teorema Vol. XXXV/2, 2016, pp. 29-39 ISNN 0210-1602 [BIBLID 0210-1602 (2016) 35:2; pp. 29-39] On Alethic Functionalism’s (Absurdly?) Wide Applicability Marc Champagne RESUMEN El funcionalismo respecto a la verdad es una tesis propuesta por Michael Lynch según la cual la verdad tiene diferentes formas de manifestarse que sin embargo contienen suficiente unidad como para evitar el pluralismo a secas. Es una tesis con muchas virtudes. No obstante, dado que la estrategia de “la unidad en la diferencia” de Lynch podría aplicarse a otros debates, mostraré que su argumentación sirve para solucionar, no sólo la controversia acerca de la verdad, sino cualquier controversia generada por una pregunta del tipo “¿qué es X?” PALABRAS CLAVE: verdad, funcionalismo, argumento. ABSTRACT Alethic functionalism, as propounded by Michael Lynch, is the view that there are different ways to be true, but that these differences nevertheless contain enough unity to forestall outright pluralism. This view has many virtues. Yet, since one could conceivably apply Lynch’s “one and many” strategy to other debates, I try to show how his argumentative steps can be used to solve—not just the controversy pertaining to truth—but any controversy that surrounds a “What is X?” question. KEYWORDS: Truth; Functionalism; Argument. I Functionalism, in philosophy of mind, is motivated in part by the idea that two creatures from different species could exhibit all the markers of pain, without having the same brain configuration and/or physical makeup [Putnam (1975), p. 436]. Drawing inspiration from philosophy of mind, Michael Lynch (2004) proposes that “truth” is a common function, realized differently in different discursive domains. There are presumably scientific truths, ethical truths, mathematical truths, aesthetic 29 30 Marc Champagne truths, and so on. Even so, these varieties are said to share a set of core features. Thus, armed with the functionalist notion of multiple realizability, Lynch holds that “[w]e can be monists about the concept of truth while being pluralists about its underlying nature” [Lynch (2001b), p. 727]. Yet, one can ask whether there is any principled way to stop one from making a similar “one and many” move to resolve other disagreements. I do not think there is, so I will try to show how the generic steps of Lynch’s argumentation can yield a solution to any philosophical controversy –– not just the controversy pertaining to truth. This wide applicability can be grounds to suspect that something is amiss in Lynch’s proposal. Indeed, one could argue that the functionalist arguments developed by Lynch, once generalized, become too successful. A modest reductio ad absurdum critique would show that the consequence of wide applicability is unpalatable/improbable, whereas a more ambitious reductio would show that the consequence is contradictory. I will attempt neither of those things. While I find the wide applicability of alethic functionalism to be absurd (in the modest sense just outlined), my title quarantines that adjective in parentheses and guards it with a question mark. I will conclude by gesturing at what I think might be wrong. But, as far as the bulk of this article is concerned, maybe a “one and many” stance can settle every debate. II According to Lynch, the realizers of truth “are always realizers for a discourse” [Lynch (2001b), p. 741; emphasis in original]. Lynch believes that most theories of truth eventually founder because they try to impose their account on inapt discursive domains. On his diagnosis, the major theories of truth tend to over-exploit a limited range of examples (e.g., “The cat is on the mat”) at the expense of more challenging cases drawn from other domains. Lynch thus maintains that “[t]he usual options [...] differ chiefly in the intuitions they privilege” [Lynch (2011), p. 2]. It is not so much that traditional theories like correspondence and coherence are wrong in what they advocate. Rather, they are wrong in inferring that since what they advocate works so well for one particular area, it ought to apply across the board. Following Cory Wright (2012), p. 89, we can call this the uniformity assumption. An alethic functionalist rejects this assumption. Discarding the uniformity assumption does not entail relinquishing uniformity altogether. Lynch argues that we should fashion a theory that i) respects the diversity of truth claims, while ii) holding on to the com- 31 On Alethic Functionalism’s (Absurdly?) Wide Applicability mitments that these claims purportedly share. From a programmatic standpoint, then, alethic functionalism purports to have