicenna answers this question in the Book of the Soul 5.5. Here, he explains the need for the assistance of a separate immaterial intellect, which provides the immaterial non-​individuated intelligible forms to the human intellect. Using Avicenna’s terminology, the material intellect (that is, our particular intellect) is able to retain the sensible forms and needs to be actualized in order to understand dematerialized forms, that is, the intelligible forms.4 127

128

Luis Xavier López-Farjeat

Hence, to complete the cognitive process, an external intellect, what Avicenna calls the ‘active intellect’, needs to provide intelligible forms. Given the limitations of the material intellect, which depends on organic faculties and could not therefore dematerialized forms by itself, the active intellect is described as an entirely immaterial and separate intellect that is always in actuality, which stores every intelligible form. Avicenna’s explanation for the relationship between the material and the active intellect is complex and raises some problems that, as will be explained in the following section, have been discussed in depth among scholars in the field. In general terms, Avicenna’s explanation of the intellective process not only needs to be understood from an epistemological perspective, but also from a cosmological-​metaphysical one. In his cosmology, Avicenna outlines an emanation scheme (‘emanative’ insofar as it is brought about through an ‘overflow’ of God’s generosity), in which, through its self-​contemplation, the necessary being (God) causes the existence of a series of intellects. These intellects, through thinking themselves, the intellect above them, and God, cause the existence of subsequent intellects (as well as the respective souls and bodies of the heavens). The lowest of these immaterial intellects, the tenth intellect proceeding from God, is the active intellect, which emanates on us (Avicenna 2005: 318–​326). This emanation includes the emanation of forms to the world, rendering the active intellect’s role as the storehouse of intelligible forms (Avicenna 2005: 337). According to Avicenna, actualization of intelligible forms is only possible when the material human intellect and the active intellect interact. This relation is explained as a conjunction (ittiṣāl) between both intellects. From this perspective, everything indicates that the active intellect functions as the efficient cause for the human intellection of immaterial intelligible forms. That said, the precise role the active intellect plays in the process has generated much debate.

Abstraction and emanation The previous section explored the role of sense-​perception in Avicenna’s description of the cognitive process. It has been shown that the cognitive process culminates in the knowledge of intelligible forms, which remains ambiguous without appealing to something outside of sense-​ perception, namely, the active intellect, whose role is not entirely clear. It appears as if what is usually called the ‘pre-​noetic’ or pre-​conceptual level of abstraction, that is, the extraction of forms and intentions, involving bodily powers both external and internal, is only a kind of preparation for the intellectual apprehension of intelligible forms via the active intellect. Put simply, the apprehension of intelligible forms is prompted or stimulated by sense-​perception but comes about through the conjunction with the separate active intellect. What is called ‘intellectual understanding’ involves both sense-​perception and the conjunction with the active intellect. And, given that the active intellect is the storehouse for all the intelligible forms, this separate intellect delivers forms to the human intellect through an emanation (fayḍ). This doctrine is difficult to explain and, of course, is a matter of controversy. What has been prepared at the ‘pre-​noetic’ level is potentially intelligible. When this content is prepared and almost dematerialized through the cogitative faculty, the active intellect actualizes the human material intellect by emanating the intelligible form. Thus, human cognition involves a conjunction, that is, a combination between, on the one side, the dematerialized form abstracted by the individual intellective soul by its own powers, and, on the other side, the active intellect emanating the intelligible form. This suggests that, even though the levels of abstraction occurring at the pre-​noetic level are necessary for cognition to take place, in a strict and complete sense, cognition is the reception of intelligible forms through an emanation. This was the commonly accepted interpretation of Avicenna among his Medieval Latin receivers, 128

129

Avicenna on information and abstraction

like Thomas Aquinas, as well as the interpretation of numerous contemporary scholars, for example, Étienne Gilson (1930), Fazlur Rahman (1958), Herbert Davidson (1992), Deborah Black (2005; 2014), Richard C. Taylor (2005; while in 2019 he has reconsidered his position emphasizing the influence of Themisitus in the understanding of abstraction), and Olga Lizzini (2010, while emphasizing Avicenna’s Neoplatonic sources). Other interpreters, like Dag Hasse (2000; 2001) and Dimitri Gutas (2001; 2012; 2013; 2014), have emphasized the importance of abstraction from an empirical point of view, limiting the role of the active intellect to that of a storehouse for intelligible forms.5 Gutas’ (2012) approach is particularly interesting. He holds that the rational soul of the newly born infants is a tabula rasa that, as it grows, acquires information about the external world through experience (mushāhada), that is, sense-​perception and reflection or, in other words, information from the external and inner world (that is, self-​awareness). All of the content received in the material intellect arises from experience. Thus, the operation of the intellect only consists in the methodic ratiocination of finding middle terms for syllogistic thinking. The intellect (or the mind) has no innate or a priori content but has a natural disposition (fiṭra) to receive content from experience. Therefore, according to Gutas, the active intellect does not emanate intelligible forms to the material intellect. At first sight, this appears to be a strange interpretation, given that, certainly, Avicenna thinks that the active intellect intervenes in the cognitive process. Gutas does not deny that Avicenna’s ‘empirical epistemology’ is embedded in an emanationist cosmological and ontological framework, according to which, as previously explained, the world is generated by successive emanations from the intellects originating from the necessary being (God). He clearly sees that there is a tension between Avicenna’s ‘empirical’ explanation of cognition and his emanationist cosmology. However, he thinks that ‘the concept of the emanation of the intelligibles from the active intellect has its place in his cosmology and it serves to solve essentially an ontological problem, not an epistemological one, which is the location of the intelligibles’ (Gutas 2012: 411). In this sense, from the epistemological perspective, the function of the active intellect is minimally relevant, limited to storing the intelligible forms. The storage of the intelligible forms or, in other words, intellectual memory is an intriguing topic. Recall that for Avicenna, as previously explained, memory, as a part of the internal faculties, stores only intentions (which are associated with images). But intellective knowledge is different. How can immaterial, intelligible forms (concepts) be retained in a material body? Could one simply credit it to a power of the mind? No, because our mind retains, in the internal faculties located in the brain, sensible forms, images, and intentions, which are divisible and combinable because they are still linked to the sensible realm. The intelligible forms, in contrast, are immaterial, so they cannot reside in a material receptacle. The only alternative, thus, is that they are in a separate, immaterial intellect, namely, the active intellect, whose function is to actively and eternally think the intelligible forms. However, according to Gutas’ interpretation: The active intellect itself, as the storage of the intelligibles, is completely passive and never initiates the ‘effluence’; Avicenna is quite explicit about this: ‘The active principle [i.e., the active intellect] lets flow upon the [human rational] soul form after form in accordance with the demand by the soul; and when the soul turns away from it [the active intellect], then the effluence is broken off’. [Avicenna 1959: 245–​246] (Gutas 2012: 412) Taylor, who disagrees with this assessment, admits that Gutas has provided relevant remarks concerning the description of cognition as a naturalistic process. However, he rejects the 129

130

Luis Xavier López-Farjeat

interpretation of Avicenna as an empiricist, particularly the diminishment of the role of the active intellect for the cognitive process to a mere storehouse, rendering its importance practically null. According to Taylor’s interpretation, the actualization of intelligible forms in human knowing can only be accomplished through the intervention of the active intellect (Taylor 2016: 278–​279). In a recent contribution, Taylor (2019) points out that Avicenna describes the active intellect as pure actuality. Hence, the notion of ‘storage’ cannot be taken literally, because this would imply that the active intellect has the potency to receive the intelligible forms abstracted by the rational soul. From this inconsistency, Taylor critically revises Hasse’s and Gutas’ interpretations (that is, the active intellect as storehouse), and he raises relevant objections to both. He argues that they do not determine what kind of activity (and consequently, the kind of causality) the active intellect carries out in the cognitive process. In a later article, as Taylor observes, Hasse (2013) explains that the active intellect is Avicenna’s solution to the impossibility of storing intelligible forms in a material memory. Therefore, the rational particular soul has the disposition to cause forms, stored in the active intellect, to emanate to itself whenever it needs them. However, according to Taylor, this position leaves unclear the causal connection between the human intellect and the intelligible forms of the active intellect. Taylor also discusses, in my view, a relevant article by Tommaso Alpina (2014), who tries to reformulate how the two models could complement each other, that is, the abstractionist (or empiricist) and the emanationist models. In line with the traditional emanationist interpretation, Alpina holds that the active intellect provides the condition of possibility for the actualization of the human intellect, so its presence is necessary to actualize the potentiality of the human intellect. At the ontological level, maintains Alpina, the active intellect is the ‘collector’ of intellectual forms. In line with Gutas and Hasse, Alpina states that since Avicenna denies memory the capability to store intellectual forms, a repository of already acquired intelligible forms is required. In this way, it is not necessary to acquire a form anew every time we know something, rather human beings simply recover the forms stored in the active intellect (as if recovering information from ‘the cloud’, to use a computing analogy). This is only possible because there is a correspondence between imaginative particulars in the human intellect and intellectual forms in the active intellect. Alpina explains Ultimately, imaginative particulars and intellectual forms can be said to be the same with respect to their essence, namely with respect to the formal core they share, while they differ according to their way of existence, which is connected with particularity in one case and with universality in the other. (Alpina 2014: 165) Alpina’s interpretation, as Taylor sees, shares the same problem as that of Gutas and Hasse: the active intellect ‘is a fully actual intellect without potentiality and receptivity, so it cannot literally “collect” intellectual forms abstracted by the individual soul’ (Taylor 2019: 71). Taylor’s proposal is to interpret Avicenna as drawing upon Themistius’ solution to this problem. Taylor shows that, in his Paraphrase of the De Anima of Aristotle (1996), Themistius also posits the existence of an active intellect that plays an intrinsic role in the formation of intelligible forms. According to Themistius, while the human soul is capable of collecting the images provided by external senses, the active intellect assists the human soul (the material or potential intellect) in actualizing itself as a knower of intelligible forms in act, so it can become an actual intellect. This assistance, however, does not consist in emanating forms upon the individual’s potential intellect, but in guiding the human intellect: 130

131

Avicenna on information and abstraction

in its organization of images towards a sound abstraction on the intelligible in accord with the corresponding intelligibles in act present in the Agent Intellect itself. This assures that all abstracted intelligibles are correlated with those in the Agent Intellect and that science is universal, veridical, and eternal for all. In this way there is a perfect identity of the form or essence of the intelligibles in the human intellect with the form or essence of the intelligibles in the Agent Intellect, with the intelligibles in each remaining ontologically distinct in distinct subjects. In this, Themistius is not in accord with Avicenna in denying intellectual memory to the rational soul. But the Avicennian notion of a connecting with the Agent Intellect for the understanding of intelligibles in act which had previously been abstracted can be explained using the Themistian notion of the collaboration and correspondence of the human intellect and the Agent Intellect in forming the intelligibles in act in the human. (Taylor 2019: 75) This is a reasonable way to understand the idea of conjunction with the active intellect. It is well known that both Themistius and Alexander of Aphrodisias were sources for Avicenna’s reconstruction of Aristotle’s On the Soul (Gutas 2013). Taylor is not the first to notice this influence.6 However, as far as I know, he is the first to identify these influential passages. In my view, Taylor’s approach successfully combines and harmonizes the two apparently divergent perspectives in Avicenna’s conception of cognition. Another recent interpretation is that of Stephen Ogden (2020), who, in the line of Jon McGinnis7 (2007: 130–​137; 2013: 52–​57) and Alpina (that is, the active intellect as a condition of possibility for the actualization of the human intellect), tries to reconstruct a via media between the two perspectives, arguing that the active intellect emanates a power for abstraction into human beings. In this sense, the abstractionist (or empirical) and the emanationist approaches are necessary and co-​efficient causes of human cognition. Ogden holds that the key to reconciling abstractionism and emanationism is to understand that the active intellect moves our potential intellect into actuality by emanating, not the intelligible forms, but the ‘power of abstraction’. He provides textual evidence from the Book of Demonstration and from the Book of the Soul. Here, I only quote the latter passage: For when the intellective faculty examines the particulars that are in the retentive imagination, and the active intellect sheds light in us upon them (which we discussed), the things abstracted (mujarrada) from matter and its associations are transformed and impressed upon the rational soul. Being transformed not in the sense that [the particulars] themselves are transferred from the imagination to our intellect, and not in the sense that the intention (al-​maʿnā) immersed in the attachments [of matter]—​ which in itself and in consideration of its essence is abstract (mujarrad) [from matter]—​ makes something like itself. Rather, [it is] in the sense that the revision of [the things abstracted from matter and its associations] prepares the soul for something abstract from the active intellect to emanate upon them … Thus, when some relation to this form takes place for the rational soul by means of the light shed by the active intellect, then from [the relation to the form] there comes to be in [the soul] something that in one way is of its genus and in another way is not, just as when light falls on colored objects, producing an effect that is not in every way [reduced] to their sum. Hence the things in the imaginative faculty, which are potentially intelligible, become intelligible in actuality, not they themselves but rather what is collected from them. But just as the effect coming, by the mediation of light, from the sensible forms is not itself 131

132

Luis Xavier López-Farjeat

those forms but rather something related to them, engendered by the mediation of the light in the recipient facing [the light], so also when the rational soul examines these imaginative forms and the light of the active intellect comes into a kind of conjunction with them, [the soul] is prepared for the abstractions (mujarradāt) of these forms [free] from [material] taints. (Avicenna 1959: 5.5, 235–​236)8 Ogden is right to focus on this passage. As he notes, emanationists and abstractionists have used this same passage to support their own views. Note that the passage describes both abstraction and the active intellect’s emanation as necessary for the actualization of the intelligible forms. According to Ogden’s interpretation, in this passage, Avicenna is arguing that the active intellect provides, not the intelligible forms, but the power to abstract them, a provision he compares to light. If we interpret Avicenna this way, with Ogden, abstraction of intelligible forms comes about through both sense-​perception via the individual and voluntary efforts of the human agent and, at the same time, the efficient causal role of the active intellect via emanation. The role of the active intellect is neither simply a storehouse for intelligible forms nor the sole causal source of intelligible forms for the human intellect. It is, rather, responsible for activating the power of abstraction, so the human intellect can actualize the intelligible forms by itself (even while requiring a ‘conjunction’ with the active intellect for its abstractive power). This is what it means to say that abstraction and emanation are co-​efficient causes of human cognition. What I find attractive about Ogden’s interpretation is his emphasis on the substantial proximity between Avicenna and Alfarabi (d. 950), another great thinker of the Islamic philosophical tradition, better known as ‘the second master’ (Aristotle being the first). Also influenced by Alexander of Aphrodisias and Themistius, Alfarabi describes the relationship of the active intellect and the human intellect as similar to that between the sun and sight (Alfarabi 1938: 24–​27; McGinnis and Reisman 2007: 74–​75). Following Aristotle’s theory of vision, Alfarabi equates the sun, as the source of light, with the active intellect; he does not equate the active intellect with the light itself. In this sense, following Davidson’s (1992: 50) description, the light from the sun does four things: (1) it enters the eye and turns its potential vision into actual vision; (2) it enables potentially visible colors to become visible in actuality; (3) it itself becomes visible to the eye; and (4) it renders the sun, its source, visible to the eye. The analogy works as follows: (1) the active intellect turns the human intellect into intellect in actuality; (2) it transforms (istaḥāla) potential intelligible forms (sense impressions stored in the imagination, in the case of Alfarabi, and the potentially intelligible forms provided by the cogitative power, in the case of Avicenna) into actual intelligible forms; (3) it itself becomes an intelligible object for the human intellect; and (4) it renders the active intellect as an intelligible object for the human intellect. This analogy suggests that the active intellect sheds light or provides some sort of illumination by which the human intellect is actualized, enabling it to grasp the forms by abstraction. In the end, abstraction cannot take place without the assistance of the active intellect. Regardless of whether this resemblance between Alfarabi and Avicenna is correct or not, what can be seen is that both involve the interaction between the human intellect and the active intellect in their explanations of human cognition. (In Ogden’s words, both intellects are co-​efficient causes.) This would explain why Avicenna gives such an active role to perception and, at the same time, he finds that perception needs assistance to grasp immaterial, non-​individuated intelligible forms. In my view, it is not necessary to consider abstraction and emanationism as two divergent explanations. Avicenna understands both abstraction and emanation as complementary processes. 132

133

Avicenna on information and abstraction

I think the abstractionist or empiricist model of Avicenna’s theory of cognition privileges the role of non-​intellectual content (in contemporary terminology, non-​conceptual content), that is, forms, images, and intentions. These contents come from sense-​perception as information in the mind (the human intellect) for cognition. Hence, according to the abstractionist or empiricist model, perception plays an active role in providing the human and animal minds non-​conceptual content, for example, in the case of intentional cognition. At this level of the cognitive process, namely, the pre-​noetic, it is possible to describe Avicenna as an empiricist. It is clear, however, that for Avicenna, perfect knowledge is necessarily attained outside of bodily powers and mental representations, that is, knowledge transcends the embodied mind. As shown, the emanationist model emphasizes the role of the active intellect, which can be understood in several different ways: as that which provides intelligible forms, as that which provides first principles, or that which actualizes the power of abstraction. I am inclined to think that the most conciliatory path to Avicenna’s texts blends the influences of Themistius and Alfarabi, that is, the active intellect as guiding the human intellect and the active intellect as actualizing the power of abstraction, respectively. In other words, I think Taylor (active intellect as guiding human intellect) and Ogden (active intellect as emanating the human power to abstract) provide valuable insights into the contributions of Rahman, Black, Lizzini, Hasse, Gutas, and others. Apart from those scholarly debates on whether Avicenna’s philosophy is internally coherent, could his understanding of cognition have a place in contemporary epistemology?

Avicennian issues in contemporary epistemology As mentioned, at the pre-​noetic level, Avicenna raises issues which are still relevant in contemporary epistemology and philosophy of mind, namely, whether intentional contents are already knowledge. Hasse (2000: 127–​ 141) has argued that Avicenna’s explanation of intentional contents is not a theory of intentionality, as understood in contemporary (post-​Brentanian) terms, given that Avicennian intentional states are part of the process of sense-​perception, not the content of mental acts more generally.9 In my view, although in his explanation of ‘intentions’ Avicenna is referring to the capacity to perceive the connotation of a particular perceptible object, I think he conceives perception as playing an active role in cognition, that is, as a provider of cognitive contents that become mental states (emotions, decisions, beliefs) which allow humans and even some animals to react to states of affairs in the external world and direct their behavior according to particular situations. In this sense, I think that Avicenna’s theory of perception could be considered in the current discussion of intentional states and the cognitive value of non-​conceptual contents.10 An emanationist cosmology could sound odd for contemporary epistemologists, but the problem, in the context of immaterial knowledge known by material beings, renders a separate intellect to complete the cognitive process more palatable. Avicenna is concerned with the problem of thinking intelligible forms, that is, concepts, by ourselves. He needs the assistance of an external agent or principle to explain how the dematerialization of mental images takes place. Avicenna is posing a metaphysical question: can human knowledge be reduced to a simple relation between the mind and the world, or is there another kind of relation that allows the human mind to transcend materiality to obtain universal knowledge? To put it in other words: can knowledge be described in purely empirical terms? Would that even count as knowledge? It certainly would not count as knowledge in a strong sense within Avicenna’s own Neoplatonic and Aristotelian tradition. However, this is a legitimate epistemological question. Avicenna, as do Kant and others, thinks that in order to transform sense-​perceptual experience 133

134

Luis Xavier López-Farjeat

into knowledge, that is, intelligent consciousness, the intervention of something transcendent (or transcendental) to sense-​perception is required. Avicenna is convinced that human knowledge cannot be explained without the experience provided by sense-​perception, however, he raises the same question many ask today: is sense-​perception the only source for human knowledge?11

Notes 1 I use Rahman’s edition (Avicenna 1959). Sometimes I use the translation by McGinnis and Reisman (2007) slightly modified to better capture the point I am making. 2 Hasse is referring to Avicenna’s mature works of his middle period including ‘The State of the Human Soul’ in the al-​Maʿād (The Destination), the De anima of the al-​Shifāʾ, al-​Najāt (The Salvation), and the Mashriqiyyūn (The Easterners). 3 Avicenna calls this intellect ʿaql hayūlāni, that is, ‘material intellect’. The terminology is confusing since one might think that it is called ‘material’ because the intellect is itself material. However, the intellect for Avicenna is immaterial. The name ‘material intellect’ is applied because is the one who receives the material forms. 4 Avicenna distinguishes in the cognitive process four modes of interaction between the human intellect and the intelligible forms. He refers to the material intellect (ʿaql hayūlāni) (the capacity of the rational soul to receive immaterial intelligible forms), the dispositional intellect (ʿaql bi al-​malakah) (when the intellect has the fist principles provided by the active intellect), the actual intellect (ʿaql bi-​l-​fiʿl) (when the intelligible forms have been attained but they are not actively thought), and the acquired intellect (ʿaql mustafād) (when the intellect has received the intelligible forms and is actually thinking them). 5 In Hasse (2013) there is a different perspective regarding the relevance of the active intellect in abstraction. 6 See Rahman (1952: 101) and Afnan (1958: 145). 7 Ogden agrees with Jon McGinnis that both the emanationist and abstractionist interpretations are partially right; however he is not entirely convinced by McGinnis’s resolution. McGinnis thinks that the human intellect prepares the intelligible forms through abstraction, while, at the same time, the active intellect emanates the ‘intellectualizing forms’ upon the forms in the human intellect. These ‘intellectualizing forms’, he proposes, are the intelligible accidents mentioned by Avicenna in the introduction to the logic of the Shifāʾ. In other words, through abstraction, the intellect abstracts the intelligible form, itself, of a perceptible object from the external world, while, at the same time, the active intellect provides the intelligible accidents central to the conceptualization of the form, that is, the differentiating species of the intelligible forms. Ogden thinks, following Hasse (2013), that McGinnis’s theory lacks textual evidence. In his psychological works, Avicenna never identifies the intelligible forms with the intelligible accidents mentioned in the Logic, nor does he say that those concepts emanate from the active intellect. Nevertheless, Ogden thinks that, despite this inconvenience, McGinnis is seeking a conciliatory way. For his part, Taylor argues that McGinnis is wrong because ‘an essence or form is classified as particular only subsequent to its realization in a subject’ (Taylor 2019: 60), therefore, there cannot be particular accidents emanated from the active intellect and received by the sensible form found in the human mind (specifically in the cogitative faculty). 8 Ogden presents his own translation. I have kept some of his phraseology but I have made several modifications. 9 Certainly, the proximity between Avicenna’s theory of intentions and modern and contemporary theories of intentionality would imply a discussion on the definition of ‘intentions’. Hasse is right in his ontological description of ‘intentions’. He is also right in stating that the relational character of intentions does not imply that intentions do not have an ontological status. I think, however, that intentional content is linked to emotions and decisions that could be considered, in contemporary terminology, ‘mental states’. In this sense, it is possible approach Avicenna to contemporary theories of intentionality. 10 For Avicenna’s contributions to the concept of intentionality see Kaukua (2014a). 11 I am extremely grateful for the comments of Nicholas Oschman, Stephen Ogden, and Jörg Tellkamp.

134

135

Avicenna on information and abstraction

References Afnan, S.M. (1958) Avicenna: His Life and Works, London: George Allen & Unwin LTD. Alpina, T. (2014) ‘Intellectual Knowledge, Active Intellect and Intellectual Memory in Avicenna’s Kitāb al-​Nafs and Its Aristotelian Background’, Documenti e studi sulla tradizione filosofica medievale 25, 131–​183. Alfarabi (1938) Risalat fī al-​ʿaql (Letter on the Intellect), M. Bouyges (ed.), Beirut: Imprimerie Catholique. Avicenna (1959) Avicenna’s De anima, being the Psychological Part of Kitāb al-​Shifāʾ, F. Rahman (ed.), London: Oxford University Press. Avicenna. (2005) The Metaphysics of the Healing, M. E. Marmura (tr.), Provo, UT: Brigham Young University Press. Black, D. (1993) ‘Estimation in Avicenna: The Logical and Psychological Dimensions’, Dialogue 32, 219–​258. Black, D. (2000) ‘Imagination and Estimation: Arabic Paradigms and Western Transformations’, Topoi 19, 59–​75. Black, D. (2005) ‘Psychology: Soul and Intellect’, in P. Adamson and R.C. Taylor (eds.), The Cambridge Companion to Arabic Philosophy, Cambridge: Cambridge University Press, 308–​326. Black, D. (2014) ‘How Do We Acquire Concepts? Avicenna on Abstraction and Emanation’, in J. Hause (ed.), Debates in Medieval Philosophy, New York: Routledge. D’Ancona, C. (2010) ‘Degrees of Abstraction in Avicenna: How to Combine Aristotle’s De Anima and The Enneads’, in S. Knuuttila and P. Kärkkäinen (eds.), Theories of Perception in Medieval and Early Modern Philosophy, Dordrecht: Springer, 47–​71. Davidson, H. (1992) Alfarabi, Avicenna, and Averroes on Intellect, Oxford: Oxford University Press. Gilson, É. (1930) ‘Les sources gréco-​arabes de l’augustinisme avicennisant,’ Archives d’Histoire Doctrinale et Littéraire du Moyen Age 5, 1–​107. Gutas, D. (2001) ‘Intuition and Thinking: The Evolving Structure of Avicenna’s Epistemology’, in R. Wisnovsky (ed.), Aspects of Avicenna, Princeton: Markus Wiener. Gutas, D. (2012) ‘The Empiricism of Avicenna’, Oriens 40, 391–​436. Gutas, D. (2013) ‘Avicenna’s Philosophical Project’, in P. Adamson (ed.), Interpreting Avicenna; Critical Essays, Cambridge: Cambridge University Press, 28–​47. Gutas, D. (2014) Avicenna and the Aristotelian Tradition, 2nd ed., Leiden: Brill, 2014. Gutas, D. (2016) ‘Ibn Sina [Avicenna]’, in E. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Fall 2016 Edition) URL https://​plato.stanford.edu/​archives/​fall2016/​entries/​ibn-​sina/​. Hasse, D.N. (2000) Avicenna’s De Anima in the Latin West: The Formation of a Peripatetic Philosophy of the Soul, 1160–​1300, London and Turin: Warburg Institute. Hasse, D.N. (2001) ‘Avicenna on Abstraction’, in R. Wisnovsky (ed.), Aspects of Avicenna, Princeton: Markus Wiener. Hasse, D.N. (2013) ‘Avicenna’s Epistemological Optimism’, in P. Adamson (ed.), Interpreting Avicenna: Critical Essays, Cambridge: Cambridge University Press. Kaukua, J. (2014a) ‘The Problem of Intentionality’, Documenti e studi sulla tradizione filosofica medievale 25, 216–​242. Kaukua, J. (2014b) ‘Avicenna on the Soul’s Activity in Perception’, in J. F. Silva and M. Yrjönsuuri (eds.), Active Perception in the History of Philosophy: From Plato to Modern Philosophy, Dordrecht: Springer, 99‒116. King, A. (2018) Language Between God and the Poets: Ma‘nā in the Eleventh Century, Oakland: University of California Press. Lizzini, O. (2010) Fluxus (fayḍ), Bari: Edizioni di Pagina. López-​Farjeat, L.X. (2012) ‘Self-​awareness (al-​shuʿūr bi-​al-​dhāt) in Human and Non-​human Animals in Avicenna’s Psychological Writings’, in A. Vigo (ed.), Oikeiosis and the Natural Bases of Morality, Hildesheim: Georg Olms Verlag. López-​Farjeat, L.X. (2016) ‘Avicenna on Non-​conceptual Content and Self-​Awareness in Non-​human Animals’, in J. Kaukua and T. Ekenberg (eds.), Subjectivity and Selfhood in Medieval and Early Modern Philosophy, Switzerland: Springer, 61–​73. McGinnis, J. (2007) ‘Making Abstraction Less Abstract: The Logical, Psychological, and Metaphysical Dimensions of Avicenna’s Theory of Abstraction’, Proceedings of the American Catholic Philosophical Association 80, 169–​183. McGinnis, J. (2010) Avicenna, Oxford: Oxford University Press.

135

136

Luis Xavier López-Farjeat McGinnis, J. (2013) ‘New Light on Avicenna: Optics and Its Role in Avicennan Theories of Vision, Cognition and Emanation’, in L.X. López-​Farjeat and J.A. Tellkamp (eds.), Philosophical Psychology in Arabic Thought and the Latin Aristotelianism of the 13th Century, Paris: Librairie Philosophique J. Vrin, 41–​57. McGinnis, J. and Reisman, D. (2007) Classical Arabic Philosophy: An Anthology of Sources, Indianapolis: Hackett. Ogden, S. (2020) ‘Avicenna’s Emanated Abstraction’, Philosophers’ Imprint 30, 2–​26. Rahman, F. (1952) Avicenna’s Psychology: An English Translation of Kitāb al-​Najāt, Book II, Chapter VI with Historico-​ Philosophical Notes and Textual Improvements on the Cairo Edition, London: Oxford University Press. Rahman, F. (1958) Essence and Existence in Avicenna, London: Routledge. Taylor, R. (2005) ‘al-​Fārābī and Avicenna: Two Recent Contributions’, MESA Bulletin 39, 180–​182. Taylor, R. (2016) ‘The Epistemology of Abstraction’, in R.C. Taylor and L.X. López-​Farjeat (eds.), The Routledge Companion to Islamic Philosophy, London and New York: Routledge, 273–​284. Taylor, R. (2019) ‘Avicenna and the Issue of the Intellectual Abstraction of Intelligibles’, in M. Cameron (ed.), The History of the Philosophy of Mind, Vol. 2: Philosophy of Mind in the Early and High Middle Ages, New York: Routledge. Themistius (1996) On Aristotle’s On the Soul, R.B. Todd (transl.), Ithaca: Cornell University Press.

Further reading Alpina, T. (2014) ‘Intellectual Knowledge, Active Intellect and Intellectual Memory in Avicenna’s Kitāb al-​ Nafs and Its Aristotelian Background’, Documenti e studi sulla tradizione filosofica medievale 25, 131–​183. Black, D. (2014) ‘How Do We Acquire Concepts? Avicenna on Abstraction and Emanation’, in Debates in Medieval Philosophy, ed. J. Hause, New York: Routledge, 2014. McGinnis, J. (2010) Avicenna, Oxford: Oxford University Press. Taylor, R. (2019) ‘Avicenna and the Issue of the Intellectual Abstraction of Intelligibles’, in M. Cameron (ed.), The History of the Philosophy of Mind, Vol. 2: Philosophy of Mind in the Early and High Middle Ages, New York: Routledge.

136

137

8 THOMAS AQUINAS ON COGNITION AS INFORMATION Cecilia Trifogli

Cognizant beings The notion of information plays a central role in Aquinas’s theory of cognition. It is this notion that he uses in his account of the distinction between beings that are capable of cognition, like animals and human beings, and beings that are not capable of cognition, like non-​living things and plants. Cognizant beings are characterized as those capable of receiving information about the world, whereas non-​cognizant beings lack this capacity. In Aquinas’s own words: (1) Cognizant beings are distinguished from non-​cognizant beings in this respect, that the non-​cognizant beings possess only their own form, whereas a cognizant being is naturally apt to have also the form of some other thing. For the species of the thing cognized is in the one cognizing. Therefore, it is clear that the nature of a non-​ cognizant being is more contracted and limited, whereas the nature of cognizant beings has more breadth and extension; therefore, the Philosopher says (De Anima III) that ‘the soul is in a sense all things.’1 Aquinas does not use the term “information” in this passage. He does, however, use ideas that correspond to the etymology of this term: taking in a form, receiving a form. Cognizant beings are those capable of receiving information about something in the sense that they are those capable of receiving the form of that thing. Accordingly, it is the notion of form that is the crucial one in Aquinas’s account of cognition as information. In Aristotle’s ontology, the form of a thing is what ultimately accounts for the nature of that thing. Thus, both cognizant beings and non-​cognizant beings have a form. For example, a stone has a form in virtue of which it is a stone just like a human being has a form in virtue of which it is a human being. As Aquinas puts it in the passage above, all beings possess their own form. The difference is that some beings, namely, the cognizant beings, in addition to their own form, can receive, and so be informed by, the forms of other things, while the non-​cognizant beings cannot do this. For example, a stone only has its own form, that which makes it a stone, but not also the form of any other thing, whereas a human being that cognizes a stone also has the form of the stone in addition to its own form.

137

138

Cecilia Trifogli

But what does this mean? That a human being cognizing a stone and thus receiving the form of a stone becomes a stone? Obviously not. The form of a stone makes only the stone a stone, but does not also make the human being that cognizes it a stone. The form of a stone makes a human being able to cognize a stone without making it a stone. The crucial idea here is that the form of a stone exists in a stone in a different way from the way in which it exists in the human being cognizing it. In general terms, the idea is that the form of a thing cognized exists both in the thing cognized and in the thing that cognizes it but in different ways. Myles Burnyeat offers a beautiful illustration of this idea: (2) For an Aristotelian, both sensible and intelligible forms are present to the world in two irreducibly different ways, one of which is cognitive of the other. The form of tiger, for example, is active in the forests as the organizing principle of the life of the tiger, but it may also be present, differently, in the intellect of a zoologist who has reached a principled understanding of that kind of life. Similarly, the orange and black colouring of a tiger’s striped coat will also be present, differently, in the eye of its mate as they hunt together.2 By sensible and intelligible forms Burnyeat means the objects of sensory and intellectual cognition respectively, which are the two main kinds of cognition distinguished by Aristotle. In Aquinas’s influential interpretation of Aristotle’s theory of cognition, the objects of sensory cognition are particular accidental forms of substances, like the orange and black colour of a particular tiger in Burnyeat’s example, whereas the (primary) objects of intellectual cognition are the universal natures of substances, natures somehow common to all the substances of a given kind, like the form of tiger (what makes something a tiger) in Burnyeat’s example. To the distinction between sensible and intelligible forms as objects of cognition there corresponds a distinction between two cognitive faculties, namely, the senses and the intellect: the senses are powers of receiving sensible forms whereas the intellect is a power of receiving intelligible forms. Both the senses and the intellect receive their respective forms but in a different way from the way in which these forms exist or are received in the objects of cognition. One standard linguistic device with which Aquinas signals this difference is by using the term “species” to refer to the form existing in the cognitive faculty, as in text (1) above. According to this terminology, it is the species of the form of a tiger and the species of its black and orange colour that exist in the intellect and the senses (the eyes) respectively. The species of sensible forms are also called sensible species and the species of intelligible forms are called intelligible species.3 These terminological clarifications, however, do not shed light on the central question of what exactly is the difference between the way in which sensible or intelligible forms exist in the objects of cognition and the way in which they exist in the cognitive faculties, that is, the difference between a form (either sensible or intelligible) and the species corresponding to it.

“Receiving forms without the matter” In the case of sensory cognition, Aristotle himself apparently provides an answer to this question in a passage of his De anima where he gives an explicit description of the special way in which sensible forms are received and exist in the senses: (3) It is necessary to grasp, concerning the whole of perception generally, that perception is what is capable of receiving perceptible forms without the matter, as wax receives the

138

139

Thomas Aquinas on cognition as information

seal of a signet ring without the iron or gold. It acquires the golden or the metallic seal, but not insofar as it is gold or metal. In a similar way, perception is also in each case affected by what has the colour or taste or sound, but not insofar as each of these is said to be something, but rather insofar as each is of a certain quality, and corresponding to its proportion.4 Aristotle’s explicit description of the special way in which sensory faculties receive forms is condensed in the celebrated formula that they receive “the forms without the matter”. Aristotle illustrates this formula with the example of the wax and the signet ring: wax only receives the seal of a signet ring, that is, its shape or form, but not also the material (the gold or iron) of which the signet ring is made. This example is relatively clear, but it is not easy to figure out how it applies to the cognitive context. It is clear that the relevant forms that the senses receive are those of the corresponding sensible objects. For example, sight is what receives colours, and hearing is what receives sounds. The difficulty is with the qualification “without the matter” added to describe the peculiar way in which the senses are receptive of sensible forms. This qualification is obviously crucial in Aristotle’s mind but the passage offers little help when we try to understand this qualification.5 An eminent Aristotelian scholar, D. W. Hamlyn, pointed out a basic problem even with the part of the formula that Aristotle seems to take for granted, namely, the claim that the senses receive the sensible forms. Contrary to what Aristotle says at the beginning of text (3), Hamlyn remarked that the formula does not seem to be applicable to all cases of perception, rather “the range of applicability of the present formula is limited and it fits touch best. It is not easy to see how the eye can receive the colour when we see, or the ear sound when we hear”.6 The objection raised by Hamlyn originates from the assumption that, according to Aristotle, the sense-​organs are informed by the forms of the sensible objects in exactly the same way as the sensible objects themselves are informed by the sensible forms, that is, that sensible forms affect the senses in exactly the same way in which they affect the objects. This assumption explains why Hamlyn thinks that the Aristotelian formula fits touch best: when the organ of touch is acted upon by a hot body, it becomes hot just like the hot body; the organ of touch is informed by heat in the same way in which the hot body acting on it is informed by heat. This assumption, however, seems highly implausible in the case of other senses and in particular of sight. Do my eyes (or more precisely the eye-​jelly, which is supposed to be the organ of sight) become actually red when a see a red apple? Hamlyn thinks that this is not the case. But this is not the only view among Aristotelian scholars. In more recent times, Richard Sorabji has defended the apparently implausible view that my eyes do become red when I see a red apple and more generally that this is true for all senses. His contemporary Myles Burnyeat has strongly criticized Sorabji’s view on more than one occasion and taken the opposite view that the eyes do not become red in seeing a red apple and this is true for all other senses, including touch (this requires some explanation though). The contrast between Sorabji and Burnyeat has shaped the debate about Aristotle’s theory of perception for many years and it still does. The contrast is nowadays standardly described as a contrast between the letter and the spirit of Aristotle’s passage, or between Literalism and Spiritualism. In accordance with Sorabji’s view, Literalism maintains that a sensory organ literally takes on the sensible form of the corresponding sensible object. On the contrary, in accordance with Burnyeat’s view, Spiritualism denies that a sensory organ literally takes on the sensible form and posits that it takes on such a form in an alternative way, labelled “spiritual”.7

139

140

Cecilia Trifogli

Spiritual/​intentional vs natural existence The term “spiritual”, which is now commonly used in the Aristotelian literature, is actually taken from Aquinas’s characterization of perception as a spiritual change and of the mode of being of a sensible form in the sensory organ as a spiritual being. It is this characterization that Aquinas offers in his attempt to explain the Aristotelian formula about the reception of forms without the matter. Aquinas starts by considering a basic objection against the validity of the Aristotelian formula. The problem is about the qualification “without the matter”, which is supposed to capture what is special about the way in which the senses receive the sensible forms, but it seems to fail to do so. As Aquinas points out, in any kind of change, not only in sense perception, the patient (i.e., what is subject to change) receives from the agent of the change only its form and not also its matter. The relevant agents are in fact material substances, that is, composite of matter and form, but their matter does not have any role in the changes they produce. It is the form that is the active principle in a change and the change itself consists in the transfer of the form only from the agent to the patient, so that the patient receives only the form of the agent and not also its matter. For example, when fire acts on air and makes it hot, it gives air the form of heat only and not also its matter. But obviously air does not perceive the heat. Thus, perceiving the heat of fire cannot simply consist in receiving heat without the matter of fire.8 In Aquinas’s view, however, this problem is not a real one, because it originates from a wrong interpretation of the contrast that Aristotle intends to draw between receiving a form with the matter and receiving a form without the matter. The contrast is not about the entity received in a change: in one case a form together with the matter and in the other case a form only. For in all changes, for example both in becoming hot and in feeling hot, it is a form only that is received. Rather, the contrast is between two modes of reception of the same kind of thing or form, so that “with the matter” and “without the matter” qualify the mode of reception and not the thing received. Thus, every patient, and not only the senses, receives a form only from the agent. The difference is that while the senses receive a form in an “immaterial way” (without the matter), the patient of an ordinary change receives it in a “material way”. A sensible form, like heat, exists in a “material way” in the object that produces the sensation, for example fire, but in an “immaterial way” in the sensory organ. The terms “material” and “immaterial” are those suggested by the Aristotelian formula. The standard terms that Aquinas himself uses to describe this contrast are instead “natural” and “spiritual” or “intentional”. As Aquinas puts this: (4) For the form has a different manner of being in the sense and in the sense object: for in the sense object it has natural being, whereas in the sense it has intentional or spiritual being.9 The terms “intentional” and “spiritual” are used interchangeably by Aquinas in his theory of cognition. The term “intentional” is that more frequently used. The notion of intention is indeed a fundamental one in Aquinas’s theory of cognition. It is not an Aristotelian notion, however, and it is not Aquinas’s invention either. It is commonly used by medieval Latin philosophers, but its origins are to be found in medieval Arabic philosophy, and especially in the writings about cognition of the two major Islamic philosophers known to the West: Avicenna and Averroes.10 The Arabic background, however, does not offer much help when one tries to understand Aquinas’s own view about intentionality. More generally, a major difficulty is that despite its centrality the notion of intention is left without any adequate explanation by 140

141

Thomas Aquinas on cognition as information

Aquinas. But one important thing that Aquinas’s use of the term “intentional” in his interpretation of the Aristotelian formula makes clear is that he views intentionality in the first place as a mode of being or existence, a property that defines a kind of entity: it is the form of the object of cognition as existing in the cognizer that is an intention, or, equivalently, has intentional being. Using this terminology, then, cognition can be defined as an intentional information. But what does Aquinas mean by intentional being?11 The fact that Aquinas treats this term as synonym of “spiritual” may suggest that a sensible form has intentional or spiritual being when it is received in something immaterial or incorporeal. This suggestion arises from the ordinary understanding of the notion of spiritual, which contrasts the spiritual with the corporeal. This suggestion, however, is clearly wrong. As Aquinas himself points out, the intentional or spiritual being that he ascribes to a sensible form is the being that this form has in a sensory organ, and the sensory organ is a body and thus a material thing. For example, both a flag in which the form of red has natural being and the eyes in which it has intentional or spiritual being are bodies.12 Thus, the contrast between natural and spiritual/​intentional is not equivalent to the contrast between material/​corporeal and immaterial/​incorporeal. How then should this contrast be understood? Some light on this is shed by the following passage from the Summa Theologiae: (5) Change is of two kinds: one natural, the other spiritual. Natural change is that by which the form of the thing that produces the change is received in the thing that is subject to the change according to natural being, just as heat is received into the thing heated. Spiritual change instead is that by which the form of the thing that produces the change is received into the thing subject to the change according to spiritual being, just as the form of color is received into the pupil which does not thereby become colored.13 In this passage the distinction between natural and spiritual being is used to draw a parallel distinction between two kinds of change: natural change, which is the reception of a form in its natural being, and spiritual change, which is the reception of a form in its spiritual being. Cognitive changes, both sensory and intellectual, the changes that cognizant beings undergo in so far as cognizant, are spiritual changes and not natural ones. Thus, Aquinas here introduces another piece of the technical language of his theory of cognition. The abstract definitions of the two kinds of change, however, are not helpful because the notions of natural and spiritual being are left unexplained. More helpful are the examples that Aquinas uses to illustrate the two kinds of change. The example of natural change is being heated: the form of heat belonging to the agent, say fire, is received in the thing subject to heating, say water, in such a way that the thing is heated, that is, becomes actually hot. Thus, when heat exists in its natural mode of being in a thing, it makes that thing hot. The example of spiritual change is taken from sight: the colour of the thing seen is received in the pupil in a spiritual mode of being; this implies that the pupil does not become coloured as a result of receiving the colour of the thing seen. More generally, in a spiritual change the form received in the patient does not have the kind of natural effect that it has on the agent. Accordingly, Aquinas is explicit here that the spiritual change involved in perception is not an ordinary physical change. Unlike Sorabji and his followers, Aquinas believes that the eye-​jelly does not become red in seeing a red flag. Seeing a red flag requires the intentional information by the form of red but in so far as intentional this information does not result in the cognizer’s becoming actually red. Aquinas’s appeal to the notion of spiritual or intentional change to account for sensory cognition seems plausible in the case of sight, but not in the case of all the other senses as 141

142

Cecilia Trifogli

well. Consider, for example, touch. It is empirically evident that a hand (the organ of touch) becomes hot when it touches a hot body. Thus, the hand undergoes a natural change and not an intentional one when it perceives something hot. Aquinas concedes that a hand in touch with a hot body becomes hot. He points out, however, that this is not all there is to perceiving hot. Perceiving something hot, feeling hot, also necessarily involves a spiritual change, a change in which the form of heat is received in the hand in such a way that the hand does not become hot in virtue of this particular kind of reception. More generally, in Aquinas’s view, all senses, with no exception, require a spiritual change. Indeed, a spiritual change is the one distinctive of percipient beings. As Aquinas puts it: (6) For the operation of the senses, a spiritual change is required, whereby an intention of the sensible form is produced in the sense organ. Otherwise, if a natural change alone sufficed for the action of the sense, all natural bodies would feel when they undergo alteration.14 What the fact that the hand becomes hot when it touches a hot body shows, for Aquinas, is not that there is no intentional or spiritual change in touch but rather that touch involves a natural change in addition to the spiritual change. Actually, for Aquinas, all external senses involve a natural change with the only exception of sight. Sight is the “most spiritual” of the senses because it only involves spiritual change. The other four senses do involve a natural change either in the object of sense (hearing and smelling) or in the organ of sense (taste and touch). The natural changes, however, are simply necessary concomitants to perceiving but not essential to it in the sense that they are not constitutive of what perceiving is.15

Intentional species in the medium A general principle about agency in Aristotle’s natural philosophy states that there is no action at a distance, so that for an action or change to occur, the agent and the patient cannot be at a distance one from the other; rather they must be spatially together, that is, in contact. Aquinas, like the great majority of medieval philosophers, accepts this principle. Furthermore, although the principle was originally formulated for what he calls natural changes, Aquinas thinks that it is valid for spiritual changes too. In particular, the spiritual changes involved in the acts of the external senses conform to the “no action at a distance” principle, so that a sense must be in contact with the sensible object for a sensory act to occur. In the case of taste and touch it is clear that this contact-​condition is verified. But what about sight? Isn’t the case that we see things at a distance? Aquinas replies that, of course, we see things at a distance, but this does not violate the “no action at a distance” principle. And the reason for this is that the contact between the object of sight and the eyes of the perceiver looking at it is achieved by the object somehow moving towards the eyes. For example, seeing the red colour of a flag at a distance implies that the red colour of the flag propagates itself through the medium, so that it somehow covers the distance between the red flag and the perceiver by existing throughout the medium. Thus, the red colour of the flag exists not only in the flag but also in the medium, but in different ways: in the flag it has natural existence, whereas in the medium it has intentional/​spiritual existence. A clear sign that the kind of existence of the red colour in the medium is intentional/​spiritual rather than natural is that the air, just like the eyes, does not become red as a result of the colour red existing in it. In Aquinas’s view, then, it is the existence of intentional species of sensible objects in the medium that explains how some sensory acts satisfy the contact-​condition.16 142

143

Thomas Aquinas on cognition as information

That there are intentional species in the medium was a common belief at Aquinas’s time. Actually, this belief reflects the original context in which the theory of species was formulated.17 When this idea is incorporated within a theory of cognition, however, it gives rise to the question of how to account for the different effects that a species has when it exists in the medium and when it exists in a cognitive faculty. In our example, if the colour red of a flag has intentional existence not only in the eyes of the perceiver but also in the air, why is it the case that only the perceiver and not also the air sees the red colour of the flag? Aquinas himself does not explicitly raise this question in its generality, but he does offer a reply to it. Referring to the case of a scent producing an intentional change in the air, he maintains that air is not affected by a scent in such a way as to smell it because air does not have a sensitive power.18 Thus, in our example, the air does not see the red color of the flag, because, unlike the eyes, it lacks sight, that is, the relevant cognitive power. More generally, the media of perception do not perceive because they do not have the perceptual power. Aquinas’s answer is highly praised by Burnyeat for its simplicity and for being very Aristotelian.19 On the contrary, Pasnau is not happy with this answer and argues that we should not take it as Aquinas’s final answer to the question. The problem with this answer is that it is in conflict with Aquinas’s considered criterion for being a cognizant in text (1) above. According to this criterion, what makes the difference between a cognizant being and a non-​cognizant being is that a cognizant being is capable of intentional information while a non-​cognizant being does not have this capacity. Crucially, it is the capacity of intentional information only that accounts for the difference and not also a cognitive capacity. If we want to save Aquinas’s criterion, then, it seems that we should concede that the air and more generally the media of perception in so far as they are capable of intentional information are somehow cognizant beings. And this is the original view that Pasnau supports.20

Intelligible species According to Aquinas, intellectual cognition, just like sensory cognition, is an intentional information. The forms with intentional existence involved in intellectual cognition are called intelligible species. They differ from their sensible counterparts, that is, the sensible species, in two main respects: (i) the subject in which they are received and (ii) the way in which they are produced. (i) Sensible species are received in something corporeal, that is, in the sensory organs, which are bodies endowed with the relevant sensitive powers (or in the media). There is not, however, something analogous to a sensory organ in the case of intellectual cognition. This would be a body endowed with the intellectual power, that is, the power for intellectual cognition. We believe that the brain is the body in which the power for thought resides. But this is not how an Aristotelian conceives of thought. For an Aristotelian, thought is something essentially incorporeal, in the sense that the power responsible for thought or intellectual cognition is not the power of a bodily organ, so that there is not a bodily organ of thought. The power of intellectual cognition, which Aristotelians call the intellect, resides in the soul itself. Thus, there is not a corporeal recipient for intelligible species. They are received in the intellect itself. Aquinas’s main argument for the incorporeality of the intellect is a version of Aristotle’s famous argument in De Anima III.4, which is still matter of debate among Aristotelian scholars.21 The fundamental feature of this argument is that it is based on an inference “from content to constitution”, that is, it moves from a consideration about the content of thought—​the range of things we can think of—​to an ontological conclusion about the constitution of the intellect as power of thought. In Aquinas’s version of the argument the inference is: since we can think of bodies of any kind, then the intellect cannot be a body or have a bodily organ. The idea is that, 143

144

Cecilia Trifogli

if the intellect had a bodily nature, its bodily nature would restrict or somehow affect the range of bodies we can think of, it would “obstruct”, as it were, our intellectual grasp of some kinds of body. Aquinas illustrates how this “obstruction” principle works with the example of sight. The objects of sight are colours. But in order for sight to perceive every colour, as it actually does, the organ of sight—​the pupil or the transparent eye-​jelly—​must be without any colours. By analogy, in order for the intellect to have cognition of every corporeal nature, the intellect itself must be incorporeal. The “obstruction” principle is too general and vague to be convincing. It is even less convincing when it is interpreted within Aquinas’s theory of intentional information. Take the example of sight, which Aquinas considers a clear case in which the principle holds. Even if the pupil is colourless, it is not at all clear that its being colourless is a necessary condition for receiving every colour, as the obstruction-​principle requires. Indeed, if the red colour of a flag were received in the pupil in its natural being so that the pupil would become actually red, then the pupil would lose its original colour if it had one or its original colour would combine with the new one and this combination would somehow affect the capacity of sight to see the red colour of the flag. For Aquinas, however, the red colour of the flag is not received in the pupil in its natural being but in its intentional being, which implies that the pupil does not become red. Why then would the colour of the pupil affect the vision of the red of the flag if the red flag only produces an intentional change in the pupil? There is indeed a gap in Aquinas’s explanation of why the obstruction-​principle applies. In addition to Aristotle’s argument, Aquinas also has his own argument for the immateriality of the intellect.22 This argument too is based on an inference from the content of intellectual cognition to the constitution of the intellect. The relevant content is identified this time with universals as contrasted with particulars: for example, the nature of a stone rather than an individual stone. The inference in this case is: since the intellect thinks of universals, it cannot be material. The implicit assumption here is that, for Aquinas, matter is the principle of individuation of material substances. Accordingly, if the intellect were material, the species received in it would be individual rather than universal. This argument too, however, suffers from problems arising from the intentional reading of Aristotle’s theory of cognition.23 (ii) The contrast between individuals and universals is also fundamental in Aquinas’s account of the difference in the ways in which sensible species and intelligible species are produced. Sensible species are produced by the corresponding objects alone, so that the senses do not have any active role in this. The senses are simply receptive powers, not active powers, that is, powers to receive sensible species but not powers to produce them. For example, it is the colour red of the flag that is the only cause of the production of its sensible species in the eyes, while sight simply is the power that enables the eyes to receive this sensible species. Intelligible species instead are not produced by the corresponding objects alone but their production also requires the active contribution of the intellect. Thus, the intellect is not only a receptive power but is also an active power. In Aristotelian terms, there is both a possible intellect and an agent intellect: the possible intellect is the one that receives intelligible species whereas the agent intellect is the one that produces them. In Aquinas’s interpretation of Aristotle’s fragmentary remarks,24 the action of the agent intellect is necessary to bridge the gap between the way in which the object of intellectual cognition exists in extra-​mental reality and the way in which it exists in the intellect. For example, the nature of the stone does not exist as a self-​standing Platonic item over and above individual stones outside the mind. It only exists within individual stones as one of their essential components together with the principle of individuation of a stone. It exists in the intellect, however, as universal, that is, in separation from the individual features with which it is joined in extra-​mental reality.25 In terms of Aquinas’s distinction between natural and 144

145

Thomas Aquinas on cognition as information

intentional existence, the object of intellectual cognition is individual in its natural existence, but it is universal in its intentional existence. It is the intellect that makes it universal. And the intellect does this by abstraction: by abstracting a nature from the individual features, that is, by separating it conceptually from the individual features.26 Thus, an intelligible species as the intentional counterpart of the nature of a material substance is produced not only by this substance but also by the abstraction of the agent intellect.

