The Republic of Science: the Emergence of Popper's Social View of Science 9042015152, 9789042015159

Citation preview

SERIES IN THE PHILOSOPHY OF KARL R. POPPER AND CRITICAL RATIONALISM SCHRIFTENREIHE ZUR PHILOSOPHIE KARL R. POPPERS UND DES KRITISCHEN RATIONALISMUS Herausgegeben von Kurt Salamun

BAND XV

Amsterdam - Atlanta, GA 2001

Ian C. Jarvie

The Republic of Science The Emergence of Popper's Social View of Science 1935-1945

The paper on which this book is printed meets the requirements of "ISO 9706:1994, Information and documentation - Paper for documents Requirements for permanence". ISBN: 90-420-1515-2 ©Editions Rodopi B.Y., Amsterdam - Atlanta, GA 2001 Printed in The Netherlands

For JB d.d.d.

CONTENTS Preface

5

Chapter 1 Introduction: Science as an Institution

9

1. From Science to Politics 10; 2. The Problem of Demarcating Science 11; 3. Science Demarcated Socially 16; 4. Consequences of the Social Turn 19; 5. The Underlying Model of the Social 23; 6. Problems of Interpretation 31 Chapter 2 Popper's 1935 Proto-constitution for the Republic of

Science

35

1. The Received View of Popper on Science 36; 2. Methodology and Social Practice 40; 3. Methodology in Detail 51; 4. Assessment 63; 5. Johansson's Critique 70; 6. Polanyi, Kuhn, Feyerabend 73; 7. Hattiangadi's Critique 78; 8. Autocritique 81; 9. The Social Aspects of Scientific Method 86

Chapter 3 Problems in a Science of Social Institutions

88

1. The Logic of Social Scientific Discovery 88; 2. Johansson's Claim 94; 3. Social Institutions 105; 4. Individuals 123; 5. Rationality 129 Chapter 4 An Enriched View of Institutions

138

1. The Story The Open Society Tells 140; 2. Popper's Ideal Types: Closed and Open, Concrete and Abstract 143; 3. A Critical Difficulty: Nature and Convention 153; 4. Rationality and Social Reform: Piecemeal versus Utopian Social Engineering 167

Chapter 5 Science and Society as Learning Institutions

178

1. The Nationalist View of Social Life 182; 2. The Collectivist View of Social Life 188; 3. Knowledge as a Social Institution 199; 4. Rationality in Science and in Society 207

Chapter 6 Conclusion: The Republic of Science

212

1. Democratic and Authoritarian Republics 213; 2. Rational Reform of Science 221; 3. Science as Socially Exemplary 224; 4. Science Among the Institutions 227 Sources and References

232

Index of Names

244

Index of Subjects

248

PREFACE Sir Karl Popper's originality as a philosopher of science is usually attributed to his stress on criticism and falsification as routes to knowledge. His originality as a philosopher of society and politics is usually presented as an extension of those epistemological and methodological ideas into what might be termed a defensive view of political life. It was prudent, he famously argued, to act as if tendencies to despotism and tyranny were endemic; to treat all government, indeed all power, with suspicion; and to create a political system in which the government could be changed by the governed without resort to violence. While providing no guarantee of freedom and openness, this was the best available insurance against their erosion. Without denying the truth of this usual account of Popper's originality, I shall argue in this book that it sells short Popper's positive and original ideas about science and about society, ideas that fit snugly together with his negativist epistemology. Taken as a package, his ideas offer us means of thinking about society and the special place of science in it that do not reduce to, or explain away one by, the other. The Introduction (chapter 1), sets out the main lines of debate about the relation between science and the social order. To present what I hope is my enriched account of Popper's ideas on these matters I proceed by examining with some care (chapter 2) the social picture of science presented in his classical work on scientific method, The Logic of Scientific Discovery (LScD) , seeking in particular to bring out his thinking about social relations and the character of social institutions. His argument, in brief, is that traditional problems of epistemology need to be reformulated socially. Most intriguing, I believe, is the necessary connection he draws between social interaction and the acquisition of knowledge. Acquisition of knowledge is not

6

THE REPUBLIC OF SCIENCE

possible without the interaction of social life, and social life is itself a fonn of acquiring, preserving, and transmitting knowledge. Anned with that result, I then tum (chapters 3, 4, and 5) to the two works on history, politics, and social science that Popper completed towards the end of the Second World War, the article series entitled, "The Poverty of Historicism" (1944-45; book version 1957 (PoHIPoH)), and the two-volume treatise, The Open Society and Its Enemies (1945 (OS&I£»). My aim is to show how over ten years of fruitful work he remained engaged with the social aspects of the quest for knowledge, working out, supplementing, and generalising ideas about society that were embryonic in his philosophy of science. A fully fleshed-out vision of politics and society emerged which today would be tenned the knowledge-using, knowledge-gathering, knowledge-embodying society. It is, in many ways, a magnificent vision and an intellectual tour de force. Popper presents society as a learning machine, and the learning machine as a Socratic seminar writ large. Wholehearted though my admiration is for this vision, I take seriously Popper's injunction that we learn from criticism. In criticism I argue that there are serious deficiencies in holding up the Socratic seminar as a model for life in society in general, and for scientific work in particular. The republic of science Popper envisions is susceptible to the corruptions of power just as is the broader body politic, and its citizens need the kind of checks on power that Popper demanded on governments. Even more troubling: the central Socratic and scientific value of truth conflicts at times with other values important for social life. These difficulties lead us back to a reconsideration of Popper's philosophy of science, both its adequacy and its emphasis: in particular its abstraction from and typification of the real historical facts about science, and its ambiguous status as both nonnative and descriptive. More narrowly, we can revisit the relationship between the embryonic institutionalism of LScD and the full-fledged institutionalism of PoH and OS&IE. Institutions do not maintain themselves; still less do they refonn or improve themselves. If

PREFACE

7

science is necessarily institutional then OS&IE in particular is a plea for a wholly new attitude to it. If science is institutional it faces all the difficulties of society in general. If objectivity, progress, rationality, and experience require institutions, those institutions need the same tender loving care as any other institutions. Science as institutions is not self-maintaining. It is not the product of great men. It has no authority. It is vulnerable to all the weaknesses of our other social arrangements. Hence the skeletal sketch of its social aspects in LScD was not, by Popper's own later social thought, critical enough. If they are to fulfill the tasks we set them, it is vital that scientific institutions get improved design, better staffmg and maintenance, and, in some cases, major reform.

*

In working on this book over the years I have many debts to acknowledge. Joseph Agassi, Peter Munz, and Jeremy Shearmur read a very early draft and gave me many critical suggestions. Fragments of the argument were tried out in papers given on divers occasions, but especially in Mexico (Coloquio International Karl Popper at UNAM) , Sweden (the Olaf Erik Burman lecture at Umea University), Vienna (a Ringvorlesung Lecture at the University of Vienna), Prague and Budapest (lectures at the Central European University). Jeanette Bicknell, Joseph Agassi, and Malachi Hacohen gave the penultimate version a close, critical reading. However the result is judged, it would have been much worse without the thoughtful attention it has received.

*

For a sabbatical leave from York University in 1993-1994 I am most grateful. In the three years preceding I was greatly assisted with research assistance and travel to the Hoover Institution at Stanford, where Popper's manuscripts and correspondence are archived, by a research grant from the SSHRC (410-91-1148). Two research assistants, Doug Gardner and Stephanie Morgenstern, compiled a secondary bibliography and translated some materials. Subsequent work to bring it to completion was partially done while a Visiting Professor at Doshisha University, Kyoto, Japan, an exceptionally congenial environment in which

8

THE REPUBLIC OF SCIENCE

to work. To those colleagues and students at various institutions in Japan and Sweden who gave me the chance tryout these ideas, and who gave me their critical reactions, many thanks. Toronto, March 2001

Chapter 1 INTRODUCTION: SCIENCE AS AN INSTITUTION Science is a special kind of endeavour. Philosophers have been seeking to specify in what· ways it is special for almost four hundred years (at least since Francis Bacon). Karl Popper is known for his attempt to characterise it negatively, as refutable conjectures about the world. But this is not the sum of what he said. What is and is not refutable requires discussion, which is a social process; if conjectures about the world are to be refuted a social framework is needed for their testing. Popper's negative method requires institutions and traditions. This book takes a close look at Popper's social view of science. He viewed science as a republic of sorts with specialised institutions to keep inquiry free and critical. Free and critical inquiry was good, he claimed, not just for science but also for society at large, especially when considering social reform. This idea is both exciting and flawed. In criticism of Popper I argue that the wider context of society at large differs from the narrower context of the republic of science in ways that nUllify some of the latter's republican virtues. Moreover, by application of his own arguments, institutional science, just as much as society at large, needs maintenance and reform. Although he wrote little or nothing to this effect, his principles of social and political thought can be applied to the task. I work out these claims in the form of a commentary on selected parts of Popper's work: 1935-1945. First, I show how his book on scientific method, The Logic of Scientific Discovery (LScD) , has a social tum at its centre, and also contains a rudimentary sociology of science. Second, I show how the 19441945 works, "The Poverty of Historicism" ("PoH" /PoH)1 and 1. "PoH" designates the original journal articles; PoR the 1957 book version. There are often small improvements of wording between the two. These will be

10

THE REPUBLIC OF SCIENCE

The Open Society and Its Enemies (OS&IE) , go far beyond Popper's earlier, rather skeletal, analysis of the social aspects of scientific method. Third, I show that as an unintended consequence of his new ideas on society and politics he develops what amounts to a critique of his own earlier thought. In particular, science viewed socially is by no means a model for the rational approach to society and politics generally, and, reciprocally, the problems he identifies in society and politics are present in science and need remedy. I. From Science to Politics In 1935 Popper published the German original of his of his great work on scientific method, Logik der Forschung (LdF).2 Austere and terse, its main focus appeared to be the narrowly philosophical problems of demarcation and induction. Only a couple of years later Popper emigrated from Austria to the comparative safety (for someone of Jewish descent) of New Zealand. When the unification of Germany and Austria was declared in 1938, he tells uS,3 he decided to write about a very different topic: the background to the European political catastrophe. Over seven years later the result was published, a two-volume treatise on history, politics, and society entitled The Open Society and Its Enemies (1945). Unlike LScD, it is a long, passionate book that presses the Enlightenment project of maximising the use of reason in improving social and political life. Although the two books might seem very different, they were connected at a deep level. Throughout OS&IE Popper takes as his inspiration the critical rationalism of Socrates: egalitarian, cooperative, open-minded, critical, reasonable, and submissive only to the truth. Popper found the best modem realisation of this model of the rational way of life in science. Indeed, he claims that science institutionalised cooperative critical discussion in a noted only when they seem substantial. Changes of wording in successive reprints of the book will be ignored. 2. Popper made a point of the fact that his book appeared in late 1934, with the imprint 1935. 3. Popper 1945, Preface to the Second Edition (1952), p. xviii.

INTRODUCTION

11

way that set it apart from and perhaps above other social institutions. This amounted to generalising a central plank of The Logic of Scientific Discovery. Thus, in a major work of social and political theory, Popper proposed to take science as a guide even as he allowed that science was special amongst social institutions. That science is special has been claimed to be the case since the Scientific Revolution, and even before. How it is special has been the subject of much debate. Among the features of science that have been nominated as demarcating it from other endeavours are its subject matter (Nature), the character of its results (universal, repeatable, reliable, authoritative), its methods (empirical, inductive). Popper's stress on its special character as social institution was a move already in the air. Unknown to him, in the 1930s the young American sociologist Robert K. Merton was also examining the special character of the institutions of science, founding thereby a tradition of sociological research which, in an unexpected twist, thirty years later would call forth attempts to normalise science into a social institution like any other (the Strong Programme). 4

2. The Problem of Demarcating Science Let us analyse the problem of demarcation a little more, since it was central to Popper's concerns. Science consists of ideas; the institutions of science consist of people and their organised social relations. Ideas are formulated in statements or sentences; statements or sentences are possible because of the social institutions of language and of concepts. Thus we can put matters 4. Steve Fuller intriguingly suggests that institutionalism was in the air among the logical positivists: "it is clear that the positivists regarded the emergence of cumulative knowledge growth as a remarkable occurrence in the history of humanity that requires special institutions whose long-term maintenance cannot be taken for granted but must be subject to eternal vigilance" (Fuller 2000b, 29-30). As evidence he cites his own work (1996) and there points to some remarks by Schlick and Carnap which he takes to reveal a hidden part of the history of the logical positivist movement. If Fuller is correct, then Popper shared those concerns and took them much further, but he and his commentators failed to bring out how early on he was institutionalist, and how deeply.

12

THE REPUBLIC OF SCIENCE

a little more sharply: there are at least two distinct problems of demarcation of science from non-science. There is the logical or methodological problem of demarcating scientific statements from other statements, including those bearing a superficial resemblance to scientific statements (pseudo-science), and those that antedate science (folk notions). There is also the sociological problem of demarcating science as an institution from other social institutions. Because of the influence of Sir Francis Bacon (c. 1620), these two problems have often been fused. Bacon solved the methodological problem of demarcation by attributing to science a special method, a qualified form of induction. He solved the sociological problem of demarcation by arguing that scientific investigation required building specialised, cooperative, and democratic social institutions - what we would now call research teams. 5 Though Bacon's influence on the organisation of science and on views of scientific method was extensive, the social aspect of his work was hard to appreciate until method could be seen as social, and the social character of science could be seen as part of its method. Both ideas go to the credit of the near-contemporaries Karl Popper (LdF, 1935) and Michael Polanyi (The Logic of Liberty, 1947; Personal Knowledge, 1958).6 Credit for the idea that method was part of the sociology of science must be shared with Merton. His early researches (Merton 1936; 1938) are traditionally sociological; the sociology of science is traced to his 1942 claim ("Science and Technology in a Democratic Order") that the scientific community was to be 5. Bacon thought that a few years' work by several coordinated teams would succeed in writing down the complete book of knowledge. This equates the writing of science with reconstructing the Book of Nature, whose author was, of course, God. 6. Polanyi developed his view in the wake of Popper's and in contrast to it. The development was not conscious: though Popper made the change to a social approach explicit in The Open Society and Its Enemies (1945, ch. 24), and though Polanyi stressed that his views of scientific method were not abstract but related to the concrete scientific community, Polanyi' s leading follower, Thomas S. Kuhn, whose 1962 monograph The Structure of Scientific Revolutions had enormous impact, confessed surprise when told his view amounted to a sociology of science.

INTRODUCTION

13

characterised by its methods, viz. universalism, disinterestedness, communism of ideas, and organised scepticism. (He referred to these as "values".) If science is to be characterised by its methods, and its methods are embodied in forms of social organisation, then we have a complementary convergence. Popper, Polanyi, and Merton do not reduce science to social factors; none of them holds science to be just another social institution. Popper offered neither sociological nor historical generalisations about science. He offered methodological rules as legislated norms, not empirical ones. Both Popper's originality and certain problems of his reception are illuminated, I contend, by viewing him as a sociologist of science in spite of himself. Popper discovered that in purely logical or methodological terms the problem of demarcation was insoluble. Yet the distinctiveness of science as intellectual activity was an undeniable (social) fact. It followed that the problem of demarcation was misformulated and needed to be recast. Only by taking account of the fact that there were inescapable choices involved in methodology, choices that structured the social institution(s) of science, was Popper able to get around the fundamental logical impasse. Although Popper was astute at working out many of the consequences of his central ideas, I believe that he never sufficiently emphasised this social tum. Almost in passing he allows that science is social because it is necessarily couched in language (PoH, p. 156; Economica 12, 1945, p. 87). This was by no means an uncontroversial position. Empiricists, for example, sought to anchor science in subjective experience. Popper's positivist colleagues, as part of their linguistic tum, anchored it to reports of subjective experience, or protocol statements. Both groups persisted in naturalism. What Popper did not quite say is that the distinction nature/convention was standardly taken to be the same as law-governed (rational)/arbitrary (not rational). His own later views make this a false contrast. He treats social institutions as conventional but by no means arbitrary and so steers a middle way: institutions are not natural, they are not conventional in the sense of arbitrary, they are rational and even

14

THE REPUBLIC OF SCIENCE

subject to laws of a sort.7 Popper might not have been too happy about my highlighting his social tum, and especially with my choice of terminology. Yet I am convinced that this view of matters does his ideas justice, and opens the way to a better understanding of the incomprehension with which they have so often been met, on the part of his critics as well as many of his self-styled followers. He discovered that sociology was transcendentally necessary. This was too bold and radical a move for most philosophers (cp. Bunge 1999 for different arguments). More important, once seen as a sociologist of science malgre lui, Popper can also be seen as a sociologist tout court, one with a bold and elegant vision of the social world. His vision is deeply informed by Marx's dictum that consciousness does not determine social existence but social existence determines consciousness. By the standard set in Popper's later discussion of the importance of institutions his skimpy analysis of the institutions of science is a flaw in The Logic of Scientific Discovery. I am alluding to the penetrating discussion in The Open Society and Its Enemies of the design and maintenance of institutions, Popper's argument that all long-term politics is institutional, and his firm methodological and moral commitment to the primacy of the individual human being in social calculation. Yet throughout his writings Popper refrained from developing the implications of these points for science. Instead, his ideas on the social character of science emerge glancingly, in a critique of another attempt to demarcate science, that of Karl Mannheim (1936, 1940). In chapter 24 of The Open Society and Its Enemies, Popper discussed "the sociology of knowledge". This phrase was a misnomer applied to sociological attempts to explain (not knowledge but) folk ideas on the way the world works, including and especially the way the political world works. Between the wars Mannheim, the target of most of the discussion, had, along with Max Scheler, suggested that people's opinions on social and political matters were conditioned by the 7. In both PoH and OS&IE Popper vacillates over the status of sociological laws. See also ch. 4, §3.

INTRODUCTION

15

social atmosphere and in particular by their class position. This conditioning is invisible to those who hold the opinions which, indeed, they take to be obvious and true. Yet persons positioned elsewhere in the social system, not to mention those from other social systems, will see nothing obvious or true in these opinions. Each socially determined system of ideas about social and political matters they called a 'total ideology' . Merely articulating Mannheim's doctrine presents some difficulties, since it offers an opinion on how the social and political world works and so should, by its own lights, be itself treated as a product of social and political conditioning. No surprise then that Mannheim makes the intellectual classes to which sociologists and other scientists belong an exception to the social determination of ideas. This 'freely poised intelligentsia' has come loose from the social and political conditioning that works on the rest of mankind and so its social, political and scientific opinions are no longer part of the natal 'total ideology'.8 A central task facing this liberated intelligentsia is unmasking and demolishing the total ideologies to which the rest of humanity is subject. Popper refers to Mannheim's sociology of knowledge ironically as a kind of socio-analysis, even socio-therapy. He points out that the sociology of knowledge so conceived is no more than an unacknowledged reiteration of Hegel and hence a prime example of an unconscious total ideology. In other words, Popper claims that the originators of the sociology of knowledge did not demarcate science, natural or social, in a manner that immunised it from the argument that it is just another total ideology subject to the social and political conditioning which governs all ideas and opinions. It is not too surprising, then, that Popper's semi-ironical argument has been taken quite seriously by proponents of the so-called strong programme in the sociology 8, Fuller (2000b, 19041) offers an interesting explanation of why Mannheim exempted mathematicians and natural scientists from the social determination of knowledge. He did not want the sociology of knowledge "to be associated with doctrines of racially grounded science" that were in the air. This does not quite explain why Mannheim exempted the free-floating intelligentsia as a whole nor does it bring out that the price of the exemption was the loss of consistency.

16

THE REPUBLIC OF SCIENCE

of knowledge (SPSK). Merton's move into the sociology of science stemmed from a modification of Mannheim: if folk ideas and common sense have a social background, then so do scientific ideas. This allows the sociology of knowledge to include the sociology of science. Did Merton, then, anticipate the Strong Programme? Only if social background is sufficient to explain scientific ideas, not if it is merely necessary. Merton hesitated; the proponents of the Strong Programme had no hesitation. The phrase 'total ideology' politicises the thought of ordinary people. Had Mannheim, Merton, or Popper, faced the consequences of the fact that there are schools of thought and political parties in science they could have preempted the Strong Programme. Popper never discusses the politics of science although its existence makes no difficulties for his system. Taking extra inspiration from Durkheim's attempt to show that the categories of time, space, and causality are social constructions, SPSK has held that all ideas and opinions, including natural and social science, including the sociology of knowledge and SPSK, are part of total ideologies subject to social, political, and especially class, determinations (Bloor 1976, 1981; Barnes 1977, 1982; Rudwick 1985; Shapin 1994).

3. Science Demarcated Socially

Now let us look at matters socially and historically rather than from the philosophical perspective, beginning with Merton's work. Proceeding less abstractly and more historically than Popper, Merton examined the relation between military imperatives and scientific/technological invention in the seventeenth century (Merton 1938), and then turned his attention to the current embodiment of scientific endeavour in American society (Merton 1942). He pictured science as being distinctive less in its institutional structure and more in its governing methodological ethos, the components of which he identified as: Universalism (the use of preestablished, impersonal criteria of assessment) Communism (findings are the product of social collaboration and are assigned to the community) Disinterestedness (in its usual meaning)

INTRODUCTION

17

Organized scepticism (methodological and institutional mandate) The four are not wholly distinct, and they mix norms and ideals with their institutional embodiment. Nevertheless, this list has not been bettered, even by Merton himself, in more than fifty years. His claim that the norm of disinterestedness is exemplified by the relative absence of fraud in science may have been overoptimistic. It remains a fact that both the statistical and the ethical norm is disinterestedness, even if there are those who disregard it. 9 Merton displays no awareness of Popper's 1935 work, even though he published an earlier version of his 1942 work in the journal Philosophy of Science in 1938, which suggests that he was browsing the literature of that field. Popper did not know of Merton's work either, and made no allusion to it in any of his published works. Of course, much the same goes for Popper's treatment of the history of science, his references seldom engaging with particular historians and their discoveries, rather with received material about the scientists themselves. Be all that as it may, my claim is that both Popper and Merton worked on the problem of demarcating science from other social institutions, even though each began from a different problem-situation. While there can be little doubt that Merton, both as sociologist and as observer of science, did not think his own results were merely the expression of some form of total ideology, his four components of the ethos of science were such that the possibility remained that each could be found elsewhere than in science, and even that the combination might be found in other social institutions. So Merton's analysis left open the possibility that science might be just one more social institution, and thus vulnerable to the strictures of the SPSK. A very good reason why Popper did not think of himself as a sociologist of science was precisely that he thought there were 9. Fuller (Fuller 1997, 2000a, 2000b) has been critical of Merton for typifying, perhaps even idealising, science in his account of its ethos. Fuller chastises Merton for basing his analysis upon conversations with scientists, rather than on extensive empirical research to check on whether scientists accurately report their practice. See my 2002 for discussion.

18

THE REPUBLIC OF SCIENCE

logical aspects as well as social aspects to scientific method, and that while the social aspects were by no means unimportant, the logical problems were deeper. I will now take this into my account. In characterising Popper's position as a republic of rules I must face the question of how this republic differs from others, e. g. the republic of letters. It is a question not amenable to a crisp logical solution. A sociological demarcation of science cannot but be open-ended, since institutions can and do change. For another, past institutions of science scarcely resemble today's Big Science. Despite these difficulties, Popper's position needs a sociological demarcation. Although science is a social institution, it is not just another social institution. The social institution of science is obedient to special imperatives because it undertakes to explore and explain the world, to in some sense use the world as the ultimate test of its results. Science institutionalises what we know as empiricism, the attempt to learn from experience. The spine of learning from experience is logic: if we have two inconsistent statements, one of them must be false, its negation true. This uncomfortable logical fact, so often allowed to be disregarded or even flouted by other social institutions (e.g. in politics and religion), has to be taken seriously in science. The republic of science makes it difficult to evade or flout contradiction, hence the detection of one is an incipient crisis. 10 Notice, further, that science qua social institution embodies, indeed gives central place to, a logical fact: a fact about language. Scientific ideas and results are articulated in language, and those statements are subjected to the laws of logic. Logical fact is not social fact, even though logicians are human, and systems of logic are social institutions. Logical fact is no more social than 10. Were the republic of science as easy-going about contradictions as other social institutions it would lose much of its distinctness. Kuhn (1962) argues only accumulated contradictions when judged excessive constitute a crisis. Lakatos (1970, 1974) argues that only contradictions that hit the 'hard core' of a research programme create a crisis. These and other authors are offering political models of science where contradictions are negotiated pro tern. They confuse intellectual crisis with means of coping with, including agreeing not to recognise, crisis.

INTRODUCTION

19

mathematical fact is social or physical fact is social. ll In this, as in much else, Popper is the only sociologist of science who articulates deep, abiding, and unanswerable objections to the Strong Programme. The logical facts of science were for Popper the most important ones since they were the most puzzling and most disputed ones. They help make science socially special. Further consideration of the special character of science among social institutions will be given in chapter 6, §4. To sum up, I argue that Popper worked out a rudimentary sociology of science in the first instance because of the important logical discovery mentioned earlier: insuperable difficulties thwarted any purely logical solution to the demarcation problem. Instead, science had to be demarcated as the institutionalised choice of certain rules. The methodological choices we make (under the constraint of logic) constitute an institutional arrangement. The mode of inquiry, method, or social institution we call science is an emergent from those choices. 4. Consequences of the Social Turn Although Popper often tied together aspects of his own work that had originally been pursued in isolation, he never brought together the sociological character of his demarcation of science with his own later argument that there could be no solitary or Crusonian science. He made this argument in his 1945 book The Open Society and Its Enemies, in the course of criticising sociologists of knowledge for overlooking the social aspects of scientific method: especially free criticism and the avoidance of talking at cross purposes - both being forms of institutionalised cooperation that foster objective and public scientific results. (Recall Merton's four value-norms of science.) If science could be demarcated solely by means of logic, then any person who respected that demarcation could, in principle, pursue science by, for example, observing. Popper holds that such solitary endeavour is impossible: the public and objective character of scientific work is made possible by its being a social enterprise 11. Most authors who subscribe to SPSK would demur, and affirm just those equations my sentence denies.

20

THE REPUBLIC OF SCIENCE

and it is not sustainable otherwise (the argument against Crusonian science is discussed in detail in ch. 5, below). What characterises science on this view is a social decision to allow the world to correct our ideas, and collateral decisions to eschew attempts to insulate those ideas from inconsistency and counter-example. Those decisions are implemented socially in institutions that entrench rules of procedure or method. Thus and only thus is a demarcation established between folk wisdom, social and political opinions and the like, all equally institutional, and science. The difference is not in the results -- the claims about the world that issue from time to time from actors in the institutions. Indeed, these may at times look (misleadingly) similar. It lies in the methods by which work proceeds, where method here is construed not simply as a bunch of methodological rules, which is how it is usually presented, but as the deliberate choice and tailoring of methodological rules to specified aims. The choice is institution-building rather than personal. Science is to be identified with an institutionalised aim or set of aims, together with the institutional rules and practices that have been devised to foster pursuit of these aims. These rules serve specifiable purposes and are subject to revision depending upon whether they accomplish such purposes. (Yet Popper did not discuss the reform of scientific institutions, even though he sometimes deplored what science had become (Popper 1972).) It is these rules that enable science to pursue the goals of being objective, realistic, in touch with experience, explanatory, progressive, and causal. Obedience to these rules is what makes the institutions of science different from other social institutions, including those of pseudo-science, magic, logic, and mathematics. It is not doctrines, ideas, or sentences that are scientific or non-scientific; it is certain institutionalised ways of handling doctrines, ideas, and sentences that are scientific or nonscientific. The outcome of this effort is not guaranteed truths, but hypotheses that stand at the end of a chain of conjecture and refutation. The main result of the investigation is that we have a deeper and more humbling sense of the underlying problems than that- with which we started. In the view just outlined we have two results: first, a solution

INTRODUCTION

21

to the problem of demarcating science. We also have a rational, egalitarian, and democratic vision, a way of looking at institutions as means of coordinating the various tasks of human living, from the mundane ones of food and shelter, through to the elevated and even rarefied ones of exploring the world and making art out of it. Popper proposed an institutional approach to the philosophy of science and an intellectual approach to the reform of institutions. This formulation reverses the received wisdom which would suggest an intellectUal approach to the philosophy of science and an institutional approach to the reform of institutions. What I call Popper's 'intellectual approach to the reform of social institutions' is plainly articulated in The Poverty of Historicism and The Open Society and Its Enemies. Popper gives central place to the role of ideas in the explanation of social problems, and in efforts to correct social ills. Ideas are miniature institutions, linguistic entities to which we ascribe truth values; ideas inform and coordinate individual actions. Individuals and ideas are central to the governance and success of social institutions, thus is the locus of responsibility placed where it belongs, on acting individuals, rather than on abstractions such as social forces, economic pressures, and the like. The engine of change is not destiny or the forces of history, but shared ideas which conduce individuals to struggle to achieve specific goals, to implement certain values. These activating ideas are not, for Popper, mere 'subjective beliefs'. They are social institutions, and we can (subjectively) entertain ideas only because of the social institutions of language, discussion, and criticism. Thus does social existence determine consciousness. My argument will be that thinking socially (rather than logically or psychologically) is central to Popper's philosophical enterprise beginning with The Logic of Scientific Discovery, continuing for his ten most creative years, and emerging sporadically after that. Popper's consistent ability to think socially also does much to account for his originality, since it is hard to do and its difficulty is attested by how often readers and critics of Popper do not grasp that this is what he is doing. While I claim novelty for the present interpretation of

22

THE REPUBLIC OF SCIENCE

Popper's main intellectual contribution, novelty is not what recommends it. There are other qualities that recommend interpretations. Discussing in The Open Society how to assess differing interpretations of history, Popper offered stimulation and fruitfulness as positive qualities. Taking my cue from that, I suggest that the present social interpretation of his thinking is stimulating and fruitful in several ways. For one thing, it constructs a single line of thought connecting the three central works on the philosophy of science and politics, LScD (1935), PoHIPoH (1944-45) and OS&IE (1945). It is a rather different line from the one Popper himself suggested when he drew a parallel between criticism and refutation of scientific hypotheses and the proposing and practical testing of social institutions, viewed as hypothetical solutions to social problems. For another thing, this social interpretation illuminates an aspect of Popper's philosophy of science that is too seldom discussed, namely the status and function of methodological rules. The results of these inquiries include one unexpected bonus. My initial hope was that the social interpretation of Popper's philosophy of science presented here would help explain the sense of cross-purposes in exchanges between Popper and his critics. Operating very much from within Popper's philosophy, I empathised strongly with his discouragement and frustration at the poor level of criticism he received. Talented philosophers would offer critical comments that revealed poor understanding of his views; even when it came to trying to re-state these views adequately the deficiencies were glaring (Papineau 1995). Yet Popper's own explanations of all this - that standards were falling and that people no longer read carefully - never seemed convincing to me, being psychologistic, not institutional, and being, moreover, historicist. With the present results in hand I can perhaps offer a better explanation of the problem. One of the reasons his critics did not grasp the key moves of Popper's philosophy is that he, despite his enormous labours, did not grasp them all himself. He never presented his ideas as the sociologising of central philosophical problems, de-naturalising them - despite his writings on the autonomy of sociology.

INTRODUCTION

23

Popper discovered that a naturalistic approach to philosophical questions was a maze of forking paths and dead ends. On discovering this he built a second floor to the maze, his institutionalism, from which vantage point the whole maze and its cuI de sacs could be seen. He sent word to those still in the maze to build upwards too, but they did not understand - there had to be a way out of the maze at ground level, it stood to reason; there always is. Finally, if my social interpretation is persuasive, it offers a new point of view on the development of the philosophy of science and the philosophy of society in the last half century. The cost of neglecting Popper's proposed social tum is being borne by a philosophy of science that is stagnant and scholastic. If anything, the philosophy of society is in worse shape. 'Stuck in the maze' might characterise both. By contrast, Popper's vision of society as an experimental learning machine driven by the urgencies of human suffering, with an agenda that constantly changes because the imperatives never go away, is humane, responsible, and optimistic. Established mainstream philosophy of science and of society tries to give the impression that it has absorbed what is of value in Popper and moved on. Thus he and his followers are somehow out of step, possibly even outmoded. This Kuhnian, even historicist, argument is one I shall try to put to rest. Before expanding and arguing in the body of the book the points made so far, two further preliminary discussions are required. The first is that, if I am attributing a social tum to Popper, I need to say more about 'the social' (§5). The second is that the attribution of a social tum is an interpretation and needs to be placed in the context of the general philosophical problem of interpreting thinkers in general and Popper in particular (§6). 5. The Underlying Model of the Social A model or conception of society is a very general notion about the ontology of social things and how they connect. Any list of social things should include conventions, customs, institutions, markets, relations, selves, traditions, and the like. Two standard conceptions of the social are that society can be thought of as

24

THE REPUBLIC OF SCIENCE

resembling an organism and that society can be thought of as resembling a machine. From ancient times society has been conceived of as resembling an organism, functioning perhaps a little like the human body, with a controlling head and operating limbs. In a more recent variation, this social organism is envisaged as being the product of, and modified by, an on-going process of evolution. Culture theorists in anthropology went so far as to write of culture as an 'exosomatic organ' - an organ outside the human body yet essential to its functioning. 12 Social thought in the modem period has toyed with a rival conception of society as a machine, with parts, pulleys, levers, tubes, fluid flow-through, energy sources, and so on. Twentieth century electro-mechanical devices that involve feedback, cybernetics, robotics, and computers have been used to enrich the machine conception of society. William Harvey, Thomas Hobbes and Rene Descartes envisaged the human organism, the body, in mechanical terms - a system of levers, tubes, pumps and wires. So it was perhaps the mechanising of the organism that led the way to the mechanised view of society.1 3 Since society is manifestly neither organic nor mechanical, these two conceptions have all the power and all the limitations of similes or metaphors. They have the power to captivate and intrigue us, but sooner or later the parallel breaks down or we have to commit that cardinal sin: mixing metaphors. A conception of the social that avoids these difficulties is the city, another quite ancient conception of the social. Plato and Aristotle, St. Augustine, Rousseau, and Fustel de Coulanges, all contemplated the city as the conceptual key to the ontology of society. For them, a city was concrete and relatively compact. Much social life was contained within its walls. Why is this not the perfect conception? The answer is not difficult to find. Many aspects of society and social things are not to be found in the city 12. A.L. Kroeber's "the Superorganic" is a strong version of this (Kroeber 1917). Such a strong conception of culture was taken up in the United States by Robert Lowie and Leslie White. For a recent assessment see Kuper 1999. 13. What E. J. Dijksterhuis called the mechanisation of the world picture (Dijksterhuis 1961).

INTRODUCTION

25

- peasant and nomadic life would be examples, and there is much more to the social than is readily revealed by contemplating the city. Markets and patterns of trade, to mention two, exist both in concrete form in the city, and in much less obvious, abstract ways. International markets like those in metals and commodities, in stocks and shares, or in art and antiques, require taking into account matters far beyond the city. Much the same is true for such social entities as the wool trade of the late Middle Ages, for the Silk Road, or for the Hanseatic League. Before dismissing the City, it is important to consider its strengths as a conception of the social. At least cities are actual social entities - congeries of institutions - unlike biological organisms or physical machines. Moreover, the city does contain conventions, customs, institutions, markets, relations, selves, and traditions in one form or another. The difficulty just raised about it not helping us conceive the more abstract aspects of the social, such as markets that transcend the boundaries of the city, might lead us to think that the market would be helpful for conceiving of the social. Society would then be thought of as proliferated transactions, between persons, between persons and institutions, and between institutions, and the problems would be in specifying the equivalents of supply, demand, price, market clearance, knowledge, preferences, and so on. Like the city, the market is a social entity. But also like the city, to use it as a model for all society is, if not reductive, then excessively simplified. There may be market aspects to much of social life, but there are many other aspects as well, such as intellectual aspects, legal aspects, aesthetic aspects, which do not naturally reduce to market transactions (pace Gary Becker). Also, cities and markets are constituents of social life rather than constitutive of social life. A hunter-gatherer group living in caves and foraging each day for the evening meal is a society, but one without cities or markets (in any interesting sense). All this should suggest that the market is not a suitable model for the social in general. If a conception of the social is to be plausible, we expect it to be able to account for when there are and when there are not markets, among other social entities. A better conception of the social would take the best aspects

26

THE REPUBLIC OF SCIENCE

of these prior attempts (organism, machine, city, market), jettison their inadequacies, and try to come to terms with society in general and as such. Popper treats social relations as ontologically fundamental - that is, individuals emerge from social relations and not vice versa. This may owe something to one of his heroes, Russell. In The Problems of Philosophy (1911) Russell argued relations and relata were equally onto logically fundamental. Popper, I believe, made use of this. Operating mainly with the ideas of social coordination through institutional rules in order to achieve collective goals, and institutional and personal responsibility for keeping such rules under revision, he has, in my view, put forward an original conception of the social using the polar concepts of the open and the closed societies, and of abstract and concrete social relations. Remarkably, this contribution has received almost no attention - a situation that needs remedy. A running theme of the present volume is that in his major work The Open Society and Its Enemies of 1945, Popper, perhaps under the pressure of war conditions, successfully pushed beyond those older kinds of thinking about society and introduced a much more adequate and indeed sophisticated conception of the social, one flexible enough to mediate between thinking about society and thinking about social policy. Thus not only could he make sense of ancient society and exotic society, but he could also help us think about even more taxing phenomena such as social change and social reform. It is clear that these results were unintended consequences of an intellectual effort that was directed elsewhere. The aims of Popper's 1945 work were concrete and urgent. Partly because of the sheer challenge of the problems he was working on and partly because of his utterly original earlier thinking about a major social institution of the modem world, science, he reaped a bonus. The problems he was concerned with were to explain the intellectual roots of totalitarianism, its popularity and intellectual support, and its ferocious global wars against regimes defending freedom and democracy. Popper's path-breaking work on the social had a history. In Unended Quest Popper tells us that as soon as Logik der Forschung was finished he turned his mind to parallel questions

INTRODUCTION

27

in'the social sciences, the result of which was "The Poverty of Historicism", which he first read in public in 1935. Indeed, he describes The Open Society and Its Enemies as a truly unintended consequence of an attempt to flesh out The Poverty in all its historical detail. I argue, however, that the roots of Popper's social thinking go deeper, indeed all the way back to Logik der Forschung itself. That book, concerned to treat metaphysical and other foundational questions as methodological questions, treats science operationally as a social institution or set of institutions. Neither Popper nor his commentators sufficiently stressed this or expanded upon it. Popper does not even bring it out in his threevolume self-commentary Postscript to the Logic of Scientific Discovery (1983, 1984). Yet the only framework that makes good sense of his key offerings of methodological decisions and methodological rules is that of social institutions. Some of the institutions he has in mind are quite concrete - such as learned societies, universities, seminars, journals, laboratories; some are quite abstract, such as the 'invisible college', the various intellectual traditions and, most explicitly, the rules that govern and constitute the institutions. Popper's Logik der Forschung depicts science as a self-governing and rather abstract community, constantly debating not only its first-order concerns of how the world works, but also its second-order concerns of how it should proceed about its work, and its third-order concerns of how in general we should proceed (intuitively, critically luncritically , cooperatively luncooperatively, and the like). In the absence of a ruling or governing body that can impose decisions on the collegium, all these levels of debate are resolved in the same way: by the participants in the institutions voting with their feet. Scientific institutions need a strong and ongoing consensus not so much about first -order matters concerning the way the world is, as about second-order ones concerning how to proceed with the discussion, and third-order ones concerning self-governance. Disputes about, say, black holes, will not get very far scientifically if there is no consensus on how disputes should be conducted. The members of a debating society may not all like or even agree with Robert's Rules of Order. Yet

28

THE REPUBLIC OF SCIENCE

participation will be contingent on some such rules being in place because even debates about their amendment cannot be successfully conducted without initial submission to them. What the body of scientists decide to do becomes their customary law and folklore and is for the most part passed on informally through their networks and to their students. Thus the folklore of inductivism could become entrenched even while in customary law there were only scattered instances of it being tried. Those at central nodes in networks decreed that induction was to be followed, and those dependent upon network patronage and good will felt constrained to affirm their allegiance to induction. (In this respect they resembled those Soviet social scientists who were able to publish their ideas provided they paid lip service to Marxism-Leninism.) While not proposing any change in the self-governing and self-constituting nature of the community of scientists, Popper did propose, in the Logik, a much more self-conscious and rigorous discussion of the metarules so that more informed choices could be made; choices that avoid futilities and inconsistencies, and hence avoid induction, among other things. There is nothing explicit on the social organisation of science in Logik der Forschung. This is a puzzling fact. Popper's original concerns, the problems of demarcation and of induction, were philosophical; and the remedies he proposed, methodological rules, were also philosophical. Yet such rules, together with the intersubjective cooperation involved in Popper's conception of observation and testing, presuppose social groups. Popper says nothing about the composition and structure of these groups. Be that as it may, the text is there for the alert reader who can pose the questions: what kind of groups, institutions and social processes did Popper see in science and envisage for science, and how much did that affect, and get modified by, his thinking on a broader social canvas for The Open Society and Its Enemies? I have already suggested that in proposing or articulating (methodological) rules by which science should govern itself, Popper developed an analysis which he later used to great effect on the wider problems of politics, society, and social reconstruction. In tum, those problems led him to reconsider

INTRODUCTION

29

some of what he had said in Logik, especially concerning the idea of truth, and also to generalise those ideas. He had succeeded in avoiding the contentious idea of truth in Logik der Forschung, for methodological rules were conventions and a conventional rule seemed sufficient to do the work formerly done by metaphysical notions of truth (see ch. 2, §3). Social institutions and their associated conventional rules do not prohibit pursuit of concrete ends, usually articulated as 'the good life', and changed by Popper into the minimising of avoidable suffering. Pursuit of such ends, however, does seem to demand a criterion of truth if we are to assess our success. Metaphysical and moral considerations are, thus, inescapable in the political discussion of society and in the acceptance of responsibility. The result of Popper's deliberations on the social is his later loss of inhibition about declaring pursuit of truth to be a central metaphysical and moral value of the society of scientists. A further, and subtler relation of great interest connects LScD to OS&IE: the latter is an implicit critique of the social aspect of the former. The social content of Popper's philosophy of science was institutionalisation of the decision to maximise falsifiability. The form of institutionalisation was methodological rules. These rules made possible intersubjective testability, and that flourished because of the ethos of mutual, friendly/hostile cooperation. Popper's early sociology, then, is one of institutions and their ethos, or culture. 14 Clearly there is much missing from this skeletal sociology. Created institutions can lose their way or become corrupt. Over time, a culture becomes a tradition; tradition and culture can also develop defects. In political society such problems call for maintenance, even reform. Also over time we may find that 14. Under the name "social epistemology", some philosophers in the last quarter of the twentieth century began to explore the social conditions under which knowledge is possible. Goldman, for example, uses neo-classical economics (Goldman 1992, 1999); and Kitcher uses rational choice theory (Kitcher 1993). Both seemed to mean by social "interacting individuals". As soon as one grasps institutionalism a fatal weakness in their approach is apparent: interaction between individuals and institutions, and interactions between institutions are missing. Yet these may be the most important interactions of all.

30

THE REPUBLIC OF SCIENCE

existing institutions fail to cover new situations and problems. In political society this leads to the design of various kinds of new institutions. These considerations are raised in Popper's account of political society, The Open Society and Its Enemies. All longterm politics, he argues, is institutional. Institutions must be well designed and well staffed. However special science is, it is not less prone to the problems of institutions, cultures, and traditions than any other part of society. Thus is the skeletal sociology of The Logic of Scientific Discovery found wanting. Consider, for example, power. The argument of The Open Society and Its Enemies is that wherever there is power there is the possibility of abuse. This renders the fundamental question of politics not 'who should rule?' but, 'how can we design government so that bad rulers can be removed without violence?' Consider the situation in science. In Popper's lifetime science was transformed from the esoteric concerns of a small number of politically unimportant individuals into a centre of power. Institutions such as laboratories, seminars, and scholarly meetings became sites where money and prestige were dispensed. Yet the institutions of science had not been designed to cope with these circumstances. They were not designed to settle disputes over money and prestige, and they had neither democratic controls nor formal means of dealing with corruption. Their primary design feature was to bring together a group of individuals interested in a common project, subjecting their activities to rules that forwarded that project. So I maintain that The Open Society and Its Enemies is readily interpreted as critical of The Logic of Scientific Discovery for not following up its own logic, and for failing to normalise the institutions of science as a new locus of power. Power over nature was to lead to Big Science. But even before Big Science, institutionalised science found itself doling out power in the form of publications, jobs, grants, prizes, and prestige. Given his passion for social justice, fully expressed in The Open Society and Its Enemies, Popper should have found the skeletal sociology of The Logic of Scientific Discovery politically naive. In some respects science was an institution like all of the others: it needed structures and controls.

INTRODUCTION

31

Reasoning backwards to the roots of Popper's social thought is vulnerable to the accusation of eisegesis or of 'reading in', which itself raises questions about interpretation in general. 6. Problems of Interpretation Two colleagues of Popper, writing in the Library of Living Philosophers volume devoted to his thought (Schilpp 1974) offered diametrically opposed interpretations. Watkins (1974) argued a fundamental unity to Popper's thought; Lakatos (1974) argued that Popper began shifting his position as early as his first book. A standing problem of interpretation regarding every writer of multiple works is how to deal with that multiplicity. Is there an early and a late Plato with different or changing views on the same topics? Is there an early and a late Marx? Just how many 'Bertrand Russells' are there? On such interpretative problems Popper was unequivocal: A critical interpretation ... must take the form of a rational reconstruction, and must be systematic; it must try to reconstruct the philosopher's thought as a consistent edifice. (OS&IE, I, p. 215/246.)

My argument for modifying this maxim is that the philosopher's problem-situation must be taken into account. In the present case, does 'Popper' signify a single, coherent system of thought extending over sixty years of publications, a system that can be pieced together in different ways depending on point of view? The answer to this can be 'yes' if and only if the multiple works can be interpreted consistently as addressing a single set of problems, so that a thought taken from anywhere in the oeuvre coheres with the rest. This condition is seldom satisfied in the case of authors of multiple works. Popper admits that his views have changed on various issues, and certainly his emphasis changed a great deal, as the problem-situation in which he wrote each major work differed - not least, as time went on, because his own growing body of work was partly constitutive of the problem situation. For these reasons, my view is that commentators who treat Popper's thought as a seamless whole, for example, Johansson

32

THE REPUBLIC OF SCIENCE

(1975 or Blaug 1980, or Newton-Smith 1982, or Hausman 1992) are fundamentally uncritical. Johansson writes about what he refers to as 'Popper's methodology', lumping together and quoting freely from any work published before his own work was completed, 1974. It is particularly surprising in such a critically astute author that he seeks enlightenment about The Logic of Scientific Discovery by examining later statements of Popper on the same questions, or indeed by using direct glosses Popper has offered on his own texts. Aside from anything else, Popper's commentary on his own work has, according to his own theories of interpretation, no special privilege and certainly no special authority. If one wants to treat Popper's thought as a seamless whole a fairly compelling argument would seem to be needed. Only Watkins (1974) attempted this but did not, in my estimation, succeed. Popper on Popper is only a fraction of the interpretative problem. There already exists a substantial secondary literature of interpretation, explication, and critique of Popper. Consider only the situation in English (there is more in German): besides the two-volume entry in 'The Library of Living Philosophers' series, The Philosophy of Karl Popper (which consists entirely of commentary and critique and of replies), there are book-length monographs by Wisdom (1952), Magee (1973), Johansson (1975), Ackerman (1976), Wilkins (1978), O'Hear (1980), Burke (1983), Berkson and Wettersten (1984), Munz (1985), Wettersten (1992), Simkin (1993), Agassi (1993), Miller (1994), Shearmur (1996), Stokes (1998), and Artigas (1999). The volumes of essays The Critical Approach to Science and Philosophy (Bunge 1964, reprinted 1998), In Pursuit of Truth (Levinson 1982), Popper and the Human Sciences (Currie and Musgrave 1985), Karl Popper: Philosophy and Problems (O'Hear 1995), and Karl Popper. Logik der Forschung (Keuth 1998) are entirely devoted to Popper's thought. He was accorded a major article in Paul Edwards' Encyclopedia of Philosophy (Quinton 1968) as well as Edward Craig's Routledge Encyclopedia of Philosophy (Jarvie 1998), and Popper's colleague Imre Lakatos initiated a huge volume of subcommentary by setting himself up as a uniquely authoritative commentator and corrector. Popper's students Munz, Watkins,

INTRODUCTION

33

Agassi, Feyerabend, Bartley, myself, Hattiangadi, Briskman, Miller, Musgrave, and Shearmur have spilled much interpretative and explicative ink. From this list I should single out Agassi's student Wettersten (1992) who, together with Shearmur (1980, 1985, 1996), somewhat anticipated the 'social reading' of Popper's methodology developed in this book. In this book I proceed on the assumption that Popper faced different problem-situations in his three major works, and I shall treat those as defIning contexts by taking care to stick to the original texts of 1935, 1943-44, and 1945, rather than their later, revised republications. Where I use material from later editions or self-commentary I shall give reasons for doing so. The appeal of the originals is that besides being a commentator on his own output, Popper was an inveterate tinkerer with his own texts. Seldom did he pass up the opportunity, on the occasion of reprint or of new edition, not only to correct misprints, but to improve previous formulations. Indeed, given his views on translation, according to which all translation is interpretation, I wish that my competence in German was suffIcient to enable me to work from the original 1934 text of Logik der Forschung. This is not because of any mistrust of his reformulations or translations but for simplicity's sake: I want to take the contents of my 'Popper' from the objective piece of writing he released for publication at a moment in time. 15 Reformulations, second thoughts, and selfcommentary immensely complicate the interpretative task. I shall proceed to sketch the development of Popper's ideas on the social. Their fIrst glimmerings are in LScD, when he may not have read much social science (except for psychology and Marxism) 16. By the time of PoHIPoH the impact is apparent of extensive reading about the social and, in its last rewrites, the specifIc impact of Hayek. Finally, The Open Society and Its 15. As the reader can imagine, if Popper worked constantly on material already published, he also worked constantly on drafts and versions as publication approached. However, once the proofs are passed, the type locked, as they used to say in printing, the work enters the public domain. This is a decisive qualitative moment, when thought becomes objective. T6. On these matters I defer to the researches of Hacohen 1993, 1996, 2000 and Shearmur 1998.

34

THE REPUBLIC OF SCIENCE

Enemies reveals a phenomenal amount of reading and thinking about the history and development of social thought as well as about the actual problems of explaining social regularities and of social planning to bring about desired ends. It is my interpretation that these three works hang together as a connected series and that the later two are implicit critiques of the earlier. The result, I claim, is a richer reading of the whole body of work.

