The View from Within: Normativity and the Limits of Self-Criticism, Notre Dame, 2011 (with Yitzhak Benbaji)

MENACHEM FISCH AND YITZHAK BENBAJI the view from within N OR M ATI V I TY A N D TH E L IM ITS O F S E L F- C RI T I C I S M Now available from Notre Dame Press . . . THE VIEW FROM WITHIN Normativity and the Limits of Self-Criticism MENACHEM FISCH AND YITZHAK BENBAJI ISBN 978-0-268-02904-3 $50.00฀paperback฀/฀£42.95฀•฀408฀pages Praise for The View from Within . . . MENACHEM FISCH is Joseph and Ceil Mazer Professor of History and Philosophy of Science at Tel Aviv University in Israel. YITZHAK BENBAJI is associate professor on the law faculty and in the philosophy department at Bar-Ilan University, Israel. In the US, order online at: http://undpress.nd.edu/book/P01491 Tel: 800 / 621-2736 or 773 / 702-7000 In the UK, order online at: www.eurospanbookstore.com Further Information Tel: (0)20 7240 0856 Email: info@eurospangroup.com “The View from Within is a thorough evaluation of the arguments made by contemporary philosophers about the normative character of reason and the derivative problem of relativism. Fisch and Benbaji have admirably compared and contrasted competing positions, and with a balanced critique, they have made a sustained effort to ‘save’ rationality and provide new guideposts for its philosophical evaluation. A timely and important contribution.” —ALFRED I. TAUBER, Boston University “A tour de force, this book sheds new light on many areas of philosophy. Indeed, by examining the role of familiar phenomena that philosophers often neglect, such as ambivalence and indecision, The View from Within illuminates the destabilizing as well as the creative potential of reason throughout human life.” —PAUL FRANKS, University of Toronto and Yale University “. . . a fresh and illuminating approach to the nature of rationality and normativity.” —JEFFREY STOUT, Princeton University University of Notre Dame Press • undpress.nd.edu for Hanna and Hagit THE VIEW FROM WITHIN NORMATIVITY AND THE LIMITS OF SELFCRITICISM BY MENACHEM FISCH & YITZHAK BENBAJI ii CONTENTS Chapter One - Setting the Stage: The Problem with Rationality 1. Two Concepts of Rational Action  Rationality's Authority  Second-Order Pondering 2. Normative Diversity: An Introduction  A Two-Step Argument 3. Rorty's Relativism 4. The Road Taken  The Problem 5. The Argument PART ONE: THROUGH THICK AND THIN Introduction Chapter Two - Comparative Irrealism and Community-Based Semantics: Kripkenstein and Beyond 1. Normativity Through Thick and Thin 2. Normative Diversity 3. Comparative Irrealism: The Philosophical Thesis  Could Normative Diversity be a Myth? 4. In Defense of Comparative Irrealism  Kripke’s Rule-Following Paradox  An Argument for Comparative Irrealism 5. Conclusion: Irrealism and Relativism Chapter Three - Factuality Without Realism: Normativity and the Davidsonian Approach to Meaning 1. The Argument 2. Theoretical and Normative Diversity in Davidsonian Theories of Meaning  Interpreting Descriptive Discourse: The Unity of the Observational and the Plurality of the Theoretical  The Factuality of Normative Discourse  Normative Diversity iii 3. Factuality Without Realism  The Idiolect-based Conception of Meaning PART TWO: RATIONALITY FROM WITHIN Introduction Chapter Four - The Limits of Connectiveness: Criticism From Within and the Interpretive Account of Normativity 1. 2. 3. 4. 5. 6. Introduction Truth by Interpretation The Thick and the True The Thin and the Void Comparative Irrealism and the Limits of Normative Comparison Two Objections  The Challenge of Moorian Realism 7. The Interpretive Account and the Open Question Argument  A Digression on Personal Goodness  The Essentiality of the Thick and the Normative Question 8. Taking Stock Chapter Five - Rationality as Agreement: Friedman’s Special Case for Science 1. 2. 3. 4. Introduction From Kant to Kuhn The Problem of Rationality: Carnap, Kuhn, and Conceptual Relativism Science in Context  A Kuhnian Dilemma  The Dual Normativity of Scientific Frameworks  Friedman’s Habermasian Turn  Scientific Revolutions - An Evolutionary Approach  Reasonable Motives 5. An ‘is’ for an ‘ought’ Chapter Six - Toward a Critical Pragmatism: A Brandomian Beginning 1. Friedman, McDowell and Beyond 2. Interpretivism Revisited: Enter Brandom iv  Between Normative Status and Normative Attitude 3. Making Brandom Explicit I: The Constitutive Role of Conceptual Norms 4. Making Brandom Explicit II: The Normative and the Critical  A Brandomian Doubter  Walzer’s Prophets PART THREE: NORMATIVE SELF-CRITICISM Introduction Chapter Seven - The Critical Stance 1. Introduction: Brandom on the Rationality of Norm-Making 2. Making Criticism Explicit  Between Criticism and Doubt  Problems Large and Small  Criticism’s Essential Boundedness  Criticism as Speech Act  Criticism and Self-Criticism 3. The Teleology of Problemhood  Problems Solved and Resolved  A Frankfurtian Caveat  Problems and Progress  The Two-fold Neutrality of Problems 4. The Critical Stance  Criticizing versus Testing  Comrades, Critics, and Strangers Chapter Eight - The Achievement of Self Criticism 1. 2. 3. 4. The Critical Moment Toward a Frankfurtian Analysis of Normative Criticism Normativity and "the Contingent Necessities of Love" Getting Ourselves Wrong: Frankfurt on Normative SelfCriticism 5. "Socializing Harry"  Critical Feedback  The Instrumental Value of Final ends v 6. Within You and Without You: The Self's Normative Dialogue  A Frankfurtian Normative Critic  Self-criticism's transformative moment  Thoughtful Ambivalence. 7. The Achievement of Normative Self-Criticism Chapter Nine – Science Revisited 1. 2. Taking Stock Back to Science I: Meta Frameworks, Trading Zones and Beyond  Practititoners Abroad 3. Back to Science II: The Force of Creative Indecision 4. In Conclusion Afterword and Acknowledgments Bibliography AFTERWORD & ACKNOWLEDGMENTS This book has been long in the making. A commitment to grounding agency in a capacity for normative self-criticism, had permeated our work long before we decided to explore the prospects of joining forces during in the summer of 2001. The initial idea to collaborate was Fisch's, who, as reader of Benbaji's dissertation on moral relativism, suggested that something along the lines of the generalized Popperian perspective he had been working with in the philosophy of science might prove the key to resolving Benbaji's deliberation between external and internal reasons in ethics. Benbaji, by then a postdoctoral research assistant at the Institute for Advanced Study, Princeton, agreed to try. Needless to say, what looked to us then like the outline of an elegant solution, served merely to aggravate the problem. Taking the internalist perspective seriously – along Frankfurtian lines, for instance - seemed to render normative commitment immune in principle to self-critique, certainly in the light of Popper's analytical construal. We swiftly found ourselves forced back to the drawingboard to rethink from scratch the very notions of normative commitment, and critical discourse we were working with. This proved a daunting, yet greatly rewarding task that required of us to seriously engage the work of some of the most interesting and demanding thinkers working in these areas today. The project transformed our teaching, thinking and writing, forcing us to divide the labor while bridging as best we could our own very different styles and philosophical temperaments. But it worked at least to our own, admittedly blinkered satisfaction. In the course of its long gestation, many colleagues and friends have invested time, effort and good cheer in commenting on various aspects and extracts of the project. These include Hagit Benbaji, Yemima Ben Menahem, Akeel Bilgrami, Bill Child, Lorraine Daston, Michael Friedman, Ariel Furstenberg, Noah Efron, Paul Franks, Snait Gissis, Niccolo Guicciardini, Zoe Gotzeit, Oren Harman, David Heyd, Don Howard, Andrew Janiak, Bernard Katz, Ehud Lamm, Menachem Lorberbaum, Avishai Margalit, Adi Ophir, Gideon Rosen, Carol Rovane, Eric Schliesser, Alfred Tauber, Nehama Verbin, and Michael Walzer. Good, hard-hitting and informed criticism, we argued toward the end of Chapter Seven, is one of the greatest gifts a person can give to another, because self-criticism is something none of us are very good at, and have great difficulty self-applying effectively. So thanks, dear friends, 372 for your friendly and invaluable critiques. We have also had the great privilege of receiving two extraordinarily insightful and engaging, yet alas anonymous, referee reports commissioned by UNDP that have contributed decisively to much improving this work. We thank you dearly, kind readers, and only wish we could have done so in person. Overviews of the argument, as well as more focused segments of it were presented at conferences, seminars and colloquia at the universities of Tel Aviv, Bar Ilan, Ben Gurion, Notre Dame, Virginia, Cape Town, and Bergamo, the Hebrew University, Jerusalem, Muhlenberg College, and the Bar Hillel Colloquium, where we enjoyed much lively discussion. We thank the organizers, hosts and participants of those meetings. This book is the result of a truly collaborative undertaking. We each wrote seperately, but read and commented on each other's drafts several times around. Fisch was responsible for the initial drafting and final versions of Chapters 1, 5, 6, 7, 8, and 9; Benbaji, for those of Chapters 2, 3 and 4. Credits An early version of Chapter Two appeared as “Through Thick and the Thin: a New Defense of Cultural Relativism” in the Southern Journal of Philosophy, 42 (1) 2004, 1-24. Chapter Three first appeared as “Factuality without Realism: Normativity and the Davidsonian Approach to Meaning”, Southern Journal of Philosophy, 43 (4) 2005, 505-530. A paper building on Chapter Four will be published in a volume of essays honoring the work of Michael Walzer. The main argument of Chapter Nine is anticipated in the latter part of "Taking the Linguistic Turn Seriously", The European Legacy, 13 (5) (Special Issue on "The Languages of the Sciences and the Humanities", ed. Oren Harman), 2008, 605-622, and is summarized along with some of the relevant strands of Chapter Eight in "Toward a History and Philosophy of Scientific Agency", The Monist, October, 2010. 373 **** As this book goes to print, the collaboration that sired it disbands, with each of us, well embarked on new projects, heading our separate ways to well-earned sabbatical leaves – Benbaji at the Yale Law School, and Fisch at the Institute for Advanced Study, Princeton, and Collegium Budapest. Benbaji's new project is a book on the ethics of war – the point at which normative diversity ceases to pose a constructive challenge and deteriorates into violence. Fisch's new book will further develop the account of scientific paradigm-shifts outlined in Chapter Nine, in application to a series of interrelated 19th century case-studies. We thus each return to our original fields – Benbaji to ethics, and Fisch to the history and philosophy of science. But we do so enriched and positively "ambivalated" by a study-partnership – not unlike the one described by Abot de Rabbi Nathan at the end of Chapter Seven – that yielded a project neither of us was capable of undertaking alone. The collaborative partnership we succeeded in forging would not have been possible, however, without the loving support of our families, and especially the sustaining and inspiring force of our partners in marriage – Hanna and Hagit - to whom this book is dedicated with our deepest gratitude and all our love. 1 CHAPTER ONE SETTING THE STAGE 1. Two Concepts of Rational Action Theories of rational action fall under two main headings. The first pronounces an action rational if, given the circumstances, it is perceived to be the right or proper thing to do. To act rationally is to act in ways that, given the actor’s needs, desires, and, situation, meet expert approval. On this view, what is scrutinized in assessments of rationality are the merits of the actor’s actual performance, rather than the quality of the data at her disposal, her level of expertise or the quality of her reasoning. It is the end result that counts. To be considered rational in this sense, it is enough for an act to be deemed effective or worthwhile. What is appraised is performance, not agency. Action taken or withheld thoughtlessly, instinctively or unwittingly may also be considered rational as long as it is deemed appropriate. Rationality thus construed is, therefore, not limited, in principle, to human action. Instinctive animal behavior can be deemed rational in this respect. Likewise, properly functioning computer programs, automatic pilots and thermostats, programmed to ‘sense’, ‘assess’ and respond to a situation effectively, are often said to perform rationally without doing too much violence to the term. It is a notion of rationality used widely in talk of skillful feats one trains to perform instinctively without having to ponder upon them each time anew. It matters little how the skill was originally acquired. Often, as Michael Polanyi famously argued, the 1 learning process may itself remain tacit. Most of us learn to perform highly complex feats of logical reasoning, for example, without ever having studied logic as such. The move from tacit human knowledge acquisition, to pre-programmed, hardwired animal and artificial skilled response is then quite natural. To paraphrase Bernard Williams’s well known distinction, it is not X’s reasons for - ing that are being judged, but the extent to which -ing is considered the reasonable 1 Polanyi (1958) especially Part Two. 2 2 thing for X to do. Rationality on such a showing is hence a matter of situation analysis. It has to do with the weighing of possibilities and calculation of possible outcomes. Due to the supposedly calculable, algorithmic nature of its underlying logic, this notion of rationality has attracted enormous attention in recent years, especially among decision- and game-theorists. Thus construed, rationality still requires reflection and sound assessment, but it is not the actor who is required, or, in most cases, even qualified to do so. To be considered rational she is expected to act in conformity with what is considered to be appropriate by the best means available, but without necessarily having to have worked it out for herself. Indeed, unless, the actor happens to be the referring expert herself, her own account of her actions will be of little if any relevance to their rationality. 