gleaned an important lesson from past attempts to define truth — which in turn allows it to overcome their respective shortcomings. According to Lynch, “[e]ven partisans must acknowledge that both intuitions – about diversity and unity – are pre-theoretically appealing” [Lynch (2011), p. 2]. It is the business of the alethic functionalist to achieve this balance of diversity and unity. Lynch hopes that, by watering down expectations about what constitutes the nature of truth, he will find a consensus that drives a skewer through all the theoretical candidates, without exception. The mere fact that contenders are seen as contenders, Lynch (2001b), p. 725 argues, speaks their enjoying a minimal unity. Lynch, who has edited an anthology on truth (2001a), induces three “core truisms” from all the debates: truths are mind-independent, correct to believe, and usually a worthy goal of inquiry [Lynch (2011), pp. 8-12, 70]. These are, in his words, “platitudes” we can presumably all agree on. They are said to square not just with specialized philosophical usage, but with folk beliefs as well (although, in fairness, Lynch never offers any experimental support for this). Lynch stresses that anyone who would dissent from this short list of truisms would, by that very dissension, be confessing an interest in some topic other than truth [Lynch (2011), p. 13]. As we shall see, this ends up having considerable strategic value. In any event, the rest is left open to debate: the principle of bivalence, for instance, is regarded by Lynch as an optional commitment [ibid., p. 16]. The truisms are nevertheless meant to prevent the pluralist element from devolving into outright relativism, which Lynch calls “simple alethic pluralism” [ibid., p. 54]. The alethic functionalist believes that truth can be multiply realized. In his more recent work, Lynch prefers to speak of truth as being “manifested” in different contexts. He writes that “[t]ruth is many because different properties may manifest truth in distinct domains of inquiry. [...] Truth is one because there is a single property so manifested, and ‘truth’ rigidly names that property” [ibid., p. 78]. Hence, just as pain can be realized in mammals and molluscs, so can truth be realized in normative and descriptive contexts alike. Lynch contends that “it is obvious that the proposition that two and two are four is fundamentally different in kind than the proposition that torture is wrong. But making out what this difference consists in is hardly the sole job of the functionalist about truth” [Lynch (2011), p. 79]. In keeping with the biconditional schema, the erasure of quotation marks is supposed to allow one to glean the relevant truth conditions. “Picasso 32 Marc Champagne is the greatest painter” is true if and only if Picasso is the greatest painter, “There is an elephant in Montreal” is true if and only if there is an elephant in Montreal,” “Witches are wicked” is true if and only if witches are wicked, and so on. This shows the laxity afforded by the “many” portion of the functionalist account: so long as a given proposition asserts something, either directly or at a remove [ibid., p. 130-131], it will be considered truth-apt (which of course does not mean that the proposition in question will turn out to be true). If Lynch is right, then apart from alethic functionalism, every theory of truth currently on the books has a limited scope. Most theories privilege claims about material domains (like cats on mats), but forget to accommodate claims about non-material domains (like ethical appraisals). The variability of manifestation allowed by Lynch’s functionalism means that no constraint whatsoever is placed on the sorts of truths one could seek out. “Moral beliefs, for example, might be true, and therefore have the contents they have, in virtue of being superwarranted, while beliefs about physical objects might be true by corresponding to the facts about those objects” [Lynch (2011), p. 50]. However, this “scope problem” [Lynch (2011), pp. 32-36] is not self-evidently a “problem.” A correspondence theorist, for instance, does not dwell on material truth-makers because she neglects or overlooks ethics. Rather, she tethers her theory to material truth-makers because she thinks that these are the only sort of truth-makers to be had. An ethicist may complain that this prevents claims in her domain from ever being true, but it is not clear why a correspondence theorist should be bothered with what lies outside the ambit of her commitments. “What we are faced with, then, is a dilemma: either accept the counterexamples [...] and expand our