Chapter 2 POPPER'S 1935 PROTO-CONSTITUTION FOR THE REPUBLIC OF SCIENCE It might indeed be said that the majority of the problems of theoretical philosophy, and the most interesting ones, can be reinterpreted ... as problems of method. ---Popper (LScD, p. 56)

In this chapter I argue that insoluble difficulties in the 'pure' logic of science led Popper to propose, in 1935, that we treat science as a social institution incorporating a set of rules, or methodology, designed to advance us towards agreed-upon aims. Popper's subsequent work is consistent with such a 'social' reading, but he nowhere makes it explicit. Virtually all of his critics (and some of his followers) overlook this decisive shift from the logical to the social, and as a result are unable effectively to appreciate or to criticise this original feature of his thought. Although I find a 'social tum' in Popper's thought, it should be stressed that his concerns were first and always philosophical. Popper never portrayed himself as having put forward a social view of science. The social tum in his work was not followed up by discussion of problems in the social aspects of science. Popper never undertook research into the workings of the institutions of science, or their history, or wrote a scientific biography. Indeed, his concrete examples from the history of science mostly relied on secondary sources. Despite these concessions, I think the epigraph to this chapter shows clearly the emphasis Popper places on method. Methods are social. Popper's repeated strategy is to reframe philosophical problems as methodological problems, to tum, that is, metaphysical issues into social decisions. Methodology is treated non-naturalistically: as a choice of social conventions, not a

36

THE REPUBLIC OF SCIENCE

description of a social practice. Thus Popper does not reduce science to the status of one social institution among others. It is a very special social institution, both in its aims (truth that transcends social particularity) and in its results (universal scientific knowledge). Thus Popper's 'sociologism', if I may call it that, contrasts sharply with the reductionism and relativism of 'the Strong Programme in the Sociology of Knowledge' (see ch. 1, §2; and ch 5 §4 ). That caveat entered, Popper's social turn makes him a major theorist of the sociology of science: The Logic of Scientific Discovery contains profound and original ideas on the social character and constitution of science, and, more to our purpose, these contain the germ of his later, highly influential ideas on society and politics generally. In other words, I shall be arguing against the view that Popper first philosophised about science and then applied that philosophy to social and political problems. Rather, he was thinking socially from the beginning. The relation between the earlier and the later work is more complex than one of generalisation or elaboration. When he turned to social thinking explicitly he developed ideas and arguments that supplemented and were critical of his earlier social turn. By bringing this out we can enrich his initial sociology of science by treating it critically. 1. The Received View of Popper on Science To begin with, it may be helpful to outline the relevant central ideas of The Logic of Scientific Discovery as they are usually described'! The book is in two parts. In Part I its contentions are sketched, in Part II they are worked out and defended in detail as a series of answers to objections. 2 The book focuses on two problems labelled 'the problem of induction' and 'the problem of demarcation' . The problem of induction is how to obtain conclusions of a general and theoretical character from the particularities of experience. Experience always comes in particulars: we observe 1. As expounded in such texts as Magee 1973, Ackerman 1976, Burke 1983, O'Hear 1989, Stokes 1998. 2. I follow a suggestion of Joseph Agassi in viewing the later chapters as answers to possible objections to the initial ideas.

THE 1935 PROTO-CONSTITUTION

37

something about this, that or the other. By contrast, theoretical knowledge claims are general; as we say, 'universal'. Logic allows that no number of particular statements, such as those describing the experience of observing white swans, permits the deduction of a general statement resembling scientific knowledge such as 'all swans are white'. How, if not by logic, do we get from particular experience to universal theories? The answer frequently offered - induction - is logically problematic. The problem of demarcation is how sharply to differentiate science from non-science, including pre-science, pseudo-science, folk wisdom, and metaphysics, as well as from logic and mathematics. Whereas the problem of induction seems a troublesome one, the problem of demarcation seems to be merely a matter of classification. However, this appearance is misleading. Since, according to Popper, scientific theories are statements, and statements require language, he is able to reformulate the two problems more crisply: the problem of induction becomes, how can the general statements of scientific law be validly derived from particular statements reporting our particularised experience? The problem of demarcation becomes, how, if at all, are scientific statements to be distinguished from other statements? According to his autobiographical accounts, Popper had been engaged in thinking about these issues since 1919, when he was 17.3 The first problem he says he solved fairly quickly: the relationship of scientific statements to statements of experience was deductive, not inductive. Logic allowed particular statements of experience to refute proposed universal scientific statements, and that, it turned out, was sufficient for the task at hand. Provided one did not expect science to consist of certain, or proved, knowledge, it was enough to see it as the body of general statements that had survived attempts at refutation: they were not supported or recommended by anything other than the fact that 3. Cf. the account that appears in Popper 1963, p. 34 (a chapter first published in 1957) and elaborated in the autobiography of 1974/76. Malachi Hacohen (1993, 1996) has argued that Popper's own accounts are not necessarily a good guide to the development of his ideas.

38

THE REPUBLIC OF SCIENCE

they had so far survived whatever challenges had been thrown at them, whether by testing against experience or for their own internal coherence. This testing made them candidates for classification as scientific knowledge. Thus the novelty of the solution lies not so much in the observation that refutability is a necessary condition for scientific theory, as in the view that refutability is sufficient to characterise the work of science and its claims to progress. Popper's new solution to the problem of induction thus derives its impact from also being a solution to the problem of demarcation. Popper relates that it took him some years to realise that the two problems of induction and demarcation were connected, and that the demarcation problem was the deeper of the two: induction was one solution to the demarcation problem, deduction another. The solution to the demarcation problem that Popper poses and elaborates is that scientific statements be characterised as general statements which are in principle falsifiable. Those already falsified, it should be added, belong to the history of science. Those not yet falsified are current science. Statements that are not falsifiable come in more than one type, including: statements of logic and mathematics, which are tautologous truths (e. g. 2 x 2 = 4); and statements containing existence claims (e. g. the Absolute exists). All bare existence claims, Popper added, are metaphysical. The reason science is best characterised as consisting of the body of falsifiable (but unfalsified) universal statements has to do with the manner in which science permits us to learn from experience. Puzzled by some logical conflict between (our received) ideas and statements capturing the evidence of experience, we seek resolution. Three options confront us in each case: the report of experience is faulty; the logical claim to detect a contradiction is faulty; or the (received) ideas are faulty - no matter how important they are, no matter how well-founded they seemed. Thus is science the critical engagement of ideas with experience. It is the process of culling from among the ideas thrown up by our imagination those that are worth criticising by

THE 1935 PROTO-CONSTITUTION

39

the check of experience. 4 The check of experience is not easy to meet. It requires that we guard against formulating our ideas in a vague, weak, or evasive manner which would make empirical criticism difficult, even impossible. If ideas cannot clash with experience, then we can learn nothing about the world, and that would render Popper's fallibilist realism empty. In order to foster the confrontation of ideas with experience Popper makes a revolutionary proposal. It will be best to describe it (§2) after a digression about the great Bacon. Sir Francis Bacon, founder of the view that induction is what characterises scientific method, stressed that adoption of the correct method was not, in itself, sufficient to generate science. The all-too-human impulse to find what we want, to observe what is not there, is overwhelming. The mind is filled with what he termed 'prejudices'.5 Bacon therefore proposed that researchers must first work at emptying their minds of all preconceptions and curbing the almost irrepressible desire to jump to conclusions. 6 Thus, for Bacon, being a researcher involves an inner struggle to achieve a certain psychological condition, a condition where the mind, freed from prejudice, is open and receptive to experience Nature as she really is. Only in that purified state, Bacon thinks, will the methodological canons of inductive reasoning, patiently pursued, keep one clear of error. 7 There are two features of Bacon's account that are worth noting here for their contrast with Popper's proposals. One is that Bacon's account is individualistic - the budding researcher struggles alone against her prejudices. The other is that it is 4. The use of the term 'check' for the present purpose is first found in Bartley 1962, p. 158-59. 5. The use of this term for present purposes is original to Bacon, who borrowed it from the legal literature. Much confusion was created by Popper's agreeing in LScD with Bacon that all humans are prejudiced and by his concluding that despite Bacon's recommendation to avoid prejudice, since prejudice is unavoidable, natural science must then be full of prejudice. Popper stresses that science is not inferior but special; because it is not free from prejudice its practitioners insist on being as critical as possible, even of its best products. See the Supreme Rule, below. 6. F. Bacon 1620, Book One, Aph. XIX, XX, XLVII, XLVIII, LXX, CXXX. 7. Bacon's ideas are contested territory, see Urbach (1987) and Vickers (1996).

40

THE REPUBLIC OF SCIENCE

psychologistic: bad habits of mind cause the trouble; science can only be attained after one has developed good habits of mind. Elsewhere, Bacon did envisage scientific institutions,8 but not in his account of method. From this point forward, I contend, it would be inadequate to present Popper's ideas in the usual way.

2. Methodology and Social Practice

Popper's revolutionary move is not his shift from induction to deduction - although that is a standard way to put it. In itself this is not revolutionary. Popper goes much further: he also rejects Bacon's individualistic and psycho logistic approach to readying oneself for science, i.e. personally repressing and avoiding prejudice. Prejudices are not to be purged by some sixteenth century version of psychotherapy, but by confronting them with experience in an institutional setting that functions to inhibit evasion. Scientific research begins from and tries to correct prejudices. Science is a creative endeavour; it is the search for new knowledge, so Bacon's emphasis on the old - on the past psychological state of the discoverer, on how ideas are obtained is misplaced. The ancestry and antecedents of ideas are logically irrelevant: what counts is how ideas stand up to the various checks we carry out on them, most importantly, the check of experience (LScD §2). Instead of urging us to work on our state of mind, Popper argues for a social and institutional approach. As he sees it, the cooperative social character of scientific research is indispensable if we are to check our impulse to see what we want to see, to jump to premature conclusions, to worship Bacon's idols.9 Popper refers to this as the friendly-hostile cooperation of 8. Bacon's New Atlantis is a utopian society where scientific research is practised in a secular college whose members are so prestigious that their progress through town resembles that of an archbishop (see Bacon 1624). Popper's most extensive discussion of Bacon dates from 1959, see ch. 9 of 1994. Bacon (1620) envisages research as collective in Book One, Aph. CXIII; as done by equals Book One, Aph. LXI. 9. Bacon (1620, Aph. 39ft) identified the idols of the tribe; of the cave; of the market place; and of the theatre as special/common dangers. In modem terms these are: tribe - prejudices due to the nature of our faculties and emotions; cave - prejudices due to individual idiosyncrasy; market-place - prejudices due

THE 1935 PROTO-CONSTITUTION

41

researchers and, more abstractly, as the intersubjective character of the scientific enterprise (LScD, §8, pp. 44-48). For this joint endeavour to succeed, scientific ideas must be formulated in publicly examinable ways, thus being open to checking by others as well as by their proposers. Those others may have various motives: a disinterested search for the truth, or a strongly interested dislike of the authors and/or of the ideas. But checking the ideas against experience is done according to institutionalised methodological rules - the best we can devise in an imperfect world. These rules do not ensure there will be no error, but they do create incentives to discover and to expose error, rather than to evade it and to cover it up. How does Popper move from the pure logic of science into aspects of scientific method that are social rather than logical? To find the answer we need to examine the text closely. Early in his discussion of the fallibilist logic of science Popper considers three possible objections to his view (LScD §6). The first objection is that we expect science to deliver positive information, so to characterise science negatively, by refutability, is wrong-headed. 10 Popper parries this objection by asserting that he will later show that positive information has a logical connection to refutability. (The promise is fulfilled in three stages: by showing that testability is the same as refutability; by showing that content and refutability grow together; and by extending his theory into one of degrees of testability hence degrees of content, see LScD §§ 31-46.) The second objection is to the effect that, in the same way that inductions cannot be verified, no more can falsifications. In rebuttal Popper maintains (or, rather, reiterates in modern language Bacon's observation) that there is a logical asymmetry between verification and falsification. One contradictory evidence statement is sufficient to refute a generalisation; no number of non-contradicting evidence statements is sufficient to establish a generalisation. It is sometimes claimed that his answer will not to communication misunderstandings; theatre - prejudices derived from systems of dogma and philosophy. 10. David Papineau rehearses this objection in his (1995) apparently unaware that it was Popper who had formulated it, as well as answered it.

42

THE REPUBLIC OF SCIENCE

work, as a contradiction between an evidence statement and a generalisation can always be handled by rejecting the evidence statement. Popper notes that this observation is correct but it does not give rise to any difficulty. The point at issue is that two contradictory statements cannot be true together. It is this logical fact that counts, not which statement we choose to give up. The latter cannot be settled by logic. The third objection is taken by Popper to be more serious: refutations can be handily evaded by conventionalist stratagems, such as the ad hoc introduction of auxiliary hypotheses or the redefinition of terms, or simply by refusing to look at contrary evidence. Popper finds no purely logical way to defeat such stratagems. He writes: it is impossible to decide, by analysing its logical form, whether a system of statements is a conventional system of irrefutable implicit defmitions, or whether it is a system which is empirical in my sense; that is, a refutable system ... Only with reference to the method applied to a theoretical system is it at all possible to ask whether we are dealing with a conventionalist or an empirical theory. The only way to avoid conventionalism is by taking a decision: the decision not to apply its methods. (LScD, p. 82.)

Thus Popper accepts a decisive criticism of the very position often attributed to him - what I have termed the received interpretation. It is recognition of the problem created by the third objection that moves Popper into proposing what I call his social view of scientific method.ll Many commentators, in both 11. Popper's critics fmd it hard to come to terms with his methodological conventionalism. Either they just overlook it or they launch objections that are irrelevant because they take naturalism for granted. Take the inductivist Salmon (1967) and the conventionalist O'Hear (1989), publishing 21 years apart. Salmon: Popper's fundamental thesis is that falsifiability is the mark by which statements of empirical science are distinguished from metaphysical statements and tautologies (Salmon 1967, pp. 21-22) Popper thus holds that falsifiability is the hallmark of empirical science (Salmon 1967, p. 22) Hallmarks are conventions; they are not statements of science unless we put them there - how else but by decision. O'Hear:

THE 1935 PROTO-CONSTITUTION

43

their expositions and their criticisms of Popper, overlook this point. The conventionalist objection to the refutability criterion of science cannot be overcome on its substance. If someone chooses to indulge in ad hoc manoeuvres, or to ignore refuting evidence, they can do so without committing logical or factual error. This does not mean their conduct is unobjectionable. Such manoeuvres have consequences and these can be criticised. Popper objects that such manoeuvres impoverish science because they shield its claims from the tribunal of experience. A system shielded in this way (as Bacon already noted) can easily and unnoticingly degenerate into metaphysics; that is, untestable statements. Like Bacon, then, Popper moves the discussion from intractable logical matters to questions about choice. Unlike Bacon, Popper sees the choice as social: aims, policies, and their consequences in the social organisation of scientific endeavour, rather than in individual psychic struggles against prejudice. For those concerned to keep science anchored in experience, Bacon suggests adopting a supreme or meta-methodological rule that individual researchers not affirm any theory unless they are certain it is true. Popper suggests adopting a supreme or metamethodological social rule not to avoid falsification; in order to That science should be demarcated from the non-scientific in terms of empirical falsifiability is the proposal of Popper which has most captured the imagination of the general public, who [sic] have [sic] seen in it a means of justifying their suspicions of influential pseudosciences (O'Hear 1989, p. 56). True scientists, on this view, will try to falsify their theories and will not rest content with 'anomalous' results or seek easy explanations for test failure ... Is it always unscientific to use evasive 'conventionalist' stratagems in the face of counter-evidence to a favoured theory? (O'Hear 1989, p. 64.) O'Hear mixes naturalism and conventionalism. The conventionalist tone of the opening phrase "should be", is cancelled out by the "justifying" that is made possible and by the mention of "true scientists" - which foists on Popper an essentialist view of what a scientist truly or essentially is. The question with which the quotation ends makes no sense: 'always' is naturalistic and so sidesteps discussion of how we should decide the matter. 0 'Hear goes on to discuss Kuhn's wholly naturalist objections, oblivious to the obvious point that how things are has no critical force against proposals for how they ought to be organised.

44

THE REPUBLIC OF SCIENCE

monitor this rule, co-inquirers must be free to check one another's claims (LScD, p. 54). Here, in the move from purely logical criteria for scientific statements to methodological criteria for scientific practice, we find both the roots of Bacon's view of research as psychological, and the beginnings of Popper's view of research as entwined into the social character of science. A methodology consists of methodological rules; each rule enjoins a decision, a choice to act in a certain way. We make these choices, in tum, in order to foster certain aims. Bacon presents rules and aims dogmatically; Popper opens them for discussion. Both the rationale of the methodological choices and whether the choices will in fact foster the desired aims are matters on which there can be reasoned dispute. Let me set this out in more detail. What is proposed in the opening pages of The Logic of Scientific Discovery is a theory of method. A scientific researcher is characterised as someone who puts forward statements or systems of statements and tests them step by step. The task of a logic of scientific discovery, or a logic of knowledge, is to give a logical analysis of this procedure, "that is, to analyse the method of the empirical sciences" (LScD, p. 27). Popper proposes a method of deductive testing of theories ("deductivism") by deriving their consequences and checking these by various means: against each other (are they selfconsistent?); by their logical form (are they empirical or tautological?); against other theories (are they consistent with them?); and finally against statements describing empirical experience (are they consistent with known and with newly discovered experimental facts?). In discussing the problem of demarcation, Popper challenges the idea that there exists any 'natural' boundary to science. He suggests instead that we treat the refutability criterion as a proposal for a suitable agreement or convention to help us to "be able to say of a given system of statements whether or not its closer study is the concern of empirical science" (LScD, p. 37). Reasonable discussion of the suitability of a convention is "only possible between parties having some purpose in common"

THE 1935 PROTO-CONSTITUTION

45

(LScD, p. 37). 'Parties having some purpose in common' already constitute a social group or the embryo of one. The social group in question joins in seeking knowledge of the world of our experience, the real world. How is knowledge of the world of experience to be distinguished from other claims to knowledge? By the fact that it has been submitted to tests and has withstood tests. Experience is constituted by the method. The inductive method is invalid. The deductive method, Popper proposes, admits to empirical science only those systems of statements that can be falsified by statements describing experience. Falsifiability is insufficient, however (LScD, pp. 42, 50 and 54), because it is always logically possible to evade or deny a threat of falsification, to shield theory from conflicting experience. It can be objected that this makes science no more than a personal choice to engage with, rather than to evade, experience. Popper needed to say that personal choice is necessary though quite insufficient to characterise an endeavour as scientific. Individuals checking their own illusions and delusions have no check on their own checking process (see the discussion of Crusonian science in ch. 4 below). Bacon mistakenly identified the problem: checking the checking, objectivity, does not depend on a psychological state of disinterestedness or of being free from prejudice. 12 Casting about for a better view than Bacon's psychologism, Popper draws on Kant's idea that 'objective' refers to a statement that can in principle be understood and tested by anybody. This Popper calls the "inter-subjective testability" of scientific claims. The language is moving towards a social view, but its Kantian echoes hold it back and invite misunderstanding. It suggests that science emerge from individual decisions, choices, and psychology in Popper as in Bacon. What checks are there on individual psychology? How can objective science emerge from 12. Feminist scholars such as Evelyn Fox Keller, Sandra Harding, and Carol Gould have exposed layers of sexist prejudice in science, especially biology and medicine. The Baconian view would be that they show the need for further vigilance, fiercer struggles against prejudice. The Popperian view would be that criticism of prejudice is where science begins and is one of the ways it advances.

46

THE REPUBLIC OF SCIENCE

prejudice? This line of objection is mistaken but not unfair. Popper does not adequately characterise his notion of decision in LScD. His exposition of his view as social is lacking.1 3 This point about objectivity made, Popper resumes outlining his slightly veiled social view of science. The theory of method, he avers, is concerned with the purely logical relations between scientific statements and with the choice of methods decisions about the way in which scientific statements are to be dealt with. These decisions will of course depend in their tum upon the aim which we choose from among a number of possible aims. The decision here proposed for laying down suitable rules for what I call the 'empirical method' is closely connected with my criterion of demarcation: I propose to adopt such rules as will ensure the testability of scientific statements; which is to say, their falsifiability (LScD, p. 49).

What are these rules, why do we need them, and can there be a theory of such rules? A methodology turns, he writes, on one's attitude to science. If one is interested in the advancement of science, in its constant revision and correction, one will be led to a very different answer than if one takes a narrowly logical and naturalistic view of science. Purely logical means fail to differentiate science from metaphysics (or anything else) because refutation can be evaded without violating logic. The choice to encourage and to embrace refutations has social implications. To implement that choice we set up social conventions for which we take responsibility: "by what we do with them and what we do to them. Thus I shall try to establish the rules, or if you will the norms, by which the scientist is guided" (LScD, p. 50). That Popper is thinking institutionally could not, I think, be clearer; all the more puzzling, then, that the point has been 13. In translating the work he was aware of this and added an important footnote to the English translation which says "inter-subjective testing is merely a very important aspect of the more general idea of inter-subjective criticism, or in other words, of the idea of mutual rational control by critical discussion" (LScD, p. 44), making explicit reference to his own later works The Poverty of Historicism, The Open Society and Its Enemies and Postscript to the Logic of Scientific Discovery.

THE 1935 PROTO-CONSTITUTION

47

overlooked. 14 Popper is here proposing that science is to be seen as the set of theoretical statements entertained by an interest group that shares the aim of subjecting that system to empirical test and which has legislated governing cOllventions for itself the better to pursue that aim. He does not explicitly say that his view is social, but he offers some analogies with the social institutions of games and of trial by jury, going so far as to refer to "the game of empirical science" (LScD, p. 53) and comparing its rules to the rules of chess. Certainly he seems to be arguing that science is constituted by its rules, as is chess. He also seems to be allowing that the rules of science can be debated, hence they are not immutable. Much the same goes for chess. The rules of chess have evolved and might evolve more. A rule revision would not necessarily make for a new game, especially if the rule were adopted by the International Chess Federation. The fact that, in baseball, the National League permits the designated hitter to substitute for the pitcher and the American League does not, hardly raises serious questions about which league really plays baseball. Why should we engage in the game of science, why accept the self-legislated conventions? Peter Munz (private communication) raises this question. Popper's main answer is to claim they are more fruitful (LScD, index entry "fruitfulness"). Munz's answer (Munz 1985, 1993) is that the conventions are subject to a quasi-evolutionary pressure, since they help us to cope with the way the world is only to the extent that they help us to get it right in relevant ways. Appealing though this evolutionism is, I find it over-optimistic. When Ptolemaic calculations could do as good a job as Copernican/Keplerian ones, how can we be confident about such selection processes? Munz flirts, it seems to me, with utilising Darwinian arguments to provide transcendental justification for science. In my formulations I have tried to stress the voluntaristic nature of science: only some engage in the game, and those who engage do

14. Except perhaps by Agassi (1972), Wettersten (1992), and Shearmur (1980, 1985, 1996).

48

THE REPUBLIC OF SCIENCE

so because, for knowledge purposes, it is the best game in town. no means the only game in town.)15 We can clarify further by noting that the analogy to games has a flaw: most games are frivolous activities, engaged in for recreation and play. Science construed as the search for knowledge is a rather more weighty matter. Early seekers after knowledge in the Age of Reason saw their activity as directed towards sweeping away the cobwebs of error and superstition inherited from the past, clearing obstacles from the corridors that would eventually lead to the enlightenment of humanity. The very success of this project has resulted in humankind becoming seriously dependent on science and its applications for economic livelihood and for preserving and extending its longevity. Thus the building of institutional norms or rules for science might better, I suggest, be compared not to a game like chess, but to the creation of dedicated social institutions - universities, for example, or learned societies. This might be especially apt as universities and some learned societies are subordinate institutions within the overall institutional creation we call science, not to mention the difference between the aim for which the institition was designed and the uses to which it was put. (Jeremy Shearmur (private communication) prompts me to interpolate at this point a caveat: institutions are not necessarily best understood as constituted solely or wholly by rule-following. To use a distinction to be discussed later, institutions that are designed differ decisively from those that are 'grown' in that the latter have no explicit aim. Without such an aim, rules to promote that aim are not likely to be found. An example would be the institution of the family. What it is to be a member of a family is by no means exhausted by specifying the rules followed by its members. Left out would be the structure of roles, the accumulated traditions, the external and internal social relations, and the emotional atmosphere. Science is a mixed institution: partly created and partly grown; partly organised, partly (It is by

15. At various places in LScD Popper treats the competition of hypotheses in an evolutionary manner (pp. 42, 108, 251, 278, 281), but in the last two places he draws back, noting that our theories are more than merely adaptive instruments.

THE 1935 PROTO-CONSTITUTION

49

anarchic; some parts following rules that are sometimes tacit, other parts anomic.) Popper does not pursue matters in this direction. Although methodology as he conceives of it is clearly institution-building, he does not examine any actual scientific institutions and their workings, including the methodological and other norms they prescribe. One of the few other places where Popper touches on the social nature of the view he is developing is the passage where he compares science to the institution of trial by jury (LScD, pp. 109-110). However, the context is not that of a discussion of the social character of science. Popper uses trial by jury to make a number of very important points about the theoretical context of inquiry and the way it directs our approach to the facts. Juries, he notes, decide problematic or disputed issues of fact. Which questions they are asked to decide will depend upon the actual laws in force, and the particular procedures of the legal tradition. He gives no examples. He has in mind, I conjecture, the decisions built into one legal system that distinguish it from other legal systems: for example, 'minor child' may be defined differently in different places, constraining which charges may be levelled at someone in one jurisdiction rather than the other. English-speaking readers will likely be familiar with a trial system that is adversarial, leading questions being permitted only in certain parts of the examination, a jury that must sit mute and not subsequently disclose its deliberations. What the jury does is make a finding about a matter of fact. Popper compares this to researchers deciding to accept a basic statement. From this finding of fact, together with statements about the law, consequences can be deduced (for example, that an accused has or has not committed an offence). Although the trial and the conduct of the jury are governed by rules, Popper is at pains to stress that the jury verdict never offers justification or grounds for what it finds. By contrast, he notes, the judge's judgment (on points of law or sentencing) is expected to be "reasoned". If the reasoning is unsound (or the process flawed), grounds exist for appellate challenge; normally, no comparable challenge can be made to the

50

THE REPUBLIC OF SCIENCE

substance of the verdict of the jury .16 Popper here seems to recommend a scientific parallel: the experimentalist experts decide the finding of fact, the basic statements; the presiding scientific community then tries to judge the implications of that fmding. The fmding of fact is far less frequently disputed, Popper stresses, than are the implications of these findings. Were factual reports repeatedly contested, he adds (LScD p. 104), science would not be as it is. The analogies drawn with chess and jury trial serve to highlight on the one hand the constitutive nature of rules, and on the other the embedding of crucial decisions in institutional procedures. The equivalent in science to the International Chess Federation, to the jury and the judge, is not spelled out. Popper's focus throughout the book remains on the logic of science as he endeavours to show that falsifiability is a viable criterion of scientific character provided it is embodied in a methodology. In the course of defending this view in LScD he offers many further suggestions for methodological rules. What he does not engage in, as we shall see, is any discussion of the general picture: do the rules come as a set, or can we pick and choose and yet stay within the game of science?; how are decisions made when new rules are offered or modifications to old ones suggested?; is submission to the rules constitutive of the institution of science broadly conceived, or is there more to institutions and social life under their aegis than rule-following? But this is only to reiterate that Popper did not look at science as an ethnographic or even an historical observer. Faced as late as 1965 with the ethnographic and historical observations of Michael Polanyi and Thomas Kuhn he found the actual practice of science to be woefully wanting (Popper 1970). Thus his social conception of science is best treated as a model, perhaps an ideal type, not a factual description. Before moving into detailed discussion of the rules, it is worth noting how Popper's social models - the game and the jury - are not political. As we shall see later, he adds to this pair a 16. I qualify this sentence because in Canadian law, under certain circumstances, the Crown (but not the defence), can appeal the verdict and move for are-trial.

THE 1935 PROTO-CONSTITUTION

51

third, the Socratic seminar; again, not political. Yet games are not weighty activities and, while trials and seminars are weighty, trials are bound by state-enforced rules, seminars only by local, even informal, ones. Popper's philosophy of science will not work without something stronger: he needs a social model where member conformity to the rules is mandatory and defIning: individuals under such rules could constitute the republic of science. The obvious model is that of self-government: a republic created by equal and autonomous individuals binding themselves to rules in order to pursue common aims. The expression, 'the republic of letters' has been in use since the eighteenth century. Popper is proposing a republic of science that is a good deal more formally organised than the republic of letters. Drawing out the consequences of such a political model is fruitful, since it makes a place for the disputes, divisions, parties, shades of thought, partisanship, patronage, and corruption that institutionalised science actually displays. To develop such a model would also, of course, take Popper a long way from the logic of science. When at the end of the 1930s he began to write about social and political problems, and the design of a better society and its institutions, his reference to the Socratic seminar rather than to the real institutions of science strikes me as a genuine defIciency. But all this is to anticipate; the business at hand is the actual methodological rules, the scaffolding for Popper's republic of science. 3. Methodology in Detail In the fInal section of Part I of The Logic of Scientific Discovery Popper discusses "methodological rules as conventions". He proposes a supreme or meta-rule which governs all the other rules. This reads (SR and the R numbers below are added by me):

(SR) the other rules of scientifIc procedure must be designed in such a way that they do not protect any statement in science against falsifIcation (LScD, p. 54).

52

THE REPUBLIC OF SCIENCE

This meta-rule enshrines the falsifiability criterion of demarcation as a control for the whole system of rules. It takes care of the objection that falsification can always be avoided by mandating that it shall not be avoided. 17 It exposes the error in saying that Popper characterises science by falsifiability. Plainly, he finds falsifiability in and of itself insufficient. Falsifiability is not selfjustifying like a tautological truth. It has to be adopted, a decision that is taken in order to foster certain aims. We impose falsificationism upon ourselves as a procedure, a methodology. Popper is no more a 'naive falsificationist' than Bacon is a naive inductivist. 18 Each was fully aware of the logical deficiencies of their alleged positions and did their level best to rectify them in their actual positions. Although he writes of "proceeding systematically", Popper does not draw up a complete list of the rules which, in his view, constitute experience as a method. He does not say why, but there are good reasons for his not so doing. To expect system would be like expecting a legal commentator to specify the complete list of laws. No such list can exist because, among other things, lawmaking and law-reform are on-going endeavours. Attempts are made from time to time to codify areas of the law, but never the legal system as a whole. All partial codifications need constant maintenance and up-dating. In this respect, a methodology is more like the laws of the land and less like the (more stable and finite) set of rules that constitute chess. After formulating the supreme or meta-rule (SR) , Popper proceeds to discuss the rules themselves by proposing two examples. We shall be able to extract others from elsewhere in the book. The first example is: (Rl) The game of science is, in principle, without end. He who decides one day that scientific statements do not call for any further test, and that they can be 17. This, to repeat, is the simplest way to rebut the objection of Lakatos and many others, which rests on the trivially true claim that falsification can be avoided (Lakatos 1974, pp. 246-47). 18. Discussed in Lakatos 1974, pp. 244-45; see also references given there.

THE 1935 PROTO-CONSTITUTION

53

regarded as finally verified, retires from the game (LSeD, p. 53). This rule directly attacks those philosophers who, like Bacon, envisage science as eventually resulting in a body of finally verified truths. Such spokespeople are a rare breed today, the seventh decade of the Popperian age, but they are not yet extinct. It also raises a fundamental question against all those spokespeople of science, the most recent of whom is Steven Weinberg, who think science might culminate on one grand, final, Theory of Everything. 19 The second example is: (R2) Once a hypothesis has been proposed and tested, and has proved its mettle, it may not be allowed to drop out without 'good reason' (LSeD, p. 53-54). Among the good reasons for a hypothesis being allowed to drop out of science are its replacement by one better testable, or the falsification of one of its consequences. This rule clearly addresses what is sometimes known as the 'stability of science'. The question is raised, if science is simply the set of unrefuted hypotheses, are we free to pick and choose among them? (R2) suggests that no, once a hypothesis has achieved a certain status it can only lose that status via an orderly procedure. Although the content of science changes, sometimes quite frequently, science comes as a package; the institution treats picking and choosing as unscientific. At least, in principle it does; in practice things are less clear cut. These examples of rules bear little resemblance to the rules of inductive logic as sought by other methodologists - from Mill's Methods to modem Bayesians; they can be debated and revised by the community of researchers, other interested parties, and the community at large, without special technical preparation. Consider the question of the plausibility of (R2). The analogy of law-making in the British Parliament may help (where common law would not). When a Bill (a proposed law) passes all the 19. See also Horgan (1996) on the so-called end of science.

54

THE REPUBLIC OF SCIENCE

stages and receives Royal Assent it changes its status - it becomes a Law. It gets printed in statute books, applied in the courts and cited in precedents. However, it is a fact that some laws simply fade away. They may stay on the statute books for centuries and yet not be enforced or even attended to. A good example is laws against suicide which in some countries were simply allowed to fall into disuse long before they were repealed. Does anything comparable happen in science? I suspect it does. That is, I suspect that hypotheses which proved their mettle once upon a time sometimes just fade out of sight, out of mind. Not because of the fallibility of memory (Morgan 1985), but because an entirely new way of looking at things has arisen, and it is not worth bothering to refute each established result one by one - those in the know ignore them, those not in the know go on thinking that these are still currently accepted. A good example from anthropology is the many, many stories from around the globe that Sir James Frazer collected in The Golden Bough. This best seller is still in print, but it would be hard to find anthropologists who could cite chapter and verse of where and when this or that report in it has been falsified. Almost all of them could be falsified in principle, but they may not have been in fact (cf. Lang 1901). More problematic are cases advanced by Agassi and others where a researcher in good standing resists some well-tested scientific idea, perhaps even a dominant scientific idea, one that appears to have proved its mettle. Sometimes the resistance is described as metaphysical, sometimes aesthetic, other times reasons are not articulated. Is such resistance unscientific? In so far as it motivates attempts to criticise and refute the idea, or to develop an alternative, the answer is no: motivation is irrelevant. In so far as it involves refusal to face facts, to suppress, or fail to acknowledge known theoretical strength, then, yes. So long as efforts to resist are not underhanded the scientific community suspends judgment until something interesting is produced. Popper does not deal with this situation, or others where the rules conflict. The supreme rule outlawing protection devices (SR) and the two rules of: the game without end (Rl), and no arbitrary

THE 1935 PROTO-CONSTITUTION

55

abandonment of results (R2), are offered by Popper early in the book as examples of the sorts of rules that can be used to govern science and ensure that its theories are connected with statements describing experience. The whole exercise Popper conceives of as his theory of experience (LScD, p. 57). In my view it is a social theory, a set of putative rules for the constitution and working of the Republic of Science; that is, a self-governing, egalitarian community of those dedicated to the aim of building empirical science. Nothing like a complete set of such rules exists in Popper's text. On the analogy with the law, mentioned above, this makes sense. But if the rules are to be part of the Constitution of the Republic of Science, then they deserve to be spelled out so that prospective citizens know the character of the regime by which they are agreeing to be governed. It is also possible to view the rules as a kind of common law - unwritten customs long established and only articulated and discussed at problematic moments. Some of the epistemological and other sins Popper criticises and erects rules against are perpetrated by scientists in good standing, not just by philosophers. So how are complaints and infractions to be dealt with? Popper supplies no clear decision procedure in the case of dispute or infraction. Yet it should be remembered that Popper is implicitly embedding the institution of science inside two larger entities. First, within an open civil society governed by the rule of law where general matters of criminal conduct and of dishonesty are already sanctioned. Second, science is an activity for those with time and inclination to rerum cognoscere causas. 20 Putting this another way, there are barriers around the Republic of Science other than those erected by the necessity of subscribing to its constitutive rules. Science is an institution or set of institutions embedded amongst other institutions, to some of which it is subordinate. This is the first of a number of ways in which we will see that rules and rule-following are far from sufficient for understanding institutions. 20. The second half of Virgil's line "Felix qui potuit rerum cognoscere causas" ('Happy the man who could search out the causes of things', Georgics, II, 1. 490) is the motto of the London School of Economics, where Popper spent the latter half of his academic career.

56

THE REPUBLIC OF SCIENCE

Popper concentrates mainly on rules of procedure - rules that guide action. But he mentions a few rules, like mandating falsifiability in the (SR) , which in effect define the enterprise as such. In the course of what follows we will occasionally find Popper offering meta-rules that are more like defining than procedural rules. But before continuing it is good to re-emphasise that Popper's book is not systematic. The rules themselves are sometimes described rather than fully formulated, and are sometimes couched as subjunctive conditionals. Nevertheless, let us proceed to extract the remainder of the rules so that we can examine them as a set. In discussing the metaphysical difficulties involved in ideas about causality, and the dubious standing of principles of causality (such as every event has a cause, no effect without a cause) Popper proposes to cut through them with a rule. The rule does not dogmatise about cause, but instead enjoins us. (R3) [We] are not to abandon the search for universal laws and for a coherent theoretical system, nor ever give up our attempts to explain causally any kind of event we can describe (LScD, p. 61). The resemblance of this rule to the supreme rule on falsifiability (SR) should be apparent. The implication is that those who do give up the search for universal laws, or abandon their attempts to explain events we can describe, are opting to leave the (game or) republic of science; those who merely wish to give up the search are wishing thereby to leave science. The normative value of this rule is considerable: it posits a rich overall aim for science (the search for universal laws, a coherent theoretical system, and causal explanations), an aim that ties the group together. Science can then be viewed as a special form of voluntary society. No-one insists that you dedicate yourself to the search for universal laws and causal explanations. But if that is your concern then science is probably what you are doing (depending on your rules of

THE 1935 PROTO-CONSTITUTION

57

procedure); and if that is not your concern you can aid in the fight against confusion by not calling yourself a scientist.21 Proceeding through LScD, the next occasion Popper brings up a methodological rule is when he is considering the problem of how to maintain the empirical character of an axiom system, to prevent it from becoming conventional. There is, he argues, no natural solution, only a methodological decision will do the job. Accordingly: (R4) I shall ... adopt a rule not to use undefmed concepts as if they were implicitly defmed (LScD, p. 75). This rule is a corollary, no more, of (SR) , since using concepts as implicitly defmed demands the use of conventionalist stratagems and (SR) enjoined citizens of the republic of science not to permit any evasion of falsification. Abruptly using empirical concepts as though they were conventions is a technique for protecting a theoretical system from empirical difficulty which Popper labelled the "conventionalist twist". (R4) functions specifically to help us avoid inadvertently giving matters a conventionalist twist when dealing with formalised areas of science. Generalising, Popper argues (in a passage partly quoted above) that it is impossible to decide, by analysing its logical form, whether a system of statements is a conventional system of irrefutable implicit deftnitions, or whether it is a system which is empirical in my sense; that is, a refutable system ... my criterion of demarcation cannot be applied immediately to a system of statements. .. Only with reference to the method applied to a theoretical system is it at all possible to ask whether we are dealing with a conventionalist or an empirical theory. The only way to avoid conventionalism is by taking a decision: the decision not to apply its methods (LScD, p. 82). 21. The aim of science built into (R3), insistence on seeking causal explanations, is not quite the same as that outlined in "The Aim of Science" (1957) - namely, seeking satisfactory explanations. Whether the difference is signiftcant or represents a strengthening and deepening is an exegetical point I leave to one side. Agassi contrasts the clear statement of "The Aim of Science" with the ambiguous ending of LScD, and claims (1988, ch. 27) that the 1957 paper portends a new philosophy of science.

58

THE REPUBLIC OF SCIENCE

Implicit here is a general rule which might be put as, 'avoid conventionalist stratagems' . Instead of such a general formulation, Popper went at it piecemeal. At the end of §19 of LScD he pinpointed four conventionalist stratagems that can be deployed to protect a threatened theoretical system, and for each of them he devised a blocking methodological rule. The four stratagems are: the introduction of auxiliary hypotheses ad hoc; the modification of ostensive definitions; the raising of doubts about the reliability of the experimenter [and/or the apparatus]22; the raising of doubts about the theoretician. In one of his few remarks about the social sciences in LScD Popper comments that, compared to the physicist, the sociologist and the psychologist need constantly to guard against these stratagems, and he singles out psycho-analysts as particularly gullible (LScD, p.82). The four rules Popper devises to meet the specific challenge of these four stratagems are as follows: (R5) [O]nly those [auxiliary hypotheses] are acceptable whose introduction does not diminish the degree of falsifiability or testability of the system in question but, on the contrary, increases it (LScD, p. 83). (R6)

We shall forbid surreptitious alterations of usage (LScD, p. 84.)

(R7) Inter-subjectively testable experiments are either to be accepted, or to be rejected in the light of counterexperiments (LScD, p. 84). (R8)

The bare appeal to logical derivations to be discovered in future can be disregarded (LScD, p. 84).

The wording of (R7) is not wholly satisfactory - read carelessly, it seems to say experiments can either be accepted or rejected. 22.

The phrase in parentheses is mine but is clearly implicit in the text.

THE 1935 PROTO-CONSTITUTION

59

What is plainly intended is a presumption that inter-subjectively testable experimental work be accepted. It should only be rejected in the face of counter-experiment. Thus in the debates about Pons and Fleischman's experiments on 'cold fusion' these rules would prescribe attention to their inter-subjective testability, i.e. their repeatability. Indeed, the scientific community did in the first instance try to repeat their work. Reiterated lack of success, as well as counter-experiment (thought experiments included), was what permitted challenges to come forward to the experimenters, their apparatus and their theoretical deficiencies (Dewdney 1997, ch. 6). Still, such cases as the cold fusion fiasco illustrate that the scientific republic allows for some cooling-down period. This does not refute Popper, whose views merely enjoin that the cooling-down period be minimal. Duhem, Poincare, Kuhn, and Lakatos all incline to the view that the cooling-down period can be stretched indefinitely. They were refuted. Plainly, close adherence to these rules would have a fairly devastating effect on much social science, where verbal sleightof-hand, the introduction of ad hoc excuses and ad hominem indictments of the (class or other 'interests' of the) investigators are almost standard. Moving on, we come to what I call (R9), which is anomalous in the present discussion because it is not in the original text of Popper's book of 1935 but in a starred footnote, the star signalling it as an addition to the English translation of 1959. My reason for not excluding it is that while not stated in 1935 it was implicit and actually employed, as a careful reader will plainly see. It is a very general rule that has application well beyond science. Indeed, it might be termed a general methodological rule for the conduct of critical inquiry and hence another defining rule. (R9) [A]fter having produced some criticism of a rival theory, we should always make a serious attempt to apply this criticism to our own theory (LScD, p. 85n).

60

THE REPUBLIC OF SCIENCE

An immediate consequence of this rule is that it should apply to the present endeavour. My view that Popper is offering the skeleton of a constitution for the republic of science must itself be subjected to vigorous criticism, which I will undertake once the expository section of this chapter is laid out (§8). An example of Popper utilising this rule is to be found in his discussion of what he terms the asymmetry between verifiability and falsifiability. The obvious objection to the demand that scientific statements be verifiable by experience is that the demand cannot be satisfied. Science consists of universal theories and universal theories make assertions about infinitely large classes of objects. There is neither time nor opportunity to examine all those objects and verify the scientific claim. Also, many scientific statements refer to invisible (quarks, black holes), long disappeared (big bang), or entirely abstract objects (relations), and it is utterly unclear how statements about them could be empirically verified. Under the impression that Popper was proposing a simple substitution of falsifiability for verifiability many philosophers made the criticism that falsification was in principle no more final or conclusive than verification. Conclusiveness was a red herring as far as the logical situation was concerned. Popper had already recognised that formal demarcation criteria like verifiability and falsifiability were inadequate. Falsifiability is a logical property, but falsification is not. A statement is declared falsified, hence falsification is a matter for decision. The necessity of such a decision sponsors the articulation of falsificationism into a set of rules, a methodology.23

23. As to the objection that every falsifying basic statement verifies the contradictory of the statement being tested (a basic statement about a black swan could contradict "all swans are white" and verify "Some swans are not white"). This involves an equivocation on the word "verify". All parties agree that two contradictory statements cannot be true together. Any statement that does not contradict another can be said to verify it, even if the other is false. The positivist demand for verification involved a stronger sense of "verification" as a demand that statements be "verified as true or probable", which required lots of verifying statements. A single one was of no interest. Pace Kuhn (1962), a single falsifying statement could be of great interest to any critically-minded scientist. See also Popper's own qualifications discussed in the next paragraph of the text.

THE 1935 PROTO-CONSTITUTION

61

Discussing the difference between falsifiability, which is the logical property we wish a system of statements to have if it is to encounter experience and thus be classed as science, and falsification, Popper stresses that a decision accepting a basic statement which contradicts a theory is a necessary but not a sufficient condition of falsification, because non-reproducible single occurrences are of no significance to science. This is a rule-like assertion, but it is not formulated as one. Although on p. 86 Popper says, "Thus a few stray basic statements contradicting a theory will hardly induce us to reject it as falsified", he does not get round to a rule for this situation until p. 106. (RIO) [W]e should not accept stray basie statements - i.e. logically disconnected ones - but .. . we should accept basic statements in the course of testing theories; or raising searching questions about these theories, to be answered by the acceptance of basic statements (LSeD, p. 106). A science, Popper holds, needs a point of view and theoretical problems. It is only in the context of investigations into those matters that interesting as opposed to stray basic statements are developed around reproducible effects. Four more rules can be isolated from the text of Popper's book. They involve a reiteration, the exclusion of accidents, the specification of random samples, and his equivalent of materialism. They have much to do with what sorts of theories and what sorts of estimates of theories are welcome in science. First (LSeD, p. 121), Popper brings in a rule not previously formulated about empirical content and shows its equivalence to another about severe tests. He argues that comparison of the empirical content of two statements is equivalent to comparison of their degrees of falsifiability: (Rll) This makes our methodological rule that those theories should be given preference which can be most severely tested ... equivalent to a rule favouring

62

THE REPUBLIC OF SCIENCE

theories with the highest possible empirical content (LScD, p. 121). The ellipses replace a cross-reference to the place where rules (R5)-(R8) are formulated, (R5) clearly implying that severity of tests is a positive value. What is sometimes known as positive knowledge is here shown to be equivalent to degree of falsifiability. Thus is blocked the objection that falsificationism cannot explain the positive knowledge of science. Again as a corollary, this time to (R1), Popper wants to estop one of the commonest evasion devices in practice, the appeal to accident: (R12) I propose that we take the methodological decision never to explain physical effects, i.e. reproducible regularities, as accumulations of accidents (LScD, p. 199). This rule is introduced during a technical discussion of probability in physics, where the issue is how to prevent probabilistic hypotheses from rendering the system of statements unfalsifiable. Yet it has broad implications, suggesting that in science we should refuse to be satisfied with explanations from accumulation of accidents. Indeed, the aim of science is to seek testable explanations of observed regularities. This sharply distinguishes science from technology, where an air crash, for example, can sometimes be satisfactorily explained as an accumulation of accidents (see Agassi 1985). In the same context of discussing the falsifiability of probability statements, a further rule is considered necessary: (R13) a rule ... which might demand that the agreement between basic statements and the probability estimate should conform to some minimum standard. Thus the rule might draw some arbitrary line and decree that only reasonably representative segments (or reasonably 'fair samples') are

THE 1935 PROTO-CONSTITUTION

63

'permitted', while a-typical or non-representative segments are 'forbidden' (LScD, p. 204). A final example from this context of probability statements and their confinement is: (RI4) [T]he rule that we should see whether we can simplify or generalise or unify our theories by employing explanatory hypotheses of the type mentioned (that is to say, hypotheses explaining observable effects as summations or integrations of micro events) (LScD, p. 207). Here the challenge was the claim that all observable events should be explained by micro events. Noting that this doctrine is similar to certain forms of materialism, Popper calls it a "metaphysical hypostatization of a methodological rule which is in itself quite unobjectionable" .

4. Assessment

What then is to be made of this list of 15 rules «SR) + (Rl)(RI4))? First and foremost to reiterate: it is incomplete. A Constitution for the Republic of Science would need many more rules, along with some specification of their institutional embodiment, including rules for citizenship, government, dispute settlement, and legislation. Popper makes no effort to organise the rules systematically and lay them out in a table so that they can be checked against one another and debated in relation to one another and the aims they serve. This may explain why they are seldom discussed. The occasional challenge philosophers mount to one or another rule is rarely framed in a manner that suggests appreciation of the innovative brilliance of the overall idea. My conjecture is that both Popper and his critics suffer from a myopia about the institutional tum. Let us re-frame the rules by spelling this out. What Popper's rules amount to is this. The demarcation between science and non-science cannot be stated in an abstract or logical way, only in a practical way. Science is an activity

64

THE REPUBLIC OF SCIENCE

carried on in a self-selected community and this community is obedient to a set of rules which guide its activities. The community is both real and partial: real because it consists of actual human beings, partial because their participation in this community is only one of the roles they playas social actors, and by and large their other roles are played in distinct, if overlapping communities. All scientists are, for example, citizens of nation states; virtually all scientists belong to families of orientation (their natal family), and the majority also to families of procreation. Most scientists belong to communities and groups in civil society (political, religious, voluntary, recreational). The vast majority of scientists belong to some one or other large institution that provides them with a livelihood: government departments, universities, laboratories, museums, business corporations, etc. In none of these "outside" involvements is it necessarily the case that scientists conduct themselves according to the rules of procedure that Popper has been sketching out. Notoriously, there are critical and open-minded scientists who cleave to dogmatic religious beliefs. In raising their children such scientists commonly want them to cleave to the family religion. The social institution of science and its constitutive rules is, then, only part of the overall social life of its members. Secondly, there is an ambiguity at the heart of the institutional view: science is an institution, and it consists of many separate institutions. Furthermore, while the separate institutions will tend to be concrete - laboratories, for example, which house people - the institution as a whole has an abstract character. A group of scientists based in a laboratory will have contact with peers in other laboratories both at meetings and by means of communications media. This abstract group of specialists in some research area will itself constitute a form of separate institution. This communicative aspect of science cuts across most social boundaries: ethnicity, class, income, geography, nation, age, seniority, and so on. Such specialist institutions are mostly self-constituted and self-governing, yet scientists differ from specialists in stamp collecting or movie still

THE 1935 PROTO-CONSTITUTION

65

collectors. Their work has results that may eventually interest and even affect the public in radical and in unknown ways. The distinctive character of science as an institution goes far deeper than, for example, Rudwick (1985) or Shapin (1994) allow. The very fact that the institution makes it possible for someone to do good scientific work whilst privately adhering to anti-scientific ideas shows something of its strength. Institutions, not private views, are what make science possible: they are necessary; in their absence there is no science. Institutions are not, of course, sufficient. (Since science requires, inter alia, luck, ingenuity, and creativity, one may doubt whether sufficient conditions can be specified.) The institutions embody and transmit the world view, the scientific outlook, including the transcendental aspiration to truth that some sociologists of science try hard to eliminate (see chapter 6 §4 below and Miller 1999). Does the social institution of science resemble any other? Despite his analogies of games and jury trial, Popper nowhere says anything about the matter of internal organisation. The rules give us no guidance as to how the republic is governed, or to whether there is established leadership. Indeed, although Popper formulates possible rules, he leaves open how discussions of them and their suitability are to be conducted and how decisions to amend accepted rules are to be made. Thus I admit that it stretches matters to see these rules as a 'Constitution for the Republic of Science' or even as a proto-constitution. They read more like a proposed set of procedures for debate by a body already in place, hence to some extent taken for granted. Yet questions of governance, reform, and responsibility, as we shall see, are just the questions that Popper later raises about political institutions and social reform in PoH and OS&/E. Of course, at the time Popper wrote, science was well established within western society. It was in many ways unique among social institutions. Its activities were centred in the participating institutions of the learned society, associated journals, universities, laboratories, and international conferences. Physics was understood by physicists to be a far-flung invisible college uniting colleagues from around the globe in common endeavours. But to point this out is also immediately to point out

66

THE REPUBLIC OF SCIENCE

that there was a leadership (cp. Polanyi 1962). To be the senior professor at certain institutions, or the head of a national academy, or the head of a learned society, was a concrete form of leadership for which there was fierce competition. There was also the less tangible phenomenon of intellectual leadership, distinct from even if overlapping such established posts. Accession to intellectual leadership turned solely on the esteem directed at the work of a particular person, regardless of formal post. Furthermore, without doubt, there were elements of charismatic leadership - scientists who could energise others with a vision, regardless of whether their own work was important. Formal, intellectual, and charismatic leadership can all be beneficial or dangerous, the latter in suppressing ideas or promoting nationalism, for example. These points will be pursued in chapter 5. Formal leadership was mostly exercised in prestigious institutions with constitutional means of selecting positionholders. Election to professorships, elections in national academies and in learned societies, the selection of journal editors, were all conducted according to rules that bore little or no relation to Popper's methodological or procedural rules. It is not at all clear that the weight of these offices and their electoral rules was always brought to bear to move science in the direction Popper intended, namely, critical open-mindedness (especially if Kuhn is to be believed). Indeed, the received view is that the formal leadership is not a leadership at all, merely selfless peers. Polanyi and Kuhn view them as authorities. Popper does not. Apart from his general anti-authoritarianism, there are specific reasons for not treating scientists and scientific institutions as authorities. Institutions and their office-holders develop vested interests, some but not all of an intellectual character. Thus, for example, injunctions to prefer the most severely tested theory, or always to apply a new criticism of the theory one opposes to the theory one favours, might not align with institutional pressures to act otherwise. Scepticism towards formal institutionalised authority is prudent in science as it is in social life generally.