3 Nor need the required approval be unanimous, or granted immediately. Approval, and with it assessments of rationality may be conceded by some and not by others, and, even then, only long after the event. Rationality, in this sense of the term, is time- and 2 Williams (1981, 101-2) Williams’s distinction between external and internal reasons is largely preserved in what follows, notwithstanding, though, his concern with the coherence of the former. If the rationality of a move is taken to turn exclusively on expert, rather than on the actor’s own opinion, as this view of rationality maintains it should, the reasoning involved, though external to the actor, will always, and by definition, remain internal to those who count – namely, the consulted experts. We remain unconvinced, however, with regard to Williams’s main claim that the very idea of critically reviewing and justifying action one did not actually will is inherently incoherent. 3 Imre Lakatos’s notion of scientific rationality is a good example of such an approach. Lakatos (1971) urges historians of science to classify as rational any scientific decision of the past that conforms with what they, the historians, believe science’s true method to be, regardless of whether the scientists they are studying were, or could have been, aware of it. For criticisms of Lakatos’s approach see Curtis (1986), Garber (1986) and Fisch (1994a) and (1994b). For discussion of the Hegelian aspects of Lakatos's historiography (the way rationality is conferred upon one's current position by rationally reconstructing the tradition to which it belongs), see Hacking (1979/1981). For a comparison of that to Brandom (2002a) and (2009), see below Chapter Seven, §1. 3 context-dependent, in principle, and is hence interestingly relativized despite its supposedly objective, calculable nature. Actions deemed inappropriate and hence irrational by some, may come in time to be considered fitting and perfectly rational by others – and vice versa. Although the rationality of the type of action in question may have been ascertained long before the specific act under consideration was performed (as in the case of standard countermoves in chess, or an appropriately applied well-honed mathematical technique), evaluations of this first category of rationality are always backwardlooking. What is deemed rational is forever what the actor ended up doing or refraining from doing. One does not embark on a process of deliberation that can be deemed rational before it is completed and its outcome independently assessed. Rationality, in this sense of the term, is always decided after the event. It is not a prospective evaluative category of acting, but a retrospective evaluative category of types of actions. What makes for the rational is not what goes into practical reasoning, if you wish, but what comes out. It is the move the actor actually makes that is assessed regardless of the manner in which she reasoned or conducted herself prior to making it. The second approach to rationality is less liberal. It insists, as Robert Brandom puts it, on distinguishing human rationality from “the mere reliable differential responsiveness that we share with non-conceptual creatures such as pigeons – or as far as that goes, 4 with photocells and thermostats.” It demands of moves deemed rational to be more than merely right or efficient. It is trickier than the first, and is less prone to calculation. To be considered rational, it is not enough for a move to gain retroactive approval, an action is required to be the outcome of considered deliberation. Rationality is made to turn not on what the actor ends up doing, but on the manner in which she arrives at her decision to do so - gauging the quality of reasoning culminating in the act, rather than 5 the merit or effectiveness of the act once made. Here, by contrast, agency reigns supreme. Thoughtless action, appropriate, effective and meritorious as it may turn out 4 5 Brandom (2002b, 96). See also his (2009, 182ff.). For a concise formulation of this position see, for example, Korsgaard (1997, 221). 4 6 to be, will never be deemed rational. There are other evaluative categories by which to commend action that merely fits the bill. Rationality is here construed as an evaluative category reserved exclusively for considered action; action taken self-consciously and reflectively. Rationality's Authority But by what standards are we to judge an actor’s considered deliberation? The answer is obviously, by appealing again to expert opinion. An action will be deemed rational only when taken by virtue of such feats of practical reasoning that meet authoritative approval. The canons of rationality appealed to here differ substantially from the those of the first-order situational analyses that gauge the first model. Here the actor is not expected to act merely in conformity with the outcome of expert diagnosis, but to emulate expert diagnosing. To paraphrase Bernard Williams again, to be deemed rational, it is not enough for a move to be considered reasonable by approved standards, it is required to be reasoned for approvingly. The first sense of rational conduct required actors merely to act as would an expert, the second requires of them to reason as experts do. The two notions of rationality differ significantly. At best, the class of activity deemed rational by the second, forms but a tiny subset of that deemed rational by the first. (But even that is not entirely true, for according to the second, an action may be considered rational even when unsuccessful, or deemed mistaken – which for the first would be unthinkable.) The demands on the actor are in a sense far more stringent in the second than in the first. But the two do have one important feature in common: they are both ultimately gauged by appeal to authoritative approval. The foci of approval are very different in each case, and, consequently, so are the criteria by which it is leveled. Still, gauging the rationality of a move by appeal to accepted standards renders rationality 6 Of course, “social choice” type accounts of rationality of the first kind, also foreground agency, especially those involving calculi of utility maximization. But it is not agency in acting, that is required, only agency in willing or desiring the outcome of the action, which in itself, could be performed unaware. See, for example, Seligman (2000, 15ff.) 5 inherently context dependent. It matters little if the standards invoked are the accepted canons of appropriate action or those of appropriate reasoning. In either case, rationality becomes a matter of conforming to the currently accepted, and hence, by definition, wholly dependent upon time, place, and opinion. Indeed, why not? Do not all our evaluative categories invoke standards and norms, and are not all our standards and norms culturally determined? Isn’t it obvious that action considered appropriate in one context may well be deemed unfounded in another; that conduct deemed reasonable in one cultural setting, may be considered outrageous in 7 another? Needless to say, the arguments for the relativity of the normative in this regard are powerful. With respect to rationality, however, there is a price to pay. Second-order Pondering To see this we need to take a closer look. As noted, the context-dependency of the first notion of rationality makes perfect sense. If the rationality of a move is thought to be calculable by the most recent decision-theoretic or game-theoretic approach, it should be expected to change as the theory is further developed or replaced. If rationality amounts to acting in accord with expert opinion then, by definition, it is a matter of opinion. What allegedly saves rationality from becoming merely a matter of opinion is, of course, the fact that it is made to turn, not just on anyone’s opinion, but on that of reliable assessors who are expected to form their opinions rationally. But if rational action is action taken in conformity with approved standards of performance, how is the work done on those very standards to be rated? Since standards of adequate performance of both acting and reasoning undergo changes from time to time, by what means, and with respect to which measures can their second-order scrutinization and replacement be deemed rational, arbitrary, or capricious? This is where the first notion of rationality converges on the second, and where both become problematic. Our reason for focusing on expert, rather than on, personal, common, communal, or 7 Unless, of course, one denies the very idea of such normative diversity. As will become apparent in chapters to come, we take the existence of radical normative diversity as a given. 6 majority opinion, is to lay special emphasis, if only metaphorically, on the special role played by informed specialists in devising, reviewing and occasionally revising the standards, norms and methods of their fields of expertise. Even the most avid supporters of the first notion of rationality will allow thoughtless, mindless action to be considered rational only if it can be said to merit the authoritative approval of someone who has competently thought it through. Unconsidered action, or activity performed thoughtlessly by an entire community that is merely following custom, instinct or habit, will not be considered rational by either model. The two approaches differ with respect to the level of reasoning, consciousness, and agency they demand of whoever (or whatever) performs the action in question. But both agree that to be deemed rational, an act must be actively and consciously approved. Rationality, on either showing, would seem to require more than knowingly or unwittingly following a rule or a ruling; it requires that the rule or ruling followed be fittingly approved and endorsed. An approved norm, rule or ruling, is a considered one; one that at some point was weighed and deliberated, compared and tested for situations and tasks resembling those at hand. What makes for rational action, according to both understandings of the term, is the fact that it is reasoned. Although the authors of the present study side with the second approach - insisting that for action to be deemed rational, the actor herself be required to perform the reasoning - we shall press the matter here no further. Rather than attempt to adjudicate between theories of rationality, let alone propose one of our own, this book, as stated at the outset, focuses on what we hold both approaches to rationality to have in common: namely, a requirement that the rational turn centrally on some form of second-order, deliberative assessment, and what Christine Korsgaard 8 terms "reflective endorsement" of the sort exemplified most fully in prudent expert 8 Korsgaard (1996) Chs. 2 and 3. Korsgaard's concern in that work is the nature of normative obligation, rather than that of rationality. She locates the source of normativity, late of Kant, in the essentially self-critical and autonomous self-governance of reason’s imprimatur, but contributes little to articulating its precise workings. One way of describing the project the present study has set itself is to turn Korsgaard's Kantian criterion of reflective endorsement on itself. For to discern the reflexive, second-order project of reason's very own normativity, is to ask how it is possible for the understanding, 7 evaluation. The term “have in common” could be misleading, though. For we believe that this type of reflective deliberation is more than something theories of rationality just happen to share. On the contrary, it is our contention that the various approaches to rationality exhibit wide, if at times implicit, agreement that this type of reasoning is, in the last analysis, rationality’s widely shared defining feature. It is where humankind’s capacity to reason (minimally) enters the picture to make the rational what it is and to mark it off from what it is not. The sort of evaluative approval to which we refer requires a stepping back from the well-honed ways of old with a view to considering them to some degree anew. If rational action is action taken with a view to effectively meeting a need or accomplishing a goal, then to approve such action amounts, minimally, to critically reviewing the means selected to do so. A more robust evaluation might question the 9 needs and goals themselves. But let us concentrate for the time being on the minimal case. This type of reasoning, located at the heart of rational endeavor, aims at carefully as McDowell puts it, "to reflect about and criticize the standards by which, at any time, it takes itself to be governed" McDowell (1994, 81). 