The role of species in cognition Aquinas not only insists on the general point that intentional information is essential for a cognitive act, but also addresses specific questions about how intentional information is related to the corresponding cognitive act. The natural question that arises from the Aristotelian background of Aquinas’s theory is whether a cognitive act should be identified with the corresponding intentional information, so that, for example, the act of seeing the red colour of a flag exactly consists in the intentional change produced by the red colour in the eyes, and there is nothing more to seeing red than the intentional species of red existing in the eyes. Indeed, some Aristotelian assumptions about the nature of cognitive acts and of cognitive powers suggest a positive reply to this question. Aristotle thinks of a cognitive act as the actualization of the corresponding cognitive power. Moreover, he thinks of the corresponding potentiality of a cognitive power as receptivity, so that a cognitive power is a receptive power. But a receptive power is actualized when it actually receives the things of which it is receptive, and such things are the forms of the objects of cognition: in other terms, a cognitive power is actualized by intentional information. The conclusion that seems to follow is that a cognitive act as the actualization of a cognitive power consists in an intentional information. In particular, a sensory act is the reception of a sensible species in a sense-​organ, and an intellectual act is the reception of an intelligible species in the intellect.27 This “Aristotelian” view, which identifies intentional information with a cognitive act, however, is not exactly Aquinas’s considered view.28 Aquinas agrees that both intentional information and a cognitive act are actualizations of a cognitive power, but he does not identify them. He thinks instead that intentional information and a cognitive act are two distinct kinds of actualization of a cognitive power, and that they are ordered: the cognitive act is the ultimate actualization, and it presupposes that a cognitive power has already been actualized by an intentional information. Intentional information is something that a cognitive power requires in order to elicit a cognitive act. So, according to Aquinas, intentional species are not the same as cognitive acts but have a causal role with respect to them. The standard example with which Aquinas illustrates the causal role of species in cognition is that of heating. Aquinas makes it clear that heating and a cognitive act are different kinds of action: heating is an external action, that is, an action that has an effect external to the agent, whereas a cognitive act is an immanent action, an action that “remains in the agent”, that is, with no external effect. He also points out, however, that there are relevant similarities between heating and a cognitive act. The most fundamental one is about the nature of the active principle in the two actions: (7) as heat is the formal principle of heating in the fire, so is the species of the thing seen the formal principle of sight to the eye.29 In Aristotelian physics, heat is the formal cause of the heating, while fire itself is the efficient cause of this action. The idea here is that fire is the agent of the action of heating but it elicits 145

146

Cecilia Trifogli

this action in virtue of the form of heat existing in it. Aquinas holds that the same kind of causality must be ascribed both to sensible species, as the example of sight illustrates, and to intelligible species. They are forms existing in the sensory organs and in the intellect respectively that provide their subjects with the causal power to elicit cognitive acts. In Aquinas’s view, the case of heating also shows the crucial feature of an intentional species that makes it suitable to be the formal cause of a cognitive act. This is its being a likeness of the object of cognition: (8) And as the form from which proceeds an act tending to something external is a likeness of the object of the action, as heat in the heater is a likeness of the thing heated; so the form from which proceeds an action remaining in the agent is a likeness of the object. Hence that by which the sight sees is a likeness of the visible thing; and a likeness of the thing understood, that is, the intelligible species, is the form by which the intellect understands ... This also appears from the opinion of the ancient philosophers, who said that “like is known by like.” For they said that the soul knows the earth outside itself, by the earth within itself; and so of the rest. If, therefore, we take the species of the earth instead of the earth, according to Aristotle (De Anima iii, 8), who says “that a stone is not in the soul, but only a likeness of the stone”; it follows that the soul knows external things by means of its intelligible species.30 Being a likeness of the corresponding object of cognition is a relational property of an intentional species, one that relates a species to the corresponding object. Thus, the intentional species of red produced by the red flag is related to the colour red of the flag by being a likeness of it. Similarly, the intentional species of the nature of a stone is related to the nature of the stone by being a likeness of it. It is this relational property of an intentional species that is responsible for the intentionality of a cognitive act, that is, for the content of a cognitive act: its being about a determinate object. In our examples, consider the act of seeing the red colour of a flag. What accounts for the property of this visual act of having the red colour of the flag as its object rather than another colour is that the formal principle in virtue of which the percipient elicits this act is a likeness of the red colour of the flag. Similarly, it is because the intentional species of a stone is a likeness of the nature of a stone and not of the nature of a dog that an act of thought deriving from the intentional species of a stone is about stones rather than about dogs. The likeness of an intentional species to its object then is the crucial ingredient of Aquinas’s account of the causal role of such species in cognitive acts. But what is it exactly? Aquinas leaves this crucial question without a clear answer.31 The example of heat used in the passage above is not very helpful. The heat of fire is a likeness of the heat produced by fire in a log in a very strong sense: they are the same kind of form, that is, forms of exactly the same nature. Aquinas’s standard expression for this kind of similarity is “agreement in nature” (convenientia in natura). However, Aquinas explicitly denies that this natural likeness is the one that holds between the cognizer and the object cognized. This denial is consistent with his spiritualist interpretation of Aristotle’s view: an agreement in nature of the intentional species of heat existing in the senses or in the intellect would require that these cognitive faculties become actually hot, which is not the case, for Aquinas. The relevant cognitive likeness is described as a likeness with respect to representation (repraesentatio): (9) All that is required between cognizer and cognized is a likeness in terms of representation, not a likeness in terms of an agreement in nature. For it’s plain that the form of the stone in the soul is of a far higher nature than the form of the stone in 146

147

Thomas Aquinas on cognition as information

matter. But that form, insofar as it represents the stone, is to that extent the principle leading to its cognition.32 With Aquinas’s example, the intentional species of stone as a form existing in the intellect is of a different nature from the form of the stone existing in the stone. A clear indication of this difference is that the intentional species of a stone is an immaterial form (a form existing in an immaterial subject, i.e., the intellect), whereas the form of the stone existing in the stone is a material form. Despite this difference in nature, the intentional species of a stone is a likeness of the form of the stone because it is a representation of it. The notion of representational likeness is not properly developed by Aquinas. His scattered remarks on the topic suggest that representational likeness is not only distinct from natural likeness but also independent from it. For example, God has a minimal degree of natural likeness to creatures but a maximal degree of representational likeness to them. Aquinas does not provide, however, any positive description of the nature of such likeness or determinate conditions that a form existing in a cognitive faculty must satisfy to play this representational role. Aquinas’s emphasis on the distinction between natural and representational likeness gives rise to an important question about the ontological status of intentional species. Is an intentional species really the same kind of form as the corresponding form in the object of cognition in a different mode of existence? Aquinas’s criterion for being a cognizer and his interpretation of the Aristotelian formula “receiving form without the matter” provide strong evidence for a positive answer to this question.33 For example, the form of stone existing in the stone and its intentional counterpart existing in the intellect are the same kind of form in two different modes of existence: natural existence in the stone and intentional existence in the intellect. It is not clear, however, how this positive answer can be reconciled with his claim in text (9) above that these two forms do not have the same nature. Is the difference in nature between the two forms simply a difference in their modes of existence? Aquinas does not explicitly address this question and the ontological status of an intentional species remains a matter of controversy among Aquinas’s scholars.34

Notes 1 Thomas Aquinas, ST 1a, q. 14, a. 1c (English translation of the passages from ST is based on that of the Fathers of the English Dominican Province). 2 Burnyeat 2001: 141. 3 For comprehensive overviews of Aquinas’s theory of cognition, see Kretzmann 1993; MacDonald 1993; Pickavé 2012. 4 Aristotle, De anima II.5, 424a17–​24 (translation from Aristotle 2016). 5 In this short presentation of the debate among Aristotelian scholars I closely follow my paper Trifogli 2019: 245–​7. 6 Aristotle 1968: 113. 7 For the references to the relevant articles by Sorabji and Burnyeat and an extensive presentation of the debate see Caston 2005. 8 Thomas Aquinas, InDA II, c. 24, 168b18–​169a26 (transl. Pasnau 282). 9 Thomas Aquinas, InDA II, c. 24, 169a53–​b56 (transl. Pasnau 283). 10 On the Arabic background, see Black 2010. 11 An excellent discussion of this question is in Pasnau 1997: 31–​47. 12 Thomas Aquinas, InDA II, c. 24, 169b76–​170a89 (transl. Pasnau, 283–​4). Despite Aquinas’s explicit rejection of this understanding of the immaterial reception proper to the senses, Hamlyn ascribed it to Aquinas and maintained that Aquinas thinks that what receives the sensible forms are not the sensory organs but the soul itself, which is immaterial. See Hamlyn 1966: 46. The first scholar to point out that Hamlyn’s interpretation is not the correct one is possibly S.M. Cohen 1982.

147

148

Cecilia Trifogli 13 Thomas Aquinas, ST 1a, q. 78, a. 3c. 14 Thomas Aquinas, ST 1a, q. 78, a. 3c. 15 For a defence of this reading of the relationship between natural and intentional changes in perception see Burnyeat 2001: 131–​7. 16 Thomas Aquinas, InDA II, c. 20, 152a24–​153a88 (transl. Pasnau 254–​6). 17 This context is a theory of natural agency based on the propagation of species. The most influential exponent of this theory is the 13th-​century philosopher Roger Bacon. See Lindberg 1983. 18 Thomas Aquinas, InDA II, c. 24, 171b187–​195 (transl. Pasnau 287). 19 Burnyeat 2001: 149–​51. 20 Pasnau 1997: 47–​60. 21 Aristotle, De anima III.4, 429a10–​27; Thomas Aquinas, InDA III, c. 1, 203a91–​204b221 (transl. Pasnau 344–​7). For a very extensive discussion of this argument see Caston 2000. 22 Thomas Aquinas, ST 1a, q. 75, a. 5c. 23 On Aquinas’s arguments, see Pasnau 2012: 354–​7. 24 Aristotle, De anima III.5. On the controversy about the agent intellect among Aristotelian interpreters see the comments by Shields in Aristotle 2016: 312–​29. 25 Thomas Aquinas, ST 1a, q. 79, a. 3. 26 Thomas Aquinas, ST 1a, q. 85, a. 1. 27 For a defence of this view against an alternative reading by Sorabji see Burnyeat 1992. 28 The Aristotelian view was endorsed by one of Aquinas’s most faithful disciples, Giles of Rome. See Trifogli forthcoming. 29 Thomas Aquinas, ST 1a, q. 56, a. 1c. 30 Thomas Aquinas, ST 1a, q. 85, a. 2c. 31 An excellent discussion of this question is in Pasnau 1997: 105–​13. 32 Passage quoted from Pasnau 1997: 108. 33 See pp. 137–9. 34 For a detailed and illuminating discussion of this controversial issue see Brower and Brower-​Toland 2008. See also my paper Trifogli 2019: 264‒7.

References Aristotle (1968) De anima. Books II and III, Translation, Introduction and Notes by D. W. Hamlyn, Oxford: Clarendon Press. Aristotle (2016) De anima, Translation, Introduction and Commentary by C. Shields, Oxford: Clarendon Press. Black, D. L. (2010) “Intentionality in Medieval Arabic Philosophy,” Quaestio 10: 65–81. Brower, J. E. and Brower-​ Toland, S. (2008) “Aquinas on Mental Representation: Concepts and Intentionality,” Philosophical Review 117: 193–243. Burnyeat, M. F. (1992) “Is an Aristotelian Philosophy of Mind Still Credible? A Draft,” in M. C. Nussbaum and A. Oksenberg Rorty (eds.), Essays on Aristotle’s De Anima, Oxford: Clarendon Press. Burnyeat, M. F. (2001) “Aquinas on ‘Spiritual Change’ in Perception,” in D. Perler (ed.), Ancient and Medieval Theories of Intentionality, Leiden-​Boston-​Köln: Brill. Caston, V. (2000) “Aristotle’s Argument for Why the Understanding is Not Compounded with the Body,” Proceedings of the Boston Area Colloquium of Ancient Philosophy 16: 135–175. Caston, V. (2005) “The Spirit and The Letter: Aristotle on Perception,” in R. Salles (ed.), Metaphysics, Soul, and Ethics in Ancient Thought: Themes from the work of Richard Sorabji, Oxford: Clarendon Press. Cohen, S. M. (1982) “St. Thomas Aquinas on the Immaterial Reception of Sensible Forms,” The Philosophical Review 91, 2: 193‒209. Hamlyn, D. W. (1966) Sensation and Perception, London: Routledge and Kegan Paul. Kretzmann, N. (1993) “Philosophy of mind,” in N. Kretzmann and E. Stump (eds.), The Cambridge Companion to Aquinas, Cambridge: Cambridge University Press. Lindberg, D. (1983) Roger Bacon’s Philosophy of Nature, Oxford: Clarendon Press. MacDonald, S. (1993) “Theory of knowledge,” in N. Kretzmann and E. Stump (eds.), The Cambridge Companion to Aquinas, Cambridge: Cambridge University Press. Pasnau, R. (1997) Theories of Cognition in the Later Middle Ages, Cambridge: Cambridge University Press.

148

149

Thomas Aquinas on cognition as information Pasnau, R. (2012) “Philosophy of Mind and Human Nature,” in B. Davies and E. Stump (eds.), The Oxford Handbook of Aquinas, Oxford: Oxford University Press. Pickavé, M. (2012) “Human Knowledge,” in B. Davies and E. Stump (eds.), The Oxford Handbook of Aquinas, Oxford: Oxford University Press. Thomas Aquinas (1888). Summa Theologiae, Rome: Leonine (ST). Thomas Aquinas (1984). Sentencia libri De anima, R. Gauthier, Rome-​Paris: Leonine (InDA). Thomas Aquinas (1999) A Commentary on Aristotle’s De anima, R. Pasnau (transl.), New Haven-​ London: Yale University Press. Thomas Aquinas (2017), The Summa Theologiae of St. Thomas Aquinas, transl. by Fathers of the English Dominican Province, online edition at http://www.newadvent.org/summa. Trifogli, C. (2019) “Thomas Aquinas and Giles of Rome on the Reception of Forms without the Matter,” Vivarium 57: 244–​267. Trifogli, C. (forthcoming) “Giles of Rome on Sense Perception,” Quaestio.

149

150

151

PART III

Information 1500–​1800 Control

152

153

9 LEIBNIZ AS A PRECURSOR TO CHAITIN’S ALGORITHMIC INFORMATION THEORY 1 Richard T. W. Arthur

Introduction Gregory Chaitin (b. 1947) is one of the founders of Algorithmic Information Theory (AIT), a theory that characterizes information in terms of the shortest algorithm sufficient to generate it. He is well known not only for his own seminal contributions, such as the Ω-​number (also called “Chaitin’s constant” or the “halting probability”—​informally, a number expressing the probability that an arbitrary program will halt) and for Chaitin’s Incompleteness Theorem (on which more below), but also for his vigorous promotion of the theory and its philosophical significance. One striking feature of his popular presentations of AIT has been his attribution of most of its leading ideas to the seventeenth-​century polymath, Gottfried Leibniz (1646‒1716).2 Chaitin identifies three central features of AIT that can all be traced back to Leibniz: the idea of a computer program, a definition of complexity in terms of simplicity of laws (or programs) needed to generate a set of data, and the idea of irreducibility, where an irreducible theory or program is one that is no simpler than the data it generates. My purpose in this chapter will be to subject these claims of anticipation to a critical examination. I will argue that Leibniz anticipates some of Chaitin’s views even more than he recognizes, undermining some of the criticisms he makes of Leibniz’s metaphysics; but that where they disagree, Leibniz’s philosophy of science is in fact more insightful and less naïve than Chaitin’s. I then argue, on the basis of the work of John Collier, that the application of AIT to physics, rather than being fatal to Leibniz’s metaphysics, can be seen to deliver results consistent with Leibniz’s defence of contingency. I shall begin by giving the briefest of sketches of the main features of AIT that are relevant to this purpose, and then proceed to an evaluation of Chaitin’s claims.

The rudiments of AIT: Complexity, and irreducibility AIT was developed independently by Andrei Kolmogorov (1965 and 1968) and by Chaitin (1966), with a debt to the ideas of Ray Solomonoff (1960) about algorithmic probability. Since then, Chaitin has been instrumental in developing and publicizing the theory. He is known for his definition of what is now called Chaitin’s constant, the Ω-​number or halting probability,

153

154

Richard T. W. Arthur

and for his development of an incompleteness theorem for formal theories based on halting probabilities. Chaitin relates that at the age of 15 he had the idea—​anticipated by Leibniz in 1686—​of looking at the size of computer programs and of defining a random string of bits to be one for which there is no program for calculating it that is substantially smaller than it is. (2005, 22) The source for what he attributes here to Leibniz, he remembered later, was a passage in Hermann Weyl’s Philosophy and Natural Science, a “key observation about complexity and randomness” (Chaitin 2005, 58).3 The passage from Weyl is worth quoting at length. First, he points out that Laplacian determinism would reduce “the causal law” to a triviality, since the knowledge of the state of the world at any given moment (from which its future states could be predicted) would already embody more information than the theory predicts. Thus, he reasons, The assertion of regularity becomes meaningless if complications of arbitrary degree are admitted. This was emphasized already by Leibniz in his “Metaphysische Abhandlung” (Philosophische Schriften, IV, p. 431). What is decisive and at the same time astounding is that the laws show such a simple mathematical structure, while the quantitative distribution of the state quantities in the world continuum is incredibly complicated. This has the consequence, for our knowledge, that limited experience enables us to ascertain those laws while the unique quantitative course of events remains largely unknown. This distinction, for the naïve realist only the vague one between simple and complicated, becomes one of principle, when the intuitionist or constructivist view is adopted in mathematics and physics. (Weyl 1949, 191) Given the last remark, it is plausible that this passage from Weyl also contained the seed for Chaitin’s later championing of a constructivist or empirical approach to the philosophy of mathematics. But it is perhaps not immediately obvious what this has to do with the computer programs Chaitin alludes to in this connection. That depends on regarding a scientific theory as being analogous to a computer program in that both may be regarded as “compressions of information”. As Chaitin describes it, “A scientific theory is like a computer program that predicts our observations of the universe. A useful theory is a compression of the data; from a small number of laws and equations, whole universes of data can be computed” (2006, 79). We will come back to this conception of scientific theories as compressions of data further on below. Thinking of them this way enabled Chaitin to establish a link with the central ideas of computer science established by Alan Turing (1912‒54). In regular information theory, information is quantified in terms of the number of bits needed to encode the information. By contrast, in Algorithmic Information Theory, in Chaitin’s words, information “is defined by asking what size computer program is necessary to generate the data” (2006, 76).4 This then allows for a precise characterization of complexity: it can be measured by the length of its shortest description. More precisely, given a binary computer program, the complexity or algorithmic information content of a given set of data is the minimum number of bits that can store the program for producing the data. Chaitin provides some simple examples to illustrate this point. One might intuitively think that the number π—​since it is not only irrational but transcendental—​would be extremely 154

155

Leibniz and Algorithmic Information Theory

complex. But it is easy to write a computer program that generates its digits to any desired decimal place: a program for Leibniz’s series for π/​4, for example, π/​4 = 1 –​1/​3 + 1/​5 –​ 1/​7 + 1/​9 –​1/​11 + …, would suffice, although there may well be ways of programming π that are shorter still when expressed in binary code. Its precise degree of complexity, on Chaitin’s definition of complexity, is the length (in bits) of the shortest binary program that generates all its digits, that is, the length of the string of zeros and ones needed to store the program. By contrast, a number that is truly random—​Chaitin gives as an example a “random number with a mere million digits, say 1.341285 … 64” (Chaitin 2006, 76)—​would require a program that simply delivered each digit in its proper place, and would therefore require a string no shorter than the string representing the number. The data represented by such a number are, in Chaitin’s inimitable prose, “lawless, unstructured, patternless, not amenable to scientific study, incomprehensible. In a word, random, irreducible!” (Chaitin 2005, 54). Such strings are “incompressible” and “are called irreducible or algorithmically random” (2006, 77). What interests Chaitin is that these considerations can be applied to mathematical theories. If such theories are axiomatized, then the axioms can be regarded as compressions of the information contained in the theory. Here he builds on Leibniz’s repeated exhortations to try to prove the axioms of any theory: the point is to try to derive them from axioms that are simpler. If a mathematical statement is derived from a set of axioms that is not simpler than it, then that statement is irreducible. It is in this context that Chaitin conceives the real number Ω, the halting probability, which is defined as follows. It is an infinite sum, expressed in binary notation, towards which every program of length N bits that halts, contributes exactly (1/​2)N. That is, every program that halts and has length N will contribute a 1 in the Nth position of the binary expression of Ω. But Turing’s halting problem cannot be solved; whereas if Chaitin’s number Ω were calculable, it could. It is not that several digits of the number cannot be calculated—​this is perfectly possible. But, as Breger remarks, the first N digits cannot be calculated with a program that is markedly shorter than N bits. If a particular program for calculating the first N digits in the binary notation of Chaitin’s number is given, then an infinite number of digits will always remain that the program cannot calculate. (Breger 2012, 259) It follows that Ω is irreducible. It also follows from this that from a given set of axioms of a formal theory one can only compute as many places of Ω as the complexity of these axioms permits. In order to compute more places, one must feed more information into the system in the form of further independent axioms. Thus “No formal mathematical theory whose axioms have less than N bits of complexity can enable us to determine N bits of Ω” (Chaitin 2007, 244). This is analogous to the situation with Gödel’s first incompleteness theorem, according to which, in any consistent formal system F adequate to express elementary number theory, there are statements of the language of F which are neither provable nor disprovable in F. But such statements can be proved in more powerful systems. Consequently, Chaitin has promoted this fact about Ω as an “extremely strong incompleteness result” (Chaitin 2005, 132; cf. Breger 2012, 259). For Chaitin Ω is just one real number that is uncomputable. But he connects this result with the situation regarding real numbers in general. From the fact that the number of possible programs one can run on a Turing machine is denumerable, Turing had proved in 1936 that the number of calculable real numbers is also denumerable. But the number of real numbers, as Cantor had shown, is non-​denumerable. It follows from this discrepancy that the probability 155

156

Richard T. W. Arthur

that a randomly chosen real number is not calculable is 1, that is, that the probability that it can be calculated is 0. Likewise one can show that the probability that a randomly chosen real number can be characterized—​whether constructively or non-​constructively—​in such a way that it is distinguished from other reals is also 0 (Chaitin 2005, 113). “In summary,” Chaitin exclaims, “Why should I believe in a real number if I can’t calculate it, if I can’t prove what its bits are, and if I can’t even refer to it? And each of these things happens with probability one!” (Chaitin 2005, 115; cf. Breger 2012, 263). (We will return to this scepticism about real numbers below.) This is enough, I think, to give the general lay of the land of Chaitin and AIT. Now let us turn to Leibniz.

Leibniz’s anticipations According to Chaitin, Leibniz anticipated all the main ideas of AIT, although he never put them together to produce the theory: Despite living 250 years before the invention of the computer program, Leibniz came very close to the modern idea of algorithmic information. He had all the key elements. He just never connected them. He knew that everything can be represented with binary information, he built one of the first calculating machines, he appreciated the power of computation, and he discussed complexity and randomness. (Chaitin 2006, 77) Of course, Leibniz was not the first person to build a calculating machine. He was aware of other such machines, and he became acquainted with Blaise Pascal’s calculating machine from his writings when he gained access to them in Paris during his stay there between 1672 and 1676. Pascal’s machine, the “Pascaline”, was able to perform addition and subtraction. Leibniz undertook to improve upon it, and design a machine capable of multiplication and division as well. (One could perform multiplication using the “Pascaline” or “Napier’s Rods”, but only by repeated addition.) In this he was successful. Leibniz was able to present a prototype of the device to the Royal Society of London on a visit there in February 1673, although the machine, his Instrumentum Arithmeticum, did not work well due to imperfections of craftsmanship. Later versions of the machine enabled not only multiplication and division to be performed, but even the extraction of roots. His main innovation for the mature version of his machine (after 1674) was the design of a new feature, the so-​called stepped drum, a cylinder bearing nine teeth of different lengths which increase by equal amounts around the drum. This device enabled a digit in each decimal place to be calculated by moving a cog along the drum to the appropriate gear before multiplication of the digit was effected by the rotation of a handle (see Figure 9.1). Leibniz’s hopes for his machine were high. In a manuscript from 1685 he described an earlier version of the machine as “desirable to everyone engaged in computations who, as is well known, are the managers of financial affairs, the administrators of others’ estates, merchants, surveyors, geographers, navigators, astronomers, and [those connected with] any of the crafts that use mathematics” (Leibniz 1897]”.5 Much to his frustration, however, the inability of craftsmen to produce working models on the basis of his designs hampered the machine’s commercial viability, and only a few machines were ever finished, only one of which was in good working order at his death in 1716.6 But his invention of the stepped drum principle survived for 300 years and was used in many later calculating systems. 156

157

Leibniz and Algorithmic Information Theory

Figure 9.1  A replica of Leibniz’s calculating machine. Source: CC BY-​SA 4.0 Eremeev.

The theoretical significance of the machine for computer science lies in Leibniz’s views about the mathematical underpinning of reasoning; and the key link is provided by multiplication. Inspired by Ramon Lull’s device for generating triples of letters that would correspond to valid arguments, and also by Thomas Hobbes’s conception of reasoning as calculating, Leibniz sought to model reasoning on arithmetical operations. Although he had initially followed Hobbes’s construal of reasoning in terms of the addition and subtraction of concepts, Leibniz came to see multiplication as a better model for the composition of concepts. In his logical studies he devised a method for proving formal validity of arguments by assigning numbers to concepts. Every compound concept is conceived as consisting in “a plurality of other concepts, as many positive as negative”, he wrote in one study from 1679,7 and is accordingly assigned an ordered pair of positive and negative numbers. True propositions can then be expressed in terms of concept containment: a subject concept contains the predicate concept (making the proposition true) if each of the numbers corresponding to the predicate concept divide without remainder into the corresponding numbers in the subject concept. Using such numbers, Leibniz claimed, we can work out what other true propositions follow from them, and in this way, “demonstrate all the rules of logic with wonderful reason through numbers, and show how to recognize whether certain arguments are in valid form.”8 If the numbers standing for the concepts in a syllogism (such as “All A are B, No B are C, therefore No A are C”) can be chosen in such a way as to make both premises and the contradictory of the conclusion come out true, then the argument is invalid; while if such a set of numbers cannot be found, “the argument will conclude in virtue of form alone.” We do not need to know that a specific pair of numbers represents a given concept; all that is required is that the chosen numbers represent the relations between the concepts of the premises, and from this we will be able to tell whether the conclusion follows. Moreover, Leibniz argued, this is similar to how we reason generally. When we think of a complex concept “we cannot think of all the concepts which compose it at the same time”, he argued in his first published paper in 1684, Meditations on Knowledge, Truth and Ideas (A VI 4, 587; Leibniz 1969, 292). Yet we can still reason correctly if we identify the correct relations between such concepts. If we use characters to stand for complex concepts, then “provided we observe a certain order and rule in their use, they give us results which always agree with one another” (ibid.). This is what we generally do in reasoning: we do not intuit the whole nature of a thing at once, especially in a rather long analysis, but instead of things we make use of signs, whose explication we usually omit in any actually present thought for the sake of brevity, knowing or believing that we have the power to produce it. (ibid.) 157

158

Richard T. W. Arthur

Thus when I think about a chiliagon, that is, a polygon with a thousand equal sides, I do not always consider the nature of “side”, “equality” or thousand” (i.e. the cube of ten), but in my mind I use these words (whose meaning appears only obscurely and imperfectly to the mind) in place of the ideas I have of them, since I remember that I know the meaning of these words and judge that an explication of them is not necessary for the present. (ibid.) This is what Leibniz called “blind thought”: “we use it in algebra and in arithmetic, and indeed almost everywhere” (ibid.). Putting all this together, we see that Leibniz not only had the idea of a mathematical grounding for logic, but also held reasoning generally to proceed by a kind of computation using symbols, and had designed a machine capable of performing it. These are, as Chaitin observes, the same ideas that underlie computer programming and Alan Turing’s notion of artificial intelligence.9 Here, to perform complex calculations, we set up a program, which encodes the desired relations among concepts as functions of symbols and their relations. We allow the machine to work on the symbols directly, and we know how to interpret the output, without having to understand any of the intermediate steps that the machine performs. Similarly, given that Leibniz modelled reasoning on multiplication and division, and that multiplication and division were the essential functions that his calculating machine could perform, it is no exaggeration to see him as having been the first to envisage the idea of computer-​based reasoning. Leibniz showed his own recognition of the implications of his machine for artificial intelligence when he wrote in an undated manuscript, This machine … shows that the human mind can find the means of transplanting itself in such a way into inanimate matter that it gives to matter the power of doing more than it could have done by itself: to convince via the senses those who have difficulty conceiving how the Creator could house the appearance of a mind a little more generally in a body, however furnished with many organs, since even brass can receive the imitation of an operation of reason which concerns a particular or determinate truth. (Quoted from Jones 2018, 522) As Chaitin mentions, a third feature of modern computing anticipated by Leibniz is his discovery of binary arithmetic (“dyadics”) and recognition of its advantages for computing. There exists a manuscript in the Leibniz Archives in Hanover dated 15 March 1679 (OS), that discusses the possibility of designing a mechanical binary calculator using moving balls to represent binary digits.10 Although a machine of this kind was never built during his life—​it was built twice in the twentieth century—​Leibniz did publish his ideas on binary arithmetic in his (1705). This had been occasioned by his correspondence with a Jesuit missionary in China, Joachim Bouvet, in which he had described the principles of his binary arithmetic (including an analogy between the formation of all numbers from 0 and 1 with God’s creation of the world). Bouvet had recognized a strong analogy between Leibniz’s binary arithmetic and the hexagrams of the I Ching (see Figure 9.2), which convinced Leibniz that the ancient Chinese had anticipated his discovery, and prompted him to publish. Leibniz’s publication formed the stimulus for all modern interest in and application of binary arithmetic. It is also worth noting that in his work on dyadic, Leibniz explicitly remarked on the connection between binary and hexadecimal number systems, and the greater practical utility of the latter: “the binary progression is for theory, the hexadecimal for praxis”.11 As is well 158

159

Leibniz and Algorithmic Information Theory

Figure 9.2  A diagram of I Ching hexagrams, sent to Leibniz by Joachim Bouvet.

known today, because of the precise nature of the correspondence, long strings can be treated more succinctly in the hexadecimal system without compromising the exact reference to the digits, and Leibniz recognized this. So far, these anticipations of Leibniz’s relate to computing in general. The key respect in which Chaitin acknowledges Leibniz as anticipating AIT, however, comes with his views on complexity. As mentioned above, Chaitin had seen the reference to Leibniz in Weyl’s work when he was young, and this had stimulated his own ideas in the 1960s. When he followed up the reference in 2002, he found the passage Weyl had referred to in Leibniz’s Discours de métaphysique: Not only does nothing happen in the world that is absolutely irregular, but one cannot even imagine any such thing. Suppose, for example, someone were to make a number of dots on a piece of paper completely haphazardly, as do people who practise the ridiculous art of geomancy. I say that it is possible to find a geometrical line whose notion12 is constant and uniform according to a certain rule, in such a way that this line passes through all the points, and in the same order in which they were drawn. And if someone were to trace in one movement a line that is now straight, now circular, and now of another nature, it is possible to find a notion, a rule or an equation, common to all the points on the line, in virtue of which these very changes should occur. And there is, for example, no face whose contours are not part of a geometrical line, and which could not be drawn in a single line by some rule-​governed motion. But when a rule is very complex, what conforms to it passes for irregular. (Discours §6, A VI 4, 1537‒38; Leibniz 1969, 306) 159

160

Richard T. W. Arthur

Commenting on this, Chaitin writes: Leibniz’s very simple and profound idea appears in section VI of the Discours, in which he essentially states that a theory has to be simpler than the data it explains, otherwise it does not explain anything. The concept of a law becomes vacuous if arbitrarily high mathematical complexity is permitted, because then one can always construct a law no matter how random and patternless the data really are. Conversely, if the only law that describes some data is an extremely complicated one, then the data are actually lawless. (Chaitin 2006, 76) Chaitin’s last remark evidently would not have been accepted by Leibniz. For the very heading of Discours §6 is “God does nothing that is without order, and it is not even possible to imagine events which are utterly irregular (ne soyent point reguliers)”. And yet God knows the “law of the series” of states of every existing substance, and how they all harmonize. According to Leibniz, then, there can be no data that “are actually lawless”, even though they can “pass for irregular”. “Whatever passes for extraordinary”, he writes in the same section, “is so only in regard to some particular order established among created beings. For as regards the universal order, everything conforms to it” (A VI 4, 1537; Leibniz 1969, 306). Thus Leibniz distinguishes between the universal order implicit in God’s design of the universe, on the one hand, and the physical laws that created beings manage to formulate, on the other. It is the first that underpins the determinism necessary to God’s being able to foresee how the universe will unfold.13 But, as we shall see, this does not entail any mathematical demonstration on God’s part of each state from the preceding ones, and it does not therefore entail the logical necessity of any state in the created universe. Where Leibniz’s remarks gain the significance that Chaitin is attributing to them is in the context of Weyl’s mentioning of them, where the “universal order” is interpreted in Laplacian terms. Thus in the famous passage from his Théorie analytique des probabilities of 1812, Laplace had imagined an “intelligence” (now known as “Laplace’s Demon”) comprehending the whole universe in one formula: We ought, then, to envisage the present state of the universe as the effect of its earlier state and as the cause of the state that is to follow. An intelligence that could, for a given instant, know all the forces by which nature is animated and as well as the relative situations of the beings that compose it, would comprehend in the same formula the motions of the greatest bodies in the universe and those of the lightest atom, provided it were sufficiently vast to submit all the data to analysis; to it, nothing would be uncertain, and the future, like the past, would be present to its eyes. The human mind, in the perfection that it has been able to give to astronomy, affords a feeble outline of such an intelligence. (Pierre Simon de Laplace 1812, Introduction, pp. ii‒iii; my translation) Weyl’s point, as we saw, was that such Laplacian determinism would reduce “the causal law” to a triviality, since the knowledge of the state of the world at any given moment (from which its future states could be predicted) would already embody as much information as the theory predicts. Chaitin had inferred from Weyl’s description of this state of affairs that such a formula as conceived by Laplace would contain at least as much information as the data it was supposed to 160

161

Leibniz and Algorithmic Information Theory

explain, and consequently would be scientifically useless. When Chaitin coupled this with his idea of expressing the information contained in any such formula or algorithm by the shortest program able to produce it, he arrived at his key insights for AIT. But on applying this reasoning to Leibniz’s own metaphysics, he inferred that it undermines a central pillar of that philosophy. If Leibniz had assembled all the components of AIT that he had anticipated—​reasoning as computability, data as encodable in an algorithm, computable algorithms as representable in binary code, complexity, and irreducibility—​and had thereby arrived at AIT, Chaitin contends, he would have had to abandon the Principle of Sufficient Reason. This is Leibniz’s principle that “there can be no real or existing fact, no true proposition, without there being a sufficient reason why it should be so and not otherwise, though in most cases these reasons cannot be known to us” (Monadology, §31; GP vi 612). Writes Chaitin: He knew that everything can be represented with binary information, he built one of the first calculating machines, he appreciated the power of computation, and he discussed complexity and randomness. If Leibniz had put all this together, he might have questioned one of the key pillars of his philosophy, namely, the principle of sufficient reason—​that everything happens for a reason. (Chaitin 2006, 77) For what AIT makes manifestly clear is “that certain mathematical facts are true for no reason, a discovery that flies in the face of the principle of sufficient reason” (Chaitin 2006, 76‒77).14 Here Chaitin alludes to his own and Gödel’s incompleteness theorems. As we have seen, there are always mathematical facts expressible in axiomatized formal systems that cannot be proved within those systems, and no such formal systems in which all such facts are demonstrable. And there are numbers, such as Chaitin’s Ω, that cannot be completely calculated at all. The situation is equally pressing in physics, Chaitin urges. Leibniz holds that contingent events are true for a reason. For even if “there is in fact an infinite series of reasons, an infinite chain of cause and effect” that is “utterly beyond the power of human comprehension” because of its infinite complexity, this “is not at all beyond the power of comprehension of the divine mind” (Chaitin 2005, 103‒104). But for Chaitin infinite complexity is associated with randomness: the infinite complexity of any law by which God would be able to compute a proof that a given contingent proposition is true—​say, that such and such an event should take place at a specified place and time—​would entail that the event concerned is effectively random! Moreover, “at this point in time quantum mechanics demands real intrinsic randomness in the physical world, real unpredictability” (104). Even in classical, deterministic physics, chaos theory shows that a somewhat milder form of randomness is actually present if you believe in infinite precision real numbers and in the power of acute sensitivity to initial conditions to quickly amplify random bits in initial conditions into the macroscopic domain. (104) To summarize this section: there are good grounds for Chaitin’s claims that Leibniz foresaw the idea of a computer program, and even a computer operating on a binary system, that he sketched a definition of complexity in terms of simplicity of laws (or programs) needed to generate a set of data, and suggested the idea of irreducibility, where an irreducible law is one that is no simpler than the data it generates. It is, however, perhaps a little overblown to say that if he had put these ideas together he might have come up with AIT, since he lacked the idea of 161

162

Richard T. W. Arthur

a universal Turing machine in terms of which Kolmogorov complexity is defined, from which the main results of the theory derive.15 But Chaitin also alleges that the implications of AIT are such as to undermine Leibniz’s metaphysics. He charges that, on the one hand, certain mathematical facts are true for no reason, contrary to Leibniz’s principle; and that, on the other, the infinite complexity of contingent facts posited by Leibniz makes them random and thus not susceptible to sufficient reason. These charges deserve a full discussion; answering them will give us reason to question not only Chaitin’s understanding of Leibniz’s metaphysics, but also the relation of AIT to physical theory.

Determinism, contingency and computability First let us examine Chaitin’s claim that demonstrations of contingent truths, for Leibniz, are beyond human comprehension because of their infinite complexity, but not for this reason beyond the comprehension of God. The first thing to note is that if this were true, it would make contingency an epistemological matter for Leibniz, as it was for Spinoza. The latter had maintained that we call things contingent when we are insufficiently knowledgeable about all the causes involved in their production. But if all the requisites for a thing’s existence are in place, then that thing must occur. Therefore, Spinoza concluded, everything that happens, happens out of strict necessity: “there is absolutely nothing in things on account of which they can be called contingent” (Ethics IP33S). This is perhaps what Chaitin had in mind by charging that Leibniz, in the company of other rationalists, “rejected contingency”.16 In fact, however, it was Leibniz’s realization that his philosophy was in danger of falling into necessitarianism in just this way that stimulated him into formalizing his new theory of contingency. Something is necessary, Leibniz reasoned, if it is impossible for it not to occur. In order to avoid everything being necessary, therefore, there must be some things which are possible, but which nevertheless do not occur. According to the Aristotelian view dominant in Leibniz’s time, possible events were understood as those which will occur at some time or other.17 Leibniz’s innovation was to make the notion of possibility a purely logical one: something is possible if its concept does not involve a contradiction, and an event is logically necessary if its not occurring involves a contradiction.18 But in order for an event to occur in a given world, it must be compatible with everything else simultaneously happening in that world; the thing it happens to must, to use Leibniz’s term, be “compossible” with all the other things in that world. Such events are contingent events; although they are possible, they are not logically necessary. There is no logical contradiction in their not occurring. It is not just that we humans could not find a contradiction in their not occurring: there is no such contradiction to be found. By the same token, this means not just that we are incapable of reducing a contingent truth to identicals, but that it cannot be done. As Leibniz wrote in 1689 in his De Libertate, Contingentia et Serie Causarum, Providentia, But in contingent truths, even though the predicate is in the subject, this can never be demonstrated of it, nor can the proposition ever be reduced to an equation or identity. Instead the resolution proceeds to infinity, God alone seeing—​not indeed the end of the resolution, since there isn’t one—​but the connection of terms or the enfolding of the predicate in the subject, since he sees whatever is in the series; indeed, this truth itself has arisen partly through his intellect and partly through his will; and expresses in its own way his infinite perfection, and the harmony of the whole series of things. (A VI 4, 1656) 162

163

Leibniz and Algorithmic Information Theory

God, according to Leibniz, sees contingent truths “not by demonstration”, but by a kind of a priori cognition through the reasons for truths. For he sees things as deriving from himself, possible things by a consideration of their own nature, existing things by the added consideration of his own free will and decrees, the first of which is to do everything in the best way, and with supreme reason. (A VI 4, 1658) From his knowledge of the “law” governing the series of states of each individual substance, God can discern immediately all the predicates that apply, in virtue of the fact that each such state reflects the states of all the other (compossible) substances existing in the same world at the same time. This is the case for the states of substances in any possible world. But it is the best selection of compossible substances that constitutes the world he decides to create, the actual world. The laws governing these substances are not necessary laws, in that other laws were possible for God to choose. The laws are therefore contingent; and they are actualized because they characterize the best of all possible complexes of possible substances. By way of illustration of this theory of contingent truths as ones that cannot be demonstrated, Leibniz presents the analogy with incommensurables in mathematics. It is not a question of reaching a common measure at infinity: there is no common measure to reach.19 Analogously with the case of contingent truths: it is not just that there is an infinity of steps in demonstrating such a truth; rather, there is no last step, and therefore no demonstration to be had. Now, one may think this rather odd. Leibniz, after all, knew how to compute the sums of infinite series. In fact, it was his work in this area that led him to the insights grounding his discovery of the differential and integral calculus.20 He knew, for example, how to calculate the sum of the infinite series 1 + ½ + ¼ + … (½)n-​1 + … as 2, and (as already noted) he knew how to compute π/​4 as the sum of the infinite series 1 –​1/​3 + 1/​5 –​1/​7 + 1/​9 –​1/​11 + … Surely, then, God would be able to compute the latter and arrive at the real number representing the incommensurable ratio π:4? This presumes that an exact value is reached after an infinite number of terms have been added. Leibniz, however, rejected such a conception of summing infinite series. For him, to say that the series 1 + ½ + ¼ + … (½)n-​1 + … sums to 2 is simply to say that one may obtain a number arbitrarily close to 2 by taking sufficiently many terms in any finite series with the same law and first term. As he wrote in a fragment (“Infinite Numbers”) from the beginning of April 1676: Whenever it is said that a certain infinite series of numbers has a sum, I am of the opinion that all that is being said is that any finite series with the same rule has a sum, and that the error always diminishes as the series increases, so that it becomes as small as we would like. For numbers do not in themselves go absolutely to infinity, since then there would be a greatest number. (A VI 3, 503/​LoC 98-​99) Leibniz rejects the notion that an infinite series has a determinate, infinite number of terms, that it forms an infinite whole. According to his conception, the infinite is to be understood syncategorematically, that is, in the sense that no matter how many terms are taken, there are more. But he denies that there is an infinity of terms in the sense of an infinite collection or set of them—​that kind of infinite he terms a categorematic infinite.21 Thus an infinite series is not literally an infinite sum, where the terms are added one by one to infinity. Rather the sum of 163

164

Richard T. W. Arthur

a convergent series is equal to a given number in the sense that it can be made arbitrarily close to that number by taking sufficiently many terms in any finite series with the same law. This is very close to the modern definition, where the sum of a convergent infinite series is defined as the limit of its partial sums, where “limit” is given the customary ε–​δ definition. As Herbert Breger has observed, this account of the infinite as syncatgorematic means that Leibniz is in fact closer to Chaitin’s own skepticism about real numbers than Chaitin seems to have realized (Breger 2012, 262‒263). For Leibniz the continuum is ideal: the points in it denote possible points of division, or “cuts”. In fact, in some of his formulations of mathematical continuity, Leibniz came close to the idea of a Dedekind cut.22 But the really important difference is that the “cut” for Leibniz would be a point in a geometrical line where it could be divided—​it connotes a possible division of the line into two bounded line segments. But no line can be divided at all possible such points: a line consisting of all possible divisions is something purely ideal. Thus just as a continuous line is not composed of points, so there is no corresponding set of all real numbers, in agreement with Chaitin. In fact, for Leibniz there is not even a set of all natural numbers, since he denies infinite collections. This entails that Cantor’s distinction between a denumerable and a non-​denumerable infinite (on which Chaitin based his own arguments about the uncomputability of reals) cannot even get off the ground.23 But what are the implications of this for the applicability of the Principle of Sufficient Reason in mathematics? Don’t these considerations imply that Chaitin is correct in seeing “limits of reason” in mathematics? We have seen that Chaitin gives his Ω as an example of a number whose digits are computable up to any finite number of places, but which is not itself computable, because irreducible. His and Gödel’s incompleteness theorems show that there are mathematical facts which, although expressible in some axiomatized formal system, are not demonstrable in that system. But we have also seen that it is a mistake to claim that Leibniz thought all mathematical facts are demonstrable. According to Leibniz, not even God could compute the real number representing the incommensurable ratio π:4 by a demonstration: a computer program can easily be set up to calculate the ratio to any desired degree of accuracy, but there is no program that can calculate all the digits of π. Likewise, there is a sufficient reason for any and all contingent truths, but this precisely does not mean that they are demonstrable. The reasons for such facts are known to God through his “intuition”, but although we certainly know contingent truths, the ultimate reasons for them are not “computable”.24 Regarding contingent events, we have seen that Leibniz distinguished his position from Spinoza’s, where every event is determined as a deductive consequence of preceding ones. The requisites of any existing event, Leibniz came to see, depend on the circumstances of time and place, unlike mathematical conditions for mathematical theorems.25 “Requisites”, on his understanding, are the necessary conditions for an event to take place, and the sum total of such requisites is the sufficient condition for it to take place. But given the infinite complexity of events occurring in any spatial or temporal interval, this makes for an infinite complexity of the requisites of any given event, and accounts for its lack of demonstrability (one could say, computability) from prior events.26 That is to say, predicting an event it is not just a question of a finite set of conditions and a physical law, expressed mathematically: one needs to know the initial conditions themselves, and for Leibniz these are infinitely complex. Laws not only abstract away from all this complexity to concentrate on the most relevant; they are also hypothetical. Given such and such a set of conditions, then the law prescribes that such and such an outcome is certain. But this does not make it necessary. For it is not only the initial conditions that are contingent; the laws themselves are also contingent on God’s having selected those that produce a world with the greatest possible perfection. 164