THE 1935 PROTO-CONSTITUTION

67

A few general words about intellectual and charismatic leadership in science might be in order, since these are phenomena not explicitly discussed by Popper or Kuhn. Intellectual leadership usually flows from the authority of ideas. An intellectual leader is one whose ideas capture the public (narrowly or widely construed as the case may be). Nobel Prize winners are the obvious case in science, as are those who head prestigious institutions or edit top journals. The usual view is that intellectual achievement led to the reward, and that the achievement and the reward confer authority. For Popper, however, who opposes authority, and who declared that "we must break with the habit of deference to great men" (OS&IE, I, p. v), leadership in science like leadership in general is more a matter of responsibility than of authority; it is a matter of embodying in one's person the modest, critical, and sceptical attitude appropriate to anyone who has contributed to knowledge. Charismatic leadership derives not from achievement but from an aura, the je ne sais quoi that makes some persons more attractive, more seriously regarded than others. This kind of leadership, which Max Weber calls a kind of authority, is highly troublesome for any naIve attempt to model the rationality of science. In principle, attractive, fluent, witty scientists should no more be treated with deference than dull high achievers. Yet science produces its quota of charismatic persons who may not otherwise embody the modest, critical, and sceptical attitudes so beneficial for science. All three forms of leadership - formal, intellectual, and charismatic - have the potential to subvert the critical spirit of the Republic of Science, especially if they are accorded authority. The solution clearly must be institutional. We shall see in later chapters (especially 4 and 5) how adequately Popper addresses such problems. Popper's rules and their rationale enable us to see the defects of actually existing science and to offer suggestions for its rectification. Thus Popper's work is less descriptive than it is normative sociology, one suited to the discussion of policy and planning.

68

THE REPUBLIC OF SCIENCE

Because Popper presents his methodological rules without reference to any institutional framework, the task of defending them, and of debating them in a responsible way, is made difficult. Popper has not shown how to turn them into a constitution for science, not even into a declaration or charter of rights (and responsibilities) which the constitution embodies in a system of governance. But perhaps this objection is misdirected. For the methodological rules Popper proposed are not intended for this or that university, this or that physics laboratory, nor are they intended for some kind of United Nations Organisation of physics. They are directed to science conceived of as a general and abstract republic. They enjoin: here, if you want to respect and advance these aims, is a set of proposed procedures. But 'you' is unspecified, and what 'you' are to do if you have what you think is a better idea, is also unspecified. These proposals, published as they are by a philosopher of science, help constitute an academic specialty of that name to debate them. Yet since they are proposals to guide the practice of research in its most general aspect, their testing and debate might seem to be most appropriately carried out by the community of scientists proper. At this point I want to elaborate on a notion of science already mentioned: not as a series of concrete institutions but as an invisible college, an abstract institution (rather like language). We might see Popper's rules as addressed to this wider community of science, one deriving from self-identification and not subject to institutional gate-keepers. Popper's arguments and rules will then seem directed to men and women of good will who want to advance the project of science. This could explain why individual scientists, in their autonomy, often pledge allegiance to Popper's vision of science, while the institutions to which they belong seldom do so. (Mainstream academic philosophy of science is even harder pressed to incorporate Popper's ideas because they accord it no special role or expertise.) Popper's lack of attention to the concrete institutional embodiments of science, and his efforts to formulate rules outside such constraints, betrays, I believe, a fundamental mistrust of existing institutions and practices. They were and are hierarchical and authoritarian and no longer welcoming to any man or woman

THE 1935 PROTO-CONSTITUTION

69

of good will, as well as being subject to the corruptions of power and wealth. Later in his career, commenting on Kuhn's notion of "normal science", i.e. one widespread, concrete institutional embodiment of science, Popper declared: 'Normal' science, in Kuhn's sense, exists. It is the activity of the nonrevolutionary, or more precisely, the not-too-critical professional: of the science student who accepts the ruling dogma of the day; who does not wish to challenge it; and who accepts a revolutionary theory only if almost everybody else is ready to accept it -- if it becomes fashionable by a kind of bandwagon effect. To resist a new fashion needs perhaps as much courage as was needed to bring it about. .. I admit that this kind of attitude exists; and it exists not only among engineers, but among people trained as scientists. I can only say that I see a very great danger in it and in the possibility of its becoming normal Gust as I see a great danger in the increase of specialisation, which also is an undeniable historical fact): a danger to science and, indeed, to our civilization (Popper 1968, pp. 52-53).

In this passage Popper is explicit that his philosophy of science portends a radical critique of the concrete institutions of science and also a desire to hold them to normative standards that men and women of good will can agree upon, not those authorised by the entrenched experts, or by fashion. The reading I have been developing of Popper as a scientific republican is novel. Its closest approximation is to be found in Jeremy Shearmur's work, when he notes the necessity of the methodology adopted towards a theory (1980, pp. 153-55). In fact, very little commentary has been written on the details of Popper's proposals for rules, never mind their being a protoconstitution for an abstract republic of science. One of the closest examinations of Popper's rules was carried out by Johansson, almost alone among philosophers (Johansson 1975). An alternative institutional view has been offered by Polanyi (1947, 1952, 1958, 1962) and Kuhn (1962 following him on many points). Feyerabend found Popper's rules authoritarian and seemed to think that both institutions and rules were better done without (1970). Johansson sympathises with this view. A more serious attempt to do without rules is that of Hattiangadi. Let us look at these critics in the following sections, and then go on to implement (R9).

70

THE REPUBLIC OF SCIENCE

5. Johansson's Critique The Swedish philosopher Johansson was the first to make a concerted effort to discuss Popper as a proposer of methodological rules in his published dissertation (Johansson 1975).24 His work merits serious consideration and has yet to be surpassed. By means of a close reading of the text, Johansson extracted more than twenty rules as well as a "few" he imputed from the context (Johansson 1975, p. 15).25 These rules he then shuffled and dealt according to a scheme said to be "unproblematic" and which has seven groups: (1) The Demarcation Rule (2) Rules Against Conventionalist Stratagems (3) Rules Demanding a High Degree of Falsifiability (4) Acceptance Rules for Basic Statements (5) Acceptance Rules for Theories (6) The Rule for Falsifying Probability Statements (7) Rules for the Social Sciences

The last category, (7), are rules Johansson extracted from

The Poverty of Historicism, which we will consider in the next

chapter. Johansson's aims are quite different from my own. He wishes to offer a critique of Popper's entire project: of the value of methodological rules for fostering the growth of scientific knowledge. His conclusion is that the outlook for Popper's rules is dim:

Popper's methodological rules, impressive at first sight, do not keep what they promise. Some of them are trivial, some of them cannot be applied due to vagueness or lack of an adequate measure of the degree of falsifiability, and others are simply not suited to the aim of promoting the growth of knowledge. The requirement that new theories should

24. Johansson's book is somewhat marred by its unaccountably churlish tone.

A later and shorter effort by Blaug (1980) to extract and discuss the rules suffers from the confusions wrought by Lakatos. 25. Johansson did not, however, identify or discuss what I have called the (SR) or meta-rule about the construction of rules.

THE 1935 PROTO-CONSTITUTION

71

contain some "simple, new, and powerful, unifying idea" is really the only thing left (Johansson 1975, p. 118).

It is clear that Johansson expects the rules to lead us to progress. But not only does Popper not promise this, his

negativism is an explicit disavowal of any logic of discovery. In this respect the title of his English translation, The Logic of Scientific Discovery in misleading; the original German, 'The Logic of Research', was much clearer. 26 Johansson's arguments tum again and again on whether basic statements are always primary, i.e. dominate universal statements. If they are not, Johansson thinks there are many ways out of a refutation. He also holds that refutation of theories always depends on (usually unarticulated) ceteris paribus clauses which, again, make falsification dependent on something beyond the logic of statements. 27 We need not concern ourselves with the details of his attempt to show that Popper's methodological rules flounder on these points. Otherwise expressed, it was exactly the logical inescapability of such conventionalist manoeuvres that led Popper to propose methodological rules for science. He himself showed that falsification could be logically evaded. The difference between evading difficulties and facing up to them was a matter not of logic but of convention. 'Conventionalism' was Popper's name for the convention of accommodating falsifiers; 'falsificationism' his name for the convention of maximising the impact of falsifiers. Formulating the latter convention as a rule (SR) brought it into the arena of rational debate (does our convention stand up to scrutiny?). Johansson misses this and, when pointing to a supposed deficiency in a rule, sees no need to propose a better rule. Instead, he says he is inclined to agree with Feyerabend that there are no methodological rules (Johansson 1975, p. 158). This conclusion does not follow from the difficulties he raises if falsificationism is under discussion. If the 26. Not in every respect, of course. Both titles emphasise logic rather than decision. Incidentally, the original German subtitle, "On the Epistemology of the Natural Sciences", is unaccountably dropped from the English translation. 27. Lakatos made much use of this line of argument in his critique of Popper ..

72

THE REPUBLIC OF SCIENCE

issue is a hope for a logic of discovery then Popper would be the first to agree that there isn't one. Johansson's critique of Popper's methodological rules proceeds in complete innocence of the proto-sociology of science I have shown them to contain. Regardless of the success or lack thereof of Popper's particular rules, what of the value of the programme itself? Rules are conventions, conventions are social practices. Popper is saying that science is a special form of social practice, one that exists alongside many others, one that is constantly altering and improving its self-understanding and its practices, hence one not at all easy to pin down or demarcate. Popper says clearly that falsifiability is a necessary but not a sufficient condition for demarcation and that no logical or formal criterion will be necessary and sufficient (cp. PoH, p. 156; Economica 12(1945):87). To get science we have to build a community of those intent on falsification who voluntarily submit to a regime of rules that keeps them honest. Their logical practices are not self-sufficient. Johansson criticises Popper for not providing a water-tight logic of science when it was Popper who showed that no water-tight logic of science was possible. Feyerabend's answer, anything goes, which Johansson is inclined towards, is scarcely argued. Its gist is that all rules, all institutions, tend to become authoritarian. Hence all should be resisted and subverted some of the time, and some should be resisted and subverted all of the time. This amounts to little more than Popper's anti-authoritarianism in society, politics, and knowledge dressed up in flamboyant and provocative rhetoric. Feyerabend, it should not be forgotten, translated OS&IE into German. The first detailed scrutiny of Popper's ideas on methodology, by Johansson, is, then, something of a disappointment. Although Johansson has some interest in the social sciences, as we shall see, he did not take up the social dimension to Popper's discussion of method and thus could not elaborate his criticism in ways that would indicate how we can do without methodological institutions or their equivalent.

THE 1935 PROTO-CONSTITUTION

73

6. Polanyi, Kuhn, Feyerabend Michael Polanyi offers an alternative social vision of science that insists on the necessity of institutions but not on a republic of open and debatable methodological rules. Polanyi is an authoritarian republican when it comes to the structure and functioning of science. Published the year before the English translation of LScD appeared, Polanyi's Personal Knowledge can nevertheless be reconstructed as in dialogue with it. Indeed, Polanyi is equally explicit that there is no formal solution to the demarcation problem: the part played by personal knowledge in science makes it impossible to formulate any precise rule by which [empty] speculations can be distinguished from properly conducted empirical investigations (Polanyi 1958, p. 153).

"Personal knowledge" is Polanyi's term of art for the exercise of a skill, a capacity for judgment that a scientist learns on the job, i.e. by being taught by other scientists and eventually becoming a practitioner. This skill consists of the intellectual power to recognise rationality in nature. Personal Knowledge argued that the conventions of teaching and doing science were personal and inarticulate. Scientific training was a lot like surgery, or any other practical art, it required hands-on transmission. Thus Polanyi wrote: our appreciation of scientific value has developed historically ... much as our sense of justice has taken shape from the outcome of judicial decisions through past centuries (Polanyi 1958, p. 158).

Thus Polanyi places great emphasis on the living historical tradition of scientists doing science and being the collective judges of what is or is not science, as well as of what is or is not good science. There are rules for the conduct of science, corresponding to what others call methodological rules, but which Polanyi calls 'maxims'. "Maxims are rules, the correct application of which is

74

THE REPUBLIC OF SCIENCE

part of the art which they govern", but this does not mean they are offered to the public for rational consideration. Maxims cannot be understood, still less applied by anyone not already possessing a good practical knowledge of the art (p.31).

"The art" here refers, confusingly, to science. The picture is of skill being exercised by a community of practitioners whose procedures, beliefs, and valuations are mutually determined. Meanwhile "the premisses of science are tacitly observed in the practice of scientific pursuits and in the acceptance of their results as true" (p. 161). Thus the implicit premisses and explicit maxims operating in science will change as science changes, and knowledge of them is to be found only among initiates. Outsiders cannot even understand still less apply them. All of this of course places a great responsibility on science and scientists conceived of as a community carrying on an historic tradition. For, in the last instance, it is the community of science that in practice handles the demarcation problem. Polanyi describes it as the "authority of a community of people accredited as scientists" (p. 163). He spells out that accreditation as being extended to those trained in recognised schools, those becoming members of established learned societies, those exercising editorial or refereeing responsibilities for specialised journals, and so on. All these bodies train newcomers, or engage in ongoing acts of accreditation. This is a plausible sociology and a woefully inadequate philosophy - despite Polanyi' s able defence of it by contrast with primitive beliefs (the Azande) or with the parlous state of Soviet science (Polanyi 1952, 1958). Scientists may see themselves as a priestly guild and society may endorse that status, but the transcendental claim to knowledge of the way things are, transsocially, trans-culturally, trans-temporally, is not a function of such attitudes or of such endorsement. All of Polanyi' s philosophy, except the overtones of the mediaeval guild, were taken over with quite inadequate acknowledgement into Thomas Kuhn's much more widely discussed theory of science (Kuhn 1962). Even paradigms or disciplinary exemplars, as Kuhn calls them, are a relabelling of

THE 1935 PROTO-CONSTITUTION

75

Polanyi's "formative episodes". Polanyi discusses "formative episodes" in the history of science to show the sorts of things that go into the content of personal knowledge (perhaps sliding too easily over the neglect of the history of science in the standard curriculum of all science specialties). The objections to Polanyi's views from Popper's standpoint are obvious: they license dogmatism (this is just the way it is done by the experts, that's all there is to it); and they are antidemocratic, even authoritarian. Only insiders, not outsiders, can raise questions about whether science may be in a good or a bad condition. Only insiders understand it, only insiders have the authority to be heard.28 Provided that science is good at selfpolicing, there is nothing to worry about. Judging by its continuing success, this is the case. One has to remind oneself that Polanyi published this argument in 1958, that is, thirteen years after the atomic bomb. True, the atomic bomb was the most spectacular technological application of scientific theories ever. It was also the largest ever public investment in science and technology to that time. Because of wartime conditions, neither the fact of the investment nor its scale was debated in the polity at large. The success of the project (by some measures) protected it from too much critical scrutiny. In the years since the birth of Big Science in the Manhattan Project, however, there has been large public investment in equally secret projects, and in more public ones. While the scientific community has always been closely involved in feasibility and design questions, so have politicians, informed lay persons, and big business - none of the latter willing to accept constraint to the effect that their outsider status should silence them. Most natural science research these days requires a considerable commitment of resources, often governmental. While micromanagement of the allocation of these is in the hands of scientists, global budgets, areas of concentration of resources, and monitoring of the expenditure of funds are the business of a 28. Horton (1993) argued that it is closedness to outside challenge that keeps primitive thought primitive. In philosophical terms: the circumscribed rationality of fideism is just what keeps a group from breaking through to science.

76

THE REPUBLIC OF SCIENCE

much wider community. Is there a connection between what I have called "keeping science honest" and this now necessary interface with those representing the public? Popper, we recall, introduced some of his rules to guard against the possibility of science degenerating into a system of reinforced and irrefutable conventions cut off from the impact of experience. Polanyi, it seems to me, displays some complacency in the face of this possible corruption of science, as well as in other more vulgar kinds of possible corruption. He is against it, of course, but thinks all we can do is to rely on the good sense of the leadership to prevent it. When a guild cannot get by without regular injections of public funds the issue cannot be evaded by pointing to the past record of the guild and demanding unquestioning acceptance of its authority. Even were its record unblemished (and in the case of science that would be a contentious claim), unquestioning acceptance is not a prudent way to conduct public business. The prudent way is to have checks and balances, including scrutiny by sceptical outsiders. Popper's rules accomplish that, for, by being explicit, they constitute public reference points (these issues are taken further in ch. 6). Feyerabend's (1970) objection that methodological rules give scope for authoritarianism and exclusionism is a genuine difficulty. His anarchistic solution, summed up in the slogan "anything goes", hardly follows from that concession. If a society with the rule of law is authoritarian, discriminatory, or whatever, the remedy is not to abolish laws but to reform the laws and/or their enforcement. If Popper's rules had the unintended consequence of permitting guilds to avoid the check of experience then, by his own lights, that would be a powerful argument for reform of the rules. For them to be reformed they have to be articulated and discussed, so Polanyite 'ways of doing things', and Feyerabendian anarchism are hand-waving rather than solutions. The allusion to a society living under the rule of law brings us back to our constitutional analogy. Popper's rules are articulated versions of those conventions or social practices which need discussion in the house of science because past rules as embodied in inarticulate practice may have inhibited the growth

THE 1935 PROTO-CONSTITUTION

77

of scientific knowledge. Popper was writing in the aftermath of the revolution in physics ushered in by relativity and quantum theory. There were difficult questions about the compatibility of the two, and some extraordinary claims to completeness and finality were being made on behalf of quantum physics, as well as about uncertainty relations and about the probabilistic nature of laws. What sorts of rules should govern debate and experiment in these changed times was a timely topic to which Popper was eager to contribute. What he saw, I believe, but did not quite succeed in articulating, was that the central issue was one of scientific selfgovernance. Not self-governance in the way that Robert's Rules of Order might govern, let us say, meetings of the American Physical Society. The governance was of the Invisible College,29 and the kind of governance was as much of thought as it was of concrete social interactions. The search was for rules that would prevent science becoming complacent, dogmatic, or merely true by convention, and rather keep it critical, tentative, and in contact with experience. Popper saw that the key to this search lay in social practices, and that institutionalising these in articulated rules facilitated debate, as well as counterbalancing the dangers.30 So he was being a social scientist of a kind: he was proposing design details for the commonwealth of learning which were in effect part of its constitution in matters intellectual. His 29. As I understand it, Derek da Solla Price gave modem life to this term of

Robert Boyle's, using it to refer to the invisible collegium of peers that steer, and also constitute, a specialty. Invisible because it is a network of mutual recognition which may never have seen all its members convened at one time and place. My usage widens the term to describe science as such, which, being a dispersed network on an even larger scale and over time, could not ever be convened at one time and place. Boyle simply meant dispersed intellectuals as opposed to those living in a college. 30. In a handwritten and undated draft of a letter to Polanyi, composed, judging by internal evidence, in 1951 or 1952, Popper conceded a " farreaching and interesting" parallelism between Zande religion and modem science (Polanyi 1952). The difference, he maintained, was "structural", a concept he does not further spell out. Such a structural difference, he concedes, could be epicyclically explained away. It seems that he is saying the structural difference lies just in the system eschewing such epicyclical and convenient evasions. See KRP to M. Polanyi, Hoover 339/1.

78

THE REPUBLIC OF SCIENCE

view recognised that science was not the province of geniuses and giants but of ordinary persons who combine their strengths by creating well-designed institutions. And it is by means of these institutions that we produce science and are able to distinguish it from other human endeavours which operate under different sets of institutionalised rules and different structures. (A further comparison of the republic of science according to Polanyi and Popper is made in chapter 6, §1).

7. Hattiangadi's Critique. In a remarkable series of papers, most notably "A Methodology Without Methodological Rules" , my York University colleague J. N. Hattiangadi has offered a wide-ranging critique of the project of demarcating science by means of submission to methodological rules (Hattiangadi 1978-1979). In a footnote (60) it is made explicit that Popper is one of the targets, especially his ideas on the role of falsifiability and of verisimilitude. Yet we shall find that despite the provocative title, the criticisms do not hit Popper at all, and that his methodological project as articulated in the present close reading survives Hattiangadi' s demurrals. Hattiangadi is much influenced by Polanyi and his follower Kuhn. He cites with approval the following passage from Personal Knowledge: The large amount of time spent by students of chemistry, biology and medicine in their practical courses shows how greatly these sciences rely on the transmission of skills and connoisseurship from master to apprentice. It offers an impressive demonstration of the extent to which the art of knowing has remained unspecifiable at the very heart of science (Polanyi 1958, p. 55).

The content of tacit knowledge is clearer in Hattiangadi than in Polanyi, however. Hattiangadi holds that in their apprentice training by their masters scientists absorb a sense of the structure of problems in their specialty. Hattiangadi takes his cue from an unelaborated line or two in the preface to LScD: A scientist engaged in a piece of research, say in physics, can attack his problem straight away. He can go at once to the heart of the matter: to the heart, that is, of an organized structure. For a structure of scientific

THE 1935 PROTO-CONSTITUTION

79

doctrines is already in existence; and with it, a generally accepted problem-situation. That is why he may leave it to others to fit his contribution into the framework of scientific knowledge (LScD p. 13).

Hattiangadi makes a bold expansion and generalisation: all intellectual activity takes place within historically formed traditions. These traditions crystallise around lines of debate. Lines of debate, in turn, come from solutions proposed to intellectual problems. Problems are contradictions in our system of hypotheses that threaten to allow the suppressed side(s) of the debate to become dominant. The central question for Hattiangadi has to be, what distinguishes the intellectual tradition we call 'science' from other intellectual traditions, including those that constitute religions? Hattiangadi says that he believes the difference is institutional and not intellectual (Hattiangadi 1978-79, p. 123). It seems to me he would get no argument from Popper there since, as we have seen, Popper abandoned falsifiability as an intellectual criterion and instead embedded it within a methodology, a set of decisions or choices about how to conduct enquiry articulated as rules. Popper and other empiricists want to anchor science in experience, especially repeatable experience. Thus the intellectual tradition needs to build into its institutional fabric rules that foster that relation to experience, and which discourage evasion. Revelation, the divine word, inspired prophecy, these might all be important criteria in other intellectual traditions, but they are eschewed in science. When we look at those episodes in the history of science where science was apparently held to be problematic because it conflicted with religious dogma we see where the exclusionary rules came from. Those rules are not explicit, but they are well understood. Hattiangadi's answer to all this would have to be that the very presumed separation of religion and science is itself an historically formed intellectual tradition, one by no means universally acceptable or unproblematic. The demarcation between science and non-science then becomes a contingent historical one. But history cannot validly explain current structure and function, which must be independently addressed. The beauty of Popper's rules, as we have seen, is that they sketch the aims

80

THE REPUBLIC OF SCIENCE

and constitution of the institution of science, without specifying how different the rules are from those of other institutions; without reference to any historical warrant. Independent arguments are offered for structuring the institution with this or that rule, mostly to the effect that the rule will keep the system honest, i.e. responsive to experience rather than to our prejudices and hopes. Peter Munz (private communication), echoing a point also made by Gellner (1974), points to the way beliefs and rules are used as social bonds. Acceptance of beliefs and rules is often a sign of membership and a form of solidarity. Rejection of beliefs or rules excludes one from the community. Science strives to be a rather different kind of community. Loyalty to particular doctrines, persons, or rules is not required, since they all may in due course be replaced. As Gellner remarks, Newton gave way to Einstein without seriously disrupting the republic of science, still less society in general (Gellner 1974, p. 167). Contrast the social consequences that flowed from the dispute within Christianity over adding the phrase "and the Son" to the words "and in the Holy Ghost who proceedeth from the Father" in the Apostles' Creed. Excommunication, rivalry, cultural estrangement, and war were all consequences of this Great Schism. Consonant with its transcendental aims, science strives to be the social formation that can tolerate even those who are corrupt: who seek advancement, power, advantage, to serve the cause of anti-science. It relies on its institutional strengths to overcome such individual lack of acceptance of scientific ideals and concentrates solely on the contribution, if any, such individuals make to ongoing projects. Thus venality and dishonesty - provided they are not pervasive are compatible with making a scientific contribution. Science is the most liberal version of the open society (see my 2000 and chapter 6). Hattiangadi's theory of problems does not do away with methodological rules. It simply warns that methodological rules, unlike the rules of logic, are not immutable and, more specifically, that they may vary with the theories under discussion. Rationality, for Hattiangadi, is specific to context (this is an idea with which Agassi and I are also associated

THE 1935 PROTO-CONSTITUTION

81

(Agassi and Jarvie 1987». What controls context is what Popper was later to call "the logic of the situation" in a specific intellectual tradition at the point at which a difficulty or problem is discovered in it. The logic of the situation is the actions that seem most appropriate to realise an aim, given the constraints imposed by the situation, including our appreciation (i.e. knowledge) of it. Nothing in the rules we have been discussing seems to me in conflict with this. Rationality is always relative to aims and the situation, and this includes awareness of rules and the possibility of suspending them, pro tern, for a purpose. Thus those possessed of a metaphysics (as has been alleged regarding Faraday) may choose to isolate their metaphysics from the rigours of the rules (Agassi 1971). Done consciously and critically this can do no harm. Done as Hattiangadi and Polanyi suggest, tacitly, in the training, including what Hattiangadi calls "the rituals of science" that may rest on underlying outdated views (Hattiangadi 1978-79, note 71), seems to me to muffle the critical attitude, perhaps even to discourage it. 8. Autocritique In discussing, in §3 above, (R9) , which states that we should always try to apply our criticism of someone's else theory to our own theory, I said that it should be applied to the endeavours of the present book. A number of further criticisms can be raised about the thesis of this chapter - that Popper proposed a social rather than a logical solution to the problem of demarcation - and it is time to look at them. I think they boil down to three main ones. First, about the claim that Popper's text can reasonably bear my reading. What is the scientific status of such a textual 'interpretation'? Is it falsifiable or otherwise rationally criticisable? Or is it apologetic? Second, is my interpretation of Popper not a reduction of a philosophical problem and its solution to a social problem? Would Popper have agreed to this? Should anyone? Third, if science is a social institution then a solution to the demarcation problem requires that we specify what is distinctive about it as a social institution. What are the features of science

82

THE REPUBLIC OF SCIENCE

that set it apart from such institutions as the family, Parliament, business, from metaphysics, mathematics, and pseudo-science? I respond to these in order. First, my interpretation in particular and interpretation in general. Interpretation is, trivially, not scientific and makes no pretence to be so. But interpretations can be rationally discussed. There are negative criteria to be met: do not distort, do not align authorities with your own views, do not conceal difficulties. There are also positive criteria: an interpretation should fit the text; an interpretation should not violate well-known facts about author, text, and world; and, an interpretation should throw light on some of the more difficult or disputed passages of the text. I have discussed rules of interpretation a little more fully in my (1987). Provided interpretation is approached rationally, that is, with the attitude Popper has captured with his formula: "I may be wrong and you may be right so from discussion we may both learn something", then critical assessment of interpretation may be undertaken (see ch.4, § 4). Working with these negative and positive criteria I could argue of my social interpretation of Popper on demarcation that it is clearly rooted in the text, including nuances of the text little discussed, even overlooked. I shall also claim later in the book that this interpretation gives unity to Popper's three classic works, which can be shown to be connected by their thinkingthrough of a philosophy of the social in order to make sense of the emergence, growth and maintenance of scientific knowledge. This thinking-through involves criticism of his own previous ideas. The fact that Popper has never presented any similar interpretation of his work would I think also stand in favour of mine. Auctorial intentions are tricky items to handle in interpretation. Authors may be the last persons to see, or admit, that their work has aspects, perhaps even deep aspects, of which they are unaware. All this is quite consistent with Popper's own views on interpretation and on the relation of authors to texts, including philosophical texts. Summing up, the present social interpretation of Popper's philosophy of science will help us to grasp his originality. It is a criticisable interpretation and I would welcome critical debate

THE 1935 PROTO-CONSTITUTION

83

about it. For example, it is not wholly consistent with things Popper has said and with the direction of he took in his subsequent work. Much of his post-1945 research probed further into the logic of science, not into its social aspects. What then of the other two critical arguments listed above, namely that my interpretation reduces a philosophical issue to a social one, ideas to society, and that, in doing so, I am making it impossible to solve the problem of demarcation? These are forceful objections that require direct replies. The charge of reducing a logical problem to a social problem is partly correct, partly not. It is not correct if 'reduction' means to eliminate or downplay logic. It is trivially true that without human society there would be no logic; it is trivially false that logical issues are no more than disguised social issues. No general reduction of logical issues to social issues is possible. What Popper argued is that a problem classically treated as logical (the demarcation of science) is insoluble in that form. Yet science does seem distinctive, i.e. demarcatable. The solution to its distinctness is found in its institutionalised rules of inquiry. These rules can be circumvented without violating any rules of logic. In the choice or decision not to circumvent them we fmd the hallmark of science. Such choice is not individual but social: it is constitutive of and dependent upon institutions. Moreover, being institutions, the rules of science are not like rules of inference: discoveries about the properties of logical systems. Institutional rules can be assessed against given aims, such as fruitfulness, and be discussed and modified. Science and its distinctiveness are open to modification. An attempt was made earlier in the twentieth century to shift the characterization of science from the realm of ideas (material mode) into the (social) realm of language (formal mode). The hope was to find formal characteristics of scientific statements that would demarcate them from other kinds of statements. One such attempt was the verification principle of the Logical Positivists: all and only empirically verifiable statements are scientific. So far from proposing a falsificationist version of this: all and only empirically falsifiable statements are scientific, Popper conceded that there was no formal way to characterise a

84

THE REPUBLIC OF SCIENCE

statement or a system of statements as scientific. He thus vacated any hopes raised by the linguistic tum, and he seems to have seen clearly that science had to be looked at institutionally, and that it was in the methodological institutions that the connection between science and experience would be found. After all, elsewhere he argued clearly that there was no logic of discovery: scientifically respectable statements came with no mark upon them (such as being free of metaphysics). Rather, a statement was checked by what scientists did to it, how they tested it. Testing is a procedure, a social practice. This leaves us with the third objection: if science is a social institution, then it is one among many. How then can the demarcation problem be solved? What makes science the unique social institution we take it to be: that devoted to a socially transcendent aim: knowledge, the truth? In the course of the argument science has been compared to other institutions which engage in inquiry, such as law courts and Parliament. To compare it to still other institutions of inquiry such as universities or learned societies would be futile, because many of these are component institutions of the grand institution of science as such. Pushed by the demarcation problem to say just what is so special about science, what differentiates it from philosophy, metaphysics, mathematics, pseudo-science and such like, I will at this stage simply point to the content of the methodological rules themselves: their insistence upon observability and repeatability, for example. As was urged earlier, these rules are necessarily an incomplete set. Consider how the rules are compiled. They are accumulated over time, historically. We gradually and incrementally learned from our mistakes and then tried to capture the result in a rule. Since science has historical ties to philosophy, mathematics, metaphysics, and even pseudo-science, we might expect that some of the rules will be there to save science from following those paths, while trying to keep to the path that leads to a discipline firmly anchored in experience and submissive to it. Once the problem of demarcation is sociologised, Popper's constitutional model of its special character comes into competition with other attempts to characterise it made by sociologists. It would not be inconsistent to enrich the bare bones

THE 1935 PROTO-CONSTITUTION

85

of Popper's sociology of methodological rules with some institutionalised values such as those suggested by Merton (discussed in chapter 1). But neither the republican nor the values model of science treat it as unchanging. If science changes, it is possible that its distinctiveness changes. How are changes made that preserve that distinctiveness; what, if anything, persists in the institution through such changes; how are the changes made, what procedure guards the process of reform? An alternative account of the distinctiveness of science is offered by the "strong programme in the sociology of knowledge" (also discussed in chapter 1). Arguing that all features and combinations of the social organisation of science can be found elsewhere in social life, the claim is made that what marks off science is its special status in society. It is a rationalisation to claim that science derives its status from its organisation and its products. Science is first and foremost a powerful and influential guild and its personnel and pronouncements are received accordingly. The strong programme differs from Polanyi only in its self-conscious attitude of detachment, its refusal to endorse, in some writers its tendency to debunk, the face-value of scientific claims.31 There is, I suppose, no doubt that science as an institution has high status. The question is, does the status derive from the achievements of science, or are the achievements seen to be such because of the status of the institution? The former is the common sense view, and would, I think, be the view shared by Russell, the Logical Positivists, Popper, and so on. To the strong programme we could put the question, if the status of science does not derive from its success in expanding our knowledge of the world, from whence does it come? One of the favoured answers in the discourse of the strong programme is to argue that it comes from social class (Shapin 1994). The institutions of science were gentlemanly and truthfulness was a value for gentlemen between whom there was trust. The status of gentlemen and their values were grafted onto science since, in the first instance of the Scientific Revolution in Britain, scientific 31. On its debunking tendency see Grove 1982, p. 555.

86

THE REPUBLIC OF SCIENCE

research was carried out by members of the gentlemen class. Struggling to make sense of Shapin (1994), Agassi writes: Perhaps Professor Shapin's thesis is that to date science still is in a sense a pursuit of gentlemen, of a "core-set," of a small platoon of leaders in each field of research (p. 415), since "science is a system of knowledge by virtue of its being a system of trusting persons" (p. 417, the book's fmal paragraph). Does this mean that every "system of trusting persons" generates science? No, surely not. He just wants to stress "the importance of trust" (p. 417, top). But of course. As Professor Shapin stresses, science evolves in a social setting, and this setting includes trust. But then gentlemen or gainfully employed, passive or active, they must trust. Where then does science come in? (Agassi 1997, p. 235.)

So far, all I have offered on the question of what makes science special is its rules and the republic they constitute. A fuller argument for the special character of science among social institutions is best postponed until Popper's enriched ideas about society have been discussed (chapter 6, §4). Whatever view one takes, one thing is clear. If science is not logically distinct but is rather a social organisation to enable us to proceed in a certain way, then legitimate historical and sociological questions arise. When Popper says that he tried to apply to social science his ideas on scientific method he is not writing like the Popper who pursues problems rather than fields or topics. That he spent much effort over ten years thinking about social and political organisation should, by his own philosophy, have come from problems, not applications. 9. The Social Aspects of Scientific Method In his skeletal early sociology of science Popper identifies a number of elements. There is the friendly-hostile cooperation of scientists, checking one another, competing with one another. This is how intersubjective testability works. There are points of social decision, upon, for example, whether to accept a basic statement as well-enough tested to challenge a theory. Also, for example, the choice of which of the logically possible explanations of a clash between experience and theory to pursue. Above all, there is the overarching structure of methodological rules, obedience to which constitutes science as institution. What

THE 1935 PROTO-CONSTITUTION

87

is absent is: science as community, science as culture, science as tradition, science as profession, science as one among many powerful guilds in modem society, science as power. Microsociologically there is no sociology of the interaction between science as primary group - the lab., the seminar, the meeting - and science in its more abstract form as a secondary group. There is no analysis of the role of the leadership, of the internal status system, of the role-structure, of recruitment, education, and socialisation. There is no sociology of how science commands the social, political, and economic resources its work requires, of its clashes with other institutions, including other thought-systems. We fmd, in PoH and OS&IE much thought and discussion of these matters in society as a whole, in education, in politics, and in the discussion of planning. But these ideas are not then reapplied to flesh out and correct the skeletal sociology of LScD. Instead, he rather simplifies and romanticises science by writing of it as a generalised Socratic seminar, even taking that simplified picture as something of a model for modem social and political life. He beautifies science, even in a work where he scorns the beautification of Plato. He himself tends to beautify Socrates, Russell, Einstein.32 As we delve into Popper's works of social and political philosophy we shall find that he thinks in a penetrating way about society in general. But he regularly fails to apply his criticisms of standard ways of thinking about society to his own early theory of the social aspects of science, and to his own attempts to generalise science as an exemplar for the conduct of other parts of social life.

32. For another example of the beautifying tendency, see Popper's letter to Nature about Schrodinger (1989).

Chapter 3 THE METHODOLOGY OF STUDYING SOCIAL INSTITUTIONS I do not intend to assert that there are no differences whatever between the methods of the theoretical natural and social sciences; such differences clearly exist, even between the various natural sciences themselves, as well as between the various social sciences. -- Popper (Economica 12(1945):78; PoH, p. 130)

1. The Logic ofSocial Scientific Discovery

In the same way that there is a received view of Popper's philosophy of science as primarily logical and linguistic rather than social, there is a received view of his works on the philosophy of politics and society. Accordingly, these are seen as extensions to the 'field' of the social of his ideas on the philosophy of natural science, together with, perhaps, some commentary on current events of 1938-1944. There are problems with this received view. It clashes with the biographical fact that Popper's interest in social and political problems preceded his interest in the physical sciences. Popper seems to have been a social activist and thinker perhaps even before he was aware of having original ideas on the philosophy of the physical sciences (Popper 1976, p. 113; Hacohen 1993, 1996, 1999a, 1999b). It also clashes with Popper metaphysics, which does not reify academic 'fields' as preexisting domains ('natural kinds ') in need of colonisation by the application of ideas from elsewhere. Better consonant with the facts, as well as with Popper's philosophy, is

INDIVIDUALS AND INSTITUTIONS

89

a critical account: the sufficiency of his methodological rules for science was subject to criticism from the institutionalised practices of the social sciences of history and of economics. In responding to these criticisms, Popper deepened and generalised the ideas about the social sketched in LScD, especially when he looked more critically at attempts to build institutions for gaining knowledge, such as schools, science itself, universities, and institutions for rectifying social problems. All this is better articulated in his four-part essay "The Poverty of Historicism" (Popper 1944-45). As a teenager, Popper confesses, his sense of injustice and fears about peace led him briefly into the Communist camp (Popper 1976, p. 33). After the death of some demonstrators he thought that the Party had displayed an irresponsible attitude to violence and suffering which made a mockery of their commitment to alleviating those conditions (Bartley 1987). A letter to Hayek relates that Popper undertook a serious study of Marxism in the 1920s, and completed a manuscript (presumably lost) out of which "The Poverty of Historicism" emerged. 1 He says that it took him many years to figure out that the heart of Marxism was a prophecy and a moral injunction, "Help bring about the inevitable" (Popper 1976, p. 35). This, he thought, was thin: the inevitable scarcely needed help. He found fault with Marxism in other ways. During 1935 he seems to have prepared a paper on his ideas and to have read a draft of it at Hayek's seminar at the London School of Economics and at the home of his friend Alfred Braunthal in Brussels. Yet, he says, he did not want to publish it. European social democrats were Marxists, and they were one of the main parties fighting fascist tyranny in Europe. It was only when the fascists achieved the Anschluss (uniting Germany with Austria) in March 1938 that he felt these restraints were lifted and he began to think of turning "The Poverty of Historicism" into a publishable paper. 2 1. KRP to F. A. Hayek, 14 March 1944, PP Hoover 305/13. 2. KRP to F. A. Hayek, 2 November 1943, PP Hoover 305113. (A slightly different account is given in the Preface to the Second Edition (Popper 1952) of Popper 1945.) It was not hubris that kept Popper from publishing. He did not

90

THE REPUBLIC OF SCIENCE

So far as I have been able to check it, Popper's account in his autobiography of his work on "The Poverty", the emergence of what was to become "The Open Society" as a kind of digression, and his completion of the fourth part of the former only after he had completed the latter, tallies with the scant records. The files show that it was only in 1943,3 after he had sent off typescripts of The Open Society under its then title "False Prophets - Plato, Hegel, Marx" to the United States to seek a publisher, that he was able to complete "The Poverty of Historicism". It was turned down by the first journal to which it was submitted, Mind. The reasons offered were its excessive length, especially as the editor doubted that anyone held the views it attacked (cf. Munz 1997).4 This "in spite of the rubbish which they [Mind] continuously accept".5 Popper then asked Hayek to try to place it; Hayek did in Economica, of which he was the editor. Hayek coincidentally would subsequently mediate the publication of The Open Society and also recommend Popper for a job that would bring him back from the intellectual exile of New Zealand (see also Hacohen 1996; Shearmur 1996). Popper strived mightily in the final revisions of the notes to The Open Society and "The Poverty" to stress his areas of agreement with Hayek on the nature of the social. This obscures not only their sharp differences, but also the extent to which Popper had changed his views since 1935. Unfortunately, Popper's archives at the Hoover Institution do not contain early typescripts of either "The Poverty of Historicism" or of The Open Society and Its Enemies, although expect publication of his work to change history. Rather, he considered the matter from a moral angle. He did not want to contribute, even in the smallest way, to the intellectual undermining of those engaged in the anti-fascist struggle. 3. KRP to F. A. Hayek, 2 November 1943, PP Hoover 305113. 4. Moore to KRP, 12 December 1943, rejecting "The Claims of Historicism": "1 felt from the beginning that this long exposition of views with which you do not agree is not of sufficient interest to be published by itself .. .l do not think 1 could possibly accept this paper for publication in Mind". PP Hoover 329129. 5. KRP to E. H. Gombrich, 30 November 1943, PP Hoover, 30012.

INDIVIDUALS AND INSTITUTIONS

91

there must have been many of them. 6 In the war-time years of paper shortage Popper had the habit of using the reverse side of the pages of previous versions for the carbon copies of his works. Thus it is next to impossible to reconstruct the evolution of his ideas towards the finished versions. All we have are the typescripts and proofs of the versions used by the printers, and of course the subsequently produced first editions. So, although we know that Marxism and psycho-analysis were in his mind long before Logik der Forschung was published in late 1934, we have no paper trail along which to trace step-by-step his efforts between that time and these major publications of the midnineteen forties. For purposes of this book, then, these first printed versions will have to suffice. I propose to treat "The Poverty of Historicism" as the earlier work despite its evident entangling, in the final writing, with The Open Society and Its Enemies. This is because it was begun and worked out much earlier, whatever modifications may have been made later. 7 Furthermore, OS&IE is a passionate political manifesto, whereas "The Poverty" is a methodological work, a critical survey of extant ideas about social science method. It contains a systematic attempt to diagnose and rebut the historicist method and purge its pervasive influence in history and the social sciences. Historicism is the idea that there are inexorable laws of history. Popper's essay is divided into four parts. In Parts I and II, he shows how historicism cuts across the standard dispute in German thought between those who argue for the unity of natural and social science (naturalism) and those who argue that the subject matter of social science (Geisteswissenschaft) imposes an entirely different method of approach (anti-naturalism). Historicism is common to both naturalist and anti-naturalist

6. Popper told me, in a conversation in June 1994, that upon removing to London he left these manuscripts behind in a trunk in the cellar of his New Zealand house. When the house was later sold the trunk could not be located. 7. I was anticipated in this chronology by Hacohen 1993.

92

THE REPUBLIC OF SCIENCE

parties and hence critique of it is deeper than that division. Parts III and IV are devoted to that critique. 8 In writing PoH Popper was forced to think as a social scientist, to examine critically the possibility of a social science challenge to his methodological ideas. He returned to these matters in The Open Society, but there the effort was entangled with considerations about the history of historicism as well as the concrete problems facing the post-war world and the task of social reconstruction. In much the same way as his philosophy of science had rejected induction as the characteristic method of science, his philosophy of social science rejected historicism as the characteristic method of social science. Induction was invalid and did not work, so it was relatively harmless. 9 Historicist thinking, by contrast, was an intellectual poison, one which could do harm. This is how it came about that Popper spent so much time on "historicism" - for he held it to be a false methodological doctrine that misled social scientists. Historicism as he understood it was the belief that there are inexorable laws of human destiny, and it had the methodological corollary that social scientists should engage in historical prediction. Historicism was, he believed, false, and its methodological corollary barren or worse. Since it was pervasive in the social sciences, as were various crude efforts to make the social sciences inductivist, his treatise consisted for the most part of destructive criticism of these two errors, and brief advocacy of a technological approach to social science, taking special note of the Oedipus effect (the capacity for views of the situation to affect the course of events). The Open Society is the most socially concrete of the three works and deserves treatment last, given the length and profundity of its ideas on the social. In the present chapter we shall 8. I have explored the structure and argument of PaR further in my 1982. 9. Popper maintained that whatever worked for stimulating ideas could be endorsed provided it remained a private affair. What he did not endorse was that any technique or method could obviate testing. See OS&IE, II, p. 308/324. Unfortunately, inductive thinking usually proceeds within extant intellectual frameworks and so it is lacking the boldness that Popper says is required for creative scientific thinking.

INDIVIDUALS AND INSTITUTIONS

93

be concerned with "The Poverty of Historicism" Parts I, II and III, as they appeared in Economica 1944-45. In the following two chapters, 4 and 5, we shall be concerned with the first edition of The Open Society and Its Enemies as it stood in 1945. In all three chapters there will be a running theme: Popper uses historical, social, and political arguments in a manner that presents a challenge to the adequacy of his social characterisation of science. A central social aspect of science is its governance by legislated methodological rules. Parallelling Hayek's claim about markets, Popper came to see the vast majority of social institutions as grown rather than designed. Language, the institution necessary for science, is grown and not designed. The obvious follow-up question is whether science itself was grown or designed. Historically, it no doubt grew (some scholars saying it finally emerged only in the nineteenth century). In its larger and more self-conscious embodiment in the twentieth century parts of it are certainly designed (e.g. most Big Science). Where do methodological rules belong in a grown institution? How can we manage the transition from a grown to a designed institution? One possible parallel is between a scientific community and a moral community. Moral communities certainly were in the first instance grown rather than designed. This does not inhibit us from exploring their implicit and explicit moral rules. Equally important, grown origins are no inhibition to additional moral rules being legislated, or to reform of existing ones. Criticism of the extant, whether institutions or ideas, is the high road to rational learning, and improvement is Popper's reason for singling it out. One might see the choice to live under a regime of rules that enjoin falsifiability as itself a moral decision, in so far as the very idea of pursuing the truth rather than appearances or fantasies can be construed as a positive moral value. It may seem rather odd that, in turning to the social sciences, Popper did not write a parallel work - say 'The Logic of Social Scientific Discovery'. It is odd only if we overlook Popper's adherence to the doctrine of the unity of method, the doctrine, that is, that science follows the same method, no matter what its 'subject matter' (a more Popperian phrasing would be, 'no matter what problem is under consideration'). In this respect a Kantian,

94

THE REPUBLIC OF SCIENCE

Popper held throughout his life that the divisions of the world and of phenomena were the result of inquiry and could not therefore predetermine its shape. Indeed, as a consequence of this, Popper maintained the methodological unity of all critical or rational inquiry. Thus the question in regard to the social sciences was whether there were any special methodological rules additional to or instead of those for the natural sciences. The basic requirements of falsifiablility and deductive testing were taken to be the same. Why is this so? What makes for the unity of method? According to Popper it is our general desire to know and our readiness to be critical. To this might be added, the attempt to develop comprehensive, unifying views. The more comprehensive, the more unified they are, on Popper's view, the more testable they are also. As we shall see, on occasion Popper sees the need for an additional rule for social science, but he never accepts that the social sciences follow wholly different rules. On the contrary, different rules lead to barrenness and worse. Popper's defence of his philosophy of science as extended to rational inquiry more generally pushes him, in tum, to examine social science critically. From the many topics he touched upon we shall concentrate on three: his emerging ideas on institutions and their bearing on scientific institutions; his ideas on individuals and how they bear on scientific behaviour; and his ideas on social rationality in general and their bearing on objectivity and testing in science. If the regime of rules is different for social science than for natural science then we have no unity of science, we have two (or more) republics, not one.

2. Johansson's Claim Ingvar Johansson, whose work was briefly discussed in chapter 2, in the course of his discussion of Popper's methodology lists a set of supplementary rules that Popper developed for the social sciences (Johansson 1975, ch. X). According to Johansson, Popper adds four methodological rules to those developed for science in general, not so much modifying them as specifying ways of dealing with special problems that arise within the social

INDIVIDUALS AND INSTITUTIONS

95

sciences. These could be called the Rule Against Prophecy, the Rule of Methodological Nominalism, the Rule of Interpretative Interest and the Rule of Methodological Individualism. In the following discussion I shall contest Johansson's claim that these are supplementary rules for the conduct of social inquiry. Not that they are invalid but that they are variants of rules already developed for the natural sciences; their variation helps criticise certain tendencies in social and political thought at the time Popper was writing. (Johansson works from the book version of PaR; I cite the corresponding passage in the 1944-45 original.) (SSR1) Rule Against Prophecy [A]ll theoretical or generalising sciences [should] make use of the same method, whether they are natural sciences or social sciences (PoH p. 130; Economica 12(1945):78).

As formulated this would seem to be a reiteration of the doctrine of trying to explain any event that can be described (R3 in ch. 2, above), rather than a specific ukase for the social sciences. It thus immediately sits uneasily with Johansson's claim that Popper is proposing rules specific to the social sciences. In his discussion of this rule Johansson lays stress on the method of science being the search for testable hypotheses (1975, pp. 8589); prophecies are not testable (they are verifiable but not falsifiable). Contrary to received views, Popper never declared prophecies in general or historicism in particular to be meaningless, or even false. Johansson constructs a strange argument to show that the distinction between prediction and prophecy will not hold. He attributes to Popper the position that it is "impossible to make historical predictions, i.e. predictions like 'society is developing towards socialism', 'science is developing towards its positive stage', etc." (Johansson 1975, p. 86). Since Johansson has in fact iterated these two predictions (and I have reiterated them) it cannot be impossible to make them. Did Popper claim impossibility? Did not Popper rather claim that social scientists have in the past confused prediction and prophecy? Prophecy is prediction that is irrefutably vague. 'The triumph of socialism is inevitable', would be an example.