9 As Robert Audi observes, practical rationality is addressed mainly in ethical works. “very few writers on practical reason” he observes, “have addressed the overall territory of reasons for actions, encompassing both moral and non-moral conduct” Audi (2001, vii). Without stating so explicitly, Audi’s remark obviously limits itself to works of the second kind discussed here, as few would consider a thoughtless act moral (good, perhaps, but not moral). In concentrating on moral conduct these works focus more on the rational and moral selection of goals (and in doing so often sorely conflate the two), rather than on the more ‘practical’ appropriation of means to morally valued ends. As a result, value-neutral instrumental rationality is routinely frowned upon for its irrelevance to ethics. In conflating the morally desirable and the rational, such accounts tend to ignore the very possibility of an amoral, or even an immoral rational act - a point we shall have a little more to say about in Chapter Seven. Audi himself is less concerned with separating the rational from the ethical than with globally combining theoretical and practical rationality - a commendable initiative in itself. 8 reviewing the actor’s circumstances, values, assets and goals, with a view to assessing 10 the aptness of her response to the specific task at hand. If the exercise is not an empty one, it will always involve, if but fleetingly, a critical weighing of alternatives. In routine situations it will amount to no more than a nodding acknowledgement of the selection by the actor of the standard tool for the job. But not all situations are routine. Some of the time much more will be at stake and much more will be involved. Each time the actor’s specific circumstances, needs or purposes defies standard procedure, standard procedure will itself become the object of reflection. Most often even then little more will be required than adjusting and tuning standard tools and methods to meet specific conditions in routine and well-rehearsed ways. But, often it is not. And when it is not, that is when the standards and the norms constitutive of the field are liable to be called into question. These are moments when experts are required to reflect critically upon the tools of their trade, rather than expertly apply them. In facing such outstanding problems they will often have to rely on the findings of sister disciplines. Peter Galison has drawn attention to the way well-managed, ‘trading zones’ are established where practitioners of different disciplines communicate and ‘trade’ by means of suitably watered-down, pidgin versions of their different 11 professional vocabularies. But dependent as a field may be on other fields for solving problems, it is seldom the case that the problems themselves are comprehended outside it. No one is better placed than the foremost practitioners of a field to encounter, experience and acknowledge difficulties that arise within it, just as it is they who will subsequently be charged with the task of considering, debating and eventually approving their solutions. 12 Needless to say, fields operate by means of norms, standards and procedures that are 10 11 12 This is true whether or not the actor and approver are one and the same. Galison (1997, 803-44) In our discussion of scientific trading zones in Chapter Nine below, we go a step beyond Galison in locating them as the principal sites in which scientists are liable to encounter the kind of "trusted criticism" capable of prompting normatively transformative processes of scientific rethinking within the community. 9 more easily modified than others. Some fields of activity - those that constitute language games in the more literal sense of the word such as in the sports and in some of the arts - maintain rigid rules and regulations in order to increase the challenge and 13 the competition artificially. But the norms, standards, and procedures that regulate the activity in most fields are not maintained merely for the sake of making the play more difficult or challenging, and yet they are not regarded by practitioners as sacred absolutes imposed from without. Rather, they are endorsed and maintained normatively as well-considered means to well-considered ends. 14 And just as they are seriously endorsed and maintained as such, they, too, are liable to come under serious review from time to time in response to problems encountered. It should not come as a surprise that the mode of reasoning that makes for rationality thus turns out to be the very same mode of reasoning that accounts for the studied operation and the reasoned fashioning and re-fashioning that occurs in all fields of considered human endeavor. If to act purposefully is to apply a means to an end, and if what makes for the rationality of such a move is that it be competently reviewed and approved, then the two modes of reasoning are not merely similar, but in this sense 15 fully coincide. Competent assessors of rational action are experts, by definition. And the role they are expected to play in assessing such action is precisely the same role 13 But even they most often exhibit interesting histories in which their rules and standards were occasionally changed with a view to improving the system. See, for example, Alan Richardson (2002) and John Haugeland's (1998,ch.13) insightful comparisons between the history of the rules constitutive of the game of Chess (real in Richardson's case, and imaginary in Haugeland's) and those of science. 14 This is true even the case of final ends, as we shall see, following the work of Harry Frankfurt, in Part IV Chapter Seven §3 and especially Chapter Eight §3 below. For Frankfurt's insightful problematization of the means-ends relation see his (1992) 15 "The notion of and arrangement of ends and means," writes Frankfurt, "comprehends both the purposefulness and the rationality that are essential features of our active nature; and it also facilitates considering the relationship by which they are connected." (Frankfurt (1992, 82). 10 they are expected to play within their respective fields. The point we wish to emphasize, however, is less the actual agreement between the two roles, which is pretty obvious, than the wide spectrum of activity of which they consist. This, we have seen, extends from the nodding approval of standard uses of standard tools and methods, at its lowest point, all the way up to the studied, at times anxious, rethinking of the field’s very foundations in the face of outstanding difficulties. Note, however, that we do not demand that to be deemed rational action must be approved or reasoned for correctly in any absolutist or foundationalist sense of the term - only that it be approved, reasoned, questioned, reflected upon authoritatively; that it be normatively endorsed. Rationality thus construed would hence seem quite susceptible to the cultural relativity of the normative central to much latter-day philosophy of language, mind, self and culture. Or so it would seem. We shall return shortly to the three master ideas that animate this study: (a) the central identification of an action or a belief's rationality with them being considered, i.e. embarked on or endorsed for a reason; (b) the essential normativity of such reasoning; and (c) the inevitable role of normative self-criticism in the rational modification or replacement of normative commitment. But first a word on relativism, or what we prefer to refer to as Normative Diversity. 2. Normative Diversity: An Introduction In what follows we take relativism in the sense endorsed and held in esteem by Richard Rorty as opposed to the sense he rejects (despite his objection to the very use of the 16 term ). The notion of relativism in which we are interested, and with which we sympathize, differs from the thesis sometimes termed “cultural” or “moral relativism”, 17 alleged to claim that any viewpoint is as good as any other. Such a position begs the question, if only by presupposing a transcultural notion of ‘good.’ According to the 16 17 For example Rorty (1989, 44) and (1996). On Rorty’s explicit rejection of cultural relativism see Rorty (1999, 15). Elsewhere in his work, however, he appears to come uncomfortably close to endorsing it – at least by implication. See, for example, Rorty (1990). 11 more stringent position we shall be interested in, such a “View from Nowhere” - to use Thomas Nagel’s oxymoronic metaphor from which the title of the present study takes its cue - is humanly unavailable. A culture, a language-game, a form of life, a normative outlook, it argues, can only be viewed from within or, alternatively, from the necessarily biased perspective of another. From these two humanly available perspectives (that roughly map onto Korsgaard's distinction between first- and thirdperson perspectives 18 with regard the normative) mutual understanding and normative evaluations and preferences are quite coherently forthcoming, but, by definition, are never ‘objective’ in the trans-cultural, framework-independent sense of the term. Relativism, as we refer to it here, is the position claimed to follow from the very idea by which cultural relativism is rejected, namely, the denial of the availability of a normative vocabulary that “has nothing to do with agency, values or interests,” 19 and that can, therefore, not be shared by radically different normative outlooks. It is the position we dub in what follows “Comparative Normative Irrealism;” one that denies thin normative concepts like ‘true’, ‘good’, ‘despicable’, ‘rational’ and ‘sound’ the capacity to compare and rank diverse normative systems, in a manner acceptable to both. “Dewey,” writes Rorty, thought that the Kantian notion of “unconditional obligation” … could not survive Darwin. … For him, all obligations were situational and conditional. This refusal to be unconditional led Dewey to be charged with ‘relativism.’ If ‘relativism’ just means failure to find a use for the notion of ‘context- 18 Cf. Korsgaard (1996, 96-100) and the discussion of her position by Moran (2001, 138151). 19 The formulation is Rorty’s, describing the extent of his agreement with Hilary Putnam. See Rorty (1993), quote at p. 45. The framework dependency of all normative evaluation and reasoning, as well as the multiplicity of such frameworks is also an idea central to Charles Taylor's Sources of the Self and the work leading up to it. See especially section II of Taylor (1977), and Part I of his (1989). 12 independent validity,’ then this charge was entirely justified. 20 Such a position, it is argued further, entails that meaningful discussion of, and reflection upon, one’s own normative vocabulary (or language game, “system of belief” or “form of life” or “final vocabulary” as such frameworks are variously termed) is necessarily determined by, and contained within it to the extent that, as Rorty rather forcefully puts it, “all we can allow ourselves” is to “limit the opposition between rational and irrational forms of persuasion to the interior of a language game, rather 21 than try to apply it to interesting and important shifts in linguistic behavior”. It is only “[w]ithin a language game, within a set of agreements about what is possible and important, [that] we can usefully distinguish reasons for beliefs from causes for belief which are not reasons.” 20 22 Relativism as we take it here is the view according to which, Rorty (2000a, 23-4). The point, of course, is for Rorty (as it was for Dewey) a quite general one concerning all manner of normative obligation and judgment. Rorty’s slide in the passage, from talk of ‘unconditional obligation’ in general, to that of ‘contextindependent validity’ is due the specific debate with Habermas, Davidson and Putnam of which it is part. 21 For an interesting, and equally forceful formulation of the same idea see Wittgenstein (1974, §608-612) – at least as we read him (see below Chapter Eight §6 and see also Dilman (1971).) Unlike Rorty, Wittgenstein reserves the term "persuasion" to the unreasoned variety. "At the end of reasons," he writes, "comes persuasion" (§612). 22 Rorty (1989, 47, 48). The point is a delicate and complex one, however, that Rorty himself saw fit to subtly reformulate in later work. Here is how he puts it in discussing the views of Apel, Habermas, Putnam, and especially Albrecht Wellmer several years later. On the one hand, he declares, “I whole-heartedly accept” the claim “that the very idea of incompatible, and perhaps reciprocally unintelligible, language-games is a pointless fiction, and that in real cases representatives of different traditions and cultures can always find a way to talk over their differences” (“This is the point”, he adds in an accompanying footnote, “made in Davidson’s ‘The Very Idea of a Conceptual Scheme’”) “I entirely agree with Wellmer,” he continues, citing Wellmer (1998, 150), “that ‘rationality – in any relevant sense of the word cannot end at the borderline of closed language-games (since 13 to paraphrase Hilary Putnam, that “elements of what we call ‘language’ or ‘mind’ penetrate so deeply into what we call ‘reality’ that the very project of representing ourselves as” capable of rationally reviewing and revising those elements, “is fatally compromised from the start.” 