account of truth [...] or maintain the naturalness of truth by rejecting the counterexamples” [Edwards (2013), p. 395]. Capitalizing on reigning trends, Lynch simply favours the first response. III Looking closely at the arguments used by Lynch to promote alethic functionalism, we can extract a generic sequence, almost like a recipe. First, take a philosophical debate over the nature of X, which over time has splintered into an array of seemingly irreconcilable proposals about what X might be. Line these positions up in a neat historical queue, so as to stress how far the dialectic has travelled. Then, building on an understandable retrospective sense that all avenues have been ex- 33 On Alethic Functionalism’s (Absurdly?) Wide Applicability plored with no tangible headway made, suggest, as Lynch does, that “[t]he history of attempts to identify the property” of X “has not been a happy one” [Lynch (2001b), p. 723]. Like Lynch, one need not provide independent reasons for rejecting each account of truth, as a mere allusion to presumably well-known flaws suffices. Conclude from this impasse that “[t]he usual options have come to seem [...] somewhat tired” [Lynch (2011), p. 2], such that we now need to reconsider the situation anew. It is crucial to acknowledge that past efforts bore some fruit — just not enough to settle the issue of what X is in the definitive. This helps reassure the proponent of a given theory that she has not wasted her time. Still, in order to loosen allegiance to a given -ism and prime the migration in commitments soon to come, prompt a mild sense of methodological guilt by calling attention to the fact that proponents of a previous theory of X have tended to privilege some examples over others, and have issued far too many promissory notes. Clearly, if the theory in question was truly robust, it would not have shied away from engaging with more challenging case studies, nor would it have failed to carry out its programmatic ambitions by now. Once confidence in the addressee’s favoured position has been suitably undermined, recommend a fresh start, beginning only with the most secure basics. A fresh look at the X question having been granted, catalogue a host of platitudinous observations about X. These platitudes must be carefully chosen, since each must fit without resistance in the following: “Surely no one would deny that to be X is to be [fill in blank].” Successfully negotiating this step requires an intimate familiarity with the major proposals on the issue. Even so, the functionalist needs a backup tool to help her brush aside potential disagreement with her chosen truisms. Lynch, as we saw, has an effective one: mention in an open-minded tone that anyone is free to disagree — yet insist with firmness that those who do not acquiesce before the list of truisms do so at the price of “changing the subject” [Lynch (2011), p. 13] from X to something else. With this exclusionary clause in place, real disagreement promises to pose no obstacle. Once a consensual core of unity has been induced from extant debates, propose that X is “realized” or “manifested” in multiple ways, such that apparent disagreements between thinkers are in fact just surface symptoms of differing domain emphases, wrongly taken as paramount. Since all participants in the debate were discussing the common topic X, stress that there was a latent form of unity which, until now, went unseen, such that “most players in the contemporary debates over 34 Marc Champagne [X] share an unnoticed allegiance to a certain type of monism” [Lynch (2001b), p. 725]. Just in case this unification begins to clash with the sense of historical despair fostered earlier, the functionalist must stand ready to answer those who conclude that X has no nature. Since such a “deflationary” stance has become “[t]he new orthodoxy” [Lynch (2011), p. 4], it requires special attention before being discarded. Strategies that can be used to achieve this include (but are not limited to): asking whether one can coherently state such cynical misgivings at all [ibid., p. 109]; claiming that those who disparage the X question are not in fact talking about X [ibid., p. 110]; reproaching dissenters for not “pay[ing] sufficient homage” to the intuitions people usually have about X [ibid., p. 110]; and finally issuing a vague warning that those who challenge the legitimacy of the topic “must remove [X] from the philosopher’s toolbox” [ibid., p. 113]. Should one have difficulty articulating the specifics in convincing detail, insist that the task of figuring out which manifestation of X is appropriate in a given instance