96

THE REPUBLIC OF SCIENCE

Johansson's 'Society is developing towards socialism' is not irrefutable. It is vague, but circumstances can be imagined in which most of us would agree it was false. Indeed, by the 1990s its falsity was the developing consensus. Uttered at certain other times and places this prediction might well have seemed to be true. That impression of plausibility, even truth, it turns out, can now be explained as an effect of events that look from our present perspective like a medium-term trend, cut short in the last quarter of the century. Trends are functions not of scientific laws but of (changeable) sets of initial conditions. Johansson, however, convinced that Popper denies we can make social scientific predictions, attributes to him the reason that scientific predictions are conditional, and social scientific prophecy is unconditional. Against this, the position just fabricated, Johansson affirms that both natural and social scientific predictions are unconditional. He examines Popper's classic analysis in LScD of how the prediction that a thread will break when a weight exceeding its tensile strength is placed upon it relies upon unconditional predictions in the premisses such as that the structure of the thread does not change and that the weight does not change. The prediction 'this thread will break' is testable, in other words, only if conditions do not change. Johansson is correct that most scientific explanations assume or, better, postulate, local stability in the physical world. A world without such stabilities would be unpredictable - true chaos. Stability of reference is also assumed. If 'this' thread and 'this' weight can refer to different objects moment to moment then reference is too vague for prediction. Johansson is developing sceptical arguments from Lakatos (1974) that were still further developed by Cartwright (1983).10 One possible conclusion that Johansson might have drawn is that both natural scientific and social scientific prediction are indistinguishable from prophecy. But is irrefutable vagueness a genuine problem in science? Vagueness is, for Popper, always a matter of degree. We work to make some views definite enough that they are testable. That is all there is to it. The point, we remember, is whether 10.

For criticism of Cartwright see Agassi 1995/96.

INDIVIDUALS AND INSTITUTIONS

97

some putative claim of social science is prophecy rather than prediction. Since a prediction, according to Popper, results from a deductive argument, let us look at the simplest example. Johansson's argument amounts to claiming that the syllogism 'All men are mortal, Socrates is a man, therefore Socrates is mortal' is valid only if the premisses 'all men are mortal' and 'Socrates is a man' are unconditional. Is this correct? If the 'conditions' described in 'All men are mortal' were to change, then 'All men are mortal' would be false. A valid syllogism with a false premiss is, by definition, always valid, and so will continue to provide an explanation of 'Socrates is mortal'. An explanation using one false premiss and one true premiss might seem unsatisfactory; but it is a deductively valid argument and hence it is an explanation. 'Conditions', then, is just a misleading way of discussing 'truth'. When we draw inferences we draw them on the assumption that this or that premiss has this or that truth value. So stability of reference is a red herring. The issue that Popper was trying to get at was not the dependence of valid inferences on the assumed truth values of premisses and conclusion. He was trying to get at the difference between testable and untestable assertions. It is not that untestable assertions are of no interest or value; untestable assertions have many roles to play in thinking about society. However, it is wiser not to mistake them for testable scientific assertions. Overall interpretations of history, for example, or a socially optimistic outlook, can be discussed and argued about, but not researched scientifically (see below under (SSR3». The social sciences, Popper held, had, as a matter of historical fact, been more susceptible to the siren song of prophecy than had the physical sciences. This susceptibility called for special care, that is all. It remains unclear whether Johansson thinks all science utilises prophecy (unconditional predictions), or only social science does.!1 Whatever the case may be, Popper articulated no rule against prophecy. Untestable predictions were already

11. See also Indart (1999) for the conclusion that radicals have to accept that social science claims are untestable.

98

THE REPUBLIC OF SCIENCE

covered by the other rules enjoining and preferring the testable to the untestable. (SSR2) The Rule of Methodological Nominalism [T]he task of science is only to describe how things behave, and [I] suggest that this is to be done by freely introducing new terms wherever necessary, or by redefining old terms wherever convenient while cheerfully neglecting their original meaning. For they [methodological nominalists] regard words merely as useful instruments of description (Johansson p. 89, quoting PoH, p. 29; Economica 11(1944):95).1 2

Having selected this as a rule, Johansson has then to spend several paragraphs indicating with what the technical term 'methodological nominalism' is to be contrasted, before he can launch his criticism, which is to the effect that specifying the task of science as descriptive might seem to some to preclude explanation, especially explanations that penetrate below the surface of things. It might seem to do SO, but an examination of the context will show that it does not. An explanation is a description in the sense that the general theory or theories and the initial conditions from which it is deduced are descriptive statements. 'All men are mortal' describes men (or humanity), 'Socrates is a man' describes Socrates, and together the two explain 'Socrates is mortal'. 13 Johansson goes on in his discussion of (SSR2) to argue that Popper's doctrine of methodological individualism (see below) 12. The older text reads as follows: "methodological nominalism holds that the task of science is only to describe the behaviour of phenomena, and suggests that this is to be done by freely introducing new terms wherever necessary, and by re-defining old terms wherever convenient, to the utter neglect of their original meaning, words being merely regarded as useful instruments of description" . 13. Popper always held that the simplest, if scarcely satisfactory, explanation of a state of affairs was the description of that state of affairs. The logically simplest explanation/deduction of a statement a is a itself, since 'from a, a follows' is a demonstrable meta-theorem, the inference in question not allowing for a counter-example.

INDIVIDUALS AND INSTITUTIONS

99

cannot be a direct consequence of his methodological nominalism as he claims in the following passage: the task of social theory is to construct and to analyse our sociological models carefully in descriptive or nominalist terms, that it to say, in terms of individuals, their attitudes, expectations, relations, etc. (PoH, p. 136; Economica 11(1944):80).14

The passage is certainly open to misinterpretation. Johansson is not the first careful reader to think that Popper's formulation commits him to oppose supra-individual explanations, that is to say, explanations utilising social institutions or social forces as components (cf. Ruben 1985). Given, however, that Popper utilises such entities in his own explanations it is unlikely he intends to rule them out. Yet some of his formulations lend themselves to interpretation as reductionist, to the doctrine that only individuals are real. True, he writes always of methodological individualism and not of individualism tout court, but the passage above does recommend individuals as the explanatory end point. Popper is recommending that we treat social forces, institutional pressures and such like as entities pro tem, useful for explanation, but explicable in their tum in terms of individuals and their relations. Such social entities have no independent aims, for one thing; and they have no independent moral claims, for another. Their apparent aims are those ascribed to them by individuals; and their moral claims belong only to their individual members or to the sentiments individuals direct towards the social entities. Now it is true that individualism is clearly recommended as methodologically fruitful rather than as metaphysical. Yet the misinterpretation is common to two of the earliest critics Taylor (1957; unsympathetic and apologetic) and Gellner (1956; sympathetic and not apologetic) and many subsequent critics. This matter will be discussed further below, under (SSR4).

14. The opening words are "our task is to analyse" in the 1944 version.

100

THE REPUBLIC OF SCIENCE (SSR3) The Rule of Interpretative Interest .. .introduce a preconceived selective point of view into one's history; that is ... write that history which interests us (PoH, p. 150; Economica 12(1945):85).

Johansson seems to agree with this rule, but worries that it is ambiguous (he does not say why ambiguity is such a defect). He finds two ambiguities in it. The first is whether it implies that there are events independent of our selective experience of them. The second is that Popper, he thinks, writes as if such points of view were not just multiple and arbitrary, but neither true nor false. Historical interpretations are general points of view; they are, in the language of the positivists, metaphysics. Metaphysics is a topic Popper wanted to avoid in The Logic of Scientific Discovery, preferring to transpose its problems to methodology. 15 He was still not comfortable with metaphysics at the time "The Poverty of Historicism" was completed. Yet his whole discussion of historical interpretation, in "The Poverty" and in The Open Society and Its Enemies was a discussion of metaphysical points of view. It is quite unclear to me whether the Popper of this period thought metaphysical points of view were without truth values - especially as he reports his views about the problem of truth having been much affected by Tarski's ideas. When Popper juxtaposed several points of view - that history is the history of class struggle, or of the struggle of races for supremacy, or of the history of religious ideas, or of scientific and industrial progress, or the struggle between the 'open' and the 'closed' society (PoH, p. 151; Economica 12(1945):85-86) - and pointed out how they were all more or less interesting, more or less fruitful, and as such unobjectionable, it does not seem to me that he was implying that these points of view have no truth values. Until there is some way to render them testable the question of deciding 15. However, under "fruitfulness" in the index to LScD one can fmd a number of passages that give it a positive role.

INDNIDUALS AND INSTITUTIONS

101

which points of view are true and which are false should be deferred and the criteria of interest and fruitfulness used instead. When, later in his career, Popper openly discussed the value and criticisability of metaphysics, he allowed that there was always metaphysics in the background of scientific disputes (Popper 1976, §33). In point of fact his pupil Agassi went so far as to suggest that metaphysical disputes were often what energised the search to find scientific and hence testable versions, and thus to show the value, perhaps even the putative truth, of some metaphysical theories (Agassi 1956 and 1975). Whether Popper ever accepted this view and whether, therefore, he would endorse a general rule about adopting a preconceived and selective point of view to guide one's scientific research is unclear. He vacillated about metaphysics too much for us to be sure. What seems to me true is this: Johansson has isolated a rule for history and the social sciences which, did he accept it, the logic of Popper's own position would have led him to generalise and apply to all critical inquiry, namely, make your metaphysical and other presuppositions as explicit as you can. As to the first ambiguity Johansson detects, namely whether Popper implies that there are events which subsist independently of our experience of them, he resolves his own doubt. He shows that Popper is thoroughly Kantian. Johansson says that Popper often writes as though he were a naive empiricist. This charge is, on his own showing, unconvincing. While Popper's language is at times a little misleading, he is entitled to the benefit of his own rule that we work with the strongest and most criticisable version of his views. This favours his manifest Kantianism rather than alleged traces of positivist empiricism in his language. (SSR4) The Rule of Methodological Individualism Although we have encountered this topic before, it serves clarity once more to reiterate the passage Johansson takes as articulating this rule: ... the task of social theory is to construct and to analyse our sociological models carefully in descriptive or nominalist terms, that is to say, in

102

THE REPUBLIC OF SCIENCE terms of individuals, of their attitudes, expectations, relations, etc ... (PoB, p. 136; Economica 12(1945):80).

Johansson's discussion of this rule mainly adverts to the 1950s debate over methodological individualism (O'Neill 1973). Our present concern is whether the passage quoted amounts to a supplementary rule for the constitution of science; whether we should accept Johansson's claim that the rules he discerns are additional to the rules proposed in The Logic of Scientific Discovery. So far, my overall argument has been thal they mostly relate not to the social sciences in general, but to particular trends and tendencies that existed in the social sciences at the time Popper was writing. In the event that other vices were plaguing the social sciences, a completely different set of corollaries to the basic rules of science might recommend themselves. Each of Johansson's four rules is a particularised version of one of the more general rules proposed for science in LScD. Nominalism, interpretation, and individualism are all corollaries of the rules about testability. The Rule of Methodological Nominalism (SSR2) is counsel against essentialist tendencies, the Rule of Historical Interpretation (SSR3) warns against confusing presuppositions with testable hypotheses, and the Rule of Methodological Individualism (SSR4) urges that historicism and holism are untestable. A rule of methodological individualism is a way of compelling us to make our explanations testable. My only concession is that insofar as the Rule of Historical Interpretation (SSR3) is read as enjoining explicit declaration of metaphysical presuppositions it is an addition to the original rules of the republic of science, not special to the vassal republic of social science. Before leaving this topic I want to consider whether two other important points Popper makes about the social sciences are candidates for the status of supplementary rules. One concerns the Oedipus effect, the other is the so-called zero-method and its application, the unintended consequences of actions. Johansson does not discuss either. Under the rubric of the Oedipus Effect, Popper refers to that feature of social things as opposed to other natural things which we know as awareness. In social life knowing that something is

INDNIDUALS AND INSTITUTIONS

103

anticipated can cause realisation of the anticipated state of affairs. The name refers to King Oedipus who, so appalled by the prophecy that he would murder his father and marry his mother, journeyed far and away until he lost touch with his parents and, for that very reason of their mutual unrecognisability, it was possible for the terrible prophecy to be fulfilled. Popper does not articulate this difference between the world of atoms (which have no awareness) and the world of human individuals (who do) in a rule. It is more of the nature of a discovery about the world; it is part of the facts that make human social life different from the rest of nature. One way it could be incorporated into a rule would be as a contrast to that naive form of behaviourism which attempts to treat human beings totally naturalistically, as mere stimulus-response mechanisms. As with the four putative rules Johansson isolated, such a rule would be a response to a perhaps temporary situation in the social sciences (the ascendancy of behaviourism), so not warranting a position in the basic set of rules. What Popper calls the zero-method is the idea, which he borrows from economics, that most social science explanation proceeds by constructing a typified and rational actor, deducing her actions in a model of the situation, then comparing that to what real actors do in real situations. The closer the predictions of the model come to the way things tum out, the better it explains. The typical buyer and seller, for example, act rationally in a perfect market model, when each is able to maximise. The buyer obtains the maximum possible for his dollar, the seller obtains the maximum dollar for her goods. This is the zero situation of fully informed and fully rational buyer and seller, operating in a perfect market. Real markets, real buyers and sellers, do not reach the theoretical maximums of the model. Social science hunts for explanation of those discrepancies by looking for market imperfections, for buyer and seller imperfections, and for unintended consequences. So social science explanation is less the search for laws and more the search for initial conditions that permit explanation of social phenomena that diverge from the zero coordinate. "All actors act rationally other things being equal", is a zero coordi-

104

THE REPUBLIC OF SCIENCE

nate against which we plot the actions of real actors. It is false, but we use it as a measure of the complicating factors that need to be introduced if action that is less than rational is to be explained. The zero method is what makes social events explicable. Its 'zero' name is intended to indicate its uninteresting or trivial character. We resort to the search for other, less than optimally rational, explanation because pure 'rational explanation' fails to explain real outcomes. Most commonly we argue that the zero method does not predict the outcome because what needs explanation are unintended consequences of rational action. Thus the zero method will not in itself explain the appearance of a path through the jungle, since the rationality of a single actor is insufficient to explain it. However, the explanation of why multiple typical actors acting rationally take the way through the jungle they do (for example it is the shortest, or the most practical, route) will in the end explain the path because a path will appear as an unintended result where many feet have trodden. If a typified and reiterated zero method explanation is sufficient to account for most of those feet and hence for the path, even though it was intended by no-one, then it is explained. If no general explanation can be given of why the feet trod the earth flat, for example that it was random, then we have no explanation of the appearance of the path. My point is that explanation using the zero method does not result in new methodological rules for the social sciences. Such explanations are no more than a variant of the methodological injunction to search for explanation of any event that can be described (R3). That human beings act, and can be explained as acting rationally, unlike atoms, is not a rule but a finding of social science: it is part of our theory of the world that there are parts of it that can be explained in a much simpler manner than can its inert parts. For these reasons, I disagree with Johansson, and do not find any modifications of the idea of the unity of method in Popper's comments, and still consider it fruitful for the social sciences. Popper's warnings for the social sciences are less new methodological rules than glosses on the original rules developed in LScD. They point out special ways in which the social sciences can evade the injunction to seek falsifiable assertions. They are

INDIVIDUALS AND INSTITUTIONS

105

pro tem in a way the original rules are not. To reiterate: they address, in Popper's terminology, certain trends that he observed in the social scientists he studied. In other periods other trends might emerge, and other pernicious trends would invite different glosses on the original rules. In arguing that Popper was a theorist of society almost before he knew it we have looked, in chapter 2, at the ways in which he conceived of the problem of demarcation as one of making and legislating (social) decisions. Although Popper had a longstanding interest in social and political matters and engaged in discussion and writing about them, nothing of this can be found in LScD (Popper 1976, p. 113; Hacohen 1996, 1999). Not until he wrote "The Poverty of Historicism" did Popper start to work out and put in publishable form his ideas about institutions in general, and the complementary role of individuals. Breaking off work on that long paper to write The Open Society and Its Enemies greatly stimulated his thinking. Some of the concrete issues raised in that book - the wide appeal of historicism, problems of education, planning, the trade cycle, the post-war treatment of Germany - pushed him to work out his ideas still further. Because the early manuscripts are not preserved, the intricate connection between the two works will likely never be fully worked out. The best we can do, given the evidence, is to look at the first emergence into print of explicit ideas on institutions, individuals, and social rationality, then try to work out their bearing on the sociology of LScD. The starting point is Part I of "PoH", which appeared in 1944, about eighteen months before The Open Society and Its Enemies. 3. Social Institutions Given the received view that Popper is an extreme individualist who refuses to admit that individuals are products of their social environment, I shall treat institutions first. The source of this misunderstanding, his methodological individualism, was a methodological and moral doctrine that seemed in conflict with the obvious fact that institutions precede any particular individual. Yet Popper both affirms methodological individualism and holds that the possibility of homo sapiens developing into

106

THE REPUBLIC OF SCIENCE

humanity (whether phylogenetically or ontogenetically) arrives only with the invention of the institution of language. Well into Part II of "The Poverty of Historicism" Popper announces: The term 'social institution' is used here in a very wide sense, to include bodies of a private as well as of a public character. Thus we shall call a business, whether it is a small shop or an insurance company, an institution, and likewise a school, or an "educational system", or a police force, or a Church, or a law court (PoH, p. 65; Economica 11(1944): 122).

It is striking that in this list Popper includes neither "science" as a networked institutional system, nor what are sometimes called the primary institutions of society, the family and language. In a footnote, however, discussing the dispute over whether social institutions are designed or just grown, he cites language as an example of an undesigned institution with an instrumental character. Another example we encountered in modified form earlier (and it is taken from Descartes I6): a simple track turning into a road merely as a result of many people using it. Discussing institutions, Popper proposes a general thesis, "only a minority of social institutions are consciously designed while the vast majority have just 'grown', as the undesigned results of human actions" (loc. cit.). However intuitively appealing, one problem with this thesis is how one could individuate and count social institutions, given Popper's very broad use of the term, so as to differentiate a minority from the vast majority. Furthermore, the claim that "the vast majority have just grown" has controversial implications. It intimates that institutions to serve our purposes appear without needing to be designed and planned. They grow without our articulating aims for them, which makes their rationality mysterious. It also slides over the issue of whether an institution not designed could be improved by design.

16. This attribution to Descartes (probably intended is the passage in 1637, p. 118) is introduced into PoH only in the 1957 book version.

INDIVIDUALS AND INSTITUTIONS

107

Popper gives extensive consideration to the problems of designing institutions, and of planning generally, yet he seldom focuses on the deficiencies of undesigned institutions. He notes that markets need a legal framework, for example, but does not highlight how that is to repair a serious deficiency in the institution as grown, for certainly markets preceded law. 17 The bearing of these questions on science will be taken up below. Popper's argument is not offered in a purely theoretical spirit. He presents his ideas about the design of institutions, namely, social planning, as an alternative to the planning he calls 'utopian'. Calling it 'utopian' emphasises the good will of the utopians and their lack of responsibility. He advocates instead what he calls "the piecemeal approach to social engineering" (PoH, p. 64; Economica 11:122). Against large-scale and utopian planners who want radically to reconstruct society, he urges that tinkering with institutions and the functions they serve needs to be done in a gradual and modest way. This conservative position is not based on denial that there may be much wrong with society that needs putting right. It is based rather on the obvious point that we do not want to make matters worse while trying to put them right. The danger comes not from our good will, or our radicalism, but from our ignorance. We do not know enough to foresee all the consequences of our actions and hence we would be well advised to proceed cautiously. The vast majority of the social institutions with which we shall interfere have just grown and so we are unaware of the degree of our dependence upon them, or of them upon one another. In the process of changing them our knowledge will grow as a result of our mistakes. Social mistakes have human costs, and these should be minimised. Scale is not the issue: revisability or the possibility of being revised is. Any measure which we defend by demanding ever more change, rather than revising and rethinking, is utopian. For this reason Popper proposes to impose an overall metarule on social planners and reformers that is derived not from the nature of society but from general epistemological considerations: 17. Popper discusses these problems somewhat further in a contemporaneous letter to Hayek, KRP to Hayek, 15 March 1944, PP Hoover 305/13.

108

THE REPUBLIC OF SCIENCE

our ignorance and our fallibility. The criticism of his ideas on piecemeal engineering as socially or as politically conservative may be correct about their practical effect, but the accusation offers no answer to the epistemological claim on which he rests his case. Popper thinks our knowledge of society rarely matches the ambition of our goals. In so arguing he does not endorse the status quo, but he does warn us against those attempts to alter the status quo that succeed only in making matters worse. Furthermore, if the status quo is liberal to any extent, then a 'conservative' defence of it is an affirmation of liberalism rather than of conservatism. A further argument goes to the same point. Not only our ignorance, but also the sheer difficulty of solving social engineering problems should also give planners and reformers pause. Consider the two kinds of social institutions distinguished earlier: those which arise as unintended consequences of rational actions directed to other ends (the road emerging out of the track), and those which we consciously design and create to serve a purpose. Only a minority are in the latter category. We might add that some designed institutions serve not only the purpose for which they were set up, but also perhaps others, and, in extreme cases, they serve only purposes acquired post hoc and not as their original remit. Examples would be business firms set up to make and market one product, but which over the years end up making something else, or not making anything at all. Other examples are laws passed for one purpose and retained for another. When abolition of the law that criminalised homosexual acts in private was debated in the United Kingdom it was argued that abolition would send the wrong signal, i.e. that society regarded such acts as unobjectionable. Another example that intrigued me was the trade union magazine of Canadian Customs Officers, which argued for retention of Canada's obscenity laws on the grounds that they provided employment to the Customs Service. 18 Yet it is clear that the criminalisation of homosexual acts in private was not originally an exercise in symbolic politics, and the intent of those who criminalised obscenity was not to provide public employ18. See the 1985 leaflet issued by the Public Service Alliance of Canada.

INDIVIDUALS AND INSTITUTIONS

109

ment. Each of these was an unintended (predictable and unavoidable) consequence of the original purposes embodied in legislative action. This illustrates a crucial point. According to Popper we cannot construct foolproof institutions (PoH, p. 66; Economica 11(1944): 123). The problems of quality of personnel and the need to cope with changing circumstances ensure that constant maintenance is required. It is for this reason that Popper allows himself the admittedly objectionable term "social engineering" .19 He emphasises that our fallibility imposes on us a responsibility to proceed with care and caution. Critics concerned about the conservatism of his conclusion have also argued that it is hard to implement: how to differentiate where piecemeal tinkering ends and wholesale alteration begins? Popper admits this difficulty. One obvious difference between the piecemeal and the wholesale is that the piecemeal engineer considers possible unintended consequences of any change carefully and in advance of the change, trying to devise criteria whereby the success or failure of the change can be estimated. Not only radicalists are holists. Such grandiose schemes as the war on crime, or the war on drugs, are holistic in hot being packaged with criteria of when the war is considered to be won (or lost) and what to do next in either case. Some countries deal with rising crime incidence in general, and rising drug-related crimes in particular, by creating special agencies, building more prisons, imposing draconian sentences, and so on. Such measures cannot but be disheartening to any critically-minded social 19. This term was bitterly criticised by a disciple of Wittgenstein, Rush Rhees, in "Social Engineering" (1947). Popper wrote a lengthy reply which G. E. Moore turned down for Mind with a remark to the effect that Popper was unfair to Rhees in various places, and more unfair to Plato than Rhees was to Popper. "These seem to me to be cases in which you have treated Plato even worse than you complain that Rhees has treated you." Moore to KRP, 11 November 1948, PP Hoover 28/11. When I was assisting Popper with the fmal arrangement of the pieces in Conjectures and Refutations this reply, entitled "Social Institutions and Personal Responsibility" was revised for inclusion, but eliminated in the fmal cut. It can be found in PP Hoover 36/22.

110

THE REPUBLIC OF SCIENCE

engineer. Twenty-five years of piling on such measures has not 'solved' these problems. There was obvious failure to think carefully and in advance about what is problematic about the problem. For example, just what makes a particular per capita rate of drug crime problematic? When a sanction is proposed, say, re-imposition of the death penalty, or mandatory prison sentences, by what criteria shall we judge whether it is a success and, most important, if by our criteria a measure is not a success, what steps have we taken to ensure it is withdrawn? Such quite obvious and common sense considerations are alltoo seldom heard in our public debates. Radical proposals from both the left and the right are put forward and implemented free of any demand for assessment of unintended consequences and criteria for success or failure. This shows that the view of society and social institutions that Popper was advancing had implications for our political practice; our political masters show few signs of having absorbed them. There are many reasons, of course, and muddled utopian thinking is only one. No-one seriously maintains that we can create a crime-free society, even though political rhetoric sometimes intimates that this will be delivered. More interesting is the repeated failure to build mechanisms that will assess the success or failure of initiatives and alter or rescind the measures. When Popper was writing "PoH", calls for large scale social planning came from the left. But his epistemological and moral arguments were about the blunders and irresponsibility of radicalism as such. The last quarter of the twentieth century has seen radical social change proposed by the political right. Popper's critique hits it equally hard. The radicalism shows that the force of his argument has not been appreciated - either by the left (who accuse him of conservatism in practice) or by the right (who try to appropriate him). Popper's epistemological and moral arguments for piecemeal social engineering remain salient, whether we are dealing with grown institutions or designed ones. Those that are grown, such as the family, are nevertheless subject to our interventions. No-one knows what the primeval 'grown' family system was, but we can observe in our present world that some societies are sustained mainly by the nuclear family and

INDIVIDUALS AND INSTITUTIONS

111

others by the extended family. Such family structures clearly antedate laws created to govern marriage and inheritance. Created laws obviously help to shape and consolidate the family system any society enjoys, including such measures as housing regulations, child allowances, tax deductions, laws against incest, etc. etc. Just because an institution has grown, does not mean that it may not need or receive intervention. Indeed, earlier I raised the problem of the rationality of grown institutions. One could argue that rationality operates in our choice to sustain them rather than change them. If they are unproblematic then doing nothing can be quite rational. As the saying goes, 'if it ain't broke, don't fix it'. If they are problematic then refusing to try to fix them, treating them as hallowed because old, is not to act rationally. A vivid but subtle example of contrasting practices in these matters is language, as we see with two cases of this grown institution, English and French. Language is a special form of institutional organisation developed by one social species of primate. The complexities of the development and spread of human languages is mind-boggling (Renfrew 1992). The English language, in which this present monograph is written, is a complex case. It is a hybrid of a number of influences both in structure and vocabulary. Indeed it has one of the biggest lexicons of any documented language. It has for better or worse replaced Latin and French as a lingua-franca for much of world business, diplomacy, science, and technology, and will not easily be displaced. (Its ascendency is an unintended consequence of other intentional actions.) It has many branches, dialects, creoles, and pidgins. Since Doctor Johnson's dictionary (1755) there have been numerous attempts to compile its lexicon and to map its rules. In only one area have these informal efforts succeeded: spelling has by and large been stabilised. Spelling reform, by contrast, has had small impact, despite the efforts of Melvil Dewey and George Bernard Shaw. Informal attempts of educational institutions to make certain constructions mandatory, and to create lists of excluded words (viz. slang, jargon, obscenity, etc.) have had only short-term success. The language is often spoken of as vigorous, alive, and constantly changing. Among today's specialists in English the largest party subscribes

112

THE REPUBLIC OF SCIENCE

to the view that a language is what a language does: usage dictates correctness rather than vice versa. Thus English is replete with and constantly adding foreign vocabulary - incorporated with ease. No official body oversees any of this. Contrast that with the situation in France, where a French Revolutionary creation, the Academie Fran~aise, has been handed the task of guarding and preserving of the Frenchness of French.20 The Academie undertook but has yet to complete a ninth edition 'official' dictionary of French, which at this writing has reached the letter E. The proposed dictionary is not a dictionary of the living French language as it is used; it is a purified and prescriptive dictionary which refuses its sanction to that of which it does not approve. Thus imported vocabulary, for example, is mostly rejected; slang is eschewed. France's Ministry of Education has brought the power of the state even more directly to bear on this grown institution by decreeing that official documents and communications cannot contain words or expressions that are considered foreign or unFrench by its own and the Academie' s standards. So, whereas in English there is a continuing tendency for the written and the spoken to converge, in France those taking it upon themselves to keep the language pure are helping create a situation where it can be predicted that, if the literate elites get their way, there will emerge two languages: a kind of stilted official French, purified of whatever is thought to be extraneous, and the demotic, which will, of course, continue like any other growing and changing living language to absorb from and adapt to the outside. 21 French attempts at social engineering on the 20. Further information is on the web page of the Academie: < www.Academie-Francaise.fr > . 21. The problem of how to specify the Frenchness of French in order to make operational decisions about what importations of words are to be disapproved and by what principles their replacements are to be constructed, may well be insoluble. The result is bound to be an arbitrary and elitist solution lathered over with rationalisations that appeal to subjective aesthetics and unarticulated chauvinism. The Academie resists the importation of words from English because they do not have Romance roots. No principle is stated, and certainly it

INDIVIDUALS AND INSTITUTIONS

113

language are rationalised as defensive of a state of purity, Frenchness, something that exists of course, only in the past - the past as conceived by present-day elite intellectuals. This is a utopianism of Plato rather than of Robespierre. Its failure is more or less certain; the only uncertain questions relate to whether collateral damage will be done to French education, the economy, the use of French abroad, and so on. Such damage would be the result of good will, high moral purpose, and a misplaced confidence that we know enough about this grown institution so that the benefits of our interventions do not outweigh the costs. I am less sure than Popper that the vast majority of social institutions are grown rather than designed. 22 Perhaps the vast majority began as forms of spontaneously grown order, but family law and French language legislation point to what may be the true majority: mixed cases. Which brings us back to science. In The Logic of Scientific Discovery Popper was discussing a special kind of social institution, science. In its origins, and still, in its broadest sense - namely the world-wide community of scientists - this is an institution that just grew. But the creation of specialties and departments, of laboratories and learned societies, of journals and higher degrees, of internet discussion groups, all of these participating institutions in the grand institution of science were consciously created, usually by the scientists would be false to claim that all words in a modem Romance language have Romance roots. Other difficulties are obvious. Apart from the back-log of importations that cannot but be accepted, and the importations into the living language that will be about as easy to oust by decree as would be trying to oust slang, their own examples hardly strike one as self-evident improvements: "On ne dit plus tie-break mais jeu decisi/, balladeur remplace walkman, logiciel se substitue a software, etc." The authoritative tone scarcely conceals the problem that if the aim is to communicate, a lot more speakers in the world will understand walkman and software than their French equivalent. French speakers will thus need to know the English for purposes of cross-cultural communication and so will be tempted to use it all the time. The case is one of methodological essentialism in action. Its futility speaks to the value of the rule of methodological nominalism. 22. There are difficulties of counting. For Popper, the language as a whole, and every word and every concept, are social institutions.

114

THE REPUBLIC OF SCIENCE

themselves, but also sometimes by commercial firms or by the state. Popper's book was concerned with the most general and abstract aspects of the institutions of science, namely those rules which demarcated science as an activity from all other kinds of activity. The debate over whether the rules are attempts to articulate actual practice (naturalistic), or whether they are new devisings to improve practice (normative), is misconceived. Popper rejected both the naturalistic and the normative approach and opted for methodological conventionalism. The rules offered are those that (arguably) create an institution of a certain sort. To the extent that they already exist and operate, to that extent the institution is already partly realised. Popper's approach, then, is instrumental: if your aim is to employ the method of experience, to engage critically with reality, then impose these rules on yourselves (and better ones, if you can devise them). Ten years later he had new things to say, which opened the possibility of making better sense of science as unintended consequence (grown) that could be improved (design). Popper extends his ideas about the institutions of science in two ways in The Poverty of Historicism. First, in his ideas about "the social or public character" of science (PoH, p. 155; Economica 12(1945):87); and second, in the connection he makes between this individualistic and competitive model and the possibility of progress. In order to identify the conditions for scientific progress he conducts a thought experiment on how its progress might be arrested, and singles out the closing down of research and free thought. All these things which are to be suppressed (or controlled) are social institutions. Language is a social institution without which scientific progress is hardly thinkable, since without it there can be neither science nor a growing and progressive tradition. Writing is a social institution, and so are the organisations for printing and publishing and all the other institutional instruments of scientific method. Scientific method itself has social aspects. Science, and more especially scientific progress, are the results not of isolated efforts but of the free competition of thought. For science needs ever more competition between hypotheses and ever more rigorous tests, and the competing hypotheses need personal representation, as it were: they need advocates, they need a jury, and even a public. This personal representation must be institutionally organised if we wish to ensure that

INDIVIDUALS AND INSTITUTIONS

115

it works. And these institutions have to be paid for, and protected by law. Ultimately, progress depends very largely on political factors; on political institutions that safeguard the freedom of thought: on democracy. (PoH, pp. 154-55; Economica 12(1945):87).

In this rich passage and its continuation Popper thinks socially about science in the course of thinking about the connection between scientific progress and progress in general. His vision is one that makes the very idea of ascribing authority to science absurd. It is just here that he parts ways with most philosophers of science, who just refuse to believe that he is serious. 23 Not only was he serious, he was correctly drawing the implications of his social tum. He stresses that scientific objectivity comes from social institutions rather than from individual scientists. Several of the methodological rules adumbrated in chapter 2 are designed to foster objectivity, although the key requirements are not in those rules. These key requirements are the public character of scientific activity: its openness to test by all-comers - and the challenge posed by any test failure. Popper wrote of tests being of an intersubjective character. Subjects were carrying them out, but their reports were for all to see and assess. In that lay their objectivity, not their authority. In forging this clear link between freedom of thought, progress in science, and social institutions, Popper is treating science normatively. He is viewing it not as an heroic quest by a few determined individuals, but as an institutional undertaking dependent upon all kinds of individuals linked within these institutions. We now may begin to see his rules more clearly as not simply rules for scientific method but as community-forging rules, and, just as important, as progress-fostering rules. A 23. Even his pupil Musgrave is a case in point of the yearning for reasonable authority, preferably inductive: "if bread nourished us when we ate it from Monday through Saturday, then it is reasonable to believe that bread nourishes and to deduce the prediction that it will nourish tomorrow (Sunday)." (Musgrave 1999, p. 324-25). Why is not the following sufficient for a principle of action: the hypothesis that bread will nourish tomorrow has not been refuted so far?

116

THE REPUBLIC OF SCIENCE

community devoted to the critical, impersonal, and intersubjective search for the way the world is, is a very unusual community indeed. Progress is widely taken to be important in science but only Popper, I think, explains why science deteriorates without it.24 When a theory is in conflict with evidence there may be a presumption in favour of the evidence, but that presumption is not absolute. Circumstances can and do arise where scientists may choose to challenge the evidence rather than the theory. Only when we say a theory has been refuted, that is, where a consensus has developed in favour of the evidence rather than the theory, is this option closed. (No option is closed once and for all.) So, only when a theory is refuted can we say that it was scientific, i.e. refutable. For refutability is a property that can be disputed, for example, before a test has been devised. Refutation, by contrast, implies a test was devised and has been decisive: the theory was refutable and is now refuted. Progress, that is, new ideas that are tested and refuted, is what keeps science honest. To put this another way we might want to say that it is the constant stream of tests (including machinery to do the tests) and refutations that gives us some confidence that our system of theories has contact with the empirical world. Without that check we might well be spinning a web of theory that was consistent with the facts, i.e. did not contradict them, yet was at the same time empty or otherwise out of touch. The bridge between science and reality is formed by the social arrangements which govern it. In this novel picture science is not some inexplicable miracle but a piece of social ingenuity. Humans place certain of their intellectual endeavours under collective constraints to keep them honest and to keep them in touch with real things. The institutional constraints foster cooperation and competition which are the joint engines of progress. (Contrast this with the claim that because science is social its results are merely a social construction. Rather like saying that 24 . His reasoning could be used to construct an argument against any fmalised or completed science. Without progress we should not know if our science was about the world or merely a system of coherent conventions.

INDIVIDUALS AND INSTITUTIONS

117

because a construction company is a social institution the buildings it constructs are nothing but social constructions. Bricks and mortar are social constructions, among other things; they are also bricks and mortar. Science is a social construction, among other things; it is also an attempt accurately to describe and to explain the world. Accuracy, description, and explanation are operationalised as social constructions that aim to make contact with the world. See also ch. 6, §4.) The view discussed is developed in passing by Popper as part of general reflections about the nature of the social. He stresses the errors of our forebears, especially Comte and Mill, in thinking of social life as rooted in psychological considerations simply because the unit from which we all start is the individual human being. It seems no more than common sense to say with them that because we start with the individual, social effects must come from concatenations of the individual's capacities, bodily and mental. To the contrary, Popper holds that social arrangements, social institutions, traditions, are emergent properties not reducible to concatenations of individual capacities. Our examples, the social institutions of the family and language greatly change both our bodies and our minds. They enhance and they constrain our capacities. They also have consequences some of which are inimical to our enlightened interests: they both empower us and restrict us. This clinches an oft-repeated point: Popper's methodological individualism is not to be equated with psychologism. At the beginning of his publishing career he gave a social account of science in LScD which downplayed the role of the gifted individual. This role is sometimes called the irrational factor, sometimes the 'gift' of the new idea; but for Popper, a new idea for a scientist is exactly like a poet's new idea for a poem. It is an inspiration. The same inspiration might go either way, to science or to poetry. But if it goes to poetry it enters a different set of social institutions of reception than it does if it enters the social institutions of science. Thus inspiration is pre-scientific. Science is what has happened to inspirations after they have been subjected to a special set of (social) constraints and rules. The

118

THE REPUBLIC OF SCIENCE

specialness of the set of rules is one simple reason why science is not just another institution. A piquant question previously raised should be further explored before we move on. We have seen that Popper's rules for the conduct of science are rules for constituting a social institution or sets of social institutions. To use his terms, is science an institution that has 'just grown' or is it constructed? A closely related question is, when Popper writes of 'social aspects' to scientific method, what are the other-than-social aspects? This question, 'just grown' or constructed might seem redundant, since Popper presents his rules as proposals for discussion, not as simply attempts to capture and codify existing practice, hence just grown and planned are abstract poles, like open and closed. Yet in a publication on the Pre-Socratic origins of scientific method, that appeared just before the English translation of LScD, Popper speculated that a challenge by a teacher (Thales) to his pupils (Anaximander, Anaximines) that they criticise and improve upon his ideas became a practice, then, briefly, a tradition (Popper 1958). The tradition was lost and had to be re-invented in the Renaissance before it could inspirit the newly created institutions of science. Writing Logik der Forschung in the 1930s, when science was largely conducted in universities and industrial laboratories, Popper gave scant hint of this later view that science was the institutional embodiment of a certain critical attitude towards ideas. A critical attitude to ideas is only one among a number of attitudes that can be adopted. If my reading is correct, Popper's first thought was that science was designed rather than grown. Then, in his (1958) he suggested that it grew as an unintended consequence of the discovery of the power of the critical attitude. A partial explanation of the shift is the vagueness of the term 'science'. Science considered as a system of institutions, traditions and persons was not and indeed could not have been designed by anyone.25 This or that institution, however, from the Royal 25. My stress on system is an allusion to Mario Bunge's systemism. Unlike Bunge, I do not think Popper's methodological individualism need conflict with a systemist approach (see Bunge 2000).

INDIVIDUALS AND INSTITUTIONS

119

Society to the Cavendish Laboratories, from the Manhattan Project to the Internet, were designed. The critical attitude underlies both the system and the particular institutions, may indeed precede both. But the fact that there are times and places without science, yet with the critical attitude, suggests that science is more designed than grown, even if it is a product not fully known to its producers. What are the other-than-social aspects of scientific method? Briefly they are logic, truth and falsity, and transcendence. In The Logic of Scientific Discovery Popper emphasised logic and eschewed the notion of truth (because he did not yet know Tarski's rehabilitation of the correspondence theory). It would be anachronistic as far as PoH goes to list truth as a non-social aspect of science, since it developed from a paper first read in 1935. Liberal use of "truth" is made for the first time in the opening pages of volume II of The Open Society and Its Enemies (1945). At the time of writing of "PoH" we are left only with logic. Strong Programme radicals try to connect mathematics, and, implicitly, logic, to social and economic factors. Popper's view, by contrast, is that logic is the unassailable (because tautological) organon of all inquiry. Its results were discovered slowly, and indeed stagnated into a system - Aristotelian logic which was both a barrier to further progress and permeated with important errors. From George Boole onwards there was rapid progress in exposing and overcoming these errors, so that modem logic is one of the towering achievements of philosophy. Such, at least, is Popper's (Russellian) view. The significance of logic for Popper is that it proves its results: they are certainly and finally true. This may be a difficult notion to grasp. On the one hand logic is a system of linguistic conventions, of statements arranged in sequence to yield inferences. Yet the properties of the system of conventions are not themselves mere conventions, they are necessarily true of the system. A parallel property attaches to basic arithmetic. Given the system of conventions we employ, it is necessarily true that 2 x 2 = 4. It is a property of the way we have defined "two", "4", numbers in general, and the operators "x" and "=", that 2 x 2 = 4. If you like, this is a property or implication of the conventions

120

THE REPUBLIC OF SCIENCE

we adopt, an implication that can be demonstrated. Logic is a generalisation of this, being not about numbers or atoms or green cheese, but about the drawing of implication itself. In his "Logic Without Assumptions" (1947) Popper showed how logic could be worked up from the simple notion of statements and statementpreserving transformations of statements. Again, this takes us a little beyond our dates, but it is mentioned here only for clarification, nothing of our argument turns on it. 26 Popper takes the aim of science and social science to be progress, and progress consists primarily in the discovery and overcoming of error. One theoretical system can be said to have made progress over another if it can explain all that the other explained, and if it can explain at least one thing the other cannot explain. The notion of progress here is timeless and without social context except in the most general sense that all science and language-using must have a social context. Popper's notion of progress is transcendental, even if in 1935 he avoided using the ancient word 'truth'. Similarly, the results of logic, the drawing out of implications, is a process with transcendental warrant: it is true in all, and hence independent of any, particular social contexts. Science aims not just to replace myth, prejudice, and folk beliefs with more testable claims. It aims also to teach us how to cope with a continuous scepticism towards unstable and changing claims. Folk beliefs about the world, natural and social, serve other purposes besides explanation, most notably social cohesion 26. Popper expounded his conception of logic in a number of papers which were published during the 1940s, and he clearly intended to write a text book utilising them. Difficulties in the system, which he later repaired, and his being partly anticipated by Gentzen, seem to be at least partly responsible for his plans coming to nothing. If this is so, he gave up too soon, as has been argued by Schroeder-Heister (1984). The tremendous generality of logic does not make it natural. It has conventional aspects, Popper stresses, that may invite further development. Yet its generality and triviality are clearly connected to our fallibility and openness to criticism, for Popper presents logic as the basic apparatus for criticism. The minimality and certitude of logic mean that it can be used to decide very little, so most criticism has to be creative.

INDIVIDUALS AND INSTITUTIONS

121

and identity. Language, location, religious and other ideas and practices, all sustain a group's sense of identity and distinctness from other groups. When such sustaining ideas are changed, whether from the pressures of empirical challenge, from force majeur, or merely from drift, there may be seismic shifts in many parts of the social structure. Fear of such shifts can motivate resistance to noetic change. Substituting a constantly changing body of ideas we call natural and social science for folk beliefs demands different means to achieve social cohesion and identity, ones somehow divorced from the content of ideas. To some, all this sounds very unlikely. How can we transcend our own framework? A powerful argument for Popper's transcendental position - one that might intrigue proponents of the Strong Programme - is the fact that science, mathematics, and logic are all international. Although they originate in different concrete societies, they cross natural, social, and cultural frontiers. The language of science - mathematics and logic - is an international language, even though it is not in all respects a settled one. Logic has many national origins. Chinese, East Indians and Ancient Greeks seem to have discovered much of early logic; modem contributors such as Boole, Russell and Whitehead were English, Meinong, Frege, Carnap and G6del were German, Bolzano and Peano were Italian, Tarski (and others) were Polish, and so on. Much the same story of international effort and even collaboration is to be found in the histories of mathematics and physics. The challenging question is not how to relate Tarski's ideas about truth and the metalanguage to the social conditions of Poland, but how to explain the fact that his ideas made perfect sense to logicians scattered all around the globe in widely diverse societies. Unlike the Strong Programmers, Popper does not debunk science. Indeed, the value and interest of the institution to him is that it produces knowledge that is transcendental: although discovered at particular times and places scientific knowledge is not confined to those times and places. Furthermore, there is no recipe and no guarantee that progress will be made, or even that putative progress will be vindicated. The methodological rules are

122

THE REPUBLIC OF SCIENCE

a social mechanism to help keep the work of science honest; they are no guarantee of honesty, still less of success. If science is a special institution, partly grown and partly designed, obedient to logic, truth must be an implicit aim of the 'just grown' part that can become explicit in the designed parts. Truth as an aim, then, may itself be explained as an unintended consequence of the developing of social institutions of inquiry. It is a regulatory idea and necessarily somewhat vague, since all that we are enjoined to do is to treat contradiction as a problem and never to regard any solution as fmal. These simple requirements are great discoveries. While cooperative social endeavour certainly antedates them, they make certain sorts of cooperative endeavour most efficient. One argument to which advocates of the Strong Programme resort is that from interests: mathematics or even truth are merely aspects of social class. Interest explanations are presented as scientific explanations, that is, deductions from general laws. They are offered as contributions to knowledge, confutations of idealist or intemalist ways of thinking. They offer countercommon-sense social explanation of appeals to 'proof or 'truth'. Taken seriously, such explanations from interest can explain how science trumps earlier views by rhetoric and persuasion. As putative social science the Strong Programme is successful only if its ideas dominate and are accepted. Since neither domination nor acceptance is the case, indeed it is and remains the view of a tiny minority, it follows that its status as science is questionable. Society at large and science endorse either the instrumental or the transcendental view of science (or a combination). Any claim to explain this as false consciousness is to be taken seriously only if it becomes the consensus or the view of the powerful. It has not. These brief remarks expose deep problems in any view that tries to dispense with or explain socially matters of truth and logic. Ideas being the expression of interests, their inception, discussion, triumph or defeat all become aspects of power. Explanations relying on power have a problem with explaining how power shifts to newcomers or the powerless, or does not. Popper's institutional view escapes this problem by envisioning science as a constitutional republic in which revisions to the

INDIVIDUALS AND INSTITUTIONS

123

constitution are infrequent and barriered, whilst under the constitution there is intellectual revolution in permanence. There are checks to and balances of power. For further discussion see ch. 4, §3, and ch. 6, §4.

4. Individuals

So far institutions have been stressed. Scientific research is carried out, however, by individuals: "A scientist, whether theorist or experimenter, puts forward statements, or systems of statements, and tests them step by step" (Popper 1959, p. 27). We have here the individual scientist, her statements of ideas, and her step by step procedure. We shall now look at Popper's developing thought about individuals, and on how they proceed their rationality (§4). "The Poverty of Historicism" advocates "methodological individualism". Previously we argued that this does not preclude social wholes; now we look at the other aspect, the individualism. Individualism, for Popper, is less a matter of ontology and more a matter of moral claims. We have already noted that he accepts Marx's dictum about the priority of social existence over consciousness. Historically speaking, the question which of the two came first is easy to settle. Organisms that reproduce sexually are dependent upon both sexes and so cannot survive as solitaries. It follows that such organisms in their present form cannot ever have been solitaries. So individuality emerges within, rather than aggregates into, society. No other priorities follow from this historical priority, as the former constitute independent questions. Given that human beings have only one life to expend, should we judge individual interests always paramount, never paramount, or somewhere in between? Soldier ants readily sacrifice themselves, sometimes in great numbers, to protect the nest. Similar behaviour can be seen in humans. Does this imply that the equivalent of the nest - the home, the clan, the tribe, the society, the ethnos, the nation, even humanity as a whole, has a moral claim that overrides the individual interest in staying alive? In politics this often comes down to hard choices between, for example, the civil liberties of the subject and the decision to enact conscription; i. e. the will of the majority being imposed upon the

124

THE REPUBLIC OF SCIENCE

minority. Such imposition is always claimed to serve the 'higher' social interest. In The Open Society and Its Enemies Popper accepts the priority of the social as a fact; yet when it comes to morality and politics he puts individual responsibility and rights above those of the collective. But in "PoH", as a proto-social scientist pondering rules of method for investigating and explaining social phenomena, he faces the question of whether to seek explanation at the level of the social whole - of which institutions are a part without which human life is impossible to imagine; or, in the self-conscious individuals, with their aims, values, and capacities to act. Some historians and philosophers have reached further than the social whole, to grand notions of the spirit of the age, the march of time, the path of history, the forces of production, the imperatives of the culture, the character of the nation, in order to explain wars, social progress, and other social events. Such historical holism and social holism Popper criticises primarily on methodological grounds. Against all holistic claims, Popper argues cogently that they are untestable, vague, lead to no programme of research, and fail to stimulate new ideas. These are damning objections. Furthermore, he analyses the wholes to show that they are abstractions that play a role not as forces shaping history and society, but as mental conceptions held by individuals, constructs that play a role in expectations because people believe in them and act upon them.27 If you believe you are nothing and your 'race' is everything, and if you are convinced that the race is threatened, then you may well conclude that it is your duty to sacrifice yourself in the racial cause. If humin beings think certain wholes are real, then, in explaining their actions, we must treat that holistic thinking as part of reality, as a social fact. Thus is illuminated an immediate difference between social explanation and natural explanation: there is nothing corresponding to such 27. Popper did not cite the influential American sociologist W. I. Thomas' situational analysis (1928) or his slogan "If men defme their situations as real, they are real in their consequences", though this precedes and encapsulates his own view. See Jarvie 1972b.

INDIVIDUALS AND INSTITUTIONS

125

facts as self-conception in explanations of nature. Atoms behave according to laws and not according to any conceptions they have of themselves or of wholes to which they belong. The actuality of institutions and of self-conceptions by agents means that we can do better than modelling society as a mere concatenation of individuals bumping around at will. Life in society can be thought of as a series of overlapping social situations which dictate much of the logic of individual actions. These social situations have as their elements the following: individuals, their physical surroundings, their social surroundings, including other individuals, institutions, traditions, groups, cultures, and so on, as well as the conceptions those individuals have of one another and of their surroundings. This sounds complicated, but works quite simply in practice. My presumption that you are untrustworthy, say, is a factor in the situation between us, whether my belief is true or not. Because of that belief I may act in ways substantially different from how I would have acted had I some different estimate of you. Similarly, if I believe something about the situation in general: suppose we are in a desert and I believe there is an oasis over that hill, then I will lead the march towards it. If I believe we are lost I will not. If I believe people trust me and will follow wherever I lead I may reflect carefully on any lead I choose to give. If there is no oasis over the hill we all may learn something about the leader or, again, we may fail to do so. So, contrary to the individualist tradition from Descartes through Hume to Weber, the individual in Popper's model is not an impermeable atom, a given. An individual is the usual conglomerate of mind and body, with a character, personality traits, propensities, and so on, all of which have been developed under the auspices of society. The individual also has internal maps (beliefs) about herself and about her social and physical surroundings and the other people in them, and it is in the total situation, as it is and as it is believed to be, that the individual finds the room to act. Actions once taken not only alter the total situation for the agent and others, they may alter the agent, for experience can shape our appreciation of the situation and can change personality traits and character.