23 A two step argument In other words, such relativist construals of the normative, it is argued, severely constrain the extent of rational review and revision to which a normative framework can admit from within. It is a two step argument. The first consists of the sobering neoKantian realization that all meaningful normative discourse is determined by, and hence relative to the particular normative vocabulary it draws upon. Following such 24 thinkers as Rorty , Davidson, Bernard Williams, Hilary Putnam, Michael Friedman, there is no such thing)’” (Rorty (2000a, 12). However, what (Wellmer and) Rorty could mean by ‘rationality’ in this context is quite unclear, for Rorty immediately goes on to argue, as shall we in detail in chapters to come, that, on the other hand, bilateral intelligibility is, of itself, no guarantee for the possibility of reasoned agreement: The fact that there are no mutually unintelligible language games does not, in itself, do much to show that disputes between racists and anti-racists, democrats and fascists, can be decided without resort to force. Both sides may agree that, although they understand what each other says perfectly well, and share common views on most topics … there seems no prospect of reaching agreement on the particular case in hand. (Loc. cit., 13) In this sense, that of reasoned second-order, normative debate, rationality, it appears, cannot but end at the border-line of normatively closed language games! 23 Putnam (1990, 28). Putnam’s immediate concern in this passage is realism rather than relativism. What is fatally compromised, he argues there, is “the very project of representing ourselves as being ‘mappers’ of something language-independent”. 24 As Brandom aptly observes, despite the seeming all-out rejection of Kantian representationalism in Rorty (1979), Rorty “follows Kant in sharply distinguishing issues of causation from issues of justification. Enforcing this distinction between the natural and the normative (according to the lessons he learned from Sellars’ “Empiricism and the 14 Charles Taylor, Michael Walzer and Robert Brandom, whose work we shall discuss in some detail, we fully concede this first step. In doing so, we shall be arguing against the opposition, leveled most interestingly from more orthodox latter-day Kantian quarters, such as in the work of Christine Korsgaard, John Rawls and his school, and to a more limited extent John McDowell's recent work. The second, and more radical step of the argument urges us to realize that by virtue of their constitutive role in the discourses they support, our normative vocabularies are themselves rendered largely immune to normative critical appraisal from within those discourses. In being constitutive of all rational reasoning and reckoning within their scope, it is argued, they defy becoming the objects of rational reasoning and reckoning from within. Therefore, though obviously the products of human endeavor, they cannot be considered the products of rational human endeavor. Scrutinized by their users, a normative vocabulary may be found to harbor inconsistencies, incoherence, or, from a Taylorian or Brandomian perspective, felt to be in need of further explication. Problems may also arise regarding the relative priority ascribed to norms. But by virtue its constitutive role in determining the good, the right and the appropriate, a normative vocabulary, it is argued, cannot be deemed from within bad, wrong, or unfitting. A final vocabulary - to use Rorty’s coinage – namely, the set of words with which people justify their actions, and about which “if doubt is cast ... their user has no noncircular 25 argumentative resource” , is not, and, in principle, cannot be the outcome of considered, reflective deliberation of the kind we locate at the heart of rationality. Many oppose the type of relativism of which we speak - some dismissively, some thoughtfully. But many concede it – more, perhaps, than one would have thought, as we shall argue in some detail in upcoming chapters. Among those who concede it, none, to the best of our knowledge, have taken issue with the second step of the Philosophy of Mind”) is what leads Rorty to insist that our environment can at most cause us to form beliefs, not justify them. In his reliance on this fundamental distinction,” submits Brandom, “Rorty is a Kantian, even as he deploys this tool to criticize the epistemological tradition Kant represents.” Brandom (2000b, xv) 25 Rorty (1989, 73). 15 argument, although only very few acknowledge it explicitly – Rorty and Davidson stand out in this respect. 26 Work on rationality among those we class as normative irrealists shows little interest in, and exhibits virtually no discomfort with the idea that our normative vocabularies elude rational scrutiny. The exceptions, as we shall see, are Friedman’s work on the rationality of scientific framework transitions and Brandom’s 27 common law, Hegelian model of ‘historical’ rationality. However, even their notions of rationality, as we shall argue in some detail, are wholly lacking of the normative self-critical aspect of rational evaluation on which we insist Put briefly, the aim of the present study is to explore the possibility of resisting the second step of the argument, while fully endorsing the first; to salvage, in other words, a viable notion of normative self-criticism, applicable to both inter-subjective and intrasubjective normative discourse, yet without resorting to an external or comparative normative foothold. Normative vocabularies, we shall show, are susceptible to normative criticism from within despite their essential incomparability to sufficiently diverse alternatives. But before explaining in more detail why, and outlining how we intend to do so, a little more needs to be said about step two in order to get a better sense of the difficulties involved in contesting it. We do so by way of an introduction by appealing to the work of its most outspoken and eloquent advocate, Richard Rorty 3. Rorty's Relativism Richard Rorty, a primary proponent for the two-step argument, does not regard its conclusion in any way problematic. On the contrary, the “ironist” strong poet, whose image he portrays and promotes in Contingency, Irony, and Solidarity, and with whom 26 27 On Davidson’s alignment with Rorty, see text to ftn. 40 below. Friedman (2001) and Brandom, first in his (2002a, 12ff.) and more fully in the third of his 2007 Woodbridge lectures now published as Chapter 3 of Brandom (2009), respectively. The interesting affinity between the two is explored in Part II Chapter Five below. 16 28 he fully identifies, is someone who knowingly conducts herself in accord with, and in full awareness of this constraint on her rationality. She is someone who fulfills three conditions: (1) She has radical and continuing doubts about the final vocabulary she currently uses ...; (2) She realizes that arguments phrased in her present vocabulary can neither underwrite nor dissolve these doubts; (3) Insofar as she philosophizes about her situation, she does not think that her vocabulary is closer to reality than others, that it is in touch with a power not herself. (73) She is portrayed as self-critical, but only in the sense of tracing the outer-borders of self-criticism. Aware of her final vocabulary's inherent contingency, she doubts it persistently, and spends “her time worrying about the possibility that she has been initiated into the wrong tribe, taught to play the wrong language game.” However, and this is the crucial point, “she cannot give a criterion of wrongness”, and thinks, therefore, that there “is no reason to think that Socratic inquiry into the essence of justice or science or rationality will take one much beyond the language games of one’s time.”(74-5) She, therefore, experiments in fashioning new vocabularies by blindly casting around for new metaphors to redescribe herself and her world. Unable to question, she dabbles; unable to troubleshoot, she tinkers. She gropes blindly because she assumes, with Rorty, that from the vantage point of her current vocabulary – which is the only committed vantage point available to her - it is impossible to articulate what might be wrong with it, or in what ways it might be improved, and that, subsequently, “the creation of … a new vocabulary, will have its utility explained only retrospectively”(55) – and, again, only from within. Rorty’s ironist aims ultimately to improve upon her vocabulary, rather than merely change it for another, but can only do so aimlessly. She can envisage change, but, unable to set herself a prospective goal, 28 Sliding effortlessly between reference to “the ironist” and “we ironists”. See for example pp. 79-81. 17 other than change itself, she cannot envisage progress. All she can do is to do her best 29 to be original, to keep telling her story differently. She works on her vocabulary, but 30 cannot appraise it critically. She can, therefore, not proceed rationally in improving it, in the sense we have described and shall explain further below. She can only blindly redescribe with the hope of hitting unwittingly upon an option that in retrospect she will find worthwhile. Rorty refers frequently and glowingly to the work of Thomas Kuhn whom he describes as “one of my idols”, “a great philosopher”, “the most influential ... to write in English since the Second World War.” 31 Most of the time, Kuhn is praised for redescribing science to show that “the subject of truth claims cannot be a relation between beliefs 32 and a putatively mind-independent or ‘external’ world,” thus contributing to the idea central to Rorty’s entire project that “correspondence to reality is a term without 33 content.” When he does refer to Kuhn’s description of paradigm shifts he is less clear 29 Rorty insists that as in the arts, the “anxiety of influence” – as Harold Bloom famously dubbed the fear of unoriginality – be considered the major incentive for all manner of creative reform. See Rorty (1989, 24-25) for his reference to Bloom (1973). 30 On this point Rorty's rhetoric can be misleading, as, for example, when speaking of “the gradual trial-and-error creation of … the sort vocabulary developed by people like Galileo, Hegel, or the later Yeats” - Rorty (1989, 12). However, it is clear from his description of the three men’s work in the very next passage, that he uses the term “trial-and-error” very differently from the Popperian notion of a problem driven process of conjecture and refutation. For Rorty, as we shall see shortly, a final vocabulary can only be deemed problematic from without, never from within. Errors may come to light, but only retrospectively from the new perspective of an aimlessly hit on solution. For another similarly misleading utterance, see his attribution to Kuhn of the idea “that science, like politics, is problem-solving” Rorty (1999, xxi). 31 32 33 Rorty (1997) at pp. 175, 188-9. See also Rorty (1999, 35-6). Kuhn (1993, 330), cited approvingly by Rorty (1990, 67). Rorty (1998, 68). See especially Rorty (1979, 322-42). See also Loc. Cit. pp. 25, Rorty (1989, 28), (1999, xvi and 35-36), and (1997). 18 than one would expect, however, tending to slide too easily into Kuhn’s collectivist idiom, and shift attention from the “strong poets” responsible for fashioning the paradigm shift, to the community that ends up endorsing it. This is to blur the difference between rational inquiry and rational acceptance; between perceiving and responding to problems, on the one hand, and being willing to entertain their solution once proposed. 34 Rorty thus creates the false impression that the issue is essentially sociological. “As Kuhn argues in The Copernican Revolution”, he writes: we did not decide on the basis of some telescopic observations, or on the basis of anything else, that the earth was not the center of the universe, that macroscopic behavior could be explained on the basis of microstructural motion, and that prediction and control should be the principal aim of scientific theorizing. One wonders who are the “we” to whom Rorty refers. He quickly clarifies: Rather, after a hundred years of inconclusive muddle, the Europeans found themselves speaking in a way which took these interlocked theses for granted. Cultural change of this magnitude does not result from applying criteria ... We should not look within ourselves for criteria of decision in such matters any more than we should look to world. 35 The question addressed here is to explain how a community, or a society at large (“the Europeans”) acquires and internalizes a newly fashioned final vocabulary after it was formed, rather than the very different, and far more radical question, so central to his 34 The same conflation between inquiry and acceptance, we shall argue in Chapter Five, plagues the account of scientific rationality proposed by Kuhn’s most important philosophical successor, Michael Friedman. 35 Rorty (1989, 6) italics added. 19 book, regarding the manner in which such radically new vocabularies are fashioned by 36 their authors in the first place. Rorty’s reference here to “the Europeans” rather than to the individuals who set the scientific revolutions he mentions in motion, creates the false impression that his reference to Kuhn pertains to the first rather than to the second question. 