is a difficult accomplishment, and while “[f]unctionalism does not dictate in advance how [X] will be realized in various discourses” the theory may nevertheless “act as a neutral frame for a less reductive picture [...]” [Lynch (2001b), pp. 745-746]. Alethic functionalism thus exhibits a high degree of resilience. Looking at the past, present, and future, either a new piece of information is absorbed by the alethic functionalist, or it is relegated to another topic. What if an unforeseen account of truth should surface in the future? Suppose, for example, that philosophers inspired by the study of so-called “memes” come to think of truth in terms of replication, such that we eventually see “replicationism” emerge as a new option alongside staples like “pragmatism,” “coherentism,” etc. In keeping with the argumentative instructions that have just been laid out, the alethic functionalist could argue that a replicationist account of truth, while ill-equipped to account for some cases (say, singular truth-bearers that do not involve any chain or series), is nevertheless well-suited to study certain objects. It is just a matter of matching the right theory to the right domain. Hence, a novel account of truth would add more grist to the functionalist mill. Of course, a new theory of truth would have to agree with the core truisms and the biconditional schema [Lynch (2004), pp. 405-406]. Yet, even these could conceivably be modified in order to save the functionalist account. The truisms, according to Lynch, are supposed to reflect a certain portion of our actual truth talk. Given this journalistic recording of “historically prominent folk truisms” [Lynch (2011), p. 13], the results 35 On Alethic Functionalism’s (Absurdly?) Wide Applicability gleaned could change, were the inductive base to undergo sufficient change. Indeed, “the informality of platitudes increases the base of admissible information” in a way that “would otherwise be omitted in a more stringent analysis of only necessary and sufficient conditions” [Wright (2005), p. 12]. It might be, for example, that we as speakers come to reject the idea that true beliefs are a worthy goal of inquiry. In that case, the functionalist account would still survive, provided it trimmed its truisms to reflect this. This contingent inquiry into linguistic practices thus allows for ample revisions. Not only is the “one” portion flexible, the “many” manifestations of truth are also flexible. If a given theory of truth currently under the umbrella of Lynch’s account were to suddenly lose all of its credibility, the functionalist framework would nevertheless emerge unscathed. In fact, entire discursive domains could presumably open and close while the functionalist account remains standing. We see, then, why and how “functional specifications are easy to proliferate—formally speaking, one need only begin altering the identity- and truth-conditions of the content constituting them” [Wright (2010), p. 282]. IV I think that, because the placeholder “X” could be replaced with any object of philosophical controversy, the recipe that I have just outlined can solve more problems than Lynch realizes. So, to illustrate functionalism’s (absurdly?) wide applicability, I now want to apply Lynch’s argumentative steps to another case, namely “What is philosophy?” It is widely accepted that, “[w]here the predication of distinct kinds of truth entails variability in our use of the term ‘true,’ strong alethic pluralism amounts to the promotion of a polysemic account of truth” [Wright (2005), p. 8]. Accounts of philosophy certainly seem plagued by a similar polysemy. Depending on who one asks, “What is philosophy?” gets the following answers: philosophy is the clarification of linguistic confusion, philosophy is the love of wisdom, philosophy is the attempt to understand the success of science, philosophy is the handmaiden of theology, philosophy is a preparation for death, and so on. In a move reminiscent of deflationism, Richard Rorty (1978) responded to this eclecticism by saying that, while we can certainly track how the signifier “philosophy” contingently drifted through Western cultures over time, looking for some common signified or essence would be misguided. The functionalist could reply that, just as pain can be realized 36 Marc Champagne in mammals and molluscs, so can philosophical activity be realized in artistic and scientific domains alike. Using Lynch’s moves, one could claim that functionalism succeeds in accounting for the elusive nature of philosophy “without committing the serious blunder of identifying” philosophy with a property