126

THE REPUBLIC OF SCIENCE

A serious objection can be raised: by including individual aims and situational maps as part of the situation, is not the whole conception rendered subjectivist? If valid, this would be fatal. In fact, though, an emphasis on aims and internal mapping operations concedes nothing to subjectivism. Popper, like Hayek, believes that unique behaviour is inexplicable, only typical or regular patterns are explicable. The notion of mental maps is intended as a putative coordinating device that explains why many people may react similarly: they model the situation in the same way. An example Popper gives is of a person going from A to B by a meandering and zigzag track (1994, ch. 8). It might be that the goal is to get to B. If so, why the indirect route? This can be made intelligible, given the aim to get to B, if we add that the situation was that of a person crossing a road and jay walking around vehicles, moving and stationary. A parallel example would be one of a vehicle that proceeded from A to B via a zig zag track covering n miles, when the distance from A to B as the crow flies is (n-m). Again, if we simply add that the physical situation was that the vehicle was a sailing ship at sea, powered by wind, we may have a completely satisfactory explanation of the trajectory, while holding on to the assumption that the aim was to go directly from A to B. Roads and traffic, sea and wind, are physical facts that take on social meaning when they constrain action. Underlying these examples is the zero method assumption of rationality (§2, above): that persons or vehicles will, other things being equal, proceed from A to B via the shortest practical route and hence with the least expenditure of energy and time. The examples of the pedestrian and the sailing ship are both obedient to those imperatives. Not that there are no examples of pedestrians and navigators who quite deliberately meander because they have intermediate aims, because they are enjoying themselves, because they are distracted, because they are incompetent at navigation. If action is rational why do such outcomes occur? They are unintended consequences. We would factor these into our account of the situation only if the first and more general explanations from rationality failed factual tests. Why are rational explanations primary? Because adding factors

INDIVIDUALS AND INSTITUTIONS

127

to the zero level make the explanation less general, more idiosyncratic, and certainly more complicated to formulate and test. Attacks on Popper's methodological individualism that impute to him the idea that there are no supra-individual social things, no social institutions, no social forces, and so on, have missed the richness of his conception of the logic of the situation. All of these can be incorporated as views actors take of the situation. What Popper argues is that social scientists can and should try to explain the institutional factors or the forces at work in more individualistic terms, terms that are preferable not only morally and metaphysically, but also because they deliver more satisfying and testable explanations. Of particular explanatory value is the idea of the unintended consequences of actions. Our explanations of the pedestrian and the sailing ship were examples of explanation from intentions. The simplest explanation of social action is that it is the result of an acting individual's direct action to achieve a goal, e.g. walking in a straight line to get from A to B. Intentional explanation treats the individual as a unit possessing aims and capabilities. To realise the aim of being at B the individual walks towards it. To walk from A to B is an action that has more effects than simply realising the individual's aim. The action renders the place A less crowded by one and the place B more crowded by one. The movement consumes oxygen and generates heat energy, as well as slightly wearing away the places where the feet have trodden. If many individuals undertake to go from A to B directly, as we have seen, a track and then a path may be created, although no individual need have 'creating a track' or 'creating a path' amongst their aims. In due course a path so formed may become worn and difficult to use, especially in unfavourable weather. Certainly no user of the path had 'making it impassable' as an aim. Indeed, we may now expect to see the development of alternate routes from A to B just because heavy use has made the direct route difficult to use. Hence an unintended consequence of many decisions to use the direct route may be to make the direct route impassable - a better example of unintended consequences would be hard to fmd. So let us expand on it. People may not

128

THE REPUBLIC OF SCIENCE

intend to create a path, yet welcome its appearance as a convenience. Certainly they do not intend that welcome convenience to become unusable. So in the progression from intentional actions of walking between two points, the walking created a path without anyone's explicit intention, and the further walking made the original route impassable, a consequence not just unintended but against intention. Thus we can see that the results of action force us to acknowledge social aggregates: a track or a path is also a social entity. We may also see that unintended consequences may be foreseeable or unforeseeable, desirable or undesirable. Unintended consequences are sufficient to explain a good deal of the social world. The rise of prices, for example, can be a consequence of many buyers entering the market, or of a fixed number of buyers chasing after a falling supply. By assumption, individual buyers do not want to drive the price up, yet that is just what their behaviour, in the aggregate, does. These unintended consequences do not arise because of the aim of any actor or actors. They arise out of the structure of the situation. The structure is partly made up of other actors, but also of the institutions, groups, traditions, and such like, which themselves structure the relations of social actors, together with the physical, biological, and chemical underlays of all this. The track, path, and impassable path, after all, are created by physical and chemical forces instigated by feet. Feet wear away plants, then soil, then sun and rain and frost make ground conditions adverse for travel. So we must always remember that the situation is physical as well as social, and that just as purely physical initiatives, such as natural disasters, can have immense and obviously unintended social consequences (since human agency is not their cause), so too can social actions eventually have unintended (as well as intended) physical consequences. If Popper does take social wholes seriously is he not then a collectivist? No: he allows no blind social forces, no collective destiny. Institutions and all other wholes operate through individual actions. Methodological individualism, then, is not a reduction of the reality of social wholes and forces to individuals and their aggregates. It is a recommendation to persist with the

INDIVIDUALS AND INSTITUTIONS

129

quest for explanation until the picture of the social situation is rich enough to make the logic of the typical individual choice, and action under its constraints, transparent. This recommendation is of course of debatable value. Most attempts to debate it do not even get its formulation correct and so end up debating something else, an imaginary opponent or straw figure. The real Popper offers many explanations, especially of the actions of intellectuals, that treat individuals as acting in the interests of their class, their political party, or their upbringing. These factors structure the· way those individuals construe the situation they are in, and hence what they take the logic of it to be. So strong is the prior construction they bring to the situation that it may be able to accommodate uncomfortable or refuting facts. One thinks of the believing Bolsheviks who went to their deaths in the basement of the Lubyanka prison still believing in the Revolution, the Party, and the man who had consigned them to death, Stalin. Of course, it is always possible that in his own explanatory practice Popper inconsistently applied his articulated principles of methodological individualism. I do not know of any evidence to this effect, but were it to tum up it would show a deficiency in Popper as sociologist, not tell against the principle itself. Criticism of the principle itself would have to show that it did not foster good social science and that some other principles would do the job better. In the absence of argument to that effect, Popper's ideas on individuals stand.28 5. Rationality Science, or, rather, doing science, is Popper's paradigm of rationality, of acting under the constraint of ends. Scientists aim 28. One of the very first reviews of the book version of PoH (Taylor 1958) argued as follows: Marx and other targets were not historicists, so historicism is a straw man. Furthermore, Popper's methodological individualism denies the reality of social forces and social wholes thus constituting an obstacle to radical social reform. More sympathetic attempts to explicate Popper's less than fulsome text are Watkins (1952, 1957), who is corrected by Agassi (1960) and Wisdom (1970).

130

THE REPUBLIC OF SCIENCE

to solve problems by advancing theories and testing them step by step. This activity presupposes and depends upon institutions, including a public language and fora in which mutual crosschecking can take place. The challenge for the sociologist is to reconstruct the situation as it is and as it is seen to be. These reconstructions should explain both the rational logic of action, i.e. why this action seemed appropriate to the typical agent, and what typically happened. The discrepancy between aims and results is to be explained by the difference between the situation as the agent saw it and the situation as the sociologist fmds it to be. The scientist who blows up the laboratory did not intend to do so, and clearly misconstrued the situation. Most seemingly irrational actions will be explicable by closer attention to actual aims and/or by closer attention to how agents typically saw the situation. Take the very widespread belief in social conspiracies. This results in actions that seem rational to some agents. These actions have consequences that can be explained by the belief in conspiracies rather than actual conspiracies. Popper's conception of the social should now be coming into focus. There are various kinds of institutions, both grown and designed, which serve social functions that need to be established by arguments independent of origin. There are groups, quasigroups, traditions, and other patterned activities that depend on or participate in institutions. One of the institutions, we may say, is the human individual, or the human person. Humans are everywhere individuated. They have a unique name,29 although 29. Or, to be more precise, they have a name which is capable, by qualification, of designating them uniquely. In many societies names are descriptions, 'X son of Y', or 'X daughter of Z', where the descriptive qualifier is added only to avoid ambiguity or in formal modes of speech. 'Joseph of Arimathrea' might be all right in Jerusalem, but probably not in Arimathrea where several families might have a Joseph, and still less so in written chronicles, where there may be Josephs in each generation. Joseph Kennedy's son Joseph was known as Joe Jr.; other families use roman numerals lId or IIId to ensure unique designation. All naming systems have trouble across time and space, since family names are finite sets, and because institutions and traditions create clusters of naming practices. Readers and viewers of I Claudius will know that

INDIVIDUALS AND INSTITUTIONS

131

naming systems vary greatly. Both individuals and, often enough, their kinsfolk, are treated as loci of explanatory initiative and responsibility. Thus explanation of social phenomena more often than not begins with typical individuals and their actions. However much we are interested in this individual qua individual, it is their typical rather than their idiosyncratic aspects that can serve us in scientific explanation. Any explanation from idiosyncrasies amounts to explaining action by saying, 'that's just they way they are'. Such just-so-story explanations run counter to the principle of methodological individualism. Human action brings about results: actions have consequences. A bare minimum assumption about human actors, then, is that they are aware of (some of) the results and their acting is directed to achieving (some of) them. Joe knows walking directly across the field is the quickest route to B. So, some actions and some social events are explained by their being aimed at by agents. These, however, are not very interesting; that is to say, they seldom pose explanatory problems. My point here is that we commonsensically understand both action directed to an end - going from A (wherever we are) to B (the Polling Station), there to mark our ballot, each of us with his or her aim in acting (as well as refraining from acting, i. e. not voting); and that those actions collectively may bring about results that correspond neither with our aim nor with the aim of any other actor. The view that what happens happens because it is willed by someone, or something, is what we might call the naive or magical view of society. So, it is often said that if we want to understand why something happened we should start by asking 'who benefits?' But this line of thought is misleading in its suggestion that those who benefit can be assumed to have acted to bring about the state of affairs from which they benefit. Furthermore, it is naive because it assumes that those beneficiaries can sufficiently well foresee the consequences of different courses of action so as to act prudently in their own even within the Roman Imperial family there were so many Gauises, Neros, Claudiuses, and so on, that a kinship diagram would sometimes be helpful to avoid confusion.

132

THE REPUBLIC OF SCIENCE

interests. Society is not so simple that there are powerful groups who, unlike the rest of us, have sufficient knowledge to know the outcome of events. If only it were so, life, and sociological explanation, would be much easier. Instead, quite the contrary is commonsensically known to be true, as the election example shows. Time and again politicians achieve or maintain power, less because the voters will it as in their interests, and more because they will something quite different (rejecting what they see as a worse alternative) and are prepared to live with the consequences, whatever they are.30 Postulating the typical acting individual with a typical aim is, again, the "zero method": the base line from which our explanatory thinking begins. We start with it, then see how adequately it explains matters. Only when we have exhausted the possible situations to explain the zigzagging pedestrian do we shift to the possibility that we have here an individual not acting according to the model. Assume for the moment that the zero method is followed: we get no very exciting explanations. 'Why did X (the election, the assassination) occur?', because (enough) people acted to bring it about. This is in a way trivially true. But what if we know that X is the last thing any of them desired? Polluted cities came about because enough people moved to them and did other things to bring it about. But did any of them wish to 30. A vivid example is the continued speculation about the assassination of President Kennedy in 1964. All those who are critical of the lone gunman explanation seek to ask who benefited from the assassination. Thus President Johnson, the Pentagon, the CIA, the FBI, organised crime; the USSR, Cuba, and so on, have all been suspected by one speculator or another. Particularly galling to these speculators is the thought that the lone assassin was a bit deranged and carried out the act with no clearly discernible aim, not even, obviously, the resolution to accept the consequences (he tried to slip away). Such a momentous act, it would seem to many, must have been carried out for a purpose. Yet precisely because it was momentous, neither the lone assassin nor any of the alleged conspirators could possibly have foreseen the outcome. The death of the assassin, the adulation of the fallen president, the vigorous carrying through of his legislative programme, none of these were predictable, and some of them certainly thwarted any putative aims (Posner 1993).

INDIVIDUALS AND INSTITUTIONS

133

live with stinging eyes, headaches, and air-borne chemicals that make them and their children ill? It is, to say the least, doubtful. Ah, say the conspiracy theorists, you should ask, 'who benefits?'; ask, 'how much profit can be made from motor cars and factories that pollute?'. But this assumes that the profitmakers aimed at corrupting their own lungs and those of their children for profit. This is a somewhat Dickensian view of the captains of industry, no doubt to be verified by pointing out how they do not live near their own pollutants. Yet, other explanations are available, for example, the explanation, again, from unintended consequences. I think there is no doubt that Faraday, Edison, and others, were of the view that electric current is, safe inside its wires, completely benign. Latterly, however, arguments have been made that high tension wires and substations have radiation patterns that are unhealthy, especially for the young. In this case the question of, 'who benefits?', can be considered more rationally. The electrical utilities, much like the tobacco companies, have a vested interest in not accepting the scientific claims: yet they also have a vested interest in not ending up like the Johns Mandeville Company or Union Carbide, ruined by liability suits. Note that the zero method does not assert that human beings act rationally. This would be manifestly, empirically, untrue. Popper wrote: in most social situations, if not in all, there is an element of rationality. Admittedly, human beings hardly ever act quite rationally (i. e. as they would if they could make the optimal use of all available information for the attainment of whatever ends they may have), but they act, none the less, more or less rationally (PoH, p. 140; Economica 12(1945):82).

Popper's claim is that there is an element of rationality in social situations because human beings act more or less rationally, and that this makes the task of explanation in the social sciences less complicated than in physics. Our initial explanatory modelling efforts in the social sciences can use the zero method as a first approximation. In addition to Popper's claim about human rationality being qualified, it is also obvious that an assumption, an explanatory postulate, is not thereby declared to be true.

134

THE REPUBLIC OF SCIENCE

Indeed, the very idea that the zero method provides a first approximation based on perfect rationality, which will then have to be modified to bring it in line with reality, shows that no strong empirical assertion is made about humans acting rationality. A more interesting point is that we have little choice but to impute rationality to human beings, because our notion of social explanation is tied up with intelligible, that is, rational, action. If we assume that human beings are, in general, irrational, then we abandon hope of explaining what they do. For to attribute irrationality is to allow that anything could happen. We do this when we are stumped: 'she must be mad', we say. This is less a clinical opinion than a cry of explanatory despair. In the popular understanding, with a mad person, anything goes. In this respect, the assumption of human rationality plays a similar role in the social sciences to that which the assumption of law-like behaviour in the inanimate physical world plays in the physical sciences. In both cases Popper's counsel is nm to enter the empirical/metaphysical dispute about the truth of the assumption, but rather to propose methodological rules (SSR4 and R3). Indeed, my view would be that methodological individualism is simply a consequence of (R3). No empirical assumption of human rationality is necessary for methodological individualism. Rationality is an explanatory assumption, a minimum condition for the task of social explanation. For the social thinker the interesting problems are those that have no rational explanation because they are unintended, or because the actors are acting less rationally than the situation and their knowledge would seem to allow. The first is meat for sociology proper; the latter, for psychopathology and deviance. Let us come back now to science. The scientist aims to solve problems. To this end she advances hypotheses and tests them step by step. The problem is grasped as part of a situation, a background knowledge. One kind of solution fits with the background knowledge. Sometimes, though, repeated failures can lead to questioning the background, the very terms in which the problem is posed. Galileo did this to the Aristotelian background;

INDIVIDUALS AND INSTITUTIONS

135

Einstein to the Newtonian background. The resultant revolutions can be explained without seeing them as aims. We can use the logic of the problem situation instead. The aim was to solve a problem. The problem seemed intractable. Its solution might lie in the background itself, which can be problematised. Originally, we might say, no scientific revolution was aimed at, - it was an unintended consequence of persisting in the search for a solution. Once the broader challenge is thought of, however, promoting a revolution may become an aim. This is precisely because of the depth of grasp of the background, for it is apparent that if something is awry about this, then it follows that changes have to be made in this, this, and this. Science itself, I suggested earlier, is a grown institution. The previous paragraph sketches how it could have happened. Persistent inquirers could not intend to build science when they could have no conception of such a thing. However, their persistent efforts and their collaboration could have created science, like a track through the forest. Just as the track may need maintenance or even paving over, science needed designed institutions to continue and foster its growth. So in "The Poverty of Historicism" we see Popper elaborating a conception of the social world that comprises persons, institutions, traditions, and other larger units, in a social and physical surrounding. Humans have projects which when pursued in aggregate always have unintended consequences, whether or not these are foreseen. Those not intended and not foreseen pose a challenge to the explanatory ingenuity of sociologists. Only by tracing them back to the patterns of action which cause and maintain them can humans exert some kind of control over their lives together. False appeals to psychology, like false appeals to conspiracy, simply lead us nowhere: we cannot give everyone psychotherapy (and anyway it does not always work) and we cannot stop conspiracies, they always proliferate. But conspirators in the grip of naive magical thinking are unlikely to be effective, since their conspiratorial actions miscalculate what is necessary to achieve their goals. True, as a minor point, the existence of conspiracies directed to certain ends will have real unintended consequences. We have only to think of

136

THE REPUBLIC OF SCIENCE

open conspiracies, such as the lobbying system. Lobbying in the interest of an industry or group may seem to serve that group but also may yield unintended and unwanted consequences - such as legislation against lobbying. The conception of society outlined is a wholly pragmatic one. Popper does not depict society as something deserving of love and affection. 31 He sees it in mechanical, even engineering, terms. Society and its components are there to serve human ends, and where they fail or thwart those ends they may need tinkering with. Nothing in Popper of sovereignty, of citizenship, of legitimacy, of society as the emotional mother to whom we owe all. Or is that too strong? Popper has made it clear that science is a social activity, scientific method is a set of social arrangements, science is a community which submits to those arrangements. Science, for Popper, is mankind's greatest spiritual movement (OS&IE, II, pp. 270-711283). He may reach for an engineering metaphor when discussing society as a whole, and he may propose rules for science like a mechanic doing a tune up, but the end is not part of the machine; it is a human purpose. Life in society permits the growth of science and the growth science is a luxurious adventure. Both the social institutions of science and the quest for knowledge in general arouse passions, offer identities, and designate opponents. For Popper does not confuse science with technology, although for certain purposes he does not distinguish them. Both social science and natural science have technological problems and solutions as well as a 'pure' aspect. Interestingly, in the social sciences Popper is suspicious of the pure motive of inquiry. He says that pure social science can easily degenerate into obfuscating and fruitless metaphysics. He does not say why he thinks only the social sciences seem to bear this affliction. As a consequence, he advocates a more technological approach to society and its problems than he does to contemplating the problems of the physical world. He refers to science as a great spiritual movement, but I do not know if he intended to include the social sciences in this encomium. 31. Although in Unended Quest (1976) he writes with affection of the societies he encountered in Britain, New Zealand, and the United States.

INDIVIDUALS AND INSTITUTIONS

137

One argument for encouraging pure social science is that social problems involve suffering and injustice which are more urgent than thinking up clever conceptions of society. This argument will not go through, because the physical world is also a cause of pain and death. Popper finds no problem in encouraging us to engage in such pure speculations as on the origins of the universe which yield no apparent benefit in the alleviation of pain or suffering. If social pain is urgent then so is physical pain. So a better argument is needed against pure social science and for technological social science. We will fmd in studying The Open Society and Its Enemies that Popper's sociology of the physical sciences is both brilliant and idealised, in short, beautified. One of the ways science is beautified is that it is explicitly linked to political liberalism, that is, to regimes which care about pain and suffering. In reading his social scientist contemporaries, Popper apparently found not just totalitarian ideas and policies, but methods of doing social science that were illiberal and dogmatic. This is a better explanation of his discriminatory recommendations between the social and the natural sciences. His beautification of science is sociologically false. Furthermore, according to Popper's own social and political thinking, watchfulness not beautification should be our attitude towards all powerful persons and institutions.

Chapter 4 AN ENRICHED VIEW OF INSTITUTIONS The spirit of science is that of Socrates. - -Popper (OS&IE, II, p. 2311244)

The Logic of Scientific Discovery and "The Poverty of Historicism" were methodological works, reflections on the problems involved in obtaining scientific knowledge: the first, on natural scientific knowledge, the second, on social scientific knowledge. What neither discussed was actual societies and actual social institutions. Those topics were taken up in The Open Society and Its Enemies (1945). It is there that Popper begins to work out a view of civil society that treats science and the scientific outlook as a model for individuals and institutions. It is there that Popper works out his ideas on institutions, traditions, reform, design, freedom, and the democratic control of power. His range is wide, examining the society of Ancient Greece (Athens and Sparta), the Age of Nationalism in Central Europe, the nineteenth century social and political upheavals in the wake of the industrial revolution, and contemporary social democracy. What he does not do is work out the application of all these ideas for what he calls "the social aspects of scientific method". If the scientific way of life is to be held up as a model for society in general, then the facts and deficiencies of that life are worth examining. We shall find, I think, that Popper's general social thinking led to strong criticisms of science in its social aspects, although he did not point this out. In my opinion, the institutional deficiencies of science show that we need to be wary of it as a model for social life. Popper tells us that while rewriting "The Poverty" for journal submission he elaborated certain points, especially on essentialism, partly in response to the puzzlement of test readers, and that this elaboration got out of hand (Popper 1976, p. 114). He cut the elaborations away to make a separate work, and the eventual result was The Open Society and Its Enemies of 1945.

INSTITUTIONS ENRICHED

139

He describes it as a "truly unintended consequence" of the endeavour of writing "The Poverty", although it is clear that, once embarked upon the project, he was aware that it was of considerable, and quite independent, importance. When The Open Society and Its Enemies reached the reading public in late 1945, the 'unintended consequence' had become a work in two volumes, totalling some 614 pages. (The first American edition of 1950 was in one thick volume.) Of these, the text of volume I was 177 pages, the notes a further 91 pages; the text of volume II was 267 pages, and the notes 79 pages. The notes being in smaller type, we can calculate there was almost as much reading in them as in the text proper. The notes contain the usual reference material, but also many scholarly elaborations of points made briefly in the text and miniature studies of all sorts of matters. Substantial changes and additions were made for the second (1952), third (1957) and fourth (1962) editions, so that, by 1962, the text of volume I had grown from 177 to 201 pages, the notes from 91 to 150 pages; the text of volume II from 267 to 280 pages, the notes from 79 to 87 pages. In the later editions Popper writes that he has marked some of the new notes with asterisks, but not all of them (OS&IE, 4th edition 1962, p. 202). The asterisks are misleading, as they may suggest that all that was added to new editions is marked. Not so. Not even such partial guides are provided for additions to the text proper. I will point out additions to the first edition that I consider significant for the reason that most readers refer of necessity to the fourth and later editions, the first to be issued in paperback. Once again, then, it is important to specify what will be discussed. My interpretative reading will focus on the first British edition of 1945. This has the elegant result that the present book covers a ten-year period of Popper's output. The Logic of Scientific Discovery was published in late 1934 as Logik der Forschung, and the fmal revisions and insertions to The Open Society and Its Enemies were completed in 1944. Between the ages of 32 and 42, then, Popper sustained a phenomenal and

140

THE REPUBLIC OF SCIENCE

closely integrated intellectual effort. 1 In order to concentrate on his social thought, I trace his trajectory from methodology abstractly presented as social in 1934, through the explicit acknowledgement that scientific method has social aspects in "The Poverty of Historicism" of 1944, to a fully developed conception of knowledge as an institution and institutions as knowledge, and a society of institutions as a knowledgeaccumulating entity, in The Open Society.

1. The Story The Open Society Tells Before focusing on the crucial questions of institutions, individuals, and rationality, the salient features of Popper's republic of science, let me present an overview of the work in its own terms. Originally entitled "False Prophets: Plato-HegelMarx" ,2 The Open Society was an effort to fathom the political ills that plagued society in the twentieth century, and especially those responsible for the two world wars. Or, to be more precise, Popper, an intellectual, investigated what intellectuals had contributed to these catastrophic events. Without ignoring the social pathologies of poverty, economic crises, Realpolitik, and sheer barbarism, he sought the historical roots of the animating ideas behind nationalist, racialist, and totalitarian thought. He 1. As noted in ch. 3, text to note 1, in a letter to Hayek (KRP to F.A.Hayek 14 March 1944, PP Hoover 305/13), Popper states of the discussion of Marx in The Open Society that "It is mainly the elaboration of a draft began in 1920 and brought to some kind of fmality as early as 1927!". I have found no trace of this draft in Popper's papers at the Hoover Institution. This is only one of many regrettable gaps in the paper record, some of which may be clarified now that the balance of his papers has reached the public domain. 2. Quoting from his own copies of his 1943-1945 correspondence with Popper, most of which are also in the Hoover collection, Sir Ernst Gombrich reports the preliminary titles: "A Social Philosophy for Everyman" and "A Critique of Political Philosophy", and the first emergence of "The Open Society and Its Enemies" (together with "A Social Philosophy for Our Time") in a letter dated 22 December 1943. See Gombrich 1995. A letter to Alfred Braunthal reveals that the original or working title was "False Prophets Plato-Hegel-Marx", see KRP to Braunthal21 May 1943, PP Hoover 2812.

INSTITUTIONS ENRICHED

141

endeavoured to show how these ideas had gained a degree of intellectual respectability amongst those who should have known better. He tried to tear away any veil of respectability these ideas might gain from their intellectual ancestry, even where that involved him in desecrating ancestors. This programme was not apparent from the opening of the book, which begins (as well as ends) on the question of the meaning of history - the meaning of social processes viewed over time. In the first chapter Popper introduces as typically 'historicist' the doctrine of the Chosen People, those destined to carry out God's will. An historicist doctrine in Popper's sense is one which tries to grasp the laws of history and to use them to predict the future. Historicist predictions are a basis for practical action in politics, as seeking laws of history is one way of coping with social change. It is the initial theme of the first volume that social change, especially change that threatens the life style or status of those privileged enough to be intellectuals (those who write the texts that have come down to us), is a source of corroding fear. First Heraclitus and then Plato are seen as possessed by that fear. In the second volume Aristotle, Hegel, and Marx are similarly presented as concerned with social change and the need to understand and predict it in order to exercise some control over it. Historicist doctrines, which often masquerade as rational philosophy (in Plato) or even as social science (in Marx), are exposed and attacked as little more than unscientific prophecy. The prophetic and scientific elements in the thought of Plato and Marx are analytically separated, and both are shown to have been brilliant and deep social scientists horribly misled by a proclivity to mix prophecy with science. As a result, the intellectual heirs of Plato and Marx were able to marshal their authority behind harmful ideas and policies, and the class of intellectuals abdicated its responsibility to show that some of the greatest intellectual leaders of mankind also made grave mistakes. At the conclusion of the second volume Popper returns to the philosophy of history and to the fear of social change. He insists that there is nothing to fear in social change itself, provided that society is not riddled with injustice and privilege. Furthermore, a

142

THE REPUBLIC OF SCIENCE

rational and prudent attitude to social change is possible in which we try to cope with its vagaries while making full allowance for our fallibility. History has no meaning in and of itself. Yet this does not reduce history to the mere succession of one damn thing after another. If we want a meaning in history, Popper holds, then we must put it there; if we want it to tell a story, we should strive to realise that story. So if we want history to be the story of the discovery and advance towards freedom, then it is up to us to discover and to advance freedom: thus we make history take on the meaning we give it. We can see that The Open Society is a companion piece to "The Poverty of Historicism" in that it is about the demarcation between pseudo-social science (historicist prophecy) and genuine social science, including social technology. But whereas "The Poverty" is mainly about methodological similarities and differences between the natural and the social sciences, The Open Society concentrates on the actual practice of social science and its pitfalls. Popper not only condemns the scientific pretensions of grand schemes and prophetic history, he also criticises political historians for their exclusive interest in Great Men and Great Events, which he denigrates as largely the history of great criminals and their great crimes. In addition to elaborating on his ideas concerning institutions, individuals, and situations, which we have outlined in the previous chapter, Popper introduces a number of further ideas in The Open Society. In particular there are his two polar contrasts between the abstract society and the concrete society, and between the closed society and the open society. 3 There are also his doctrines that institutions are hypotheses and concepts are institutions. Finally there is a running theme throughout the work which is that the fundamental social science, if any, is not psychology, as John Stuart Mill thought, nor economics, as Marx may have thought, but sociology (the priority is given to Marx 3. According to Unended Quest (p. 118), the passages about the abstract society were added for the first American edition of 1950, and thus appeared only in the second British edition of 1952. They were implicity in the first edition; labelled and elaborated in the second.

INSTITUTIONS ENRICHED

143

for this idea, even though he is usually associated with 'economism '). My discussion of The Open Society will be spread over two chapters. In the rest of this chapter we shall look at the central and organising distinction between closed and open society, and how institutionalised science functions as a model for social institutions in general or, rather for those that aim at a rational improvement of society. The chapter ends with a discussion of managing social change by reform, reform made modest by recognition of what is natural and must be obeyed and what is conventional and can be altered. The scale of reform, whether piecemeal or utopian social engineering, is related to a sound understanding of the limits of social scientific knowledge. In chapter 5 I discuss the two great utopianist social engineering movements of nationalism and totalitarianism. One problem they pose is that they were espoused by intellectuals, including scientists. To tum them aside Popper has to develop his view of social science and of the social organisation of science. From this effort we can extrapolate his view that science is a model for society: science is the knowledge-gaining institution par excellence. Because all institutional initiatives embody conjectures, the society of institutions embodies our aggregate conjectural attempts to realise our aims. Knowledge is a social institution, and social institutions are our attempts to apply our knowledge. 2. Popper's Ideal Types: Closed and Open, Concrete and Abstract In the Introduction to The Open Society and Its Enemies, Popper contrasts life in tribal society with life in an open society. A tribal or 'closed' society is one submissive to magical forces, as contrasted with an 'open' society which sets free the critical powers of humanity. Popper candidly confesses that an open society which aims at humaneness and reasonableness, at equality and freedom, "is still in its infancy" (OS&IE, I, p. 1) and, indeed, has not yet fully recovered from the shock of its birth. Tribalism, by contrast, he seems to think of as the ancestral form of human society. He writes of it as emphasising "the supreme

144

THE REPUBLIC OF SCIENCE

importance of the tribe without which the individual is nothing at all" (OS&IE, I, p. 7/9), i.e. a kind of collectivism. In chapter 10 of the book, eponymously entitled "The Open Society and Its Enemies", a chapter he was tempted to place at the beginning (OS&IE, I, p. 4/5), Popper elaborates. The Greeks, he says, made the first step from tribalism towards openness and humanitarianism. The early Greek tribes were not unlike, he conjectures, Polynesian peoples such as the Maoris (we recall that Popper was writing this in New Zealand). He does not deny that there are differences: There is no standardised 'tribal way of life'. It seems to me, however, that there is one distinguishing feature which is common to most, if not all, of these tribal societies. I mean their magical or irrational attitude towards the customs of social life, and the corresponding rigidity of these customs (OS&IE, I, p. 1511172).

While stressing that tribal life changes, he brings in a notion of rational attempts at change that tribalists fail to meet. The comparatively infrequent changes in tribal life have the character of religious conversions, or of the introduction of new magical taboos. They are not based upon a fully rational attempt to improve social conditions. Apart from such rare change, taboos rigidly regulate and dominate all aspects of life ... institutions leave no room for personal responsibility (OS&IE, I, pp. 151-2/172).

He stresses how even the Greek philosopher Heraclitus does not distinguish clearly between the institutional laws of tribal life and the laws of nature, "Both are taken to be of the same magical character". We must be careful not to take the changes Popper is discussing too literally; after all, magical and tabooistic thinking survive in our own (supposedly modem) social life alongside openness. A sign of the relative openness of some modem societies is the enlarged space between the unchanging laws of social life and man-made taboos. This is the space of personal responsibility, a space where people are permitted, even encouraged, rationally to ponder decisions by estimating their consequences. Sometimes thought may even alter the framework of laws to which we are subject by altering fundamental features

INSTITUTIONS ENRICHED

145

of social organisation (always assuming that some sociological laws are specific to social and economic forms). Hacohen (2000), summarising a literature, criticises Popper for idealising Periclean Athens. Popper nowhere denies that he sifts gold from much dross as he seeks the birth-ideas of the open society. Is Popper identifying magical and tabooistic thinking with irrationality? This would be a serious misattribution or error of reading. Popper does not say that magic and taboo are irrational per se. What he says is irrational is a magical or tabooistic attitude to human custom and to human law, and then not the attitude as such or even its lingering, but its revival by Hegel. Yet is not an attitude a state of mind? That is one meaning of the term, as when we confess to having an attitude of hostility to someone, we may mean no more than that we have inner feelings of hostility. However, the word has another and better-established meaning. An attitude is also a stance, or posture, or, even, a policy. When we ask, 'what is to be our attitude' towards something or other?, we are asking a question about stance. Shall we treat it with respect, with disdain, shall we ignore it, shall we investigate it, and so on. Feelings may underlie our deliberations, but they are not necessarily embodied in the attitude. Our attitude would be manifested not in some introspected mental state but in action. A magical belief, such as that this spell will tum you into a frog, or a taboo, such as that people in certain ritual states are unclean and it is forbidden to approach them, could be approached with a rational attitude. The spell could be tested and the test-results publicised; the project of terrorising people with threats of spells could be condemned. The taboo could be assessed as to whether it causes those under it to suffer and those believing in it to mistreat them. Indeed, alongside of the magical and tabooistic thinking in our own society we do sometimes engage in such rational assessment. Many taboos and superstitions have simply faded away because our attitude is not to take them seriously enough to oppose them. The crimes of witchcraft and sorcery have long been removed from our law books and serious accusations are now almost unknown. The point is of the greatest importance. Popper holds that the difference between science and magic is not their ideas but the

146

THE REPUBLIC OF SCIENCE

attitude adopted to the ideas. Magic treated as hypothesis is refuted: all known magical rites are ineffective. Being refutable, magic is scientific, but the magical attitude is uncritical. Its contrast is critical. The critical attitude is what he attempted to capture and institutionalise in methodology (ch.2). The fact that there may be ideas in science - the alchemical search for the philosopher's stone, Newton's theory of action at a distance - that resemble those behind magical practice and even that were as an historical fact inspired by them poses no difficulty for Popper. Science and its rationality are an attitude to, a way we treat, ideas. The beginnings of science are the adoption and institutionalisation of an attitude, not a new and superior kind of idea. In scientific institutions ideas are massaged into testable form, and if they fail the tests they become candidates for replacement. In the absence of suitable replacements their status is ambiguous. In chapter 2 I argued that in Logik der Forschung Popper had presented science as an invisible college governed by rationally debated rules of procedure. The attitudes embodied in this college are strongly to be contrasted with the attitudes of magic and taboo. Rules of the college are strictly secular and, indeed, optional. People are free not to join, members are free to try to alter this or that rule. The rules are not surrounded with a magical aura, or sanctioned by the dread of taboo. Science was explicitly compared to the disinterested search for truth in a jury trial. In Popper's discussion, science, magic and taboo, and jury trial are all, of course, ideal types. Traces of magical and tabooistic attitudes towards science and jury trial linger to this day. Disinterested and critical inquiry can occur in magical and tabooistic contexts. The ideal types alert us to alternatives, and allow us to be critical of our attitudes and their institutional embodiment and so be able to contemplate altering them. Under Popper's hand this model of science is transformed into a model for society in general. Popper in OS&IE is engaged in an ambitious test of his ideas about the characteristic rationality of the republic of science to society in general. He is taking the circumscribed rationality of the institution(s) of science as a model for an attitude that should

INSTITUTIONS ENRICHED

147

be extended to social life as a whole, namely an attitude of independent-minded, coolly critical, assessment - of rules, laws, institutions, policies, and so on, by an estimate of their likely consequences. A society which adopts such an attitude to itself and its components is an open society, and a paradigm for the open society is the open institution of science. Seeking a metaphor to characterise the closed society, Popper proposes the ancient one of the organism. In an organism each part plays its role and no other: the legs cannot become the brain, other parts cannot become the belly. In a society which views itself as an organism the way it is constituted is a given of nature. Popper contrasts this to the situation in an even partially open society, where we can observe competition between parts to assume other roles, members trying to displace others, for example in the class struggle, where the ruled endeavour to become the rulers. The very existence of such struggle constitutes a rejection of a natural order of parts and dooms the comparison of society to an organism. How did this transition from the closed to the open society begin in Ancient Greece? No explanation is possible, as that would make invention reproducible. It is, however, possible to specify necessary and even facilitating conditions (facilitating in the sense that conditions create an urgent problem to which the open society was one possible solution). Popper suggests that population pressure created tension within the ruling class, presumably in a struggle for the spoils, and was meliorated by the creation of daughter cities or colonies. But culture contact and commerce created even more stress, as non-traditional classes, including foreigners, appeared on the scene to disrupt the order of things. By the sixth century there were revolutions and reactions, the birth of free thought and clear evidence of the strain of civilisation. The strain comes from the demands made on the citizen by the break-down of the closed society: to be rational, to think for themselves, to be autonomous, and above all to take responsibility for the way things are. Popper concludes of the strain of civilisation, "It is the price we have to pay for being human. " This astonishing remark might seem to equate 'being human' with being 'civilised' and, perhaps, not being tribal.

148

THE REPUBLIC OF SCIENCE

Such an assertion would be unsustainable for someone with Popper's humane and enlightened views, and I would argue that it is a misreading to attribute it to him. It occurs in a passage where Popper is contrasting the open and the closed societies. These concepts are far from being fully and satisfactorily discussed, especially when they are related to concrete historical events, such as developments in Ancient Greece. Popper is explicit, however, that the process of transition from the one form of society to the other is only partial. He draws attention to the continued existence of magical and tabooistic attitudes in our present society. What perhaps he fails sufficiently to emphasise is that closedness is only partial amongst tribes. The pairs 'tribal' or 'closed' and 'open' or 'rational' are deployed as 'ideal types', not as concrete descriptions. Thus we can interpret the strain of civilisation as falling on tribesmen also, to the extent that they are civilised, that is, to the extent that their society compels individuals to be rational, to look after themselves, and to take responsibility. In this respect I think it is a fact that all known societies are to a greater or a lesser degree open; none is completely closed. 4 It could be urged against my interpretation that it is too charitable, that Popper does write this sort of thing: Thus when we say that our western civilisation comes from the Greeks, we oUght to be clear what that means. It means that the Greek[s] began that greatest of all revolutions, a revolution which started just yesterday, as it were, for we are still in its initial stage the transition from the closed to the open society (OS&IE, I, p. 153/175).

This passage appears in the 1962 edition only slightly rewritten: Popper's language at this place gives the distinct impression that genuine closed societies existed historically, societies that are like herds, societies to which the organic metaphor can properly be 4. In footnote 8 to chapter 10 of the first edition of OSIE Popper notes that the strain of civilization is clearly expressed in Heraclitus, with traces in Hesiod, long before the time Hellenic society begins to disintegrate. This note is unchanged in later editions.

INSTITUTIONS ENRICHED

149

applied, not only because of the magical or tabooistic attitude to social custom, but also because there were no phenomena like class struggle or upward social mobility. What, then, can I say in defence of my interpretation that we treat 'open' and 'closed' as ideal types? There is a decisive argument, I think, and it comes from Popper himself; we could call it 'Popper's hermeneutic rule':5 (popper's Hermeneutic Rule) Always try to reformulate the position under discussion in its logically strongest form. (OS&IE, I, p. 2151246, n. 45.) This rule is the opposite of apologetic; that is, it does not say defend your pet ideas by any means to hand. What it enjoins is that where an author has formulated a promising position in a way that is vulnerable to some obvious objection, the commentator should first try to improve upon it before subjecting it to criticism - not for the author's sake but for the sake of the inquiry. Weak positions easily knocked down do not advance inquiry. We learn more from criticising strong positions, whether they succumb to our criticism or not. To interpret Popper as treating the transition from the open to the closed society as a series of real historical events sits ill with remarks he makes at various places in the text. Furthermore, such a claim is easily refuted by the evidence of anthropology. 6 Neither tribes nor totalitarian societies are ever completely closed. The best examples of the impossibility of closure are to be found in some of the dystopias of fiction. But to conclude that the concepts of 'open' and 'closed' are therefore without analytical interest, unable to assist our inquiries, would be a mistake. These concepts are immensely illuminating. It better 5. See chapter 11 of my 1987. The rule can be abused so as to allow one to read one's own view into texts viewed as sacred. Hence the rule invites (tentative) attribution of a problem to a text and not ignoring the text itself or its background. 6. These issues were first broached by Ernest Gellner, in a generous Foreword he contributed to my The Revolution in Anthropology (1964).

150

THE REPUBLIC OF SCIENCE

accords with Popper's hermeneutic rule (PHR) to ignore his failure to state that open and closed are ideal types and proceed to treat them as such. The passages from chapter 1 that we have been discussing were extensively re-worked for the new American edition of 1950, although the line of argument was not altered in any way. There and subsequently, Popper added to his conceptual armoury another pair of ideal types of sufficient importance to warrant an exception to our concentration on the period 1935-1945. This pair is 'abstract' society as contrasted with 'concrete' society.

°

As a consequence of its loss of organic character, an open society may become, by degrees, what I would like to call an 'abstract society'. It may, to a considerable extent, lose the character of a concrete or real group of men, or of a system of such real groups. (OS&IE, US and subsequent editions, I, p. 174).

Popper offers as illustration a thought-experiment society whose citizens never meet face-to-face. Individuals go about in closed motor cars and communicate by typed letters or by telegrams, with artificial insemination for reproduction. This is a "completely abstract or depersonalised society". Yet, Popper notes, elements of this abstraction do permeate our modem society. There are many people living in a modern society who have no, or extremely few, intimate personal contacts, who live in anonymity and isolation, and consequently unhappiness. For although society has become abstract, the biological make-up of man has not changed much; men have social needs which they cannot satisfy in an abstract society (Ibid., I, pp. 174-75).

This discussion is interpolated before the passage from p. 153 of the first edition quoted above. Popper stresses that there is some element of exaggeration in his discussion of the trend towards abstraction in the open society. "There never will be or can be a completely abstract or even a predominantly abstract society - no more than a completely rational or even a predominantly rational society" (ibid, p. 175). There are, however, gains as well as losses from abstraction. Personal

INSTITUTIONS ENRICHED

151

relations are no longer determined by accidents of birth; spiritual bonds can play a role when biological bonds are less prominent. Modem open societies function largely by way of abstract relations such as exchange or co-operation. Here Popper adds a very important methodological corollary. It is, he says, with the analysis of these abstract social relations that modem social theory is mainly concerned. What he could have added, but did not, is that precisely because so many social relations are abstract it is hard for their actors to think out the consequences, intended and unintended; whereas in a face-to-face society (the preferred term in sociology for what Popper terms 'concrete'), actions and their consequences can be more readily seen. When we think of the actions of modem government, with which our relations are entirely abstract, it is obvious how the problems are compounded. Governments take such actions as writing laws or creating new institutions which have intended and unintended consequences on an enormous scale. Much effort is required to ascertain whether the actions succeeded in their aim, whether they thwarted their own aim, whether the intricacies of their implementation resulted in nothing very much at all, or perhaps in monstrous unwanted effects. Much of our social and political discussion turns on these question, about which the social sciences have surprisingly little to say. A clear example is the imposition of taxes. Sociologists and economists do not emphasise enough that the imposition of a new tax is always a social experiment, a conjectural shot in the dark. Despite the existence of elaborate economic models, outcomes are by no means sufficiently obvious to ensure success: whenever taxes are introduced, or even just reformed, either to close loopholes, rectify injustices, or seek new sources of revenue, success is uncertain. Popper clinches his advocacy of the piecemeal attitude to social experiment by drawing on his discussion of methodological nominalism. Each tax, each increment of tax, is a social experiment and accords well with Popper's recommendation that we adopt methodological nominalism. Methodological nominalism and methodological essentialism are names for two alternative

152

THE REPUBLIC OF SCIENCE

approaches to the study of society; one advocated by Popper, the other advocated by enemies of the open society. Methodological essentialism, i.e. the theory that it is the aim of science to reveal essences and to describe them by means of defInitions, can be better understood when contrasted with its opposite, methodological nominalism. Instead of aiming at froding out what a thing really is, and at defming its true nature, methodological nominalism aims at describing how a thing behaves, and, especially, whether there are any regularities in its behaviour. In other words, methodological nominalism sees the aim of science in the description of the things and events of our experience, and in an 'explanation' of these events, i.e. their description with the help of universal laws. And it sees in our language, and especially in the rules which distinguish properly constructed sentences and inferences from a mere heap of words, the great instrument of scientific description; words it considers rather as subsidiary tools for this task, and not as names of essences (OS&IE, I, p. 26/32).

The methodological nominalist will eschew questions like 'what is a just tax?' or 'what is the best society?' and replace them with such questions as, 'how can we make the tax system less onerous and more revenue-efficient?', 'how can we alleviate this or that concrete misery afflicting people in our society?' Popper claims that methodological essentialism dominates the social sciences, not to mention philosophy. A principal essentialist argument strives to show there is an essential difference between the phenomena of nature and the phenomena of society although a moment's thought would show that this distinction itself must be the result of some overall theoretical point of view. The essentialist argument rests on the claim that, even more than the world of sensible things in general, the world of human things is in flux: the human world is constantly changing and so cannot be subject to rigid laws of the kind that dominate nature. Does it not follow, the argument continues, that social events are not repeatable, because social events are defined by human beings and given their meaning by human beings; their individuation and meaning are not objectively observable natural things?

INSTITUTIONS ENRICHED

153

3. A Critical Difficulty: Nature and Convention All contentions about differences between social things and natural things presuppose (or articulate) a theory. Typically, the theory postulates a fundamental or essential difference between natural things and social (conventional) things. According to Popper, the natural attitude is to assimilate social regularities to natural ones. It is one of the characteristic features of the magical attitude of a

primitive tribal or 'closed' society that it lives in a charmed circle of unchanging taboos, of laws and customs which are felt to be as inevitable as the rising of the sun, or the cycle of the seasons, or similar obvious regularities of nature (OS&IE, I, p. 49/57).

One result of the breakdown of the closed society is the discovery that the social, the human, behaves differently from the natural. Natural laws, like those governing the movements of the sun and moon, or the law of gravity, continue unperturbed by human events; normative laws, such as legislation, cultural imperatives, entrenched custom, rules of conduct, moral injunctions, religious taboos, established authority, can alter. Nature and society present an orderly face to the child who is brought up to treat both as walls of prohibition and constraint. Altering the natural order requires resort to magical means, with success by no means guaranteed. Alteration of the normative order may seem similarly difficult. Strangers who come from different normative systems may be figures of awe and danger, for there can be no simple 'Eureka!' moment when a person or a group becomes apprised of the conventional character of its norms. We saw earlier how Popper suggested that a considerable amount of internal social change plus trade and commerce may have been needed to bring such awareness to the Greeks. Before continuing, a caveat. Popper's presentation of the nature/convention distinction is unsatisfactory. His language sometimes suggests that it was a discovery made at a particular time and place. This cannot be so, for the distinction between nature and convention is itself a convention. 'Nature' and 'convention' are not natural objects with natural boundaries like

154

THE REPUBLIC OF SCIENCE

stones, trees or islands; they are ideal types. They dwell, hence, within convention. Possibly the convention of distinguishing them has an earliest known appearance, and possibly that was in Ancient Greece. The adoption of a convention cannot be naturalised into a discovery, and language suggesting otherwise should be discounted if for no other reason than that Popper's argument does not require it.7 Popper's cautious wording about recogmsmg the conventional character of social life ("the fact that there are still many who try to avoid making this step may be taken as an indication that we are still in the midst of the transition from the closed to the open society" (OS&IE, I, p. 50/59) was prudent. Gone is the confidence of the era when the United States and Canada saw themselves as melting pots in which immigrants would shuck off one set of conventions, the conventions of orientation, within which they were raised, and slip into another set, the conventions of procreation, which they proceed to propagate. Instead we presently have multiculturalism, a doctrine of the social that contains a curious tension at its heart. Multiculturalism was proposed as the epitome of tolerance: those who had come to new lands would be given maximum freedom in their new society to perpetuate their ways of life, their norms and conventions, constrained only by the formal system of law of their new homeland (and even that law would sometimes be modified the better to accommodate them). Yet to assume that different cultures and their social conventions are mutually compatible and can live side by side in tolerant harmony, domesticates and even trivialises cultural differences. The vision of harmony between systems of conventions is a fond illusion. A liberal reading of multiculturalism treats norms and conventions as not unlike fashion or costume parades: easy to put on and easy to put aside. I hinted earlier that this is by no means the usual reaction of those coming into contact with the norms and conventions of other groups. Some norms and conventions of the others are bound to violate ours. Ours must prevail for two reasons: (1) being ours they are right and proper 7. See also Agassi 1977, pp. 222-28, 321-22.