165

Leibniz and Algorithmic Information Theory

For Leibniz these considerations fit in with a philosophy of science in which hypotheses are created in order to give intelligible explanations of the phenomena. The hypothesis that gives the most intelligible explanation of the phenomena in question is taken (defeasibly) as true. The phenomena in question are not merely “data”. For example, as Leibniz writes in an early piece from about 1676, the Copernican hypothesis is to be preferred because with the earth transferred into the place previously ascribed to the sun, many imaginary circles, many eccentric circles, and many anomalies vanish. But if also the apparent diameters of the fixed stars change, or different parts of the heavens are vertically above the same point of the earth at different times, and this agrees with the hypothesis of the annual motion of the earth, according to which the earth changes its situation to the fixed stars, this hypothesis would certainly seem to be sufficiently corroborated. (A VI 3, 105; Arthur 2013, 101) This is already a far cry from the (Machian) conception Chaitin seems to have of physical theories as experimental laws, conceived as compressions of experimental data, data which they “explain” simply by predicting them: The way I would put it is like this: I think of a scientific theory as a binary computer program for calculating the observations, which are also written in binary. And you have a law of nature if there is compression, if the experimental data is compressed into a computer program that has a smaller number of bits than are in the data that it explains. The greater the degree of compression, the better the law, the more you understand the data. (Chaitin 2005, 53) Thus on the one hand Chaitin presupposes a formal, axiomatized theory. But on the other, he equates “theory” with what Hempel and the logical empiricists would have called an experimental law; he also presupposes a model of explanation as simply deduction of the phenomena to be explained from the law (together with, presumably, initial conditions)—​the deductive-​ nomological model of explanation. This is not the place to lecture on the philosophy of science, but it should not need stressing that the field has come a long way since this logical empiricist philosophy of science held sway. Most relevant are the following points: predictions are a way of testing the theory, they are not its content. They can only be made with auxiliary assumptions and hypotheses, initial conditions, and so forth. Moreover, a theory itself is not simply a law, but a system of mathematical statements together with all kinds of idealizing assumptions, dependent on subsidiary hypotheses, and so on, and usually part of a research programme that provides metaphysical foundations, implicit norms, and so on. What Chaitin is proposing here sounds more like the curve fitting that might be appropriate to some particular law expressed in functional terms. The upshot of all this is that we need a more nuanced account of how information is involved in physics than Chaitin’s notion that physical theories or laws are simply compressions of data. An interesting proposal for making just such a connection between AIT and physical dynamics has been made by John Collier. Its pertinence to this discussion is that he attributes the fundamental insight underlying his proposal to Leibniz. 165

166

Richard T. W. Arthur

Physical laws and information Collier proposes that causation be understood as the transfer of information: more specifically, as “the propagation of form, as measured by information theoretic methods” (Collier 1999, 2). He sees a precedent for this in Leibniz’s account of causation, interpreting information as Leibnizian “form”, and form as constituted by the arrangement of the parts of a body.27 In justification, he quotes a passage from Leibniz from a draft for a book on physics from 1680: And the operation of a body cannot be understood adequately unless we know what its parts contribute; hence we cannot hope for the explanation of any corporeal phenomenon without taking up the arrangement of its parts. [Leibniz, 1969, 289]. (Collier 1999, 4) Collier then explicates the concept of “the arrangement of a body’s parts” through statistical mechanics. Here, like many others, he is following the lead of Erwin Schrödinger, who in his seminal book What is Life? (first published in 1944), characterized living organisms as sustaining themselves in existence through feeding on “negative entropy”, that is, “by extracting ‘order’ from the environment” (Schrödinger 1967, 72‒80). Let me briefly delineate some of the main features of Collier’s proposal before returning to the question of how it might have been prefigured in Leibniz’s thought. The first point to note is that “information” is hardly a univocal concept. Some, such as Edwin Jaynes, have taken information to be a property of an epistemic state (Jaynes 1979); others, like Bar-​Hillel (1964), and Barwise and Perry (1983), have taken it to be a kind of semantic property of statements (Collier 2010, 3‒4). As for entropy, on the other hand, Collier points out that “the statistical notion of entropy has allowed the extension of the idea in a number of directions, directions that do not always sit happily with each other” (1999, 12–​13). This is particularly the case with the notion of entropy used in mathematical communications theory (Shannon and Weaver, 1949), where it is identified with information. Nevertheless, these divergent conceptions of entropy and its relation to information are related, Collier argues, in the fact that underlying them there must be some physical basis for information. “Whatever allows meaningful information at the cognitive level is constructed from physical resources. Physical reality constrains the way cognitive systems can work, including how they can process and interpret information” (Collier 1990, 2). He therefore insists that “what is transmitted is information, irrespective of whether there is a cognitive receiver” (Collier 1990, 2).28 Accordingly, he adopts Schrödinger’s interpretive heuristic, the Negentropy Principle of Information (NPI), according to which the information in a specific state of a physical system is a measure of the capacity of the system in that state to do work (Schrödinger, 1944; Brillouin, 1962: 153), where work is defined as the application of a force in a specific direction, through a specific distance.17 Work capacity is the ability to control a physical process, and is thus closely related to causality. (Collier 1999, 13) Schrödinger’s idea, as further developed by Brillouin, is to express physical information, IP, as the difference between two quantities: HMAX, “the maximal possible entropy of the system (its entropy after all constraints internal to the system have been removed and the system has fully relaxed, i.e. has gone to equilibrium)”; and HACT, the actual entropy. As such, the NPI is 166

167

Leibniz and Algorithmic Information Theory



NPI: IP = HMAX –​HACT where the environment of the system and the set of external constraints on the system are presumed to be constant. (Collier 1999, 13)

Collier argues, with a nod to Atkins (1994), that the actual entropy, HACT, “is a specific physical value that can in principle be measured directly”, and that HMAX, likewise is unique “since it is a fundamental theorem of thermodynamics that the order of removal of constraints does not affect the value of the state variables at equilibrium (Kestin, 1968)” (Collier 1999, 13). Physical information is thus a well-​defined physical quantity. In Collier’s words, it is “a unique and dynamically fundamental measure of the amount of form, order or regularity in a state of physical system” (Collier 1999, 13‒14). It has a non-​zero value for a given system only when that system is not in equilibrium with its environment, giving a natural measure of the orderliness or form of the system.29 How then does this connect with AIT? Collier quotes results from Kolmogorov (1965, 1968) and from Li and Vitánvi (1993) to show that the measure of the complexity information contained in IP is identical to HMAX –​HACT, and therefore to IP itself. “This is so,” he explains, “since the physical information of a system determines its regularity and this regularity can be neither more nor less informationally complex than is required to specify the regularity.”30 The resulting identities “allow us to use the resources of algorithmic complexity theory to discuss physical information, in particular to apply computation theory to the regularities of physical systems” (Collier 1999, 18‒19). More precisely, Collier represents informational complexity by subtracting the machine-​dependent measure of the computational overhead required to run the program from the algorithmic complexity,31 yielding “a machine-​independent measure that is directly numerically comparable to Shannon information, permitting identification of algorithmic complexity and combinatorial and probabilistic measures of information” (Collier 1999, 7). Causation is now interpreted in terms of the transfer of information. A key part of Collier’s conception is that this must be some particular information that is transferred from one system to another at a later time. A causal process is then defined as follows: P is a causal process in system S from time t0 to t1 iff some part of the information of S involved in stages of P is transferred from t0 to t1. (Collier 1999, 10) Then it follows that, since the informational relations are computational, the information transferred must be in the effect as well as in the cause, so that “the existence of the identical information (token) in both the cause and effect is both a necessary and a sufficient condition for causation” (Collier 1999, 11). This implies that the effect is necessitated by the cause, but only in the sense that it is a deductive consequence given the specific information contained in the initial state, the conditions for which are contingent: We can think of a causal process as a computation (though perhaps not a Turing computation or equivalent) in which the information in the initial state determines information in the final state. The effect, inasmuch as it is determined, is necessitated by the cause, and the cause must contain the determined information in the effect. 167

168

Richard T. W. Arthur

Although the causal relation is necessary, its conditions are contingent, so it is necessary only in the sense that given the relata it cannot be false that it holds, not that it must hold… (Collier 1999, 11) The connection with dynamics is provided by the stipulation that “the information in a specific state of a physical system is a measure of the capacity of the system in that state to do work” (Collier 1999, 13). As Collier explains, “HMAX represents a possible state of the system in which there is no internal structure except for random fluctuations” (Collier 1999, 15). In such a state the system cannot do any work internally; it is uniform except for random fluctuations which cannot be harnessed to do any work. “The actual entropy, however, except for systems in equilibrium, permits internal work, since there is energy available in the non-​uniformities that can be used to guide other energy” (15). HACT therefore represents the non-​random component of the system, the form or capacity to do work. This capacity itself is resolvable into two components, “one that is not constrained by the cohesion in the system, and one that is” (16). The former, called the intropy, ι, measures the available energy to do work, while the latter, which he calls enformation, ε, “measures the structural constraints internal to the system that can guide energy to do work” (Collier 1999, 16; 1990, 7–​10). I cannot do justice to all Collier’s innovative ideas here, but this gives us the gist of his theory, and of how AIT can be applied in physics. Turning now to the case for Leibniz as precedent for these ideas, there are a number of very interesting points to make. First, regarding Schrödinger’s formulation of the NPI in the context of accounting for the physical principles underlying life, there is a delicious irony (given the low esteem in which Schrödinger held Leibniz’s philosophy32) in the similarities between his descriptions of the difference between non-​living and living matter, between “clockwork” and “organism”, and those penned by Leibniz some 260 years earlier.33 Indeed, there is a strong sense in which Leibniz was the first to see the importance of information in connection with living organisms, and indeed as an essential feature of physical reality. The substances he posited as constituting the universe, his monads, were living beings each characterized by its own “substantial form” or “law of the series” of its states, where the law is interpreted as encoding all its future states, thus embodying all the information pertaining to the future phenotypical development of its own organic body—​including such marked transitions as those characterizing the development of a caterpillar through the various stages to its emergence as a butterfly. Each such organism is conceived by him as an “organic machine” which, in contrast to artefactual or man-​built machines, possesses (1) the ability to repair itself and draw in energy by taking in nutrition from its environment, and to excrete waste products, (2) the power of self-​motion, moving itself towards nutriments and away from danger, (3) the ability to perceive, that is to be able to represent to itself its surroundings, especially sources of nutrition and danger: Nature brings it about that her Machine is able to do these very things by itself, that is, that it is now nourished, whereby its worn out parts and forces are restored, that it now moves itself towards the nutriments or other means of sustaining its functions that are to be procured, and away from impediments that are to be avoided, and now finally, that it is warned by internal or external signs, and incited to this motion when needed. (Leibniz 1680‒82; Pasini 1996, 219)34 168

169

Leibniz and Algorithmic Information Theory

The law of each individual substance is accompanied by a tendency to take it into subsequent states, and this is identified by Leibniz with a “primitive force”, from which derive all the derivative forces of physics, such as energy (“living force”) elasticity, cohesion, and centrifugal force.35 Leibniz equates a substance’s states with its perceptions, taken in a very wide sense as representations. This idea of a state as a representation is crucial to the link with information: each state of a monad represents the whole of its environment, and its more immediate environment (especially its organic body) with the greatest distinctness. There is no connotation of imagery in this idea of representation: it is general enough to include the idea of an equation representing a curve, or a code representing a message. The key is an isomorphism between representation and what is represented, such that the representation contains all the information of what it represents. The particular information necessary for the constitution of its organic body from one time to another is preserved in the substance through an isomorphism with the information encoded in the law of the series, an idea that fits well with Collier’s characterization. As he writes, “[Leibniz’s] dynamics was explained by the properties of monads, whose substantial form implied a primitive active force (encoded by the logico-​ mathematical structure of the form in a way similar to the compressed form of causal properties I will discuss later)” (Collier 1999, 4).36 No less important, however (and not explicitly noted by Collier), is the anticipation by Leibniz of his views connecting cause with work capacity, the capacity of a given physical system to act on another. Leibniz was enamoured of principles, and the founding principle of his dynamics, as he stressed on numerous occasions, was his Principle of Equipollence, namely that “the entire effect is always equal in power to the full cause” (A VIII 2, 133). According to this principle, which he first articulated in 1676, Any full effect, if the opportunity presents itself, can perfectly reproduce its cause, that is, it has forces enough to bring itself back into the same state it was in previously, or into an equivalent state. In order to be able to estimate equivalent things, it is therefore useful that a measure be assumed, such as the force necessary to raise some heavy thing to some height. … Hence it happens that a stone that falls from a certain height, can, if it is constrained by a pendulum, and if nothing interferes and it acts perfectly, climb back to the same height; but no higher, and if none of the force is removed, no lower either. (A VIII 2, 136) As has been documented in detail by Michel Fichant (Leibniz 1994), in 1678 Leibniz used reasoning explicitly based on this principle to establish (and then confirm experimentally) that the measure of the “force” described here is proportional to mv2, his famous vis viva, and the original of our concept of energy. Ignoring resistance, the effect produced in the pendulum bob at the bottom of its swing by falling through a certain height is proportional to mv2, and this “living force” is a measure of the power to raise the bob again to the same height. At that height it will have an equivalent quantity of “dead force”, a forerunner of what we now call the “potential energy”. Leibniz’s “force” is thus the capacity to do work, and just like our modern concept of mechanical work, has the dimensions of energy. Moreover, Leibniz argued that this quantity is conserved in any isolated system, and that any apparent loss of energy, such as friction, may be accounted for by increases in the internal motions of the body concerned. He also held that a violation of this energy conservation would result in a perpetual mechanical motion, which is impossible. 169

170

Richard T. W. Arthur

Here Collier’s idea of a quantitatively equivalent work capacity being transmitted from cause to effect is clearly prefigured. Leibniz’s Principle of Equipollence, moreover, is (like the NPI) a heuristic principle: one has to work out how it is to be articulated in given scenarios. It presupposes that causes can be identified in given situations; but the identification of causes must be made by other means, since work capacity by itself does not involve directedness. As Collier writes in a footnote, “Work has dimensions of energy in standard mechanics, and thus has no direction. However, since it is the result of a force applied through a distance, it must be directed. Surely, undirected force is useless” (Collier 1999, 13). I do not want to overstate the case for Leibniz’s anticipation of these modern ideas based on statistical mechanics. Of course, he did not have the concepts of entropy or negentropy. Nevertheless, he was well aware of the need for organisms to keep extracting useful energy from the environment by absorbing nutrients and excreting waste products, thereby sustaining themselves in existence. Where his views differ most from modern conceptions, though, are in his attachment to preformation in biology, to the scale invariance of matter that is further divided to infinity, and to the infinitude of the information contained in any state. These issues are connected. For Leibniz’s commitment to preformationism was caused by his not being able see how perception and sentience could emerge from inert matter; so he posited them as essential to and persisting in substances, each of which has existed in various phenotypic forms since the first days of the Biblical Creation, with its various phenotypes encoded in the law of the series of perceptions. A perception, however, involves the representation of everything else in the universe at the same time, and therefore must be infinite (assuming the infinitude of the universe). Given the isomorphism between the information contained in each state and the physical information contained in a body, this was consistent with Leibniz’s conception of an organic body as divided in such a way that each of its parts either is or contains an organized body or organic machine, with this division repeated recursively to infinity: only an infinitely complex body could physically manifest infinite information. This fractal division is the above-​mentioned scale invariance: “c’est tousjours et partout en toutes choses tout comme icy”, writes Leibniz, echoing Harlequin, Emperor of the Moon (GP III 343). A finitely divided body, he insisted, could not contain the infinite quantity of information that it would have to manifest. We now know that there is no such scale invariance, and that at the level of atoms and elementary particles it no longer makes sense to think of them as spatially divided, and certainly not into organic bodies capable of manifesting biological organization and functioning. But here there is a deeper consideration, which David Layzer (1990) has described very eloquently. This is that Boltzmann’s reduction of thermodynamics to statistical mechanics (and thus the route to Schrödinger’s NPI) depends crucially on the assumption of discreteness. For, in order to count how many indistinguishable microstates there could be in a region (macrocell) of 6-​dimensional state space (each point of which corresponds to one position-​ velocity microstate), Boltzmann had to assume microcells of a finite discrete size, each small enough to contain one position-​velocity microstate. This is an assumption contrary to the continuity of the state space as presupposed by classical mechanics, and which only received a theoretical explanation and confirmation with Einstein’s realistic interpretation of Planck’s postulate.37 Thus quantum theory, as Schrödinger was at pains to explain in What is Life? is at the heart of the understanding of order through statistical mechanics. So although Leibniz might have anticipated Schrödinger in explaining the physical basis for life in terms of information—​ specifically, through the organic body’s having a physical constitution that is sufficiently complex to contain the blueprint for the organism’s further development—​the concept of information involved in the NPI is finite, not infinite. 170

171

Leibniz and Algorithmic Information Theory

Don’t these considerations then undermine Leibniz’s account of contingency, and validate Chaitin’s claims that AIT is incompatible with the Principle of Sufficient Reason? To answer these charges it is necessary to refer back to the distinction between laws and initial conditions. Leibniz’s assumption of the infinitude of information in each state underwrote his distinction of contingency from logical necessity. He granted that if we abstract away from this complexity to isolate a finite set of conditions sufficient for the application of a law to derive predictions, then the outcome prescribed by such a law is certain (assuming the law is correct). The outcome will not be logically necessary, however, both because of the contingency of the law, and because of the complexity of the initial conditions. But in a world in which there is an objective chance element that is not reducible to human ignorance—​and such is our world, according to quantum theory—​one may maintain that the initial conditions assumed for the application of physical laws to any causal process are contingent precisely because of this chance element.38 Thus while on the one hand such objective indeterminacy violates Leibniz’s Principle of Sufficient Reason—​there being no reason, for example, why a radioactive event occurs exactly when it does—​by the same token they underpin a notion of the objective contingency of initial conditions that supports his distinction between a deterministic contingency and logical necessity. As we have seen, on the basis of his analysis of how AIT applies in physics, John Collier makes the same distinction as Leibniz does between its being certain that something will follow and its being necessary that the thing will follow: “Although the causal relation is necessary, its conditions are contingent, so it is necessary only in the sense that given the relata it cannot be false that it holds, not that it must hold” (Collier 1999, 11). As he explains in a footnote, Causal connection is necessary in the same way that computation or deduction is necessary, but it is not necessary in the sense that it is impossible for things to be otherwise. The necessity depends on contingent conditions analogous to the premises of a valid argument … Something that is necessary in this way cannot be false or otherwise than it is, but is contingently true; i.e. it is contingent that it is. (Collier 1999, 2, fn. 2) Evidently, this is in perfect accord with Leibniz’s claims.

Conclusion We have seen that there are good grounds for upholding Gregory Chaitin’s claims for Leibniz’s anticipations of some of his ideas. A discussion of his criticisms of Leibniz’s Principle of Sufficient reason in mathematics led us to a consideration of Leibniz’s views on the infinite, where we found more common ground with Chaitin’s scepticism about real numbers than he had recognized. Regarding Leibniz’s theory of contingency, however, we found that Chaitin had seriously misconstrued Leibniz’s thought. This had its origins in Chaitin’s naïve conception of physical theories as compressions of empirical data—​the very idea that he had inferred from the passage in Leibniz’s Discours, which, through the mediation of Weyl, had inspired his thoughts on AIT in the first instance. A further investigation of Leibniz’s views on contingency revealed the importance of information being encoded not just in physical laws, but in initial conditions. For a detailed account of how AIT could be applied to dynamics we turned to the work of John Collier, based on Schrödinger’s Negentropy Principle of Information. Again, we found anticipations in Leibniz’s work of Schrödinger’s way of connecting information with the 171

172

Richard T. W. Arthur

foundations of life, and more specifically of Collier’s theory of causality as the transfer of information, with the latter interpreted through AIT. Finally, it should be noted that Chaitin is among those who read Leibniz as an idealist who wished to reduce the world to information alone, as the forerunner of the philosophy of “it from bit”, to use John Archibald Wheeler’s formula. Thus in a recent lecture Chaitin characterized Leibniz as the founder of digital philosophy: “This is called digital philosophy or digital physics, but it’s really the idea that maybe the universe is built out of information and computation, and not matter and energy … Leibniz is the first person that very clearly belongs in this new school” (Chaitin 2016).39 For Leibniz, however—​ notwithstanding his views about matter being constituted by monads—​the idea of disembodied information would have made no sense. As he responds to Bayle in 1698, “thought involves an actual external material object which contains a very large number of modifications, through which it is connected with surrounding bodies and, by means of them, step by step with all others” (GP IV 545). Consequently, the order represented in a monad’s perception must correspond to the order in the external world that it is representing.40 There must be a physical basis for information in the sense promoted by Layzer and Collier, underlying the order an intelligent being can perceive; it must be embodied in a particular arrangement of matter, and energy (in the form of work capacity) is involved in its transfer, according to Collier’s Leibnizian conception of causation.

Notes 1 This paper is dedicated to John Collier (1950‒2018), from whose fertile mind I hope some information has been transferred here. 2 ​So much so that, as Herbert Breger notes, someone skimming through Chaitin’s Scientific American article (2006) about his number Ω could get the impression “that one might find all this in Leibniz’s works” (Breger 2012, 261). 3 In his essay “Leibniz, information, math & physics”, Chaitin quotes Weyl’s remark from his (1932, 40‒2): “That the notion of law becomes empty when an arbitrary complication is permitted was already pointed out by Leibniz in his Metaphysical Treatise [Discourse on Metaphysics]” (Chaitin 2007, 228‒9). He explains that he although he “actually read Weyl [1949] as a teenager, before inventing AIT at age 15 … the matter is not stated so sharply there” as in the 1932 text, which he had afterwards “stumbled on”, but only followed up when “stimulated to do so by an invitation from the German Philosophy Association to talk at their 2002 annual congress, that happened to be on limits and how to transcend them” (Chaitin 2007, 229). (Chaitin 2007) is a collection of his papers to that date; I am indebted to David Wright for equipping me with this and (Calude 2007), a Festschrift volume on Chaitin’s work. 4 As Chaitin notes (2006, 79), this would depend on the computing language used, and so that would have to be specified in a rigorous definition, which he gives elsewhere. 5 (Leibniz, 1897, 307; English translation by Mark Kormes in (Smith, 1929, 180)). The original text can be found in the 1685 Machina arithmetica manuscript [LH XLII, 5, GWLB Hannover]. 6 See Matthew Jones’s article, “Calculating Machine” (Jones 2018), for a thorough scholarly assessment of Leibniz’s machine and its significance. Klaus Badur, an engineer who built the machine some years ago, worked out the trick for getting it to work correctly (Badur and Rottstedt 2004), as Herbert Breger has kindly informed me. 7 “A Calculus of Consequences” (A VI 4, 226). See (Arthur 2014, chs. 2 and 4) for discussion of these points. 8 “On characteristic numbers for establishing a universal language” [Spring‒Summer 1679?] (A VI 4, 270). 9 An accessible entrée to Turing’s thought is his article, “Computing Machinery and Intelligence” (Turing 1950). 10 The manuscript is De Progressione Dyadica [On the Dyadic Progression], (Leibniz 1679). For a treatment of Leibniz’s dyadics, see Eberhard Knobloch (2018, 242‒4). He cites a second early manuscript,

172

173

Leibniz and Algorithmic Information Theory Summum calculi analytici fastigium per calculum algorithmicum [A treatise on analytical calculation by means of algorithmic calculation], from December 1679. 11 This quotation is from one of Leibniz’s unpublished manuscripts, LH 35 III B 17 Bl. 4. I am indebted to Mattia Brancato for this observation, and for the reference and a link to the manuscript. Mattia is writing a book on dyadics, which will include an edited version of this piece. 12 In his enthusiasm for Leibniz, Chaitin went so far as to challenge the accuracy of the Akademie editors’ reading of the manuscript, suggesting that Leibniz had written motion, not notion: see (Chaitin 2007, 230). But see (Breger 2012) for a refutation of this suggestion. (The later occurrence of “motion” in this passage translates mouvement.) Breger also corrects Chaitin’s interpretation of Leibniz on transcendentals, but I will not repeat that here. 13 I discuss this distinction between the cosmic determinism grounding God’s prevision, on the one hand, and the determinism specific to prediction by scientific laws in (Arthur 2019b), where I compare Leibniz’s views on determinism and causality to those of Mario Bunge. I also argue that Leibniz (like Bunge) is perfectly correct to take determinism as necessary for free will. This directly contradicts a commonly held view, argued for in this context by F. Walter Meyerstein: “Note that admitting a free-​will, if it is really free, is equivalent to introducing randomness into the world, as freedom of decision implies previous indetermination of the choices” (Meyerstein 2007, 283). Meyerstein claims that Leibniz was committed to “a rigidly pre-​determined fate” (2007, 284)—​an absolute travesty of his clear and subtle refutation of the Fatum Mahometanum in the Theodicy, as Meyerstein might have discovered for himself, had he read it. See Leibniz (1985, 53ff.) and Arthur (2014, 174‒8). 14 Cf. (Chaitin 2007, 234): “So the bits of Ω are irreducible mathematical facts, they are mathematical facts that contradict Leibniz’s principle of sufficient reason by being true for no reason.” 15 I am indebted to Jose Hernandez-​Orallo of the Universitat Politecnica de Valencia (private communication) for articulating this reservation, and to Dr. Siegmund Probst of Hannover for putting me in touch with him. 16 “So rationalists like Leibniz and Wolfram have always rejected physical randomness, or ‘contingent events’, as Leibniz called them, because they cannot be understood using reason, they utterly refute the power of reason” (Chaitin 2005, 103). 17 For a clear discussion, see (Nachtomy 2007, ch. 1). As he writes, “If a state of affairs has not occurred, does not occur, and will not occur, then, in the Aristotelian view, it is considered impossible. If it occurs at all times, it is considered necessary. In the end, all possible states of affairs, i.e. all potentialities, are to become actual …” (15). 18 As Leibniz succinctly expresses it: “If something implies a contradiction, it is said to be impossible, otherwise it is understood to be possible. If the contradictory of something implies a contradiction, it is said to be necessary; otherwise it is contingent” (A VI 4, 626). 19 In his essay Specimen Geometriae luciferae (GM VII 273), Leibniz writes that although there is no common measure, there is, rather “a certain resolution as if [quasi] into a common measure, albeit one smaller than any given quantity”. Here the “quasi-​common measure” is an infinitesimal, a mathematical fiction—​useful for calculations, but not a determinate part of the continuum. 20 Leibniz acknowledges the analogy: “Meanwhile, just as incommensurable proportions are subjected to the science of geometry, and we even have demonstrations about infinite series, so much more are contingent or infinite truths subordinate to God’s knowledge, and are known by him not, indeed, through demonstration (which would involve a contradiction), but through an infallible vision. But God’s vision must not at all be conceived as a kind of experiential knowledge, as if he sees something in things that is distinct from himself, but rather as a kind of a priori cognition …” (A VI 4, 1658). The passage continues as quoted above. 21 For an account of these scholastic terms, see (Arthur 2019a). 22 According to Dedekind, “If all points of the straight line fall into two classes such that every point of the first class lies to the left of every point of the second class, then there exists one and only one point which produces this division of all points into two classes, thus severing the straight line into two portions.” That point is the “Dedekind cut”. The quotation is from Dedekind’s Stetigkeit und irrationale Zahlen, in the translation of Vincenzo De Risi from his magisterial (2019), which should be consulted for an account of how close Leibniz came to the notion of a Dedekind cut. 23 For an argument to this effect, and a detailed comparison of Leibniz’s take on the actual infinite with Cantor’s, see Arthur (2019a). 24 Thus Ugo Pagallo, who uncritically follows Chaitin on this point, is simply wrong when he claims that “Randomness is what Leibniz was wrong about in, at least, the realm of mathematical truths, for

173

174

Richard T. W. Arthur we cannot think of all of them as demonstrable” (Pagallo 2007, 289). Pagallo’s analysis is also premised on taking Heidegger as offering “a canonical reading” of Leibniz’s Principle of Sufficient Reason, although he sides with Chaitin over Heidegger. 25 Contra Pagallo (2007, 288) even in his youthful De Arte combinatoria (1666), Leibniz did not claim that all truths “can be reduced to a finite set of principles or axioms considered ‘eternal truths’ ”, since he specifically exempted contingent truths. He reached his mature conception of how to distinguish contingent from necessary truths by about 1684. See my account in Arthur (2014). 26 Again, when Pagallo says of Leibniz’s favourite example of a contingent event, Caesar’s crossing the Rubicon, that “this sound fact … represents a historical truth, and, therefore, should be thought of as demonstrable” (2007, 289), the “therefore” is precisely and profoundly wrong. 27 “The view I propose is not entirely without precedent. Except for its idealism, Leibniz’ account of causation is in spirit the most developed precursor of the account I will give. The following quotation illustrates the importance of form in Leibniz’ philosophy …” (Collier 1999, 3–​4). 28 See (Collier 2011) for an extended discussion and defence of this stance on information. 29 See also (Collier 2010, 10ff.). Collier’s definition of information using the NPI is closely related to that of David Layzer (1990), according to whom “information = maximum randomness –​actual randomness” (28), or “I = Rmax –​R” (138)—​a fact noted by Collier (1999, 18). 30 More formally, “The informational complexity of the disordered part is equal to the entropy of the system, i.e. CI(HMAX –​IP) = CI(HACT) = HACT and since O = IP/​HMAX, the ordered content of S = HMAXO = IP as required” (Collier 1999, 19). 31 “Let s be mapped isomorphically onto some binary string σs (i.e. so that s and only s can be recovered from the inverse mapping), then the informational complexity of s is the length in bits of the shortest self-​delimiting computer program on a reference universal Turing machine that produces σs, minus any computational overhead required to run the program, i.e. C1 = length(σs) –​ O(1). … The second part of the measure, O(1), is a constant (order of magnitude 1) representing the computational overhead required to produce the string σs. This is the complexity of the program that computes σs. It is machine dependent, but can be reduced to an arbitrarily small value, mitigating the machine dependence” (Collier 1999, 6‒7). 32 See Schrödinger’s reference to vis viva and entelechy as “supernatural forces” in What is Life? (1967, 75) and his talk of the fearful monadology of Leibniz” in Mind and Matter (1967, 140). 33 See Ohad Nachtomy’s (2011) for a discussion of these similarities between Schrödinger’s and Leibniz’s views on the difference between living and non-​living matter. 34 Leibniz also attributes a fourth ability to organic bodies, the ability to reproduce further machines of the same species: “Nature, … when it is not readily able to conserve the individual, at least conserves the species of the Machine and of its mechanical motion as much as possible, and contrives a means by which machines of this kind could produce others similar to themselves.” These quotations are from Corpus hominis, written by Leibniz in 1680‒82. I have translated it from Pasini’s critical Latin edition (Pasini 1996, 217‒24). For further discussion, see (Pasini 1996, 107‒20) (Arthur 2016) and François Duchesneau’s recent Handbook article on Leibniz’s biology and physiology (Duchesneau 2018, 474). 35 As François Duchesneau explains, “for Leibniz, the formal perfection of substantial units consists of primitive force as a source of continuous action and diversified effects: this force expresses itself in the organic arrangement (at once mechanical and teleological) of the physical realities deriving their integrative unity from that source” (Duchesneau 2018, 474). 36 Again, this is supported by Duchesneau’s analysis of Leibniz’s physiology: “for all physical realities, particularly the more organized ones, their series of states and processes are caused by an inherent form or force that constitutes and actualizes the law of their series, as intended in the plan of nature” (Duchesneau 2018, 476). 37 See David Layzer’s discussion in C ­ hapter 5 of Cosmogenesis: The Growth of Order in the Universe (1990). 38 Layzer argues that “irreducible or objective randomness—​randomness that does not arise from human ignorance—​can exist only in a universe whose most fundamental laws are quantal rather than classical” (1990, 43). He also advocates what he calls the Strong Cosmological Principle, according to which “a complete description of the initial state contains no information at all that would serve to define a preferred position in space or a preferred direction at any point” (47), and then argues that quantal indeterminacy and cosmological indeterminacy are two sides of the same coin. 39 Cf. (Chaitin 2007, 237): “In our new interest in complex systems, the concepts of energy and matter take second place to the concepts of information and computation.”

174

175

Leibniz and Algorithmic Information Theory 40 “Thus this body of a living thing or animal is always organic; for since every monad is a mirror of the universe in its own way, and since the universe is regulated in a perfectly orderly manner, there must also be order within that which represents it, i.e. in the perceptions of the soul, and consequently also in the body by means of which the universe is represented there” (Monadologie §63, GP VI 618).

References Antognazza, M. R., ed. 2018. The Oxford Handbook of Leibniz. Oxford: Oxford University Press. Arthur, R. T. W. 2013. “Leibniz’s Mechanical Principles (c. 1676): Commentary and Translation”, Leibniz Review, 23, 101‒116. Arthur, R. T. W. 2014. Leibniz. Cambridge: Polity Press. Arthur, R. T. W. 2016. “Leibniz, Organic Matter and Astrobiology”, pp. 81‒107 in Tercentenary Essays on the Philosophy and Science of G. W. Leibniz, eds. Lloyd Strickland, Erik Vynckier and Julia Weckend. London: Routledge, 2016. Arthur, R. T. W. 2019a. “Leibniz in Cantor’s Paradise”, ch. 3 in Leibniz and the Structure of Science: Modern Perspectives on the History of Logic, Mathematics, Epistemology, ed. Vincenzo De Risi. Dordrecht: Springer, 2019. Arthur, R. T. W. 2019b. “Mario Bunge on causality: Some key insights and their Leibnizian precedents”, pp. 185–​204 in Mario Bunge: A Centenary Festschrift, ed. Michael R. Matthews. Cham: Springer Verlag. Atkins, P. W. 1994. The Second Law: Energy, Chaos, and Form. New York: Scientific American Library. Badur, K. and W. Rottstedt. 2004. “Und sie rechnet doch richtig! Erfahrungen beim Nachbau einer Leibniz-​Rechenmaschine”, Studia Leibnitiana 36, 129‒146. Bar-​Hillel, Y. 1964. Language and Information. Reading, MA: Addison-​Wesley. Barwise, J. and J. Perry. 1983. Situations and Attitudes. Cambridge, MA: MIT Press. Breger, Herbert. 2012. “Chaitin, Leibniz and Complexity”, pp. 257–​265 in New Essays in Leibniz Reception: In Science and Philosophy of Science 1800-​2000, eds. R. Krömer and Y. C.-​Drian. Basel: Springer, 2012. Brillouin, L. 1962. Science and Information Theory, 2nd edition. New York: Academic Press. Brooks, D. R. and E. O. Wiley. 1988. Evolution as Entropy: Toward a Unified Theory of Biology, 2nd edition. Chicago: University of Chicago Press. Calude, C. S., ed. 2007. Randomness & Complexity, from Leibniz to Chaitin. Singapore: World Scientific. Chaitin, G. 2006. “The Limits of Reason”, Scientific American, March 2006, 74–​81. Chaitin, G. 2005. Meta Math! The Quest for Omega. New York: Pantheon; reprinted by Vintage Books, New York, 2006. Chaitin, G. 2007. Thinking about Gödel & Turing: Essays on Complexity, 1970–​2007. Singapore: World Scientific. Chaitin, G. 2016. “COLLOQUIUM: Leibniz on Complexity”, Centre for Quantum Technologies (March 2016), accessed from www.youtube.com/​watch?v=QKKD-​VxxFA, 19 June 2019. Collier, J. D. 1990. “Intrinsic Information”, pp. 390‒409 in Information, Language and Cognition: Vancouver Studies in Cognitive Science, Vol. 1, ed. P. P. Hanson. Oxford: Oxford University Press, 1990, paginated according to a downloaded version, 1‒16. Collier, J. D. 1999. “Causation is the Transfer of Information”, pp. 279‒331 in Causation and Laws of Nature, ed. Howard Sankey. Dordrecht: Kluwer, 1999, paginated according to a downloaded version, 1‒32. Collier, J. D. 2010. “Information, Causation and Computation”, ch. 4 in Information and Computation: Essays on Scientific and Philosophical Understanding of Foundations of Information and Computation, eds. G. Dodig-​ Crnkovic and M. Burgin. Singapore: World Scientific, 2011, paginated according to a downloaded version dated 2010, 1‒17. Collier, J. D. 2011. “Kinds of Information in Scientific Use”, TripleC, 9 (2), 295‒304. De Risi, V. 2019. “Leibniz on the Continuity of Space”, ch. 4 in Leibniz and the Structure of Science: Modern Perspectives on the History of Logic, Mathematics, Epistemology, ed. Vincenzo De Risi. Dordrecht: Springer, 2019. Duchesneau, F. 2018. “Physiology and Organic Bodies”, pp. 466‒484 The Oxford Handbook of Leibniz, ed. M. R. Antognazza. Oxford: Oxford University Press. Jaynes, E. T. 1979. “Where Do We Stand on Maximum Entropy?”, in The Maximum Entropy Formalism, eds. R. D. Levine and M. Tribus. Cambridge, MA: MIT Press.

175

176

Richard T. W. Arthur Jones, M. 2018. “Calculating Machine”, pp. 509‒525 in The Oxford Handbook of Leibniz, ed. M. R. Antognazza. Oxford: Oxford University Press. Kestin, J. 1968. A Course in Thermodynamics. Waltham, MA: Blaisdell. Knobloch, E. 2018. “Determinant Theory, Symmetric Functions, and Dyadic”, pp. 225–​246 in The Oxford Handbook of Leibniz, ed. M. R. Antognazza. Oxford: Oxford University Press. Kolmogorov, A. N. 1965. “Three Approaches to the Quantitative Definition of Information”. Problems of Information Transmission, 1, 1‒7. Kolmogorov, A. N. 1968. “Logical Basis for Information Theory and Probability Theory”. IEEE Transactions on Information Theory, 14, 662‒664. Layzer, D. 1990. Cosmogenesis: The Growth of Order in the Universe. Oxford: Oxford University Press. Leibniz, G. W. 1679. De Progressione Dyadica, LH 35, 3 B 2 Bl. 1‒4, ms. 39113 in the Archives. Leibniz, G. W. 1705. “Explication de l’arithmétique binaire, qui se sert des seuls caracteres 0 & 1; avec des remarques sur son utilité, & sur ce qu’elle donne le sens des anciennes figues chinoises de Fohy”, published in Histoire de l’Académie Royale des Sciences année MDCCIII. Avec les mémoires de mathématiques, 1705 (GM VII 223‒227). Leibniz, G. W. 1849–​63. Leibnizens mathematische Schriften. Berlin: Asher; Halle: Schmidt. 7 vols.; cited by volume and page, e.g. (GM III 131). Leibniz, G. W. 1875–​90. Die philosophische Schriften von Gottfried Wilhelm Leibniz. (C. I. Gerhardt, ed.). Berlin: Weidman. 7 vols; cited by volume and page, e.g. (GP VI 282). Leibniz, G. W. 1897. Machina arithmetica in qua non additio tantum et substractio sed et multiplicatio nullo, divisio vero paene nullo animi labore peragatur (R. Jordan, ed.). Zeitschrift für Vermessungswesen 26, 301–​307; English translation in D. E. Smith (ed.), 1929. A Source Book in Mathematics. New York/​London: McGraw-​Hill, 173–​181. Leibniz, G. W. 1923–​. Sämtliche Schriften und Briefe. Deutsche Akademie der Wissenschaften. Berlin: Akademie Verlag; cited by series, volume and page, e.g. (A VI 3, 343). Leibniz, G. W. 1969. Philosophical Papers and Letters, ed. and trans. Leroy E. Loemker, 2nd edition, Hingham, MA: D. Reidel Publishing Company, Inc. Leibniz, G. W. 1985. Theodicy, Auston Farrer, ed. E. M. Huggard, trans. La Salle, IL: Open Court. Leibniz, G. W. 1994. G. W. Leibniz: La réforme de la dynamique. Edition, presentation, traductions et commentaires par Michel Fichant. Paris: J. Vrin. Li, M. and P. Vitànyi. 1993. An Introduction to Kolmogorov Complexity and its Applications, 2nd edition. New York: Springer-​Verlag. Meyerstein, F. W. 2007. “The Dilemma Destiny/​Free–​Will”, ch. 16, pp. 281‒286 in Randomness & Complexity, from Leibniz to Chaitin, ed. C. S. Calude. Singapore: World Scientific. Nachtomy, O. 2007. Possibility, Agency, and Individuality in Leibniz’s Metaphysics. Dordrecht: Springer. Nachtomy, O. 2011. “Leibniz on Artificial and Natural Machines: Or What It Means to Remain a Machine to the Least of Its Parts”, ch. 5, pp. 61‒80 in Machines of Nature and Corporeal Substances in Leibniz, eds. J. E. H. Smith and O. Nachtomy. Dordrecht: Springer, 2011. Pagallo, U. “Aliquid Est Sine Ratione: On some Philosophical Consequences of Chaitin’s Quest for Ω”, ch. 17, pp. 287‒298 in Randomness & Complexity, from Leibniz to Chaitin, ed. C. S. Calude. Singapore: World Scientific. Pasini, E. 1996. Corpo e Funzioni Cognitive in Leibniz. Milan: FrancoAngeli. Schrödinger, E. 1967. What is Life? & Mind and Matter. Cambridge: Cambridge University Press. Shannon, C. E. and W. Weaver. 1949. The Mathematical Theory of Communication. Urbana: University of Illinois Press. Smith, D. E. (ed.). 1929. A Source Book in Mathematics. New York/​London: McGraw-​Hill. Solomonoff, R. 1960. A Preliminary Report on a General Theory of Inductive Inference, Technical Report ZTB-​ 138. Cambridge, MA: Zator Company. Turing, A. 1950. “Computing Machinery and Intelligence”, Mind 49 (236), 433–​460. Available online at www.csee.umbc.edu/​courses/​471/​papers/​turing.pdf accessed February 3, 2021. Weyl, H. 1932. The Open World. New Haven, CT: Yale University Press (Reprint: Woodbridge, CT: Ox Bow Press, 1989). Weyl, H. 1949. Philosophy of Mathematics and Science. Princeton, NJ: Princeton University Press.

176

177

10 INFORMATION VISUALISATION IN THE PHILOSOPHICAL TRANSACTIONS Chris Meyns

Introduction A brief epigraph graced mathematician James Joseph Sylvester’s 1864 paper on the geometric curve they called ‘bicorne’: Turns them to shapes and gives to airy nothing A local habitation and a name. (Sylvester 1864: 579) Perhaps Sylvester merely invoked the Duke of Athens in a spirit of jolliness—​their paper does not say. Yet they may also be conveying a core insight into the implications of the act of ‘turning to shapes’, or of visualising, as I will call it. Does, or can, visualising function as a fundamental shift, or does it for many authors amount to merely a difference of presentation? In this chapter I will focus on information visualisation in one of the earliest scientific venues of scientific publishing—​the one that also welcomed Sylvester’s paper in the 1860s—​namely the Philosophical Transactions of the Royal Society, during the first two-​and-​a-​half centuries of its operation. Specifically, I will consider how in this context the nature and function of the visualisation of information is conceived. Section 2, ‘Journal visualisations’, will give you an overview of the basic characteristics of the Philosophical Transactions corpus and how visualisations occur therein. The next three sections delve into these questions for distinct type of visualisations: diagrams (section 3), maps (section 4) and graphs (section 5). Section 6, ‘Visualise for user performance’, then zooms out again to consider what, at a more general level, can be said about the nature and role of visualised scientific information in the earlier days of the Royal Society’s journal. Overall, I argue that by and large, visualisation here functions not to offer information of a distinct kind, but to optimise user (reader) performance.

177

178

Chris Meyns

Journal visualisations Commencing in 1665, the journal Philosophical Transactions: Giving some Accompt of the Present Undertakings, Studies and Labours of the Ingenious in Many Considerable Parts of the World venue was among the first scientific circulars in Europe (the Paris-​based Journal de Sçavans had started a few months prior). Initiator of the enterprise was scholar and diplomat Henry Oldenburg (1619–​1677), then the Royal Society of London’s Secretary. Justifying his actions, Oldenburg introduced the first issue by noting that nothing was ‘more necessary for promoting the improvement of Philosophical Matters’ than the communication of studies, enterprises and discoveries among its practitioners (Oldenburg 1665: 1). Initially, this consisted of any scientific material that Oldenburg saw fit to print: extracts from correspondence, accounts of experiments and discoveries, or book reports. It took nearly a century, until 1752, for the Society to formally (and financially) take over responsibility for the journal from its successive secretaries. (Moreover, the Society promptly introduced a rudimentary form of peer review, though not under that name (Fyfe 2015: 278), to curb the volume of items it published.) A little over a century later, after always having been run as a general platform, in 1887 the journal split into two specialist venues: Philosophical Transactions Part A (for physical sciences) and Part B (for biological and life sciences). During its formative period, from 1665 to 1886 (inclusive), which I focus on here, I found 7,807 distinct published items. The number of distinct items published each decade peaks in the 1750s, then decreases (see Figure 10.1). Yet, as on average the documents printed got longer and longer, the average information quantity, measured in kB per document, on average goes up; meaning that the total quantity of information published over the years broadly continues to rise (Fyfe 2015; Meyns 2019). What about visualisations that may or may not occur as part of the material published in the Transactions? That depends on what an information visualisation is. A visualisation, as I will understand it, is broadly: any representation which makes an object, situation, or information

Figure 10.1  Number of documents published in the Philosophical Transactions per decade, 1670–​1879, distinguishing between documents that include one or more visualisation (in the form of a diagram, map or graph) and documents that do not include one or more visualisation.

178

179

Information visualisation

set visible (that is, accessible by sight), where this visibility is not exclusively text-​based or narrative. During the period studied here, a growing proportion of items published in the Transactions incorporates one or more visualisation, increasing from 13.23 per cent in the 1670s to 82.67 per cent in the 1870s.1 In total, out of the 7,807 documents published, I found that visualisations of one form or other occur in just over a quarter of documents (26.45 per cent, or 2,065 documents). Imaging techniques used include wood-​ cutting, mainly for simple figures, engraving for illustrations that require finer detail, and toward the end of the period also lithography, which facilitated printing in multiple colours. While there are myriad ways to classify visualisation types, here I will concentrate on three broad, familiar categories: diagrams (or ‘figures’), which are representations, often schematic, of the appearance, structure, or workings of something; maps, which are diagrammatic representations of the spatial arrangement or distribution of something over an area; and graphs (often called ‘charts’ or ‘curves’), namely diagrammatic representations of the relation between variable quantities. Hence, a diagram might document the layout of instrumentation used in an experiment, a map, the locations from which observations were conducted, and a graph may plot the measurements taken (as, for instance, in Benjamin Thompson’s 1781 study on gunpowder). Diagrams make up by far the largest proportion of information visualisations found in the Transactions, occurring in around a quarter (25.08 per cent) of all published documents. Maps and graphs are much rarer overall, occurring in only 1.37 per cent and 1.81 per cent of all documents respectively, and they gain their foothold only later. While maps begin to rise somewhat earlier than graphs, around the 1790s, by the 1830s graphs have overtaken maps as the second most prominent type of information visualisation supplied in the journal (Figure 10.2).2 (Note that each document can contain multiple distinct visualisations, as well as visualisations of different types.) A mark of the extent to which the use of visualisations grew, was that the journal’s production turned increasingly costly. By the 1820s the cost of publishing the Transactions could consume up to 50 per cent of the Society’s average annual income, with in some years the cost of

Figure 10.2  Percentage of documents published in the Philosophical Transactions per decade, 1670‒1879, with one or more diagram, map, or graph.

179

180

Chris Meyns

engraving alone taking up around a sixth of its annual budget (nearly £600) (Fyfe, McDougall-​ Waters, and Moxham 2014: 16). (In the 1890s the Society’s Treasurer, John Evans, urged that the number of illustrations be reduced, and any plate costing more than £35 to produce would require special permission, as printing had become a drain on the Society’s resources (Fyfe 2015: 286).) If visualisations were so costly, and so effortful to produce, what status and role did they play for the authors who insisted on including them with their publications? I will consider this question for diagrams first.

Diagrams Diagrammatic visualisations are a staple of Philosophical Transactions publishing. Diagrams are included from the earliest volumes onwards. A growing literature on the history of scientific illustration has shown how, in the seventeenth century and beyond, images became standard elements in scientific practices, including in its communication in book publications and periodicals (e.g. Ford 1993; Baigrie 1996; Pauwels 2006; Burri and Dumit 2008; Kusukawa 2011, 2019; Moser 2014; Coopmans et al. 2014; Daston 2015; Bredekamp, Dünkel, and Schneider 2019). For the Royal Society, the likelihood of diagrams being present in papers steadily increases, with a marked expansion from the 1780s and throughout the nineteenth century. Not only would images accompany papers in print. In addition—​as was common procedure around ‘curious’ objects and inventions—​many drawings would already have been presented to, and scrutinised by, members the Fellowship at one of its meetings, reaching an (informal) ‘approval’ status on the reliability or authenticity of the depiction (Kusukawa 2011: 285, 288). Where, or in what sort of papers, do diagrams appear in the Transactions corpus? In terms of themes and subject matter, diagrams are particularly prominent in papers that deal with topics in physics and engineering, and they can equally often be found in themes to do with biology or archaeology. For example, in papers reporting experiments, we can regularly find diagrams detailing experimental setup, or exhibiting the instruments used in the investigation. Such is the case for one of the earliest (set of) diagrams to appear in the Transactions, accompanying an article with guidance on how to make observations and do experiments at sea, published in 1666. It shows eight distinct instruments for taking physical measurements—​including a ‘Dipping-​ needle’, a globe ‘to sound the deepest Seas’, and a ‘Glass Tube’ to find ‘different gravities of Water’—​on a single plate (Moray and Rook 1666). Experimental setup diagrams can be seen, for instance, in one of the many studies on gunpowder that the journal published (Thompson 1781, Figure 10.3). In the biological sciences, diagrams are regularly used to report on ‘curious’ species and their features, a testimony to which is Philip Carteret’s 1771 letter reporting on a ‘Camelopardalis’, today better known as the giraffe (Carteret 1771: 29). Archaeologists and exploratory travellers use diagrams to convey archaeological finds, such as coins (Garden and Plot 1685), ancient inscriptions (Swinton 1766) or fossils (Home 1818). Certainly, though, not all diagrammatic visualisations are of the realistic, substitute-​for-​seeing format that the cases mentioned invite, and that they also display. Astronomer David Gregory’s diagram in support of a proof regarding properties of a type of curve named a ‘catenary’ (Gregory 1699) is a clear example, and there are many more. In fact, schematic, abstract geometric diagrams are quite common, especially in papers dealing with topics in mathematics, physics, or astronomy. (These diagrams often get embedded directly into the running text,

180

181

Information visualisation

Figure 10.3  A representation of “the apparatus as it was put up for making the experiments” with gunpowder by Benjamin Thompson in 1781.

building upon a tradition of visual language that had become conventional at least since the time of Euclid.) A broad use of information visualisation in the form of diagrams, also found in the Transactions, is that they can show, exhibit, or in in short: make visible certain things—​whatever it is that gets represented. For quite a number of the diagrams, authors take this for granted to such an extent that they comment little further upon their aim or purpose in including the scheme. A simple ‘see Figure’ or ‘Plate I’ is understood to suffice. Or, again common, they might comment that a certain object is ‘represented in the figure’ (Cat 1752: 324); ‘shown’ (Home 1818); that an image ‘exhibits’ or ‘gives’ certain details (Stokes 1862: 606, 608). It is on a continuum with such remarks that, at some points, authors publishing in the journal get more specific on the type of information that a diagram might be thought to make available, namely when they comment on how a diagram is able to provide the form or appearance of something. For example, a paper detailing an instrument for astronomical observations notes: ‘The drawing accompanying this paper (Tab. XV.) will shew the general form of the instrument; and need very little explanation’ (Wollaston 1793: 136). Other papers refer to how their drawing ‘present[s]‌the curious with the following Form’ (Thomson 1673: 6097), or how the visualisation can help ‘to shew the form of the internal cavity’ (Home 1810: 189). Remarks about how an object’s ‘appearance’ can be conveyed through a diagram are not rare, either. Naturalist Thomas Pennant asks: ‘I beg leave to lay before you the very extra-​ordinary appearance on the thigh of a Turkey’ (Pennant 1781: 81, Figure 10.4), while a paper on coral says one of its diagrams ‘shews the appearance of the tubes’ (Ellis 1757: 193). Such comments point to diagrams fulfilling the role of providing not just any sort of information, but of them channelling information of a very specific sort: namely regarding the spatial, structural shape and appearance features of whatever is represented. That precisely this specific range or type of information is what diagrams provide, is yet underlined in how authors (for realistic work from nature, in any case) regularly comment on scale and sizing of items diagrammatically represented. They will note whether an item is shown ‘magnified’ or conversely in ‘Natural size’ (Ellis 1757: 193), from what angle something is shown (Westwood 1835: 327‒338), or may even provide a scale for the reader to determine dimensions (Brander 1755: 295). 181

182

Chris Meyns

Figure 10.4  Diagram in Thomas Pennant’s 1781 ‘An Account of the Turkey’, showing that “from the thigh-​bone issued a short upright process, and to that grew a large and strong toe, with a sharp and crooked claw, exactly resembling that of a rapacious bird” (Pennant 1781: 81).