that stays constant “across the board” [Lynch (2011), p. 180]. Let us therefore see how the “one and many” recipe could explain this case. Taking our clue from Lynch, the first step is to remark how debates over the nature of philosophy have splintered into an array of seemingly irreconcilable proposals. Building on an understandable sense that all avenues have been explored with no tangible headway made, we can call attention to the fact that proponents of a given theory tend to privilege some examples over others. Indeed, adherents to analytical definitions of philosophy are unlikely to consider voluntary starvation and selfimmolation [Bradatan (2015)], just as adherents to life-style definitions of philosophy are unlikely to consider analyses of the words “this” and “that” [Kaplan (1979)]. Clearly, if a candidate definition were truly robust, it would not ignore examples that challenge it. What is needed, then, is a fresh start, beginning only with the most secure basics. The next step is to catalogue a host of platitudinous observations about X, in this case philosophy. One must be familiar with the major proposals to select this handful of truisms. But, as a tentative list, surely no one would deny that philosophy is 1) conducted by humans, 2) expressed in language, 3) important to many people, 4) demanding, 5) related to argumentation. These “platitudes” try to square, not just with specialized usage, but with folk usage as well (although, like Lynch, I will not offer any experimental evidence to support this). Picking the right set of truisms is probably the hardest step, since some people are always bound to disagree. For instance, what if one points to the fact that some philosophers like Diogenes the Cynic or Jacques Derrida have refused to argue for their views, thereby rejecting truism (5)? Lynch, as we saw, has an effective response to such a challenge: mention in an open-minded tone that anyone is free to disagree — yet insist with firmness that those who do not acquiesce before the aforementioned truisms do so at the price of “changing the subject” from X to something else. So, following Lynch, we can stress that anyone who would dissent from our short list of truisms about philosophy would, by that very dissension, be confessing an interest in some topic other than philosophy. The rest, though, is left open to debate: specific prose style, for instance, can be regarded as an optional commitment. Paraphrasing 37 On Alethic Functionalism’s (Absurdly?) Wide Applicability Lynch, we can hold that “it is obvious that [a preparation for death] is fundamentally different in kind than [a justification of science]. But making out what this difference consists in is hardly the sole job of the functionalist about [philosophy]” [Lynch (2011), p. 79]. What if an unforeseen account of philosophy should surface in the future? Recently, philosophers inspired by the methods of the social sciences have come to think of philosophy in terms of experimentation, such that we now see “experimental philosophy” (or “X-Phi”) emerge as a new option alongside staples like analysis, love of wisdom, etc. In keeping with the generic argumentative pattern used to defend alethic functionalism, one could argue that this new experimentalist account of philosophy, while ill-equipped to account for some cases (say, claims that lie on the “ought” side of the is/ought distinction), is nevertheless wellsuited to study certain objects. It is just a matter of matching the right theory to the right discourse. Of course, any new theory of philosophy would have to agree with the core truisms that I have identified. Yet, even these could conceivably be modified in order to save the functionalist account of philosophy. The five truisms are supposed to reflect our actual talk about philosophy. Because the list is gleaned by induction, it could change, were its inductive base to undergo sufficient change. It might be, for example, that philosophers eventually come to reject the idea that expression in language is an essential part of their craft. In that case, the functionalist account could still survive, provided it trims its truisms to reflect this. Also, if a given take on philosophy were to suddenly lose all its credibility, the functionalist framework would emerge unscathed. As mentioned, functionalism is highly flexible. V Is all of this right? Has the notion of multiple realizability really allowed me to solve the long-standing debate over the nature of philosophy? If I had to bet, I would say that an error in reasoning is more likely than a functionalist panacea. Hence, to my mind, the