INSTITUTIONS ENRICHED

155

and those of the others are wrong and bad; (2) our norms and conventions are what make us what we are, without them we are nothing. To think that they are like fashion or a costume parade is deeply insulting. It is not a fashion for Catholic women to cover their heads in church, or for orthodox Muslim or Jewish women to be covered from neck to ankle and wrists. It is not a fashion for the female children of devout Muslims to be segregated from males after a certain age and educated separately. It is not a fashion that for certain North American native tribes pieces of land are sacred and interference with them sacreligious and disrespectful. So far from being fashion these may be matters of life and death: members of the out-group then become legitimate targets for murder, torture, rape, massacre, kidnap, extortion, and the like. Sometimes the target outgroup can be as vaguely designated as 'westerners' or 'fundamentalists'. Such arguments, in my opinion, are the rocks on to which Popper's critical conventionalism runs. Conventions are not freely chosen, not open to criticism at will, all the more so because they are our responsibility. Both the Roman Catholic Church and certain Muslim groups believe that, if they can gain political power, they should impose the norms and conventions of their group on the state as a whole backed by law (See Blanshard 1952; Gellner 1981). Ireland, Poland, reunified Germany, Austria, Saudi Arabia, Iran, and Pakistan are, in their different ways, illustrations. The impositions are undertaken not in the name of mere conventions, costume parades, but in the name of God who has ordained that a certain set of norms and conventions is right, proper, and to be enforced. Minorities, dissidents, and travellers have ambiguous status under such regimes, often defined ad hoc and under political pressure. God is often taken to have spoken about the norms of social life at one and the same time as he created the laws of the natural world. Particularly acute problems are obvious when we think of conventions governing the status of women, because neither the Roman Catholic Church nor Islamic fundamentalism could be said to be strong on the equality of women. Both exclude women from the centres of religious power. Both attempt to control the bodies of women so that men can use them for reproductive

156

THE REPUBLIC OF SCIENCE

purposes. Neither encourages women to be unmarried, powerful, and autonomous in all aspects of their lives. Both disguise their control under the pretence of 'protecting' women. In environments where these religions dominate, the movement for women's liberation has to proceed with the utmost caution and be satisfied with minuscule gains. Not only is the reign of virtue enforced, discussion and compromise are ruled out. It is rather startling that such doctrines have become so popular in the last quarter of the twentieth century, a time when sociologists had expected that a multicultural and basically secular outlook would spread with modernity. The usual explanations for this not happening - that the popularity of implacability reflects a fear of social change, or perhaps expresses rage at underlying injustice and inequality seem a trifle ad hoc. Neither secularisation nor revivalism is inevitable: both are merely trends. Better to remind ourselves that dogmatists are always with us, and that the magical and tabooistic attitude to convention may never permanently disappear. Different coalitions find common cause with dogmatisms at different times. All coalitions are unstable, so rationalists and liberals may only need survive until current dogmatism fails, hoping that the errors and terrors of dogmatism as such can be exposed before a new one becomes ascendant. How to make a case for controlling the dogmatists? It may be difficult to argue, but it can be and has been for hundreds of years, in Christian and Muslim contexts alike. The Kantian argument which Popper stresses is this: claims to the authority of God notwithstanding, social arrangements are created by the actions of people. People may claim that they are acting under the authority of God, but they claim that because they know there are those who oppose what they are doing. Thus they try shift the discussion from the advisability of what they are doing to whether or not they are carrying out the remit of God. Since anyone can claim to be carrying out the word of God and, since there is dispute about what exactly the word of God enjoins - a fact about all known religions - it merely obfuscates matters to make that claim. Behind the claim, 'I am carrying out the word of God', lies a much less palatable claim, 'I, not you, know what God's

INSTITUTIONS ENRICHED

157

word is and am commanded to act this way'. Now either such a person is open to reason and dispute about infallible access to the word of God, or such a person is not open to reason, which itself constitutes an argument for thinking them to be carrying out not God's will but their own. Their refusal of reason reflects a lack of confidence that they can uphold their claim. The analysis of dogmatism just presented employs Popper's idea of a critical dualism of facts and decisions which says, decisions or norms cannot be reduced to facts (OS&IE, I, p. 52/63). Of course, he noted, decisions are facts once they are made. Still, making them is not just a fact explicable by reference to brute facts (though it may be explicable by reference to other decisions). The enshrining of the Roman Catholic Church into the Constitution of the Republic of Ireland, the imposition of religious law (sharia) in some Muslim countries, were not, as a matter of fact, acts of God, they were decisions of human beings. People can and should be held responsible for their decisions. They may appeal to the word of God, but that may have no more persuasive force than a mass murderer appealing to the voices in his head. Once dogmatists cease dogmatising and present reasons for their actions, rationalists can expose their inconsistencies. All this may be clearer if put positively. Suppose some people agree that a piece of social legislation conforms, as a matter of fact, to the word of God. Also, suppose there are objectors to this measure who can clearly show that it will disadvantage them severely. Is it the will of God to disadvantage them severely? Now, since God has not spoken on this topic, a decision must be made. Deciding to pray for guidance or ask the oracle at Delphi, and such like, are human choices. Whatever answer emerges, interpreting it is a human act also. Consider: if those disadvantaged people were severely damaged and sought legal redress, would the defence that the measure had been implemented because of the word of God cut any ice? God mayor may not have said something, that is report; something was done by some people, that is fact. The usual objection to the secular view of the situation as expounded above, namely, that humans must and can be held responsible for their actions, responsibility for which cannot be

158

THE REPUBLIC OF SCIENCE

shifted over to God, is that it makes social conventions subject only to the whims and caprices of humanity: norms that are merely human norms are arbitrary. But are merely human conventions thereby arbitrary, in the sense of whimsical or capricious? A piece of music, for example, or a house, are human creations that are not arbitrary: they are purposive and we find it possible to judge one piece of music or one house better than another. Our judgments are not arbitrary. Similarly, to say that morality is a human creation is not to say that it is arbitrary: it is to suggest that it has a function and that we can judge which systems better serve their functions than others. We can even discuss, in a non-arbitrary manner, the desirability of the functions. For example, in a pluralist world, systems that aim at tolerance and fairness morally trump systems which do not. Whatever the alleged origins of conventions, they remain the responsibility of those acting them out. Wickedness in the name of God no more avoids responsibility than any other attempted excuse, such as 'I was only following orders'. This discussion of the reasonableness of secularism and the unreasonableness of dogmatism exposes a crucial asymmetry. A secular liberal society can accommodate dogmatists better than a dogmatic society can accommodate liberals. A liberal society can be agnostic towards competing dogmatic claims; a dogmatic society cannot. Science viewed socially must, therefore, be a liberal and tolerant society, and, if its condition is to be stable and protected, must be nested within a liberal and tolerant wider society. A republic of science should fear only those dogmatists who want to control it, or shut it down. Dogmatists with idees fixes or endless ingenuity at avoiding refutation can readily be tolerated (as they should be and are: see Jarvie and Agassi 1979; Jarvie 2001). Given the dualism of facts and decisions, what has become of our original problem of the difference between natural laws and normative laws? Are there no 'natural' laws of humankind in society? Decidedly not, by the very meaning we have given to 'natural', namely immutable, as it were 'untouchable by human hand'. This does not mean there are no regularities in social life, and hence laws of those configurations of regularities, themselves

INSTITUTIONS ENRICHED

159

subservient to the laws of nature, knowledge of which is essential if we wish to take control of our social life. Just because our social life and its institutions are the works of humans does not mean that we fully understand and control those works. As Hayek put it beautifully, they are, in large part, the results of human action but not of human design. 8 Despite his differentiation between nature and convention, Popper wishes to extend the notion of natural laws to sociology. In his consideration of the laws of society Popper refers to economic laws and the laws connected with the functioning of social institutions, and he immediately makes a technological metaphor. These laws, he writes playa role in our social life corresponding to the role played in mechanical engineering by, say, the principle of the lever. For institutions, like levers, are needed if we want to achieve anything which goes beyond the power of our muscles ... Furthermore, their construction needs some knowledge of social regularities which impose limitations upon what can be achieved by institutions ... institutions are always made by establishing the observance of certain norms, designed with a certain aim in mind. (Even mechanical engines are made, as it were, not only of iron, but by combining iron and norms; i.e. by transforming physical things, but according to certain normative rules, namely their plan or design.) In institutions, normative laws and sociological, i.e. natural laws are closely interwoven, and it is therefore impossible to understand the functioning of institutions without being able to distinguish between these two. (OS&IE, I, p. 57/68.)

For Popper, then, natural laws include laws of the functioning of social institutions, which he calls sociological laws. The examples cited are from economics. Thus although society is fabricated, social things do not, for Popper, coincide with conventional things. Social institutions the majority of which, we remember, he believes to be not designed, develop regularities of behaviour best classified with the natural laws of the physical world. The possibility arises that the institutions of science develop law-like regularities of unintended consequences 8. F. A. Hayek, "The Results of Human Action but not of Human Design", ch. 6 in his 1967. In a footnote, Hayek traces the phrase to Adam Ferguson.

160

THE REPUBLIC OF SCIENCE

that enable us to explain and predict some of the phenomena they display. Popper writes nothing about this, so we are left only with the piquant possibility. The duality of nature and convention, then, is not simple and straightforward; it is also seldom readily accepted. Aversion to dualism may have much to do with shirking the burden of responsibility that flows from the man-made character of norms. If humankind made the norms it imposes on itself, then humankind is responsible for them and for their consequences the bad as well as the good. Moreover, if humankind made them then humankind can unmake them: humans can choose to assess them and their consequences and, should they so decide, to alter them because something has been learned in the interim. Society, in this line of thought, is constructed to be open to learning from mistakes. Its sub-society, science, partakes in this feature. Among the most important of the norms needed in a society are those grouped under 'justice', a complex of ideas and ideals much debated down the millennia. Popper believes that Plato is a great social thinker. Plato's most famous work, The Republic, centres its discussion on the question, 'what is justice?'. The great social thinker has not, however, proposed an answer that Popper finds convincing or just. Popper develops his own ideas in the course of a searching critique of Plato's work, as much to neutralise its influence as to win an intellectual battle. He begins by offering some relatively uncontroversial remarks about justice as he viewed it in the mid-twentieth century. What humanitarians (Popper's term of art for himself and the like-minded) mean by justice is (at the very least): (a) an equal distribution of the burdens of citizenship, i.e. those limitations of freedom which are necessary in social life; (b) equal treatment of the citizens before the law; (c) provided the laws themselves neither favour nor disfavour any individual, group or class; (d) impartiality of the courts of justice; (e) an equal share of the advantages which membership of the state may offer to the citizen (OS&IE, I, p. 77/89).

These features are framed by a self-confessed 'critical dualist', one who holds that facts and norms are separable. The

INSTITUTIONS ENRICHED

161

questions of what justice really or truly is, or what 'justice' really or truly means, are ignored. Instead, Popper advances a set of demands for certain kinds and qualities of social arrangement. A recurrent word in (a), (b) and (e) is 'equal'. Popper is not saying that justice 'means' equality; he is treating 'justice' as a label for a series of demands for equal treatment of all citizens, and equal sharing of responsibility amongst all citizens. This by-passes Plato and two millennia of discussion and focuses on the concrete problems of (re-)arranging society. Putting matters as Popper does has many advantages. It avoids the justificationist trap of trying to found or ground claims about justice. Instead, it advances bold claims that shift the burden to the opponent to offer criticism. Criticism to the effect that no foundations are offered for (a)(e) would seem irrelevant at best and irresponsible at worst depending on the urgency of the suffering traceable to any lack of (a)-(e). Effective criticism would need to address three points: consequentialist assessment of whether the proposed features would bring greater justice, and, particularly, whether the demands are consistent with one another; discussion of whether the just goal aimed at is worthy of support; something which may in tum depend upon other goals and changes undertaken or proposed. 9 Whether the list (a)-(e) truly or historically captures what we mean by 'justice' is of little or no critical interest. What label we attach to the quality of society aimed at is unimportant; what society is like, all-important. What the list (a)-(e) proposes is a reform of social institutions in order that they should accord equal status to all citizens. The responsibilities and advantages of citizenship are to be equally shared, and, above all, laws must be constructed, and the process of implementation of the laws conducted, so as to recognise that equality of status. Institutional arrangements that fail to conform to these demands need to be constructed or re-constructed. Popper does not apply (a)-(e) to the republic of science. Given the centrality of science to his thinking it will be interesting to make that application. Should (a)-(e) range over 9. My reworking of this discussion owes much to Jeremy Shearmur.

162

THE REPUBLIC OF SCIENCE

scientists only? That is, should scientists be equal in treatment, rewards, and responsibilities? Equality would be a radical move in such a structured hierarchy. What other considerations are relevant? For one thing, science is not a concrete society, community, or group. It is a paradigm case of an abstract society. Despite its fierce gatekeepers (Merton 1973, p. 521-22), science does have space for outsiders and laypersons.1 0 Should, then, general equality apply to science - especially equality of responsibility? One argument to this point would be that it avoids having one rule for the republic in general and another for the republic of science. In tum, this raises the question of the overall relation between the republic of science and the wider society. We shall take it up in chapter 6. The topic of institutions and their alteration brings us to the problems of politics, for Popper holds that "all long-term politics is institutional" (OS&I£, I, p. 110/126).11 Institutional here is contrasted to personal. I shall discuss politics (and education) in general before applying the arguments to science. Popper criticises Plato for personalising politics with his seductive question, 'who should rule?' Implicit in the question is the suggestion that some person or other ('the best', or 'the enlightened') or some group ('the aristocrats' or 'the just' or 'the people') is the best choice. By contrast, Popper holds that almost all rulers range along a scale from evil through mediocre to incompetent. Even the least harmful rulers have had their incompetent and mediocre days. It follows, he thinks, that it is less important to ask 'who rules?' and most important to ask what institutional protections we have erected against our masters, be they good or evil, brilliant or mediocre, competent or incompetent, just or unjust. Institutional protections that will work unaided cannot be devised: "Institutions are like fortresses. They must be well designed and manned." (OS&I£, I, p. 110/126). Institutions, he holds, can hardly provide the reasoning 10. For severe criticism of education as gatekeeping see OS&/E, I, pp. 11819/135-36. 11. The verb is plural in later editions.

INSTITUTIONS ENRICHED

163

which citizens bring to the task of strengthening, improving and designing new ones; institutions cannot refonn themselves. Political action thus (like all social action) is always a mix of the institutional and the personal. In the long-tenn we need institutions to guide our course; in the short tenn both the problems of politics and the problems with present institutions will be largely personal, namely: where does the shoe pinch?; has anyone any good ideas about repairs?; how can we get those repairs carried out? While Popper is plainly an interventionist, a partisan of social engineering to ameliorate misery, he sharply limits state power over an important aspect of civil society - education. He criticises Plato's moral authoritarianism, namely the view that ordinary citizens, not knowing how little they know, do not know how much education they lack and so may act unwisely out of ignorance. This argument licenses (OS&IE, I, p. 114/130) the Platonic and Aristotelian demand that the state should oversee the moral life of its citizens. How can the less educated and the less good judge for themselves, let alone for their educated betters? In modem tenns this is often phrased as the view that the 'fabric' of society or the 'character' or the 'moral foundations' are at risk. Persons using these tenns seldom feel personally threatened; what are threatened are the longer-tenn or truer interests of those being spoken for. Such moral oversight is enshrined in modem judicial thinking, as emerges in the famous debate between H. L. A. Hart and Lord Devlin, both of whom take it for granted that the state has some kind of independent interest in the moral regulation of the citizenry .12 Popper points angrily to how those in search of the "man of proven probity" (Richard Crossman's phrase) had removed Bertrand Russell from a teaching post at the City University of New York in 1944 (OS&IE, I, pp. 115/131). A difficulty with Popper's view is that most acts criminalised by the law are also moral crimes, so that the state, in criminalising murder, assault, theft, fraud, and the like, 12. There is a very large literature on the debate between Patrick Devlin and H.L.A. Hart, including their own writings. A useful overview is Peter A. Bittlinger 1984. See also Agassi 1978.

164

THE REPUBLIC OF SCIENCE

undertakes a moral supervision of the citizenry. Popper views this as a necessary evil which should not be presented as moral, still less be moralised. Education should foster autonomy, not teach morality. In the educational system we seek to foster a critical attitude to, among other things, the criminal law and its provisions. Teaching what the law says is one thing (factual); discussing why and whether the reasons given are persuasive, is another (critical). Limiting state oversight of morality in education is a means of keeping its oversight of morality through the criminal law under citizen supervision. The difference between viewing the law as a moral agency or as a necessary evil is best illustrated by Popper's contrast between Plato's view of the role of the state in education with his own. While not recommending complete laissez-faire in education, Popper explicitly prefers that to allowing state officials to mould the minds of the younger generation. (He seems to have overlooked the problem posed by what some parents take to be education, namely, endeavouring to make their children carbon copies of themselves.) He is especially against teaching science as authoritative doctrine, rather than as the outcome of sceptical, subversive, Socratic questioning. In a bitterly ironical closing discussion Popper argues that many generations have survived the devastation visited upon them by a system of schools and universities 'invented' by Plato (OS&IE, I, p. 119/136) Democracies have grown complacent in that they tolerate more assignment of arbitrary moral powers to the state than is wise according to the Popperian view of the state as the main danger to an open society. The issue is less one of present abuse, more of the potential for abuse. Laws rarely have built-in mechanisms for testing, and expiry dates in case of failure. On the social engineering view, when a legal remedy fails it should be reformed, possibly abandoned. A measure of the irrational failure of our democracy to be self-critical is how little we hear of such reform and how seldom poorly made, ineffective laws are removed from the books. Popper is a classical liberal committed to state protection of the rights to life, liberty and the pursuit of happiness. He also wants to minimise state power. Life and liberty need state

INSTITUTIONS ENRICHED

165

protection; the pursuit of happiness can be left to the individual. A weakness of his argument is that in pursuing happiness some cultivate virtue and repudiate vice, and hold that the virtue of others, as it were the collective virtue of the community, is a legitimate concern. Puritans of all stripes refuse to accept that the low morals of others does them no harm. They make the principle of harm virtually limitless and need acknowledge no sphere of privacy - in words, in religion, in politics, in education. Popper's principle of tolerance, "Tolerance towards all those who are not intolerant and who do not propagate intolerance" (OS&IE, I, ch. 5, note 6) clashes headlong with puritanism. Why, the puritan asks, should I tolerate what offends me (offends God) (Talmon 1952)? At this point we should come back to science and ask how the problems of politics, of morals, of education, and of limits on state action, apply to it. The first point to make is that science is clearly seen by Popper as part of civil society. Its institutions are joined voluntarily; indeed, the whole enterprise is voluntary. Apart from general framework considerations of freedom of thought, speech, communication, and movement, as well as the rule of law, it would seem that the less state intervention the better. But given the close connection between science and technology, and the dependence of modernised societies on technology, no such separation is likely to be realised. Many forms of science and technology now require high levels of investment, levels that only the state and the largest corporations can provide. State influence on science policy is exercised through this financial dependence, even if there is no direct policy or plan in place. Moreover, preparation for science involves long-term, specialised, and expensive education, something almost all modem states monopolise and fund. This situation is far removed from that Popper considered optimal. The traditional solution to these sorts of problems of funding and education, the creation of self-governing professional organisations, has not been achieved in science. Royal Societies and Academies do not usually have the disciplinary functions of lawyers' and doctors' professional organisations, partly because they do not have the same kind of legally created monopoly with

166

THE REPUBLIC OF SCIENCE

its attendant responsibilities. Even more important, lawyers and doctors organised themselves on a local as well as a national basis. The professional body governs a community of practitioners. By contrast, scientific bodies are as often transnational as local, as often abstract networks as concrete face-toface groups. Furthermore, their ethos is of openness. Openness is not the ethos of the traditional professions, which retain many of the features and outlook of closed guilds. Professionals do not want any outsiders, government or lay, telling them 'Nhat to do. Despite tendencies towards the closed shop, that is not the general ethos of science - where the spirit of Socrates still lives. All of which means that science is high profile, highly politicised, and yet amorphous when it comes to exercising power in defence of its interests. Scientific education is a case in point. The deficiencies Popper singles out in humane education, namely, disregard for science, failure to educate in intellectual honesty, lack of respect for truth, and authoritarianism, are all, in fact, part of current science education (OS&IE, II, pp. 270-71(n6)/283-84(n6); I, pp. 114-15/130-31; II, pp. 260-64/273-77). I pointedly include disregard for science as Popper envisages it - as a great spiritual movement that comes from autonomous criticism. Science education proceeds rather in a pragmatic vein: science and technology are fused, so that truth can be replaced with success, there is much stress on authority, and little intellectual humility. How is this situation to be remedied? Popper's arguments against the state undertaking moral supervision of the citizens are prudential: let us not grant the state power that will make it more difficult to control. Yet he is an interventionist or, as a prefers to style himself, a 'protectionist'. Protectionist measures are to be directed at limited goals. His argument for limits is less prudential than epistemological: hypotheses should be testable, social reforms are hypotheses about change, testers need to control conditions, testing too much at once vitiates the whole exercise. This recommends piecemeal rather than large-scale social engineering.

INSTITUTIONS ENRICHED

167

4. Rationality and Social Reform: Piecemeal versus Utopian Social Engineering We have encountered this pair of concepts before, in "The Poverty of Historicism". In OS&IE (I, p. 138/157) Popper declares piecemeal engineering the only rational kind of social engineering. He begins, however, by presenting the opposite case: the rationality of Utopian social engineering. It is relatively uncontroversial that to be rational an action must have an aim, and is rational "in the same degree as it pursues its aim consciously and consistently, and as it determines its means according to this end" (OS&IE, I, p. 138/157). It follows that if we want to act rationally in social and political matters we must first articulate our final aim, for only with that out in the open can we begin to act consciously towards it and to select our means appropriately. Action on preliminary or subordinate aims can only be organised and ranked by priority if we have our final aim clearly in view. By this reasoning, the most rational initiatives of social change will be those undertaken with a clear vision of the desired end state: the Good Society, for example. By contrast, the agenda of the piecemeal approach is less ambitious. There is no fmal aim. We have instead a general principle - "minimise avoidable suffering" is the candidate Popper suggests - under which we seek out sources of suffering and aim to remedy them piecemeal and without particular regard to one another and certainly without regard to any overall or final aim. This sounds less goal-directed than utopian social engineering, hence less rational. Surprisingly, upon further analysis it turns out to be more rational. Rationality is assessed by measuring the effectiveness of action towards goals. A concrete social goal to relieve particular suffering - Popper gives the examples of health and unemployment insurance, arbitration courts, counter-cyclical economic measures, and educational reform - is something most people can grasp; they can understand the measures being directed towards it; and therefore they can make their own assessments of our success or lack of it in reaching that goal. The rationality of

168

THE REPUBLIC OF SCIENCE

action directed to such concrete goals can be assessed by many citizens. An obvious objection to the claim that piecemeal social engineering is the more rational approach goes like this. If we deal with social suffering piecemeal we may (l) treat symptoms not causes; (2) take measures that mutually interfere with one another or, in the extreme case, cancel one another out. Poverty is a case in point. Its symptoms include uncomfortable housing, malnutrition, substandard health, inadequate education and training, lower life expectancy, high levels of stress, family breakdown, and the like. Anyone of these could be tackled separately without alleviating the others or preventing the emergence of new disabilities from the underlying condition. Would it not be most rational in this case to take aim at relieving the underlying condition of poverty rather than its symptoms? Indeed, would it not be self-defeating and hence irrational to proceed otherwise? Were we to extend life expectancy, for example, without alleviating any other condition, would we not increase the amount of suffering? Better education can also exacerbate suffering if it includes information on relative deprivation, i.e. on how much better off some people are. Most generally, the social pathologies of poverty are a waste of human capital, they inhibit the capacity to create wealth. Palliative measures waste resources by draining wealth rather than creating more of it. All of this is predicated, however, on a metaphor: that poverty is a disease and its manifestations are symptoms. This is, of course, a variant of the Marxist view. Social structures can be healthy or sick, on the rise or on the wane. All hitherto existing societies are sick with class struggle, property relations, and exploitation. Only radical social treatment will cure the patient. Marxism in practice ("actually existing socialism") has, however, been prone to make the patient sick with new diseases of its own, whilst failing to eliminate poverty, class structure, or even control of property (Djilas 1957). I have already argued that social formations are not organic formations; hence the metaphors of health and disease can seriously mislead. Marx himself distinguished structure from

INSTITUTIONS ENRICHED

169

superstructure and tried to find the laws that governed the progress of structure. Reconciling his historical determinism with any kind of political action he reached for a graphic but perhaps even more misleading organic metaphor: lessening the birth pangs. Without claiming that there is one correct approach to thinking about society and its problems, one can say that relying on organic (or mechanical) metaphors, when we know social formations are sui generis, would seem less than fully rational. Poverty is not a disease; and social change is not a birth. Given our ignorance," Popper's philosophy commends methodological nominalism. 'Poverty' is shorthand for a cluster: persons and families without sufficient resources to reach a certain social minimum. Poverty can be defined absolutely, in terms of the resources needed to purchase basic housing, nutrition, health services, education, and the like. Poverty can also be defined relatively, say as a percentage of average expenditures on the same array. It may be possible to devise one single measure of reform that will relieve much poverty - the negative income tax, for example. Yet it is far from clear how that measure will cope with fecklessness and vagrancy. How, for example, to ensure that transfers for children's nutrition and education will not be diverted? Poverty is a label that bunches social problems we wish to relieve but which are not necessarily all connected, and we are ignorant of what connections there are, which are certainly neither mechanical nor organic. In this situation of ignorance the most rational course of action is to deal with each problem as we identify it, and consider consolidated action only when we have some well-tested theory about how society as a whole works. What the discussion above shows is that both the utopian and the piecemeal approaches to social engineering can be presented as rational. What, then, is there to choose between them? Vision of a utopian society is likely to be restricted to a very few, and possibly to only one person, since the logic of the situation requires one single goal and it may be necessary to eliminate alternative goals and even, perhaps, those who propose them. The result will be measures which citizens in general cannot

170

THE REPUBLIC OF SCIENCE

assess. Whether or not there has been progress with the problems is something they have to take on trust from the leadership. A more serious problem afflicts utopian social engineering. It is to be expected that over the period of implementation (such as a five-year or ten-year plan) of a large-scale trans formative improvement of society, ideas and ideals will change. The possibility immediately arises that the values the final outcome was supposed to fulfil may no longer accord with changed ideas on value. The leadership may continue to insist on the original aim, and thus be in some kind of irrational internal conflict with current values or priorities, or may attempt to alter the aim, in which case all possibility of rational assessment gets lost. The leadership's revised aim may well suffer from the defects of the original aim plus the confusion and incoherence created by the shift of ideas and ideals. Such confusion and incoherence may be irresoluble by methods of rational discussion or even persuasive propaganda. The resultant breakdown of rationality gives an excuse for resort to the ultimate methods of violence. Many cannot resist that temptation, some accept it with relish. 13 Some of Popper's critics have complained that there is no way to draw a clear line between large-scale and small-scale changes.14 In anticipation of this objection Popper wrote of limiting the scale of change in recognition of how our limited experience makes the practical consequences of action hard to calculate. He concludes that the sociological knowledge required for large scale planning is non-existent (OS&IE, I, p. 142/162).1 5 13. Such an attitude in Lenin is now documented in Richard Pipes, ed., 1996. Russell spotted it already in 1920 (Russell 1920, ch. III). 14. This is among the arguments in Shaw's 1971. Although it looks serious, it is not. Large and small are relative terms, not essential features. They serve to draw attention to scale, to the scale of our ignorance, and to counsel assessment of costs. No algorithm could accomplish these tasks. All this being obvious, the objection has to be seen as defensive. 15. In the first edition of Popper 1945 chapter 9 is entitled "Aestheticism, Radicalism, Utopianism", in later editions "Aestheticism, Perfectionism, Utopianism". The removal of "radicalism" strikes me as a mistake, since it is a

INSTITUTIONS ENRICHED

171

It is worth noting that Hayek had been developing a stronger argument against large-scale planning, the consequences of which he spelled out in his near-contemporaneous work The Road to Serfdom (1944). The argument was worked and reworked numerous times (and is itself a reworking of an argument of Hayek's teacher, Ludwig von Mises).1 6 This is an epistemological argument to the effect that no person (or computer) can in principle come into possession of the knowledge that rational planning would require. For, according to Hayek, such knowledge is distributed and changing rapidly. The model here is prices in the market. The preference schedules of buyers and sellers are constantly adjusted as conditions in the market alter. No one person or thing has all of this knowledge or can in principle have it. Only price is truly sensitive to these changes, hence price embodies this knowledge - price is the only form in which it can be captured and communicated. No observer or gatherer of data or preferences could in principle match it. It follows that the data required for planning cannot be obtained and that planning therefore is necessarily irrational, hence, by the reasoning given above, potentially violent (Hayek 1935). To another objection, that large scale experiment is essential to gain sociological knowledge of how to transform society as a whole, Popper answers that there is no reason why experiment should not be limited and localised, given that that is what we do all the time. Every new law, new business, new social programme, is a social experiment. It is from these experiments, and their failures, that we learn. Popper's epistemological argument is, then, negative. Because we learn from error, being able to detect error is essential for us to learn, for us to gain knowledge. The argument above showed that large scale experiment is unwise precisely because it creates such a jumbled situation that no assessment of error (or success) is possible: the distinct form of intellectual error. Its substitute, "perfectionism" subsumes both "aestheticism" and "utopianism". 16. See Ronald Hamowy's review of Hayek on Hayek (1996). Perhaps Hayek's pithiest expression of the idea is in his 1945 paper "The Use of Knowledge in Society" (1949).

172

THE REPUBLIC OF SCIENCE

very standards we would apply will have been swept away in the many changes, including changes of aim and of evaluation. Popper's point amounts to a caveat: when designing social experiments we should take heed of the cost, especially when all planning rests on serious error. In the absence of general learning we would be left with the personnel who initiated change. They would have an interest in declaring the outcome a success no matter what, since their own careers are tied into it. Lacking clear criteria of failure, citizens would be unable to challenge critically the leadership's claim of success. Utopianism thus leaves citizens vulnerable to the fundamental aestheticism of many social planners - their impulse to tidy up the world, to make it neat and orderly, and to take care of all loose ends. Such an impulse is glimpsed in Marx and strongly present in Plato. In our own age the planners of cities for beauty's sake (OS&1E, I, p. 145/165) have been architects and city planners, some of whose schemes have gone horribly wrong)7 Popper protests vehemently against aestheticism in politics, because human lives are involved: these demand that we act reasonably, responsibly, and with humane impulses above all. Distress and injustice must take precedence over prettification or orderliness, and only the urgent reform of institutions can accomplish that relief. Furthermore, radical aestheticism is, Popper writes, unrealistic and futile. 18 There are two arguments here. One is that planners and all those who cooperate with them, as well as all the social institutions that make life possible, are part of the social system which is to be altered, even wiped clean, by the utopian planner. Thus the prescription seems to involve 17. I have in mind the way post-war public housing became slums, how the scale and bleakness of many examples of modernism have come to seem repulsive, how garden cities have come to be sprawling suburbs. It is a failure noone defends, yet it is not clear whether we have learned from it. 18. It makes the world unsafe for the neighbour we think ugly, Santayana claims in his 1916.

INSTITUTIONS ENRICHED

173

repudiating the planner and his plans as a prerequisite for radical planning. This is incoherence with a vengeance. The second argument is that, since we are fallible, we should expect that much of what we have planned will not work, and will, then, need modification and replacement. Such alterations will necessarily be piecemeal, being made as we learn from experience. But the logic of radical planning will lead to the demand that failure be followed by starting again, wiping out all we have tried. The cycle will then repeat. Thus we see that sweeping measures of planning are from the start futile and in so far as they do damage to individuals and individual lives, irresponsible and inhumane. Such cycles of damaging failure can only lead the dedicated radicalist to hope for miracles and abandon reason. All historical examples of societies that opted for radical planning are also examples of the failure of that planning and of damage and devastation, human and environmental, as illustrated by the collapse of the Soviet Union and its empire. According to Popper's sociological analysis there is nothing in the least accidental about this outcome. Large scale planning creates a situation in which noble ends and sincere believers will find themselves unable to continue acting rationally. It is rather like the situation in war: once a certain number have died in the cause who will then declare the aims bankrupt and the sacrifice worthless? With this sociological insight we can predict that conditions under the totalitarian regimes yet to fall - preeminently China, North Korea and Cuba - will be exposed as much worse than is yet known. The irrationality and violence will likely prove to be utterly all-pervasive once events from 1949 and 1961 are fully open to public inspection. Even without the human failings of megalomania or corruption, then, irrationality and violence will present themselves as tempting alternatives to selfknowledge, epistemological fallibilism, and humility. These qualities are not new to morality; they are new to politics. Popper argues that we should moralise politics. His mix of morals, politics, and epistemology runs directly opposite to Plato's: humanism and fallibilism instead of aestheticism and infallibilism.

174

THE REPUBLIC OF SCIENCE

Mixing epistemological and sociological arguments, Popper has tried to show how social reform can best proceed in the face of our ignorance and our fallibility. More often than not it is refusal to admit our ignorance and our fallibility that makes experimental social change go wrong. Ill-will is far less often the problem. Plato and Marx are heroes of Popper's book, despite his severe criticism of them. In chapter 10 Popper makes the case for Plato as fundamentally well-intentioned, a victim of, as well as the first to diagnose, the condition called "the strain of civilisation". This is a form of social distress, as we have seen, characteristic of the transition from the closed to the open society, a distress that afflicts even the closed society, because no society (of humans, anyway) is completely closed, completely unchanging. This strain, or uneasiness, is a direct consequence of the shock due to the breakdown of the closed society; a shock which I do not doubt has not been forgotten even in our day. It is the strain of the demand that we should be rational, look after ourselves, and take immense responsibilities. It is the price we have to pay for being human (OS&IE, I, p. 1541176).

Popper compares the loss of bearings when class and status are disrupted by the breakdown of the closed society with the strain on children caught in the break-up of their family and home situation. While exonerating Plato and Marx, Popper does target some intellectuals who fought tenaciously to preserve the closed society and its privileges. He says that the pamphlet by the Old Oligarch, Constitution of Athens was "the oldest monument of the desertion of mankind by its intellectual leaders" . Unwilling and unable to help mankind along their difficult path into an unknown future which they have to create for themselves, the 'educated' tried to make them tum back into the past. Incapable of leading a new way, they could only make themselves leaders of the perennial revolt against freedom (OS&IE, I, p. 164/188).

Instead of exhibiting his hostility to reason, Plato captured by his brilliance all intellectuals, flattering and thrilling them by his

INSTITUTIONS ENRICHED

175

demand that the learned should rule (08&IE, I, p. 175/199). Indeed, the invention of philosophy itself, usually seen as some kind of noble achievement of 'thought', is taken by Popper to be a reaction to the breakdown of the closed society, an attempt to replace the lost magical faith by a rational faith. The Ionians and Pythagoras invented a new social institution, the school, based on tribal models. They were reformers. Socrates was the critic who died for this reform. Democrats welcome criticism, totalitarians do not, for every criticism of authority challenges the principle of authority itself. All attempts to return to the harmonious society, to arrest political change, fail in Popper's view. The more we try to reverse course the faster we arrive at the Inquisition, the secret police, and romanticised gangsterism. Suppression of reason and truth is the first step towards allowing the most violent and brutal methods of rule. Given these trenchant views on reform, what of science? Many obvious questions arise. In what does rational reform of science consist? What should we do with grown institutions that falter? How should we monitor designed institutions? What are the limits of reform? Should reform be left to internal mechanisms, as in the professions, given the status of science, indeed the dependency of modem societies on the growth of scientific knowledge? How can science be defended against Stalinist impositions like Pavlov and Lysenko? How did a Sakharov emerge in Soviet science? What is the role of leadership in science? These questions about the institutions of science show the inadequacies of Popper's skeletal sociology in L8cD, and perhaps his reluctance to look at science as it actually is in modem society. I find no clear address to these obvious questions in Popper's work from the period under consideration, 1935-1945. That many prominent scientists of the twentieth century have been utopians and supported Stalinism seems to raise questions about whether the humble spirit of Socrates did indeed inform the institutions of science. Scientists readily attach themselves to the cult of experts. Popper says that he know of no war that was "waged for a 'scientific' aim, and inspired by scientists" (08&IE, II, pp.

176

THE REPUBLIC OF SCIENCE

2301244).1 9 By his own reasoning, however, that cannot be the

result of the personal virtues of scientists; it must be institutional. Hence the urgency of the questions he overlooked. Perhaps science was not so economically, politically, and militarily important in the period 1935-1945 as it has become. Perhaps its manifest institutional deficiencies and need for reform and control were less apparent. When Popper wrote to Hayek of the Stalinist British scientists that they would learn (from experience) he made a sociological mistake. 20 The mistake was to treat science as a total institution; that is, an institution that governed all a person's actions. In fact, science is not a total institution. All scientists play roles in other institutions. Hence there is potential roleconflict and there can be no argument to the effect that scientific values, such as humility and critical-mindedness, will dominate in all areas of role-playing. It follows that scientific education, scientific work, and scientific success can co-exist with dogmatism and closed-mindedness on non-scientific but also scientific matters. This is an intriguing sociological problem. How can the ethos fail to inform the practitioners? Is it possible that Popper is over-optimistic about scientific institutions or even insufficiently institutional as regards science? Scientists have been devoted Stalinists; they have also been war-mongers (that charge has been levelled against, e.g. Teller). It seems to me that this is not a criticism of Popper's sociologism. Rather, it is a call for it to be carried further and to substitute defensive prudence for optimism towards science and scientists just as towards all powerful institutions. In the next chapter we shall tum to Popper's diagnosis of nationalism and of totalitarianism, social phenomena that overlap, 19. But see Fuller (1996) on how the complicity of scientists in world war II gaivanized the Logical Positivists. :.W. " ... your very understandable reaction against Blackett, Hogben, Haldane, Hemal, etc etc seems to arouse your pessimism (that of the natural scientist Clabbling in Economics), while I am more optimistic. They will learn, in fact they may be most ready to learn. (Viz. M. Polanyi.) Such articles as yours may prove invaluable eye-openers. (Crowther is no real scientist.)" KRP to FAH, undated [1944], PP, Hoover, 305/11.

INSTITUTIONS ENRICHED

177

and yet differ in their content and their emotional resonance. Nationalism is a word its believers readily apply to themselves. Totalitarianism is a scholar's or an analyst's word, largely pejorative in overtone. Both nationalism and totalitarianism have afflicted science as well as society in general. Following that we shall look at Popper's own ideas about the possibilities of rational action. We shall then be in a position to deepen the institutional critique of Popper's institutionalism.

Chapter 5 SCIENCE AND SOCIETY AS LEARNING INSTITUTIONS In The Open Society and Its Enemies Popper envisages society as a cooperative endeavour between individuals, a cooperation facilitated by social institutions, some emergent, some invented. Invention less often is the result of solitary reflection, more often the outcome of divers social processes. Social processes precede, nurture, and to an extent produce the individual, rather than vice versa. Popper, we recall, holds the vast majority of social institutions to have just grown. The (small) minority are those institutions that were consciously devised and implemented. Language, markets, pathways through the forest, are institutions that emerge without being articulated as aims. Popper took for his first model of social life the friendlyhostile cooperating community of scientific inquirers, a group with shared aims, operating under a regime of methodological rules that foster those aims, rules which offered scaffolding for the building of the social institutions of science. Popper did not, in The Logic of Scientific Discovery, give an analysis of actual scientific institutions, their diversity, their rituals, their shortcomings. Perhaps this is just as well: in certain obvious respects the social aspects of science are a poor model for society in general. For one thing, human communities are not in general groups that have congregated around stated aims. Most human communities existed long before aims were articulated or adopted (this is one of the respects in which the countries of the New World differ sharply from those of the Old World). For another thing, a person's membership in the most critical social institutions and groupings is involuntary: family of orientation, ethnicity, linguistic group, citizenship at birth, are all given rather than chosen, and the first two are not malleable (which is not to deny that people can resort to deception). Lack of community, lack of institutional aims, involuntary membership are other differences between the social aspects of science and society in general. They are less decisive than the

LEARNING INSTITUTIONS

179

overall, asymmetric dependence of science on a wider social order. Science is a specialised set of institutions dependent upon prior and external social conditions. The institutions of science can emerge and flourish only when there is division of labour, surplus wealth, and a degree of civil peace and civil liberty. It is, of course, possible that in the process of building institutions for science we can learn lessons that will help us to improve the basic institutions of society. It is this Popper has in mind, I believe, when he holds up generalised Socratic dialogue as a model for all human learning, possibly for all human cooperation. According to Popper, human history is largely a history of power politics, i.e. mass murder and other acts we would not normally hesitate to call cardinal crimes, and has afforded only a limited place to Socratic dialogue (OS&IE, II, p. 2571270). Popper offers arguments to the effect that it is in the interests of everyone to expand rather than to contain or restrict in any way the use of dialogue. So long as openness is a matter of degree, every society being partly closed, we cannot identify science with openness. Notoriously, scientific research, abstract and applied, continued in the Soviet Union (Sakharov 1990). Since research depends upon being rational, i.e. critical, we cannot identify the ideal type of the closed society with the completely closed society: it was earlier established that no society is completely closed. In chapter 3 Popper's developing ideas on institutions, on individuals, and on rationality were discussed. In chapter 4 Popper's overall vision of human history as a struggle to achieve the transition from a closed, magical, tabooistic world, to an open, rational, scientific world was discussed. Popper admits that no society is ever fully closed. A conventionalist attitude to society and social life is difficult to discern in a closed society; is resisted by reactionary elements in an open society, since it legitimates change. But once invented, such an attitude is applicable to both the open and the closed ideal types. Let us briefly apply it. The first fruit of openness is the ability to see social arrangements as humanly malleable, thus raising questions about the limits of such malleability and the most effective ways of utilising it. Once social arrangements are seen as a human product under human control they are also seen as a human responsibility, and

180

THE REPUBLIC OF SCIENCE

their flaws and deficiencies no longer a visitation from the fates but something that could be otherwise. A person in a closed society struck by this realisation will become an anomalous individual, one with a different outlook. Such an outlook is evidence of openness; indeed, like openness, individuality is a continuing process, not a finite accomplishment. Closed societies change in different ways from open societies. In open(ing) societies individuals may cause social change; in closed societies change must come from outside because members are enjoined to block change. One exception to this, which Plato utilised, is change dictated from above, by acknowledged leaders. The other exception is response to external events, natural and social: floods, droughts, seismic disruptions, plagues, invasions, wars. Where change has come from human invention and cooperation - the neolithic revolution, agriculture, animal husbandry, trade, writing - we see evidence of rationality, but little of individuality or of a conventionalist attitude to social institutions. These changes possibly came from below but took root only where the leadership endorsed them. This leaves scant room for conventionalism. Indeed, in explaining any of these great changes in human history the usual evolutionist assumption is that the new arrangements gave advantage to those who adopted them over those who did not. Groups that did not change are lost without trace. Present day open societies are strikingly new in that they show awareness of some of the processes they are going through, such as the medical revolution, the atomic revolution, global population increase, the electronic communications revolution, the extension and equalisation of civil rights to the disadvantaged, the alteration of the environment. All such processes can be assessed on their mixture of benefits and costs, and rational discussion about taking responsibility for minimising the costs is possible. Less grandly, human social divisions can be seen in overview. Humanity is sub-divided in many different ways, by language, by religion, by nation, by state, by even more amorphous yet still important measures such as ethnicity and culture. Each is a set of conventions, institutions, traditions whose very diversity in overview reinforces the view that they are not nat-

LEARNING INSTITUTIONS

181

ural. Not being natural, they invite a different rational assessment from the rational assessment of nature. Challenging much of humanity to think this way is an unfinished task. The rational assessment of nature consists of trying to master its laws and calculating ways in which some of these can be used to defend ourselves against others - methods of storage, for example, can be devised as hedges against the inevitability of winter and the lesser inevitability of flood, drought, plague. What about the laws governing social life? Maybe a few of these are iron, though we would be hard pressed to spell out the members of the set. The vast majority are laws that will apply to this social system and not to that. Laws of prices, to take an example, will not necessarily govern a barter system; principles of kinship will not govern an abstract society. On this point Popper's ideas are to say the least unclear. He cites examples from neoclassical economics as sociological laws comparable to natural laws, incautiously failing to mention that, unlike natural laws, such sociological laws may be far from universal. If they are not universal, then their capacity to explain is called into question. Chapter 4 presented Popper's division of societies into open and closed as contrasted to the classical division of laws and truths into those of nature and those of convention. The classical division does not allow for gradation; Popper's division demands it. We saw that he may have over-estimated the conventional character of social arrangements, even though he allowed for sociological laws comparable to natural laws. We will return to these issues later in this chapter when discussing nationalism. We also saw that Popper argued that openness invited over-optimism towards solving all social problems, or, atleast, did not discourage it. Indeed, our very ability to envisage the whole panorama of human social history inspires the impulse to reconstruct society wholesale as a means of eliminating exploitation, violence, and injustice. Alas, such attempts at holistic alteration had the unintended consequence of vitiating the ends for which they were undertaken, an irrational outcome in the name of rational reform. In the present chapter we will revisit some of these topics as we consider 1) nationalism, 2) the origins of totalitarian social forms, and 3) Popper's view of institutions and their rationality,

182

THE REPUBLIC OF SCIENCE

as well as 4) the rationality of individuals made possible by institutions. Nationalism may seem remote from science, but not if we take science as the model for all social institutions, as is explored here. Moreover, the two interact, e.g. when the achievements of researchers are credited to the spirit of the nation, thus causing a clash between the national and the scientific spirits, rather than mutual indifference. There is a direct clash between the Enlightenment ideal of cosmopolitanism or internationalism and a particularist, expressive, view of science. There should, on this Enlightenment view, be no nationalism in the social aspects of science. Popper, taking Enlightenment science as his social model, argues that there should be no nationalism, period. 1. The Nationalist View of Social Life All nationalism and racialism is evil, and Jewish nationalism is no exception. --- Popper (1976, §21)

A striking feature of Popper's conception of the social is that it is utterly demystified: society and its social institutions are machinery for cooperative social living, mechanisms to coordinate attempts to achieve aims. Absent from his vision of society are ideas of national identity, national character, the soul (or essence) of a people, collective destiny, mystical connections to the land, the ancestors - a whole range of mumbo-jumbo indulged by even the sanest of his predecessors. Popper's distaste for these positions is palpable, quite aside from his rational arguments against all of them as essentialist. He consciously aims to minimise the role of such ideas. Most striking is his treatment of nationalism. Since nationalism is modem, it enters only in the second volume of The Open Society and Its Enemies. It is distinct from tribalism and parochial feelings. Yet advocates of nationalism, Popper claims, appeal to such tribal feeling, which he describes as "always strong" (OS&IE, II, p. 48/50). He traces the political expression of modem nationalism to Prussia in the 1840s, a

LEARNING INSTITUTIONS

183

Prussia populated largely by Slavs. (Later scholarship on nationalism has not agreed with him on this origin claim, see Smith 1971, 1996.) He claims that the principle of the national state has become all-pervasive since then, despite its reactionary origins, which have been forgotten. (Gellner 1983 and Smith 1971 argue that early political nationalism was liberal, not reactionary.) Thus liberals of the ilk of Woodrow Wilson and Jan Masaryk were convinced that the principle of national self-determination was a basic postulate of political ethics. How anybody who had the slightest knowledge of European history, of the shifting and mixing of all kinds of tribes, of the countless waves of peoples who had come forth from Central Asia and split up and mingled when reaching the maze of peninsulas called the European continent, how anybody who knew this could ever have put forward such an inapplicable principle, is hard to understand (OS&IE, II, pp. 48/50-51).

Popper's explanation follows Russell and others in suggesting that intellectuals fell victim to the metaphysical political theories of Plato and Hegel and the nationalist movement anchored in those ideas. Popper was writing in the period 19381944. From the perspective of the twenty-first century his question is even more poignant and his solution even more sorely tested. The pervasiveness of the reactionary thought of Plato and Hegel (not to overlook Heidegger) among the intellectuals of East Central Europe (cf. F. Tudjman of Croatia, and R. Karadicz of the Bosnian Serbs) and the former USSR is plain enough. 1 Even civil war does not make these leaders re-think their intellectual loyalties, any more than it leads those who elect them to repudiate the nastier consequences of their nationalist ideas. Political discourse being closed to any non-nationalist talk, voters must be under the impression that the next nationalist will be the one who puts matters right. No number of previous failures shakes them. I am irresistibly reminded of Melanesian cargo cults, where it was 1. Even V. Havel of the Czech Republic, an eminent liberal, takes his cue from such reactionary philosophers.

184

THE REPUBLIC OF SCIENCE

almost as if the participants were 'imprisoned' in a thought-set that could not be undermined by refutation of its individual tenets.2 None of the theories which maintain that a nation is united by a common origin, or a common language, or a common history, is acceptable, or applicable in practice. The principle of the national state is not only inapplicable but it has never been clearly conceived. It is a myth. It is an irrational, a romantic and Utopian dream, a dream of naturalism and tribal collectivism. (OS&IE, II, p. 49/51)

The question for the agenda is, then, in what consists the attraction of this myth? Which is another way of asking, what functions it serves. Subordinate questions are: how is its mythic status concealed and a semblance of rationality given to stateidentity doctrines? What sorts of contexts make nationalism the rational option? How relevant is nationalism to science? Possibly it is less that these myths appeal, more that people are averse to what they see as the only alternative. The alternative to nationalism is pluralism: states as legal and political entities that encompass within their borders any number of nations. Under pluralism, tribal considerations of kinship, resemblance, common customs, and language are no guide to managing relations with one's fellow citizens. Those relations are governed by public law and regulation and demand a good deal of personal responsibility in fulfilling the abstract demands of citizenship. If the law requires that national minorities get equal treatment then individuals cannot with impunity allow themselves to act on instinct or prejudice. The law may thus demand that people live up to higher standards than many customary societies apply to the treatment of out-groups. The law may demand that powerful majorities curb any tendency to convert their advantages into permanent and discriminatory political power. It is not hard to see how the older ways can be presented as fairer and more natural, and demagogic 2. See my 1963, 1964, 1966, 1970, 1972a. In these works I applied situational logic to give a rational reconstruction of cult behaviour that had been all-tooeasily dismissed as 'irrational'.

LEARNING INSTITUTIONS

185

accusations made against those who advocate pluralism. Yet it is also clear that nationalisms can coexist within pluralism. Nationalism is always dangerous, but not always without value. Popper does not oppose national liberation movements against colonial oppression, for example. But Kantian universalism is best: that is, he sees relief and protection from present ills and pains, such as exploitation and injustice, best secured by extending full and proper human rights and dignity to all. Thus, if the Hungarians in the Austro-Hungarian Empire felt themselves to be getting a raw deal, Popper's principles would lead to the suggestion that they fight for a better deal - for all citizens of the Empire. The idea that Hungarians can only secure decent treatment by carving out a separate state run by and for Hungarians falls foul of Popper's argument. As the quotation on p. 274 shows, his decisive argument is that separation makes no historicalor practical sense, since Europe is criss-crossed by national differences, most of which are also lines of discrimination and mistreatment. (Asia, Africa, and the Western Hemisphere are no different.) He also has strong objections to the carved-out result: nationalist ideas give states license to go beyond the rectification of grievance, the relief of genuine human pain. What he sees as Hegelian reactionary nationalism is rooted in phoney history, the history of 'Spirit' unfolding in history. The history of a nation is the history of its essence (Spirit again) asserting itself on the stage of history. This is a ready-made excuse for crimes against fellow citizens and foreigners, for discrimination, ethnic cleansing, war, genocide. Popper's assault on nationalism has not been met. He refers with distaste (which he shares with Bertrand Russell and the Logical Positivists) to the intellectuals' appetite for war in 1914. His argument that nationalist thought relies on indefensible distinctions of race, tribe, ethnicity, and language has also not been met. His charge that nationalism is rooted in historical and other myths is widely admitted. Gellner has said bluntly that the intellectual content of nationalist ideas is beneath discussion (Gellner 1983, p. 123-25). The existence of the ideas and sentiments is, however, a sociological fact, and their appeal a sociological problem.