In fact, expectations of accuracy rise to such an extent, that when a diagram has not represented everything proportionally, this must explicitly be flagged, as John Sackette does when identifying a ‘sinking of the Earth’ in Kent: I hope Sir, you will understand the Situation of the Place pretty well, tho’ I have not observed exact Proportion in the Sketch; which the Paper would not allow after I have taken the Rise of the Cliffs so high, which I thought proper for the more particular Describing of them. (Sackette 1714: 470) A step further, epistemically, is the thought that in certain instances, a diagram could epistemically function as a substitute for access to the thing represented itself. Sachiko Kusukawa describes this as the image as a form of ‘witness’, valued for its ability to contain, store and subsequently (re-​) induce knowledge (instead of for its aesthetic assessment) (Kusukawa 2011: 288; 2019: 359). Perhaps exemplifying such thoughts, authors may emphasize both the accuracy and the reliable source of the images adjoined to their articles. For example, authors may note they: ‘are enabled to give an accurate drawing of the apparatus’ (Allen and Pepys 1808: 250); that a depiction produced ‘with the greatest accuracy and minuteness’ the relevant nervous structures (Lee 1849: 47). Deducing accuracy from method of production, J. Lockhart Clarke points out: The drawings have been executed by myself with the greatest care, and may be relied on as faithful and exact delineations of what was seen under the microscope. The outlines of all were taken by means of a camera lucida, in order to ensure correctness. (Clarke 1851: 608) 182

183

Information visualisation

To enhance credibility, authors might flag that a drawing was made ‘from life’ (Carteret 1771: 27), that they made it themselves or alternatively, who else they got it from; leaving a clear trail of accountability for the accuracy of the representation. (Perhaps surprisingly, engravers, those who translated scientists’ drawings to plates suited for mass-​printing, were far less often credited, their names generally relegated to a quiet spot in the bottom right of a plate—​‘J. Basire. Lith’.) (The Basire family regularly served the Royal Society in its image production since the 1770s (Fyfe, McDougall-​Waters, and Moxham 2014: 16).) Counter-​intuitively, maybe, precisely when someone urges readers not to trust diagrams, these remarks indirectly affirm that such practice of trusting was expected and common. (If people need to be cautioned not to do something, that suggests that at least some people were doing precisely that thing.) Astronomer Charles Piazzi Smyth instructs readers not to rely too much on single sources (which, they thought, had led to confusion as to the appearance of Saturn): A single drawing should not be looked on, by itself, as of importance in the present state of astronomy; for how can others than the artist prove the reality in nature of anything they may find in that one document, when this alone is before them? (Smyth 1858: 484) Draughtspeople may (unconsciously) make mistakes. To make astronomical diagrams as trustworthy as ‘numerical observation’, we must eliminate errors in representation arising from individual people and instruments (Smyth 1858: 485) (and consider: Daston and Galison 1992, 2007). This suggests that, as information visualisations, diagrams did to a certain extent function in the Philosophical Transactions as a source of spatio-​structural and appearance information regarding their objects. Epistemically, authors (and, we may presume, their readers) do on occasion take them to as ‘stand-​ins’ or witnesses to that object, and will employ various methods—​such as emphasising accuracy, source, or method of production—​to have the diagrams fulfil that information transfer function.

Maps Information contained in map-​based visualisations is bound always to involve some aspect of the organisation of elements in space—​be it land, water, or the heavenly regions. Yet even within that broad characterisation, not all maps inform alike. One category is the plain topological map documenting a region, instances of which begin to appear in the Transactions within the first decade of publication (for instance a map of mining locations in the Darthmore Hills in Devon (Anonymous 1671)). In particular up to the late eighteenth century, a good number of publications sets out to map (to them) lesser known terrains (such as a 1674 map of, among other locations, the island of Jesso—​today’s Hokkaidō—​near Japan (Rembrantz van Nierop 1674), or known terrains in greater detail. This includes many trigonometric surveys of parts of the British Isles and the European continent (see Figure 10.5), stimulated significantly by a succession of papers by major-​general William Roy (1726–​1790) and collaborators (whose campaigning eventually lead to the establishment of the Ordinance Survey of Great Britain) (Roy 1777: 788, 1785: 385, 1787, 1790). As a sub-​category of the basic topological maps we find ‘special interest’ maps, concentrating not so much on the geography itself, but on the location or distribution of a selective category of features in the mapped space. Examples of such maps in the journal include ones detailing the locations of archaeological finds of bones, fossils or inscriptions (Williams 1740; Boddington 183

184

Chris Meyns

Figure 10.5  A “sketch” accompanying Henry Kater’s (1828) paper on trigonometrical operations to determine the difference of longitude between observatories of Paris and Greenwich.

and Hunter 1771; Buckland 1822), or the maps produced by physicist John Goldingham of the locations at which where they carried out experiments measuring the velocity of sound (Goldingham 1823). Yet information visualised in maps can be still further removed from showing rivers and towns when these maps start displaying the spatial distribution of quantitative information, be it physical quantities or other statistical information in so-​called ‘thematic maps’ (Beniger and Robyn 1978: 2; Friendly and Palsky 2007: 208; Young, Valero-​Mora, and Friendly 2011: 19). In fact, the first such map ever to be published appeared in the Philosophical Transactions, with Edmond Halley’s (1656–​1742) ‘An Historical Account of the Trade Winds, and Monsoons, observable in the Seas between and near the Tropics, with an attempt to assign the Phisical cause of the said Winds’ of 1686 (Halley 1686). As the title indicates, this paper included a weather map, with dotted lines indicating the direction of ‘prevailing winds’ on a schema of the earth. Halley provided the map to aid the reader in grasping the conceptual move of tying ‘prevailing’ or average physical quantities to specific geographic regions (Figure 10.6): To help the conception of the reader in a matter of so much difficulty, I believed it necessary to adjoyn a Scheme shewing at one view all the various Tracts and Courses of these Winds; whereby ’tis possible the thing may be better understood, than by any verbal description whatsoever. (Halley 1686: 162–​163) 184

185

Information visualisation

Figure 10.6  Edmond Halley’s 1686 meteorological (thematic) map, showing the “the various Tracts and Courses” of trade winds and monsoons on the geography of the world.

Information regarding physical quantities appears in the journal with some regularity over the following centuries, stimulated by enterprises of gathering vast amounts of data on these matters—​in the guise of weather maps, or maps of magnetic declination in specific parts of the earth (Schlagintweit 1863; Evans 1872; Sabine 1872; Blanford 1874; Sabine 1875). Far less common in the Transactions is the mapping of statistical information relating to human affairs, such as demographic, economic, or transport data. Such information was increasingly being collected from the late eighteenth century onward, gathering speed in the early nineteenth century, but mostly as a state-​related enterprise rather than by scientific societies (hence, etymologically, the label ‘statistics’), and it fell largely outside the inclination of the Society’s membership toward topics in the physical and biological sciences. William Roy had deemed surveys to be ‘universally admitted to be works of great public utility’: useful for infrastructure maintenance in peacetime, a help in defending against enemy invasions during war (Roy 1785: 385). Was this also how map-​based information visualisations function in the Royal Society’s journal? As with diagrams, authors regularly flag that adjoined maps show a certain subject, for instance when they talk of a map that ‘shows’ a location (Martin 1806: 342ff); of geographies being ‘represented’ or ‘shewn’ (Farey 1811: 247, 253); or when they note that locations referred to ‘may also be seen in the Mapp here joyned’ (Rembrantz van Nierop 1674: 202; Goree 1710: 373). They equally regularly emphasise questions of sourcing: who drew the map, and where? (For instance, we find comments such as that a map was ‘the result of operations I have made on the spot’ (Pictet 1791: 110), or ‘drawn from a sketch made on the spot by Professor Buckland’ (Buckland 1822: 236, cf. Valvasor 1687: 412; Hamilton 1795: 92; Sabine 1843: 171; Sabine 1872: 357; Sabine 1875: 162)). But there is far less, if any, emphasis on the information being made available concerning appearance or looks features, or of the map being a substitute or stand-​in for seeing the actual thing. Rather, authors may often point to certain information having already been made available to the reader (a descriptions of a region, enumerations of locations), and remark that a map may simply help them grasp what has been said. For instance, the author contributing a 1671 report on Devon mines (only identified as ‘an Inquisitive person’) comments: That this mysterious underground-​way of working, may the easier and sooner be apprehended, be pleased to cast an eye on the annexed Scheme, in Tab. II … which may give some information to those that have not been conversant in Mines. (Anonymous 1671: 2102) 185

186

Chris Meyns

Others note that if the reader is willing to consult the map (to see where specific volcanic eruptions had happened), ‘it will add much to these Relations’ (Sloane, Alvarez de Toledo, and Morley 1694: 80), or that the sketch given in a map ‘will give some account of the particulars’ referred to (Pictet 1791: 107). As one realises that, in many cases, people would not have been able to see the regions in question from a top-​view or as a cross-​section in the manner in which many of them were mapped, it is less surprising that the point of structural information transfer should be central, and not specific appearance features. More strongly, while matters of sizing and scale are mentioned for diagrams as well as for maps, in the latter case the emphasis often falls on aspects of keeping just proportions and the projection chosen. (We can find comments on scale and projection in many articles, including: Foley and Molyneux 1694; Strachey 1724: 397‒398; Pictet 1791: 109; Hitchins 1801: 164; Frankland 1861: 673.) Engineer and surveyor Henry Beighton (1687–​1743), who had developed an instrument for map production, noted that in making a general map, one’s aim should be ‘to give the Form, Idea, and Proportion, that the Parts bear to the Whole, and one another’, and at such a scale that the whole fits on the intended sheet (Beighton 1740: 756). That Beighton further emphasises not only exactness, but also economy of design (that is, omit anything that is not essential to a plan of the area) (Beighton 1740: 751), signals that it is specifically structural elements, not specific looks, that are to be communicated through the use of maps.

Graphs While ineliminable from scientific publications today, information visualisation in the form a graphic recording was not prominent in Philosophical Transactions from the outset. Such paucity in graphing during the first one-​and-​a-​half centuries of the journal’s existence is not, as such, especially surprising when considered historically. Graphing had simply only recently begun to make some conceptual and practical strides: conceiving of functions of the continuous variables of mathematics as graphable in geometric curves in a coordinate system had been conceptually available at least since René Descartes’s Géométrie (1637). Yet it would still be a further step to extend such practices to the graphic visualisation of observed physical variables; a practice which did not become commonplace until the eighteenth century (cf. Shields 1937: 68; Beniger and Robyn 1978: 2). In the Transactions, graphs only begin to rise from the 1820s onward, with the increase becoming near unstoppable from the 1850s (at no point again decreasing during the period studied).3 Where do graphs occur in the journal? Graphing is most common in subjects to do with physics and physical measurement, though they do also appear to a certain extent in papers in astronomy and chemistry. Common themes around which this form of information visualisation can occur in the Transactions are, indeed, in graphed barometer or thermometer readings to detail temperature or atmospheric pressure. In fact, the very first graph (or, more specifically, histogram) in the journal is one by naturalist Robert Plot, charting observations of ‘the rise and fall of the Mercury in the Barometer’ (Plot 1685: 930) (Figure 10.7). (In their work Plot used an instrument and template developed by Martin Lister, the use of which resulted semi-​ automatically in graphic output; a practice which, as Laura Tilling notes, was fairly common for graphic recording before the nineteenth century (Tilling 1975: 194).) Further prominent themes are registrations of terrestrial magnetism (such as in the work of Sabine 1851; Sabine, McClintock, and Maguire 1863; Airy 1863; Chambers and Chambers 1875), tide graphs (such 186

187

Information visualisation

Figure 10.7  A histogram of atmospheric pressure based on barometer readings by Robert Plot (1685), using a registration template developed by Martin Lister.

as: Whewell 1836; Lubbock 1837; Haughton 1863), chemical analyses (Bunsen and Roscoe 1857; Crookes 1881), and quite some research on gunpowder (Thompson 1781, 1797). That graphs come up in these fields, around this time, has, to an important extent, to do with an increase, in the seventeenth and eighteenth century, in the systematic collection of empirical data (regarding variables of time, distance, space, intensity, and the like) (Young, Valero-​Mora, and Friendly 2011: 18–​19). Often such registrations would be collected in tables. In fact, even some observations that were ‘born graphically’, so to say, with automated recording devices, 187

188

Chris Meyns

Figure 10.8  Two graphs with ‘Life-​Tables’ for ‘Healthy Districts’ of England, as constructed by William Farr (1859). The top figure shows deaths in each year of age; the bottom shows the number of people living in each year of age.

would still be converted back to tabular format for scrutiny (Tilling 1975: 194). Indeed, we see that tables occur more frequently in the Transactions than graphs (or, for that matter, than diagrammatic or map-​based forms of information visualisation). Authors regularly cross-​reference their graphs with tables (noting, for instance, that ‘By means of these numbers the curve fig. 1 in Plate II. has been constructed’ (Airy 1843: 49), or that certain figures ‘represent graphically’ the ratios given in a table (Sabine, McClintock, and Maguire 1863: 654)). Moreover, in the journal, the inclusion of one or more graphs in a paper is also strongly predictive of inclusion of one or more tables: 90.78 per cent (128 of 141 documents) of papers with graphs also included one or more tables. (The converse was not the case: only 6,08 per cent of documents with tables contained one or more graphs.) As with maps, we see fewer graphs associated with information concerning human affairs, such as demography or economics (be it wealth distributions, imports and exports, or transportation). One of the rare instances of a graph in the Transactions in this field is William Farr’s (1807–​1883) article ‘On the Construction of Life-​Tables, illustrated by a New Life-​Table of the Healthy Districts of England’, published in 1859, in which Farr plotted both the ‘decrement of life’ (that is, numbers dying plotted against age, ranging 0 to 100) and the total number of people living at different ages (Farr 1859; Figure 10.8). It might not be a huge surprise to learn that Farr, in addition to their membership of the Royal Society, was also active at the UK General Register Office, where in 1800–​1842 they worked on the UK Census, and hence had acquaintance with this demographic information. While observational information may have been omnipresent in graphs, not all that was graphed in the journal concerned observations. Here I am not merely thinking of papers in, say, geometry that plot abstract mathematical functions (such as Sylvester’s 1864 graph of a

188

189

Information visualisation

‘bicorn’). Also papers that do work with empirical information may regularly be found to plot theoretical or calculated values. For instance in a paper ‘On the Tides’, John William Lubbock graphs lines for both ‘Observations at the London and Liverpool Docks’ and for values ‘as deducted from Bernoulli’s Theory’ (Lubbock 1837). Moreover, authors regularly make use of practices of interpolation, where non-​observed—​calculated or estimated—​intermediate values are inserted into a series; for instance Henry Enfield Roscoe comments of certain intensity tables that they ‘are obtained by careful graphical interpolation from the above numbers’ (Roscoe 1865: 613) (cf. Noble and Abel 1875: 113; Dixon 1884 :643). Hence, graphing practices in the Transactions were significantly, but not exclusively, tied to the gathering of observational information. What role or function, then, does graphing fulfil in the journal? As with diagrams and maps, when authors comment on their graphs, they will often note in general terms that these can make certain information visible or accessible to the reader. For instance, they may explain a ‘curve’ (as the graphs are generally referred to) was added ‘For the purpose of exhibiting at one view the facts pointed out by these observations’ (Christie 1823: 352), or because, quite efficiently, ‘a glance at these curves show how closely the measurements made by the two methods agree’ (Roscoe 1865: 618–​619). As physiologist Henry Newell Martin puts the rationale in a study of the effects of temperature on heart rate: ‘as curves present very quickly and accurately to the apprehension the general outcome of long columns of figures, charts have also been constructed’ (Martin 1883: 674). The general assessment that graphs can ease the absorption of information was shared more widely among those using this form of visualisation in the period. Florence Nightingale—​a member of the Statistical Society of London—​explained that in their work they had graphed information for that exact purpose: ‘To make the figures which have been adopted in Tables I, II, and III more intelligible’ (Nightingale 1859: 6; viz. Brasseur 2005: 167), where William Playfair appreciated the efficiency of graphing, noting that while words or figures can take ‘some pains’ to impress onto memory, ‘It is different with a chart … what in the usual mode was attended with some difficulty, becomes not only easy, but as it were unavoidable’ (Playfair 1801: 6). Yet at several points in the Transactions authors go beyond this, and appear to regard graphs as particularly suited to convey a specific type information: namely information regarding relations or laws. For instance, one may find remarks such as: ‘The curve drawn on Plate 13 … shows the relation’ between density and pressure in a gas (Noble and Abel 1875: 52), or: ‘That curves may be said to represent the relation’ between susceptibility to magnetisation and wire stress (Ewing 1885: 609). In a paper on combustion, chemist Edward Frankland boasts: ‘This uniformity of relation between pressure and luminosity will be more clearly seen from Plate XIX’ (Frankland 1861: 643). Perhaps more remarkable is that some authors assert that graphs can directly show law-​like relations, or laws. Engineer Jacob Perkins comments that they made a graph: ‘For the purpose of representing the law of condensation of water’ (Perkins 1826: 544). A study of the tides can include a graph titled ‘Curve representing the law of rise and fall of the tide at Southampton’ (Airy 1843); and in a chart of magnetic declination consisting of multiple graphed lines, some solid and some dotted, Charles Chambers and collaborator note that ‘the thick curves must be taken as best representing the true law’ (Chambers and Chambers 1875: 383). These comments are not, we can suppose, loose talk. Royal Society Fellow William Whewell, who published regularly in the Transactions, motivates the law-​capturing use of ‘the Method of Curves’ (as they called it) in their stand-​alone publication Novum Organon Renovatum:

189

190

Chris Meyns

The Method may be used to detect the Laws which the observed quantities follow; and also when the Observations are inexact, it may be used to correct these Observations, so as to obtain data more true than the observed facts themselves. (Whewell 1858: 202) Not only are graphs, in cases like these, understood to be able to directly exhibit a specific type of information—​namely, concerning relations and laws—​in addition, publications in the journal take graph-​like information visualisations to fulfil a stronger epistemic function. Akin to how diagrams were, on occasion, thought to be able to function as (epistemic) substitutes for whatever they represented, some authors take graphs as themselves able to provide evidence, to demonstrate, or even to establish the truth of something. Consider what astronomer and geophysicist Edward Sabine, who conducted many studies in terrestrial magnetism, notes on a chart of annual variation of magnetic declination: ‘On directing our attention to this Plate, it is perceived at the first glance, that the range of variation at all the stations is considerably greater during the hours of the day than during those of the night’ (Sabine 1851: 639). Commenting on a graph of the tides, Samuel Haughton remarks: ‘From an inspection of diagrams 1 and 2, Plate XII., it is plain that the conditions of the tide are different at New and Full Moon’ (Haughton 1863: 257). Perceiving that, and it being plain that, something is the case is factual; so in these circumstances, whatever is perceived or plain from the chart must at least be so. Hence, inspecting a graph can enable observation of facts. Rarely does the exact attitude that attending to graphs can get one to facts, or even the truth, get expressed more clearly than in William Ramsay and Sydney Young’s statement, in their joint article on vapour-​pressure, that: ‘On inspection of the curves representing the above numbers ... the truth of Professor Thomson’s theory is evident’ (Ramsay and Young 1884: 467), or in John William Lubbock’s dry remark that ‘the diagrams in Plate II. appear to confirm the truth of this passage only at neap tides’ (referring to the included graph of tidal observations as a ‘diagram’) (Lubbock 1837: 100). More strongly, not irregularly the graphs themselves are referred to as fulfilling an evidential function. Lubbock, in that same article, extends their claims, showing themselves happy not only to state that certain things can be ‘evident from the diagrams’ (Lubbock 1837: 102), but also that some of their graphs are ‘confirming what is stated by Laplace in the Exposition du système du Monde’ (Lubbock 1837: 99). And compare William Crookes who, in a paper on radiation, notes of a curve with observational data and another with theoretical values, that: ‘The agreement is sufficiently close to prove that the force of radiation varies inversely with the square of the distance of the source’ (Crookes 1876: 336). These statements are testimony to how, for at least some authors publishing in the Philosophical Transactions, graphs can hold significant epistemic force: being able to prove, demonstrate, or show the truth of theses and views.

Visualise for user performance Information visualisations in one of the first scientific journals during the first 231 years of its existence, we have seen, are by their producers understood, and used, directly for the purposes of communication. For all three forms of visualisations studied here—​diagrams, maps, and graphs—​we find authors emphasising how they ‘give’, ‘exhibit’ or allow certain features to ‘be seen’. But can anything be said about the wider functioning, overall? In literature on information visualisation, we may distinguish two opposing stances on how the overall role of visualisation can be captured, diverging on whether information visualised 190

191

Information visualisation

could or could not also have been communicated to the reader by means other than through visualisation. [exclusivity] Visualisations offer information that could not have been provided in a manner other than through visualisation. Though they do not present it as a formal thesis, for instance the statement of Michael Friendly and Gilles Palsky: ‘Data and information visualization are fundamentally about showing quantitative and qualitative information so that a viewer can see patterns, trends, or anomalies, constancy or variation, in ways that other forms—​text and tables—​do not allow’ (Friendly and Palsky 2007: 208) could be interpreted as signally exclusivity. Non-​exclusivity, it follows, must then be: [non-​exclusivity] Visualisations offer information that could have been provided in a manner other than through visualisation. One could find a version of non-​exclusivity with those who claim, as Colin Ware does (for the role of information visualisations today), that visualisations assist in accessing huge quantities of information more efficiently, in less time (Ware 2012: 3‒4): not the content, but the manner of delivering the content is what distinguishes visualisations. What can be said about the position of visualisations in the Philosophical Transactions, on the basis of what has been discussed so far? (Of course I am here not concerned with identifying a uniform single ‘doctrine’ among all authors contributing to the journal across centuries, but rather with what one can call an institutional tendency or inclination in the journal.) Occasional statements might be taken to point in the direction of the first option, exclusivity. Perhaps we consider Hans Sloane’s note that a map they report on ‘will add much’ (Sloane, Alvarez de Toledo, and Myorle 1694: 80) to what they had already said; Halley’s claim that with a ‘Scheme’ added, their point ‘may be better understood, than by any verbal description whatsoever’ (Halley 1686: 163), or James Dunlop’s suggestion that their drawings of a nebula (an outer space cloud of gas and dust) ‘convey a better idea of it than I could possibly hope to do by words’ (Dunlop 1828: 147). Does a visualisation’s exceeding verbal description not mean that certain types of information are exclusively visual, and communicable through visualisation? I suggest that no such inference needs to follow for visualisations in the Transactions. This is because often a description of what is displayed on a map, or in a diagram or graph, is also given (this is the, from a certain point onward, ubiquitous ‘Explanation of the Plates’, which gets added at the end of papers). That signals that visualisation and description can illuminate the same subject matter, though in different guises. More strongly, for thematic maps and for graphs, numerical tables with measurements (or sometimes calculated values) are regularly included, too, and cross-​referenced. Hence for that reason I suggest that, at least for the Royal Society’s journal, there is no reason to infer that what gets visualised must concern information of a distinct kind, exclusively suited to being transferred visually. Instead, I propose that the use and function of information visualisations in the Philosophical Transactions is better understood as concerned with (what can be called) ‘user performance’. Visualisations in the journal help ‘user performance’, first of all, because the advantages authors often cite in including a diagram, map or graph are often performance related: they concern the speed, efficiency and convenience of using such a visualisation over the effort of having to offer (potentially elaborate) description. We witness Halley and Dunlop explicitly 191

192

Chris Meyns

noting that their schemes convey things ‘better’ than words. And Whewell, in the same work cited earlier, praised the ‘efficacy’ of graphing, which he explained was due to ‘the faculty which the eye possesses, of readily detecting regularity and irregularity in forms’ (Whewell 1858: 202). (Performance features of visualisation are also regularly picked up in scholarly literature on the history of the subject matter, such as Tilling’s claim that a graph ‘may convey in a few seconds information that could only be gleaned from a table of measurements by hours of close study’ (Tilling 1975: 193), or Fienberg’s point that visualisations can provide access to information in summarised form (Fienberg 1979: 1); where such summary should be understood to increase efficiency.) Second, I argue that the performance benefits of visualisations are best understood to lie on the part of the user of the visualisation. (Where ‘user’ should be understood as a relative or functional category: designating whoever consults the diagram, map or graph.) As Tilling notes, this can also include the original scientist who, say, glances at a graph to quickly check their data for outliers (Tilling 1975: 193). To understand this user-​focus in the functioning of information visualisations, we would do well to recall the practical point, as has been laid out in section 2 on journal visualisations, that woodcuts, engravings and lithographs were comparatively costly to produce—​a testimony to which is the Society’s Treasurer’s attempts to curtail the journals production costs. Including a visualisation would place an extra burden on the part of the author (who would generally produce the original drawings or sketches), the printer-​illustrator (who would produce blocks or plates for mass-​production), as well as on the journal’s owner-​ publisher (who would shoulder the additional cost of having visualisations printed), compared to setting regular type for printing narrative text. On the producer side of the journal, adding visualisations to papers would come with additional effort. Yet for the user of the volumes of the journals, that is, those gaining information from the various papers, reports, and correspondence printed there, the information processing benefits can be expected to ensue. Hence, I conclude that in the Philosophical Transactions between 1665 and 1886, there is good indication that such visualisations of scientific information functioned in providing information in a manner that enhances user performance, or the user’s efficacy in accessing information.

Conclusion Visualisations of scientific information occur with increasing frequency on the pages of the Philosophical Transactions during its early period as a generalist scientific journal, between 1665–​ 1886. During the early days, added schemes could sometimes be presented rather apologetically, as some frivolous add-​on, ‘for the satisfaction of the Curious’ (Anonymous 1667: 487). However, when we take a broader view at how some of the core categories of visualisation are commented on in the journal—​I considered diagrams, maps, and graphs—​one gets a much sharper indication of how authors conceived the nature and function of information visualisation to operate in their printed work. Here I have shown that, while all three categories of visualisation are, naturally, understood to ‘show’, ‘exhibit’ or allow the reader to ‘see’ certain information, the epistemic positioning can vary across the different formats. Diagrams are occasionally employed to function epistemically as a ‘stand-​in’, substitute for, or witness to, whatever they represent. With maps, the emphasis is generally much more on the visualisation (often on large, fold-​out formats) as a useful and trustworthy tool for planning and decision making. And not irregularly graphs are presented as fulfilling an enhanced epistemic role, of being able to prove, offer evidence, or even to confirm the truth of certain statements. 192

193

Information visualisation

Stepping back, I have argued that there is no need to infer that authors publishing in the Transactions saw visualisations as being able to fulfil these functions in virtue of them providing access to a distinct, exclusive kind of information. Instead, we can see information visualisation in the journal as geared, broadly, toward aiding user performance: cumbersome and costly to produce, yes, but so much easier and efficient to work with on the part of the reader.

Notes 1 Here I work with the proportion of documents which include one or more visualisations, not an absolute number of distinct visualisations. This is because, in particular in the earlier years of the journal, liberties in printing can leave significant ambiguity over what should be counted as one single visualisation (figures are not always numbered, and are frequently bundled on a single “Plate”). 2 For comparison: tables, understood as systematic displays of sets of facts or figures, often using columns, are yet more prevalent than any of these forms of information visualisation, occurring in 26.94 per cent of documents. I call tables an “in-​between” category of visualisation, as they remain importantly text-​ or number-​based, but clearly use spatial arrangement of visible features to communicate information. Scholarly literature features both those who do count tables as visualisations (such as Beniger and Robyn 1978: 1) and those who do not (such as Friendly 2008: 502). 3 As M.C. Shields (Shields 1937: 69‒71) had already noted in an early study on the subject, during the period from circa 1780 to 1840, few graphs were published in scientific journals more generally. Where graphs occured, they were often restricted to charts of observations of physical quantities, such as barometer and thermometer readings, observations of the tides, rainfall, and terrestrial magnetism.

References Airy, George Biddell. 1843. ‘On the Laws of Individual Tides at Southampton and at Ipswich’. Philos. Trans. R. Soc. Lond. 133: 45–​54. https://​doi.org/​10.1098/​rstl.1843.0007. Airy, George Biddell. 1863. ‘On the Diurnal Inequalities of Terrestrial Magnetism’. Philos. Trans. R. Soc. Lond. 153: 309–​16. https://​doi.org/​10.1098/​rstl.1863.0013. Allen, William and William Hasledine Pepys. 1808. ‘On the Changes Produced in Atmospheric Air, and Oxygen Gas, Respiration’. Philos. Trans. R. Soc. Lond. 98: 249–​81. https://​doi.org/​10.1098/​ rstl.1808.0018. Anonymous. 1667. ‘Account of Athanasii Kircheri China Illustrata’. Philos. Trans. R. Soc. Lond. 2 (26): 484–​ 88. https://​doi.org/​10.1098/​rstl.1666.0030. Anonymous. 1671. ‘An Accompt of Some Mineral Observations Touching the Mines of Cornwal and Devon’. Philos. Trans. R. Soc. Lond. 6 (69): 2096–​113. https://​doi.org/​10.1098/​rstl.1671.0004. Baigrie, Brian Scott. 1996. Picturing Knowledge: Historical and Philosophical Problems Concerning the Use of Art in Science. Toronto: University of Toronto Press. Beighton, Henry. 1740. ‘A New Plotting-​Table for Taking Plans and Maps, in Surveying’. Philos. Trans. R. Soc. Lond. 41 (461): 747–​61. https://​doi.org/​10.1098/​rstl.1739.0131. Beniger, James R. and Dorothy L. Robyn. 1978. ‘Quantitative Graphics in Statistics’. The American Statistician 32 (1): 1–​11. https://​doi.org/​10.1080/​00031305.1978.10479235. Blanford, Henry Francis. 1874. ‘The Winds of Northern India’. Philos. Trans. R. Soc. Lond. 164: 563–​653. https://​doi.org/​10.1098/​rstl.1874.0017. Boddington, John and William Hunter. 1771. ‘Account of Some Bones Found in the Rock of Gibraltar’. Philos. Trans. R. Soc. Lond. 60: 414–​16. https://​doi.org/​10.1098/​rstl.1770.0037. Brander, Gustavus. 1755. ‘An Account of a Remarkable Echinus’. Philos. Trans. R. Soc. Lond. 49: 295–​95. https://​doi.org/​10.1098/​rstl.1755.0048. Brasseur, Lee. 2005. ‘Florence Nightingale’s Visual Rhetoric in the Rose Diagrams’. Technical Communication Quarterly 14 (2): 161–​82. https://​doi.org/​10.1207/​s15427625tcq1402_​3. Bredekamp, Horst, Vera Dünkel, and Birgit Schneider. 2019. The Technical Image: A History of Styles in Scientific Imagery. Chicago: University of Chicago Press. Buckland, William. 1822. ‘Account of an Assemblage of Fossil Teeth and Bones’. Philos. Trans. R. Soc. Lond. 112: 171–​236. https://​doi.org/​10.1098/​rstl.1822.0017.

193

194

Chris Meyns Bunsen, Robert and Henry Enfield Roscoe. 1857. ‘Photo-​Chemical Researches’. Philos. Trans. R. Soc. Lond. 147: 381–​402. https://​doi.org/​10.1098/​rstl.1857.0019. Burri, Regula Valérie and Joseph Dumit. 2008. ‘Social Studies of Scientific Imaging and Visualization’. The Handbook of Science and Technology Studies 297: 297–​317. Carteret, Philip. 1771. ‘A Letter on a Camelopardalis Found about the Cape of Good Hope’. Philos. Trans. R. Soc. Lond. 60: 27–​29. https://​doi.org/​10.1098/​rstl.1770.0003. Cat, Claude Nicholas le. 1752. ‘A Letter from Mons Le Cat’. Philos. Trans. R. Soc. Lond. 47: 324–​29. https://​doi.org/​10.1098/​rstl.1751.0052. Chambers, Charles and F. Chambers. 1875. ‘On the Mathematical Expression of Observations of Complex Periodical Phenomena’. Philos. Trans. R. Soc. Lond. 165: 361–​402. https://​doi.org/​10.1098/​ rstl.1875.0014. Christie, Samuel Hunter. 1823. ‘On the Diurnal Deviations of the Horizontal Needle When under the Influence of Magnets’. Philos. Trans. R. Soc. Lond. 113: 342–​92. https://​doi.org/​10.1098/​rstl. 1823.0026. Clarke, Jacob Augustus Lockhart. 1851. ‘Researches into the Structure of the Spinal Chord’. Philos. Trans. R. Soc. Lond. 141: 607–​21. https://​doi.org/​10.1098/​rstl.1851.0029. Coopmans, Catelijne, Janet Vertesi, Michael E. Lynch, and Steve Woolgar. 2014. Representation in Scientific Practice Revisited. Cambridge, MA: MIT Press. Crookes, William. 1876. ‘On Repulsion Resulting from Radiation’. Philos. Trans. R. Soc. Lond. 166: 325–​ 76. https://​doi.org/​10.1098/​rstl.1876.0013. Crookes, William. 1881. ‘On the Viscosity of Gases at High Exhaustions’. Philos. Trans. R. Soc. Lond. 172: 387–​446. https://​doi.org/​10.1098/​rstl.1881.0007. Daston, Lorraine and Peter Galison. 1992. ‘The Image of Objectivity’. Representations, 40: 81–​128. Daston, Lorraine and Peter Galison. 2007. Objectivity. New York: Zone Books. Daston, Lorraine. 2015. ‘Epistemic Images’. In Vision and Its Instruments. Edited by A. Payne. Philadelphia: Pennsylvania State University Press, pp. 13–​35. Dixon, Harold Baily. 1884. ‘Conditions of Chemical Change in Gases’. Philos. Trans. R. Soc. Lond. 175: 617–​84. https://​doi.org/​10.1098/​rstl.1884.0022. Dunlop, Jacob. 1828. ‘A Catalogue of Nebulæ and Clusters of Stars in the Southern Hemisphere’. Philos. Trans. R. Soc. Lond. 118: 113–​51. https://​doi.org/​10.1098/​rstl.1828.0010. Ellis, John. 1757. ‘An Account of a Red Coral from the East-​Indies’. Philos. Trans. R. Soc. Lond. 50: 188–​ 94. https://​doi.org/​10.1098/​rstl.1757.0024. Evans, Frederick John Owen. 1872. ‘On the Present Amount of Westerly Magnetic Declination [Variation of the Compass] on the Coast of Great Britain’. Philos. Trans. R. Soc. Lond. 162: 319–​30. https://​doi. org/​10.1098/​rstl.1872.0014. Ewing, James Alfred. 1885. ‘Experimental Researches in Magnetism’. Philos. Trans. R. Soc. Lond. 176: 523–​ 640. https://​doi.org/​10.1098/​rstl.1885.0010. Farey, John. 1811. ‘An Account of the Great Derbyshire Denudation’. Philos. Trans. R. Soc. Lond. 101: 242–​ 56. https://​doi.org/​10.1098/​rstl.1811.0015. Farr, William. 1859. ‘On the Construction of Life-​Tables’. Philos. Trans. R. Soc. Lond. 149: 837–​78. https://​doi.org/​10.1098/​rstl.1859.0034. Fienberg, Stephen E. 1979. ‘Graphical Methods in Statistics’. The American Statistician 33 (4): 165–​78. https://​doi.org/​10.1080/​00031305.1979.10482688. Foley, Sam and Thomas Molyneux. 1694. ‘An Account of the Giants Causeway in the North of Ireland’. Philos. Trans. R. Soc. Lond. 18 (212): 170–​82. https://​doi.org/​10.1098/​rstl.1694.0036. Ford, Brian J. 1993. Images of Science: A History of Scientific Illustration. Oxford: Oxford University Press. Frankland, Edward. 1861. ‘On the Influence of Atmospheric Pressure upon Some of the Phenomena of Combustion’. Philos. Trans. R. Soc. Lond. 151: 629–​53. https://​doi.org/​10.1098/​rstl.1861.0027. Friendly, Michael. 2008. ‘The Golden Age of Statistical Graphics’. Statistical Science 23 (4): 502–​35. Friendly, Michael and Gilles Palsky. 2007. ‘Visualizing Nature and Society’. In Maps. Edited by James R. Akerman and Robert W. Karrow, 205–​51. Chicago, IL: University of Chicago Press. Fyfe, Aileen. 2015. ‘Journals, Learned Societies and Money’. Notes and Records 69 (3): 277–​299. Garden, George and Robert Plot. 1685. ‘Extracts of Two Letters’. Philos. Trans. R. Soc. Lond. 15 (175): 1148–​58. https://​doi.org/​10.1098/​rstl.1685.0060. Goldingham, John. 1823. ‘Experiments for Ascertaining the Velocity of Sound’. Philos. Trans. R. Soc. Lond. 113: 96–​139. https://​doi.org/​10.1098/​rstl.1823.0013.

194

195

Information visualisation Goree, Pr. 1710. ‘A Relation of New Island’. Philos. Trans. R. Soc. Lond. 27 (332): 354–​75. https://​doi. org/​10.1098/​rstl.1710.0034. Gregory, David. 1699. ‘Responsio ad animadversionem ad Davidis Gregorii Catenariam’. Philos. Trans. R. Soc. Lond. 21 (259): 419–​26. https://​doi.org/​10.1098/​rstl.1699.0076. Halley, Edmond. 1686. ‘An Historical Account of the Trade Winds, and Monsoons, Observable in the Seas between and near the Tropicks’. Philos. Trans. R. Soc. Lond. 16 (183): 153–​68. https://​doi.org/​ 10.1098/​rstl.1686.0026. Hamilton, William. 1795. ‘An Account of the Late Eruption of Mount Vesuvius’. Philos. Trans. R. Soc. Lond. 85: 73–​116. https://​doi.org/​10.1098/​rstl.1795.0006. Haughton, Samuel. 1863. ‘On the Tides of the Arctic Seas’. Philos. Trans. R. Soc. Lond. 153: 243–​72. https://​doi.org/​10.1098/​rstl.1863.0011. Hitchins, Malachy. 1801. ‘Account of the Discovery of Silver in Herland Copper Mine’. Philos. Trans. R. Soc. Lond. 91: 159–​64. https://​doi.org/​10.1098/​rstl.1801.0010. Home, Everard. 1810. ‘On the Gizzards of Grazing Birds’. Philos. Trans. R. Soc. Lond. 100: 184–​89. https://​doi.org/​10.1098/​rstl.1810.0011. Home, Everard. 1818. ‘Additional Facts Respecting the Fossil Remains of an Animal’. Philos. Trans. R. Soc. Lond. 108: 24–​32. https://​doi.org/​10.1098/​rstl.1818.0005. Kusukawa, Sachiko. 2011. ‘Picturing Knowledge in the Early Royal Society’. Notes and Records 65 (3): 273–​94. https://​doi.org/​10.1098/​rsnr.2010.0094. Kusukawa, Sachiko. 2019. ‘The Early Royal Society and Visual Culture’. Perspectives on Science 27 (3): 350–​94. Lee, Robert. 1849. ‘Postscript to a Paper “On the Ganglia and Nerves of the Heart”’. Philos. Trans. R. Soc. Lond. 139: 47–​48. https://​doi.org/​10.1098/​rstl.1849.0004. Lubbock, John William. 1837. ‘On the Tides’. Philos. Trans. R. Soc. Lond. 127: 97–​140. https://​doi.org/​ 10.1098/​rstl.1837.0011. Mallet, Robert. 1862. ‘Appendix to the Account of the Earthquake-​ Wave Experiments Made at Holyhead’. Philos. Trans. R. Soc. Lond. 152: 663–​76. https://​doi.org/​10.1098/​rstl.1862.0033. Martin, Edward. 1806. ‘Description of the Mineral Bason in the Counties of Monmouth, Glamorgan, Brecon, Carmarthen, and Pembroke’. Philos. Trans. R. Soc. Lond. 96: 342–​ 47. https://​doi.org/​ 10.1098/​rstl.1806.0019. Martin, Henry Newell. 1883. ‘The Direct Influence of Gradual Variations of Temperature upon the Rate of Beat of the Dog’s Heart’. Philos. Trans. R. Soc. Lond. 174: 663–​88. https://​doi.org/​10.1098/​ rstl.1883.0021. Meyns, Chris. 2019. ‘ “Data” in the Royal Society’s Philosophical Transactions, 1665–​1886’. Notes and Records, 1–​21. https://​doi.org/​10.1098/​rsnr.2019.0024. Moray, R. and Mr. Rook. 1666. ‘Directions for Observations and Experiments to Be Made by Masters of Ships, Pilots, and Other Fit Persons in Their Sea-​Voyages’. Philos. Trans. R. Soc. Lond. 2 (24): 433–​48. https://​doi.org/​10.1098/​rstl.1666.0009. Moser, Stephanie. 2014. ‘Making Expert Knowledge through the Image: Connections between Antiquarian and Early Modern Scientific Illustration’. Isis 105 (1): 58–​99. https://​doi.org/​10.1086/​ 675551. Nightingale, Florence. 1859. A Contribution to the Sanitary History of the British Army during the Late War with Russia. London: John W. Parker and Son. Noble, Andrew and Frederick Augustus Abel. 1875. ‘Researches on Explosives’. Philos. Trans. R. Soc. Lond. 165: 49–​155. https://​doi.org/​10.1098/​rstl.1875.0002. Oldenburg, Henry. 1665. ‘An Introduction to This Tract’. Philosophical Transactions of the Royal Society of London 1 (1): 1–​2. https://​doi.org/​10.1098/​rstl.1665.0002. Pauwels, Luc. 2006. Visual Cultures of Science: Rethinking Representational Practices in Knowledge Building and Science Communication. Lebanon, NH: University Press of New England. Pennant, Thomas. 1781. ‘An Account of the Turkey’. Philos. Trans. R. Soc. Lond. 71: 67–​81. https://​doi. org/​10.1098/​rstl.1781.0028. Perkins, Jacob. 1826. ‘On the Progressive Compression of Water by High Degrees of Force’. Philos. Trans. R. Soc. Lond. 116: 541–​47. https://​doi.org/​10.1098/​rstl.1826.0036. Pictet, Marc-​Auguste. 1791. ‘Considerations on the Convenience of Measuring an Arch of the Meridian’. Philos. Trans. R. Soc. Lond. 81: 106–​27. https://​doi.org/​10.1098/​rstl.1791.0009. Playfair, William. 1801. The Statistical Breviary. London: J. Wallis.

195

196

Chris Meyns Plot, Robert. 1685. ‘A Letter from Dr. Robert Plot of Oxford’. Philos. Trans. R. Soc. Lond. 15 (169): 930–​ 43. https://​doi.org/​10.1098/​rstl.1685.0022. Ramsay, William and Sydney Young. 1884. ‘Influence of Change of Condition from the Liquid to the Solid State on Vapour-​Pressure’. Philos. Trans. R. Soc. Lond. 175: 461–​78. https://​doi.org/​10.1098/​ rstl.1884.0019. Rembrantz van Nierop, Dirick. 1674. ‘A Narrative of Some Observations Made upon Several Voyages’. Philos. Trans. R. Soc. Lond. 9 (109): 197–​208. https://​doi.org/​10.1098/​rstl.1674.0044. Roscoe, Henry Enfield. 1865. ‘On a Method of Meteorological Registration of the Chemical Action of Total Daylight’. Philos. Trans. R. Soc. Lond. 155: 605–​31. https://​doi.org/​10.1098/​rstl.1865.0012. Roy, William. 1777. ‘Experiments and Observations Made in Britain’. Philos. Trans. R. Soc. Lond. 67: 653–​ 787. https://​doi.org/​10.1098/​rstl.1777.0035. Roy, William. 1785. ‘An Account of the Measurement of a Base on Hounslow-​Heath’. Philos. Trans. R. Soc. Lond. 75: 385–​480. https://​doi.org/​10.1098/​rstl.1785.0024. Roy, William. 1787. ‘An Account of the Mode Proposed to Be Followed in Determining the Relative Situation of the Royal Observatories of Greenwich and Paris’. Philos. Trans. R. Soc. Lond. 77: 188–​469. https://​doi.org/​10.1098/​rstl.1787.0019. Roy, William. 1790. ‘An Account of the Trigonometrical Operation’. Philos. Trans. R. Soc. Lond. 80: 111–​ 614. https://​doi.org/​10.1098/​rstl.1790.0015. Fyfe, Aileen, Julie McDougall-​Waters, and Noah Joseph Moxham. 2014. ‘Philosophical Transactions: 350 Years of Publishing at the Royal Society (1665–​2015)’. Report. University of St Andrews. http://​hdl. handle.net/​10023/​6058 Sabine, Edward. 1843. ‘Contributions to Terrestrial Magnetism—​No. V’. Philos. Trans. R. Soc. Lond. 133: 145–​231. https://​doi.org/​10.1098/​rstl.1843.0011. Sabine, Edward. 1851. ‘On the Annual Variation of the Magnetic Declination’. Philos. Trans. R. Soc. Lond. 141: 635–​41. https://​doi.org/​10.1098/​rstl.1851.0032. Sabine, Edward. 1872. ‘Contributions to Terrestrial Magnetism—​No. XIII’. Philos. Trans. R. Soc. Lond. 162: 353–​433. https://​doi.org/​10.1098/​rstl.1872.0016. Sabine, Edward. 1875. ‘Contributions to Terrestrial Magnetism—​No. XIV’. Philos. Trans. R. Soc. Lond. 165: 161–​203. https://​doi.org/​10.1098/​rstl.1875.0004. Sabine, Edward, Francis Leopold McClintock, and Robert Maguire. 1863. ‘Results of Hourly Observations of the Magnetic Declination’. Philos. Trans. R. Soc. Lond. 153: 649–​63. https://​doi.org/​ 10.1098/​rstl.1863.0030. Sackette, John. 1714. ‘A Letter of the Reverend Mr. John Sackette’. Philos. Trans. R. Soc. Lond. 29 (349): 469–​73. https://​doi.org/​10.1098/​rstl.1714.0058. Schlagintweit, Hermann de. 1863. ‘Numerical Elements of Indian Meteorology’. Philos. Trans. R. Soc. Lond. 153: 525–​42. https://​doi.org/​10.1098/​rstl.1863.0025. Shields, Margaret C. 1937. ‘The Early History of Graphs in Physical Literature’. American Journal of Physics 5 (2): 68–​71. Sloane, Hans, Pr. Alvarez de Toledo, and Christopher Love Morley. 1694. ‘A Letter from Hans Sloane’. Philos. Trans. R. Soc. Lond. 18 (209): 78–​100. https://​doi.org/​10.1098/​rstl.1694.0019. Smyth, Charles Piazzi. 1858. ‘Astronomical Experiment on the Peak of Teneriffe’. Philos. Trans. R. Soc. Lond. 148: 465–​533. https://​doi.org/​10.1098/​rstl.1858.0025. Stokes, George Gabriel. 1862. ‘On the Long Spectrum of Electric Light’. Philos. Trans. R. Soc. Lond. 152: 599–​619. https://​doi.org/​10.1098/​rstl.1862.0030. Strachey, John. 1724. ‘An Account of the Strata in Coal-​Mines, &c’. Philos. Trans. R. Soc. Lond. 33 (391): 395–​98. https://​doi.org/​10.1098/​rstl.1724.0073. Swinton, John. 1766. ‘Remarks on the Palmyrene inscriptions at Teive’. Philos. Trans. R. Soc. Lond. 56: 4–​ 9. https://​doi.org/​10.1098/​rstl.1766.0002. Sylvester, James Joseph. 1864. ‘Algebraical Researches’. Philos. Trans. R. Soc. Lond. 154: 579–​666. https://​ doi.org/​10.1098/​rstl.1864.0017. Thompson, Benjamin. 1781. ‘New Experiments upon Gun-​Powder’. Philos. Trans. R. Soc. Lond. 71: 229–​ 328. https://​doi.org/​10.1098/​rstl.1781.0039. Thompson, Benjamin. 1797. ‘Experiments to Determine the Force of Fired Gunpowder’. Philos. Trans. R. Soc. Lond. 87: 222–​92. https://​doi.org/​10.1098/​rstl.1797.0014. Thomson, William. 1673. ‘A Description of a Bee-​House’. Philos. Trans. R. Soc. Lond. 8 (96): 6076–​78. https://​doi.org/​10.1098/​rstl.1673.0038.