unforeseen consequence of wide applicability should prompt a recoil — a healthy suspicion that something must have gone wrong. What might this error be? Shapiro writes: “Surely, the functionalist takes the functionalist theory to be true. What is its truth-realizer?” [Shapiro (2011), p. 44]. Leaving this interesting bootstrapping to the side, I would flag a sensible maxim that can easily generate a false sense of 38 Marc Champagne success: either there are different views on a common topic or, if the differences are too deep, there are different topics. Obeying this simple disjunction can easily fool one into thinking that one has unearthed a triumphant solution. As we have seen, unusual cases like Diogenes and Derrida cannot topple a truism about philosophy requiring argumentation, because the very failure to realize this core truism ejects those figures from the rank of philosophers. Yet, even with these ejections, philosophy remains incredibly diverse. This ought not bother the functionalist however, since the mere fact that rivals rally under a common banner shows that a common ground can be had: “After all, the idea that there is more than one way for propositions to be true just implies that there is something these ways have in common that makes them all worth calling ‘ways of being true’” [Lynch (2005), p. 42]. That may be, but topical unity – which can be glossed merely as membership to an arbitrarily assembled set [Pedersen (2010), p. 94] – is arguably the least interesting sort of unity to be had. It is uninformative, then, for Lynch to assert that “most players in the contemporary debates over truth share an unnoticed allegiance to a certain type of monism” [Lynch (2001b), p. 725]. Of course they do –– if we relax monism enough. A major selling point of alethic functionalism, according to Lynch, is that it does not require one to abandon any theory, provided that one recognizes the domain-specific limitations of that theory. Yet, a falsificationist reflex in me is not overly impressed by what a theory includes, since I also want to be told what it excludes/forbids. With its shifting admixture of monism and pluralism and relegation of all serious challenges as off-topic, it is hard to see how alethic functionalism could ever be shown wrong. Now, for all I know, my (bracketed) sense that this is absurd might itself be wrong, and rallying all disagreeing parties with multiply realized truisms might well turn out to be a panacea. Either way, this would make the wide applicability of functionalism worthy of further scrutiny. Department of Philosophy, Trent University 1600 West Bank Drive, Peterborough, Ontario, Canada, K9L 0G2 E-mail: marcchampagne@trentu.ca 39 On Alethic Functionalism’s (Absurdly?) Wide Applicability ACKNOWLEDGEMENTS For helpful feedback on earlier versions, I want to thank Henry Jackman, Dilip Ninan, anonymous reviewers for Teorema, and audience members at the University of St. Andrews. REFERENCES BRADATAN, C. (2015), Dying for Ideas: The Dangerous Lives of the Philosophers; London, Bloomsbury. EDWARDS, D. (2013), “Naturalness, Representation and the Metaphysics of Truth”; European Journal of Philosophy, vol. 21, no. 3, pp. 384-401. KAPLAN, D. (1979), “On the Logic of Demonstratives”; Journal of Philosophical Logic, vol. 8, no. 1, pp. 81-98. LYNCH, M. P. (ED.) (2001a), The Nature of Truth: Classic and Contemporary Perspectives; Cambridge, MA, MIT Press. –– (2001b), “A Functionalist Theory of Truth”; in The Nature of Truth: Classic and Contemporary Perspectives, edited by Michael P. Lynch, Cambridge, MA, MIT Press, pp. 723-749. –– (2004), “Truth and Multiple Realizability”; Australasian Journal of Philosophy, vol. 82, no. 3, pp. 384-408. –– (2005), “Functionalism and our Folk Theory of Truth”; Synthese, vol. 145, no. 1, pp. 29-43. –– (2011), Truth as One and Many; Oxford, Oxford University Press. PEDERSEN, N. J. L. L. (2010), “Stabilizing Alethic Pluralism”; The Philosophical Quarterly, vol. 60, no. 238, pp. 92-108. PUTNAM, H. (1975), Mind, Language and Reality; Cambridge, Cambridge University Press. RORTY, R. (1978), “Philosophy as a Kind of Writing”; New Literary History, vol. 10, no. 1, pp. 141-160. SHAPIRO, S. (2011), “Truth, Function and Paradox”; Analysis, vol. 71, no. 1, pp. 38-44. WRIGHT, C. D. (2005), “On the Functionalization of Pluralist Approaches to Truth”; Synthese, vol. 145, no.1, pp. 1-28. –– (2010), “Truth, Ramsification, and the Pluralist’s Revenge”; Australasian Journal of Philosophy, vol. 88, no. 2, pp. 265-283. –– (2012), “Is Pluralism about Truth Inherently Unstable?”; Philosophical Studies, vol. 159, no. 1, pp. 89-105.