186

THE REPUBLIC OF SCIENCE

What is Popper's view of the temptations of this evil? Possibly it seduces intellectuals because it guarantees them a special, even a political, role as articulators of the national culture and Popper is clear that intellectuals are tempted by power-worship. If 'the best and the brightest' are offering a set of ideas as a solution to major political ills then we have the beginning of an explanation. But we can go no further. Popper gives us no clues as to why the populace would eagerly embrace this set of ideas, even when they lead to war. Popper's hostility to nationalism rests on his assumption that it cannot be reconciled with political liberalism. Is this true? It certainly cannot be reconciled with Fichte and Hegel's variants. But why not with Masaryk's? In practice it often is; that is, in the liberal democracies of the world there is some nationalism, but usually it is contained. The question is whether the combination can be made intellectually coherent. Friends, even co-critical rationalists, may divide over the issue. Agassi, in his book on Israel, tries to offer such a coherent view (2000). Hattiangadi, in his paper in Agassi's Festschrift, suggests that no coherent case for liberal nationalism is possible (Hattiangadi 1995). Nationalism in general being pernicious, it follows that nationalism in science is to be condemned. Encouragingly, outbreaks of nationalism in science come and go, being much more common before, than after, the second world war. Most often, scientists are partisans of science as such, and Popper has emphasised how their love of a problem, and their tenacious defence of a pet theory, are part and parcel of friendly-hostile cooperation (see ch. 2). Scientists even defend their turf, that is, they invest their work with emotion and partisanship. All such failings fall short of the excesses of nationalism. Notoriously, loyalty to science is loyalty to an endeavour that crosses national boundaries; attempts to appropriate scientific achievements to the credit of national culture are crass. (Like universities and countries boasting of their numbers of Nobel Laureates.) The Enlightenment in general and Kant in particular celebrated the universalism of science. One attraction of Marxism for scientists has always been its self-proclaimed internationalism and universalism. Yet, in both world wars scientists passionately contributed their scientific exp-

LEARNING INSTITUTIONS

187

ertise to assist their countries in waging war. Powerful scientists could and did orchestrate expressions of nationalism by scientific institutions. It remains true, however, that in the normal course of things the institutions of science do not need and do not seek to generate nationalist sentiments. These institutions function well enough by pursuing their scientific aims under their methodological rules. Science is a curious republic: it is not based in anyone nation and has no fixed body of citizens. Indeed, its citizens all possess dual nationality: of their native land, and of the republic of science. The surrounding communities that sustain science may be deeply beholden to worthless ideas and ugly sentiments especially nationalism. Whether these ideas and sentiments are subjected to scientific criticism or not, a potential conflict exists. Popper offers no easy answer or indeed alternative to nationalism. He does not offer a sociological explanation of its appeal to scientists and intellectuals or to the general public. Without an explanation we do not know how nationalism can be combated and minimised in society in general or science in particular. This lack points once more to a criticism I ventured when discussing Popper's conventionalism, namely, the persistence and emotional hold of some conventions, regardless of recognition of their conventionality. One must assume that scientists who embrace nationalism do not confuse it with the physical nature they study. Knowledge that it is conventional does little to blunt its force. Society is not a Socratic seminar writ large precisely because of the emotional power of many ideas and the weaknesses of controlling institutions and traditions. The argument from emotions should not, however, be taken to be psychologistic. It is a sociological variant of the psychological experience, since many social institutions and traditions, perhaps even the wider society itself, are sites of emotional bonds, sustenance, identity and the like, so that, in a broader sense of rationality, the attempt to intellectualise this ingredient away is not necessarily rational. Gellner has taken this critique one step further, by pointing to the social disruption entailed by the concept of disinterested and supra-human truth. Since so many social bonds tum on loy-

188

THE REPUBLIC OF SCIENCE

alty to ideas, a general scepticism towards ideas is already highly disruptive. Refusing all loyalty except to truth, and placing truth beyond human control, is still more disruptive. Gone are all guarantees of loyalty to actual existing social bonds. Such bonds are themselves now hostage to the vagaries and vicissitudes of nature. This may be the most decisive objection to taking science as a model for social institutions in general. I shall return to it in chapter 6. 2. The Collectivist View of Social Life Nationalists try to appropriate the status of science, nationalistic scientists are caught in a conflict of ideals. Keeping nationalism out of science is not a simple political problem. Scientific selfgovernance is a useful antidote only so long as the governing ethos of science is not itself nationalistic and so long as funding from outside is not controlled by nationalists. Nationalism in its most virulent form is fascism, and fascism presents further problems for science in its emphasis on the collective and its will over the individual. Fascism takes all social activity to be at the service of politics and requires total homogeneity of the population. The latter proviso rescues the theory of collective will from refutation: where society is homogeneous, the collective will works perfectly. This raises the spectre that there is science worthy of a nation (or 'race') and science that is not, that the social/political attachments of the individual scientist are relevant to the worth of the product. Fascism grew out of the breakdown of Marxism. Popper argues that because social democracy failed to oppose war in 1914, was unable to alleviate unemployment and depression after the war, and defended itself only half-heartedly against fascism, its commitment to freedom and equality came into question. It lost the initiative to fascism, a much more militant and activist movement that tapped into yearnings for freedom and equality and (by adding nationalism to the brew) twisted them into aggression. Since its roots, Marxism and social democracy, were secular, fascism, unlike many traditional revolts against freedom

LEARNING INSTITUTIONS

189

and reason, could not ally with a traditional religion. 3 Instead it utilised a form of nineteenth-century evolutionist materialism. Popper says the formula was: Hegel plus a dash of nineteenthcentury materialism. Racialism involved substituting for Hegel's 'spirit' the quasi-biological concept of blood or race. The result was the following set of ideological tenets. (a) The state as the incarnation of the nation. (b) All states are in competition.

(c) The interests of the state override morality; only history is its judge. (d) War of young or rising nations against others is ethical, for war, fame, and fate are desirable goods. (e) Leadership belongs to great men, i.e. those who seize power if) Heroic life (live dangerously) is a higher value than the shallow mediocrity of petty bourgeois safety, as it better serves the state. Having "tried to show the identity of Hegelian historicism with the philosophy of modem totalitarianism" (08&IE, II, pp. 75178), Popper cleared the decks for his long discussion (ten chapters) of what he saw (1938-1944) as the main totalitarian danger facing the post-war world, Marxism. It is not necessary that Marxism be totalitarian: the democratic variant of it could have triumphed and this would have gained Popper's approval. Instead, Marxism's social democratic failure spawned fascism and perhaps, we might say, a fascist form of Marxism for long known as Marxism-Leninism-Stalinism. In the midst of a world war against militant fascism it was prescient of Popper to discern the element in Marxism that tipped the scale in favour of totalitarianism: a form of the revolt against freedom waiting to make a come-back, as it were . Yet he also insisted that Marxism was much more intellectually interesting and perhaps just as appealing as its fascist rival. 3. Popper's anlysis here is too particular to Germany. It fails to account for the variation we know as 'clerico-fascism'.

190

THE REPUBLIC OF SCIENCE

It is not my purpose in the present book to rehearse and discuss Popper's critique of Marx, although in fact I share his own estimate that it was "devastating" (OS&IE, 1952 and subsequent eds., p. viii). Popper held out the hope that intellectuals of good will, opposed to exploitation, injustice, unemployment and war, could put aside the errors of Marx and concentrate on the problems to hand, which Marx had done so much to expose. 4 Out of the debacle of Marxist ideas he wanted to rescue for philosophical individualists one of Marx's greatest ideas (OS&IE, II, p. 85/88): the autonomy of sociology; the idea, that is, that society is not simply an aggregate of individual human beings and their actions, it is, rather, a system of institutions and individuals, and every individual is preceded by the institutions. Actions are always initiated by individuals, but the outcomes of action cannot be explained without reference to an autonomous social level. Society as an aggregate of individuals is in some ways a common sense view. Popper credits John Stuart Mill with articulating the strongest version - which Popper calls psychologistic individualism, or methodological psychologism, the view that the study of society must, in the last analysis, be reducible to the study of psychology and that human nature, the laws of the mind, are the fundamental components of a science of society. The processes and entities postulated in those laws are what makes up society. Science is then no longer envisaged as only possible in an institutional matrix, but merely as an aggregate of individual efforts, individual attitudes; scientific knowledge merely some well-entrenched beliefs of the leadership. Popper offers two main arguments against Mill's psychologism. First, that it has to operate with a psychological origin of society, i.e. a point in time where rules or institutions were introduced. This assumes a human nature and a human psychology prior to society - an assumption that is both an historical and a methodological myth. Second, that our actions have social repercussions, often unwanted, and hence hardly explicable in terms of 4. This sentiment was strongly iterated in a letter to Hayek of 11 January 1947, PP Hoover 305/13 where he calls for a reconciliation of liberals and socialists as fellow democrats.

LEARNING INSTITUTIONS

191

our motives or even as results of our intentional actions. Unintended and unwanted consequences are, we have seen, an important object of social studies. A third argument against Mill's psychologism stresses that situations are vital to understanding actions, for rationality is assessed in relation to situations. Psychological analysis presupposes some standard of what actions are rational in the situation in question (OS&IE, II, pp. 89-90/93-96). What psychologism gets right is that actions are individual. It follows that social explanation may be satisfactory only when we can point to action, specific or typical, that could lead to what we want to explain. This is Popper's "methodological individualism": all social phenomena, and especially the functioning of all social institutions, should always be understood as resulting from the decisions, actions, attitudes, etc., of human individuals, and that we should never be satisfied by an explanation in terms of so-called 'collectives' (states, nations, races, etc.). The mistake of psychologism is its prejudice5 that this methodological individualism in the field of social science implies the programme of reducing all social phenomena and all social regularities to psychological laws. The danger of this prejudice is that it leads to historicism ... That it is unwarranted is shown by the need for a theory of the unintended repercussions of our actions, and by the need for what I have described as a logic of social situations (OS&IE, II, pp. 91-92/98).

Two points in this passage are particularly striking. One is Popper's view of the individual; the other, is his strong reliance on social situations. He views the individual as a pivotal element in the explanatory model: "social phenomena ... should always be understood as resulting from ... human individuals". Even though individuality may be a social construct, it is the locus of thought, of morality, of value. This is Popper's Kantian Enlightenment view. Even if Marx is correct that social existence precedes and determines consciousness, for moral and intellectual reasons it is preferable to treat individuals as autonomous actors/decision 5. "Prejudice" is changed to "presumption" in later editions.

192

THE REPUBLIC OF SCIENCE

makers and as nodes of value. To attribute autonomy or moral value to social entities is to license them to override recalcitrant individual human beings for 'higher' purposes. Or rather, it is to license some human individuals to override others in the name of supposed 'higher' purposes. Popper stressed that social situations are not mere aggregates of individuals: they are these plus the conditions in which these aggregates find themselves, and these conditions are the constraints on human action as well as the grooves and channels down which the consequences of social actions flow. Included in the notion of the social situation as the social scientist understands it is the actor's model of it, as well as the adequacy of that model. How the actor sees the situation constrains both her aims and the actions to be directed at realising the aims. The actual outcome of the actor's initiative requires for explanation a correct model of the situation, one that reveals the deficiencies of the model used by the actor. When we apply these general considerations to the social aspects of science they constitute clear reasons for rejecting collectivist, be they nationalist or totalitarian attempts, to co-opt science, as they lead to the prediction that science may not survive attempts to control it or recruit it. It will be useful here to revert to Popper's original discussion of science as a social institution constructed around methodological rules. Popper explicitly argued against an individualist or psychologistic view of science. Robinson Crusoe, alone on his island, whatever physical and mental equipment he possesses, lacks the social institutions of science and so cannot continue to do science, no matter what he may think he is doing. His situation makes the doing of science impossible and his grasp of that situation is a contingent matter that does not alter the fact of it. He needs the friendly-hostile cooperation of others, and this will not be forthcoming even were he joined by others, unless they institute science. Without the institutional methodological rules of science, Crusoe's appreciation of the problem-situation cannot be checked, so that, instead of debating others and learning from his mistakes and those of others, he will be talking to himself. He will be like Bacon's

LEARNING INSTITUTIONS

193

spiders, "which spin webs out of their own bowels" (Bacon 1620, Aph. XCV). (See also below, §3.) Popper's positivist opponents of 1935 still operated with a psychologistic model of science and scientific knowledge. Even the Marxists among them did so too, if not consistently. Knowledge was a special kind of (verified) belief; beliefs are states of individual minds. Scientific knowledge is therefore the sum of verified beliefs. Methodological issues are simply those of methods of verifying beliefs. (Current jargon would specify methods of generating more reliable and fewer unreliable, beliefs.) The psychologistic individualism was so deep that the positivists in their discussions of Popper's book did not notice that he was challenging and overthrowing this model of scientific knowledge. The 'knowing subject' was already gone from his philosophical system in 1935, though Popper did not express it that way until 1967.6 His critical study of Marxism in the 1920s had convinced Popper that psychologism would not do. So we have seen that in The Open Society he quotes Marx approvingly: "It is not the consciousness of man that determines his existence rather, it is his social existence that determines his consciousness". This acceptance of Marx's view already deeply affects Popper's 1935 conception of science: it sociologises it. The state of scientific knowledge is what determines beliefs (if anything does), not the other way around. The state of scientific knowledge is a social fact, in fact a social situation, a given that is not reducible to an aggregate of individual beliefs. Despite his own at times individualistic-sounding language (" A scientist, whether theorist or experimenter, puts forward statements, or systems of statements, and tests them step by step", LScD, p. 27), it is an interpretative blunder to read Popper's model of science as individualistic, as inept as to read his methodological individualism as individualism tout court, i.e. as ontological (the claim that institutions do not exist). Social institutions do not exist if no 6. My allusion is to his paper "Epistemology Without a Knowing Subject" read at a conference in 1967 and published in the proceedings in 1968. The most accessible version is as chapter 3 of his 1972. For an interesting if supportive critique see Agassi 1998.

194

THE REPUBLIC OF SCIENCE

people exist. Ontological individualism rests on this trivial truth.

It rests uncomfortably, however, since institutions are indepen-

dent of the people who inhabit them. Action is individual but aims, knowledge, and outcome are all socially given. Thus Popper's methodological rules are not rules to purify the mind (Bacon), not a method to improve the understanding (Spinoza), not laws of thought (Boole). They are procedural and policy rules that strive to constitute a particular kind of communal endeavour: a set of learning institutions. The learning known as science can only be accomplished within such institutions. Marx saw that social institutions are independent powers which shape and constrain; although he was not wholly at ease with this knowledge because it conflicted with his materialism. Consequently, he sought to connect social power to the power of machinery. In chapter 17 of The Open Society and Its Enemies Popper discusses Marx's idea that real power came from the evolution of machinery, and then from the economic system of classes machinery spawns. The least important influence is politics. This reduction of the social to the material conflicts with Marx's own sociologism, his sense of social institutions as determining consciousness. Popper reverses Marx and points to the supremacy of politics and the necessity to extend politics to economic matters. He argues that Marx never took seriously the paradox of freedom, namely that no form of freedom escapes the tendency to defeat itself if it is not limited. Marx instead assumed that progress was inevitable. Marx was unaware of the totalitarian dangers in his own arguments. Popper, a good student of Marx, avoids the paradox and carries through Marx's sociologism by presenting a picture of politics as consisting of what is elsewhere known as checks and balances, especially the check of the people on the government. Indeed, so far from social power deriving from machinery and its associated economic power, economic power is entirely dependent on political and social power, because the state and its laws protect property and wealth. Marx and Marxists dismiss these as merely 'formal' freedoms because laws may not be enforced. Better to rely on the inexorable laws of history.

LEARNING INSTITUTIONS

195

What Marxists denigrate as 'merely formal' freedoms are our means of controlling coercion and exploitation. Once we have freedom from coercion we have no business blaming other people or sinister economic demons behind the scenes for our troubles. It is up to us to create institutions that accomplish our goals. Any failure is ours, not that of the system. So-called formal freedoms are a necessary condition for science - absence of coercion being a sine qua non of the institutions devised to suit its purposes. In this respect, at least, science is a paradigm case for civil society in general. Institutions are, of course, developed by people with aims and knowledge who form an appreciation of the situation, an appreciation liable to error. The institutions we create, even the aims we set for ourselves, may be affected by the errors we make in our institution-building. Each error discovered can reverberate all the way back to our models of the social situation and of ourselves. Popper's model of science is not individualistic but social, not materialist but political. Science is not subservient to the polity from necessity; only from choice. Science, like civil society in general, needs protection by as well as from the state. State intervention in civil society is required to counter injustice and exploitation. Combating injustice and exploitation seem modest enough goals, very far from totalitarianism. Popper, interventionist though he is, points to the dangerous charter for state intervention created by these modest goals. He fmds a new paradox of state power, in that even piecemeal interventionism extends the power of the state and thus requires extra vigilance if that power is to be controlled. To minimise the monitoring and controlling effort required, intervention should be strictly limited to what the problem demands. We earlier saw that the usual rebuttal to this view (offered by Marxist apologists, among others) is that small, seemingly discrete problems may be linked and that dealing with one may even exacerbate another. For example, poor relief will simply meliorate discontent and allow opportunities for exploiters to regroup. Only a coordinated broad frontal assault on all connected problems will eradicate social ills. Here is the germ of the idea that alleviating social distress adds up to a prescription for social revolution.

196

THE REPUBLIC OF SCIENCE

Popper's counter-argument is, as we saw earlier, that social revolution is a prescription for chaos and disorder and an open door for those who take over to impose new forms of injustice and exploitation. 7 Furthermore, for Popper avoidable suffering is real and factual, whereas explanations and nostrums are theories, and theories need to be tested. The claim that two or more social problems are linked is also a theory and also needs to be tested. The tests are not purely tests of truth; they are also to explore whether the proposed remedies make the situation worse. Some social problems we declare to be more urgent than others, depending upon our assessment of the amount of avoidable suffering involved. An attempt to meliorate suffering in one place at a time will help us see if the less urgent suffering elsewhere is connected - if it increases then we must tum our attention to a problem newly moved to the top of the agenda. However untidy and even risky such a piecemeal approach seems, Popper argues (a) that any alternative approach is much more risky; and (b) that despite holistic rhetoric, this is in fact how we always proceed. The broad frontal assault is a misleading metaphor. Social reform differs from a battle front. We always give priority to one action or another. If we do not proceed cautiously and self-critically we can conceal our priorities from ourselves and act irresponsibly by our own lights. In the course of his critique of Marx, Popper endeavours to show that not only do his conclusions not follow from his premisses, but, worse, that Marxism inspires some of its followers to behave in ways that make democracy unworkable, thus facilitating the rise of fascism. His shocking charge is not new, indeed between the wars leftists sometimes argued that fascism should be encouraged because its advent would speed the

7. In his great book The New Class, Milovan Djilas (1957) showed how Stalinism was objectively the entrenchment of a new exploiting class recruited pseudo-meritocratically and enjoying the traditional perks of the ruling class it had usurped. In some ways the new class has more power to exploit because its regime is lawless and the rulers' power is therefore effectively boundless. This point is forcefully made in Solzhenitsyn 1973.

LEARNING INSTITUTIONS

197

revolution. What it sped instead was the destruction of democracy and of much of the left. Spurning the totalitarianism of the right and the left, Popper tries to specify the alternative, the open society which cherishes freedom, democracy, and equality. A society in which science can flourish. Popper says democracy needs a number of conditions to work, which he specifies thus: (l) General elections are an insufficient condition for democracy, as the majority may become tyrannical or elect tyrants. "If the men in power do not safeguard those institutions which secure to the minority the possibility of working for peaceful change then their rule is a tyranny" (OS&IE, II, p. 150/161). (2) Democracies have protective institutions for minorities, all other forms of government which lack them are tyrannies. (3) A consistent democratic constitution should exclude only change that would endanger its democratic character. (4) "In a democracy, the full protection of minorities should not extend to those who violate the law, and especially not to those who incite others to the violent overthrow of democracy" (OS&IE, II, p. 150/161).8 (5) Institutions should always be framed on the assumption that anti-democratic tendencies are latent among the ruled and the rulers. (6) If democracy is destroyed all rights are destroyed; such as survive are only on sufferance. 8. Given the possibility of unjust laws, the first part of this provision is controversial. On the other hand, to announce that breaking the law was permitted provided only that the breaker consider it unjust, would offer a general licence and excuse to law-breakers. It is difficult to see anything other than an ad hoc approach as possible here, one in which a substantial body of opinion and doctrine has to hold that the law in question is arguably unjust. Basic or consitutionallaws might thus seem to be what we should be mandated to obey, but the possibility of unjust elements in a constitution (or its interpretation) must also be allowed for, and so the same caveat applies. In brief, a seriously defective legal system jeopardises democracy. This is a strong argument for keeping the legal system under constant review.

198

THE REPUBLIC OF SCIENCE

(7) Since democracy permits reform without violence, its preservation is always the first order of business. It is notable that Popper equates democracy with institutional

machinery that protects freedom. Freedom is for him the value that trumps democracy and equality. Marxists, by contrast, treat democratic systems as class tyrannies no less than despotisms: this encourages blaming democracy rather than particular democrats for unprevented evils; encourages thinking of the state as belonging to the rulers rather than to the ruled; encourages the idea of seizing complete power, when the paradox of power enjoins restricting and counterbalancing power. These Marxist errors are found in all radicalisms and calls for revolution. Many of the enemies of freedom and reason have convinced themselves that they are its 'true' friends. Politically, of course, Popper disagrees with those who are opposed to freedom. But the thrust of The Open Society and Its Enemies is less to argue with those political opponents and more to try to show allies who consider themselves true friends of freedom that in thinking about what ails society they are misled by some deep-seated and plausible mistakes. Freedom is not only in danger from new incarnations of Ghengis Khan, but more avoidably from those with seemingly the best of intentions and deeply flawed ideas: Marx and his followers especially. Apart from the general purgative of fallibilism, the more specific remedy Popper offers is a vision of free society as a fragile entity for which we, its citizens, are responsible. Our efforts to improve it should take account of what we know of the general learning process. The way we learn from experience is by trial and error, conjecture and refutation. Conjectures are putative truths, not established truths; responsibility requires that before attempting to implement a conjecture about improving society we critically evaluate it against current knowledge by thought experiment. In order to detect the possibility of refutation we must strive to test one conjecture at a time. Our means of testing are institutional initiatives of design or reform. Institutional creations or changes are crystallised conjectures, putative social

LEARNING INSTITUTIONS

199

knowledge, and society is a knowledge-storing and a knowledgegaining structure. Popper has used the social character of science to illuminate knowledge as a social institution; and knowledge as a social institutions to illuminate the public character of scientific knowledge. 3. Knowledge as a Social Institution In the course of a critique of the sociology of knowledge in ch. 23 of OS&IE, Popper raises the question, which he had discussed sociologically in LScD, of in what the objectivity of science consists. He argues that the self-described 'sociologists of knowledge' failed to present a sociological explanation, since they conceive of scientific knowledge as in the mind or 'consciousness' of individual scientists. 9 They then devise arguments to the effect that, consciousness being a social or cultural product, science is socially or culturally determined. Such a sociology of science is a Marxist elaboration of Bacon's theory of prejudice. It being impossible to avoid all prejudice, what passes for knowledge are sophisticated reifications of the prejudices of the knower's age, nation, culture, or class, the emphasis depending upon the particular sociologist of knowledge. Only by attributing science to a class of deracinated intellectuals can it be exempted from this argument. Popper argues against this doctrine of prejudice that, if it were true, scientific objectivity would be impossible, even incomprehensible. Individual scientists, like the rest of us, are the 'sites' of various predispositions (for example towards their own ideas), prejudices (for example those of their time or training), and of political partisanship (for example loyalty to parties within their scientific organisations). They can be self-critical, but such inner dialogue is an intemalisation of true dialogue with others that is much more powerful. Scientific objectivity does not dwell in the minds or consciousnesses or unique class position of scientists. As argued in The Logic of Scientific Discovery, we must look for its source in social institutions. Objectivity is a 9. It is standard in analytic philosophy to identify knowledge with justified, true, belief

200

THE REPUBLIC OF SCIENCE

property or rather an outcome of certain social institutions, or arrays or systems of institutions, not of individual minds or attitudes. Scientific method emphasises the organised cooperation of many scientists in a common project, as follows. Scientific suggestions are open to free criticism, including attempts at replication, by other scientists. Scientists make a great effort or, rather, their institutions foster efforts, to avoid talking at cross purposes, to forge a common and even international language. Such communicative arrays permit experience institutionalised in method to do its work as impersonal arbiter. Since the community is clear about what is proposed, it can also be clear in submitting it to the test of nature, and be clear about the results. Proposal, test, and outcome are also open to free criticism by the scientific community (and, sometimes, by outsiders). In an endeavour to clarify his sociological notion of objectivity, Popper asks whether there could be 'revealed science'. Suppose a clairvoyant were to produce, by automatic writing, a book identical to one produced years later by a great scientist. Would the clairvoyant's book be science? Scientists at the time of its production would have found the clairvoyant's writing partly incomprehensible and partly fantastic. Not produced in accord with the scientific method, the results contained in such a book would be 'revealed science'. The label is unimportant. What is important is that the clairvoyant's work would not be taken seriously because it was· not under the auspices of the institutions of scientific method. Moving closer to reality, Popper brings up the thought-experiment of Robinson Crusoe, imagined to be stranded on his island with various scientific equipment for physical, chemical, and astronomical experiments, and time on his hands. He produces and writes down new science far ahead of his contemporaries, results which, discovered among his papers, are found to have anticipated much that came later. Crusoe, unlike the clairvoyant, used the scientific method - or what he remembered of it. Popper argues that this Crusonian science is still 'revealed science'. His point is that only a small part of scientific method consists of rules which the unaided individual can implement: Crusonian science has not been exposed to the critical challenge

LEARNING INSTITUTIONS

201

of the cooperating scientific community, and so the results are accidental or miraculous, like the clairvoyant's; Crusoe's writings would not have been honed on trying to explain the work in an agreed scientific language. His astronomical results would depend on his discovering his own 'reaction time'; but he could not have discovered that without other observers to check it against, so his research must be treated as 'revealed': it may be said that what we call 'scientific objectivity' is not a product of the individual scientist's impartiality, but a product of the social or public character of scientific method; and the individual scientist's impartiality is, so far as it exists, not the source of but rather the result of this socially or institutionally organised objectivity of science (OS&IE, II, p. 208/220).

This argument is the culmination of a social philosophy of science, and a scientific philosophy of the social that Popper began articulating in The Logic of Scientific Discovery ten years before. Marx's idea about the social determination of consciousness is taken to mean not that the content of science is determined by social factors but that the impartiality or objectivity of science is determined by social factors. The clairvoyant is a powerful example; that of Robinson Crusoe even more troubling. Crusoe lacks all social context, yet we can imagine that he goes on writing and speaking his native language and we would not be tempted to think of his notebooks as not being in English because he had for many years been denied its social context. It is of course possible that Crusoe will have used English idiolectically, but it is not a difficulty that seems serious. We assume later readers can and will decode any such entries in the notebooks, as they usually can with entries written in private code. Why then is the same not true of Crusoe's scientific results? Crusoe is equipped both physically and mentally. There is a possibility that he will make errors, both of substance and of method. The absence of social institutions to detect and check those errors would only render his efforts unsatisfactory as science if we had science to take their measure. We learn from errors. With no social peers to detect them, we cannot learn. Yet

202

THE REPUBLIC OF SCIENCE

in criticism of Popper's argument it can be said that there are cases of hermit-like or otherwise solitary contributors to science, most notably Gregor Mendel. Indeed, we know that during the voyage of The Beagle Darwin was a kind of Robinson Crusoe who used his stock of mental and physical equipment to explore and to question the contradictory evidence in the natural world for which he sought explanation in a manner he and we consider paradigmatically scientific. Reading new publications by his teacher Lyell, sporadic correspondence with him, and discussions with the ship's Captain and others are no substitute for scientific society. Mendel and Darwin had: nobody but himself to check his results; nobody but himself to correct those prejudices which are the unavoidable consequence of his peculiar mental history; nobody to help him to get rid of that strange blindness concerning the inherent possibilities of our own results which is a consequence of the fact that most of them are reached through comparatively irrelevant approaches (OS&IE, II, p. 207/219).

Only by one criterion does each escape doing revealed science: And concerning his scientific papers, it is only in attempts to explain his work to somebody who has not done it that he can acquire the discipline of clear and reasoned communication which too is part of scientific method (OS&IE, II, pp. 207-07/219).

Mendel and Darwin did write in this way but then, in the thought experiment, so did Crusoe. Popper is no doubt correct to hold that a surround of social institutions, traditions, and practice is the optimum condition for the progress of science. His Crusonian thought experiment is, however, too strong: it consigns too much to the dump of revealed science. Like the capacity to use a language, some aspects of scientific method may be internalised and sustained over long periods of isolation. It is sufficient for Popper's point to say that scientific method, being social, can only be learned in social surroundings and cannot be indefinitely sustained in their absence. Over time the unaided individual's grasp of scientific method is liable to decay. Decay is simply one more source of

LEARNING INSTITUTIONS

203

error, and error in itself is no Bad Thing; it is the refusal or indeed the inability to learn from error that is bad. Since for Popper society itself is a learning machine, even those isolated from scientific institutions can benefit from the error-checking. Social surroundings are hardly sufficient, of course. Such endeavours as alchemy, astrology, and racialism were and continue to be practised in social surroundings; they even developed specialised institutions for their cultivation and propagation. Yet they are pseudo-scientific endeavours. By Popper's lights they are not pseudo-scientific results or ideas, for that would be a category mistake. Alchemical, astrological, and racial ideas are either false or unfalsifiable. If the former, they are part of the history of science and hence uncontroversially scientific; if the latter, they are extra-scientific and mayor may not enjoy intimate relations with the history of science. Alchemy, astrology, and racialism are examples of pseudo-scientific methods rather than pseudo-scientific results. Their methods were the wrong kind of social institutions. Instead of the ideas being subordinate to falsificationist methodological rules, their partisans utilised the methodological rules of dogmatic schools. The aim of dogmatic schools is the care and preservation of a doctrine, including development of it and its cult, and defence of it from dissenters. Such partisans do not readily submit to the regime of the republic of science. Whether or not dogmatic schools are a threat to the overall liberalism of the society, and not just to its scientific part, is a contingent matter. Today's companies of alchemists, astrologers, and racialists, being small, are relatively comfortably tolerated. In former times they were near-dominant, and probably were a brake on scientific progress. Fascist and communist dogmatisms illustrate the dangers inherent in militant, dogmatic, pseudo-scientific schools. The history of Europe is littered with examples of the struggle between the scientific and liberal spirit and the demand for hegemony of dogmatic schools of religion. With this critical clarification of revealed and Crusonian science in hand we can continue our investigation of Popper's ideas on the social conditions of knowledge. The 'sociology of knowledge' accepted an individual rather than a social model of

204

THE REPUBLIC OF SCIENCE

science, and tried to explain false ideas sociologically while envisaging science as the work of those who are somehow free from all ideological background prejudices. In his discussions of the sociology of knowledge Popper's main source and target was Karl Mannheim, later his colleague at the L.S.E. Mannheim was inconsistent on where ideology ended and science began. Later practitioners of the sociology of knowledge, adherents of the socalled 'Strong Programme', ironed out the inconsistency by treating all thought systems as ideology, science and their own works included. Science becomes just another set of social institutions. Popper's view is that this underestimates the power of a social discovery: the rules of the empirical method are an institution for creating ideas of a different calibre from all other thought systems. Although deeply entrenched, presuppositions (and ideology) do not vitiate all our scientific thinking. Einstein's scientific questioning of our most general presuppositions (about space and time), shows that the empirical method is not crippled by ideology; that even the deepest presuppositions can be exposed, criticised, and overcome. The effectiveness of the empirical method is that it proceeds piecemeal, it does not try to eradicate all prejudice at one go. Scientific theories depend upon our experiments and our prejudices (08&IE, II, p. 209/221). No theory is immutable; all theories may, in due time, yield to criticism. Scientific results are relative to our state of knowledge at a given time and so are liable, in due course, to be superseded. Scientific theories are hypotheses liable to revision at any time. But it certainly has to be admitted that, at any given moment, our scientific theories will depend not only on the experiments, etc., made up to that moment, but also upon prejudices which are taken for granted, so that we have not become aware of them (although the application of certain logical methods may help us to detect them). At any rate, we can say in regard to this incrustation that science is capable of learning, of breaking down some of its crusts. The process may never be perfected, but there is no fixed barrier before which it must stop short. Any assumption can, in principle, be criticised. And that anybody may criticise constitutes scientific objectivity (OS&IE, II, p. 209/221).

LEARNING INSTITUTIONS

205

Popper suggested that the social sciences badly needed to learn the scientific method of free criticism - a common language, advancing hypotheses in a spirit of trial and error, and testing them by piecemeal social engineering. Only the test of practice would cure the irrationalism in social science and avoid scholasticism. By contrast, the sociologist of knowledge naIvely suggests that we can be rid of prejudices directly and so is burdened with the most ingrained prejudices. Only practical actions that lead to democratic institutions that guarantee freedom of critical thought and the progress of science will make any headway with this situation. Only in an intellectual atmosphere of free exchange and criticism can we expose our prejudices, one by one. Claiming that the revolt against reason may be the most important intellectual and even moral issue of our time, Popper extends his Crusonian example to language and to argument. He suggests that Robinson Crusoe could not invent language or the art of argumentation, which are intrinsically social, even though we may develop the internal analogues of talking and arguing with ourselves. "Thus we can say that we owe our reason, like our language, to intercourse with other men" (OS&IE, II, p. 2131225). Directing our intellectual attention to the argument rather than the person allows us to see everyone we meet as a potential source of argument and reasonable information and thus actualises the rational unity of humankind. Popper's is an interpersonal, not a collectivist, theory. It repudiates the Platonic view of reason as a faculty. "We not only owe our reason to others; but we can never excel others in our reasonableness to a degree that would establish a claim to authority" (OS&IE, II, p. 2141226). For Popper, egalitarianism is not only a moral demand, it is an epistemic (social) fact. Inegalitarians characteristically qualify the rational unity of humankind. One line is to claim that all are rational but some are more, much more, than others. A stronger claim is that only some humans are rational, the rest are feeble of mind and dependent on their emotions and passions. Such inegalitarianisms, Popper argues, open the way to legitimating criminal attitudes

206

THE REPUBLIC OF SCIENCE

towards those who disagree. For one thing, if a dispute cannot be settled by reason or by the constructive emotions of reverence, love, devotion to a common cause, then the irrationalists effectively hand it over to less elevated emotions or to power. For another thing, the inequality of humanity is not relevant to politics. Equality before the law is a political demand based upon a moral decision and independent of the false theory that all men are created equal.1O Complacent affirmation of irrationalism, by contrast, can lead to collectivist sociology of knowledge, to antiegalitarianism, to dividing people into 'us' and 'them' and ignoring the rational equality and contribution of anyone, and hence to justifications of inequality which are criminal. But of all political ideals, that of making the people happy [out of love] is perhaps the most dangerous one. It leads invariably to the attempt to impose our scale of 'higher' values upon others, in order to make them realise what seems to us of greatest importance to their happiness; in order, as it were, to save their souls. That is to say, it leads to Utopianism and romanticism (OS&IE, II, p. 223-24/237).

Popper makes strong connections between rational attitude, impartiality, egalitarianism, tolerance, our dependence upon social institutions, and piecemeal social engineering. Furthermore, our decision to adopt rationalism implies a common medium of communication and establishes something like a moral obligation to keep up its standards of clarity as a vehicle for argument; to use language to communicate rather than selfexpress. Translatability of the human languages into one another sufficiently accurately to permit dialogue implies the unity of humankind. Insofar as irrationalism decays into dogmatism it is anti-imaginative. The rationalist attitude of, 'I may be wrong and you may be right', demands a real effort of imagination. Consider the social and other woes of the machine age. The ills of the machine age will be cured by better machines, better 10. Gellner notes that what is taken to be a self-evident truth in Thomas Jefferson's 'Declaration of Independence' is in fact "heretical or unintelligible in most other cultures" (1992, p. 52).

LEARNING INSTITUTIONS

207

understanding of machines and their limits, more general ideas about machines, not by yearnings for the lost, machine-free past (mythical anyway). There is a close link between rationalism and humanitarianism, stronger than the link between irrationalism and anti-egalitarian, anti-humanitarian attitudes. 4. Rationality in Science and in Society Now to sum up. Science is the rational enterprise par excellence according to Popper. Its participants aim at the growth of knowledge and act cooperatively to pursue that aim, as well as constantly to improve their cooperation. The human capacity to be rational is what undergirds the unity and equality of humankind. In parallel to his views on the nature of scientific method, Popper, as might be expected, offers a social rather than an individual or psychological view of rationality. He finds rationality not in results but in the methods by which we approach problems and handle results. In offering such a model he is running in the face of almost 2500 years of philosophising, where reason is taken to be a mental faculty, one often contrasted with another, emotion. Popper instead fmds our rationality in our approach, in our policy towards problems and solutions. The question immediately becomes whether the policy can be formulated in a manner that avoids paradox. Popper sharply contrasts two rationalist approaches: uncritical or comprehensive rationalism, and critical or limited rationalism. Comprehensive rationalism is the policy of not being "prepared to accept anything that cannot be defended by means of argument or experience" (OS&IE, II, p. 2171230). This policy principle is clearly inconsistent since it cannot, in its tum, be defended by argument or by experience. Thus an announcement of a commitment to argument and experience flounders on an argument from inconsistency. Generalising, Popper maintains that since "all argument must proceed from assumptions, it is plainly impossible to demand that all assumptions should be based on argument" (loc. cit). It follows that the decision to choose to listen and to give weight to arguments must precede argument. A choice not made on the basis of argument or experience is by definition an irrational choice. So a comprehensive rationalism

208

THE REPUBLIC OF SCIENCE

rests on irrationalism and so is untenable. There is a distant echo here of Popper's argument that conventionalism in philosophy of science cannot be defeated by argument. Irrationalism is logically similar. It cannot be defeated by the rationalist tool of argument: we have to choose to tum away from it. Once we formulate it as a choice, a decision, an action, we look not at the logical situation but at the aims we are pursuing and whether or not the adoption of a rationalist policy will better serve to achieve those aims than some other. From such a policy choice certain consequences flow. Policies need to be translated into institutional form so that they are permanent, and so that they can scrutinised, assessed, and reformed. This in itself may be an argument that weighs heavily in our decision to opt for rationalism. There may also be other arguments, some of a moral nature. If we hold that "I may be wrong and you may be right, and by an effort, we may get nearer to the truth" (OS&IE, II, p. 213/225) then maximum inclusiveness is the most efficient use of resources, and our equality is seen to lie in our potential to contribute. It is hard to envisage any other policy, and especially not a decision to choose irrationalism, as favouring these values. In all of this Popper is proposing that we weigh the policy decision by drawing out the consequences of a proposed decision and seeing whether we find them consistent with our other aims and policies, and whether they further our aims in the most efficient way. Drawing out the consequences is a logical exercise and so logic and argument seem to be presupposed. Since comprehensive rationalism is inconsistent and comprehensive irrationalism is not, complacent and over-claiming rationalists can easily get hung by their own petard. Instead of capitulating to irrationalism, Popper suggests a minimum concession: a decision to engage in argument precedes argument, hence is unargued, hence may be called irrational. This is Popper's "minimum concession to irrationalism" (OS&IE, II, p. 219/232) which has been characterised by Bartley twice, once correctly as an exaggeration (it is a concession from the classical not the critical rationalist viewpoint); and once incorrectly as fideist, no less, i.e. there is a rational excuse for irrationalism in the failure of comprehensive rationalism (Bartley 1984, p. 215n). 'Excuse',

LEARNING INSTITUTIONS

209

perhaps; but not a very convincing one) 1 No inference to irrationalism is possible from the failure of comprehensive rationalism because it and irrationalism are not mutually exclusive and exhaustive alternatives. All there can be is a choice. Choices can be made irrationally or rationally - provided we do not read the last occurrence of 'rationally' to mean 'comprehensively rational'. A rational choice means a choice influenced by the balance of argument and experience; it is not determined by those things; other considerations are not excluded. Rationality, then, is a choice of policy, a fundamental policy for conduct of social life. It is less a mental act than a mode of cooperation with others, a rudimentary institution. It gives rise to other social institutions, institutions which teach us to argue and which enjoin us to argue before making further decisions. We might conceive of rationality as the most general constitutional provision of scientific method and interpret Popper as social philosopher as offering it as the most general constitutional provision for the open society itself. In view of the fact that many opponents of rationalism consider it a cold and inhumane ideal (cf. Feyerabend 1970, 1995) compared to, say, love or justice, it may be difficult to accept it as a founding principle for a society to which we expect people to be loyal, to the extent of defending it with their lives. Why would people suffer and even die for rationalism? Apart from the historical fact that many philosophers of the seventeenth and eighteenth century found this thought thrilling because of the equality and emancipation that rationalism entailed, we may also argue as follows. We already find people willing to sacrifice for freedom, democracy, and the rule of law. All three are facets of a choice of rationality over irrationality. Freedom includes freedom from the arbitrariness of irrationalism; democracy is an attempt to 11. In a path-breaking analysis Bartley (1961) showed that the 'minimum concession' was unnecessary. He showed that reason need not be chosen on faith, but in a critical spirit. By Popper's own principles reason was a critical attitude, not a justifying one. If reason was chosen in a tentative and openminded way then it was not chosen irrationally.

210

THE REPUBLIC OF SCIENCE

give concrete shape to a regime of rational discussion before decisions are made about laws and public policies; the rule of law is an attempt to crystallise the social framework as a whole into one that enjoins a rational approach to the relation between citizens and the society as a whole. Rationalism underlies all these traditional values, values which have been proclaimed as worthy of sacrifice on innumerable occasions, and this has not been found absurd; far from it. That being so it seems to me that a society that has made a decision for rationality is one more than worthy of the loyalty of its citizens, and if necessary their sacrifice. What irrationalists sense is that the ideal offered is strictly secular rather than transcendent. Rationalism as an ideal is this-worldly rather than other-worldly. Dogmatisms, above all religious dogmatisms, including secular communism, demand loyalty to ideas, to a known truth, something grander than mere people and their calculations of advantage and efficiency. Rationalism offers instead a regime that is self-critical and self-correcting, openminded rather than dogmatic, secure in learning rather than in possessing the truth, and which suggests these attributes are more than sufficient for loyalty and identity. On the empirical question of whether the citizenry can do without the props of irrationalism, which continues to flourish and even in some cases to undergird the open societies of the world, the jury is out. Before discussing some of these matters further in the final chapter, a word about them as critical of Popper's early, skeletal, sociology of science. It goes without saying that if society in general is a learning machine then science functions as a template for it. If society in general can be stricken by the afflictions of nationalism, collectivism, dogmatism, and irrationalism, then so can institutional science. If responsibility and vigilance towards the design, staffing, and maintenance of the institutions of the body politic are vital, then the same goes, pari passu, for science. It is not sufficient that the social surround of science should be a liberal society. Liberal society can tolerate within it dogmatic institutions. It is also necessary that we cultivate attitudes of responsibility and vigilance towards science, and oversee the design, staffing, and maintenance of the institutions that keep it

LEARNING INSTITUTIONS

211

honest. The obvious model is the self-governing professions, lawyers, doctors, architects, etc. However, the scattered and above all the international character of science offer no obvious organisational parallel. Yet the power and wealth devolved on science invite abuse and corruption. Perhaps all of this is alarmist. Just because scientific institutions are hard to control, it does not follow that they are out of control. Perhaps they have already evolved their own ways of staying honest. After all, science is not doing too badly in its work, and its scandals are the exception not the rule. Something along these lines is argued by Michael Polanyi and we shall begin with his alternative view of the republic of science in the next and final chapter.

Chapter 6 CONCLUSION: THE REPUBLIC OF SCIENCE The argument of this book is almost complete. I have tried to show how Karl Popper discovered that some philosophical questions about science resolved into social questions, namely decisions and their institutional implementation. On my reading, his extension and application of his ideas to the social sciences are best read as criticisms of his own earlier social thinking. His criticisms were powerful, but he did not push them through. It seems to me that at this point lowe the reader a broader perspective on these criticisms, one that helps us see where they could have been taken. Of all the problems surrounding science considered socially, it seems to me the most basic are those concerning governance. To exploit one of my central metaphors, the most basic questions concern the nature and provisions of the Constitution of the Republic of Science, including its relation to the Constitution governing the whole polity. Not wanting to begin another book on the subject (Fuller 1997, 2000a), I shall proceed to throw light on Popper's ideas by comparing them with those of Michael Polanyi. Karl Popper and Michael Polanyi are, in my view, the two pre-eminent republicans in the philosophy of science. Using conventional political labels, Polanyi articulates a conservative and traditionalist outlook; Popper a reformist and social democratic one. Both were sharply critical of the political radicals in science in the nineteen-thirties and 'forties. Polanyi endorsed the status quo in science because he could find no competent higher authority (Polanyi 1958, 1961, 1962). Self-regulation is the only option for science and hence is as good as it is going to get. His arguments are interesting and insightful; his conservative

CONCLUSION

213

conclusions open to challenge. In particular, he inconsistently exempts science from the application of his own values of freedom and democracy. Popper's first moves towards a social account of science (LScD) simplified and beautified it. Later, when working out his critical and democratic social philosophy (1944-45, 1945), he condemned complacency towards social institutions, and so, tacitly, condemned complacency towards institutionalised science. Yet he also held up science - beautified as Socratic seminar - as a model for rational conduct in social matters. In not working out the implications for science of his critical and democratic ideas, Popper faltered. His beautification of science is inconsistent with his critical philosophy of institutions. In comparing Popper's republican model with that of Polanyi my focus will be on these overlapping questions: should science be exempt from the democratic control of the wider society as Polanyi maintains? (§1); is critical and democratic reform of institutionalised science possible along Popperian lines? (§2); is real existing science a model for society in general along either Polanyiite or Popperian lines? (§3); and, is science just one more social institution, or, as both Polanyi and Popper claim, does its pursuit of truth make it special? (§4)

1. Democratic and Authoritarian Republics Polanyi's principal argument for exempting science from the democratic control of the wider society relies on his special view of scientific authority. Polanyi found lessons for democratic society in the social relations of science (1961, 1962). He combines reliable ethnographic reports with controversial social and political analysis. Popper, being sceptical of all authority, offers arguments to rebut Polanyi. He wrote little about science as social institution and offered no alternative ethnography; however, he did work out a political philosophy that can be applied to the issues at hand. Polanyi beautifies and elevates science: in the free cooperation of independent scientists we shall fmd a highly simplified model of a free society, which presents in isolation certain

214

THE REPUBLIC OF SCIENCE basic features of it that are more difficult to identify within the comprehensive functions of a national body (1962, 54).

Polanyi describes science as "coordination by mutual adjustment of independent initiatives" much like a group working on a large jig-saw puzzle. Every puzzler keeps in mind the pieces held, the elements of the picture so far assembled, and remains watchfully aware of the progress of others. The jig-saw puzzle metaphor harbours a difficulty. Unlike the picture cut up by the jig saw, the world-picture is not a prior given. 1 Polanyi gets round this difficulty by arguing that sClenfific work is guided by an 'invisible hand' towards "a joint result which is unpremeditated by any of those who bring it about" (1962, 55). There being no end in mind or to hand, progress can only be stepwise, each step being decided at the time by the most competent members of the profession. Any attempt to centralise decisions would violate such local competence. Self-coordination of independent initiatives by the most competent "assures the most efficient possible organisation of scientific progress" (1962, 56). Polanyi draws an explicit parallel with the market, where, he holds, price functions in the same way as does taking note of published results functions in science. He also adds a qualifier: making efficient use of money in the market is very different from responding to new intellectual situations "motivated by current professional standards" (1962, 56). For example, choice of problems in science, Polanyi writes, is influenced by the plausibility, scientific value, and originality of their prospective solutions. Plausibility and scientific value tend to favour conformity, while originality encourages dissent. The professional standards of science must impose a framework of discipline and at the same time encourage rebellion against it. They 1. The metaphor of the jig saw puzzle is misleading in a host of ways. A contemporary epistemologist, Susan Haack, explicates knowledge by means of the crossword puzzle. She acknowledges one of its misleading aspects, there being no analogue to checking one's solution against that of the setter (Haack 1993, p. 214). A much deeper and more misleading disanalogy she actually celebrates: its justificationist character whereby correct answers interlock and confirm one another.

CONCLUSION

215

must demand that, in order to be taken seriously, an investigation should largely conform to the currently predominant beliefs about the nature of things, while allowing that in order to be original it may to some extent go against these. Thus, the authority of scientific opinion enforces the teachings of science in general, for the very purpose of fostering their subversion in particular points (1962, 58-9).

Expertise is distributed over individuals, authority lies in the network as a whole. Each scientist has an area of local competence, but also some competence in overlapping areas. The network of overlapping competencies covers all the domains of science. The network is the seat of scientific opinion. Scientific opinion is an opinion not held by any single human mind, but one which, split into thousands of fragments, is held by a multitude of individuals, each of whom enforces the other's opinion at second hand, by relying on the consensual chains which link him to all the others through a sequence of overlapping neighbourhoods (1962, 59-60).

Newcomers affiliate to a network that extends far beyond their horizon. This is much the way "artistic, moral or legal traditions are transmitted", from master to apprentice: "the authority of science is essentially traditional" (1962, 69). Following the model of the free market as closely as he can, Polanyi views science as a "highly simplified model of a free society" (1962, 54). What concerns me most is Polanyi' s claim that science stands apart; it needs to be protected from political and economic interference, and must in tum protect itself from the incompetent, such as "cranks and dabblers" (1962, 61). Polanyi might be described as an authoritarian with a human face. Scientific institutions enforce legitimate traditional authority. They consist of self-organised bodies of experts, "masters". His stress is on network and diffusion; the powers of hierarchy and leadership are played down. Elaborating on his giant jig-saw puzzle metaphor, Polanyi writes that cooperation can advance only stepwise, and ... total performance will be the best possible if each step is decided upon by the person most competent to do so .. .if each helper watches out for any new

216

THE REPUBLIC OF SCIENCE opportunities arising along a particular section of the hitherto completed patch of the puzzle, and also keeps an eye on a particular lot of pieces, so as to fit them in whenever a chance presents itself (1962,55).

Polanyi sees science advancing "by independent initiatives adjusting themselves consecutively to the results achieved by all the others" (1962, 56). Unlike the jig-saw puzzle, which has a single, complete, solution, scientific endeavours are governed by an "invisible hand", because knowledge is distributed through a wide network and not concentrated in anyone place. The network is a community abiding by standards that set it off from other social endeavours. Unlike lay citizens, not all scientist citizens are equal: some have more authority than others do, especially the senior over the junior. Authority is differentially exercised in training, appointments, refereeing, endorsing textbooks, popularising, and setting school curricula. Most important, there is an authoritative consensus on where there should be expansion and where there should be curtailment of scientific effort. Arguing that attempts to coordinate such changes from outside science must fail, Polanyi stresses how scientific initiative needs seclusion, insulation from the general public (1962, 67). Strong and united scientific authority reassures and gains the support of the public for the independence of science and the unhindered publicising of its results. Such authoritative coordination of individual initiative, despite its flaws, is the only principle of governance that could have promoted the unprecedented advance of science in this century. If the wider society wants flourishing science it should leave it alone. Polanyi warns us that scientific authority may be used to suppress evidence that contradicts current views, as well as to welcome discoveries that deeply modify those views. Nevertheless, no outside authority is competent to oversee scientific authority. The Republic of Science is a Society of Explorers. Such a society strives towards an unknown future, which it believes to be accessible and worth achieving. In the case of scientists, the explorers strive towards a hidden reality, for the sake of intellectual satisfaction ...

CONCLUSION

217

A free society may be seen to be bent in its entirety on exploring selfimprovement - every kind of self-improvement. This suggests a generalisation of the principles governing the Republic of Science. It appears that a society bent on discovery must advance by supporting independent initiatives, coordinating themselves mutually to each other. Such adjustment may include rivalries and opposing responses which, in society as a whole, will be far more frequent than they are within science. Even so, all these independent initiatives must accept for their guidance a traditional authority (1962, 72).

Originality is the quality that keeps the whole structure honest, for original work may subvert currently accepted ideas. Just how it is that incremental subversion ever becomes revolutionary subversion is as obscure in Polanyi as in his follower Kuhn (1962).2 If one sets aside reverence for tradition and authority, especially authority that inculcates hierarchy and suppresses nonconformity, then critical and democratic control of science by the wider society seems reasonable enough. Furthermore, Polanyi's view draws some of its strength from the idea that what is or is not science is unproblematic from a philosophical and from an institutional point of view. On this, the so-called 'demarcation problem', Popper is less sanguine. Polanyi's demarcation is clear and crisp, if not always palatable. Science is what institutionalised scientific authority recognises as such. Candidate work is assessed for its plausibility against current science, for its scientific value (i.e. its accuracy, systematic importance, and intrinsic interest in the eyes of current science), and, finally, for its originality. This amounts to saying that the current institutions of science themselves decide what they will admit and what they will exclude (or suppress). Clearly, such a demarcation will not work politically. Science cannot be insulated from the wider society and at the same time be extended the total freedom to decide its own boundaries. Quite intentionally, Popper gives at best a partial solution to the problem of demarcation, utilising a logical criterion, 2. Polanyi is silently endorsing the historiographic continuity theory of Duhem (see Agassi 1963).