196

197

Information visualisation Tilling, Laura. 1975. ‘Early Experimental Graphs’. British Journal for the History of Science 8 (3): 193–​213. https://​doi.org/​10.1017/​S0007087400014229. Valvasor, Johann Weikhard von. 1687. ‘An Extract of a Letter’. Philos. Trans. R. Soc. Lond. 16 (191): 411–​ 27. https://​doi.org/​10.1098/​rstl.1686.0074. Ware, Colin. 2012. Information Visualization. Waltham, MA: Elsevier. Westwood, J.O. 1835. ‘On the Supposed Existence of Metamorphoses in the Crustacea’. Philos. Trans. R. Soc. Lond. 125: 311–​28. https://​doi.org/​10.1098/​rstl.1835.0019. Whewell, William. 1858. Novum Organon Renovatum. London: J. W. Parker and son. Williams, Stephen. 1740. ‘An Attempt to Examine the Barrows in Cornwall’. Philos. Trans. R. Soc. Lond. 41 (458): 465–​84. https://​doi.org/​10.1098/​rstl.1739.0079. Wilkinson, Leland. 2013. The Grammar of Graphics. New York: Springer. Wollaston, Francis. 1793. ‘A Description of a Transit Circle’. Philos. Trans. R. Soc. Lond. 83: 133–​53. https://​doi.org/​10.1098/​rstl.1793.0013. Young, Forrest W., Pedro M. Valero-​Mora, and Michael Friendly. 2011. Visual Statistics: Seeing Data with Dynamic Interactive Graphics. Hoboken, NJ: Wiley.

197

198

11 ‘DWINDLED INTO CONFUSION AND NONSENSE’ Information in a copyright perspective from the Statute of Anne to Google Books Stina Teilmann-​Lock

‘Information wants to be free’.1 This has long served as an aphorism for Internet freedom activists who have been pointing to the system of intellectual property rights as a main obstacle for access to knowledge and culture. The Free Software Foundation used it as a slogan for open-​ source software initiatives and John Perry Barlow popularized the phrase in his 1994 article in Wired, ‘The Economy of Ideas (Everything you know about intellectual property is wrong)’, which was an attack on the proprietary logics of copyright and patent law and an insistence on their inapplicability in the digital domain (Barlow 1994). Famously, the free-​software activist Richard Stallman explained the ‘freedom’ of freedom of information to be: ‘free’ as in ‘free speech’, not ‘free’ as in ‘free beer’.2 In the early digital culture of the 1990s and 2000s the perceived conflict between freedom of information and copyright protection served to unite hackers, left-​wingers, libertarians, artists and others. Key threats to Internet freedom of the day were considered to include Walt Disney, Warner Bros., SONY, Microsoft and other multinationals who dealt in copyright protected materials that the public had an interest in ‘using’, ‘mashing-​up’ and ‘remixing’ online. This led to one of the paradoxes of the early Internet, which was the phenomenon of corporations suing their own customers, celebrities suing their fans (Coombe 1998). The Open-​Source Movement and the Creative Commons license system came into being as attempts to strengthen the digital public domain by enabling creators to make their work available to the public with ‘some’ or ‘no rights reserved’ and thereby keeping the Internet free and online users on the right side of the law (Lessig 1999). Today the situation has changed in many ways: new stakeholders, new alliances and new differences of interest have appeared with respect to the relationship between freedom of information and copyright law. The history of shifting relationships between information and copyright law will be the focus of this chapter. Importantly, adopting a historical perspective is a means to a better-​informed conception of the role played by copyright law today in relation to the dissemination of information. In particular, we shall attempt to demonstrate that it is reductive to see copyright merely as a constraint on the free flow of information as the ‘information wants to be free’ campaigners have it. Rather, as will be argued, copyright has served, historically, to incite the publication of learned volumes and to promote accuracy and authenticity as 198

199

Information in a copyright perspective

qualifiers of the category of information; in that capacity copyright law has itself been a promoter of the propagation of information. Importantly, information is not protected by copyright. The idea/​ expression doctrine, which is a universally recognized principle in copyright law, prescribes that only the specific expression of any given content is copyrightable; ideas that get conveyed through the means of such expression are not protectable. Today, in legal terms, information belongs to the category of ideas. The divide between idea and expression is a cornerstone of copyright law that dates back to the eighteenth century. Notably, it serves to limit the scope of copyright and to thereby strike a balance between a common—​a public domain of ideas, knowledge, information, culture and so forth—​free for all and a domain of restricted content that originators have exclusive rights in.3 Only the unauthorized taking of ‘expression’, the reproduction of an original work in any material form, constitutes an infringement of copyright law. As such, the idea/​expression doctrine came into being as an anti-​monopolistic measure: the British lawyer William Blackstone used the distinction in an early copyright infringement case, Tonson v. Collins (1762), to ensure that only the words ‘in which an author has clothed his ideas’ would be monopolized. The ideas contained in a book were to be free for readers to use (Tonson v. Collins (1761) 1 Black W. 301). Notwithstanding, as a principle in law, the idea/​expression dichotomy has been criticized for being unhelpful and unclear (Samuels 1989). As a Swedish copyright lawyer, Gunnar Karnell, has put it: ‘an idea is nothing other than its expression or expressions. Ask for an idea and you will get an expression!’ (Karnell 1989: 17). To be sure, the doctrine of the idea/​expression dichotomy is difficult to define clearly. This has been opportune for Big Tech, in particular Google. In so much as the distinction between copyrightable expression and uncopyrightable information is elusive it lends itself to strategic usage by parties in litigation. In particular, as shall be discussed below, the distinction between expression and information was tactically redefined by Google’s lawyers in a way to render Google Books’ snippet view services lawful. We shall go through a series of recent copyright infringement cases over the Google Books project heard by United States courts. The American Authors Guild and the American Association of Publishers were plaintiffs of the cases and Google was the defendant. The rulings were in favour of Google and the outcome of the cases have had notable implications for global information searchers on the Internet. The Google decisions should be understood in the context of three centuries’ worth of having copyright law serve as a tool for regulating the book market and its actors. To fully appreciate the extent to which the distinction between expression and information may serve to advance the ideological interest of different stakeholders it is necessary to consider the historical evolution of literary copyright. We shall look into both the French authorial rights tradition and the British common law tradition for copyright to distinguish ways in which copyright law has directed the flow of information. As we shall see, eighteenth-​century legislators of copyright were deeply motivated by a desire to enable the dissemination of ‘learning’ in the shape of knowledge, ideas and information. As we shall see, historical rationales for introducing copyright are instructive for the sake of appreciating the original intentions behind the introduction of copyright law. Apart from being a means to regulate the commercial market for books, copyright was also employed as an instrument for the regulation of information embedded in printed volumes. Thus, in the French authorial rights tradition, copyright as a regulatory tool implied the protection of so-​called moral rights of authors, defined by copyright law as a bundle of non-​economic rights that serve to, for example, preserve the integrity of published texts. History shows us that the relationship between copyright and information has been both a complex and a productive one. 199

200

Stina Teilmann-Lock

Early copyright: ‘An Act for the Encouragement of Learning’ The world’s first copyright act was the British Statute of Anne of 1710 entitled ‘An Act for the Encouragement of Learning, by Vesting the Copies of Printed Books in the Authors or Purchasers of such Copies, during the Times therein mentioned’.4 While today it is normal for copyright to last for the lifetime of the author plus 70 years, the Statute of Anne offered the more limited term of 21 years of protection to members of the London booksellers’ guild, the ‘Company of Stationers’, as well as 14 years of copyright protection (plus the option of an additional term of 14 years) to non-​members. Obviously copyright law was introduced as (and remains) a form of restriction of the printed (and, today, the digitized) word; as such it is a regime of control over the circulation of knowledge. However, the introduction of copyright went hand in hand with the liberation of the press in Britain and the rest of Europe (Teilmann-​Lock 2016). Before copyright, and since the advent of printing technology, European book markets had been regulated by systems of book privileging (Loewenstein 2002; Armstrong 1990; Feather 1994). Book privileges were temporary monopolies to print books issued by the crown or the state to printers or booksellers who were members of booksellers’ guilds. As such, privileges were emanations of royal favour contrary to copyright, which is a legal right. In the seventeenth and eighteenth centuries, the royal grant of exclusive permissions to publish certain works or types of works for a limited period had played a central role in state censorship: monopolies to print books came with the obligation to function as pre-​publication censors. In this way the book privilege system was part and parcel of state surveillance of the book trade. By the end of the eighteenth century the book privilege systems came to an end in several countries in Europe (Alexander and Gomez-​ Arostegui 2016). The introduction of copyright was the result of Enlightenment thinking in tandem with the advent of a liberal economy. As such, the introduction of the copyright system represented the creation of new economic and information infrastructures to replace the guild system and by this means also the regime of pre-​publication censorship. The copyright historian Ronan Deazley has made the case that, above all, the legislators of the Statute of Anne were concerned with securing a ‘free market of ideas’ (Deazley 2004: 46). In this way, arguably, safeguarding authors’ and publishers’ property rights and economic interests may be said to have been a means to the end of creating a book market that would facilitate the free flow of new knowledge and information by continued printing and distributing of books on learned subjects useful to the public.5 By the same token the revolutionaries in France introduced the first authors’ rights act, La loi du 19–​24 juillet 1793, in the wake of the Declaration of the Rights of Man and Citizen of 26 August, 1789 which had introduced the freedom to speak, write and print. The Declaration had effectively targeted and suppressed the system of book privileges and the control of the press that had existed under the ancien régime. This had paved the way for copyright law and a book market no longer controlled by the Paris Book Guild (the Chambre Syndicale de la Librairie et Imprimerie de Paris).6 Early United States legislators were similarly aware of using copyright as a vehicle for the creation a free market of ideas. The first United States copyright act of 1790 was modelled on the British Statute of Anne, the rationale of which been to support the spread of ideas. Thus, Article 1, Section 8 (8) of the 1787 United States Constitution states that ‘The Congress shall have power … To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Inventions’. To be sure, copyright law established a new type of temporary monopolies on books by granting to authors and their publishers exclusive printing and distribution rights. In that sense 200

201

Information in a copyright perspective

copyright installed a new system for controlling the press. However, given that the backdrop against which copyright law was first introduced was a regime of pre-​publication censorship and a centralized stranglehold on the book market it should be seen as a legislative means to the effect of distributing and decentralizing control of the book market. Thus, while some actors have since used copyright to concentrate power over information, its initial drive was one of decentralization. Significantly, at the time of the introduction of copyright, commentators framed it as a means to curb misinformation. Prominent proponents of early copyright, including Daniel Defoe and Immanuel Kant, stressed the use of copyright law as a tool for making the printed word a carrier of authentic content.

Kant’s theory of copyright Kant, in an article of 1785 entitled ‘Von der Unrechtmäßigkeit der Büchernachdrucks’ (‘On the unlawfullness of the unauthorized publication of books’), made the case for a personality rights based theory of copyright (Kant 1785). The article was a contribution to a debate on book piracy where Kant argued that piracy should be prohibited because it is a violation of the personal right of authors to speak in their own name to the public. Books are here conceptualized as dual-​natured, consisting of, respectively, the material copy, ‘das Exemplar’, and the expression, ‘die Rede’. The material copy like any other kind of material property may be bought and sold by anyone. By contrast, the expression of the book remains the property of the author to whose name the book is tied. (This corresponds to the notion of ‘expression’ in the idea/​expression doctrine referred to above: the author owns the particular expression of a book.) An author must answer for his expression, for what he utters in public. At the same time, authorial accountability must be matched by an exclusive right to authorial expression. A publisher serves merely as a mediator and, as such, as the author’s representative in the bringing forth what the author seeks to communicate to the public. According to Kant, the publisher is under an obligation not only towards the author but also towards the public that the author’s address to them should reach the intended audience. In this way the publisher is a means, authorized and trusted by the author, for conveying the message of the author to the readers by printing and disseminating the author’s writings in book form. As follows, pirate publishers who do not have the authorization of the author to speak in his name must be outlawed. By implication, Kant’s argument in support of copyright creates economic monopolies on the printing of books. Nevertheless, the rationale for copyright is not presented by Kant as the just reward and protection of economic investments (as it usually is today). Rather copyright is presented by Kant as a way to sustain freedom of communication (under responsibility) in a modern-​day Republic of Letters.

Daniel Defoe and the freedom of the press In Britain, Daniel Defoe—​who was deeply involved in the British copyright debates of the turn of the eighteenth century—​presented similar ideas. He was much concerned with the sustenance of the Republic of Letters. Around the time of the passing of the Statute of Anne Defoe had been speaking out against book pirates less because of economic loss and more due to the harm done by pirate publishers to the conveying of accurate information. In A Review of the Affairs of France: And of All Europe, as Influence’d by That Nation (1705) he had argued that the greatest problem with piracy was the diffusing of distorted communication. He pointed out that pirate editions were full of ‘innumerable Errors, by which the Design of the Author is often inverted, conceal’d, or destroy’d, and the Information the World would reap by a curious and 201

202

Stina Teilmann-Lock

well studied Discourse, is dwindled into Confusion and Nonsense’ (Defoe 1705: 26). Publishers are expedient for authors in their capacity to assist them in communicating with the public, thereby creating a free market for information. As Defoe frames it, the publishing industry may be instrumental in the ambition to ‘polish the Learned World, make men Polite, and encrease the knowledge of Letters, and thereby all useful Arts and Sciences’ (Ibid.: 3). Pirate publishers, however, disrupt the free exchange of ideas by offering for sale abridgements or reprints in poor quality that may be inaccurate or misleading: An Author prints a Book, whether on a Civil or Religious Subject, Philosophy, History, or any Subject, if it be a large Volume, it shall be immediately abridg’d by some mercenary Bookseller, employing a Hackney-​writer, who shall give such a contrary Turn to the Sense, such a false Idea of the Design, and so huddle Matters of the greatest Consequence together in abrupt Generals, that no greater Wrong can be done to the Subject; thus the sale of a Volume of twenty Shillings is spoil’d, by perswading People that the Substance of the Book is contain’d in the Summary of 4s. price, the Undertaker is ruin’d, the Reader impos’d upon, and the Author’s perhaps 20 Years Labour lost and undervalued. (Ibid.: 26) Against this background Defoe proposed to see literature as a type of intellectual property protected by copyright law to thereby ensure the integrity of writing and the conveying of knowledge and information the accuracy of which authors would be accountable for. Importantly, as we shall discuss below, the ‘moral rights’ that came out of the French authorial rights tradition of copyright law serve in a similar vein to protect the integrity of works of authorship.

The right to the integrity of the work In the nineteenth century, French courts began to recognize so-​called ‘moral rights’ of the author. Moral rights are personal, non-​economic and inalienable rights of authors, including the author’s right to respect for his name, his qualities and his work.7 Today moral rights have been implemented into copyright law in most countries. Historically, the author’s right to respect for his or her work—​also labelled ‘the right to the integrity of the work’—​has served as a protection against many types of unauthorized modifications of texts thus regulating the integrity of information. Examples from early French case law illustrate this aspect of the relationship between copyright and information. While publishers in the early day of publishing were considered to have a right of revision, a shift took place in the nineteenth century in France. In 1814, a French court decided that books were generally to be printed as delivered. Corrections were to be limited to typographical, orthographical and punctuation errors (Trib. Civ. Seine, 17 août 1814). This principle matured into a full-​blown authorial right to the integrity of the work.8 Two examples from French case law will serve as a demonstration of this right. An appeal case heard by the Court of Lyon, Marle c. Lacordaire (Cour de Lyon, 17 juillet 1845. D.1845.2.128), concerned the abbé Jean-​Baptiste-​Henri Dominique Lacordaire‘s (1802–​ 1861) right in his sermons. The case revolved around the publication, in May 1845, by the writer Charles-​Louis Marle, of an edited volume of Lacordaire’s sermons entitled Les Conférences de l’abbé Lacordaire. Without Lacordaire’s consent Marle had taken down in shorthand sermons given by the abbé. The Court of Appeals in Lyon found that unauthorized publication of the

202

203

Information in a copyright perspective

sermons was in breach of copyright law. Neither the religious character of the work nor the ecclesiastic status of the abbé served as a defence. The author of sermons has the right not only to profit from his own work; given the existence of an author’s ‘personnalité morale’ the court of appeal affirmed that ‘the author should always preserve the rights to revise and correct his own work, to survey the fidelity of the reproduction and to choose the time and mode of publication’.9 The court maintained that any authorial expression was the property of its author and, as such, was protected against unauthorized reproduction that would imply the risk of unauthorized modifications. The originator of a sermon, according to the court, had a right in his work because in effect the orator delivers his speech only, without giving up the power of disposal of his thought through printing; on the contrary it is essential for him to preserve the fruits of his labour, to remain the sole judge of the opportune moment for its publication and to keep guard against potentially dangerous alterations of his work.10 Against this background, the defendant, Charles-​Louis Marle, was found guilty of copyright infringement. He was ordered to pay damages and the infringing copies were recalled. Another case, Delprat c. Charpentier, which was heard by various instances in the years from 1864 to 1868, was a milestone case in the history of the moral rights of authors. It concerned a disagreement between a news editor and the publisher of the journal Revue Nationale. The French Supreme Court, the ‘Cour de Cassation’, had annulled a decision by an appeal court, which had gone against the authorial right of integrity, and had returned the case to the Court of Orléans. Thus, in 1868, in the final ruling by the Court of Orléans the author’s right of integrity was confirmed. Charpentier was the director and owner of the journal, Revue Nationale, and Delprat the political editor of the journal. The dispute began when Charpentier made a number of suppressions and modifications to a too-​long manuscript delivered by Delprat without informing the latter. Delprat maintained that the changes had distorted his article and demanded a disclaimer stating that due to cuts and alterations the article was not his authentic work. Charpentier refused to insert a disclaimer in the journal. Delprat took the dispute to court and the first court decided that while, in principle, a publisher has a right to edit contributions, the authors must be notified thereof. As pointed out by the Court, French law ordered published articles to be signed by their authors. Accordingly, for the sake of the execution of the law there would have to be some assurance that a piece of writing was authored by the person whose name was affixed. On these grounds the court ruled in favour of Delprat. Charpentier appealed and the higher court overturned the decision, finding that inasmuch as, in this court’s opinion, the modifications did not prejudice the meaning and character of Delprat’s writings, his responsibility and reputation had not been compromised (Charpentier c. Delprat, Cour de Paris, 16 mars 1865, D.1865.2.213). However, as mentioned, the Cour de Cassation annulled the second judgment, confirming that a writer is absolute master of his or her work. Accordingly, alterations must be authorized by the one who signs as author (Delprat c. Charpentier, Cour de Cassation, 21 août 1867, D.1867.1.369). As held by the Cour de Cassation, in denying Delprat control over his text the court of appeal had failed to recognize his authorial rights. Hence the case was referred to the Court of Orléans that confirmed Delprat’s right to have a disclaimer printed by Charpentier. As the two cases serve to illustrate, moral rights have set a standard for the credibility of the contents of books. Moral rights have served as a security for authors (not least in nineteenth-​ century Europe where strict blasphemy and libel laws were in force) that they would be held

203

204

Stina Teilmann-Lock

accountable only for their own views. Conversely, moral rights were the readers’ guarantee that a named author was there to answer for the plausibility of ideas expressed in books. By this means, copyright has long served as an infrastructure for securing that what gets communicated in books is authorized. In a nutshell, the copyright system has served as an authorization system where the authoritativeness of content relies on the authority of authorship (supported by publishers). This again has been linked to the fact that well into the twentieth century works of authorship have been the epistemic norm for knowledge expression under copyright law (remembering how, as Foucault noted, ‘the author function does not affect all discourses in a universal and constant way’ (Foucault 1989)). With the advent of the Internet an Enlightenment vision of a universal encyclopedia came alive.11 However, the author-​centric copyright system seemed a stumbling block for the building up of a digital infrastructure for ubiquitous access to information. This was the environment in which ‘information wants to be free’ became a refrain. The World Wide Web seemed to offer the setup for realizing a global library with free access for all. In hindsight—​with the ad-​ driven Web of today—​any hope for a global commons was short-​lived. Yet, in the tragedy of the infinite commons that digitalization has brought about, copyright has not played the part predicted by ‘Information wants to be free’ campaigners. In significant ways, copyright has yielded to the driving logics of digital infrastructures. A prominent example is the litigation over Google Books where Google’s successful legal defence relied on a new distinction forged between copyrightable expression and un-​copyrightable information.

Google Books The Google Books project was launched in 2004. The aim of the project was to digitize books from partners including a large number of libraries and archives and then create a searchable digital database (Thylstrup 2018). Users of the Google Books database are able to make word searches in the whole corpus of digitized books and the search results, comprise meta-​data about books, bibliographic data as well as various analytical data. For example, if a user makes a word search the search result will list which books the word appears in and how many times the word occurs. An associated service is the Ngram Viewer, which provides diagrams of the historical occurrence of a word or phrase in the corpus of Google Books. Books that are out of copyright may in principle be viewed in toto (but often they are not) whereas users are not allowed full access to copyright works. Rather, users of copyright works are permitted to see a number of ‘snippets’ of text wherein the search term appears in highlighting (except from ‘blacklisted’ pages where snippet view is not available). Snippets are usually restricted to an eighth of a page. Thus Google Books is structured around four access levels: ‘full view’, ‘limited preview’, ‘snippet view’ and ‘no preview available’. The legal dispute over Google Books began in 2005 when the American Authors Guild and the American Publishers Association filed a class action against Google. The snippet view was at the centre of the dispute. In 2008–​2009 the parties were working on a settlement agreement but the attempt was unsuccessful. In the process of working out a settlement agreement Google’s lawyers introduced the notions of ‘non-​consumptive research’ and ‘non-​ display uses’.12 A definition for legal purposes of ‘non-​consumptive research’ was presented in the amended settlement agreement between the Authors Guild, the American Publishers Association and Google: ‘Non-​Consumptive Research’ means research in which computational analysis is performed on one or more Books, but not research in which a researcher reads or 204

205

Information in a copyright perspective

displays substantial portions of a Book to understand the intellectual content presented within the Book. (Amended Settlement Agreement, The Authors Guild, Inc. & Association of American Publishers, Inc., et al. v. Google Inc., Case No. 05 CV 8136-​DC, S.D.N.Y. Signed on 13 November, 2009: sec. 1.93) In the amended agreement a list of examples of ‘non-​consumptive research’ included computational image and text analysis, text and information extraction, analysis of historical and synchronic linguistic patterns, automated translation, indexing and searching. By the same token, the definition of what would amount to ‘non-​display uses’ included uses that do not display Expression from Digital Copies of Books or Inserts to the public. By way of example, display of bibliographic information, full-​text indexing without display of Expression (such as listing the number or location of search matches), geographic indexing of Books, algorithmic listings of key terms for chapters of Books, and internal research and development using Digital Copies are all Non-​ Display Uses. (Ibid.: sec. 1.94) The strategic effect of introducing the two concepts of ‘non-​consumptive research’ and ‘non-​ display uses’ was significant (Thylstrup and Teilmann-​Lock 2017). They served to strengthen Google’s defence based on fair use in the cases that came to be heard after the failed attempt for a settlement agreement between the parties. United States copyright law defines a number of limitations to copyright law including ‘fair uses’ of copyright works that do not require the authorization of the copyright holder.13 In the Google Books case the fair use defence came to rely on the ‘transformativeness’ of the uses of copyright books in the Google Books project. An important development in the application of the fair use defence had taken place in the 1990s where ‘transformative use’ began to be given major weight in favour of fair use.14 Based on a set of governing principles for the fair use doctrine defined by Judge Pierre Leval in an article of 1990 in the Harvard Law Review courts took the view that the answer to the question of justification [of the use of a copyright work] turns primarily on whether, and to what extent, the challenged use is transformative. The use must be productive and must employ the quoted matter in a different manner or for a different purpose from the original. (Leval 1990: 1111) In a milestone decision, Campbell v. Acuff-​Rose Music (1994), the United States Supreme Court ruled that transformative use is a use where a work ’adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message’ (Campbell v. Acuff-​Rose Music, Inc., 510 U.S (1994): 579). In the context of mass-​digitization of literature and art the ‘transformative use’ argument became a powerful defence in United States copyright infringement cases. This was also the case in the dispute over Google Books. In 2015, the appeals court ruled in favour of Google. Crucially, the court decided that the website functionality of Google Books including the snippet view was adding ‘importantly to the highly transformative purpose of identifying books of interest to the searcher’ (Authors Guild v. Google Inc., 804 F.3d 202, 2d Cir. 2015: 218). As such, it fell under fair use as defined by US Copyright law. Moreover, the 205

206

Stina Teilmann-Lock

fair use defence for snippet view was strengthened by the fact that although snippets display ‘a fragment of expressive content’, any snippet would have been ‘arbitrarily selected’ (by the searcher’s search term) and as such, as was found by the court, ‘the snippet function does not provide searchers with any meaningful experience of the expressive content of the book’ (Ibid.: 227). On these grounds, the court declared that the search function of Google Books was a non-​infringing use of the copyright works of the Authors Guild’s members, affirming that: the purpose of Google’s copying of the original copyrighted books is to make available significant information about those books [italics in original], permitting a searcher to identify those that contain a word or term of interest, as well as those that do not include reference to it. In addition, through the ngrams tool, Google allows readers to learn the frequency of usage of selected words in the aggregate corpus of published books in different historical periods. (Ibid.: 217) Crucially, the court presented the website functionalities of Google Books as means for turning ‘expression’ into ‘information’, upholding that no copyright infringement had taken place since Google Books does not allow for the ‘reading’ of books. Users of Google Books do not get any meaningful experience of the ‘expressive content’ of books. Rather, as was held by the court, they are provided with access to ‘information’ about the books. The case Authors Guild v Hathi Trust (2014), that also concerned mass-​digitization of literature, had reasoning along similar lines. The Hathi Trust Digital library was sued by the Authors Guild for digitizing and making available the collections of a series of university libraries, including works that were in copyright. The Hathi Trust had collaborated with Google on this project where library users were presented with searchable full texts of books, the books not being eligible for reading. The court ruled against the Authors Guild finding that the use made by the Hathi Trust of the copyright works was fair use as no new ‘human-​readable copies’ of books were brought into circulation. In its analysis the high court stated that we conclude that the creation of a full-​text searchable database is a quintessentially transformative use … the result of a word search is different in purpose, character, expression, meaning, and message from the page (and the book) from which it is drawn. Indeed, we can discern little or no resemblance between the original text and the results of the HDL full-​text search. There is no evidence that the Authors write with the purpose of enabling text searches of their books. Consequently, the full-​text search function does not ‘supersede … the objects [or purposes] of the original creation’, Campbell, 510 U.S. at 579. (Authors Guild v. Hathi Trust, 755 F.3d 87, 2d Cir. 2014: 18) The court of appeal maintained that inasmuch as authors write in order to create human-​ readable texts—​and not in order to create texts that lend themselves to text searches—​the digitized texts that constituted the material of a full-​text searchable database did not amount to infringing copies of the originals. Similarly, as was ruled, the perception of the result of a word search is different from reading the page of a book. Importantly, the digital transformation afforded by Google Books’ website functions was key to Google’s legal defence. The snippet view—​along with the other user design features of Google Books—​are types of ‘knowledge designs’ where the embodiment of knowledge is integral to knowledge itself (Drucker 2014: 79). As such, the knowledge contained in books 206

207

Information in a copyright perspective

became divided into, on the one hand, the ‘information about those books’ which is inextricably linked with ‘non-​consumptive research’ and ‘non-​display uses’ and, on the other hand, ‘expressive content’ linked with reading and understanding. Arguably, this divide takes the Kantian theory of copyright to its ultimate consequence: copyright applies to what authors have to answer for, not to what we may perceive as information. To be sure the American Authors’ Guild and the American Publishers’ Association as well as authors around the world whose works have been digitized and made available on the Internet without their consent consider Google Books an infringement of the copyright in their work. A French ruling confirmed as much, ruling in favour of authors and publishers and against Google (Les Editions du Seuils contre Société Google INC, 18 December, 2009, T.G.I. Paris). Equally the case, Internet users around the world enjoy and use the ‘information about books’, which was made free by the limitation of copyright by the Unites States court. The Google Books case is symptomatic of the ways that we, the Internet users, typically treat online content as if it were information about something—​and, as such, a means to an end—​and not as expression lending itself to scrutiny and interpretation—​and thus an end in itself. Online we perform a mixture of what Katherine Hayles categorizes as different modes of reading: close reading, hyper reading and machine reading. Arguably, what we do most on the Internet is to ‘hyper read’: we skim and scan on screen assisted by a computer that filters by word searches, hyperlinking and picking fragments of texts, and we rely on ‘machine reading’ which is human supervised pattern recognition, summarizing and systematizing by computers (Hayles 2012). Affordances of digital technology are shaping our way of taking in text when reading online. In that way, in the age of Google Books, arguably, we perceive of online books as vessels of information, which in significant ways have been freed from its status as authorial expression.

Conclusion As a product of Enlightenment thinking, copyright law was installed to create a legal infrastructure for the dissemination of knowledge, ideas and information to the public under authorial guarantee of authenticity. The rationales of eighteenth-​century copyright law make this clear. Moral rights have further reinforced this purport of copyright. Under the analogue regime of copyright, publishers served as the trusted intermediaries linking authors with their readers, information with its public. With the arrival of the Internet and mass-​digitization projects, including Google Books, the regulatory logics and logistics of copyright have been disabled in various ways. Google Books as an intermediary is not equivalent to a publisher and the books are not books as we know them but items promoted by a service containing information but no expression, we are told. Searchers are not supposed to read the digitized ‘books’: Google Books is a means for providing information about books according to the American rulings. Notably, this type of freedom of information—​which implies freedom from its status of copyrightable expression—​is indicative of the current situation on the Web. It is a space where whoever can post whatever. Effectively, information has become cheap (because indiscriminate) rather than free. In the analogue world copyright secured the integrity of expression in books and information was conveyed through the means of expression. Defoe and Kant argued as much. Google Books (and, more broadly, the existence of online services that use information not as an end in itself but as a means for generating user traffic) contributed to a new norm for information in digitized books as conveyed independently of copyrightable expression. The rulings by United States courts that affirmed the lawfulness of Google Books added to the creation of the World Wide Web as an infrastructure not governed by copyright law in any significant ways. By the same token they may have added to creating a state of affairs similar to 207

208

Stina Teilmann-Lock

what Defoe described as a Republic of Letters dominated by pirates where, as cited above, ‘the Information the World would reap by a curious and well studied Discourse, is dwindled into Confusion and Nonsense’.

Notes 1 The saying is a section of a longer sequence. At Hackers’ Conference 1984 Stewart Brand is reported to have said that ‘On the one hand information wants to be expensive, because it’s so valuable. The right information in the right place just changes your life. On the other hand, information wants to be free, because the cost of getting it out is getting lower and lower all the time. So you have these two fighting against each other.’ See https://​tech-​insider.org/​personal-​computers/​research/​acrobat/​8505-​ a.pdf (Retrieved on 27 March 2019). 2 Richard Stallman was a prominent promoter of open source software. See https://​www.gnu.org/​philosophy/​free-​sw.en.html. Amusingly, the open source movement inspired the artist group Superflex to create, tongue-​in cheek, the World’s first ‘open source beer’ in the project Free Beer. See https://​www. superflex.net/​tools/​free_​beer/​image (Retrieved on 27 March 2019). 3 As framed by Bernard Edelman, defending the usefulness of the distinction which serves to preserve a public domain of ideas: ‘l’auteur, qui utilise la nature humaine comme son unique matériau, ne peut s’en approprier l’essence. C’est pourqoui le droit nous dit, d’une part, que la nature humaine ne peut être appropriée, d’autre part, que l’idée, qui exprime cette inaliénabilité, est banale et, partant, non protégeable’ (Edelman 1983: 76). 4 (1709) Anne c. 19. (The Statute of Anne 1710). 5 Censorship had existed since 1487 when the Court of Star Chamber was established, one of its main purposes being to control the recently invented printing press. The Company of Stationers had been instrumental in the exercise of state censorship since 1557 when a Royal Charter was granted to them, authorizing their control over the book trade in England. The control of the Stationers’ Company over printing was consolidated by the Licensing Act of 1662, which decreed that every printed item was to be licensed and registered at Stationers’ Hall. In this way, the Licensing Act had worked to regulate printing by the combined measures of making licenses a legal requirement for printing books and of confirming the printing privileges of the Stationers. Since the lapse of the Licensing Act in 1695 the booksellers had been asking for legislation that would restore their former monopoly over printing. See Primary Sources on Copyright (1450–​1900). Available at www.copyrighthistory.org. 6 For sources on French copyright history see Primary Sources on Copyright (1450–​1900). See www. copyrighthistory.org. In France, a system of book-​privileges had served to control the printing of books since 1498. By an act of 1547 the granting of privileges was linked to the exercise of prepublication censorship (Edict fait par le Roy très-​chrestien Henry deuxieme de ce nom, sur les livres censurez par la faculté de Theologie de Paris.) Thus the Paris Book Guild (Chambre Syndicale de la Librairie et Imprimerie de Paris)—​which held a royal privilege and monopoly to book printing in Paris—​operated as a prepublication censor in the seventeenth and eighteenth centuries. (Lettres Patentes du roi, pour le reglement des libraries, imprimeures & Relieurs de cette ville de Paris 19 juillet, 1618.) In 1777 a royal decree limited the duration of printing privileges, thereby restricting the control of booksellers over the book market. By the same decree authors’ privileges were made perpetual as long as they were not transferred. (Arrest du conseil d’état du roi. Portant Réglement sur la durée des Privilèges en Librarie du 30 Août 1777.) 7 Today article L. 121–​1 of the French Intellectual Property Code defines the right of integrity as follows: ‘L’auteur jouit du droit au respect de son nom, de sa qualité et de son oeuvre. Ce droit est attaché à sa personne. Il est perpètuel, inaliénable et imprescriptible. Il est transmissible à cause de mort aux héritiers de l’auteur. L’exercise peut être conféré à un tiers en vertu de dispositions testamentaires.’ 8 See also Trib.comm.Seine, 29 déc. 1842; Trib.comm.Seine, 16 sept. 1858; Cour de Lyon, 23 juin 1847, D. 1847. 2. 152; Trib.civ. de la Seine 14 déc. 1859. D.1860.3.16 and Trib. Civ. de la Seine, 14 mars, 1860. D.1860.3.16. 9 ‘Au point de vue de sa personnalité morale et dans l’intérêt même de ses doctrines, l’auteur doit toujours conserver le droit de revoir et de corriger son oeuvre, d’en surveiller la fidèle reproduction, et de choisir le moment et le mode de la publication’ (128). 10 ‘Qu’en effet, l’orateur livre seulement sa parole, sans donner le pouvoir de disposer de sa pensée à l’aide de l’impression; qu’il lui importe, au contraire, de conserver le fruit de son travail, de rester juge de l’opportunité de sa publication, et de se mettre en garde contre une altération dangereuse’ (129).

208

209

Information in a copyright perspective 11 Carla Hesse has studied how among other the French Enlightenment thinker Condorcet imagined ‘an authorless world of free manipulation and circulation of information and ideas’ where the book market was one of ‘ideas rather than authors, substance rather than style’ and structured ‘according to the principles of periodical rather than book publishing, as was the publication of proceedings of the Académie des sciences or the Encyclopédie: through reader subscription to a genre of knowledge rather than through the marketing of unique works’ (Hesse 1991: 116). See also Condorcet (1847). 12 The concept of ‘non-​consumptive research’ is comparable to the concept of ‘non-​consumptive reading’ used in Digital Humanities. See Schreibman (2014). 13 § 107 Limitations on exclusive rights: Fair use. Notwithstanding the provisions of sections 106 and 106a the fair use of a copyrighted work, including such use by reproduction in copies or phonorecords or by any other means specified by that section, for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright. In determining whether the use made of a work in any particular case is a fair use the factors to be considered shall include—​(1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes; (2) the nature of the copyrighted work; (3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and (4) the effect of the use upon the potential market for or value of the copyrighted work. The fact that a work is unpublished shall not itself bar a finding of fair use if such finding is made upon consideration of all the above factors. 14 As will be discussed, Pierre N. Leval, who had been the judge of numerous copyright suits where fair use claims had been made, published a game-​changing article in the Harvard Law Review. The article defined a number of governing principles of the fair use doctrine that, in 1994, were applied by the US Supreme court in the decision Music Campbell v. Acuff-​Rose Music (92–​1292), 510 U.S. 569 (1994).

References Alexander, I. and Tomas Gomez-​Arostegui, H. (eds.) (2016) Research Handbook on the History of Copyright Law, Cheltenham: Edward Elgar Publishing. Armstrong, E. (1990) Before Copyright: The French Book-​Privilege System 1498–​1526, Cambridge: Cambridge University Press. Barlow, J. P. (1994) ‘The Economy of Ideas: A Framework for Patents and Copyrights in the Digital Age. (Everything You Know About Intellectual Property Is Wrong)’, Wired 2 (3): 85–​129. Condorcet, M. C. (1847) ‘Fragments sur la liberté de la presse (1776)’, in M.-​F. Arago (ed.), Oeuvres de Condorcet, Paris: Didot: 253–​314. Coombe, R. (1998) The Cultural Life of Intellectual Properties: Authorship, Appropriation, and the Law, Durham: Duke University Press. Deazley, R. (2004), On the Origin of the Right to Copy: Charting the Movement of Copyright Law in Eighteenth Century Britain (1695–​1775), Oxford: Hart. Defoe, D. (1705) A Review of the Affairs of France: And of All Europe, as Influence’d by That Nation, vol. 2. London. Drucker, J. (2014) ‘Knowledge Design: A Conceptual and Curricular Challenge’, Design and Culture 6 (1): 65–​84. Edelman, B. (1983) ‘Création et banalité’, Recueil Dalloz Sirey (Chronique): 73–​77. Feather, J. (1994) Publishing, Piracy and Politics: A Historical Study of Copyright in Britain, New York: Mansell Publishing. Foucault, M. (1989) ‘What Is an Author?’, in R. C. Davis and R. Schleifer (eds.), Contemporary Literary Criticism, New York, London: Longman: 263–​275. Hayles, K. N. (2012) How We Think: Digital Media and Contemporary Technogenesis, Chicago: Chicago University Press. Hesse, C. (1991) Publishing and Cultural Politics in Revolutionary Paris, 1789–​1810, Berkeley: University of California Press. Kant, I. (1785) ‘Von der Unrechtmäßigkeit der Büchernachdrucks’, Berlinische Monatsschrift 5: 403–​417. Karnell, G. W. G. (1989) ‘The Idea/​Expression Dichotomy—​A Conceptual Fallacy’, Copyright World 7: 16–​17. Lessig, L. (1999) Code and Other Laws of Cyberspace, New York: Basic Books. Leval, P. (1990) ‘Toward a Fair Use Standard’, Harvard Law Review 103: 1105–​1136.

209

210

Stina Teilmann-Lock Loewenstein, J. (2002) The Author’s Due: Printing and the Prehistory of Copyright, Chicago: The University of Chicago Press. Samuels, E. (1989) ‘The Idea-​Expression Dichotomy in Copyright Law’, Tennessee Law Review 56: 32–​462. Schreibman, S. (2014) ‘Non-​Consumptive Reading’, in N. Segal and D. Koleva (eds.), From Literature to Cultural Literacy, London: Palgrave Macmillan, 148–​165. Teilmann-​Lock, S. (2016) The Object of Copyright: A Conceptual History of Originals and Copies in Literature, Art and Design, London: Routledge. Thylstrup, N. B. (2018). The Politics of Mass Digitization, Cambridge, MA: MIT Press. Thylstrup, N. B. and S. Teilmann-​Lock (2017) ‘The Transformative Power of the Thumbnail Image: Media Logistics and Infrastructural Aesthetics’, First Monday 22 (10).

210

211

12 INFORMATION IN THE PURSUIT OF SOCIAL REFORM Lynn McDonald

Introduction: the need for adequate information in the pursuit of social and democratic reform Long before the existence of Information as a field of study or any sense of an “information revolution” occurring (Weller 2008), some of the greatest contributors to social and political reform gave considerable attention to empiricism, both to its philosophical foundations or epistemology and to practical research methods. By “empiricism” is meant the use of real-​world data in the social and natural sciences, such that if the data contradict a theory, the theory is wrong. The term “methodology” will be used here to embrace both components of empiricism. The time span runs from the 17th century to the early 20th century. Eight British writers are at least briefly noted: John Locke, Mary Astell, Catharine Macaulay, David Hume, John Stuart Mill, Harriet Martineau, Florence Nightingale and Beatrice Webb. Five French writers and one French-​speaking Belgian are, again at least briefly, discussed: Voltaire, Saint-​Simon, Denis Diderot, Condorcet, Germaine de Staël and L.A.J. Quetelet. This chapter, then, covers the emergence of scientific method and its use in social and political reform, through the great Enlightenment writers, both British and French, to the articulation and employment of empirical methods in numerous works of social science and social and political advocacy in the nineteenth and early twentieth centuries. The writers chosen for discussion include women who are still scarcely known in the political science, sociology or philosophical literature, but all of whom published substantial books and papers and were well recognized in their day. They have been called “missing persons” in the development of social science methodology (McDonald 1994: ­Chapter 5; see also McDonald 1993). It is gratifying to see 21st-​century sociologists coming to challenge their exclusion, to protest also the exclusion of non-​white writers—​Eurocentrism and androcentrism (Alatas and Sinha 2017: 1–​16). They describe the history of Sociology and sociological theory as “steeped in androcentrism,” listing and reproducing the contributions of male scholars, whose work dominated (and still dominates) course material and textbooks (Alatas and Sinha 2017: 8–​9). They advocate the incorporation of women’s scholarship, especially in discussion of comparisons and contrasts among theorists.

211

212

Lynn McDonald

In this chapter, two writers are pursued in some detail: Germaine de Staël, of the French Enlightenment, French Revolution and the First Empire, for her advocacy of a precise “geometric” political science; and Florence Nightingale, a British 19th-​century figure, for her foundational work in applying statistical methods to health care advocacy. Both women were and remain well known, but largely for their other achievements, Staël for her novels and literary commentary, and Nightingale as the major founder of the modern profession of nursing and heroine of the Crimean War, 1854–​6. That war has been described as the first “media” war, as the first to have had journalists reporting back by telegraph, and photographers providing vivid illustrations of the war’s horrors.

Requisites for advances in social science Advocates of democratic institutions and various social reforms need high quality information to advance their concerns. Five key components are readily identifiable as important for these goals: 1 A methodology (empiricism), both its philosophical base and practical research methods and statistics, as a source for data collection in the real world, as opposed to recourse to accepted authorities, be it Aristotle, Thomas Aquinas, the Vatican or the Faculty of Theology at the Sorbonne. 2 An adequate educational level among the citizenry to make informed discussion possible, meaning state support for education, for women as well as men. 3 Availability of quality data as a basis for policy deliberations: a reliable Census, regular production of vital statistics and other political, social, economic, health and environmental data. 4 Freedom of expression, especially a free press, to ensure that new and controversial ideas can be expressed and debated; freedom of religion, in some times and places, would also be essential. 5 Quality newspapers and journals, desirably with investigative journalists and access by researchers through letters to the editor and comments; public policy requires public input, and royal commission reports and regular reports from government departments need scrutiny in civil society; many of the great thinkers also produced short, accessible, opinion pieces. Those are the positives. In the process of examining information, we must also be sensitive to issues of sexism and sex-​role stereotyping in the choice of what theorists deserve inclusion and prominence, and how their work is reported. The contributors discussed here made the case for various of the five components. It should be noted that, along with their largely progressive orientation, there were also some significant lapses. The exclusion of women in education and civil life was acceptable to many. Locke considered slavery justifiable. However, all made significant contributions both to methodology, in its foundations or the articulation of research methods or both, and to core social and political reforms advancing democratic measures, what would come to be called “human rights,” and early steps to the welfare state. All saw the need for reliable, accurate, information as essential to their reform goals. The philosophers discussed here, and many more, were advocates of the great principles of the 18th-​century Enlightenment: liberty of expression, a free press, sovereignty of the people (against the rights of the monarch), the right to representation, extension of the right to vote 212

213

Information in the pursuit of social reform

to non-​property holders, working-​class men and (all) women, the right to education, freedom of religion, freedom from torture and an end to slavery. A key unifying theme is sympathy, that is human-​to-​human sympathy, sometimes extended to the “whole sensitive creation,” as Jeremy Bentham (1970 [1780]) called it. This resulted in a social bond and obligations to fellow citizens, which could be extended to all nations, for a ground-​up society, as opposed to a top-​ down, in the divine right of kings.

Major sources on methodology The next three sections set out the major sources, for the 17th century, 18th-​ century Enlightenment and the 19th century, with the emergence of detailed social science research. For Staël and Nightingale, their brief discussion here is followed up with a more detailed comparison below.

The 17th century, emergence of scientific method John Locke (1632–​1704): An Essay concerning Human Understanding, 1690a, the major source on the foundations of empiricism, used by all the important Enlightenment writers, and containing a theory of education based on the “blank slate” (tabula rasa), as opposed to innate ideas; his Two Treatises of Government (1960 [1690b]), were a defence of liberal social contract theory, as opposed to Thomas Hobbes’s authoritarian version, and A Letter concerning Toleration, 1689, advocated religious toleration when torture, execution and war were often the means of settling religious disputes. Hobbes’s social contract theory (1968 [1651]) did not allow subjects to choose their own religion, but each must accept the sovereign’s. Locke’s theory accorded power to the people over the monarch, and called for a separation of powers. Mary Astell (1668–​1731): her A Serious Proposal to the Ladies for the Advancement of their True and Greatest Interest, 1694, was a plea for education for women, based on Locke’s “blank slate” concept, insisting that (the apparent) inferiority of women in intellectual pursuits was not the product of an inferior brain, but lack of education, that they might achieve as much as men if given the same experience, as both started with a blank slate; Some Reflections Upon Marriage, 1700, set out the numerous injustices women face in marriage, and asked “if all men are born free, how is it women are born slaves?” (Astell 1730 [1700]: 107–​8). In The Christian Religion, 1705, she advanced such Enlightenment principles as responsibility for one’s neighbour and the collective good, in an age of growing individualism. Astell published a pamphlet supporting inoculation against smallpox, which another underestimated woman author, Mary Wortley Montagu, had learned of when in Turkey, and advocated its use back home. This is an early demonstration of evidence-​based health care advocacy. Astell’s promotion of it shows her familiarity with the limited social data available in her time, the “bills of mortality,” which gave comparative mortality rates. One in six people with smallpox died, but only one in 50 of those inoculated, she reported (Astell 1724: 241).

18th-​century Enlightenment, French and British Voltaire (1694–​1778), in Lettres philosophiques, 1733, published in English as Letters concerning the English Nation, promoted diversity, even respect for Quakers and other “sects,” and flagged the methodological advances of Bacon and Locke, on observation and scepticism, as opposed to the top-​down system of certain knowledge and censorship of his day. 213

214

Lynn McDonald

David Hume (1711–​76): his An Enquiry concerning Human Understanding, 1748, paid tribute to ancient (constructive) scepticism; his Inquiry concerning the Principles of Morals, 1751, defended utility theory, the greatest good of the greatest number, a key concept in the defence of democratic rights. Denis Diderot (1713–​84): Pensées philosophiques [Philosophical Thoughts], 1746, Lettre sur les aveugles (Letter on the Blind) 1749, both relate Lockean empiricism; as editor of the famous French Encyclopédie, ou dictionnaire raisonnée des sciences, des arts et des métiers [Encyclopedia, or Reasoned Dictionary of the Sciences, Arts and Occupations], Diderot covered a large number of reforms, such as the abolition of torture, written by Voltaire; opposition to capital punishment; the end to slavery and forced labour for peasants; and for freedom of religion and expression; promotion of health care, including for the mentally ill; education, job creation, support for the poor and women’s rights. Catharine Macaulay (1731–​91): her Letters on Education, 1790a, were based on Lockean empiricism, which she used to argue for education for women, as had Astell earlier. The book was also a plea for civil liberties and opposition to the “deification” of government, with (brief) material on environmental concerns. Macaulay drew on Hume and Smith on the origins of ethics. Her eight-​volume History of England, 1763–​83, was a liberal, republican history, in contrast with Hume’s earlier (1754–​62) conservative history; hers defended the “Glorious Revolution” of 1688, which overthrew the divine-​right King James, and justified his execution; her Loose Remarks on Hobbes’s Philosophical Rudiments, 1767, refuted his authoritarian social contract; “Short Sketch of a Democratic Form of Government” (1767: 21–​8) outlined the principles for a new, democratic constitution for Corsica, when it seemed it would become independent (it did not), a document she shared with George Washington. Her “Address to the People of England, Scotland and Ireland,” 1775, was a defence of the American Revolution; her Observations on the Reflections of the Rt Hon Edmund Burke on the Revolution in France, 1790b, was an early rebuttal to his attack on the French Revolution. Macaulay was well known in her day but did not make it into “the canon” of political philosophy. Ironically, Mary Wollstonecraft would later be admitted, if marginally, yet it was Macaulay’s Letters on Education which influenced Wollstonecraft to become an advocate for women, which she did in Vindication of the Rights of Woman, 1792. Jean le Rond Caritat, marquis de Condorcet (1743–​94), a mathematician, contributed entries on natural science to the French Encyclopedia. His probability theory allowed that, while we could never have grounds for certainty, we could know how firm are our grounds for belief, in Elémens du calcul des probabilités [Elements of the Calculation of Probabilities], 1795. He also advocated the establishment of a national data bank; his “social mathematics,” or a mathematically advanced social science,” would contribute to “the happiness and perfecting of the human species.” His “Report on the General Organization of Public Instruction,” 1793, advocated public education, girls included. He opposed capital punishment, and voted against it for Louis XIV. His (posthumous) Esquisse d’un tableau historique des progrès de l’esprit humain [Sketch for a Historical Picture of the Human Mind] tracks the development of the human mind, notably on methodology, insisting on the same, probabilistic, status for social science as for natural science, with attention to the role of observation in theory development and such related institutions as education, publishing and learned institutes. He was one of, if not the first, writer to use the term “social science.” Condorcet had the advantage of marriage with a philosophically astute wife, Sophie Grouchy de Condorcet (1764–​1822) (McDonald 1994: 131–​4). Henri (comte de) Saint-​Simon (1760–​1825) gave sociology and social policy some of its basic vocabulary, “social physiology” and “industrialization,” in Introduction des travaux scientifiques du 214

215

Information in the pursuit of social reform

dix-​neuvième siècle [Introduction to Scientific Works of the Nineteenth Century], 1807, and Social Physiology, 1813. Saint-​Simon influenced both mainstream social scientists and Karl Marx. He advocated the use of social science for social welfare and international co-​operation, using “society” in a broader sense than the nation-​state, to embrace “European society,” realized in the European Union. Germaine de Staël (1766–​1817) produced several important works of social science, several novels and much literary criticism. Her Influence of the Passions upon The Happiness of Individuals and Nations, 1796, with English translations in 1798 and 1813, was a work of social science. Building on Adam Smith’s Theory of Moral Sentiments, 1759, it stressed the importance of emotions in society, particularly in the framing of morality and laws. Her Circonstances actuelles [On the Present Circumstances that Can End the Revolution and the Principles which must Found the Republic in France] written in 1798, but not published until 1906, and never in English translation, contains much advocacy of the use of political science, as well as her considered analysis of what went wrong in the French Revolution. Her On Literature and its Relations with Social Institutions, 1799, again is social science—​including philosophy and science in “literature.” Her 3-​volume Considerations on the Principal Events of the French Revolution, posthumously published in 1818, was a history, but initially written in defence of her father, Jacques Necker, who had been the King’s minister of finance. Her De l’Allemagne [On Germany] has much comparative analysis of social and political institutions, published in London first, in both French and English, in 1813 and 1814, in France not until after the fall of Napoleon.