37 The passage immediately preceding the one quoted, goes beyond Kuhn to include paradigm-shifts outside the sciences, but remains similarly misleading: Europe did not decide to accept the idiom of Romantic poetry, or of socialist politics, or of Galilean mechanics. That sort of shift was no more an act of will than it was a result of argument. Rather, Europe gradually lost the habit of using certain words and gradually acquired the habit of using others. (ibid) Discerning how professional communities and the educated public at large lose and acquire such habits of thought and speech is a complex and fascinating story in itself, of course, to which we return if briefly in Chapter Nine. But the real challenge to rationality is that of the poets, the social reformers, and Galileo themselves, who spent a lifetime deliberating, contriving and perfecting these new vocabularies? Can Galileo's creative enterprise be deemed “no more an act of will than it was a result of argument”? Can his lifework also be described as a matter of merely losing “the habit of using certain words” and acquiring “the habit of using others”? When these early passages of Contingency, Irony, and Solidarity are reread in the light of the self-portrayal of Rorty’s ironist developed later in the book, the answer turns out to be a clear and 36 This is not to say that the first question is purely sociological. For an early and lively discussion of the philosophical significance and ramifications of what he dubs “the sociological problem of induction” see Harry Collins’s now classic (1985). We discuss Michael Friedman's equally 'collectivist' reading of Kuhn in some detail in Part II, Ch. Five below. 37 Elsewhere, though, he is clearer. See, for instance, Rorty (1989, 20) and (1998, 339 ftn.14). 20 disturbing “yes!”. And indeed, a little later in the first chapter, he says so explicitly. Wittgenstein’s analogy between a vocabulary and a tool, he says “has one obvious drawback”: The craftsman typically knows what job he needs to do before picking or inventing tools with which to do it. By contrast, someone like Galileo, Yeats or Hegel … is typically unable to make clear exactly what it is that he wants to do before developing the language in which he succeeds in doing it. His new vocabulary makes possible for the first time a formulation of its own purpose. 38 Only after a new vocabulary is up and running as it were, can its creators appreciate retrospectively what it is good for and how it fares better than the vocabulary it replaced. Although few would disagree that there is always more to a tool than what goes into it, and that it is never possible to anticipate all its uses before it is actually 38 Pp. 12-13, italics added. In some of his later work, especially when writing in favor of Habermas’s distinction between “subject-centered” and “communicative” accounts of rationality and preference for the latter (cf. Habermas (1987, Lecture XI)), Rorty might seem to be less decisive on this point than he was in 1989. In Rorty (2007, 85-6), for instance, he characterizes his position thus: “On the view of culture I am suggesting, intellectual and moral progress is achieved by making claims that seem absurd to one generation into the common sense of the later generations. The role of the intellectuals is to effect this change by explaining how the new ideas might, if tried out, solve, or dissolve, problems created by the old ones. … by explaining how the new institution or the new theory might solve problems that the old institutions or theories could not handle.” But the change of heart is but apparent. There is nothing to indicate that his position had changed in this respect between the publication of Contingency, Irony, and Solidarity, and that of his most recent collection of essays. The give-away phrase in the passage quoted is of course “if tried out”. Only from the vantage point of the new, can the problems created by the old be appreciated, and the new not “seem absurd.” 21 employed, Rorty’s position is far more radical. Only after an alternative vocabulary is endorsed and employed, can the problems solved by its endorsement be even formulated. A functioning final vocabulary can only be deemed normatively wanting from the perspective of another. In other words, a final vocabulary cannot be normatively trouble-shot and modified from within, but only by someone working from the outside; someone whose final vocabulary it no longer is! A person can be ‘ironically’ suspicious of her final vocabulary, and desire to experiment with other ways of life, as Rorty’s ironist is. But she can never do so for a reason. One is unable ever to state what is, or what might be wrong with one’s normative commitments (other than to find them incongruous with other of one's norms or in need of further clarification). In short, a language game cannot be reflected upon, it cannot admit of normative shortcomings, cannot be criticized and cannot be improved upon from within, except in the trivial sense of troubleshooting for consistency, coherence, or clarity. Language games may be fashioned by humans, but never for a purpose other than for their sheer novelty. And thus, as Rorty is the first to admit, precious as it may be, our admirable capacity for self-criticism and self-improvement is required by his philosophy to stop short of its seemingly profoundest, to be replaced by the aimless experimentation of clueless irony. At one point Rorty briefly explains why. Quoting Davidson’s “Paradoxes of Irrationality”, 39 and alluding to the distinctions between thick and thin normative concepts as well as to that, late of Frankfurt, between first- and second-order desires, Rorty presents a strictly coherentist account of in-house critical reflection. He notes that if we reserve the term ‘rational’ to mean something like ‘internal coherence’, ... then we shall be forced to call ‘irrational’ many things we wish to praise. In particular we shall have to describe as ‘irrational’ what Davidson calls ‘a form of self-criticism and reform which we hold in high esteem, and that has even been thought to be 39 Davidson (1982). 22 the very essence of rationality and the source of freedom.' 40 The process by which “a person forms a positive or negative judgment of some of his own desires”, and acts to change them, he explains, is limited to acts of self-criticism and self-improvement pertaining to a “higher-level” desire capable of mediating and rationalizing the contesting lower first- and second-level desires. However, working up the hierarchy one necessarily reaches a point where the only candidates for such highest-level desires are abstract and empty as to have no mediating powers: They are typified by ‘I wish to be good’, ‘I wish to be rational’, and ‘I wish to know the truth.’ Because what will count as ‘good’ or ‘rational’ or ‘true’ will be determined by the contest between the first- and second-level desires, wistful top-level protestations of goodwill are impotent to intervene in that contest. (Ibid) The possibility of rational appraisal doing real normative work swiftly evaporates as normative concepts become ‘thinner’. And thus, just at the point that our highest and most relished norms and standards ‘thicken’ out to acquire the precise meanings they have for us, they freeze and fossilize, and for lack of independently meaningful higher standards by which to judge them, are rendered immune to the normative scrutiny of those whose form of life they govern. Given these constraints on rational self-criticism and self-improvement, Rorty, we have seen, is left no choice but to portray the self-consciously critical path-breaking work of a Galileo or a Hegel as the outcome of the aimless tinkering of a Rortian ironist. He is aware of how outrageous he must sound, but sees no reason to reconsider. He voices the natural objections, only to dismiss them, not by taking anything back, but by rejecting the vocabulary by which they are voiced. Going as it does to the heart of the problem, his comments here are worth quoting at length. 40 Rorty (1989, 49). 23 To accept the claim that there is no standpoint outside the particular historical conditioned and temporary vocabulary we are presently using from which to judge this vocabulary is to give up on the idea that there can be reasons for using languages as well as reasons within languages for believing statements. This amounts to giving up the idea that intellectual or political progress is rational, in any sense of “rational” which is neutral between vocabularies. But because it seems pointless to say that all the great moral and intellectual advances of European history - Christianity, Galilean science, the Enlightenment, Romanticism, and so on - were fortunate falls into temporary irrationality, the moral to be drawn is that the rationalirrational distinction is less useful than it once appeared. Once we realize that progress, for the community as for the individual, is a matter of using new words as well as arguing from premises phrased in old words, we realize that a critical vocabulary which revolves around notions like “rational”, “criteria”, “argument”, and “foundation” and “absolute” is badly suited to describe the relation between the old and the new. 41 41 Loc. cit pp. 48-9 - emphasis added. Charles Taylor's notion of personal normative advancement (limited in his case to the intra-subjective level of what he terms "strong evaluation") is more carefully analyzed than Rorty's, but is also made to turn exclusively on self-(re)interpretation, and, therefore, just like Rorty's, lacks critical bite. The constant attempt to make explicit one's inchoate desires and motivations inevitably lends them new form, meaning and direction, and is constitutive of what will by hindsight be deemed progressive. See especially his (1971) and (1977), and (1985a, Ch.2). We chose to focus on Rorty at this point for two reasons. First because of how, unlike Taylor (but squarely in line with the concerns of the present study), he explicitly raises the question of the rationality of normative change while unequivocally identifying rationality with selfcriticism. And second, for the full generality of his account, which unlike Taylor's, by no means excludes the natural sciences. For a pointed critique of Taylor along these lines see Geertz (1995). 24 The problem regarding our critical vocabulary is, of course, not at all that of describing “the relation between the old and the new”, but that of describing the effort invested in forming and fashioning the new as an activity thoughtfully and meaningfully performed on, and from within the old. It is not that of overcoming the incommensurability between two normatively incongruous existing frameworks, but that of deeming the search or creation of one as rationally motivated while committed to the other. Nor has the question, as we pose it and shall elaborate further in upcoming chapters, anything to do with ‘foundations’ or ‘absolutes’. ‘Rational,’ as we insist on employing the term, is a category of action and agency not of the relations obtaining between statements, languages or paradigms. 42 The question is not whether the new can be understood or inferred from the old, but whether its fashioning can be described as the outcome of the reasoned deliberation of anxious practitioners committed to and working from within the old? Rorty flatly denies that it can. He is clearly aware that in saying so he is doing violence not only to the term ‘rationality,’ but to everything we know about the way in which creative individuals critically reflect upon, not merely toy with, their norms, their standards, their tools and their methods. It strikes us not merely as irrelevant or superfluous - or as pointless, as Rorty has it - to describe such “great moral and intellectual advances of European history” as the fortunate falls of aimless, blind tinkering, but as downright absurd! Rorty, one feels, should not be allowed to get away 43 with it by merely switching terminologies or refusing to use certain words (a move he 42 Which we locate in Chapter Five, as the major drawback of Friedman's account of the rationality of scientific framework replacement. 43 In an essay published shortly after Contingency, Irony, and Solidarity, Rorty (1992) (later re-titled: “Rationality and Cultural Difference” and republished as Ch. 10 of his (1998), Rorty defines rationality anew, rather than merely drop the term. He distinguishes three senses of the term, none of which remotely relates to the notion of critical appraisal discussed here. What he terms “rationality 1 " is “an ability that squids have more than amoebas”, namely, “the ability to cope with the environment by adjusting one’s reactions to environmental stimuli in complex and delicate ways”. Rationality 2 is “the name of an extra added ingredient that human beings have and brutes do not”, an ingredient that “sets 25 makes quite purposefully, one should add, not as a result of blindly toying.) The question, however, is not a matter of sticking with, or abandoning certain words, and certainly not that of approving or disapproving such evasion tactics. The problem is a major one. Rorty’s problem is that, absurd as it may seem to some, the philosophy of language and mind to which he subscribes will not allow him to view normative vocabularies as subject to reasoned modification or replacement (except, again, in the uninteresting sense of testing for consistency, coherence or clarity). For him, dropping the troublesome terminology is not an evasion tactic. His philosophy leaves him no choice. Does our commitment to the rational appraisal of the standards and norms we employ leave us no choice but to drop Rorty’s troublesome philosophy? The answer is neither a simple yes nor a simple no. So far all we have established is that Rorty’s normative relativism is incompatible with the notion of rational appraisal when applied reflexively to the normative vocabulary by which it speaks. If that was all we were claiming, it would have been a matter of deciding to which horn of the dilemma we were more deeply committed. But the matter is not so simple for the simple reason that Rorty’s normative relativism is not an option easily set aside. So far we have stated our own preferences, and described Rorty’s contrasting account of the dynamics of normative vocabularies, and contrasting it with our own. We not yet argued for either. It is now time to say a little more about our point of departure, and to examine a little more closely the theories of thought, meaning and commitment that ground the type of position Rorty represents. Only then will we be in a position, to take fuller stock of the dilemma, and the prospects of confronting it anew be means of more than a change of words. goals other than mere survival”. Rationality 3 , according to Rorty, is “roughly synonymous with tolerance”, and ability “not to be overly disconcerted by differences from oneself, not to respond aggressively to such differences” (Rorty (1992, 186-7)). All three of Rorty’s senses of rationality have little to do with the self-conscious, self-doubting, reflective use of one’s reasoning ability. 