218

THE REPUBLIC OF SCIENCE

falsifiability. For it to work, he argued, it has to be institutionalised, and institutions, he observes, are in principle imperfect. The conclusion is obvious: falsifiability, and hence demarcation, are ideals. Borderline cases and ambiguities are bound to arise. Still further ambiguity must arise because Popper allows that, historically, science has emerged from a background of myth and uncontrolled speculation. Compounding the issue is his concession that to some extent or other magical and tabooistic thinking are present in modem scientific society. We might sum up by saying that, for Popper, demarcation questions need critical debate that can only ever achieve temporary closure. Demarcation, like science itself, is conjectural. This being so, it is debatable just where the writ of scientific leadership runs, just how distinct science is or can be from the rest of society, just how vague it is to exempt science from critical and democratic control. Most philosophers of science, Popper included, have demarcated science by its method; in Popper's version, the scientific citizenry's submission to methodological rules. Polanyi does not assign any role to methodological rules, whether of induction or of falsification. Provocatively, he compares the learning of science to the learning of the arts: the method of scientific inquiry cannot be explicitly formulated and hence can be transmitted only in the same way as an art, by the affiliation of apprentices to a master. The authority of science is essentially traditional (1962, 69).

The effect of this is to make science institutionally esoteric: only its practitioners can understand it. The wider society is asked to extend to science privileges without scrutiny, simply to go by results. By contrast to Polanyi, Popper offers no arguments for excluding science from critical and democratic control. Polanyi's apparatus of dispersed networks, hierarchical structure, and the authority of tradition are very different from Popper's Socratic seminar and methodological rules. Polanyi's are developed and

CONCLUSION

219

structured; Popper's are at best proto-institutions) We fmd further contrast in the different consequences each draws for power, and reform. In the skeletal sociology of LScD (1959) Popper used the analogies of games and jury trial to show how institutions serve to coordinate the attempts of individuals to realise the aim of scientific progress. The coordinating rules (methodology), we might say, become institutional elements of his Constitution for the Republic of Science. In his later work Popper fleshes out his view of institutions as devices that, through language, make possible the selves we call individuals, and the articulation of the projects we call aims. These aims emerge in a situation bounded by a few factors: the physical environment, other individuals, institutions, and the articulated (possibly institutionalised) ideas about all of the above. Difficulties due to contradictions in the articulated ideas constitute problems, their surround constitutes the problem-situation. Rational action consists in a critical appreciation of a problemsituation and proposing solutions that are to be tested step by step. Note how abstract all of this is. Popper does not specify any institutional structure. New difficulties arise when we consider the external relations of science to the wider society. Polanyi and Popper share the idea of science as a model institution, a miniature of society, but differ greatly over the details. Popper has no equivalent to the demand for science being esoteric and set apart. Polanyi makes two points about relations to the wider society. One is that science is a special kind of institution that embodies a special kind of authority. The other is that this institution and its authority cannot be controlled or planned by outsiders. Any attempt to do so will fail, and possibly damage the institution beyond repair. For Polanyi, scientific authority is the authority of competence, expertise, and their transmission from master to apprentice. The test of success is not solely instrumental, it is also philosophical:

3. They contribute to what John Hall has labelled the "sociological deficit" of The Open Society (1999). It is also a political deficit since a Socratic seminar, viewed politically, tends towards a self-governing anarchy.

220

THE REPUBLIC OF SCIENCE for the last three hundred years the progress of science has increasingly controlled the outlook of man on the universe, and has profoundly modified (for better and for worse) the accepted meaning of human existence. Its theoretic and philosophic influence was pervasive (1962, 67).

In return for exemption from outside control science exports influential new ideas about the world we all inhabit. Indeed, unless it is self-governing, science will be subject to the attempts of political or religious authorities to interfere. Polanyi's Republic of Science thus is not under the control of its host society, indeed it demands autonomy from it. He compares science to a republic that enjoys extraterritorial status within society in general: The soil of academic science must be extraterritorial in order to secure its control by scientific opinion (1962, 67).

What is Popper's view of how the Republic of Science interacts with the wider republic? He discusses this question very little if at all. He offers the metaphor of science as a structure built over sand, with piles driven down far enough to support its present load and what we can anticipate given by our present limited foresight (Popper 1959, 111). The locus of decision and issues of cost and control are ignored. But his ideas can be applied, as we shall see in what follows. Popper's overall concern is to maximise rationality in the management of society and in general. He wanted us to view institutions as instruments to achieve our aims as humanely and with as few constraints on freedom as possible. His conservatism is not the kind that confers authority on what is established, but rather of the kind that demands, before things are changed, that we try to assess whether we have enough knowledge to know what we are doing, i.e. avoid making things worse. Social change has human costs, and he wants those minimised. Intellectual change has social costs also, but Popper does not pursue that implication. On the contrary, he sometimes wrote as if fighting with words is cost-free compared to fighting with swords. Of

CONCLUSION

221

course this is correct in the immediate case, but not in a wider perspective. In politics Popper calls for a universally critical attitude from autonomous citizens. Does he apply this to science? Only indirectly: Popper's social and political ideas imply a critique of concretely existing science by the standards of the wider polity in which it is embedded. His link between science and social reform is the idea that institutions are hypotheses: conjectural attempts to solve problems. Science is granted no special exemption from maximising rationality by means of criticism, debate, and reform. The source of a criticism or proposal is irrelevant, only its validity matters. I conclude that Polanyi's case for a hands-off approach to science is weak. 2. Rational Reform of Science Is rational reform of scientific institutions possible along Popperian lines? Polanyi thinks not, since he restricts or denies the values of open discussion, transparent process, and a critical attitude to institutions. He treats authority and hierarchy as simply how things are, without discussing improvement. That is left to insiders; it is not the concern of the wider society. Issues of corruption and abuse of power are best handled internally. Popper, by contrast, nowhere says that science's problems are solely science's business. In fact he has a quite general philosophy of institutions. "Institutions," he wrote, "must be well designed and well manned" (OS&IE, I, p. 110/126). Applied to science as Socratic seminar this would highlight structure and leadership. A Socratic seminar led by Thrasymachus would not much resemble Popper's critical ideal. So led, it could easily decay into a dogmatic school, the opposite of science on Popper's view. Again, a Socratic seminar could not flourish in a hostile society. The design of the institutions of science can hardly ignore the design of other institutions or the structure of relationships between science and the rest. It follows that reform cannot be carried out purely internally. As social institution, science is subject to all the normal failings. Problems of governance are politically primary and upon their solution depend the possibilities of rational reform. Some

222

THE REPUBLIC OF SCIENCE

scientific institutions have considerable status, power, and influence. Given that science in general is socially important, and that it receives public monies, is there not a reasonable expectation of public scrutiny and accountability? Issues of scientific fraud, for example, may need experts to make certain kinds of assessment, but their discovery and consequences, including reforms due to such cases of fraud, are of wide social concern. Where there is power there will be abuse of it; where there are rewards there will be corruption. When science was semi-private or small-scale its failings were of little concern to outsiders. The present prominence of science, and our dependence upon its technological fallout, alters the situation. Science is too important to escape the twin necessities of structural subordination to the governance of the wider society, and adopting an internal structure of governance that assures accountability within and without. To expand: part of being a free and open society is obedience to the rule of law. No institution can be exempt from that rule. Science benefits from the order and freedom brought by the rule of law. Indeed, the civility of civil society is one of the preconditions of effective scientific work. It thus makes no sense to exempt science from democratic control within and without; its authorities can and should be challengeable not just from inside but from outside as well. Polanyi defends scientific autonomy with the argument that an essential feature of science, its method, cannot be articulated. It can only be transmitted hands-on, from master to apprentice. The masters have competence and in this lies the legitimacy of their authority. No other authority is competent. This picture of the master/apprenticeship relationship is incoherent. Imagine a situation in which the local item of science that the neophyte overthrows is the work that elevated her master to that status. Can that master go on taking apprentices? Is he still to be regarded as competent? Since there is no articulate content to what the master transmits to the apprentice, it is difficult to differentiate errors that call competence into question from honest mistakes. It is true that most of us need teachers, but autodidacts exist, in science as elsewhere (I have in mind Schrodinger's

CONCLUSION

223

mastering of biochemistry in mid-life (Schr6dinger 1944, Moore 1989). The contributions of autodidacts and other outsiders throws into question the ineluctability of Polanyi' s apparatus of hierarchy, class divisions, and rigid exclusion of cranks and dabblers. That in tum allows that valuable reforms may come from outside. Popper's approach at least permits raising the question, 'is there value in others criticising the judgment of the scientific leadership?, Reforms aiming to help science to live up to its own standards might initially be handled internally, but the record is not promising. Consider again reforms to do with science's relation to the wider society. For example, under war conditions, scientists agreed to do a lot of research under government secrecy controls. Which group has more competence, to use Polanyi's standard, when it comes to judging what needs to be shrouded? I doubt if there is a definitive answer. In practice, many scientists accepted the authority of representatives of the wider republic. If, as a body, in the name of openness and democracy, scientists were to refuse to work without more control over their own situation, doubtless they could achieve it. Instead, the bureaucrats can lift clearance so that scientists may become officially unable to possess copies of, or read, their own papers. What has the scientific leadership done about such absurd overrides of their competence? Was it galvanised, as on Polanyiite lines it should have been, to paralyse government research until its own competence was respected? It was not. Yet perhaps that was wise. Whole areas of scientific endeavour were put under political and social control by the wider society. The competent accepted this for the sake of their solicitations of money. Big Science depends upon big flows of money. So organised science has much work to do just to realise the Polanyiite values of maximising openness and sharing discoveries. Although he was aware of such objections, Polanyi simply reiterated the special character and irreplaceability of the scientific leadership. In my view, the leadership could benefit from outside counsel. Science education is a topic that straddles any demarcation between science and the wider society. All rational reform of

224

THE REPUBLIC OF SCIENCE

scientific education shows up another incoherence in Polanyi's master/apprentice view. Experts and the wider society have overlapping responsibilities in education. Science education has faults such as its isolation from arts education. Thus science education can produce students without any history of science or sense of the wider impact of science on the arts or the outlook of citizens. One minor result of all this is that the culture of scientists is somewhat Philistine. Less minor, the public relations of science systematically confuse it with technology and the practical, with knowledge and authority, rather than with risk, speculation, and the immense but rewarding difficulty of discovering new knowledge. Worst of all, children are told early that they have no capacity for science. Thus it is far from clear that the best talent is recruited, and it is very clear that many intelligent and cultivated citizens are alienated from science. Hence they treat it with too little and too much respect, and never dream that they can, perhaps should, influence it. Reform of this situation involves too many social interests to be handed over to the "competent."

3. Science as Socially Exemplary

Is real existing science - viewed along either Polanyiite or Popperian lines - a suitable model for society in general? Both Polanyi and Popper think so. The values they both claim for science - freedom and cooperation - are uncontroversial. But Polanyi defends privilege, hierarchy, and authority not for society in general, rather for science in particular. An obvious problem for his view is that the competent authorities may be mistaken and their authority endorse error. Polanyi admits this danger and sees it as the price of excluding cranks and dabblers. From a wider social and political point of view it might be too high a price for an impossible gain: citizenship in a free society cannot exclude cranks and dabblers. In order for any society to be, indeed, free and open the threshold of competence for participation must be set very low. Polanyiite science is a closed society, not an open one. Popper's deficiencies on this question stem from his failure to apply his social and political ideas to science. These ideas can

CONCLUSION

225

correct his own tendency to beautification. Popper holds that prudential citizenship in an open society includes a vigilant attitude towards institutions. It was his stress on criticism, and his view that science institutionalises the critical attitude, that led him to recommend science as a model for the open society. Polanyi and Popper agree on two further features of science that strike me as unsuitable for society in general. These are its anomic character (A); and its taking truth as an aim (B). (A) According to both Popper and Polanyi the society of science is abstract; its citizens need meet rarely if at all, and need share no ties, no overlapping system of roles beyond science. In the second edition of The Open Society and Its Enemies Popper notes that there are human needs that can only be satisfied in concrete or face to face society. Hence, if an open society is to satisfy human needs it cannot be purely abstract; it must have concrete aspects and institutionalised aims other than the pursuit of truth. There must be affective bonding, family structure, civil society, friendship, and so on, to make life tolerable, to make science worthwhile. By Popper's own arguments, science can only be a part of social life, not the model for all of it. The role of scientist can only be one amongst several roles that a citizen plays, and conduct appropriate in one role set is not necessarily appropriate in another. Ergo, an abstract association is not sufficient as a model for social life in general. It is also not necessary, since human social life long antedates science or anything like it. (B) What about the social consequences of pursuing the truth? Polanyi lauds the way science has altered our world view. Popper also affirms this when he suggests that no educated person should be ignorant of science. Yet truth, like death, is no respecter of person, status, or hierarchy.4 Truth may demote the master to the status of apprentice-to-a-former-apprentice. This is feasible within an abstract institution. Is it feasible in the family, the law, politics, friendship, religion? These, alas for the claim that science is socially exemplary, are aspects of society from 4. I am supplementing Popper with some points made by Gellner who argues that truth is no respecter of custom, of convention, of culture, of religion (1992,71-72).

226

THE REPUBLIC OF SCIENCE

which we draw much comfort, orientation, and even identity. There is no knowing where pursuit of truth will lead. The entire vision of humanity in the universe can be upset. At this point the search for truth has yielded a cold, comfortless vision, in which neither each one of us, nor human life in general, has any meaning whatsoever. Humanity has come into existence, and it will go out of existence. It is likely that it will leave no trace in a universe conjectured to be vast in time and space (Gellner 1987, 152-65). Under the impact of the scientific world view traditional beliefs can be exposed as false; traditional expertise can be exposed as worthless; traditional institutions can be shown to be harmful, even self-defeating. Far from always being socially positive, the pursuit of truth can be socially destructive. This reasoning led Gellner (1992) to a curiously Polanyi-like compromise view that we might seek to insulate the elements that give us comfort, roughly, the cultural, from the depredations of truth. We would need to reject the feasibility of Descartes' project to dispense with custom and example and begin again from next to nothing. We would try to have the cozy comforts of our cultural world co-exist in apartheid with science. The social stability of Gellner's solution is debatable, as he was the first to acknowledge. Neither Popper nor Polanyi thought through the social consequences of the pursuit of the truth. Science as viewed by both Popper and Polanyi is a goal-directed activity. Its theories are devised to solve problems. Its network institutions may be just grown; its hierarchical institutions were certainly designed to achieve its goals. Human social life in general is not, however, in that sense, a goal-directed activity. It may even be the case, as Popper claims, that the vast majority of social institutions are just grown; if so, the purposes they serve may be inchoate. Among the grown institutions are, of course, language, myths, legends, and folk beliefs - all of which differ from society to society. Except in some broad Social Darwinist perspective, they are, despite differences, more or less equally functiona1. 5 When we 5. Gellner says that they come before the Big Ditch (the Scientific Revolution), and that whichever set of them a society adopts makes little difference to its viability (1974, 149-167).

CONCLUSION

227

replace grown institutions with designed ones we articulate aims; when we replace grown outlooks with coherent, problem-solving ones, we may well fail to replace the non-directed aspects of the originals. This constitutes a further argument against science as a model for society in general. Some social activity, such as solving a problem, can benefit from the rational or critical or scientific approach. But we have to go case by case. There are well known cases of where the social disruption of axes, new seeds, or irrigation projects is more severe than the 'problems' they came to solve. East End slum clearance in London after the Second World War seemed a worthy end. The result, people marooned among strangers in faulty high rises, suffering both anomie and the rapid creation of new slums, illustrates the multidimensionality of social problems, and the difficulty of isolating them for 'solution' (Young and Willmott 1957). For simplicity I have framed the argument technologically. But I reiterate that pure science is just as socially problematic. Berkeley found central ideas of Newton to be a threat to religion. Since his time debates about the threat scientific ideas pose to religion, morality, hierarchy, tradition, and so on, have grown exponentially. From these arguments it follows, I claim, that the social institutions of science are far from being exemplars for society in general.

4. Science Among the Institutions

Is science just an institution like all the others or, as Polanyi and Popper claim, is it special? Like them, I hold it to be special as a means for coordinating efforts to approach the truth, at which it has not been ineffective. Yet I would also concede that science is indeed a normal social institution in a host of a ways. It has a structure of roles, and it has functions, some of which were handed to it, to some of which it arose in response, some of which it has despite anyone's intention. The same goes for aims. People engage in scientific research for many reasons. In so far as science is a set of grown rather than designed institutions, it is without a predetermined aim. As a set of institutions of long standing it has associated traditions and statuses. Science and its claims are social facts, like any other social facts.

228

THE REPUBLIC OF SCIENCE

Let me now try to point to features of science that would restore the plausibility of its traditional exceptionalism; its claim to be able, with great effort, to transcend the social and economic background, to overcome the weight of prejudice, of custom and example, and to formulate statements that in some way or another capture how the world is. On this transcendence Polanyi and Popper are agreed. Science alters our overall vision of the world and of our place in it and claims this to be progress, that is, to be a better, truer vision of the way things are. On a second point they are also agreed: fallibilism. Scientific findings are formulated as statements, but each statement is understood in context to be tentative. That is, scientific claims are always revisable. Changes to, revisions of, extensions to, and alterations of, its pronouncements are a permanent feature of science. When Polanyi endorses the authority of science he does not endorse this, that, or the other current judgment of science; he endorses the institutional authority of science to monopolise pronouncing on those questions it has appropriated. In addition to agreement on the superiority of the scientific vision and its revisability, Popper and Polanyi agree that science employs the power of institutional cooperation to probe the truth about the world. It does this not by postulating this or that idea as the truth, but rather by showing that things cannot be both this way and that, and to claim so would be inconsistent. When we choose to say that one side of the contradiction is true and the other false, we are, of course, conjecturing rather than pronouncing final judgment. But we are conjecturing a truth; that is, we conjecture that in our best collective judgment, i.e. science, things are that rather than this. The quality in question is truth, truthlikeness, or closeness to the truth. Truth so understood is a regulative idea. When doing science we do not need to envisage having the whole truth or the final truth to hand. We need not be in this respect holists, rather, piecemealists. Eliminating contradictions is an urgent matter because contradictions are embracing and thus threaten the very possibility of deploying meaningful language, the foundation of social cooperation (Popper 1940, 1943).

CONCLUSION

229

It could be objected that all of this is mere social necessity in linguistic disguise. If human life necessitates social cooperation, if social cooperation requires viable language, and if viable language demands the elimination of contradictions, then truth is a social construction, a label for some manoeuvres required to keep society functioning. I see no way to defeat this response, or indeed any necessity to do so. All I will claim is that it is not helpful. Truth is not an empty place-marker for social presuppositions. Rather, truth is the aim that governs attempts to articulate a conception of the relation between the organism and its environment. No organism before the language-using animal could do this. It is a powerful notion: our ideas about the environment, conjectural reactions to the various ways we bump against it, can approximate it better or worse. That is, seemingly benign environments can be lethal; safe ones unsafe; dangerous situations not dangerous at all. We try to express this by envisaging a perspective that is outside or in front (what Quine mocks as cosmic exile): we look back at the organism and its means of mapping and coping, and we assess its success. For example, the organism thought it lived on a flat surface and acted accordingly. That turned out to be a mistake, but not a lethal mistake, merely a confining one. We can view truth as the outcome of attempts to generalise this insight: as organisms we conjecture our environment all the time in order to cope with it, to act. Some of the time we fail, we are surprised. We conclude that the environment was not the way we conjectured it to be. If it was not the way we conjectured it to be, then it was some other way. We need no assumption of final or complete truth. We proceed, as Polanyi said, stepwise, working on a jig-saw puzzle with the hope that parts of it make some sort of picture from time to time. The final product, the (uncompletable) jig-saw completed, is what we call truth. Popper's original notion of truth was modest, perhaps too modest. At the end of his LScD he says that all notions of truth share the same notion of falsity and that this is sufficient for the logic of science. It is a superb idea. In such a modest notion of truth that regulates science there is no finality, and hence no particular judgment is endorsed or enforced. The content of

230

THE REPUBLIC OF SCIENCE

'trying to get it right' is almost exhausted by the injunction to avoid contradictions, though not quite. Science thus conceived resembles institutions like schools, universities, libraries, which do not stand or fall with their current contents. It is perhaps no accident that these institutions are affiliates of science and properly seek only the authority to teach and transmit, not the authority of what is transmitted. This differentiates them from religious seminaries, schools of thought, and political parties. The authority of science is less like that of a party and more like that of a legislature. Legislative authority is qualified: current laws must be obeyed. Legislation, however, is conventional and coercive. Science eschews coercion for reason, and uses conventions to transcend convention: it discovers errors in our conventional representations of the world, thereby providing tantalising glimpses at how things may be. Therein lies its transcendence. I want to suggest one final feature that sets science off from other social institutions, and which is not unconnected to its exceptional social status. This I will call its universalistic aspiration. The authority of science, unlike the authority of other institutions, is not bounded by national jurisdiction. What is truth on this side of the Pyrenees remains truth on the other side, whether or not those over there are ready to admit it. This is where science, as is often remarked, resembles the catholic, monotheistic religions. To view it as just another evangelical church WOUld, however, be most misleading. It seeks not converts, but trainees; demands loyalty not to doctrine, but to process (which therefore had better not be inarticulable if its errors are to be detected); it rewards autonomy and initiative, not conformity and obedience, pace Polanyi. Of course, all sorts of institutions are international, including families, extraterritorial laws, international institutions, international law, and other legal instruments. Viewed internationally, science resembles these institutions. What differentiates science is that it seeks transcendental truth while denying that it possesses it. Not unlike universalistic religions, science makes an institutional and standing claim for freedom from local institutions, and even from specific (national) scientific

CONCLUSION

231

ones, as these embody prejudice. This remains a powerful aspiration even if is not currently fashionable to endorse it. Moreover, what is transcended by science is not this world and this life. Rather, what is transcended are cultures and human differences. It is hard to know how these aspirations, which are evidently not so very unsuccessful, can be reconciled with fashionable attempts to explain them as the product of local connections like this or that class structure, in this or that political economy at a particular time. Science is distinct amongst institutions in being an organised attempt to transcend what Descartes dismissed as the local truths of custom and example and to replace them with universal, secular truths that correct and replace those of local origin. Science as institution is its own primary reference group, and its membership over time and space is diverse along all social and economic parameters. Here lies the true extraterritoriality of science. Just how such an international organisation should be structured and governed and just what degree of special treatment should be extended to it remain, therefore, questions on the agenda.

SOURCES AND REFERENCES All primary source documents are in the Archives of the Hoover Institution, Stanford University. They are cited as "PP Hoover" with a box number/file number, e.g. "329/13". The finding aid (Register) for those papers accessed in 1985 is posted on the web at . At the time of going to press, the balance of the papers acquired after Popper's death in 1994 was still in the process of being registered. BIBLIOGRAPHY Abse, Joan, ed. 1977. My LSE. London: Robson Books. Ackerman, Robert. 1976. The Philosophy of Karl Popper. Amherst, MA: University of Massachusetts Press. Agassi, Joseph. 1956. The function of interpretation in physics. Ph.D. dissertation, University of London, unpublished. ------------------. 1960. Methodological individualism. British Journal of Sociology 11:244-70. ------------------. 1963. Towards an Historiography of Science. History and Theory, Beiheft 2. Reprint, Middletown, CT: Wesleyan University Press 1967. ------------------. 1971. Faraday as a Natural Philosopher. Chicago: University of Chicago Press. -------------------. 1972. Sociologism in philosophy of science. Metaphilosophy 3: 103-22. -------------------. 1975. Science in Flux. Dordrecht: Reidel. -------------------. 1977. Towards a Rational Philosophical Anthropology. The Hague: Mouton. -------------------. 1978. Liberal forensic medicine. Journal of Medicine and Philosophy 3:226-41.

REFERENCES

233

-------------------. 1985. Technology: Philosophical and Social Aspects. Dordrecht: Reidel. -------------------. 1988. The Gentle Art of Philosophical Polemics. La Salle: Open Court. -------------------. 1993. A Philosopher's Apprentice. Amsterdam: Rodopi. ------------------. 1995/96. Blame not the laws of nature. Foundations of Science 1:131-54. ------------------. 1997. Truth, trust, and gentlemen. Shapin on Boyle. Philosophy of the Social Sciences 27(2):217-30. ------------------. 1998. Science real and ideal: Popper and the dogmatic scientist. Proto Sociology. 12:297-304. ------------------. 2000. Liberal Nationalism for Israel. Jerusalem and New York: Gefen. ------------------ and I. C. Jarvie, eds. 1987. Rationality: The Critical View. The Hague: Martinus Nijhoff. Artigas, Mariano. 1999. The Ethical Nature of Karl Popper's Theory of Knowledge. Bern: Peter Lang. Bacon, Francis. 1620. Novum Organum. In Spedding and Ellis 1857-1874. -------------------. 1624. New Atlantis. In Spedding and Ellis 1857-1874. Barnes, Barry. 1974. Scientific Knowledge and Sociological Theory. London: Routledge. ------------------. 1977. Interests and the Growth of Knowledge. London: Routledge. ------------------. 1982. T.S.Kuhn and Social Science. New York: Columbia University Press. Bartley, III, W. W. 1962. The Retreat to Commitment. New York: Knopf. -------------------------. 1984. The Retreat to Commitment (2nd edition). LaSalle, IL: Open Court. -------------------------. 1989. Music and Politics. A short extract from Rehearsing a Revolution. Karl Popper: A Life. Prepared for the Pacific Regional Meeting of the Mont Pelerin Society, Christchurch, New Zealand, November 27-30. Berkson, William and John Wettersten. 1984. Learning from Error. La Salle, IL: Open Court.

234

THE REPUBLIC OF SCIENCE

Bittlinger, Peter A. 1984. Government Enforcement of Morality: A Critical Analysis of the Devlin-Hart Controversy. Ph.D. dissertation, University of Massachusetts 1975. Ann Arbor, MI: University Microfilms. Blanshard, P. 1952. Freedom and Catholic Power. London: Seeker. Blaug, Mark. 1980. The Methodology of Economics. Cambridge: Cambridge University Press. Bloor, David. 1976. Knowledge and Social Imagery. London: Routledge. -----------------. 1981. The strengths of the strong programme. Philosophy of the Social Sciences 11:199-213 Bunge, Mario. 1999. The Sociology-Philosophy Connection. New Brunswick: Transaction. ------------------. 2000. Ten modes of individualism - none of which works - and their alternatives. Philosophy of the Social Sciences 30(3):384-406. ------------------, ed. 1964. The Critical Approach to Science and Philosophy. New York: The Free Press. Burke, T.E. 1983. The Philosophy of Popper. Manchester: Manchester University Press. Cartwright, N. 1983. How the Laws of Physics Lie. New York: Oxford University Press. Currie, Gregory and Alan Musgrave, eds. 1985. Popper and the Human Sciences. The Hague: Nijhoff. Descartes, Rene. 1637. Discourse on Method. In, The Philosophical Works of Descartes, trs. John Cottingham, Robert Stoothoff and Dugald Murdoch. Cambridge: Cambridge University Press 1985, vol. 1. Dewdney, A.K. 1997. Yes, We Have No Neutrons. New York: Wiley. Djilas, Milovan. 1957. The New Class. New York: Harcourt Brace Feyerabend, Paul. 1970. Against method: outline of an anarchistic theory of knowledge. In M. Radner and S. Winokur, eds., Analyses of Theories and Methods of Physics and Psychology. Minneapolis: University of Minnesota Press, pp. 17-130.

REFERENCES

235

-----------------------. 1995. Killing Time. Chicago, IL: University of Chicago Press. Frazer, J.G. 1890. The Golden Bough. London: Macmillan. Fuller, Steve. 1996. Social epistemology and the recovery of the normative in the post-epistemic era. Journal of Mind and Behavior 17(2):83-98. ----------------. 1997. Science. Minneapolis, MN: University of Minnesota Press. ----------------. 2000a. The Governance of Science. Buckingham: The Open University Press. ----------------. 2000b. Thomas Kuhn: A Philosophical History for Our Times. Chicago: University of Chicago Press. Gellner, E.A. 1959. Words and Things. London: Gollancz. -----------------. 1974. Legitimation of Belief. Cambridge: Cambridge University Press -----------------. 1981. Muslim Society. Cambridge: Cambridge University Press. Goldman, A. I. 1992. Liaisons. Cambridge, MA: MIT Press ------------------. 1999. Knowledge in a Social World. Oxford: OUP Gombrich, E. H. 1995. The Open Society and Its Enemies: Remembering Its Publication Fifty Years Ago. Discussion Paper no. 11. London: LSE Centre for the Philosophy of the Natural and Social Sciences. Reprinted in Jarvie and Pralong, eds., 1999, pp. 17-27. Grove, J.W. 1982. The sociological denigration of the rationality of science. Minerva 20(3-4):550-56. Haack, Susan. 1993. Evidence and Inquiry. Oxford: Blackwell. Hacohen, Malachi. 1993. "The Making of the Open Society: Karl Popper, Philosophy and Politics in Interwar Vienna". PhD dissertation, Columbia University. Ann Arbor: University Microfilms. -----------------------. 1996. Karl Popper in exile. Philosophy of the Social Sciences 26(4):452-92. -----------------------. 2000. Karl Popper - The Formative Years, 1902-1945. Politics and Philosophy in Interwar Vienna. Cambridge: Cambridge University Press.

236

THE REPUBLIC OF SCIENCE

Hall, John A. 1999. The sociological deficit of The Open Society, analysed and remedied. In Jarvie and Pralong 1999, pp.83-96. Hamowy, Ronald. 1996. Review of Hayek on Hayek. Philosophy of the Social Sciences 26(3):417-21. Hattiangadi, J.N. 1978-79. The structure of problems, I and II. Philosophy of the Social Sciences. 8:345-65; 9:49-76. ----------------------. 1983. Three contemporary problems of methodology and a methodology without methodological rules. In R. S. Cohen and M. W. Wartofsky, eds., Language, Logic and Method, Dordrecht: Reidel, pp. 103-51. ----------------------. 1995. The first world war and 1991. In N. Laor and I. C. Jarvie, eds. Critical Rationalism, the Social Sciences and the Humanities. Dordrecht: Kluwer, pp. 59-78. Hausman, Daniel M. 1992. Essays on Philosophy and Economic Methodology. New York: Cambridge University Press. Hayek, F.A. 1935. Collectivist Economic Planning. London: George Routledge. -----------------. 1941a. The counter-revolution of science. Economica 8:9-36;119-150;281-320. -----------------. 1941b. Planning, science and freedom. Nature 143:580-84. -----------------. 1942-44. Scientism and the study of society. Economica 9:267-91; 10:34-63; 11:27-39. -----------------. 1944. The Road to Serfdom. London: Routledge. -----------------. 1949. Individualism and Economic Order. London: Routledge. -----------------. 1960. The Constitution of Liberty. London: Routledge. -----------------. 1967. Studies in Philosophy, Politics and Economics. London: Routledge. Horgan, John. 1996. The End of Science. Reading, MA: Addison-Wesley. Horton, Robin. 1993. Patterns of Thought in Africa and the West. Cambridge: Cambridge University Press. Indart, Rafael. 1999. "Karl Popper's Philosophy of Social Sciences and the Problem of Tyranny." Ph.D. thesis, York University, Social and Political Thought. Unpublished.

REFERENCES

237

Jarvie, I.C. 1963. Theories of cargo cults: A critical analysis I and II. Oceania 34:1-31;108-36. ----------------------.1964. The Revolution in Anthropology. London: Routledge. ----------------------. 1966. On the explanation of cargo cults. Archives Europeennes de Sociologie 7(2):299-312. ----------------------. 1970. Cargo cults. In Man, Myth and Magic. No. 15. London: British Publishing Corporation, pp. 409-12. ----------------------. 1972a .. The Story of Social Anthropology. New York: McGraw Hill. ----------------------. 1972b. Concepts and Society. London: Routledge. ----------------------. 1982. Popper on the difference between the natural and the social sciences. In Levinson, ed., 1982, pp. 83-107; reprinted in Jarvie 1985 as ch. 5. ------------. 1985. Thinking About Society: Theory and Practice. Dordrecht: Reidel ------------. 1987. Philosophy of Film. London: Routledge. ------------. 1998. Popper, Sir Karl. In Edward Craig, ed., Routledge Encyclopedia of Philosophy. London: Routledge, vol. 7, pp. 533-40. ---------------------. 2001. Workshop rationality and the reasonable persistence of dogmatism. In D.M. Johnson and C. Emeling, eds. The Mind as a Scientific Object: Between Brain and Culture. New York: Oxford. ---------------------. 2002. Fuller on Science. Philosophy of the Social Sciences Jarvie, I.C. and Joseph Agassi. 1979. The rationality of dogmatism. In Th. Geraets, ed., Rationality Today. Ottawa: University of Ottawa Press, pp. 353-62. Jarvie, I. C. and Sandra Pralong, eds. 1999. Popper's Open Society after 50 Years. London: Routledge. Johnasson, Ingvar. 1975. A Critique of Karl Popper's Methodology. G6teborg: Akademif6rlaget. Keuth, Herbert, ed. 1998. Karl Popper: Logik der Forschung. Berlin: Akademie Verlag. Kroeber, A.L. 1917. The superorganic. American Anthropologist. 19:163-213.

238

THE REPUBLIC OF SCIENCE

Kuhn, Thomas S. 1962. The Structure of Scientitific Revolutions. Chicago: University of Chicago Press. Kuper, Adam. 1999. Culture. The Anthropologists' Account. Cambridge, MA: Harvard University Press. Lakatos, I. 1970. Falsification and the methodology of scientific research programmes. In Lakatos and Musgrave 1970, pp. 91-195. -------------. 1974. Popper on demarcation and induction. In Schilpp, ed., 1974, pp. 241-73. Lang, Andrew. 1901. Magic and Relgion. London: Longmans. Levinson, Paul, ed. 1982. In Pursuit of Truth. Atlantic Highlands: Humanities. Magee, Bryan. 1971. Modern British Philosophy. London: Secker. -------------------. 1973. Popper. London: Fontana. Mannheim, Karl. 1936. Ideology and Utopia. London: Routledge. -----------------------. 1940. Man and Society in an Age of Reconstruction. London: Routledge. Merton, Robert K. 1936. Puritanism, pietism, and science. Sociological Review 28:1-30 -------------------------. 1938a. Science and the social order. Philosophy of Science 5:321-37 -------------------------. 1938b. Science, Technology and Society in Seventeenth Century England. Bruges: St. Catherine Press. -------------------------. 1942. Science and technology in a democratic order. Journal of Legal and Political Science 1:11526. (Also known as Science and democratic social structure, and as The normative structure of science.) -------------------------. 1973. The Sociology of Science. Chicago: University of Chicago Press. Miller, David. 1994. Critical Rationalism: Restatement and Critique. La Salle: Open Court. ------------------. 1999. Being an absolute skeptic. Science 284 (5420): 1625-26. A longer version 'Sokal and Bricmont: Into the Frying Pan' is at

REFERENCES

239

Moore, Walter. 1989. Schr6dinger. Life and Thought. Cambridge: Cambridge University Press. Morgan, P.E.D. 1985. Chemical reincarnation? Bulletin of the American Ceramic Society 64:929. Munz, Peter. 1982. Transformation in philosophy through the teaching methods of Wittgenstein and Popper. In The Search for Absolute Values and the Creation of the New World. Proceedings of the Tenth International Conference on the Unity of the Sciences, 9-13 November 1981, Seoul, Korea. New York: International Cultural Foundation, vol. 2, pp. 1235-62. ---------------. 1985. Our Knowledge of the Growth of Knowledge. London: Routledge. ---------------. 1997. The Quixotic element in The Open Society. Philosophy of the Social Sciences 27:39-55. Musgrave, Alan. 1999. Essays on Realism and Rationalism. Amsterdam: Rodopi. ---------------------. and Gregory Currie, eds. 1985. Popper and the Human Sciences. The Hague: Nijhoff. Newton-Smith, W. 1982. The Rationality of Science. London: Routledge and Kegan Paul. O'Hear, Anthony. 1980. Karl Popper. London: Routledge. ----------------------. 1989. Introduction to the Philosophy of Science. Oxford: Oxford University Press. O'Neill, John, ed. 1973. Modes of Individualism and Collectivism. London: Heinemann. Papineau, David. 1995. Open society, closed thinker. Times Literary Supplement #4812 (23 June), pp. 4-5. Pipes, Richard, ed. 1996. The Unknown Lenin. New Haven: Yale University Press. Polanyi, Michael. 1947. The Logic of Liberty. Chicago: University of Chicago Press. -----------------------. 1952. The stability of beliefs. British Journalfor the Philosophy of Science 3:217-37. -----------------------. 1958. Personal Knowledge. London: Routledge and Kegan Paul. -----------------------. 1962. The republic of science: its political and economic theory. Minerva 1:54-73.

240

THE REPUBLIC OF SCIENCE

Popper, Karl R. 1935. Logik der Forschung. Vienna: Springer Verlag. -------------------. 1940. What is dialectic? Mind 49:403-26 -------------------. 1943. Are contradictions embracing? Mind 52:47-50 -------------------. 1944-45. The poverty of historicism, I, II, III. Economica 11:86-103; 119-37; 12:69-89. -------------------. 1945. The Open Society and Its Enemies. London: Routledge and Kegan Paul. -------------------. 1947. Logic without assumptions. Proceedings of the Aristotelian Society 67: 251-92. -------------------. 1957. The Poverty of Historicism. London: Routledge and Kegan Paul. -------------------. 1958. Back to the Presocratics. Proceedings of the Aristotelian Society 59: 1-24; reprinted in Popper 1963 as ch.5. -------------------. 1959. The Logic of Scientific Discovery. London: Hutchinson. -------------------. 1963. Conjectures and Refutations. London: Routledge and Kegan Paul. -------------------. 1970. Normal science and its dangers. In Imre Lakatos and Alan Musgrave, eds. Criticism and the Growth of Knowledge. Cambridge: Cambridge University Press, pp. 51-58. -------------------. 1972. Objective Knowledge. Oxford: Oxford University Press. -------------------. 1976. Unended Quest. La Salle: Open Court. -------------------. 1983/4 Postscript to the Logic of Scientific Discovery. London: Hutchinson. ------------------. 1989. Popper on Schrodinger. Nature: 342 (23 November):337. -------------------. 1994. The Myth of the Framework. London: Routledge. Posner, Gerald. 1993. Case Closed. Lee Harvey Oswald and the Assassination of JFK. New York: Random House. Public Service Alliance of Canada. 1985. Keep our customs inspectors. Keep out drugs and porno. Broadsheet.

REFERENCES

241

Renfrew, Colin. 1992. World languages and human dispersals: a minimalist view. In John A. Hall and I. C. Jarvie, eds., Transition to Modernity. Cambridge: Cambridge University Press, pp. 11-68. Rhees, Rush. 1947. Social engineering. Mind 56:317-337. Ruben, David-Hilel. 1985. The Metaphysics of the Social World. London: Routledge. Rudwick, M.J.S. 1985. The Great Devonian Controversy. Chicago: University of Chicago Press. Russell, Bertrand. 1911. The Problems of Philosophy. London: Williams and Norgate. ----------------------. 1920. The Practice and Theory of Bolshevism. London: Allen and Unwin. Sakharov, Andrei. 1990. Memoirs. New York: Knopf. Salmon, Wesley. 1967. The Foundations of Scientific Inference. Pittsburgh, PA: University of Pittsburgh Press. Santayana, George. 1916. Egotism in German Philosophy. New York: Scribner's. Schilpp, P.A. ed. 1974. The Philosophy of Karl Popper. LaSalle: Open Court. Schr6dinger, Erwin. 1944. What is Life? Cambridge: Cambridge University Press. Schroeder-Heister, P. 1984. Popper's theory of deductive inference and the concept of a logical constant. History and Philosophy of Logic 5:79-110. Shapin, Steven. 1994. A Social History of Truth. Chicago: University of Chicago. Shaw, P.D. 1971. Popper, historicism, and the remaking of society. Philosophy of the Social Sciences 1(4):299-308. Shearmur, Jeremy. 1979. Abstract institutions in an open society. In Wittgenstein, the Vienna Circle and Critical Rationalism. Proceedings of the 3rd International Wittgenstein Symposium. Vienna: HPT, pp. 349-54. ------------------------. 1980. The religious sect as a cognitive system. Annual Review for the Social Sciences of Religion 4:149-63. 1985. Epistemology socialized. Etc. , 42(3):272-82.

242

THE REPUBLIC OF SCIENCE

------------------------. 1996. Popper's Politics. London: Routledge. ------------------------. 1998. Popper, Hayek, and the Poverty of Historicism Part I: A largely bibliographic essay. Philosophy of the Social Sciences 28:434-50. Simkin, Colin. 1993. Popper's Views on Natural and Social Science. Amsterdam: Rodopi. Smith, Anthony D. 1971. Theories of Nationalism. London: Duckworth. Second edition 1996. Solzhenitsyn, Aleksandr I. 1973. The Gulag Archipelago 19181956, I-II. Trs. Thomas P. Whitney. New York: Harper and Row. Soble, Alan. 1995. In defense of Bacon. Philosophy of the Social Sciences 25: 192-215. Spedding, J.R.L. Ellis, and D.D. Heath, eds. 1857-1874. The Works of Francis Bacon, vols. I-XIV. London: Longmans. Stokes, Geoffrey. 1998. Popper. Oxford: Polity. Talmon, J.L. 1952. The Origins of Totalitarian Democracy. London: Secker. Taylor, Charles. 1958. The Poverty of the Poverty of Historicism. Universities and Left Review. Summer, 77-78. Urbach, Peter. 1987. Francis Bacon's Philosophy of Science,' an Account and a Reappraisal. LaSalle, IL: Open Court. Vickers, Brian, ed. 1996. Francis Bacon. Oxford: Oxford University Press. Watkins, J.W.N. 1952. Ideal Types and Historical Explanation. British Journal for the Philosophy of Science 3 :22-43 ----------------------. 1957. Historical explanation in the social sciences. British Journalfor the Philosophy of Science 8:10417. ----------------------. 1964. Confirmation, the paradoxes and positivism. In M. Bunge, ed., 1964, pp. 92-115. ----------------------. 1974. The unity of Popper's thought. In P. A. Schilpp, ed., vol 2, pp. 372-412. ---------------------- 1977. Chapter in Abse, ed., My LSE, pp. 65-82. ----------------------. 1995. Sir Karl Popper at LSE. LSE 5(2) October.

REFERENCES

243

Wettersten, John. 1992. The Origins of Critical Rationalism. Amsterdam: Rodopi. Wilkins, Burleigh Taylor. 1978. Has History Any Meaning? A Critique of Popper's Philosophy of History. Hassocks: Harvester. Wisdom, J.O. 1952. The Foundations of Inference in Natural Science. London: Methuen. ------------. 1970. Situational individualism and the emergent group properties. In R. Borger and F. Cioffi, eds., Explanation in the Behavioural Sciences. Cambridge: Cambridge University Press, pp. 271-96. ------------. 1993. Philosophy of the Social Sciences. Aldershot: Avebury. Wittgenstein, L. 1922. Tractatus Logico-Philosophicus. London: Routledge. Young, Michael and Peter Willmott. 1957. Family and Kinshiop in East London. London: Routledge.

INDEX OF NAMES n indicates that the reference is in a note Ackerman, Robert, 32, 36n Agassi, Joseph, 7, 32, 33, 36n, 47n, 54, 62, 81, 86, 96n, 101, 129n, 141, 154n, 158, 163n, 196, 193n, 217n

Anaximander, 118 Anaximines, 118 Aristotle, 24, 119, 134, 141 Artigas, Mariano, 32 Augustine, St., 24 Bacon, Sir Francis, 9, 12 and n, 39 and n, 40 and n, 41, 43, 44, 45 and n, 52, 53, 57n, 192-94, 198, 199 Barnes, Barry, 16 Bartley, III, William Warren, 33, 39n, 89, 208, 209n

Bayes, Th., 51 Becker, Gary, 25 Berkeley, Bishop George, 217 Berkson, William, 32 Bernal, J.D., 176n Bicknell, Jeanette, 7 Bittlinger, Peter A., 163n

Blackett, P.M.S., 176n Blanshard, Paul, 155 Blaug, Mark, 32, 70n Bloor, David, 16 Bolzano, Bernard, 121 Boole, George, 119, 121, 194 Boyle, Robert, 77n Braunthal, Alfred, 89, 140n Briskman, Larry, 33 Bunge, Mario, 32, 119n Burke, T.E., 33, 36n Carnap, Rudolf, lIn, 121 Cartwright, Nancy, 96 and n Comte, Auguste, 117 Coulanges, Fustel de, 24 Craig, Edward, 32 Crossman, R. H. S., 146 Crowther, J.G., 176n Crusoe, Robinson, 192, 20002,205 Currie, Gregory, 32 Darwin, Charles, 47, 202 Descartes, Rene, 24, 106 and n, 125, 226, 231 Devlin, Lord Justice Patrick, 163 and n

INDEX OF NAMES Dewdney, A.K., 55 Dewey, Melvil, 111 Dijksterhuis, E.J., 24n Djilas, Milovan, 168, 196n Duhem, Pierre, 59, 217n Durkheim, Emile, 16 Edison, Thomas A., 133 Edwards, Paul, 32 Einstein, Albert, 76, 83, 135,204 Faraday, Michael, 81, 133 Ferguson, Adam, 159n Feyerabend, 33, 69, 71-73, 76, 209 Fichte, J.G., 186 Fleischman, Martin, 59 Frazer, Sir James G., 54 Frege, Gottlob, 121 Fuller, Steve, lIn, 15n, 17n, 176n, 212 Galileo, Galilei, 134 Gardner, Doug, 7 Gellner, Ernest, 76, 100, 149n, 155, 183, 185, 187, 206n, 225n, 226 and n Gentzen, Gerhard, 120n G6del, Kurt, 121 Goldman, Alvin, 29n Gombrich, E.H., 90n, 140n Grove, J. W., 86n Haack, Susan, 214n

245

Hacohen, Malachi, 7, 33n, 37n, 88,90, 92n, 105, 145 Haldane, J.B.S., 176n Hall, John A., 219n Hamowy, Ronald, 171n Hart, H.L.A., 163 and n Harvey, William, 24 Hattiangadi, J.N.H., 33, 69, 78-81, 186 Hausman, Daniel, 32 Havel, Vaclav, 183n Hayek, F.A., 33, 89 and n, 90 and n, 93, 107n, 126, 140n, 159 and n, 171 and n, 176, 190n Hegel, G.W.F., 15, 90, 140 and n, 141, 145, 189, 183, 185, 186 Heidegger, Martin, 183 Heraclitus, 141, 148n Hesiod, 148n Hobbes, Thomas, 24 Hogben, Lancelot, 176n Horgan, John, 53n Horton, Robin, 75n Hume, David, 125 Indart, Rafael, 97n Jarvie, I.e., 32, 81, 124n, 158 Jefferson, Thomas, 206n Johansson, Ingvar, 32, 69 andn, 70-72, 94-105 Johnson, President Lyndon, 132n

Johnson, Dr. Samuel, 106

246

THE REPUBLIC OF SCIENCE

Kant, Immanuel, 45, 186 Karadicz, R., 183 Kennedy, President John F., 132n

Keuth, Herbert, 32 Khan, Ghengis, 198 Kitcher, Philip, 29n Kroeber, A.L., 24n Kuhn, Thomas S., 12n, 18n,

Mises, Ludwig von, 171 Moore, G.E., 90n, 109n Moore, Walter, 223 Morgan, P.E.D., 54 Morgenstern, Stephanie, 7 Munz, Peter, 7, 32, 47, 76, 90 Musgrave, Alan, 32, 33, 115n

23,43n,50,59,60n,66,

Newton, Sir Isaac, 76, 135, 227 Newton-Smith, William, 32

Lakatos, Imre, 18n, 31, 32, 52n, 59, 70n, 71n, 92 Lang, Andrew, 54 Lenin, V.I., 170n Levinson, Paul, 32 Lowie, Robert, 24n Lyell, Charles, 202 Lysenko, T.D., 175

Oedipus, 103 O'Hear, Anthony, 32, 36n,

67, 69, 73, 74, 78, 217 Kuper, Adam, 24n

Magee, Bryan, 32, 36n Mannheim, Karl, 14-16, 214 Marx, Karl, 14, 15, 31, 90, 123, 129n, 140 and n, 141, 142, 168, 172, 174, 183, 186, 190, 191, 193, 198,199,201 Masaryk, Jan, 183, 186 Meinong, Alexius, 121 Mendel, Gregor, 202 Merton, Robert K., 11-13, 16, 17 and n, 85, 162 Miller, David, 32, 33, 65 Mill, John Stuart, 53, 117, 142, 190, 191

42n,43n

Old Oligarch, 157 O'Neill, John, 102 Papineau, David, 22, 38n Pavlov, I.P., 175 Peano, Giuseppe, 121 Pipes, Richard, 170n Plato, 24, 31, 90, 109n, 113, 140 and n, 141, 16065, 172-74, 180, 183 Poincare, Henri, 59 Polanyi, Michael, 12 and n, 13, 50, 66, 69, 73-76, 77n, 78, 81, 85, 176n, 211, 212-31 Pons, Stanley, 59 Posner, Gerald, 132n Price, Derek da Solla, 77n Pythagoras, 175 Quine, W.V.O., 229

INDEX OF NAMES Quinton, Lord Anthony, 32 Renfrew, Colin, 111 Rhees, Rush, 109n Robespierre, Maximillian, 113 Rousseau, Jean-Jacques, 24 Ruben, David-Hillel, 99 Rudwick, M.J.S., 16,65 Russell, Bertrand, 26, 31, 85,87, 119, 121, 163, 170andn, 183, 185 Sakharov, Andrei, 175 Salmon, Wesley, 42n Santayana, George, 172n Scheler, Max, 14 Schilpp, Paul A., 31 Schlick, Moritz, lIn Schr6dinger, Erwin, 87n, 222-23 Schroeder-Heister, P., 120n Shapin, Steven, 16, 65, 85, 86 Shaw, G.B., 111 Shaw, P.D., 10, 170n Shearmur, Jeremy, 7, 32, 33 and n, 47n, 48, 69, 90, 162n Simkin, Colin, 32 Smith, Anthony, 183 Socrates, 10, 51, 87, 97, 94, 132, 135, 166, 175 Solzhenitsyn, A. I., 196n Spinoza, Baruch, 194

247

Stalin, Josef, 129 Stokes, Geoffrey, 32 Talmon, Jacob, 165 Tarski, Alfred, 100, 119, 121 Taylor, Charles, 100, 129n Teller, Edward, 176 Thales, 118 Thomas, W.I., 124n Thrasymachus, 221 Tudjrnan, Franjo, 183 Urbach, Peter, 39n Vickers, Brian, 39n Virgil,55n Watkins, J.W.N., 31, 32, 129n Weber, Max, 67, 125 Weinberg, Steven, 53 Wettersten, John, 32, 33, 47n White, Leslie, 24n Whitehead, Alfred North, 121 Wilkins, Burleigh Taylor, 32 Willmott, Peter, 227 Wilson, Woodrow, 183 Wisdom, John 0, 32, 129n Wittgenstein, L., 109n Young, Michael, 227

INDEX OF SUBJECTS

n indicates that the reference is in a note; q that it is in a quotation; t that a term is discussed abstract society, see society Academie Fran