19th-​century social physics and social science L.A.J. Quetelet (1796–​ 1874), a Belgian astronomer and expert in probability theory, first published on “social physics” in French in 1835, Sur l’homme et le développement de ses facultés: physique sociale [On Man and the Development of his Faculties: Social Physics], which soon appeared in English as A Treatise on Man, 1842; he reversed the title and sub-​title for a much expanded French edition in 1869, Physique sociale, ou Essai sur le développement des facultés de l’homme. Quetelet’s social physics takes up the project of Condorcet and Staël to operationalize variables, pioneering what would become “the standard demographic variables”; he was prominent in the design of the first modern censuses, wrote on the causes of crime, exposing the faults of the usual “hard on crime” explanations, instead to advocate addressing the causes of crime. He explored the connection between exposure to sunlight and mental difficulties. Nightingale was so impressed with his analysis that she sought to have a chair on social physics created at Oxford University, the university that trained the most public servants and politicians—​the people she thought most needed social data for making policy (McDonald 2003: 105–​28). She regretted that an “enormous amount of statistics” sat in people’s “pigeon holes”: Cabinet ministers, army officials, members of Parliament, the majority of whom had received a university education. Unlike Staël, Nightingale did not want “a great arithmetical law,” but “to know what we are doing in things which must be tested by results” (Letter 3 January 1891, cited in McDonald 2003: 109–​10). Quetelet’s Social Physics influenced her notably on his point that the administration of a hospital did more to determine outcomes than the nature of the treatment given, this at a time when hospitals had very high death rates. Quetelet was well known in Britain and throughout Europe through his work at the International Statistical Congress; he presided over its meetings in London in 1860. Karl Marx, at the other end of the political spectrum, was another theorist who used his work (Bottomore and Rubel 1961: 234–​5). 215

216

Lynn McDonald

Harriet Martineau (1802–​76): her How to Observe Morals and Manners (1999 [1838a]) could be described as the first book of sociological method, setting out practical research methods, which she employed in her American travels. It predates Durkheim’s Rules of Sociological Method, 1895, by 67 years; her Society in America, 1837, reports her travels in the United States, with a critical analysis of American institutions, notably on women, politics and slavery; “The Martyr Age of the United States” (1838b) was a substantial journal article on slavery and the abolition movement. Her Positive Philosophy of Auguste Comte, 1864, contains both a translation and abridgement of his ungainly Cours de philosophie positive and her commentary on it. He was so impressed with it that he substituted it for his original six volumes, and had it translated back into French! (Comte 1894–​5). Martineau’s Society in America is much better researched and more comprehensive than the better-​known Democracy in America, 1835, by Alexis de Tocqueville. Martineau of course had the advantage of being an English speaker (Tocqueville’s English was rudimentary), and she, on account of her poor hearing, travelled with a research assistant. As well, she went prepared for conducting research, as is evident in her how-​to book, How to Observe Morals and Manners. Martineau also assisted Nightingale by producing a short, popular, account of her Crimean War research (see Figure 12.1). England and her Soldiers, 1859, which included Nightingale’s pioneering charts on war mortality, and its reduction. These “polar area” or rose diagrams enormously influenced the presentation of data by government, industry and throughout society. John Stuart Mill (1806–​73): his System of Logic, Ratiocinative and Inductive (1882 [1843]) was a major defence of induction in empiricism; his Principles of Political Economy (first edition, 1848) was the central source on political theory at the time, building on utility theory; On Liberty, 1859, a joint work with his wife, Harriet Taylor Mill, was an eloquent plea for the rights of the individual relative to political authorities; his On the Subjection of Women, 1869, related

Figure 12.1  Diagrams showing the causes of mortality in the Crimean War (April 1854–​March 1856) Source: published in Nightingale (1858b), produced with assistance from Martineau.

216

217

Information in the pursuit of social reform

sympathetically the tyranny routinely experienced by women; Mill was the first person to move an amendment in Parliament to admit women to the vote, on his birthday, 20 May 1867. Florence Nightingale (1820–​1910) acquired her methodological principles from two earlier sources, J.S. Mill on the underlying philosophy and Quetelet on the use of statistics. Her expert analysis of mortality and morbidity data, from the Crimean War appears in her (written) evidence to the royal commission on the Crimean War (1858a), and her comprehensive “confidential” report (1858b), the charts from which she also published in a shorter Contribution to the Sanitary History of the British Army, 1859). The work won her nomination as the first woman fellow of the Royal Statistical Society. She continued to use data in social policy advocacy, such as the failure of repressive laws (against women, suspected prostitutes) to prevent syphilis in the army (Nightingale 1862). Her numerous reports and articles on India use the best available data (Vallée 2007). Beatrice Webb (1858–​1943), in her My Apprenticeship, 1926, recounts how she came, early in life, to applied social science research methods; her Methods of Investigation, 1906, gives a succinct outline for the conduct of research; while a later book, with her husband, Sidney Webb, Methods of Social Study, 1932, gives more detail; her Case for the Factory Acts, 1901, argues for laws to provide safer working conditions and shorter hours of work; her “Abolition of the Poor Law,” 1918, called for an end to England’s punitive approach to the poor (the Poor Law was not ended until 1946). She co-​published much on social reform with Sidney Webb, beginning with Industrial Democracy, 1902 [1897], on trade unions. She and her husband were among the founders of the British Labour Party and the London School of Economics.

Mme de Staël and the pursuit of “geometric” political science Germaine de Staël’s family connections—​her father was the Minister of Finance for Louis XIV and her mother’s salon attracted leading Enlightenment intellectuals—​as well as her marriage and other partners, ensured that she would be a figure of great interest. Those connections would admit Mme de Staël to the great events of the day, but would also lead to her work being demeaned, her views attributed to others or personal motives, and her originality downplayed. The title of a major biography of her begins with Mistress to an Age (Herold 1959). Her confidence in the possibility of useful, applicable, prediction occurs in several of her works. In Influence of the Passions, Staël asserted the predictive possibilities of quantitative social science. Events occurred in fixed proportions when the numbers were large. It was not possible to predict what any individual would do, but it was possible to predict percentages of regularly occurring events, such as births, deaths, crime and suicide, as events occurred in fixed proportions. Thus one could predict the number of murders in Italy and divorces in Berne—​ her examples—​g iven their stable numbers over the preceding decade (Staël 1813 [1796]: 7). Nations, similarly, could be studied with regard to their birth, history and demise. Political science would one day have the “force of geometric evidence” (Staël, 1813 [1796]: 7). She also used the term “human sciences.” For these points, Staël drew on Montesquieu’s Spirit of Laws, 1748, as well as Smith’s Theory of Moral Sentiments (Staël 1813 [1796]: 312). In The Influence of the Passions, Staël described revenge as the passion destructive of happiness, and “even the existence of free countries.” France had to overcome this animosity, to prevent new revolutions (Staël, 1813 [1796], 165). France needed reconciliation; it could only be saved by “mutual forgiveness,” meaning by both royalists and republicans (Staël 1813 [1796]: 166). The reaction to the Revolution would not end until persecution ended. In On Literature in Relation to Social Institutions, Staël looked at the relationship between science, technology and social institutions. Here she expressed concern, that, with scientific 217

218

Lynn McDonald

progress, further moral or social progress was needed. Increasing “human power” made it necessary “to strengthen the brakes that prevent its abuse.” Scientific progress made “political progress” necessary. More enlightened government entailed “greater respect for public opinion” (Staël 1871: 199). In On Literature, as well, Staël affirmed the causal influence of emotions in shaping moral views. To guide the new republic, Staël saw social science as essential. The term “social art” had been in use for some time. Now she asserted the need for a social science, “science morale,” or the “science of moral happiness” (1813 ed. 30) to ground the “social art.” With the benefit of hindsight, her ambitions seem naive, that disputes could be settled by “calculation,” passions by reason. Mistakes were inevitable, she acknowledged; even “extravagant excesses,” so that it was impossible that the first leaders should preserve their power (Staël 1813 [1796]: 197). Yet it was essential that France not go back to its royalist past; France must remain a republic. The “calamities” of the revolution were over, so there could be hope for the future (Passions: 28–​30). However, as we know, after Napoleon’s final defeat, the Bourbon monarchy was restored. Two kings reigned before Louis-​Napoleon, a nephew (or not) of Napoleon Bonaparte, made himself “emperor of the French,” now the Second Empire, and he lasted until the German defeat of France in the Franco-​Prussian War, 1870–​1. Just as Descartes applied algebra to geometry, Staël observed, it was now necessary to apply calculation to politics. Moreover, when this was achieved, political quarrels would cease: Human passions are as susceptible to calculation as the movements of machines, given a certain number of cases the repetition of the same events is certain. The passions of a nation can then be calculated by a legislator, as are births, deaths and marriages. (Staël 1813 [1796]: 27) Staël’s confidence in “geometric” social science being applicable to real-​life problems came out even more strongly in Circonstances actuelles. There was great advantage to founding government on “geometrically true principles,” Staël held, for they could “bring the peace of demonstration to the most terrible cause of war.” She looked to nations adopting the “same theory of government resembling each other in political associations, regardless of their diversity in development and their differences in individual morals” (Staël 1979 [1798]: 27). She likened Montesquieu, a social theorist, to the physicist Isaac Newton, who “discovered a theory which serves as a guide for experiments and which in turn confirms them.” Montesquieu, no less, “came to examine the laws in all the combinations destiny brought.” He explained motives, and gave a reason for all chance happenings: “Theory without experience is nothing but a phrase; experience without theory is but prejudice” (Staël 1979 [1798]: 32). Laws, as in laws of nature, cannot be reversed. “Injustice comes to an end as a rock falls” (Staël 1979 [1798]: 134). “For a long time the art of governing the republic was a military science. But today we must consider opinion.” Now the concern is commerce and the public debt, not military power (Staël 1979 [1798]: 135). The Present Circumstances was written after the Revolution, and contains analysis of what went wrong in it. Like the British feminist writers of the time who discussed the Revolution, Catharine Macaulay (1790b) and Mary Wollstonecraft (1794), Staël blamed the faults of the ancien régime for the horrors that ensued. The republic was created before the enlightenment required to prepare for it. If a constitutional monarchy had lasted ten years, a republic could have been achieved “by consent,” without bloodshed (Staël 1979 [1798]: 35). Instead, there was recourse to criminals. 218

219

Information in the pursuit of social reform

She asked: “Can the majority impose highly despotic laws on the minority?” one might observe a more useful question than recourse to the general will. She then observed that “With the furor of parties, the majority and the minority have become … two different nations, or rather, the slaves and the oppressors” (Staël 1979 [1798]: 252). In Present Circumstances, Staël held that people had to be attracted to the new regime, not threatened by it. “Moral force” was needed, and a stake in the economy (Staël 1979 [1798]: 201). “Prosperity is the price of republic,” she affirmed (252). She sought the “requisites” of stable democracy, calling for “enlightened republicans” (135). Staël sought political equality, equality before the law, but not a levelling. Aristocratic privileges should be abolished for a merit system. But a levelling of wealth would not necessarily provide for poverty, while “natural” inequalities, as in ability, would enrich the whole society, through talented enterprise. She deplored the lack of freedom of the press, asking: What power do governments not have, when freedom of the press is suspended and yet publications continue every day? In despotic governments, no freedom of the press is allowed, but all are silenced, governed and governors. Opinion is left to fend for itself. But in France, all the newspapers argue, invent, make false accusations in one direction as if that were permitted in free countries. (Staël 1979 [1798]: 201)

Abolition of the slave trade After the (final) defeat of Napoleon, while a peace treaty was being negotiated in Paris, Staël took advantage of the meetings to call for the abolition of the slave trade: “An Appeal to the Sovereigns Convened in Paris to Grant the Abolition of the Slave Trade” (Staël 1994 [1814]: 157–​9). Slavery itself was no longer allowed in any European country, but Europeans could make money transporting slaves to the West Indies, and the French and Portuguese profited greatly from this. Le Havre, Nantes and Bordeaux were the chief French examples. Liverpool and Bristol were the top English examples until Parliament abolished the slave trade in 1807, or seven years before the congress. Staël now sought Europe-​wide abolition. She was familiar with the British abolition movement and had met William Wilberforce, its leader, while she was in exile in England. She wrote an introduction for a translation of an anti-​slavery book of his (Staël 1994 [1814]: 159–​62). As social reformers typically do, Staël emphasized the feasibility of the reform—​it would do good without causing undue harm. In the case of the slave trade, she argued, abolition had had “no effect on the prosperity of the English colonies. The blacks have multiplied sufficiently to do the necessary work.” Thus millions of people and whole nations were saved “from all manner of wrongs, without commerce suffering financial losses.” She gave a vivid description of the suffering endured by the slaves in their capture and transportation. She quoted numbers, citing William Pitt, the anti-​slavery prime minister, of 80,000 slaves transported through the U.K. every year (Staël 1994 [1814]: 158). She even suggested that America owed it to England, which had resisted Napoleon alone for ten years, then finally conquered him. Thus France should adopt “the great act of humanity she [Britain] recommended to all the governments of Europe.” Staël ended the Appeal on a stirring note: “There is no country on earth which is not worthy of justice.” Staël was bold enough to raise the slavery issue with Thomas Jefferson, a slave owner, by then past president of the United States. She had known him from his time in Paris, 1783–​89, 219

220

Lynn McDonald

when he was ambassador for the United States. The letter in question declared no less than: “If you should succeed in destroying slavery in the south there would be at least one government in the world as perfect as the human mind can conceive” (Letter 6 January 1816, in Kimball 1918: 70). It was a spirited correspondence, neither convincing the other. Information, clearly, can fail to persuade, especially when there are great interests at stake.

Other correspondence with Thomas Jefferson Most of Staël’s correspondence with Jefferson concerned Napoleon. In a letter in 1812, when in exile in Stockholm, she told Jefferson that he confused tyranny with liberty, so that people re-​established “the old despotism” while professing to be liberal. She sought a third way, a constitutional, liberal, republic, like that of the United States: “For ten years, England has been the sole barrier against this singular despotism which unites all the means barbarism and civilization can furnish to debased humanity.” If England succumbed to Napoleon, she warned, Napoleon would turn on America. Napoleon was the “emperor of France or rather of Europe” (Letter 10 November 1812, in Kimball 1918: 66). Staël did not profess to know the circumstances that gave rise to the differences between America and England, but took a broader approach. She recalled Jefferson’s prediction, when in Paris at the outbreak of the French Revolution, that “demagogic principles would lead to despotism.” This was fulfilled: “Europe and the human race are bent under the will of a single man who wished to establish a universal monarchy,” even the “conqueror of the earth.” Already Germany, Italy, Holland and Denmark were its provinces. Napoleon had early offered to England to make America an “appanage of an English prince,” which England rejected. Jefferson had said that America had nothing to do with Europe; she asked how he could be indifferent, as the most republican country of all (Letter 10 November 1812, in Kimball 1918: 66). Jefferson countered by referring to England’s tyranny: its object was “the permanent dominion of the seas” and “monopoly” of world trade. Its king, “poor manic” George III, would die, but a nation lived (Letter 28 May 1813 in Kimball 1918: 67). Staël understood that the French republic was a world first; the American Revolution had established a new country, a republic, but George III remained on the throne in the U.K. and its possessions. She looked, ambitiously, to a future when information would play a key role in government. As to her hopes, governments clearly have become great creators and users of data. Their departments routinely publish vast amounts of statistical data. They commission opinion polls and conduct focus groups to the end of winning public support for their policies. Auditors and special departments produce comprehensive reports reviewing the above. But nothing that has developed in the intervening centuries has come close to the “geometric” certainty that Staël predicted. Staël was both a political polemicist and a social scientist, and went back and forth without apology between the two fields. Some of her arguments were simply moral—​r ight and wrong—​ as on slavery and executions. Some were pleas for the new methodology—​that practical reforms for human betterment could be made by careful research. These claims were based on faith—​ she had no successful examples to cite. Enlightenment optimism did lead to advances, made later, notably by Florence Nightingale in the 19th century.

Florence Nightingale: Information for social reform Florence Nightingale became a legend from her nursing and the hospital improvements she made during the Crimean War (McDonald 2010b). Exaggerated reports of her efforts 220

221

Information in the pursuit of social reform

flourished, so that she was said to have personally, virtually single-​handedly, brought down the enormous death rates at the war hospitals by her nursing and other reforms. In practice, she credited the Sanitary and Supply Commissions with the great reduction of mortality, rightly, it is clear, with the benefit of later research. Nor did she collect the hospital data herself, as has often been stated, but she used the statistics collected by the War Office. No doubt she is worthy of an “assist,” but a vast number of secondary sources are wrong, including both nursing academics and military historians. Post-​Crimea, Nightingale became an astute user of data and an influence on improved data collection and presentation. She paid careful attention to the Census, and even tried to get questions on health status and housing added to the 1861 Census (she failed) (McDonald, 2003, “Proposals for the 1861 Census”: 95–​103). Her landmark reports of 1858 on the Crimean War have already been noted above. She continued to conduct quantitative studies (McDonald, 2010a: Table 9.1). As well as being an avid researcher, Nightingale was a convinced believer that reports are not “self-​executive,” but require what would later be called a media campaign to win support for them. This meant material for political leaders, experts and the general public. Getting the right reviewer for the right journal was important. Lengthy and complex reports, like those of a royal commission, require short, accessible, versions for a wide readership. Throughout her working life, Nightingale had an excellent relationship with the most influential newspaper in the U.K., The Times, whose reports from the Crimean War brought her fame. Post-​war, she took full advantage of her reputation, sending it letters to the editor on her various projects, which they published. She and her team of experts understood that visuals were important to convey information. To persuade government to spend money on health care in the army required persuasive information that better results could be obtained, in lower rates of hospital admissions and deaths by changed practice. She and Dr William Farr, superintendent of statistics and a leading medical statistician, produced “polar area charts” that made statistical history. They showed how death rates went down, month by month, when sanitary reforms were put in place. In her numerous reports and articles on India, Nightingale used individual stories and illustrations and advocated the use of both to colleagues. She was an avid promoter of “magic lantern” shows, meaning slide shows. She sent a projector to use in lectures to soldiers during the Crimean War (“Turkey” 1856). Later, in an article promoting village sanitation in India, she urged the use of slides to illustrate the cholera bacillus, “the noxious living organisms in foul air and water.” Seeing the slides would inspire villagers to strenuous efforts at clean-​up (Nightingale 1891, in McDonald 2010a: 103; Letter to the Joint Secretaries, in Vallée 2007: 363). Nightingale and her team continued to promote improved health among the army at home, in barracks, where the death rates were twice those of men of comparable age, although soldiers, who had to pass a physical, should have been healthier. The death rates of soldiers in home barracks were in fact radically reduced. On the death of the leader of army reforms, Sidney Herbert, she published bar charts showing the reductions his work achieved, Army Sanitary Administration and its Reform under the late Lord Herbert, 1862. A few years after the war, Nightingale persuaded the senior war minister, the Duke of Newcastle, by then Secretary of State for the Colonies, to have a quantitative study done of disease and death in colonial, aboriginal, schools and hospitals. He had her questionnaires circulated, and she wrote up the results for the National Association for the Promotion of Social Science. Her paper showed disease and death rates roughly double what they should be, that is, of similar people not in an institution (Nightingale 1863). The data she was able to collect were of poor quality, but she thought adequate to draw at least rough conclusions and start the 221

222

Lynn McDonald

discussion. She was not successful even in getting the ministry to continue to collect data. She did get wide press attention to the issue, and to the question of the “disappearance” of aboriginal populations (Nightingale 1864). She made the strategic decision to concentrate her work after that on India, where she had better chances of effecting change. Nightingale and other reformers also encouraged the development of credible information sources on India, which was then ruled from London by the British government. When civil society organizations emerged with Indian national leaders, she, with other progressives, encouraged them. She began to publish articles in their journals, such as the Transactions of the Bengal Social Science Association and the Quarterly Journal of the Poona Sarvajanik Sabha (Vallée 2007). Nightingale seldom published anonymously, but did on the delicate subject of sexually transmitted disease and prostitution. She opposed the highly sexist Contagious Diseases Act of 1862, which subjected women suspected of venereal disease to compulsory treatment in a “lock hospital,” nothing done to bother the men. Her “private and confidential” paper for the government was printed without her name in it (Nightingale, 1862b). Later, when Dr Elizabeth Garrett defended the compulsory measures, Nightingale published two anonymous articles in the Pall-​Mall Gazette, using the pseudonym “Justina.” Here she compared data for different years and types of hospital, calling Garrett’s statistics “worthless” (Nightingale 1870a and 1870b). The Contagious Diseases Acts were repealed in 1886, after much agitation led by Josephine Butler, Nightingale assisting in the background. Comparing the contributions of Nightingale and Staël, the great difference in time is obvious. Staël looked to the use of quantitative data to solve problems and achieve reforms in the future, with excessive confidence in the results possible. Nightingale, living when much progress had been made in methodology and actual data collection, could actually achieve results her predecessor sought. She, too, sometimes, exaggerated the possibilities of application, but she did see enormous decreases in morbidity and mortality rates, and problems solved in many areas by using what would later be term evidence-​based health care.

A cautionary conclusion The fact that women theorists who produced excellent work have been omitted from the “canon” in university teaching and books on theory shows the ongoing influence of sexism. Numerous instances are apparent, affecting the treatment of all women theorists, the classical theorists most of all (McDonald 1993, 1994 and 1998). In recent years, authors of textbooks in sociological and political theory have tended to add “women theorists,” presumably to show inclusion and not be labelled as sexist, but without doing the necessary research to find appropriate contributors. Thus there are books on “classical theory” with (brief) comments on late 20th-​century or early 21st-​century women, for example ethnomethodologist Dorothy Smith and Simone de Beauvoir, for The Second Sex (Longhofer and Winchester 2017) and queer theorist Judith Butler (Ritzer and Goodman 2004), novelist Nellie McClung and racial integration activist Rosa Parks (Bratton, Denham and Deutschmann 2009) and birth control advocate Margaret Fuller (Kimmel and Mahler 2007). Still, the real, serious, theorists are omitted. It is difficult to see how change might proceed. Publishers continue to publish sexist textbooks with shoddy coverage of women theorists and authors collect the royalties no matter the quality of the work. So far, students have not protested, but then how can they know about theorists never mentioned in course work or the assigned textbook? Can peer reviewers and 222

223

Information in the pursuit of social reform

publishers’ reviewers demand adequate coverage? Sociological theory remains “the last bastion of sexism in Sociology” (McDonald 2019). Finally, misinformation, “alternative facts” and “fake news” have attained a new notoriety with the American presidential election of 2016. Accurate information, care for sources and attention to sexist bias have always been challenges, perhaps now more than ever before.

Bibliography Alatas, S.F. and V. Sinha (2017) Sociological Theory Beyond the Canon, London: Palgrave Macmillan. Astell, M. (1694) A Serious Proposal to the Ladies for the Advancement of their True and Greatest Interest, London: J.R. Astell, M. (1730 [1700]) Some Reflections Upon Marriage, London: Parker. Astell, M. (1705) The Christian Religion, as Professed by a Daughter of the Church of England, 3rd ed., London: W. Parker. Astell, M. (1724) The Plain Dealer No. 30:239-​48. Bentham, J. (1970 [1780]) An Introduction to the Principles of Morals and Legislation, eds. J.H. Burns and H.L.A. Hart, London: Athlone. Bottomore, T.B. and Rubel, M. eds. (1961 [1956]) Karl Marx: Selected Writings in Sociology and Social Philosophy, Harmondsworth: Penguin. Bratton, J., Denham, D. and Deutschmann, L. (2009) Capitalism and Classical Social Theory, Toronto: University of Toronto Press. Comte, A. (1894‒95) La Philosophie positive d’Auguste Comte condensée par Harriet Martineau, trans. C. Avezac-​Lavigne, 2 vols., 2nd ed., Paris: L’Bahi. Condorcet. J.-​A.-​N. de Caritat (1932 [1793]) “Report on the General Organization of Public Instruction,” in F. de la Fontainerie (ed.) French Liberalism, New York: McGraw-​Hill. Condorcet. J.-​A.-​N. de Caritat (1966 [1794]) Esquisse d’un tableau historique des progrès de l’esprit humain [Sketch for a Historical Picture of the Human Mind], Paris: Editions sociales. Condorcet. J.-​A.-​N. de Caritat (1805) Elémens du calcul des probabilités [Elements of the Calculation of Probabilities], Paris: Royez. Diderot, D. (1746) Pensées philosophiques [Philosophical Thoughts], ed. M. Delon, The Hague. Diderot, D. (1963 [1749]) Lettre sur les aveugles à l’usage de ceux qui voient [Letter on the Blind for Those Who Can See], ed. R. Niklaus. Geneva: Droz. Diderot, D. and J. le Rond d’Alembert (1751‒66) Encyclopédie, ou dictionnaire raisonné des sciences, des arts et des métiers [Encyclopedia, or reasoned dictionary of the sciences, arts and occupations], 28 vols. Paris: Briasson 1751‒72. Durkheim, E. (1938 [1895]) The Rules of Sociological Method, trans. S.A. Solway and J.H. Mueller. New York: Free Press. Herold, J.C. (1959) Mistress to an Age: A Life of Madame De Staël, London: Hamilton. Hobbes, T. (1968 [1651]) Leviathan, ed. C.B. Macpherson, Harmondsworth: Penguin. Hume, D. (2007 [1748]) An Enquiry concerning Human Understanding, ed. P.F. Millican, Oxford: Oxford University Press. Hume, D. (1957 [1751]) An Inquiry concerning the Principles of Morals, Indianapolis: Liberal Arts Press. Hume, D. (1862 [1754-​62]) History of England, 4 vols., Philadelphia: Philadelphia Library. Kimball, M.G. (1918 July) “Unpublished Correspondence of Mme. de Staël with Thomas Jefferson,” North American Review, 208(752): 63‒71. Kimmel, M.S. and Mahler, M. (2007) Classical Sociological Theory, 2nd ed., New York: Oxford University Press. Locke, J. (1955 [1689]) A Letter concerning Toleration, Indianopolis: Bobbs-​Merrill. Locke, J. (1959 [1690a]) An Essay concerning Human Understanding, ed. A.C. Fraser, 2 vols., New York: Dover. Locke, J. (1960 [1690b]) Two Treatises of Government, ed. P. Laslett, Cambridge: Cambridge University Press. Longhofer, W. and Winchester, D., eds. (2017) Social Theory Re-​Wired: New Connections to Classical and Contemporary Perspectives, New York: Routledge. Macaulay (Graham), C. (1767) Loose Remarks on Certain Positions to be found in Mr Hobbes’s Philosophical Rudiments, with a Short Sketch of a Democratic Form of Government, in a Letter to Signor Paoli, London: Johnston, Davies & Dilly.

223

224

Lynn McDonald Macaulay (Graham), C. (1763-​83) The History of England from the Revolution to the Present Time, 8 vols., London: Nourse. Macaulay (Graham), C. (1775) An Address to the People of England, Scotland and Ireland, on the Present Important Crisis of Affairs, London: Edward & Charles Dilly. Macaulay (Graham), C. (1790a) Letters on Education: With Observations on Religious and Metaphysical Subjects, London: C. Dilly. Macaulay (Graham), C. (1790b) Observations on the Reflections of the Rt Hon Edmund Burke on the Revolution in France, London: Dilly. Martineau, H. (1837) Society in America, 2 vols., New York: Saunders & Otley. Martineau, H. (1999 [1838a]) How to Observe Morals and Manners, ed. M.R. Hill, New Brunswick, NJ: Transaction. Martineau, H. (1838b) “The Martyr Age of the United States,” Westminster Review (December, 1838), 1‒44. Martineau, H. (1859) England and her Soldiers, London: Smith Elder. Martineau, H. ed., intro. and trans. (1864) The Positive Philosophy of Auguste Comte, 2 vols., London: Kegan Paul. McDonald, L. (1993) The Early Origins of the Social Sciences, Montreal: McGill-​Queen’s University Press. McDonald, L. (1994) The Women Founders of the Social Sciences, Montreal: McGill Queen’s University Press. McDonald, L. ed. (1998) Women Theorists on Society and Politics, Waterloo: Wilfrid Laurier University Press. McDonald, L. ed. (2003) Florence Nightingale on Society and Politics, Philosophy, Science, Education and Literature, Waterloo: Wilfrid Laurier University Press. McDonald, L. (2010a) Florence Nightingale at First Hand, London: Continuum Publishing. McDonald, L. ed, (2010b) Florence Nightingale and the Crimean War, Waterloo: Wilfrid Laurier University Press. McDonald, L. (2019) “Sociological Theory: The Last Bastion of Sexism in Sociology.” American Sociologist (30 April 2019) DOI: 10.1007/​s12108-​019-​9410-​z Mill, J.S. (1882 [1843]) System of Logic, Ratiocinative and Inductive, Being a Connected View of the Principles of Evidence and the Methods of Scientific Investigation, 8th ed., New York: Harper & Brothers. Mill, J.S. (1848) Principles of Political Economy, With Some of their Applications to Social Philosophy, 2 vols., London: John Parker. Mill, J.S. (1859) with Harriet Taylor Mill. On Liberty, London: John Parker. Mill, J.S. (1869) The Subjection of Women, London: Longmans, Green, Reader & Dyer. Montesquieu, Secondat de. (1848 [1778]) The Spirit of Laws, translated from the French, 2 vols., 7th ed., Edinburgh: Alexander Donaldson. Nightingale, F. (1858a) “Answers to Written Questions Addressed to Miss Nightingale by the Commissioners,” Report of the Commissioners appointed to Inquire into the Regulations affecting the Sanitary Condition of the Army and the Treatment of the Sick and Wounded, London: HMSO, 361‒94. Nightingale, F. (1858b) Notes on Matters Affecting the Health, Efficiency and Hospital Administration of the British Army, London: Harrison. Nightingale, F. (1859) A Contribution to the Sanitary History of the British Army during the Late War with Russia, London: Harrison. Nightingale, F. (1862) Army Sanitary Administration and its Reform under the late Lord Herbert, London: McCorquodale. Nightingale, F. n.d. [ca. 1862] Note on the Supposed Protection Afforded Against Venereal Diseases, by recognizing Prostitution and Putting It under Police Regulation. Private and Confidential. Nightingale, F. (1863) “Sanitary Statistics of Native Colonial Schools and Hospitals,” Transactions of the National Association for the Promotion of Social Science, 475‒88. Nightingale, F. (1864) “Note on the Aboriginal Races in Australia” in Transactions of the National Association for the Promotion of Social Science, 552‒58. Nightingale, F. (1870a) Justina. Letter to the Editor, “The Contagious Diseases Acts,” Pall Mall Gazette (3 March 1870): 3. Nightingale, F. (1870b) Justina. “Miss Garrett on the Contagious Diseases Acts,” Pall Mall Gazette (18 March 1870): 6. Quetelet, L.A.J. (1835) Sur l’homme et le développement de ses facultés: physique sociale [On Man and the Development of his Faculties: Social Physics], Bruxelles. Quetelet, L.A.J. (1842) A Treatise on Man (trans.) R. Knox. Edinburgh. Quetelet, L.A.J. (1869) Physique sociale, ou Essai sur le développement des facultés de l’homme [Social Physics, or Essay on the development of human faculties], 2nd ed., 2 vols., Brussels: Muquardt.

224

225

Information in the pursuit of social reform Ritzer, G. and D.J. Goodman (2004) Sociological Theory, 6th ed., Boston: McGraw-​Hill. Saint-​Simon, H. de. (1807) Introduction des travaux scientifiques du dix-​neuvième siècle [Introduction to the Scientific Works of the Nineteenth Century], Paris: Scherff. Saint-​Simon, H. de. (1965 [1813]) La physiologie sociale: textes: choisis [Social Physiology: Selected Texts], ed. Georges Gurvitch, Paris: Presses Universitaires de France. Smith, A. (1976 [1759]) The Theory of Moral Sentiments, eds. D.D. Raphael and A. Macfie, Oxford: Clarendon. Staël, G. de. (1813 [1796]) The Influence of the Passions upon The Happiness of Individuals and Nations, trans. from the French, London: Henry Colburn. Staël, G. de. (1871) Oeuvres complètes, Paris: Firmin Didot. Staël, G. de. (1994 [1814]) “An Appeal to the Sovereigns Convened in Paris to Grant the Abolition of the Slave Trade,” trans. S. Bell, in D.Y. Kadish and F. Massardier-​Kennney (eds.), Translating Slavery: Gender and Race in French Women’s Writing, 1783–​1823, Kent, OH: Kent State University Press, 157‒59; “Preface to the Translation of a Work by Mr Wilberforce on the Slave Trade,” 159‒62. Staël, G. de. (1820‒21) Oeuvres complètes, 17 vols., Paris: Truettel & Würtz. Staël, G. de. “Appel aux souverains,” Oeuvres complètes 17, 376‒82. Staël, G. de. (1821 [1814]) “Préface pour la traduction d’un ouvrage de M Wilberforce sur La Traite des Nègres” [Preface for the Translation of a Work by M Wilberforce on the Slave Trade of Negroes, in Vol. 17, Oeuvres complètes, Paris: Treutell & Würtz. Staël, G. de. (1998 [1799]) De la littérature considérée dans ses rapports avec les institutions sociales [On Literature Considered in its Relations with Social Institutions], ed. A. Blaeschke, Paris: InfoMédia. Staël, G. de. (1979) [1906] Des circonstances actuelles qui peuvent terminer la Révolution et des principes qui doivent fonder la République en France [On the Present Circumstances that Can End the Revolution and the Principles which must Found the Republic in France], ed. L. Omacini, Geneva: Droz. Staël, G. de. 1813 Germany, trans. from the French, 3 vols., London: John Murray. Staël, G. de. 1918 [posthumous] Considerations on the Principal Events of the French Revolution, 3 vols., London 1918. Translated from the French. Tocqueville, A. de. (1945 [1835]) Democracy in America, 2 vols., trans. H. Reeve, New York: Knopf. “Turkey” (1856, March 5) Times, 9B. Vallée, G., ed. (2007) Florence Nightingale on Social Change in India, Waterloo, ON: Wilfrid Laurier University Press. Voltaire [François-​ Marie Arouet] (1994 [1733]) Letters concerning the English Nation, ed. N. Cronk, Oxford: Oxford University Press. Webb, B. ed. (1901) The Case for the Factory Acts, London: Grant Richards, 1‒38. Webb, B. (1918) “The Abolition of the Poor Law,” Fabian Tract No. 185. Webb, B. (1906) Methods of Investigation, London: London Sociological Society. Webb, B. (1950 [1926]) My Apprenticeship, 2 vols., London: Longmans, Green. Webb, B. and S. Webb (1902 [1897]) Industrial Democracy, 2 vols. in one, London: Longmans, Green. Webb, B. and S. Webb (1975 [1932]) Methods of Social Study, Cambridge: LSE/​Cambridge University Press. Weller, T. (2008) Information History: ​An Introduction: Exploring an Emergent Field, Oxford: Chandos. Wollstonecraft, M. (1845 [1792]) A Vindication of the Rights of Woman, With Strictures on Political and Moral Subjects, New York: Vale. Wollstonecraft, M. (1794) An Historical and Moral View of the Origin and Progress of the French Revolution, London: J. Johnson.

Further reading McDonald, L. (1994) The Women Founders of the Social Sciences. Montreal: McGill Queen’s University Press. Sets out the case for including Catharine Macaulay, Germaine de Staël, Florence Nightingale, Beatrice Webb and Jane Addams as theorists who should be part of university courses and textbooks on sociological theory, political theory and philosophy. Alatas, S.F. and V. Sinha (2017) Sociological Theory Beyond the Canon, London: Palgrave Macmillan. The “Introduction: Eurocentrism, Androcentrism and Sociological Theory” argues for the inclusion of women (focussing on Nightingale and Martineau) in the “canon,” and chapters on them and non-​ European thinkers follow.

225

226

227

PART IV

Information in the nineteenth century (Dangerous) systems

228

229

13 THE NINETEENTH-​C ENTURY INFORMATION REVOLUTION AND WORLD PEACE Edward Beasley

In Victorian London, the post came a dozen times a day. In 1882 Charles Dickens’s son, also called Charles Dickens, published compendia of useful information. He detailed how in central London the mail was delivered 12 times every day on weekdays and six times in the inner suburbs. And most cities and towns around England received between two and four deliveries from London each day of the work-​week. The largest cities had six or seven deliveries. The younger Dickens included a multi-​page table so one could know exactly when the mail was to be picked up or delivered in every town around the country. He also provided ample details about the telegraph system, which provided near-​instant communication. But the extraordinary thing is that none of this seemed out of the ordinary. The younger Dickens let pass without comment what might seem to us to be an incredibly frequent delivery of physical messages, from the 12 daily postal deliveries to the underground pneumatic tubes connecting the telegraph offices of the City of London with the stock exchange. Apparently this rapid flow of information which was made possible by recent technological changes was so important there was no need for Dickens to justify or explain it.1 Obviously this system of frenetic communication had a great deal to do with the information needs of a British Empire, and with the needs of the world financial and trading system centred on London. Military orders had to be sent around the world, and stock traders needed their market news. But in my view much of the importance of what was going on lay in the cultural area. People were very conscious of the speed and multiplication of information. The new world of communications changed the way they thought about their own civilization and its place in the world. For some observers, it seemed that exciting vistas had opened. All the new information would make life better for the poor, bring nations closer together, and promote peace. One is reminded of more recent thinkers who have imagined that computers and the internet would knit us together and bring in an age of direct democracy.2 Yet some Victorians took a different view. The same new achievements could feed a sense of cultural superiority. Some nations had built the post offices and railways and telegraph cables that all might benefit from, and some nations had not built them. Perhaps these places needed British help and British control.3 The new information order of the nineteenth century contributed in very direct, practical ways to the monitoring and governing of a world-​wide imperial and financial system; on another level it also contributed to the growth of imperialism.

229

230

Edward Beasley

The example of what the West had achieved reinforced the sense of an imperial mission of civilization and modernization and intellectual ordering. Is this a surprise? Does having more information about the world always lead to more tolerance for other cultures, or more international understanding? In the eighteenth century, Samuel Johnson had been sceptical about whether any quickly-​shared observations about foreign peoples could be of much use in the creation of deeper knowledge. His friend and biographer James Boswell came in one day excited by meeting Captain Cook. Boswell was ready to ship off and learn all about the cultures of the Pacific until Johnson reminded him ‘how very little he can learn from such voyages’. Boswell summed up Johnson’s position: a great part of what we are told by the travellers to the South Sea must be conjecture, because they had not enough of the language of those countries to understand so much as they have related. Objects falling under the observation of the senses might be clearly known; but every thing intellectual, every thing abstract—​politicks, morals, and religion, must be darkly guessed at.4 This chapter will begin with the place of information in early Victorian culture. Then I will look at Rowland Hill’s postal reforms and their effects, and how the mail came to be delivered so frequently. Finally this chapter will consider some of the larger questions about the information order which had been constructed over the course of the nineteenth century. How did the new network change the way people thought about the world? I will conclude by asking whether all the new information led the British toward any real understanding of other cultures, or toward any greater philosophical sophistication in thinking about them.

An all-​encompassing revolution To some degree, managing information was always a necessity. In 1825, two culturally conservative observers were already complaining about ‘the progress of wealth, the increase and condensation of the population, the facilities of loco-​motion, the quick circulation of intelligence’—​caused by ’commerce, manufactures, steam-​boats, stage-​coaches, newspapers’.5 There was plenty of noise back then, at least in the cities—​literal noise from the hucksters and barkers and people yelling in the streets, as well as the informational noise. And so the need to seek out some quiet—​and some better-​sorted information—​also has its pre-​Victorian analogues. In the eighteenth and early nineteenth centuries, one could repair to coffeehouses that specialized in keeping libraries of plays; other coffeehouses were the centres of intelligence about the trade and politics of one or another part of the world, whether it might be South America or the Baltic, and these establishments kept backfiles of the newspapers from that area; others yet had the shipping news—​such as Lloyd’s.6 To understand how much further Victorian changes in the information order would go, perhaps we should start with changes in publishing. Steam-​powered printing presses began to be used in the 1820s and became widespread in the following decade. A new world of cheap, mass-​audience newspapers, pamphlets, and books could develop. Before this time, the poor only had access to broadsheets—​single sheets of sensationalist news—​and also to the three standard works of English Protestant civilization, namely the Bible, Fox’s Book of Martyrs, and Pilgrim’s Progress. Now for the first time, substantial quantities of other printed material could reach the common people. Among the new offerings were the part-​by-​part publications of the novels of Charles Dickens (the father).7 The elder Dickens’s career would have looked very different two decades earlier, before steam-​powered print. 230

231

Information revolution and world peace

There was also Charles Knight’s Penny Magazine.8 In backing the Penny Magazine (1832–​ 46), the Penny Cyclopaedia (1833–​44, and published in parts so the humble could afford it), and a number of popular monographs on scientific and cultural subjects, the Society for the Diffusion of Useful Knowledge (SDUK), founded in 1826, tried to exploit this explosion in print to bring information about science and culture to the masses. The Society’s publications also brought into the homes of those of modest means, for the first time, cheap reproductions of pictures of art and travel from around the world. This gave a visual education to those whose rough clothes and hard, poorly-​paid work left them no access to galleries or finely printed volumes.9 The publications of the Society went around the world, being translated and republished from Mexico to China, and forming much of the basis of the modern liberal education order in many countries, not least British India.10 And then there were the other revolutionary changes in communication, such as the building of the railways, the telegraphs (set up alongside the railways), or, for that matter, the creation of a bewildering number of displays about current events and the wonders of the world that literary scholar Richard Altick described in his oversized volume The Shows of London.11 Meanwhile the poor themselves were taking action to link up their local networks and educate themselves about the world. Working-​class activists actively corresponded with each other, often in letters sent in bundles or by courier rather than through the post. The repressive ministries during the Napoleonic Wars tried to stop them but they could choke off the mails only for a time.12 The importance of communication and the spread of knowledge was well understood by the radicals of the Chartist era of 1838–​48. The great political leader of the Chartists, Feargus O’Connor, was the proprietor of the weekly newspaper, The Northern Star, bringing news from around the globe to the working classes. Many of the consumers of this news were illiterate and had the newspaper read to them by their fellow workers, but in this way its circulation was huge. Chartists looked forward to a world of greater literacy and learning when rich and poor would be able to reason together in the halls of power.13 If even the poor could look forward to a new era of communications and education, so too could the government itself make new attempts to use information on an imperial scale. The desire on the part of rulers to use information to control their people was hardly a new one. During the Enlightenment, monarchs in Prussia and Austria had created the ‘police state’ (Polizeistaat) in its original sense. House-​numbering and the registration of the population allowed the police force, operating as the agency of government administration, to monitor and promote the health and well-​being of the people, including their proper ventilation and diet—​ as well as allowing the state to find military recruits with a new efficiency.14 In France, Alexis de Tocqueville was surprised to discover in his researchers in the pre-​Revolutionary archives that the old French monarchy—​however bad it was at following through with specific plans and however hamstrung it was because of a lack of financial resources—​had been monitoring the country very closely and trying to promote local welfare and development everywhere.15 Information was sought by the government in England, too. In the eighteenth century, postal riders gathered intelligence for the government of the day, and in troubled times private correspondence was regularly opened by the authorities.16 But the Information Age improvements made possible in England starting in the 1830s and 1840s, with the inauguration of mass publishing, cheap mails, and telegraphy, were so much greater in scale than in earlier years that they were also different in kind. The electric telegraph system was very well developed from one end of Europe to the other in the 1850s, with cables under the Channel as well. The telegraph was a key part in defeating the so-​called Indian Mutiny (in fact a national revolution) in 1857–​58, despite the continual cutting of the lines. 231

232

Edward Beasley

The system was opened to Indian use soon after the Mutiny. In 1870 a successful Red Sea Cable (there had been earlier unsuccessful attempts) united Europe to India.17 Historian Daniel Headrick and others have examined the creation of the transatlantic cable and other long-​distance undersea cables. Transatlantic service began in 1866; Australia was connected by cable in 1872, and New Zealand to Australia and thus the rest of the world in 1876.18 It wasn’t just cables that were tying the world together. In a 1981 work, Headrick noted that the development of the ocean-​going steam ship—​making much faster progress and following its own schedule rather than the winds—​made global travel and communications more systematic, more predictable, and more legible than ever before.19 Not only did information move about like never before, but it was also gathered and deployed in new ways. The inauguration in 1801 of what became the decennial British census—​part of the need to more carefully administer a country at war—​came not long before the publication of Parliamentary Debates by William Cobbett and soon by T.C. Hansard, beginning in 1803. (Although voluminous, Hansard was not a perfect record. It contained only a selective account of what was said, and it included revised remarks and written answers that were inserted later).20 After the third census, in 1821, there were three data points (1801, 1811, 1821), and therefore curves (speed-​ups and slow-​downs) rather than merely trendlines could be defined. To explore such data the London Statistical Society was founded in 1834. (The information useful for governing a state, including at that time maps and the descriptions of ports, had begun to be referred to as ‘statistics’ in the 1790s—​in England and on the Continent—​with the more strictly numerical sense coming later.) So two major dimensions of information, then, namely the statistical information that government had and what the legislators were saying about it each day, were increasingly matters of public record and reported in the newspapers. So too, the ‘Blue Books’—​testimony before parliamentary commissions and reports by government bodies—​were now being made public and collected for reference in a numbered sequence; this material radically increased in the 1830s because of investigations on the Poor Law and other matters. As historian Oz Frankel notes, there were complaints from MPs and others that there was so much of this material coming out that it was hard to assess or store.21 Meanwhile government correspondence was multiplying. In 1867, the former Chancellor of the Exchequer and soon to be Prime Minister, William Ewart Gladstone, estimated that it had multiplied by nine times in the preceding quarter-​century.22 All these letters led to challenges for policy makers, and not least for the Colonial Office, much of whose work was carried out through handwritten missives from around the globe. The Colonial Office’s Permanent Undersecretary of State, James Stephen (he held the position from 1836 to 1848), tried to rationalize paper flow within the office, as did his successor, Herman Merivale, who had been Professor of Political Economy at Oxford (1837–​47). While at Oxford, Merivale had written a volume attempting to systematize knowledge of the many different parts of the world subsumed under the name British Empire.23 Systematization also meant that Colonial Office increasingly required officials from around the globe to send in reports with standard narrative and statistical elements. Historian Zöe Laidlaw argues that an ‘information crisis’ in this period soon led all government departments to remodel their data collection according to the practises pioneered in the Colonial Office.24 Creating concise, usable digests of correspondence was increasingly important. An example of the problems that might arise should government business continue to depend on haphazard collections of private letters came in 1839. The Home and War Offices found that they did not have any accessible record of where the commander of the troops in the north of England had ordered his forces to go in meeting the Chartist threat. The commander, Richard Jackson, 232

233

Information revolution and world peace

could not be replaced by Charles Napier until well into the next year because too much key information about where the troops were located lay in Jackson’s head and nowhere else. His reports to London had been in the form of individual letters subject to little in the way of a system in either their sending or their receiving.25 Other reforms, as well, could be made in the way government information was recorded and presented. To demonstrate the need for changes in the administration of the army, the author Harriet Martineau collaborated with a mathematically trained observer who developed new ways of showing information visually—​the exploding pie chart (or ‘polar chart’), which could reveal the patterns in large amounts of data. The data in question were mortality figures for the army in the Crimean War. What the charts made clear was that far more soldiers died from what should have been preventable diseases than from enemy action. The mathematician with whom Martineau worked was Florence Nightingale, who provided the illustrations for Martineau’s 1859 work, England and Her Soldiers.26 Nightingale had invented a new way to bring government information to life—​a new way of looking at the world.