26 4. The Road Taken Those who accept (some version of) the first, neo-Kantian step in Rorty’s argument, but, like us, are unwilling to accept his radical conclusions, have one of two nondismissive options. One, is to argue for there being some kind of foothold external to one’s normative vocabulary 44 capable of facilitating its critique from within. The modest absolutism premised, for instance, by Dworkin’s Rawlsian critique of Walzer’s interpretivist ethics, is one such approach. 45 The external foothold, as we shall see, need not be absolutist, however. Michael Friedman’s Dynamics of Reason, though limited exclusively to science, belongs in this more subtle category, as we shall see in some detail in Chapter Five below. The second option is to meet the challenge head-on by opposing Rorty’s conclusions without contesting the premises that support them (or, as described above, by conceding the first, and contesting the second step of the two-step argument). This, as we have stated, is the option opted for here. 44 Our use of ‘external’ in this context requires some qualification. We use the term here to describe positions similar to Korsgaard's for whom, following Kant, certain foundational values, whose source remains wholly internal to individual normative reflection, are nonetheless viewed as transcending individual reflection or communal commitment. They are thus normative absolutists in the manner described, while wholly rejecting the idea (often dubbed "moral realism") that normative commitment be attributed to an independent normative reality. For further discussion of moral realism within a Kantian framework see Korsgaard (1997, 240-3). However, there is a fundamental difference between attributing transcendental, and hence trans-cultural value to certain norms, and merely acknowledging the existence of a "minimal code" of values contingently shared by all communities by virtue of their members' shared human nature. There is nothing transcultural, and hence no reason to bestow normatively supra-cultural status to a value merely because it happens to be shared. The difference between acknowledging a minimal code and resisting all forms of normative absolutism (from an interpretivist perspective) is explored in greater detail in Chapter Four below. 45 Cf. Walzer (1987) and (1996, Ch.1) and Dworkin (1983). See also Dworkin (1986, 424-5) and (1996). 27 Our interests naturally lie beyond the specifics of Rorty’s own position. It is the type of challenge his work poses to rationality that animates this study, one we wish first to reformulate in a way that we find most general and effective. Properly constructed, it is a challenge we find deep and far-reaching, to a large extent valid, and not at all easily dismissed. In our reading, it comprises two oft-conflated theses that, we believe, can, and need to be properly formulated, analyzed, and distinguished. The first, mentioned briefly above, is the denial of a normative scale on which two maximally coherent, yet sufficiently diverse normative outlooks can be ordered in a manner acceptable to both. It claims, in other words that there is no purely normative relevant comparative property that is shared by significantly conflicting normative outlooks, and, therefore, capable of ranking them. We call this meta-normative position “Comparative Irrealism.” Comparative Irrealism, however, entails that, because (unbiased) normative comparisons of sufficiently diverse (and sufficiently coherent) outlooks are not forthcoming, it follows that an outlook cannot be deemed normatively wanting from within in comparison to a superior alternative. 46 The second claim, to which the present work devotes itself to contest, is that since an outlook cannot be found normatively wanting by comparison, it cannot be found normatively wanting at all! The logic is simple: if for lack of a suitable comparative normative vocabulary, an outlook cannot be deemed by its adherents normatively 46 That is not to say, however, that while engaging a form of life radically different from her own, a person cannot become attracted, and eventually commit herself to new, and formerly unanticipated normative possibilities that she comes to realize her own vocabulary lacks (e.g. the value of hospitality in Muslim societies, forms of equality found in some single-sex marriages, or some of the so-called "knightly virtues"). This is certainly a commonplace, unproblematic, and undeniably rational form of normative enrichment. But, as we argue below, such cases of framework supplementation, however, are not in violation of Comparative Irrealism because they do not require normatively faulting any part of one's functioning framework, and therefore are irrelevant to cases of norm replacement. We thank Lorraine Daston for forcefully raising this point in private communication, but beg to friendly disagree with her suggestion that all forms of framework transition are explicable along such lines. 28 inferior to another, then, for the very same reason, neither can they find it normatively wanting in itself. For the question would always be: wanting in comparison to what? A normative outlook may be judged from within to be inconsistent, incoherent, or unacceptably inexplicit, and, to some extent, even lacking in comparison to another. But, lacking a comparative dimension, the argument goes, it cannot be deemed wrong, false or unjust. And if an outlook cannot be deemed normatively problematic by its own lights, its present normative state can never be deemed to be the outcome, or the subject-matter of rational scrutiny as defined at the outset. The ultimate aim of the following pages is to drive a wedge between Comparative Irrealism and the claim claimed to follow from it, according to which all normative criticism is by nature comparative. We thus seek to salvage a robust notion of normative self-criticism while fully conceding the broadly conceived neo-Kantian and neo-pragmatist premises of the first step of Rorty's argument. Before outlining the argument unfolded in the following chapters, let us reiterate our point of departure and motivation for taking the road that we take. We set forth from what we characterized above as rational action’s defining feature (even for those who do not require agency in the actual performance of the acts they consider rational), namely, it being (at least in the last analysis) considered action; 47 action taken on reflection, for which a person is liable to be prompted for, and be willing to give his reasons for taking. 48 In other words, we take rational action to be action deemed appropriate. Which, in yet others words, is to take it to be inherently norm-governed. From which follows that for an act to be deemed rational, standards of propriety must be in place at least prior to the deeming. It is in this sense that we take 47 In talking of action we include all manner of endeavor, including such voluntary mental acts as endorsing a belief or committing oneself. 48 Hence the immediate relevance to this study of the sharp distinction (late of Kant, and central to the work of Sellars, Brandom, McDowell and many others) between exercising one’s agency in the 'space of reasons' and being merely subject to the 'space of law'. 29 normative commitment to be constitutive of rationality. Hence the problem of revising normative commitment rationally. Moreover, although the point is little stressed in the literature, we insist in addition, as noted above and as the language of the previous paragraph implies, on taking rationality to be, in essence, a prospective category of acting. There is a crucial difference between action opted for for a reason, and action taken thoughtlessly and justified only after the event; between the exercise of reason in evaluating a move and in making it. 49 Retrospective justification, or rationalization can justify an act only in the sense of deeming it in retrospect to have been the right thing to do, but it cannot count as the actor’s reason for so acting. If to act rationally is to act for a reason, because prompted, or motivated by that reason, then action reasoned for only by hindsight, for which reasons are assigned only after the event, falls short of fullblooded rationality. 50 The problem that animates this study arises when commitment to the prospective nature and inherent norm-governed-ness of rationality comes to bear in considering the possibility of rational norm revision. No such problem arises for those who deny normative diversity and/or the very possibility of rationally contrived normative realignment. But for those of us who do not, the problem of rational norm revision, as we shall explain immediately, is a major one, deemed by most insurmountable. 49 Here, of course, we show our colors as firm supporters of the second approach to rationality. However, as we argued at the outset, although the two approaches sorely differ with regard to first-order rational endeavor, they converge (on the second approach) with respect to second-order acts of rational norm revision, on which this study primarily focuses. 50 Korsgaard clearly formulates what she calls "the problem of the normative" in terms of this sense of the priority of reason to action. However, Sellars, Brandom, and McDowell locate rationality in the essentially timeless inferential relations that obtain at any one moment within a person's set on commitments (or as Brandom now calls it, her "synthetic unity of apperception (2009, pp.41ff.). 30 But first two points of clarification. First, in what follows normative diversity, i.e. the plurality of incompatible normative outlooks, will be assumed, not proven, or even justified. We take normativity, the self-committed, and commitment-driven nature of human agency, to be a human universal, in the sense of being definitive of sapience. And we join Taylor, Brandom and others in viewing normativity further as grounded in our discursive capacities. But, bracketing contingently shared areas of "minimal code", we take no normative content to be fixed or universal. The second point concerns the possibility of rational norm revision. It is important to understand that the aim of this study is not to prove that framework transitions can be genuinely motivated by rational argument and deliberation. We know that they can. The problem we set ourselves is primarily not that of convincing our readers that norm revision can be rational, but that a compelling philosophical account of their rationality is forthcoming – passé Rorty - from the broadly neo-Kantian and neo-pragmatist perspective from which we write. Of all the thinkers whose work is engaged in what follows, Friedman's Dynamics of Reason steers closest to us in this respect. Friedman does not ask whether framework transitions in physics are rational, but rather, how to make philosophical sense of their rationality from the essentially Kuhnian perspective to which he is committed? Like Friedman's, our work too is primarily a work of philosophy. The Problem Revisable Normative frameworks are adjusted and changed in a variety of ways, some surreptitious and unwitting, some contrived and deliberate. Concentrating on the rationality of framework revision, we focus naturally on a sub-set of the latter category: namely, cases in which a functioning framework is modified or replaced for a reason. Now, the only prospective reason practitioners might have for wanting to alter or replace their framework is to find it in some sense sufficiently wanting to be reconsidered (as opposed to cases, of which, Rorty makes much, in which replacing the old framework is justified in retrospect from the now-perceived superior perspective of the new). 31 There are three senses in which a functioning normative framework can, in principle, be faulted from within: it could be found incoherent, in some sense lacking, or in some sense wrong. The first two are relatively unproblematic because they do not involve deeming any part of it to be normatively flawed. They are also, as we shall see, largely uninteresting. Formal failings - inconsistency, incoherence or insufficient clarity – may well require modifying or even relinquishing lower-level commitments in the light of higher-level 51 standards. But doing so proves the framework to be considered right, not wrong! All the more so in cases where, in engaging other cultures, one’s framework is supplemented by unforeseen normative possibilities encountered. Here too, amendment is achieved without any part of the original framework being normatively 52 condemning. In the first case, lower-level commitments may well be relinquished, but they'll be dropped because found expendable, not because they are deemed normatively wanting. In the second, a commitment is added, but none renounced. Both cases obviously represent rational forms of normative adjustment. But they are powerless to account for the rationality of transitions between incommensurable, or normatively incompatible alternatives; cases in which practitioners contrive to locate or to fashion frameworks significantly different from their own – as in the case of scientific revolutions. Troubleshooting for inconsistency or incoherence can result in normative re-prioritizing within a given system, and in extreme cases, in pruning certain of its expendable elements. Awakening to new possibilities can at most result in supplementing the existing system (but never, as we shall argue at length, in rationally motivating replacing one of the system's existing elements. That would be in blatant violation of Comparative Irrealism.) It is, therefore, to the third option that we direct our attention – where frameworks are revised opt replaced because deemed normatively wanting; an option that would seem to characterize processes of rational norm revision far better than the other two, yet is considered notoriously problematic to the point of incoherence. 