The Post Office But the greatest information revolutionaries were those who sought to multiply the world of useful information for the widest audiences, above and beyond the comparatively rarefied few who might be interested in government statistics. What did these reformers think they were doing, and how were their hopes realized in practice? The early Victorian generation of government improvers—​the Society for the Diffusion of Useful Knowledge members who backed the Penny Magazine and the Great Exhibition—​had their greatest success with the introduction of the Penny Postage, a success far greater than the invention of the exploding pie chart or the government Blue Book. Here the visionary was one of the younger members of the group, Rowland Hill. Having taught for a time in his family’s school, Hill became the paid secretary to the South Australia Commission, working on the foundation of the first colony there. And then his real life’s work began. In 1837 he circulated a pamphlet on how the British Post Office might be improved.27 The main problem was that, at the time, letters were charged by distance carried, in a very complicated system that cost a great deal to administer. This cost accounted for most of the postage charged. The government’s desire to turn a profit on the mails accounted for most of the rest. Postal rates were high, and the postage was paid by the recipient, who might not have the money to accept the letter. The minimum charge was fourpence for the first 15 miles. Postage then rose steeply for greater distances, so that a one-​page letter from London to Manchester cost 11d, and double that if it had two sheets; while one page sent from London to Edinburgh cost 16 1/​2d. The average letter cost 8 1/​2d, an amount that represented perhaps a fifth of a poor family’s weekly income. With rates so exorbitant there was massive evasion. Some people sent parcels full of letters by coach, for example, instead of through the Post Office, or they wrote many small letters on one sheet to save postage, trusting the recipient to cut the sheet apart and deliver the fragments to the other addressees by hand.28 Hill found that whatever the distance, the actual cost to the Post Office of carrying a letter was only a fraction of a penny. This excluded the cost of collecting the postage on delivery, plus the cost of balancing the accounts of the postage collected against the accounts from the post offices where the mail had originated. So he suggested abolishing those extra costs. He wanted to charge a uniform postal rate of one penny per letter, paid in advance so there would be no such accounts to keep and no distances to measure; and when the letters were delivered, 233

234

Edward Beasley

there would be no need for the postmen to move slowly on their rounds, collecting money. He envisioned using prepaid covers or postage stamps to show that postage had been paid.29 Richard Cobden, Henry Cole, and others helped to build a national campaign of occasional newspapers, pamphlets, placards, and public speeches to promote the penny post.30 Hill believed the amount of mail carried by the Post Office would be multiplied by five or six times, and this would make up for the loss in revenue resulting from the lower charges for each letter. Charles Knight told Hill that ‘if the rate of postage were reduced to a Penny, many hundreds of thousands of Prospectuses of new books, and of Publishers’ Catalogues, would be annually circulated … putting such information effectually before the inhabitants of rural districts especially’. Knight and other publishers could place their ‘lists practically in the hands of Country Booksellers, Professional Men, and Literary Societies’.31 Direct mail at an affordable rate would then bring the latest in literary life much more immediately into the countryside. As John Dillon wrote to Hill: ‘I suspect that the tax on postage is more emphatically a “tax on knowledge”, at least on civilization, than at least the tax on newspapers was’.32 Above everything else, the new plan would allow servants and other workers to write to their distant relations. Family life would be strengthened across the country.33 For the labouring classes, the cheapest letter under the old system cost one-​third of a day’s wage. Much of their mail, even letters containing dire family news, went unclaimed because they could not afford the postage.34 Now, mail would simply be delivered. Family news would also travel far more quickly. As things stood, wrote Hill, ‘[a]‌letter which shall arrive in London between six and seven o ’clock by a morning mail, would not be delivered in Hampstead, or any other place equally distant, till eleven or twelve o ’clock’35 Shocking! But why did this level of service seem so bad to him? Perhaps the need for speed that Hill seemed to feel was inspired by the easy circulation of ideas in Scotland. For before Rowland Hill was ever heard of, the mails were already being delivered four times a day in Edinburgh, Glasgow, and several other cities north of the border.36 In any case, in framing his plan, as Hill tells us, he drew his ‘data’ from the official statistics about the mails in the Blue Books,37 and from a study of the subject presented at the annual meeting of the British Association for the Advancement of Science.38 (This gathering, held each year in a different city outside London, dates from 1831.39) And so an acceleration in the information flow had itself allowed Hill to propose accelerating the flow of information still further. His own plan was intended to make possible many even greater analytical triumphs of this kind. New knowledge would come from the rapidly increasing desire for the collection of statistical and general scientific information. Thus the Society for the Diffusion of Useful Knowledge, by means of an extensive correspondence, collected a mass of most valuable matter connected to benefit societies … [M]‌any other inquiries relating, for instance, to education, to the practice of medicine; to various departments of science, as astronomy, meteorology, and geology; as also to general statistics, are, beyond all doubt, suppressed at present by the cost of postage. It is needless to enlarge upon the importance of such inquiries to commerce, science, and good government.40 What Hill had in mind was an expanded horizon for social science, one which would extend the part statistical, part correspondence-​based research that he himself had carried out in order to craft his proposal.41 A second daily delivery in towns, as Hill noted in 1838, would be extended across the country ‘according to the increase of population, wealth, and information’—​the word ‘information’ is present here. Even if no one at the time discussed an 234

235

Information revolution and world peace

‘information age’ (the earliest reference to the phrase in the Oxford English Dictionary is from 1960), Hill and his fellow-​campaigners were going out of their way to construct it.42

After Hill Hill’s arguments about Penny Postage carried the day, bolstered by petitions with more than a quarter of a million signatures. His plan was adopted by Parliament in 1839.43 The volume of mails radically increased, more than doubling to over 168 million letters from 1839 to 1840, the first year of the new system, and more than doubling again by 1850.44 In much of the rest of Europe postal rates were lowered and standardized on Hillian lines.45 In the UK, further advances were made. Postal districts and mail codes were instituted; the new railways were employed to replace coaches; sorting-​on-​the-​move began on the mail trains; and the country was dotted with the new phenomena of mail slots and pillar boxes. The latter idea was borrowed from France by the Post Office official and novelist Anthony Trollope. Inventions multiplied. There were mechanical arms and nets so sacks of letters could be thrown off high-​ speed trains; and a contract fleet of the new ocean-​going steamships (replacing the old Post Office packet ships) moved the Irish, imperial, and foreign mail.46 The age of railways and cheap postage produced a massive growth and change in the flow of information. In 1837, two years before his plan was adopted, Hill had suggested increasing London’s mail deliveries to every half hour.47 By 1864, when Hill was fully in charge in the Post Office, there were already 11 deliveries in central London, with the first between 7:30 and 9:00 in the morning, and eleventh beginning at 7:45 at night.48 It wasn’t only the Square Mile, where trading information had to move quickly. By 1867, Paddington, Notting Hill, Kensington, and Brompton had seven-​times-​a-​day mail delivery, although there were sometimes unplanned delays.49 And as we have seen, by that date the centre of London had its 12 daily deliveries, essentially an hourly visit from the postman. Meanwhile, as Charles Knight pointed out, the use of the railways had meant that by the 1860s, ‘it would … be impossible for a traveller in England to set himself down in any accessible situation where the post from London would not reach him in twelve hours’50 Telegrams, while expensive, were also having their effect on society. In 1880, the blind Postmaster General Henry Fawcett, citing Bentham, would make arguments for their further increase that were similar to Hill’s for the Penny Post—​arguments for the importance of the government doing whatever it could to increase communication. Fawcett could point out that since the Post Office had taken over the telegraph service in 1870, the number of telegrams had risen from six million per year to 26 million. He rejected any means testing for the cheapest rates, which should be open to all with no discrimination by class or wealth.51 I am not the first historian to quote Prime Minister Benjamin Disraeli’s 1880 novel Endymyion on the results of these reforms: It is difficult for those of us who live in an age of railroads, telegraphs, penny posts, and penny newspapers, to realize how uneventful, how limited in thought and feeling, as well as in incident, was the life of an English family of retired habits and limited means only forty years ago.52 The ramifications of the change spread everywhere. Scholars have explored what lower postal rates and the invention of postage stamps meant for the literacy of the people.53 Others have focused on the development of the Post Office itself: Its world-​wide shipping fleet; its savings bank; or how it took over the private telegraphs in the 1870s.54 By the end of the 235

236

Edward Beasley

century, it had become the largest employer in the country, as historian Duncan Campbell-​ Smith notes, with a larger workforce by 1914 than the army, navy, and shipyards of the British Empire all around the world, all combined.55 But it was at its heart a system of information exchange, and the amount of information moving around was I think the most impressive result of all. This aspect of things certainly impressed contemporaries. The revolution in information which had been accomplished came to be central to the self-​image of the nation. By 1851, the journalist and economist Walter Bagehot could base his case for the superiority of British civilization over the French version on the free discussion that self-​controlled Englishmen were capable of, because they communicated with each other and accumulated vast numbers of facts rather than rushing into theories.56 He would later argue that modern England was the country it was because it had entered ‘The Age of Discussion’—​the title of the sixth chapter of his Physics and Politics (1872).57 In On Liberty, published in 1859, John Stuart Mill described an England where the industrial revolution and a satisfying intellectual life both stemmed from the way orthodoxies were challenged in open debate. The more opinions and courses of action that were tried and thought about, the better. There would be ever more progress if there were ever more variation in opinion and in style of life, and ever more trials of the variations against each other—​assuming people were paying attention.58 Another book that was published in the same year made essentially the same argument—​that the more variations there were, and the more contests of strength and influence between varieties, the more the most successful varieties would make themselves felt, and the more progress and improvement and wonder there would be. This book was On the Origin of Species by Means of Natural Selection, or the Preservation of Favoured Races in the Struggle for Life.59 Mill’s argument for maximizing freedom of expression was also informed by what he and other officials of the East India Company thought of the cultures they ruled over. But his views on the matter were not to receive their fullest development. Mill died before he could write his projected work on ‘ethology’. In that work he had hoped to resolve the contradiction between his individualist theory of human psychological development and his equal commitment to another idea, that the different cultures of the world stood at different levels on a civilizational ladder.60 Could individuals educated in the West rise above their non-​Western cultural backgrounds? Did they deserve to be free individually before all their compatriots completed their centuries-​long schooling in the ways of civilization? The imperial context in which Mill worked and thought—​giving him some inkling of the diversity of the world’s cultures, and also the temptation to rank them—​provided him with a perhaps insuperable philosophical challenge as an individualist. Could England’s amazing world of information exchange be made global, thus combating racism? Here the crusader was John Henniker Heaton, an Australian who served in the UK as MP for Canterbury from 1885 to 1910. His first idea was that one penny should take a letter all over the British Empire and the United States. He began writing pamphlets on the subject in 1886, campaigning in the newspapers and in Parliament, and corresponding with the great and good across party lines.61 Heaton travelled to the United States in 1890, where he worked with Mark Twain and others for a Penny Postage there, which was instituted in 1908.62 In the United Kingdom, the postal bureaucracy frustrated many of his improvements, as he told Alfred Russel Wallace (who had suggested some improvements of his own).63 But eventually he had his way. In an era when there was a greater focus on ideas about race than had been the case in Rowland Hill’s day, Heaton had stressed that his plan would transcend such divisions. Arguing for a world-​wide penny post in 1905 (the British Empire Penny Post was almost complete by this point64), he envisioned a system whereby 236

237

Information revolution and world peace

any inhabitant of the planet, white, black, or yellow, may be enabled for the sum of one penny to communicate with any other, at the lowest possible rate and at the highest attainable speed: Englishman with Frenchman, German, Italian, or Russian; European with American; Asiatic with Australasian or African; so that when one soul has something to say to another neither colour, nor religion, nor creed, nor diplomacy, nor national antipathy, nor latitude, nor longitude, nor poverty, nor any other barrier shall stand between them.65 What was at stake, he argued the year before, was nothing less than world peace: ‘Nations, like individuals, are less likely to quarrel if in free communication, than if they wrap themselves in savage isolation.’66

The cultural effects of imperial communication Was this high rhetoric justified? Rowland Hill had predicted that the Penny Postage would bring changes to ‘the practice of medicine … various departments of science, as astronomy, meteorology, and geology; as also to general statistics’. Was that saying too much about the effects of the changes in information exchange upon thought? In fact Hill was right. Charles Darwin’s life, hardly an insignificant one, was one illustration of this point. As is well known, after the voyage of the Beagle Darwin sat for much of the rest of his life at his desk in Down House, using the post to ask questions and request specimens from specialists and friendly onlookers around the globe, and to spar with and discuss scientific questions with friend and foe. Darwin also reached out to a former crewman of the Beagle to send him barnacles from Australia, carefully explaining what barnacles were. And he wrote to a great many obscure, work-​a-​day naturalists with whom he would co-​develop experimental procedures and shared data.67 As his modern biographer Janet Browne puts the matter, ‘[w]‌ith pen and ink and postage stamps … in his country estate in a tiny village in Kent, he was in turn manager, chief executive, broker, and strategist for a world-​wide enterprise’. She adds: ‘Such a life obviously depended on the postal system, the preeminent collective enterprise of the Victorian period.’ She points out that by the middle of the century, ‘600 million letters were being despatched every year’.68 The way Darwin lived his life and the research and writing he did were made possible—​not made easier, but made possible—​by the Post Office and the global network of mail ships. If a few days’ correspondence stretched instead to weeks, or a few weeks’ worth stretched to months, then inescapably the correspondence of decades would have taken several times longer than Darwin’s busy life could ever have lasted. The new information order of the nineteenth century, and the postal reforms not least among them, had many other profound effects on the very nature of writing and thought. There was an unprecedented level of awareness regarding the world outside the British Isles, and also led to new ways of thinking about it. Books appeared on what the British Empire was and how it might be given a tighter political structure or even turned into a world state, with a single Parliament in London rather than the almost self-​governing polities which had existed in growing numbers in Australia, New Zealand, British North America, and South Africa since the 1850s. Little would come of this, not least because self-​governing colonies did not want to lose their political autonomy. But it suggests the kinds of dreams that late Victorians might indulge in over their morning newspaper.69 But not all of the effects of this new awareness of empire on the part of the British people were so benign. As the mid-​to late Victorians came to categorize the world and carry pictures of it in their minds, new, harsher ways of thinking about race spread throughout British society 237

238

Edward Beasley

after mid-​century, as I have discussed elsewhere.70 New fields of activity were opened for foreign adventures to please or placate the public. This was the era of imperial growth in Africa and inter-​imperial rivalries. As Robert M. Young has argued, another result of the growth in knowledge and information was that after 1870 or so, newly specialized professional journals carried more specialized, less broadly accessible articles. New challenges arose in how to index and access this burgeoning technical literature.71 Meanwhile, the popular journals which had once featured the greatest and most intellectually ambitious writers of the day, those men and women who were trying to fit all of life and knowledge together at the highest level, could not keep up. People writing for the broader journals were no longer theologically informed scientists aiming for the highest levels of intellectual connection and unity; and many of the less specialized periodicals now aimed no higher than topical political coverage, or at what was easily digestible and popular.72 In examining the Colonial Society, which was organized in 1868 to promote the empire, I found that many of its early members had written on some specific part of the world when they were younger, basing their views on personal experience and careful research. But as the years went by and they were faced with information about other places, they set themselves up as all-​purpose imperial experts. In their later works many of them—​there were some exceptions—​moved toward pontification and generalities, commenting on the issues of the day in their well-​tried phrases.73 Popular newspapers—​and the more popular illustrated periodicals, from Charles Dickens the novelist’s Household Words (1850–​58) and All the Year Round (1859–​95) to the non-​Dickensian Illustrated London News (1842–​2003) and the highly informative and topical humour magazine Punch (1841–​1992, 1996–​2002)—​presented to the public their own take on notable events, whether the doings of General Gordon in Africa or what might be done about Oom Paul Kruger in the Transvaal. What did it all mean? The underlying question of whether the common people of Great Britain were even paying attention to the empire and the information streaming in from it has been raised by historian Bernard Porter. He looks sceptically at the appeal of each specific imperial phenomenon to various audiences. He argues that much of the time the British people as a whole were not paying attention. But I think the question needs to be reframed—​more in terms of whether the common people paid attention to the world as a whole, of which their empire formed so large a part. So, what would be at stake is not the popular support for British Imperialism, but the effects of the global information revolution which the United Kingdom was at the centre of.74 Indeed, the growing scholarly literature on globalization in the modern era presumes and further documents a new global information order.75 Certainly there was ever more information about the empire for interested parties to sort through if they so desired; and because of the parcel post, cultural and biological specimens were piling up, too.76 Without some principle of selection and arrangement, the ‘imperial archive’ that literary researcher Thomas Richards discussed in a 1993 book threatened to turn into an imperial cacophony, and the museums which presented it to the public threatened to become unmanageable and devoid of meaning.77 One might simply let all the information about the peoples of the world pile up in heaps. One example of that was the Pitt Rivers Museum in Oxford, where anthropological gatherings from all the world are piled up together by form—​headdresses, combs, guns, votive statuary, and what have you, each kind of thing in its own case or cases. This arrangement was designed to allow for cross-​cultural comparisons of similar objects, rather than highlighting any particular cultural context. As an irreplaceable Victorian vision, this arrangement and the huge crowded display of objects that it features have been preserved to this day, with some changes in rationale and detailed arrangement.78 238

239

Information revolution and world peace

The Superintendent of the British Museum’s natural history collections, Richard Owen, faced a similar challenge of displaying all in one place the collections gleaned or taken from the whole world. In the crowded natural history galleries of the British Museum in Bloomsbury he hit up an idea that he took to the new British Museum (Natural History) when it opened in South Kensington in 1881. This was the ‘index collection’, where in the great hall samples of each major order of life would make the biological classification clear to the more casual visitors, and introduce them to the detailed displays in the individual galleries. Owen’s resistance to Darwinian evolution helped guide the categories he chose, but as a great comparative anatomist, he did offer a set of groupings that made sense.79 Those charged with cultural artefacts tried different means of classification, too, for the growing fund of knowledge about contemporary peoples—​their arts, their religion, their languages—​coming in from around the world. The principle might be geographical, or it might be a time series running from primitivism (the dark-​coloured people) to civilization (the light-​coloured people). This developmental paradigm reflects Victorian ‘social evolutionism’. With the idea of the ‘missing link’, physical brutes could be conjured up and sorted at one end of a progression that ended with what we might call Homo londinium modernus.80 Imposing categories in this way is exercising power. Having grown up in this era, the British imperial officials of the first half of the twentieth century would employ anthropology for their own purposes, sponsoring anthropological investigations in order to better administer their colonies. They used anthropology to unearth and co-​opt African power structures. The degree to which this may have suborned or changed the direction of anthropological study has been a subject of much discussion among scholars.81 Understanding the many hundreds of cultures in sub-​Saharan Africa was indeed a challenge to anthropological thinking, but attempts were made to pay attention to the specifics, anyway, and some well-​placed Africans spoke for themselves.82 More could have been accomplished in the construction of scholarship if not for the marginalizing of voices because of European racism.

Conclusion The innovations that we have been reviewing had made possible a new, wider gaze. Anti-​ European resistance efforts around the globe became connected to one another by post and telegram, as historian Antoinette Burton details. An Indian revolutionary group, the Ghadar Party, was founded in California in 1913, and its newspapers and pamphlets were seized in British colonies throughout the world. Most likely, this was not the kind of communication between the continents that Henniker Heaton had in mind when he imagined a world-​wide penny post in 1905. Yet in a way it illustrated his dream: The subjects of the anthropological studies were now writing to one another. To some British colonial officials there seemed to be a world-​wide conspiracy against them.83 Indeed, detecting conspiracies or hidden patterns was one way of trying to recapture a sense of understanding or control. As the historian Carlo Ginzburg has noted, a cultural trope developed in the late nineteenth century which centred on finding the subtle clues that might unlock a hidden order underneath all the masses of data. The clues might be—​in the burgeoning world of detective fiction—​the gleanings of Sherlock Holmes through his magnifying glass; or they might be the Freudian slip that could be used to unlock the hidden depths of the personality; or (extending Ginzburg’s analysis), they might be the fingerprints employed—​in the first instance in British India—​to identify one individual among the millions.84 If there was a global conspiracy of the Others against the British, it was made possible by the information order which the British themselves had created, and which they exploited as well. 239

240

Edward Beasley

Colonial authorities in the nineteenth and twentieth centuries used the information infrastructure for their own purposes, and spy networks and coded cables were employed in the service of the British state at home and abroad.85 The efficiency of the British information order is described in Rudyard Kipling’s 1901 novel, Kim. The novel begins by extolling the displays of religious art and scholarly acumen of the staff of the Lahore Museum, ​in fact run by Kipling’s father, John Lockwood Kipling.86 It then proceeds to dramatize the superior skills of the British spies and couriers in northwest India, and the superiority of the British communications system generally. Kipling imagines a Tibetan Lama who—​when asked how he was going to use the railways of India to travel from Lahore to Varanasi—​responds this way: ‘Oh for that one but asks a question and pays money, and the appointed persons despatch all to the appointed place. That much I knew in my lamassery from sure report.’87 For all of Kipling’s somewhat ambiguous pride in the British information order in India, from its presentation of Indian culture back to the Indians in the Lahore Museum to the use of the communications and spying network as a means of imperial administration, the administered peoples themselves probably had other ideas about who they were and how information might be used, as this novel itself suggests. The new information system was too Protean to be controlled from the top, a fact which the foundation and spread of the Ghadar Party makes clear. It was simply too hard to keep track of all the information. Indeed, despite the multiple expansions and reorganizations of the Colonial Office in the era of jingoism and the New Imperialism at the turn of the twentieth century, there were never enough staff to keep up with all the reports, the letters, the telegrams, and (after the turn of the century) the telephone calls that came in from a world empire.88 Imperial officials may have had the finest communication facilities at their beck and call, but the new information order was never entirely under their control. Information regimes are seldom so benign as their originators expect them to be, nor so controllable as their regulators would like. Nor did the new information order of the nineteenth century bring in a period of world peace. This was especially clear as the telegrams flew back and forth in July 1914.

Notes 1 Charles Dickens, Dickens’s Dictionary of London, 4th edn (London: Macmillan, 1882)—​ ‘Postal Regulations’, 220–​3; ‘Mails’, 170–​5; Élisa Le Briand, ‘Pneumatic Post Networks in Europe’, in Muriel Le Roux (ed.), Post Offices of Europe, 18th‒21st Centuries: A Comparative History (Brussels: Peter Lang, 2014), 225–​40. Sunday deliveries existed only in the provinces—​Howard Robinson, The British Post Office: A History (Princeton: Princeton University Press, 1948), 330–​1. 2 Martin Ceadel, The Origins of War Prevention: The British Peace Movement and International Relations, 1730–​1854 (Oxford: Clarendon Press, 1996); Paul Laity, The British Peace Movement, 1870–​1914 (Oxford: Clarendon Press, 2001); Åke Grönlund, ‘Introduction to the Special Issue on E-​democracy in Practice’, e-​Service Journal 2, 1 (Fall 2002), 3–​8. 3 Francis G. Hutchins, The Illusion of Permanence: British Imperialism in India (Princeton: Princeton University Press, 1967), 119–​24; Michael Adas, Machines as the Measure of Men: Science, Technology, and Ideologies of Western Dominance (Ithaca, New York, and London: Cornell University Press, 1989). 4 James Boswell, The Life of Samuel Johnson [1791] (New York: Alfred A. Knopf, 1992), 2, 5. Entry for 3 April 1776. 5 [J.T. Coleridge and John Miller], ‘Mechanics Institutes and Infant Schools’, Quarterly Review 32 (1825), 410–​28 at 415. 6 See John Timbs, Clubs and Club Life in London: With Anecdotes of its Former Coffee-​houses, Hostelries, and Taverns, from the Seventeenth Century to the Present Time (London: Chatto and Windus, 1872; reprint, Detroit: Gale Research Co., 1967); Brian Cowan, The Social Life of Coffee: The Emergence of the British Coffeehouse (New Haven and London: Yale University Press, 2005), 164–​76.

240

241

Information revolution and world peace 7 Aileen Fyfe, Steam-​Powered Knowledge: William Chambers and the Business of Publishing, 1820–​1860 (Chicago: University of Chicago Press, 2012), 35–​8; Richard D. Altick, The English Common Reader: A Social History of the Mass Reading Public, 1800–​1900 (Chicago: University of Chicago Press, 1957), 262, 277; Patricia Anderson, The Printed Image and the Transformation of Popular Culture, 1790–​1860 (Oxford: Clarendon Press, 1991), ­Chapter 1. 8 See Charles Knight, Passages of Working Life During Half a Century, 3 vols (London: Bradbury & Evans, 1864–​5); Valerie Gray, Charles Knight: Educator, Publisher, Writer (Aldershot: Ashgate, 2006), 46–​7, 50–​3. 9 Anderson, The Printed Image, ­chapter 2. 10 Thomas Palmelund Johansen, ‘The World Wide Web of the Society for the Diffusion of Useful Knowledge: On the Global Circulation of Broughamite Educational Literature, 1826–​1848’, Victorian Periodicals Review 50, 4 (Winter 2017), 703–​20; Eugenia Roldán Vera, ‘Useful Knowledge for Export’, in Maria Frasca-​Spada and Nick Jardine (eds), Books and the Sciences in History (Cambridge: Cambridge University Press, 2000), 338–​53; Ian J. Barrow, ‘India for the Working Classes: The Maps of the Society for the Diffusion of Useful Knowledge’, Modern Asian Studies 38, 3 (2004), 677–​702. 11 Richard D. Altick, The Shows of London: A Panoramic History of Exhibitions, 1600–​1862 (Cambridge, MA: Harvard University Press, 1978). 12 E. P. Thompson, The Making of the English Working Class (London: Victor Gollancz, 1963); Derek Gregory, ‘The Friction of Distance? Information Circulation and the Mails in Early Nineteenth-​ Century England’, Journal of Historical Geography 13 (1987), 130–​54. 13 David Jones, Chartism and the Chartists (London: Allen Lane, 1975). Henry Weisser, British Working-​ Class Movements and Europe, 1815–​48 (Manchester: Manchester University Press, 1975); Malcolm Chase, Chartism: A New History (Manchester: Manchester University Press, 2007), 44–​5, 294, 333. 14 See Rudolf Braun, ‘Taxation, Sociopolitical Structure, and State-​ Building: Great Britain and Brandenburg-​ Prussia’, in Charles Tilly (ed.), The Formation of National States in Western Europe (Princeton: Princeton University Press, 1975), 243–​327 at 272, 299–​301; Markus Krajewski, Paper Machines: About Cards and Catalogs, 1548–​1929, trans. Peter Krapp (Cambridge, MA: MIT Press, 2011), 27–​31. 15 Alexis de Tocqueville, The Old Régime and the French Revolution, Vol. 1: The Complete Text, François Furet and Françoise Mélonio (eds), trans. Alan S. Kahan (Chicago: University of Chicago Press, 1998), book 2, c­ hapters 2–​5. 16 Patrick Joyce, The State of Freedom: A Social History of the British State since 1800 (Cambridge: Cambridge University Press, 2013), 62–​3. 17 Saroj Ghose, ‘Commercial Needs and Military Necessities: The Telegraph in India’, in Roy MacLeod and Deepak Kumar (eds), Technology and the Raj: Western Technology and Technical Transfers to India, 1700–​1947 (New Delhi: Sage, 1995), 153–​76; Daniel Headrick, The Tentacles of Progress: Technology Transfer in the Age of Imperialism, 1850–​1940 (Oxford: Oxford University Press, 1988), 97–​122. 18 Daniel Headrick, The Tools of Empire: Technology and European Imperialism in the Nineteenth Century (New York: Oxford University Press, 1981), ­chapter 11. 19 Headrick, The Tools of Empire, ­chapters 9 and 12. See also Douglas R. Burgess, Jr., Engines of Empire: Steamships and the Victorian Imagination (Stanford: Stanford University Press, 2016). 20 H. Donaldson Jordon, ‘The Reports of Parliamentary Debates, 1803–​1908’, Econonica 34 (1931), 437–​49. 21 Michael J. Cullen, The Statistical Movement in Early Victorian England: The Foundation of Empirical Social Research (Hassocks, Sussex: Harvester Press, 1975), 9–​24; Kathrin Levitan, A Cultural History of the British Census: Envisioning the Multitude in the Nineteenth Century (Basingstoke: Palgrave, 2011); Oz Frankel, ‘Blue Books and the Victorian Reader’, Victorian Studies 46, 2 (2004), 308–​18. 22 William Ewart Gladstone, Draft Minute on the Pension for Rowland Hill [March 1867], Gladstone Papers, British Library Add. MS 44753, fols 44–​7 at fols 44–​5. 23 Herman Merivale, Lectures on Colonization and the Colonies (London: Longman, Brown, Green, and Longmans, 1842). 24 Zöe Laidlaw, Colonial Connections, 1815–​ 45: Patronage, the Information Revolution, and Colonial Government (Manchester: Manchester University Press, 2005), ­chapter 7. See also Cullen, Statistical Movement; and John Cell, British Colonial Administration in the Mid-​Nineteenth Century: The Policy-​ Making Process (New Haven and London: Yale University Press, 1970), 32–​6. 25 Edward Beasley, The Chartist General: Charles James Napier, the Conquest of Sind and Imperial Liberalism (London: Routledge, 2017), 137, 139.

241

242

Edward Beasley 26 Harriet Martineau, England and Her Soldiers (London: Smith, Elder, 1859); Lynn McDonald, ‘Florence Nightingale, Statistics and the Crimean War’, Journal of the Royal Statistical Society A 177, 3 (2014), 569–​86. 27 Rowland Hill and George Birkbeck Hill, The Life of Sir Rowland Hill and the History of the Penny Postage, 2 vols (London: Thomas de la Rue, 1880). 28 [Matthew Davenport Hill], ‘Postal Reform’, Edinburgh Review 70 (1840), 545–​73; [Pearson Hill], The Post Office of Fifty Years Ago (London: Cassell, [1887]), 4–​5; Henry Cole, Fifty Years of Public Work, 2 vols (London: George Bell and Sons, 1884), 1:51. The steep rise in postage with distance was instituted during the Napoleonic Wars as a revenue measure—​Gregory, ‘The Friction of Distance?’, 132, 135, 139. 29 Rowland Hill, Post Office Reform: Its Importance and Practicability (London: Charles Knight, 1837). 30 Bonython and Burton, The Great Exhibitor, 56–​ 61; Cole, Fifty Years of Public Work, 1:34–​69, 2:94–​142. 31 Charles Knight to Hill, 2 March 1837, Hill Papers, BL Add. MS 31978, fols 29–​30. 32 John Dillon to Hill, 17 January 1837, Hill Papers, BL Add. MS 31978, fols 9–​10. There had been a campaign against stamp and paper duties as ‘taxes on knowledge’. The last of these taxes was repealed in 1861—​Collet Dobson Collet, History of the Taxes on Knowledge: Their Origin and Repeal (London: Watts, 1933), 159–​65. 33 Thornley to Hill, 21 January 1837, Hill Papers, BL Add. MS 31978, fols 11–​14 at fol 13; Hill, Post Office Reform, 92. 34 [Matthew Davenport Hill], ‘Postal Reform’, 554–​6. 35 Hill, Post Office Reform, 52. 36 Testimony of Robert Wallace, M.P., 14 August 1835, given in Great Britain, Parliament, Appendix to the Sixth Report of the Commissioners appointed to Inquire into the Post Office Department, No. 135 (1836), Appendix H, 256. 37 Hill and Hill, Life of Sir Rowland Hill, 1:246. 38 Hill, Post Office Reform, quotation, 64. 39 Jack Morrell and Arnold Thackray, Gentlemen of Science: Early Years of the British Association for the Advancement of Science (Oxford: Clarendon Press, 1981). 40 Hill, Post Office Reform, 94. 41 See also Hill, Requisites, where Hill stresses that the Penny Post had facilitated Government research as well, 12. 42 Rowland Hill, draft testimony [February 1838], Hill Papers, BL Add. MS 31978, fols. 126–​31 at 130. 43 [Matthew Davenport Hill], ‘Postal Reform’; Campbell-​Smith, Masters of the Post, 125–​32. 44 Daunton, Royal Mail, 23. 45 Olivia Langlois, ‘The English Impact on the Adoption of a Unique Tax in France and Other European Countries’, trans. Kenneth Berri, in Muriel Le Roux (ed.), Post Offices of Europe, 18th–​21st Centuries: A Comparative History (Brussels: Peter Lang, 2014), 429–​39. 46 Campbell-​Smith, Masters of the Post, 139–​47, 157–​60; Daunton, Royal Mail, 46, 154–​79; James Wilson Hyde, The Royal Mail: Its Curiosities and Romance (Edinburgh and London: William Blackwood and Sons, 1885), 145–​54. 47 Hill’s testimony, 23 February 1837, given in Great Britain, Parliament, ‘Ninth Report of the Commissioners Appointed to Inquire into the Management of the Post-​Office Department’ (1837), No. 99, Appendix, 36. 48 William Lewins, Her Majesty’s Mails: An Historical and Descriptive Account of the British Post-​Office (London: Samson Low, Son, and Marston, 1864), 328. 49 J. W. A. Lowder, A Postal History of London, 1635–​ 1960 (corrected proofs, Postal Museum), Chapter IX, 4. 50 Charles Knight, Knowledge is Power, new edn (London: Bell and Daldy, 1866), 165. 51 Postmaster-​General Henry Fawcett, ‘Postal Telegrams—​Deputation to the Postmaster-​General,’ Journal of the Society of Arts, vol. 27, no. 1444 (23 July 1880), 735–​9. 52 Benjamin Disraeli, Endymion [1880], c­ hapter 12, quoted in Robinson, British Post Office, 301. 53 John Vincent, Literacy and Popular Culture: England 1750–​1914 (Cambridge: Cambridge University Press, 1989), 32–​52, 230–​2; David Vincent, ‘The Progress of Literacy’, Victorian Studies, 45, 3 (Spring 2003), 405–​431. 54 Daunton, Royal Mail, ­chapter 3; Campbell-​Smith, Masters of the Post, 175–​96. 55 Campbell-​Smith, Masters of the Post, 169.

242

243

Information revolution and world peace 56 Walter Bagehot, ‘Letters on the French Coup de État of 1851’, Letter 4 ‘On the Aptitude of the French Character for National Freedom’ [29 January 1852, published in Inquirer, 31 January 1852], in Norman St. John-​Stevas (ed.), The Collected Works of Walter Bagehot, 15 vols (London: The Economist, 1965–​86), IV, 54–​62 at 55–​61. 57 Walter Bagehot, Physics and Politics, in Norman St. John-​Stevas (ed.), The Collected Works of Walter Bagehot, 15 vols (London: The Economist, 1965–​86), VII, 106–​33. 58 John Stuart Mill, On Liberty (London: John W. Parker and Son, 1859). 59 Charles Darwin, On the Origin of Species by Means of Natural Selection, or the Preservation of Favoured Races in the Struggle for Life (London: John Murray, 1859), final chapter. 60 L. S. Feuer, ‘John Stuart Mill as Sociologist: The Unwritten Ethology’, in John M. Robson and Michael Laine (eds), James and John Start Mill: Papers of the Centenary Conference (Toronto: University of Toronto Press, 1976), 86–​100. 61 John Henniker Heaton to William Ewart Gladstone, 29 October 1886, Gladstone Papers, BL Add. MS 44459, fols. 106–​7; John Henniker Heaton, ‘Postal and Telegraphic Reforms’, Contemporary Review 59 (March 1891), 328–​42; idem, Imperial Penny Postage: Speech at Manchester on May 2nd, 1892 (London: Harrison and Sons, [1892]), 28. 62 Mrs Adrian Porter [Rose Henniker Heaton], The Life and Letters of Sir John Henniker Heaton, Bt. (London: John Lane at the Bodley Head, 1916), 191, 195–​7. 63 Alfred Russel Wallace to John Henniker Heaton, 17 March 1889, Wallace Papers, BL Add. 46441, fols 90–​1. 64 Porter, Life and Letters of Sir John Henniker Heaton, 178–​82, 215–​21; Bruce, The Broad Stone of Empire, 2:287–​293. 65 Heaton’s public appeal, 10 August 1905, given in Porter, Life and Letters of Sir John Henniker Heaton, 182–​4. 66 Great Britain, Parliament, ‘Penny Postage: Copy of letters addressed to the Postmaster-​General by the Hon. Member for Canterbury, dated October 1904 and February 1906, on the subject of universal penny postage (in continuation of Parliamentary Paper, no. 34, of session 1887)’ (1906) Cd. 117. 67 Paul White, ‘Correspondence as a Medium of Reception and Appropriation’, in Eve-​Marie Engles and Thomas F. Glick (eds), The Reception of Charles Darwin in Europe, vol. 1 (London: Continuum, 2008), 54–​65. 68 Janet Browne, Charles Darwin: The Power of Place (New York: Alfred A. Knopf, 2002), 13. The number of letters was close to three and a half billion per year by 1914, or six billion items if one counted everything, including postcards and parcels—​Campbell-​Smith, Masters of the Post, 167. 69 The earliest major example: Charles Dilke, Greater Britain: A Record of Travel in the English-​Speaking Countries during 1866 and 1867, 2 vols (London: Macmillan, 1868)—​running to many subsequent editions. 70 Edward Beasley, The Victorian Reinvention of Race: New Racisms and the Problem of Grouping in the Human Sciences (London: Routledge, 2010). 71 Alex Csiszar, ‘Seriality and the Search for Order: Scientific Print and its Problems During the Late Nineteenth Century’, History of Science 68 (2010), 399–​434; Susan Sheets-​Pyenson, ‘Darwin’s Data: His Reading of Natural History Journals, 1837–​42’, Journal of the History of Biology 14, 2 (autumn 1991), 231–​48. 72 Robert M. Young, ‘Natural Theology, Victorian Periodicals, and the Fragmentation of a Common Context’, in idem., Darwin’s Metaphor: Nature’s Place in Victorian Culture (Cambridge: Cambridge University Press, 1985), 126–​ 63, especially 127–​ 8, 152–​ 61; T. W. Heyck, The Transformation of Intellectual Life in Victorian England (London: Croom Helm, 1982), 75–​6, 120–​1, 184–​6, 197–​202. 73 Edward Beasley, Empire as the Triumph of Theory: Imperialism, Information, and the Colonial Society of 1868 (London: Routledge, 2005); idem., Mid-​Victorian Imperialists: British Gentlemen and the Empire of the Mind (London: Routledge, 2005). 74 See Bernard Porter, Absent-​Minded Imperialists: Empire, Society, and Culture in Britain (Oxford: Oxford University Press, 2004). 75 A. G. Hopkins (ed.), Global History: Interactions between the Universal and the Local (Basingstoke: Palgrave Macmillan, 2006). 76 Lewins, Her Majesty’s Mails, 135–​6; natural history specimens were being sent in increasing numbers in the original Penny Post as well: [Henry Cole], ‘The Postage Stamp’, Westminster Review 65 (March 1840), 263–​71 at 265.

243

244

Edward Beasley 77 Thomas Richards, The Imperial Archive: Knowledge and the Fantasy of Empire (London: Verso, 1993); Sophie Forgan, ‘The Architecture of Display: Museums, Universities and Objects in Nineteenth-​ century Britain’, History of Science 32 (1994), 139–​62 at 149–​51. 78 George W. Stocking, Jr, Victorian Anthropology (New York: The Free Press, 1987), 263–​5, 273; William Ryan Chapman, ‘Arranging Ethnology: A.H.L.F. Pitt Rivers and the Typological Tradition’, in George Stocking (ed.), Objects and Others: Essays on Museums and Material Culture (Madison: University of Wisconsin Press, 1988), 15–​48; David K. van Kueren, ‘Cabinets and Culture: Victorian Anthropology and the Museum Context’, Journal of the History of the Behavioral Sciences 25 (January 1989), 26–​39; Chris Gosden, Frances Larson, and Alison Petch, Knowing Things: Exploring the Collections at the Pitt Rivers Museum, 1884–​1945 (Oxford: Oxford University Press, 2007); Bishnupriya Basak, ‘Collecting Objects: Situating the Indian Collection (1884–​1945) in Pitt Rivers Museum, Oxford’, Studies in History 27, 1 (2011), 85–​109. 79 Nicolaas A. Rupke, Richard Owen: Victorian Naturalist (New Haven and London: Yale University Press, 1994), 34, 151, 257. 80 David M. Wilson, The British Museum: A History (London: British Museum Press, 2002), 159–​61, 194; Stocking, Victorian Anthropology; Tony Ballantyne, Orientalism and Race: Aryanism and the British Empire (Basingstoke: Palgrave Macmillan, 2002); Colin Kidd, The Forging of Races: Race and Scripture in the Protestant Atlantic World, 1600–​2000 (Cambridge: Cambridge University Press, 2006), 176–​8, 186. 81 See Talal Asad (ed.), Anthropology and the Colonial Encounter (Atlantic Highlands, New Jersey: Humanities Press International; Reading: Ithaca Press, 1973). 82 See Kwame Anthony Appiah, In My Father’s House: Africa in the Philosophy of Culture (Oxford: Oxford University Press, 1992); and Russell H. Bates, V. Y. Mudimbe, and Jean O’Barr (eds), Africa and the Disciplines: The Contribution of Research in Africa to the Social Sciences and Humanities (Chicago: University of Chicago Press, 1993). 83 Antoinette Burton, The Trouble with Empire: Challenges to Modern British Imperialism (Oxford: Oxford University Press, 2015), c­ hapter 3, especially 191–​7. 84 Carlo Ginzburg, Clues, Myths, and the Historical Method, trans. by John and Anne C. Tedeschi (Baltimore: Johns Hopkins University Press, 1989); G. S. Sodhi and Jasjeet Kaur, ‘The Forgotten Indian Pioneers of Fingerprint Science’, Current Science 88, 1 (10 January 2005), 185–​91. 85 Campbell-​Smith, Masters of the Post, 144–​5, 215; Neil Pye, The Home Office and the Chartists, 1838–​ 1848: Protest and Repression in the West Riding of Yorkshire (Pontypool, Wales: Merlin Press, 2013), 51–​3, 59; Beasley, The Chartist General, 143–​4, 165, 209; C. A. Bayly, Empire and Information: Intelligence Gathering and Social Communication in India, 1780–​1870 (Cambridge: Cambridge University Press, 1996). 86 James Bryant and Susan Weber (eds), John Lockwood Kipling: Arts and Crafts in the Punjab and London (New Haven and London: Yale University Press, 2017). 87 Rudyard Kipling, Kim (New York: Doubleday, Doran, 1901), c­ hapters 1–​2, quotation at 16. 88 Robert V. Kubicek, The Administration of Imperialism: Joseph Chamberlain at the Colonial Office (Durham, North Carolina: Duke University Press, 1969), 22–​34.

244

245

14 CHARLES BABBAGE’S ECONOMY OF KNOWLEDGE Renee Prendergast

Introduction Charles Babbage started out as a mathematician and held the Lucasian chair of mathematics at the University of Cambridge from 1828 to 1839.1 However, from the early 1820s onwards, most of his energies and a large part of his fortune were devoted to the design and construction of his difference engines. Although partly constructed, no difference engine was ever completed by Babbage. The reasons that have been adduced include the limitations of precision engineering in the early nineteenth century,2 the withdrawal of government support and Babbage’s own limitations as a manager and negotiator. Babbage’s modern fame rests not on the difference engine but on his design of an analytical engine which, in retrospect, can be seen to have anticipated many of the architectural features of the modern computer.3 However, to focus only on this aspect of Babbage’s contribution to human knowledge would do him a considerable injustice. He provided his own assessment of his contribution in two chapters of his autobiographical, Passages from the Life of a Philosopher [1864]. The topics covered in these chapters indicate broad intellectual interests ranging from pure and applied mathematics, investigations into natural phenomena, inventions including his difference and analytical engines, mechanical notation and issues in political economy. There was a strong connection between Babbage’s interests in mathematics, his activities as an inventor and his work in political economy. His earliest publications were on mathematical topics and included his contribution to the translation of an influential work on integral and differential calculus by Lacroix [1816].4 Involvement with plans to set up a life assurance company which was eventually abandoned led to the production of life tables and A Comparative View of the Various Institutions for the Assurance of Lives [1826]. Babbage’s interest in mathematics led to an awareness of the shortcomings of the available mathematical tables and of the work being done in France to apply the principles of division of labour to the production of accurate sets of tables. This, in turn, led him to consider the potential for cost reduction and the elimination of errors through mechanization of calculation.5 His work on the design and construction of his calculating engine required that he familiarize himself with production processes and their limitations through visits to workshops and factories in Britain and on the continent. The generalizations arising from these visits supplemented by other material such as

245

246

Renee Prendergast

reports of House of Commons committees provided the basis for his examination of the role of machinery in manufacturing and his broader consideration of industrial economics in his most popular work On the Economy of Machinery and Manufacturing [1832].6 Even Babbage’s venture into theology in the Ninth Bridgewater Treatise, A Fragment [1837] in which he envisaged God as a programmer was connected to his work on the analytical engine (Hyman 1982, Ch. 10). Babbage approached issues relating to knowledge and information from a number of different standpoints. The early chapters of Economy contain interesting discussions of the way in which capital equipment economizes on information and skill while later chapters examine both production costs and the transaction costs which arise due to imperfections in information. Babbage was deeply interested in the relationship between science and its practical applications. These relationships were at the centre of Babbage’s reforming efforts in relation to the Royal Society and are discussed in Reflections of the decline of Science in England and some of its Causes [1830] and in Economy. Other features of innovation are discussed in detail in Economy and touched on in Babbage’s autobiography. The former also includes advice on appropriate methods for collecting facts about manufactures.7 Transaction costs in the form of search costs are discussed in Babbage’s The Exhibition of 1851 [1851]. As a student in Cambridge, Babbage and his colleagues read and discussed works by Francis Bacon and became adherents of Bacon’s inductive approach. They were impressed by Bacon’s views on the importance of accurate experimental data and also by his view that knowledge should bear fruit and transform the conditions of life (Snyder 2011: 37). Although Whewell and Jones, two of Babbage’s friends from his Cambridge days, were involved in efforts to create an inductive political economy in opposition to the deductive approach of Ricardo, there is no evidence that Babbage was directly involved in this enterprise (Henderson 1996 de Marchi and Sturges 1973). Nonetheless, he is likely to have had some sympathy with their position, since he wrote in Economy that political economists were often ‘reproached with too small a use of facts and too large an employment of theory’ (Babbage 1935: 156). Having argued that the relevant factual material could be supplied by merchants and manufacturers and having acknowledged that erroneous deduction could be made from recorded facts, Babbage went on to insist that ‘the errors arising from the absence of facts were far more numerous and far more durable’ (ibid.). Babbage’s dedication to accurate measurement and data collection is also evident in his involvement alongside Malthus and Sidgwick in the setting up of the statistical section, Section F, of the British Association for the Advancement of Science in 1833 and the London Statistical Society (forerunner of the current Royal Statistical Society) in 1834. However, Babbage was no mere empiricist. As Rosenberg (1994) notes he ‘saw everything through the inquiring eyes of someone searching for more general underlying principles, categories and commonalities’. This will be clear from our account of Babbage’s discussion of the role of machinery in manufactures and also his discussion of the associated transaction or verification costs.

Standardizing and economizing on knowledge A useful place to start is Section I of Economy which Babbage saw as covering the mechanical part of the subject. He begins by considering the sources of the advantages arising for machinery and manufacturing (Ch. 2). These advantages arise from three sources: (i) the addition which they made to human power; (ii) the economy they produce of human time and (iii) the conversion of apparently common and worthless substances into valuable products. Machines, themselves, fell into two broad categories: those which were employed to produce power and those whose role was to transmit force and execute work. Subsequent chapters classify and provide examples of machines based on the general purpose they fulfill. The chapters 246

247

Charles Babbage’s economy of knowledge

which are most germane to our interest in knowledge and information are Ch. XIII on the registration of operations, Ch. X which bears the title ‘On the Identity of the Work when it is of the Same Kind and of its Accuracy when it is of Different Kinds’ and Chapter XI which is entitled ‘On Copying’. These are discussed in greater detail below. Chapter VII also merits a brief mention. The title of the chapter refers to the exertion of forces too great for human power or operations too delicate for human touch but our interest here is in its discussion of problems of co-​ordination. Following a brief allusion to the difficulty of monitoring effort in cases where large numbers were required to act in unison, the first part of the chapter looks at how simple instruments can assist with the co-​ordination of the actions of large numbers of workmen relative to some goal. Most of the examples given in this instance involve the use of sound for signalling purposes. The chapter then considers how innovations involving the use of large-​scale machinery eliminated the need for the coordinated effort of large numbers of workmen. In the chapter on the registration of operations, Babbage is concerned with various devices that had been contrived for the purpose of measurement. Because measurement offers more precise information about objects and operations, it acts as a check against the ‘inattention, idleness and dishonesty of human agents’. Babbage enumerated a variety of instruments which could be used for counting series of operations and which in turn allowed the measurement of such things as distance travelled. He described the simple mechanism by which the amount of fluid contained in flasks and barrels could be measured. He discussed the merits of gas meters which, by making each customer pay only for what he consumes, would encourage the reduction in waste and he advocated the extension of metering to water with the same objective of reducing waste in mind. A variety of other instruments and innovations relating to them are described. These included clocks and watches, instruments for measuring atmospheric pressure, the quantity of rain and the direction and intensity of the shocks arising from earthquakes. In Chapter X, Babbage discusses what he regards as one of the most important results of the use of machinery, namely, the accuracy with which it executed its work and the identity of things manufactured by the same tool once the appropriate setting was arrived at by trial. The fact that, once properly set up, a machine could consistently produce accurate work economized on skill and saved time. In addition, it meant that the consistency of the output could not be compromised by the inattention or unskillfulness of labour. Even the addition of the rudest of tools extends human capabilities. Each improvement in tools makes existing operations easier and brings new, more difficult operations into view. Without tools, many things would be impossible to make but even if they were technically possible, they might not be feasible economically. With the aid of files and polishing substances, it would be possible for a skillful workman to fashion a cylinder out of a piece of steel. But the time required would be so considerable and the number of failures so large that the task would be impossible for all practical purposes. With the aid of a lathe and a sliding rest, the same process had become the everyday employment of hundreds of workmen. The processes by means of which the physical duplication of objects is achieved using information embodied in machinery and tools is discussed further in the chapter on copying. Copying was one of the most important sources of excellence in work produced by machinery because it allowed a great deal of effort and expense to be lavished on the original of the item to be replicated. Babbage described eight different processes by means of which copying could be accomplished. The first two of these were printing which could be from cavities (as in copper plate) or from surface as in wooden blocks or moveable type. These processes were important not just for the production of books and newspapers but also for the printing of calico, oilcloth and china. The third process described was that of casting. A pattern of the item to be cast is 247

248

Renee Prendergast

used to make a mould into which molten metal or other substance in a fluid state is poured. The mould retains the substance until it solidifies, at which point the mould is destroyed. The thing produced should exactly resemble the mould from which it was formed. The fourth process considered is that of moulding which is used in the production of things such as bricks and tiles; embossed china and calico and jewellery. The substances employed must be in a soft or plastic state at the outset of the process. They are compressed by mechanical force into a mould of the required form. The fifth process considered is stamping which is generally carried out on hot or cold metal by means of large presses. Items produced using this process include buttons, coins, medals and nail heads. Punching, which involves driving a steel punch through the substance to be cut, is the next copying process considered. Uses include the punching of iron plate for boilers and the creation of ornamental patterns in various metals. The final processes considered to constitute copying are unusual in that there is little resemblance between the copy and the original. The first of these is elongation and here the similarity lies in the fact that the cross-​section of the thing to be produced is the same as that of the tool through which it passes. Processes for which this is the case include wire drawing, pipe drawing, iron rolling and vermicelli. The last set of processes considered are entitled copying with altered dimensions. Under this heading, Babbage includes a wide variety of processes, many of them involving the use of a lathe for turning operations such as screw cutting. As noted by Rosenberg, Babbage’s chapter on copying brings together a wide array of industrial processes in which cheapness results from the skill devoted to some original instrument or tool becoming the basis for many thousands of copies (Rosenberg 1994). These instruments or tools can be thought of as embodying the information necessary to reproduce the objects in question. As is evident from the mechanical descriptions, the use of machinery had important implications for the nature and composition of the skill required for production. The questions relating to this were discussed in Section II of the Economy, ‘On the Domestic and Political Economy of Manufactures’, particularly in the chapters relating to the division of labour to which we now turn.

The division of labour and the Babbage Principle8 Like Adam Smith before him, Babbage recognised that the division of labour was not the result any foresight relating to its consequences for the welfare of the community. However, he did not follow Smith in attributing it to an original propensity to exchange, suggesting instead that individuals would specialize when they found it profitable to do so. Smith and other writers on the division of labour had attributed its benefits to three factors: the increase in the dexterity in each individual workman; the saving of time in passing from one kind of work to another, and the invention of machinery which facilitated and abridged labour (Smith 1976, I.i.5). Babbage added an additional factor, namely, reduction in the time required for learning. This had been discussed by Smith in the context of a critique of long apprenticeships but not as part of his discussion of the advantages of the division of labour. Babbage agreed that the time required for ‘the acquisition of any art will depend on the difficulty of its execution; and that the greater the number of distinct processes, the longer will be the time which the apprentice must employ in acquiring it’ (Babbage 1835: 170). However, he argued that the length of apprenticeships depended both on the time required for learning and that required to allow the master to recoup the cost of training. As the former was reduced by the division of labour, the benefit to the master would increase but this, in turn, could be reduced by competition amongst the masters. With regard to reductions in the time lost when changing from one occupation to another, Babbage suggested that this factor could be important in cases where the change 248

249

Charles Babbage’s economy of knowledge

required the use of different tools which themselves involved substantial set-​up costs. Babbage also accepted that specialised workers would increase their dexterity more quickly than non-​ specialised workers but he did not regard this as a permanent source of advantage because, although it would take them longer to do so, non-​specialised workers would eventually reach the same level of dexterity (Babbage 1835: 173). As might be expected, Babbage had most to add on the topic of innovation. While he accepted that specialization could facilitate workmen in contriving new tools and simplifying processes, he was strongly of the view that the development of new machinery required a variety of skills and experience which were unlikely to be the province of the specialized workman. Different experience and skills including extensive knowledge of machinery and the power of making mechanical drawings were necessary in order to combine the scattered arts into a single machine (ibid.: 174–​5). Having thus engaged with the discussion of the division of labour amongst writers of political economy, Babbage argued that there was one additional principle without which an explanation for the cheapness of manufactured articles owing to the division of labour would be incomplete. This was: That the master manufacturer, by dividing the work to be executed into different processes, each requiring different degrees of skill or of force, can purchase exactly that precise quantity of both which is necessary for each process; whereas, if the whole work were executed by one workman, that person must possess sufficient skill to perform the most difficult, and sufficient strength to execute the most laborious, of the operations into which the art is divided. (Babbage 1935: 175–​6) Although Babbage discovered this principle independently, he acknowledged that it had been already stated by Melchiorre Gioja (1767–​1829) in Nuovo Prospetto delle Scienze Economiche published in Milan in 1815. Nonetheless, the principle is generally associated with Babbage and is considered to provide an explanation for the de-​skilling of labour that takes place under capitalism (Braverman 1974). Babbage, himself, provided a detailed illustration based on the process of pin making as carried out in England and in France. With an appropriate division of labour, wages per day ranged from 4.5 pence (paid to a boy) to 6 shillings (for the most skilled workman) whereas if one person were to carry out all the work, he would be required to have the skill necessary to perform the most demanding parts of the work even though it would occupy only a tiny proportion of his time. Babbage estimated that this would increase the cost of production by 3¾ times. As Babbage made clear, the principle applied not only to manual labour but also to mental labour. In fact, it may well have been first suggested to Babbage by his knowledge of the system used by the French mathematician and engineer Gaspard de Prony (1755–​1839) in the production of a series of new mathematical tables in France. This was a major undertaking the purpose of which was to facilitate the establishment of the metric system as the universal standard of measurement thereby distancing France from its pre-​revolutionary past and the plethora of units of measure which were seen as a legacy of protected interests and privileges of class (Swade 2003: 57). Inspired by a chance reading of chapter I of the Wealth of Nations, Prony had introduced a division of labour in which the work of constructing the tables was divided amongst three sections. The first section included some of France’s premier mathematicians, including amongst them Legendre and Carnot who decided on the appropriate functional forms; the second were people who had a good acquaintance with mathematics who converted the formulae into numbers while the skill required of members of the third and largest section 249

250

Renee Prendergast

was the ability to do simple addition and subtraction (Babbage 1935: 194–​6). As might be expected, Babbage saw the potential for the mechanization of the routine work conducted in the final stage through the use of the difference engine (Babbage 1935: 196–​9). The relevant details had been set out by him in 1822 in a letter to Sir Humphrey Davy, President of the Royal Society.9 The letter, which was part of Babbage’s campaign for public support for the construction of his difference engine, outlined the benefits of the proposed machine. These included the elimination of the ‘intolerable labour and fatiguing monotony of continued repetition of similar arithmetical calculations’; the elimination of errors in the computation of tables; and the reduction in the labour required to construct a set of tables. Referring to the work carried out in France under the guidance of Prony, Babbage suggested that his difference engine would allow the elimination of the 80 or so persons in the third and largest section who were employed to carry out the routine work of calculation. Hence, according Babbage, the total number of calculators employed could be reduced from 96 to 12. Babbage also noted that the errors made by those who were employed to copy the figures presented by the engines were a further source of inaccuracy. In order to eliminate this source of error, it was necessary that the machines themselves printed the relevant numbers as they were calculated. Babbage was aware that a large output and capital were necessary in order to fully utilize the most skilful human labour. The introduction of machinery also made for larger plants. The first machines devised for any particular production process tended to be very expensive and their economical use required 24-​hour operation of the plant. Although increasing demand for the machines led to improvements and reductions in cost, the need to provide expensive maintenance capability and the desire to utilize it fully created further pressures for expansion as did the setting up of separate accounts, payroll, procurement and sales departments. Babbage also noted that because of their scale of operations, large plants could afford to use agents to source rare materials from abroad, to conduct surveys into the tastes of consumers in different countries and to carry out the experiments necessary to invent new products and to improve existing ones (Babbage 1935: 221–​4).10 Finally, Babbage noted that the concentration of production in large factories had led to the elimination of the class of middlemen that had been previously interposed between the maker and the merchant (Babbage 1935: 218–​21). As is discussed in greater detail below, the reason for this was that in situations of large-​ scale business, problems of asymmetric information about quality could be resolved though reputational effects. Babbage’s discussion of the division of labour and the advantages of large firms was historically important and was drawn on by John Stuart Mill, Alfred Marshall and Karl Marx (Mill 1973: 110–​12, 132–​5; Marshall 1959: 219–​20, 1919: 224–​6; Marx 1977: 227). Mill drew on Babbage to support his argument that substantial advantages were derived in business transactions from proved trustworthiness and that these advantages were sometimes linked to the size of firms. Marshall credited Babbage with working out some of the chief ideas of Scientific Management and suggested that his extension of specialization to mental labour had provided the basis for Taylor’s doctrine on the separation of planning and execution (Marshall 1919: 224–​5). Marx’s discussion of the factors making for the ever-​increasing size of manufacturing establishments also drew on Babbage (Marx 1977: 227). Marx also worked out the implications of ‘Babbage’s principle’ in terms of de-​skilling and disciplining labour: ‘The habit of doing only one thing converts him [the labourer] into a never failing instrument, while his connection with the whole mechanism compels him to work with the regularity of parts of a machine’ (Marx 1977: 330). Marx also attended to the wider sociological implications of the division of labour, noting that ‘Manufacture … develops a hierarchy of labour powers, to which there corresponds a scale of wages’ (Marx 1977: 330). 250

251

Charles Babbage’s economy of knowledge

Innovation Babbage saw the tendency to economize on the input of the skilled labour at work in every aspect of productive activity. The division of mental labour and skill described by Prony in the context of mathematical tables also applied to innovation in an ordinary manufacturing context. A person seeking to set up a cotton or silk mill who had designed some new machinery which could be successfully employed in that pursuit would make sketches and drawings of his plans. This corresponded to Prony’s first section. To take matters further would require the assistance of operative engineers capable of executing the machine that had been designed. These corresponded to Prony’s second section. When sufficient machines had been produced, a large number of persons of a lower degree of skill would be employed to use them. However, the work performance of members of this third section would be supervised by members of the 2nd class (Babbage 1835: 195–​6). As to the construction of machines themselves, in the letter to Davy mentioned above, Babbage noted that his difference engine was constructed from relatively few parts frequently repeated and that, in this respect, it resembled the arithmetic to which it was applied (Babbage 1922: 6–​7). Babbage was of the view that the invention of new machinery did not require any rare or special gifts. Given the variety of contrivances already in use, a lot of relevant knowledge was readily available. With any new machinery, an important first step was to make drawings of the machine. This could involve a certain amount of trial and error until a coherent design was produced. While much experimentation could be conducted on paper, some physical and chemical aspects of design could only be tested by means of direct trials. When these experiments were conducted, the next step was the construction of the actual machine. The essential ingredients at this stage were good working drawings, good tools and good methods of working. Failure of inventions was often due to the inadequacy of early trials though inventions that experienced no mechanical problems could fail for economic reasons (Babbage 1835: 264–​ 5). It was also the case that inventions that were abandoned at one point in time might later succeed either because of improved trials or because of developments in the art of making machinery. The time was ripe for the introduction of machinery whenever there was demand for a great quantity of things, all exactly of the same kind. Despite his claim that there was nothing special involved in inventing new machinery, Babbage held that very superior intellectual powers were required in the case of complex machines such as his own difference and analytical engines. As he explained in his autobiography, the greatest difficulty in contriving his difference engine arose not from any difficulty in contriving mechanisms to execute each individual movement but rather from the complexity of the whole. Several different modes could be devised for executing each individual movement with the result that the number of combinations of these contrivances was so innumerable that no human mind could examine them all (Babbage 1991: 133–​5). With a view to alleviating this problem, Babbage invented a scheme of mechanical notation, the purpose of which was to make clear in a drawing the action of all the moving parts of a piece of machinery (Morrison and Morrison 1989: xxvi). I have invented and brought to maturity a system of signs for the explanation of machinery, which I have called Mechanical Notation, by means of which the drawings, the times of action, and the trains for the transmission of force, are expressed in a language at once simple and concise. Without the aid of this language I could not have invented the Analytical Engine; nor do I believe that any machinery of equal complexity can ever be contrived without the assistance of that or of some other 251

252

Renee Prendergast

equivalent language. The Difference Engine No. 2, to which I shall presently refer, is entirely described by its aid. The first details of the notation were set out in a paper communicated to the Royal Society on 17 January 1826.11 An overview of a later version was provided in a paper given away by Babbage during the Great Exhibition of 1851.12 However, despite his attempts to promote his invention, Babbage’s hopes for wide adoption of the notation were not fulfilled. The main reason appears to be that, given the relative simplicity of the bulk of mechanical undertakings, ordinary drawings proved to be perfectly adequate. However, recent developments relating to the design of modern, large-​scale integrated circuits have caused Babbage’s mechanical notation to be seen in new light. It is suggested that his parallel notations showing the geometry, the timing, the causal chains and the abstract components of his machines, have a direct parallel in the Hardware Description Languages developed to aid the design of modern, large-​scale integrated circuits. Research on the issue is ongoing but preliminary results suggest that, in the case of the second of Babbage’s difference engines, the notations were a description of the physical relationships of mechanical parts rather than an abstract description of the engine’s function. However, new material found in the archive suggests that while this might be true for the second of Babbage’s difference engine, the 2,700 notations for the analytical engine appear to embody higher-​level logic. This possibility has led to the suggestion that Babbage should be recognised not only as a pioneer of computing but also as a pioneer of Hardware Description Languages.13 Babbage argued that the quality of decision making with regard to the introduction of new machinery could be improved by the use of good factual information. Where a breakdown of the costs associated with any process was available, it would provide an indication of where innovation effort was likely to be most profitably employed (Babbage 1835: 203–​4). Babbage recommended that a full investigation of the associated costs and benefits be carried out before setting up any new manufacturing establishment or adopting a new machine (ibid.: 265–​6). However, he recognised (presumably based on his own experience) that it was often difficult to estimate these costs especially in the case of complicated new machinery. Experience also suggested that in the case of a new invention, the cost of the first machine could be as much as five times that of subsequent copies. Given Babbage’s awareness of the costs and risks associated with the first introduction of new machinery and indeed his own interest in invention, it might be expected that he would have been a strong advocate of patent protection.14 He accepted that each inventor should have the sole use of his invention, until such time as he had ‘been amply repaid for the risk and expense to which he has been exposed, as well as for the talent he has exerted in completing it’ (Babbage 1835: 360–​61). However, he was of the view that it was difficult to frame a law on intellectual property which was not open to serious objections. Besides, it was known that patents could be very difficult to defend. Consequently, instead of regarding a patent as a means of securing a monopoly price, some manufacturers ‘sell the patent article at such a price, as will merely produce the ordinary profits of capital; and thus secure to themselves the fabrication of it, because no competitors can derive a profit from invading a patent so exercised’ (Babbage 1935: 361).15 Babbage argued that progress in the arts and manufactures was intimately connected with progress in the abstract sciences and that continued progress required that the connection be rendered more intimate (Babbage 1835: 379). He accepted that innovators would normally receive adequate rewards through the market mechanism and that, in general, the public were better judges of the merits of any new product than the government would be. However, this 252

253

Charles Babbage’s economy of knowledge

mechanism could not be relied upon to adequately reward those who made advances in the pure sciences. Long intervals often elapsed between the discovery of new scientific principles and their practical application. Moreover, since the talents required for advancing the pure sciences were not the same as those required for practical applications, those reaping rewards through the markets were not those who made the initial discovery. Given that those who made advances in sciences could not expect to achieve adequate remuneration through the market mechanism, government support was necessary to promote the pure sciences. Some argued that no such intervention was necessary given the fact that making discoveries was a highly agreeable occupation. Babbage disagreed, arguing that if the only recompense was the intellectual pleasure felt in the pursuit, it was likely that scientific effort would be devoted to the more abstract areas of scientific endeavour rather than the practical, albeit, more expensive branches of science (Babbage 1968: 242). Babbage’s suggestions for the encouragement of the pure sciences included curriculum reform in the universities, better remunerated professorships, honours and prizes for discovery and greater access to public appointments for men of science (Babbage 1830: 2–​39). He directed strong criticism at the Royal Society16 but held out hope that the newly formed bodies such as the British Association for the Advancement of Science would provide a platform both for the stimulation of scientific activity and greater interaction between industry and science. The men of science would derive practical information from the great manufacturers and persons of wealth and property would become acquainted with the manufactures of the country and the importance of science for their further development. Babbage appreciated that without the support of public opinion, no administration was likely to provide support for scientific endeavour. Moreover, it would lack the knowledge base to do so appropriately (Babbage 1835: 382–​4). Babbage was of the view that technological progress would be encouraged by a law permitting limited partnerships involving people with moderate amounts of capital and inventive workmen. Because the pay of workmen would be linked to the success of the firm, such partnerships would offer better incentives for innovation to ordinary workmen (Babbage 1835: 254–​9). Later, in The Exposition of 1851, Babbage argued that, because it involved unlimited liability, exiting partnership law represented a major obstacle to the advance of the mechanical arts in England. If instead, partnerships could be entered into, in which liability was limited, persons with knowledge and capital could invest small sums in a number of schemes.17 By this, spreading risks over an extensive field, they would have more certain profits. Inventions which otherwise would be lost for want of capital would then be able to attract the necessary funds to go forward (Babbage 1968: 136–​7). In the period 1780–​1824 Britain sought to protect its superior technological knowledge from foreign competitors by means of restrictions on migration of skilled workmen and on the export of machinery (Jeremy 1977). Babbage viewed prohibitions on the migration of workmen as both ineffective and at variance with every principle of liberty (Babbage 1935: 362–​3). He also argued strongly against the continuing restrictions on the export of machinery.18 Such restrictions involved a sacrifice of the interests of the makers of machinery for the imagined benefit of those who use it. In Babbage’s view, rather than resulting in a loss of British advantage, machinery export would increase the class of skilled workmen and allow the benefits of greater specialization to be achieved. In the context of this discussion, Babbage considered the question of whether, in each kind of machinery, the potential for improvement would continue indefinitely. His view was that, although the incremental improvements to a particular innovation eventually grew smaller and smaller, the impact of small improvements could be very considerable where a large amount of machinery was employed (1835: 373).19 253

254

Renee Prendergast

Transactions costs In the later decades of the twentieth century, the problem of verifying the quality of goods and services was explored in economics as a problem of asymmetric information (Arrow 1963; Akerlof 1970; Spence 1973, 1976; Stiglitz 1987). However, the problem has a much longer, largely forgotten, history (Lupton 2015). The issue of quality was widely discussed in the seventeenth and early eighteenth century by authors including Barbon, Child, Swift and Mandeville, all of whom understood that the information available to buyers of goods was generally inferior to that available to the sellers (Prendergast 2007). In France, Colbert had introduced a strict regulatory regime with a view to improving the quality and reputation of French manufactures. However, from the late eighteenth century, authors such Turgot and later Say opposed the Colbertian system of regulation, arguing that low quality goods were still valuable, that the regulatory system was itself costly, and that in most cases the market mechanism could force producers to maintain the quality of goods because to do otherwise would eventually lead to a loss of reputation and business (Geary and Prendergast 2008). By the late eighteenth and early nineteenth centuries most British economic writers showed little interest in issues of quality. Reviewing On the Economy of Machinery and Manufactures for the Edinburgh Review, McCulloch dismissed Babbage’s discussion of quality arguing that adulteration was not carried out to the extent that was commonly supposed (McCulloch 1833: 320–​2). However, Babbage, himself, was clearly intrigued by the subject and, in his autobiography, listed his identification of the cost of verifying that an article was what it professed to be as one of his important contributions to human knowledge (Babbage 1864 [1991]: 329–​30). Not all of Babbage’s insights on the matter were new but what set his contribution apart from that of his predecessors was the relationships he uncovered between matters of quality, industry structure and technology. The costs of verifying quality could influence industry structure through decisions about vertical integration but the means by which quality issues could be resolved themselves depended on the prevailing industry structure. Babbage accepted that the classical theory according to which market prices were determined by the proportion between supply and demand and natural prices by the cost of production including normal profits was broadly speaking true. However, he argued that it required some modification in two cases: (i) when it was applied in conditions of monopoly and (ii) where the quality of an article could not be easily determined. In the latter case, the cost of the article to the purchaser was not just the price he paid for it but also the cost of verifying its quality (Babbage 1835: 134–​5). Babbage argued that these extra costs could be an important determinant of the decision to make or buy a particular product. For example, although it was generally the case that government could purchase most articles at a cheaper rate than that at which it could manufacture them itself, it might still be more economical to set up its own operations when account was taken of the cost of verifying the quality of each unit purchased and in devising methods of detecting the new modes of adulteration which were constantly resorted to (ibid.). Babbage quoted extensively from House of Commons Select Committee discussions of quality and information issues relating to clover seed, the Irish flax trade, the lace trade, the stocking trade and the watch trade. Based on these reports and other instances with which he was familiar, he argued that since the majority of purchasers lacked the skill to verify the quality of products, it was necessary for them to pay a higher price to some person with the skill and integrity to supply articles of the required quality (ibid.: 139–​40).20 The problem of verification was not confined to physical goods. When it was difficult to evaluate what exactly was being purchased, there was considerable scope for fraud in transactions involving services. In the conveying of parcels, for example, it was difficult for the customer 254

255

Charles Babbage’s economy of knowledge

to verify the appropriateness of a charge and extremely costly to recover any excess charges once they had been paid. In these circumstances, Babbage suggested that government should experiment with the provision of a parcel service along the lines of its exiting postal service (ibid.: 144). A consequence of the difficulty of charging for services could be seen in the case of Apothecaries who instead of being paid for their services and skill were remunerated by being allowed to charge a high price for their medicines. This provided an inducement to the Apothecaries to prescribe more than was necessary. In fact, in most cases, it was only by doing so that the apothecary could achieve fair remuneration for his effort (Babbage 1935: 140–​2). One of the possible consequences of poor quality was the collapse of trade. Babbage noted that a report investigating the grievances of framework knitters had pointed out that differences in the quality of lace were not visible to the naked eye but that the inferior product deteriorated rapidly on washing. The report claimed that where the inferior product was sold as first quality, it led to a subsequent collapse of the demand for any sort of lace from the district in question (ibid.: 137–​8). Such a result is indeed possible but, as we now know, following the work of Akerlof (1970), it is also possible that the market for inferior quality lace would continue while that for that of lace of the first quality disappeared. Babbage also reported that, in watchmaking, it was alleged that the marks and names of respectable manufacturers were forged by producers of inferior goods who sold these in foreign markets thereby leading to the collapse of orders. Such products were also sold in the countryside by pedlars who had departed by the time the fraud became known. As noted above, Babbage held that vertical integration might take place where the costs of verifying the quality of purchased inputs was high. Consistent with this view, he considered that outsourcing was most likely ‘where there exists a sure and quick method of ascertaining that the terms of the contract had been fulfilled’ (Babbage 1835: 213). Babbage also noted that the move from cottage to factory production in the calico trade led to the demise of a class of middle men and to a reduction in the need to monitor the measure and quality of cloth. In the case of the cottage industry, while the majority of workmen could be depended on, the fraud of a few rendered it necessary to examine each piece. Any single cottager had an incentive to cheat the merchant provided that fraud on his part, even if detected, did not become known to other buyers. However, once the institutional setting became one involving the merchant and a single great manufacturer, there was no longer an incentive to cheat because the parties to the transaction will know that any ‘loss or impeachment of the manufacturer’s character would be attended with greater injury to himself than any profit upon a single transaction would compensate’ (Babbage 1835: 218–​19). The value of character, though great in all circumstances of life, can never be so fully experienced by persons possessed of small capital, as by those employing much larger sums: whilst these larger sums of money for which the merchant deals, render his character for punctuality more studied and known by others. Thus it happens that high character supplies the place of an additional portion of capital; and the merchant, in dealing with the great manufacturer, is saved from the expense of verification, by knowing that the loss, or even the impeachment, of the manufacturer’s character, would be attended with greater injury to himself than any profit upon a single transaction could compensate. (Ibid.: 219) Babbage argued that well-​ grounded confidence in the character of its merchants and manufacturers was an important advantage that an old manufacturing country always possessed 255

256

Renee Prendergast

over its rivals. A consequence of this was that, in some of the largest English towns, sales and purchases on an extensive scale were made daily without any of the parties ever exchanging a written document (ibid.: 219–​21). The idea that repeated transactions and the expectation of further dealing could provide a mechanism for ensuring honesty in commerce was not in itself new (Geary and Prendergast 2008). What was new was Babbage’s explicit linking of the issue of honest dealing with the use of modern factory-​based organisational forms and his idea that the value of reputation was linked to firm size. Babbage also understood that well-​g rounded confidence in the character of merchants had a wider social value in reducing the transactions costs associated with trade. This aspect was subsequently emphasised by Mill who argued that the aggregate effect of labour was built on trustworthiness and that all labour expended on watching that people fulfilled their engagements was ‘so much withdrawn from the real business of production’ (Mill 1973: 111).

Search costs and middlemen Whereas in the Economy of Machinery, Babbage focused on the costs associated with ascertaining quality, in The Exposition of 1851 (Babbage 1851 [1968]) his focus was on the lack or asymmetry of price information. This arose because the organising committee of the exposition refused to allow prices to be posted on the various exhibits. Babbage argued that, in general, the failure to display prices defeated the effects of competition and imposed search costs in terms of a loss of time on both buyers and sellers. He also claimed that it led to price dispersion, increased prices and reduced demand (ibid.: 67–​9). To further support his argument on the importance of price information, Babbage turned his attention to the evolution of market days. Hawkers attending market days lost the opportunity of gaining extraordinary profits from isolated customers but gained more regular, less fluctuating profits and both they and their customers saved time (ibid.: 73–​4). As more regular markets evolved to facilitate the trade of merchants, larger dealers and international exchanges, a class of men emerged whose business it was to buy and sell on commission and whose presence provided a means of reducing the cost of obtaining market information. To reinforce his point, Babbage considered a market with 100 buyers and 100 sellers. In order to become fully acquainted with the state of the market, each buyer has to ask each seller at least two questions, namely, what is the price? and what quantity have you for sale at that price? This would give 20,000 questions in all whereas if there is a broker in the market each agent would provide the information on prices and quantities only to his own broker, making 400 questions in all. Brokers would then pool the information and arrange the necessary exchanges in a manner which was equitable to all parties. While the advantage just described was an important reason for the emergence of brokers, it was not the only one. Babbage noted that, in bargains between two individuals, each side would seek to secure advantageous terms by misrepresenting the price at which they are willing to buy or sell. Babbage suggested that the time required for bargaining before agreement was reached was longer and the agreement itself often less equitable than that which could be achieved by a broker. A broker was generally paid in proportion to the amount of their completed sales but they have no interest in favouring one class of customers more than another. Babbage was clear that a party with perfect information would have no need for a broker because he could acquire no new information. He also recognised that market exchange could take different forms and that the forms which were successful at any point in time were likely to be those which best succeeded in reducing the cost of transactions (ibid.: 76). 256

257

Charles Babbage’s economy of knowledge

Knowledge and progress As Jones 2016 notes, Babbage like many of his predecessors, had ambitions to realize a philosophical theory of invention or as he, himself, put it: ‘the principles that guided the intellect as well as the hands of those by whom immortal … works were executed’ (Babbage 1968: 133). Babbage’s description of the process involved in the discovery of new principles and the invention of new methods is an eloquent one but it also captures some of the difficulties and uncertainties involved in creating new knowledge. To investigate the laws by which the human intellect picks with caution its uncertain track through those obscure and outlying regions of our knowledge which separate the known and the certain from the unknown; -​to teach us how to cast as it were an intellectual and temporary connecting line across that chasm, by which a new truth is separated from the old-​confident that when arrested by that isolated truth it will have fixed itself upon one solid point, amidst a floating chaos of error, -​confident also that, when once the fixity of that single point has been assured, it is always possible, however formidable the task, to link it with innumerable ties to established knowledge, and thus to fill up the intervening space even to the very boundary of its enlarged domain: -​to achieve such a conquest in any science surpasses all other discoveries, for it supplies tools for the use of intellect, and enlarges the limits and powers of human reason. (Babbage, 1968: 134) While Babbage was certainly confident of his own merits, and ever mindful that his genius and his effort did not achieve the recognition that it deserved, he was never entirely guilty of forgetting that progress was a collective and cumulative enterprise. His reference to concentrating in one life the accumulated labour of many lives suggests familiarity with the theories of progress based on the accumulation of knowledge articulated by Blaise Pascal and Giorgio Baglivi in the seventeenth century and Bernard Mandeville and Adam Ferguson in the eighteenth (Prendergast 2014, 2017). In arguing for the importance of knowledge accumulation, Babbage noted that few in his time believed in the existence of a past golden age. However, a larger number were admirers of the past and ‘the wisdom of our ancestors’. Babbage accepted that our ancestors were indeed wise inasmuch as they applied their energies to their existing wants and necessities. They availed of the knowledge bequeathed to them by their predecessors and struggled to add to it for the benefit of themselves and their children. It was the duty of each generation to purify the knowledge which it had inherited and to add to it as much as possible. Babbage continued: Remember that accumulated knowledge, like accumulated capital, increases at compound interest: but it differs from the accumulation of capital in this; that the increase of knowledge produces a more rapid rate of progress, whilst the accumulation of capital leads to a lower rate of interest. Capital thus checks its own accumulation: knowledge accelerates its own advance. Each generation, therefore, to deserve comparison with its predecessor, is bound to add much more largely to the common stock than that which it immediately succeeds. (Babbage 1968: 210–​11)21 The argument that capital checks its own accumulation was commonplace amongst classical economists and is taken for granted by Babbage. While intimations of the argument that 257

258

Renee Prendergast

knowledge accelerates its own advance could be found in the work of French writers on progress, the comparison between the rate of knowledge accumulation and capital accumulation is unique to Babbage. Science and knowledge differed not only from capital but also from material things in general: Science and knowledge are subject, in their extension and increase, to laws quite opposite to those which regulate the material world. Unlike the forces of molecular attraction, which cease at sensible distances … the further we advance from the origin of our knowledge, the larger it becomes, and the greater power it bestows upon its cultivators, to add new fields to its dominions. (Babbage 1935: 386–​7) Each advance in science opened the way for further advance. There was no prospect that the advance of science would be exhausted. Babbage’s views can be seen as linking to the theories of development based on knowledge accumulation which predated the Smithian focus on capital accumulation but which had been largely displaced by it in the early nineteenth century (Prendergast 2010; Elmslie and Criss 1999). He was not alone in attributing independent importance to technical progress and knowledge accumulation but he alone emphasised the independent importance of scientific activity while at the same time appreciating the importance of its practical application. In Babbage’s grand vision of the division of labour, competition drove producers to constantly economize on skill and the costs associated with acquiring it. Machinery further economised on skill though it also extended human power and saved time. The result as Marx noted was a hierarchy of labour powers with those at the bottom of this hierarchy constantly at the mercy of being displaced by machinery. Babbage had glimpses of an international of division of labour in which countries like England which had obtained a technological advantage captured a greater share of the highly skilled jobs. He would have seen the contemporary international division of labour as the rational consequence of the forces he identified.

Notes 1 See Schaffer 2003 for details. 2 This explanation has been rejected by Bromley, 1990 who argues that the precision achievable at the time was adequate for Babbage’s purposes. Bromley ascribes the failure to complete the construction of the difference engine to problems arising in Babbage’s relationship with the British government over funding and especially to the difficulties in his relationship with the engineer Joseph Clement (Bromley, 1990: 81 and 96). This is in line with the views expressed the writer of Babbage’s obituary in The Engineer (1871: 284) who observed that that great fame would have been achieved by Babbage had ‘his noble brain been united with more common sense notions of managing mankind’ (ibid.). 3 Although it is accepted that Babbage’s analytical engine anticipated the architecture of the modern computer, it is also accepted that ‘his work, however brilliant and original, was without influence on the modern development of computers’ (Wilkes, 1977: 415). Indeed it has been suggested that knowledge of Babbage’s failure may have discouraged others from thinking along similar lines thus delaying the advent of the modern computer (ibid.). 4 Lacroix, S.F. 1816. An Elementary Treatise on the Differential and integral Calculus, translated from the French by Messrs. Babbage, Hershel and Peacock. 5 As noted by Hyman (1982: 49), Babbage gave several descriptions of the origin of the idea of constructing the difference engine but all of these related one way or another to the production of mathematical tables. See also accounts in Moseley (1964: 65) and Snyder (2011, Ch. 4). Bromley (1990: 96) has expressed doubts about the value of the engines for the calculation of mathematical tables.

258

259

Charles Babbage’s economy of knowledge 6 Henceforward Economy. References are to the fourth enlarged edition published in 1835. The work was an immediate success with 3,000 copies sold on publication. French and German translations were published in 1833. An Italian translation soon followed (Hyman 1982: 122). 7 See Prendergast 2017: 280 for a brief summary or Babbage (1835, Ch.12) for full details. 8 The Division of Labour is the subject of Ch. 19 of Babbage 1835. 9 It perhaps provides some insight into Babbage’s personality that having sought and received Sir Humphrey Davy’s support when lobbing the government to fund the construction of his difference engine, Babbage was later to launch a vehement attack on him in his Reflections on the Decline of Science in England. 10 In The Exposition of 1851, Babbage noted that while separate machines were used for each process in large workshops, smaller workshops such as those of jobbing masters who repaired machinery or made models of the inventions of others could not afford this arrangement. He suggested that the development of a multipurpose machine to meet the needs of these skilled craftsmen as an innovation opportunity (Babbage 1968: 102–​3). 11 On a method of expressing by signs the action of machinery. By Charles Babbage Esq. FRS. Communicated January 17 1826. Read 16 March 1826. 12 ‘Laws of Mechanical Notation’, By Charles Babbage. Given away by Mr Babbage during and after the Great Exhibition of 1851. 13 See S. Gee, 23 April 2015, ‘Plan 28 Makes Progress in Understanding Babbage’s Mechanical Notation’, /​www.i-​programmer.info/​news/​82-​heritage/​8506-​plan-​28-​makes-​progress-​in-​understanding-​ babbages-​mechanical-​notation-​.html; See also reports from Adrian Johnstone, Doron Swade and others in the Plan 28 Blog and lecture by Adrian Johnstone, ‘Notions and Notations: Designing Computers before Computing’, Ada Lovelace Symposium, University of Oxford: 9–​10 December 2015, published 6 January 2016, www.youtube.com/​watch?v=CsereRIRGQ0. 14 A patent is an exclusive privilege or monopoly granted to inventors or first introducers of some improvement for a period of time on payment of certain fees. Its purpose is to encourage new inventions and improvements by repaying the costs of invention and rewarding the inventor for his effort and talent (Babbage 1835: 359–​61). 15 For further discussion of this issue, see Romano 1982 and Prendergast 2017. 16 Babbage was concerned with the apparent indifference to science in the country with the most advanced industrial technology of the period. His many criticisms of the Royal Society included the fact that it had grown too large, that many of its members were not practicing scientists, that it was insular and that much of the work presented at the society fell below the standards achieved elsewhere. Babbage was also concerned with the governance of the society and the unwillingness of those in charge to implement necessary reforms (Hyman 1982: 88–​102). 17 The UK’s limited liability law was enacted in 1855. 18 Legislation prohibiting the export of machinery from Britain was partially relaxed in 1825 but finally repealed only in 1843. Jeremy (1977 1981) indicates that the prohibition was largely ineffective in any case. Elmslie and Criss (1999) provide details of debates on the issue including arguments for prohibition of machinery export. 19 Babbage regarded his recognition that a cause which at first sight might appear to be insignificant could have important consequences as one of his important insights which he highlighted in the introduction to the The Exposition of 1851. Babbage illustrated his point with the example of a workman raising his shovel an inch or two higher than was optimal. In the course of a day, this would produce a very sensible difference either in fatigue or in the amount of work done (Babbage 1968: 3). 20 For an earlier version of this insight, see the chapter ‘Of the Quality and Quantity of Wares’ in Nicholas Barbon’s A Discourse of Trade. 21 This could be regarded as an anticipation (from a different theoretical standpoint) of Solow’s finding that the permanent rate of growth of output per unit of labour input is independent of the saving (investment) rate and depends entirely on the rate of technological progress in the broadest sense (Solow, Nobel Lecture, 1987 in Solow 2000: xii).

References Akerlof, G. 1970. ‘The Market for Lemons: Quality Uncertainty and the Market Mechanism’, Quarterly Journal of Economics, 89: 488–​500.

259

260

Renee Prendergast Arrow, K. 1963. ‘Uncertainty and the Welfare Economics of Medical Care’, American Economic Review, 53(5): 941–​73. Babbage, C. 1822. A Letter to Sir Humphry Davy, Bart. ... on the Application of Machinery to the Purpose of Calculating and Printing Mathematical Tables. London: J. Booth. Babbage, C. 1826. A Comparative View of the Various Institutions for the Assurance of Lives. London: J. Mawman. Babbage, C. [1830] 1970. Reflections on the Decline of Science in England and Some of its Causes. New York: Augustus M. Kelly. Babbage, C. [1832] 1835. On the Economy of Machinery and Manufactures. Fourth enlarged edition. London: Charles Knight. Babbage, C. [ 1838] 1837. Ninth Bridgewater Treatise, A Fragment. Second edition. London: John Murray. Babbage, C. 1826. On a method of expressing by signs the action of machinery. By Charles Babbage Esq. FRS. Communicated January 17. Read March 16 1826. Babbage, C. 1851. ‘Laws of Mechanical Notation’, Given away by Mr Babbage during and after the Great Exhibition of 1851. https://​history-​computer.com/​Library/​CharlesBabbageLawsOfMechanicalNotat ion.pdf (accessed 5 February 2021). Babbage, C. [1851] 1968. The Exposition of 1851. London: Frank Cass. Babbage, C. [1864] 1991. Passages from the Life of a Philosopher, ed. M. Campbell-​Kelly. London: Pickering and Chatto. Barbon, N. 1690. A Discourse of Trade. London: Printed by Tho. Milbourn for the author. Braverman, H. 1974. Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century. New York: Monthly Review Press. Bromley, A.G. 1990. ‘Difference and Analytical Engines’, in W. Aspray, ed. Computing before Computers. Ames, IA: Iowa State University Press, pp. 29–​98. De Marchi, N.B. and Sturges, R.P. 1973. ‘Malthus and Ricardo’s Inductivist Critics: Four Letters to William Whewell’, Economica, 40(160): 379–​93. Elmslie, B and Criss, A.J. 1999. ‘Theories of Convergence and Growth in the Classical Period: The Role of Science, Technology and Trade’, Economica, 66: 135–​49. The Engineer 1871. ‘Murchison and Babbage’. 27 October: 284. Geary, F. and Prendergast, R. 2008. ‘Philosophers and Practical Men: Charles Babbage, Irish Merchants and the Economics of Information’, European Journal of the History of Economic Thought, 15(4): 571–​94. Gioja, M. 1815. Nuovo prospetto delle scienze economiche. Milan: Presse Gio. Pirotta. Henderson, J.P. 1996. Early Mathematical Economics: William Whewell and the British Case. London: Rowman & Littlefield. Hyman, A. 1982. Charles Babbage: Pioneer of the Computer. Oxford: Oxford University Press. Jeremy, D. 1977. ‘Damming the Flood: British Government Efforts to Check the Outflow of Technicians and Machinery, 1780‒1843’, The Business History Review, 51(1): 1–​34. Jeremy, D.J. 1981. Transatlantic Industrial Revolution: The Diffusion of Textile Technologies between Britain and America 1790–​1930s. Cambridge, MA: MIT Press. Jones, M.L. 2016. Reckoning with Matter: Calculating Machines, Innovation, and Thinking about Thinking from Pascal to Babbage. Chicago: University of Chicago Press. Lacroix, S.F. 1816. An Elementary Treatise on the Differential and Integral Calculus, translated from the French by Messrs. Babbage, Hershel and Peacock. Cambridge: J. Deighton and Sons. Lupton, S. 2015. ‘Quality Uncertainly in Early Economic Thought’, History of Political Economy, 47(3): 511–​34. Marshall, A. 1959. Principles of Economics. Eighth edition. London: Macmillan. Marshall, A. 1919. Industry and Trade. London: Macmillan. Marx, K. (1867) [1977]. Capital. Volume I. London: Lawrence & Wishart. McCulloch, J.R. (1833). ‘Babbage on Machinery and Manufactures’, Edinburgh Review, 56: 313–​32. Mill, J.S. [1948] 1973. Principles of Political Economy. Clifton: Augustus M. Kelley. Morrison, P. and Morrison, E. 1989. ‘Introduction’, in Charles Babbage, On the Principles and Development of the Calculator. New York: Dover Publications. Moseley, M. 1964. Irascible Genius: A Life of Charles Babbage, Inventor. London: Hutchinson. Prendergast, R. 2007. ‘Knowledge and Information in Economics: What Did the Classical Economists Know?’ History of Political Economy, 39: 679–​712. Prendergast, R., 2010. ‘Accumulation of Knowledge and Accumulation of Capital in Early Theories of Growth and Development’, Cambridge Journal of Economics, 34: 413–​31.

260

261

Charles Babbage’s economy of knowledge Prendergast, R. 2017. ‘Charles Babbage 1791–​1871’ in R.A. Cord (ed.) The Palgrave Companion to Cambridge Economics, Vol. 1. London: Palgrave, pp. 275–​96. Romano, R.M. 1982. ‘The Economic Ideas of Charles Babbage’, History of Political Economy, 14(3): 385–​404. Rosenberg, N. 1994. ‘Charles Babbage: Pioneer Economist’, in N. Rosenberg, Exploring the Black Box: Technology, Economics and History. Cambridge: Cambridge University Press, pp. 24–46. Schaffer, S. 2003. ‘Paper and Brass: The Lucasian Professorship 1820–​39’, in K.C. Knox and R. Noakes (eds) From Newton to Hawking. Cambridge: Cambridge University Press, pp. 241–​93. Smith, A. 1924. ‘An Inquiry into the Nature and the Causes of the Wealth of Nations’, 2 Vols. (edited by E.R. Seligman). London: J.M. Dent &Sons Ltd. Smith, A. 1976. An inquiry Into the Nature and Causes of the Wealth of Nations, general editors R.H. Campbell and A.S. Skinner textual editor W.B. Todd. Oxford: Clarendon Press. Snyder, L.J. 2011. The Philosophical Breakfast Club. New York: Broadway Paperbacks. Spence, M. 1973. ‘Job Market Signalling’, The Quarterly Journal of Economics, 87(3): 355–​74. Solow, R.M. 2000. Growth Theory: An Exposition. Oxford: University Press. Stiglitz, J. 1987. ‘The Causes and Consequences of the Dependence of Quality on Price’, Journal of Economic Literature, 25(1): 1–​48. Swade, D. 2003. Calculation and Tabulation in the Nineteenth Century: Airy versus Babbage. Thesis submitted for the degree of PhD University College London. Wilkes, M.V. 1977. ‘Babbage as a Computer Pioneer’, Historia Mathematica, 4: 415–​40.

261

262

15 MENDEL ON DEVELOPMENTAL INFORMATION Yafeng Shan

Introduction1 To date, Gregor Mendel (1822–​1884) is still widely credited as the father of genetics and his paper, Versuche über Pflanzen-​Hybriden (1866), is viewed as the founding document of the modern study of heredity. However, the historiography of Mendel has changed dramatically for the past five decades. Mendel’s paper is no longer simply viewed as an attempt to study the problem of heredity. It is now a consensus that Mendel’s concern, literally speaking, is about the development of hybrids in their progeny (e.g. Müller-​Wille and Orel 2007; Gliboff 2013; Zhang, Chen, and Sun 2017). It has also been widely received that one of Mendel’s most important contribution to the history of genetics is his novel work on developmental series (e.g. Olby 1997; Dröscher 2015). This view is well evidenced by the fact that much of early Menelians’ work in the 1900s focuses on the retrodiction (i.e. the re-​analysis of the pre-​exist data with Mendel’s approach) (e.g. Müller-​Wille 2005). However, many issues on Mendel and his work remain puzzling: There is no consensus on what Mendel meant by development (e.g. Gliboff 1999; Sandler 2000). Nor is there an agreement on the interpretation of Mendel’s laws of developmental series (Orel 1996, 1998; Gliboff 1999, 2013; Wood and Orel 2005; Szybalski 2010; Dijk, Weissing, and Ellis 2018). This chapter revisits three issues: What is developmental information meant by Mendel? In what sense is Mendel’s conceptualisation of developmental information novel and important? Were Mendel’s laws of developmental series about heredity? The second part explores the research context of Mendel’s notion of development. The third part examines Mendel’s notion of developmental series. The fourth part analyses the novelty of Mendel’s work on developmental series. The fifth part discusses the implication of Mendel’s laws of developmental series to the history of genetics.

Mendel and Gärtner on development (Entwicklung) Mendel explicitly stated his concern in the introductory remarks (Einleitende Bemerkungen). Artificial fertilization undertaken on ornamental plants to obtain new color variants initiated the experiments to be discussed here. The striking regularity with which the

262

263

Mendel on developmental information

same hybrid forms always reappeared whenever fertilization between like species took place suggested further experiments whose task it was to follow the development of hybrids in their progeny. … That no generally applicable law of the formation and development of hybrids has yet been successfully formulated can hardly astonish anyone who is acquainted with the extent of the task and who can appreciate the difficulties with which experiments of this kind have to contend. A final decision can be reached only when the results of detailed experiments from the most diverse plant families are available. Whoever surveys the work in this field will come to the conviction that among the numerous experiments not one has been carried out to an extent or in a manner that would make it possible to determine the number of different forms in which hybrid progeny appear, permit classification of these forms in each generation with certainty, and ascertain their numerical interrelationships. It requires a good deal of courage indeed to undertake such a far-​reaching task; however, this seems to be the one correct way of finally reaching the solution to a question whose significance for the [developmental]2 history of organic forms must not be underestimated. The paper discusses the attempt at such a detailed experiment. (Mendel 1866, 3–​4, 1966a, 1–​2) It seems that the problem of heredity was not Mendel’s concern. Vererbung, the German word for “inheritance” or “heredity” does not appear in the introductory remarks at all. More surprisingly, it is absent in the rest of the paper, except that Mendel used the verb “vererbt (inherited)” once.3 In contrast, there are two other key words I found. The German word Hybriden (hybrids) remarkably appears 101 times. In addition, Entwicklung is another key word, appearing 45 times in the paper.4 These two key words are highly suggestive. They reflect the objective of the paper: to study “the development of hybrids in their progeny”. More precisely speaking, the objective is an attempt to formulate a generally applicable law of the development of hybrids in their progeny by a detailed experiment. The key words also suggest the research context of Mendel’s work. As many historians (e.g. Olby 1979, 1985; Brannigan 1979; Müller-​Wille and Orel 2007) have already argued, Mendel’s work was well within the tradition of hybridism.5 This is also well corroborated by the references that Mendel made in the paper. In the paper, there are only five scholars whose works are mentioned: Kölreuter, Gärtner, Herbert, Lecoq, Wichura. Remarkably, all of them were important figures of hybridism. In Mendel’s words, they all had “devoted a part of their lives to” the problem of the development of hybrids in their progeny (Mendel 1866, 3, 1966a, 1–​2). What is more, in the concluding remarks (Schluss-​Bemerkungen), Mendel himself clearly identifies that his work on Pisum (i.e. peas) is within the “field” of the hybridist tradition, led by “two authorities” Kölreuter and Gärtner,6 and makes a lengthy comparison of his work with theirs (see Table 15.1). Although it is now a received view that Mendel’s concern was about the development of hybrids in their progeny, there is no agreement on the interpretation of Mendel’s concern. Given its significance, I find it necessary to make clear the meaning of Mendel’s Entwicklung at first. In Mendel’s paper, Entwicklung appears 22 times as an independent noun, and 23 times as an element in a compound word (e.g. Entwicklungsgeschichte, Entwicklungsreihe(n), and Entwicklungs-​Gesetz). As a noun, it is usually translated as development, though occasionally as formation (Mendel 1966a, 33, 43) or constitution (Mendel 1965, 19, 38). Remarkably,

263

264

Yafeng Shan Table 15.1  Cited Scholars in Mendel’s paper (1866)

Cited scholar

Number of occurrences

Kölreuter Gärtner Herbert Lecocq Wichura

6 18 1 1 3

Entwicklungsgeschichte is not accordingly translated as developmental history or the history of development. Rather it is typically translated as evolutionary history (Mendel 1966a, 2, 41) or the history of evolution (Mendel 1965, 35).7 Such a translation leads to a once popular but mistaken reading of Mendel’s work and its context. For example, based on the English translations of the same passage in which Entwicklungsgeschichte is translated as evolutionary history, Margaret Campbell (1982, 40) and L. A. Callender (1988, 51) take for granted that Mendel’s work should be understood within a Darwinian evolutionary context. Such a reading is too arbitrary, however. It is problematic to conflate the 19th century German word Entwicklungsgeschichte with the 19th century English word evolution without argument. As Robert Olby (1997) points out, “it is very misleading to transpose Mendel’s work from its source in the Austro-​Hungarian empire to the world of Darwinian debates in Victorian England and America”. Thus, in order to figure out the very meaning of Mendel’s Entwicklung, it seems necessary to locate it in its intellectual context. Sander Gliboff (1999) proposes that Mendel’s use of Entwicklung was directly influenced by Franz Unger.8 For Gliboff, the connection between Unger and Mendel is both intellectual and sociological. It is recorded that Unger taught Mendel botany at University of Vienna in 1851–​1853. Unger is also thought to be one of the people to whom Mendel sent an offprint of his paper.9 Both of Unger and Mendel were connected to some same academic associations (e.g. the Society of Naturalists in Brno and the Zoological-​Botanical Society in Vienna) and involved in some academic activities (e.g. a project of surveying the sprawling Habsburg Empire). Thus, Gliboff (1999, 226) argues that under the influence of Unger, Mendel refers Entwicklung to both “the individual ontogeny and the evolution of lineage”.10 A glimpse of Unger’s work seems to be compatible with Gliboff’s conclusion. Entwicklung is also a key word in Unger’s work, appearing 21 times in Botanische Briefe (Unger 1852a) and in 47 times in Versuch Einer Geschichte der Pflanzenwelt (Unger 1852b). More surprisingly, I found that the phrases Entwicklungsgeschichte (nine times),11 Entwicklungsgesetze (once), and Entwicklungsreihe (once) were already used by Unger. Thus, it seems that Gliboff’s argument that Mendel’s use of Entwicklung was Unger-​oriented is plausible, given the connection between Unger and Mendel. If so, another puzzle arises. If Mendel’s concern was directly influenced by Unger, why was Unger not cited or mentioned at all in Mendel’s paper? A more careful reading of Unger’s work (1852a, 1852b) and Mendel’s paper (1866) suggests something more complicated. It is clear that Unger and Mendel used the terms Entwicklung, Entwicklungsgeschichte, Entwicklungsgesetze, and Entwicklungsreihe differently. Although both of Unger and Mendel used Entwicklung to designate individual ontogeny in some cases, Unger talked much of it at the cellular level (e.g. Unger 1852a, 104, 106, 112) while Mendel at the morphological level (Mendel 1866, 8, 11). Unger (1852a, 110) referred Entwicklungsreihe to a

264

265

Mendel on developmental information

series of developmental process, while Mendel (1866, 5, 20, 21, 22, 24, 29, 31, 35, 39, 40) refers Entwicklungsreihe to the numerical relationships of hybrid forms. Unger’s Entwicklungsgesetze (laws of development) was about the whole plant world (Unger 1852b, 282), while Mendel’s was specifically about hybrids (Mendel 1866, 18, 32). Mendel (1866, 4) was explicit on the point that his work on Pisum is significant for “die Entwicklungs-​Geschichte der organischen Formen (the developmental history of organic forms)”, but it is too hasty to conclude that this was related to Unger’s general interest of “die Entwicklungsgeschichte der Pflanzenwelt (the developmental history of plant world)”, especially given that Me