51 52 Or as Taylor would have it: ordinary goods in the light of "hypergoods". See below fn. 53. See fn. 46 above. 32 Framework transitions to normatively incompatible alternatives, involve, by definition, substituting major elements of the system by others - which to be deemed rational must be undertaken for a reason. Here, in particular, the difference between prospective and retrospective justification is crucial, because the norms governing the reasoning are replaced in the course of the transition. Finding reason retrospectively for having relinquished norms or standards to which one is no longer committed is matter quite different from finding normative reason to do so while they still bind one. Hence the problem. To insist that rationality be both prospective and norm-governed, and that norm revision can be rational, is hence to insist that rational agents not only exercise their norms and standards in self-criticism, but make them the object of such criticism, that one can coherently deem wrong one’s very standards of right and wrong! No viable philosophical account of personal identity makes room for such feats of selfnegation. On the other hand, no amount of mere prioritizing or supplementation, we insist, can get one rationally from Aristotelian to Newtonian physics, for example. It follows that if troubleshooting (prospectively) for coherence or (retrospectively) for lacunae are held to exhaust the range of rational normative revision, then Rorty must be granted his point that the classical examples of framework replacement in science and philosophy, or the type of deep social reform explored by Walzer, can simply not be considered rational – a conclusion we firmly contest. No work to date has addressed this dilemma openly and systematically. This is our aim in what follows. But we do so with a bias. If one is committed, as we are, to the possibility of rational norm revision (passé Rorty), one has one of two options: (a) to argue for a viable philosophical account of normative self-criticism capable of meeting the obvious objections, or (b) to argue that the rationality of radical framework transition can be fully and viably accounted for without resort to normative self-criticism, on the basis solely of troubleshooting for normative coherence, and openness to non-critical normative supplementation. 33 Many, including Brandom, Friedman and McDowell whose work we discuss and build upon, opt for (b). The bias of the present study is firmly in favor of (a). 5. The Argument The remainder of the book consists of three Parts. Part I, "Through Thick and Thin," purports rigorously to make the case for Comparative Normative Irrealism: the principle, to which this study remains committed throughout, according to which two sufficiently diverse normative outlooks are normatively unrankable. Chapters Two and Three argue for Comparative Irrealism on semantic grounds, from the different perspectives of two major theories of meaning that dominate approaches that take seriously the neo-Kantian idea of the constitutive role of language. By utilizing the distinction introduced by Bernard Williams and Michael Walzer between thick and thin normative concepts, we are able to chart a robust framework-dependent notion of normative realism, as well as to make a compelling case for Normative Diversity with reference to the former (e.g. concepts such as modesty, valor or holiness). The idea is to show that with respect to all three theories – those that ground meaning in communal norms, John McDowell's special brand of the communal approach, and idiolect-based theories late of Davidson – the very assumption of Normative Diversity entails Comparative Irrealism – i.e. the unrankability in principle of sufficiently alien normative outlooks. The upshot of Part I, is, therefore, that if normative self-criticism is at all possible, it cannot be comparative. The three chapters comprising Part II, entitled "Rationality from Within," assess and engage three important latter-day philosophical positions, all of which share premises significantly akin to our own, and bear directly on issues central to the question we raise. Normative Diversity, the incomparability in principle of sufficiently different normative outlooks, and the transformative role of “connected” normative criticism from within, are the very principles on which the so called interpretive approach to ethics builds its entire approach. Chapter Four purports, first, to substantiate the 34 53 interpretive account (in its Walzerian, rather than Taylorian form ) in terms of the thick/thin distinction developed in Part I, and in a manner applicable to normativity in general. The second and more critical objective of Chapter Four is to expose and analyze the constraints interpretivism seems inevitably to impose on the kind of normative self-criticism it seeks to promote, which we ultimately find too restrictive. Under an interpretivist construal, normative criticism can venture no further than to challenge other articulations of a given way of life, but is powerless to challenge the way of life itself. Michael Friedman’s work on the rationality of framework transitions in science is the topic of Chapter Five. Though limited to science, Friedman's Kuhnian point of departure steers closest to the presuppositions that animate the present study: forcefully conceding the framework-dependency of science late of Kant, as well as the Normative Diversity and Comparative Irrealism of a science's successive frameworks – at least to a certain extent. Also appealing is his willingness to deem rational framework transitions exceedingly more radical than those allowed by interpretivists. And yet, as we argue in some detail, unlike interpretivists, Friedman's account succeeds in steering 53 We prefer Walzer's interpretivism to Taylor's because of Walzer's explicit focus on the transformative effect of normative criticism, and the nature of its communal disputative context. Taylor's interpretivism is primarily intra-subjective and, as previously mentioned, his early work locates its normative transformative force, not unlike Rorty, in essentially non-critical acts of self-redescription and rearticulation. In Sources of the Self a decidedly self-critical form of normative self-adjustment is added to his earlier account of strong evaluation in the form of what he terms "hypergoods" - goods that "are not only incomparably more important than others but provide the standpoint from which these must be weighed, judged, decided about" (1989, 63). Taylor admits, however, that such normative criticism can only convince those whose hypergoods they are. Their normative force is hence inevitably internal, and, therefore, begs the question of the rationality of their own endorsement and modification. "A hypergood," he maintains (not unlike Brandom’s Hegelian Wiggism), "can only be defended" genealogically "through a certain reading of its genesis" (73), namely, from the necessarily circular perspective of those who are "morally moved" by it. 35 wide of any mention of the problem of normative self-criticism by focusing exclusively on the so-called problem of incommensurability, namely, that of explaining how, once formed, a new scientific framework can be considered retrospectively a "live option" by practitioners of the old. Lacking from Friedman's picture is any attempt to account prospectively for the rational incentive to seek or form an alternative to a functioning scientific framework in the first place – without which, we argue, no account of the rationality of scientific paradigm shifts can be considered complete. To do so, we insist, the problem of normative self-criticism is unavoidable. 54 Chapter Six is devoted to Robert Brandom's inferentialist normative pragmatism (appropriately extended from conceptual norms to normativity in general, as suggested by the work of Jeffrey Stout, as well as by some of Brandom's own more recent writing) in which, we argue, the interpretivist position receives powerful (if unacknowledged) articulation. Brandom’s system offers a rich account of rational action as reasoned action that, not unlike Friedman’s, appears to make no room for criticism. But unlike Friedman's account, we find in Brandom’s “deontic scorekeeping” 55 (especially when its Kantian and Hegelian underpinnings are sufficiently exposed) an 54 Robert DiSalle, who otherwise fully concedes Friedman's account, calls him to task for overlooking how practitioners keenly engage in "dialectical confrontation with prevailing conceptions" DiSalle (2002, 204) with a view explicitly to "revealing the hidden presuppositions of the old conception, and exhibiting the internal difficulties that must be resolved by the new" (2010, 528). DiSalle, however, shows no awareness of the difficulties involved in adopting such a self-critical stance, especially from the neoKantian perspective he supposedly shares with Friedman. 55 Following Sellars, Brandom, as is well known, centrally identifies rationality, or sapience, with the capacity and willingness to engage in the game of giving and asking for reasons, which, taking his cue from Lewis (1979), he further identifies in Making it Explicit (ch.3, §IV) with the ability to keep score of the deontic, inferential structure of one's fellow players' commitments and entitlements. "Being rational", he writes there, i.e. "understanding, knowing how in the sense of being able to play the game of giving and asking for reasons – is mastering in practice the evolution of the score. Talking and thinking is keeping score in this kind of game." (1994, 183). Here, we argue, the self- 36 articulation of reasoned discourse that tacitly attributes to its participants an essentially self-critical motivational stance. Making Brandom explicit in this regard, so to speak, paves the way for the constructive account of normative self-criticism we propose and develop in Part III. In Chapter Seven, the first of the three concluding chapters of the book, we propose, in outline, an essentially pragmatist phenomenology of (prudent) criticizing as a discursive move, or speech-act clearly distinguished from doubting, questioning, testing, and being merely pressed for one’s reasons. Criticism is analyzed as an addressed act of speech, directed to alerting its addressees to the existence of a problem or shortcoming within the compass of their responsibility, and, in doing so, prompting them to take action. When properly distinguished from, say, mere testing, all criticism is shown to contain an element of rebuke. To be criticized is to be held, to some extent, responsible for the shortcomings exposed; to be prompted not only to attend to those shortcomings, but in some sense also to mend our ways. Hence, all criticism, it is argued, comprises some measure of normative criticism. And because the aim of all prudent criticism is to convince its addressee to take action, to achieve its desired transformative effect, it must be endorsed by its addressee as self-criticism. Normative self-criticism is, therefore, at least to some degree, of the very nature of all criticism. Placing the onus of rationality on criticism, as we do, ultimately locates the transformative locus of rational reckoning in the realm of intra-personal deliberation, rather than inter-personal discourse. Participation in the public game of giving and critical motivation for doing so remains wholly implicit. But in later work, most notably in his Woodbridge Lectures, the emphasis is shifted away from an ability to keep score of other players' reasoning (which is dropped altogether), to the responsibility for articulating and integrating one's own commitments into an inferentially structured "synthetic unity of apperception" (e.g. 2009, 37) in dialogue with others. Here, we argue, the critical and selfcritical aspects of rational engagement become far more apparent in Brandom's exposition, but in a manner that wholly sidesteps the problem of normative self-criticism, because the critical dimension of normative change and development is reduced to troubleshooting for coherence.. 37 asking for reasons is certainly the sure sign of rational engagement. But the outward critical scrutiny of one another’s reasoning will have its rational transformative effect only when accompanied by a parallel, resonating self-scrutinizing of one's own commitments and entitlements. Indispensable as participation in the public game of giving and asking for reasons is to demonstrating one's rationality, it is only in the intra-subjective processes of normative self-scrutinizing that rationality can properly be said to reside and assert itself. In Chapter Eight we, therefore, turn our attention inward. Focusing on the so-called hierarchical account of personhood, identity, and normativity developed by Charles Taylor, Christine Korsgaard, and especially Harry Frankfurt, Chapter Eight explores the prospects and especially the limitations it imposes a person’s capacity for normative self-criticism. We find Frankfurt’s intra-personal, volitional account of human agency to resonate well, and to an extent to reflect, if implicitly, central aspects of the neoKantian, pragmatist and interpretivist pictures of inter-personal normative discourse explored in Part II. The problem is that no account of human selfhood will suffice of itself to adequately account for the possibility of genuine normative self-criticism – not even Frankfurt's for the simple reason that left to her own devices, a person is indeed incapable of taking genuinely critical normative stock of her own normative commitments. The central claim of Chapter Eight, which is the central claim of the entire study, is that a viably philosophical account of normative self-criticism does indeed present itself, but only when the two pictures are amalgamated: when a detailed Frankfurtian picture of the reflective self is grafted upon an equally detailed picture of the kind of critical discursive environment pictured by Brandom and Walzer. This has so far never been seriously attempted. Those, like Taylor, Korsgaard and Frankfurt, whose interest lies in the kind of reflective, intra-subjective self-reckoning central to questions of self, agency, and practical reasoning, have given little if any thought to the possible bearings on it of the external inter-subjective discursive contexts in which the selves they study constantly partake. Conversely, thinkers like Brandom, McDowell and Walzer, whose interests lie in the social, dialogical arenas of communal critical discourse, show no interest in following through their possible transformative impact on the personal 38 56 commitments and self-reasoning of individual discussants. Intra-subjective reflection and inter-subjective discourse are studied by philosophers quite independently, as if the extent and quality of a person's dialogical dealings had no bearing on his reflective capacities, and vice versa. Combining the two realms of discourse does not, however, yield as straightforward a solution to the problem of normative self-criticism as one might expect. For the very same reason it is impossible argue normatively against the norms to which one is committed, it is impossible to be convinced to do so by others. If normative selfcriticism is unthinkable, it remains unthinkable when leveled by others. But if normative criticism is incapable of convincing in what sense can it be said to be rationally internalized and endorsed? The answer proposed in Chapter Eight, follows on the heels of the conclusions of Chapter Seven. For their criticism to be endorsable by those they criticize, we argued there, prudent critics must base their case on premises their addressees hold true. But in the case of normative criticism, this is never an option. No premises exist to which a person is liable to agree, that entail a denunciation of his very norms. Sensing this, normative critics tend, we argue, to argue from premises close to what their addressees hold true, yet sufficiently different to allow their arguments to stick. (Arguing from the left, critics will surreptitiously premise certain liberal norms to make their case, while arguing from the right, they will tend to smuggle in just enough conservative value to make their point.) Normative criticism thus typically conveys a portrayal of it addressees' relevant commitments that differs minimally, yet crucially from their own self-image. Because normative criticism challenges heartfelt norms, it is powerless to convince, but if trusted, we argue at length, the discrepancy between the two portrayals, as in the case of a disturbing playback (an analogy explored in some detail), is capable of undermining its addressees' initial commitment, and rendering them ambivalent toward those norms. Herein, we argue, lies the destabilizing or 'ambivalating' potential 56 This, we argue, remains true of Brandom's later work in which a supposedly intrasubjective Kantian account of apperception is placed within an inter-subjective Hegelian context of communal appraisal. 39 of trusted normative criticism. And, as Frankfurt famously argues, norms to which one becomes ambivalent lose their wholehearted volitional grounding, and normative hold, and demand to be reassessed. Chapter Nine purports to take stock of the book's project by revisiting the dynamics of scientific framework transitions in the light of the conclusions of Chapter Eight. If, as McDowell puts it, an ability to "reflect about and criticize the standards by which, at any time, [we take ourselves] to be governed. … from the midst of the way of thinking 57 [we are] reflecting about" , becomes a philosophically articulable option only when a critically motivated version of Brandom's inter-personal deontic interrogation is allowed to bear on a Frankfurtian normative dialogue of self (or some similarly hierarchical equivalent), then, to be considered rational, Friedman's Kuhnian picture paradigm shifts requires substantial supplementation on both counts. Friedman, as argued in Chapter Five, reduces the problem of rational framework replacement in science to that of communicating rationally across an already existing divide. We take it primarily to be that of the rational incentive to create a new framework in the first place. To answer this question, one cannot avoid asking, what fault such a person could have found in the framework to which he was still committed? - which in the light of the conclusions of Chapter Eight, we argue, can now be addressed from within Friedman's approach, in ways that have so far been unavailable from that perspective. It is not an easy fit, however. As we have seen, the sort of self-critical disposition needed for setting a paradigm shift in motion, requires the challenging environment of trusted, potentially ‘ambivalating’ criticism. With all due respect to the importance Walzer assigns to the "connectedness" of effective critics, from the neo-Kantian perspective we share with Friedman, such criticism will clearly not be forthcoming from within the 'normal,' paradigm-governed discursive settings of one's own field. The initial source of destabilization has to be external. On the other hand, a scientific field can only be transformed from within. Only physicists can change physics! Accordingly, Chapter Nine proposes in rough outline a two-stage format for studying scientific paradigm shifts, in which practitioners of voice and standing are first "ambivalated" by 57 McDowell (1994, 81). 40 external critics, and later (inadvertently) succeed in promulgating their newfound indecision within their communities by means of a particular form of published work. Both stages represent irreducible moments of intense intra-subjective deliberation, that do not admit easily to the collectivist modes of explanation that have come to dominate science-studies since Kuhn. To account for the first stage, we propose an extension and reworking of Peter Galison's novel notion of the scientific "trading zone" – the 'locales' of professional engagement outside one's home community, as when bidding for financial support, "trading" with neighbor disciplines for techniques or instruments, offering scientific opinion, engaging students, and so forth. The point we stress is how when "trading" abroad, one is frequently exposed to the friendly and trustedly bemused questioning of genuinely curious professionals, on whose thinking the framework constitutive of one's own may have far less of a grip. Here, we show, lies the destabilizing and "ambivalating" potential of science's external critics. The second stage, is achieved when the doubts and indecision of duly "ambivalated" individuals come tacitly to inform their creative efforts to overcome them. Such efforts are shown to typically take the form of uneasily split, hybrid attempts to re-represent their field's basic assumptions by retaining some of the old while groping imaginatively toward other possibilities. When analyzed prospectively, such works can be seen to represent anxious, yet highly creative departures from heartfelt commitments capable of motivating others to seeker cleaner, more radical breaks with the old. Classical examples we mention briefly are Tycho’s planetary theory, Galileo’s analysis of projectile motion, and Poincare’s geometrical conventionalism, all of which we show unwittingly preserved, and, therefore, propagated the keen ambivalence that begot them, prompting others to take a firmer stand. The philosophical problem this study raises and purports to address – that of articulating the possibility of normative self-criticism from a broadly neo-Kantian perspective – is ruled by those who deny Normative Diversity to be misconceived, by others (Popper's school, for instance) trivial, but is taken by most (explicitly or implicitly) to be insurmountable. Because we concede Normative Diversity, and refuse 41 to limit self-criticism to instrumental reason, in what follows the work of the first two 58 groups is largely ignored . We contest the third group's conclusions, but not by accusing them of misconceiving the problem or of reasoning it through mistakenly. For we fully concede their claim (or assumption, or tacit presupposition) that, of her own accord, a person is indeed incapable of mounting a normative argument against her very norms, and consequently, of being convinced to do so by others. If we accuse them of anything, it is of oversight, of failing to notice how the destabilizing impact of rational discourse can extend rationally beyond the realm of reasoned exchange. This is the modest undertaking of the chapters that follow. 58 With regard to science, Popper and his school arguably fall in the first group - denying Normative Diversity in science, and treating all elements of scientific knowledge as unproblematically criticizable. Lakatos, whose philosophy of science retains much of Popper's, departs from him sharply on this point by at least acknowledging the existence of a potentially different "hardcore" of empirically irrefutable beliefs at the heart of every "scientific research programme". Lakatos, however, (not unlike Quine) overlooks the constitutive role of the hardcore with regard to the programme's succession of theories, concentrating exclusively on its immunity to empirical test. In his view, therefore, an entire research programme can be rationally faulted - hardcore included - by force of instrumental considerations of fruitfulness operating between programmes. Interestingly, in the Open Society, Normative Diversity is not only acknowledged by Popper with regard to ethical and political ways of life, but is prized for its critical, rationally, self-correcting potential. But because his philosophy of science motivates his political theory, rather than the other way round, the problem of normative self-criticism remains unacknowledged. The View from Within Normativity and the Limits of Self-Criticism Menachem Fisch and Yitzhak Benbaji "This is a bold and wide-ranging book that offers a novel solution to a central problem of philosophy: if there is no normatively neutral language in which to compare normatively distinct vocabularies, how can transitions from one vocabulary to another ever be rational? Combining great analytic subtlety with deep knowledge of the history of science, Fisch and Benbaji argue that a central role is played by the ambivalence induced in insiders when they engage external critics within the "trading zones" of discourse. A tour de force, this book sheds new light on many areas of philosophy. Indeed, by examining the role of familiar phenomena that philosophers often neglect, such as ambivalence and indecision, The View from Within illuminates the destabilizing as well as the creative potential of reason throughout human life." —Paul Franks, Yale University "How can one change one’s mind about the very standards one applies as a critical thinker without losing a grip on one's reasons? Fisch and Benbaji assess the state of the question in a remarkably wide range of fields: Kuhnian philosophy of science, interpretative social theory, pragmatism from Rorty to Brandom, and Frankfurt's philosophy of personal identity. Then they offer an answer of their own, which integrates a social account of rationality as a trait agents exhibit when exchanging reasons with one another and a subject-centered account of rationality as a trait agents exhibit when criticizing their own commitments from within. The result is a fresh and illuminating approach to the nature of rationality and normativity." —Jeffrey Stout, Princeton University “The View from Within is a thorough evaluation of the arguments made by contemporary philosophers about the normative character of reason and the derivative problem of relativism. Fisch and Benbaji have admirably compared and contrasted competing positions, and with a balanced critique, they have made a sustained effort to ‘save’ rationality and provide new guideposts for its philosophical evaluation. A timely and important contribution.” —Alfred I. Tauber, Boston University Menachem Fisch is Joseph and Ceil Mazer Professor of History and Philosophy of Science at Tel Aviv University in Israel. Yitzhak Benbaji is associate professor on the law faculty and in the philosophy department at Bar-Ilan University in Israel.