Williams’s Pragmatic Genealogy and Self-Effacing Functionality

Author: Matthieu Queloz
Published in: Philosophers’ Imprint 18 (17): 1-20. 2018.
DOI: 2027/spo.3521354.0018.017
Canonical entry: https://www.matthieuqueloz.com/entries/williams-s-pragmatic-genealogy-and-self-effacing-functionality/
Published PDF: https://philpapers.org/archive/QUEWPG.pdf

Machine-readable text companion generated from the PDF. Page markers follow the journal pagination.

[p. 1]

INTRODUCTION

Bernard Williams’s last book, Truth and Truthfulness, left posterity much to puzzle over. Here is the arch-critic of utilitarianism seeking to vindicate the intrinsic value of truth in terms of its instrumental value, thereby entrenching himself squarely in the traditional territory of indirect utilitarianism. What is more, he proposes to do so using the method of genealogy, which suggests a historical approach — but he also denies that the concept of truth has a history, and prefaces his historical remarks with an avowedly fictional state-of-nature story. Unsurprisingly, this has raised questions. What separates Williams’s instrumental vindication from indirect utilitarianism?1 And how can genealogy vindicate anything, let alone something which does not have a history?2 These questions — and Truth and Truthfulness as a whole — have not received the attention they deserve. One reason for this may be that the book was untimely. Upon its appearance in 2002, it seemed ostensibly directed against over-enthusiastic forms of postmodernism which soon appeared to be already on the wane. But a decade and a half later, the advent of “post-truth” politics vindicated Williams’s sense that the value of truth needed defending. And this rich exploration of the point of valuing the truth anyway rewards engagement with the puzzles it raises. My aim in this paper is to resolve some of these puzzles in order to develop an understanding of Williams’s genealogical method that reveals it to be uniquely suited to dealing with what I call self-effacing 1. McGinn (2003).

2. Nagel (2009, 134) expresses puzzlement over the project of vindicating through genealogy. Koopman (2013, 20, 64–5, 74, 87) even charges Williams with committing the genetic fallacy. McGinn (2003) finds the genealogical story redundant, given that instrumental considerations are supposed to vindicate. Rorty (2002) confesses himself unable to see the relation between the fictional and the historical parts of the book. In general, what exactly the book’s “circuitous” (Elgin 2005, 343) argument is supposed to be has been contested. Reactions have ranged from hailing it as “the most interesting set of reflections on the values of truth and truth-telling in living memory” (Hacking 2004, 137) to questioning whether the book is more than “a collection of loosely related essays on truth” (Fleischacker 2004, 382). The first monograph on Williams calls it “a collection of interesting intellectual tributaries feeding a somewhat elusive main channel” (Jenkins 2006, 163).

[p. 2]

functionality: the phenomenon whereby a practice is functional, but we do not engage in it for its functionality, and it is only functional because we do not engage in it for its functionality. Key to this reading will be, first, to get clear about what exactly Williams’s genealogy is a genealogy of;3 and second, to read Williams as a type of pragmatist — notwithstanding the fact that ‘the pragmatists’ is the label he gives to his opponents. Williams’s genealogy merits the label of a ‘pragmatic genealogy’ in two respects: it focuses on the practices revolving around truth instead of asking what truth itself is; and it explains these practices in terms of their point for creatures like us. Viewing Williams’s genealogy as a pragmatic genealogy allows us to see it not as a piece of erudite historical stage-setting, but as a direct answer to Richard Rorty’s question: Why should we value the truth? While Rorty concludes that we are better off dropping the notion of truth altogether,4 Williams’s genealogy offers a contrary answer by displaying the instrumental value of valuing the truth intrinsically. The genealogy is a perspicuous derivation, from needs we have anyway, of the need for an intrinsic value of truth — and in showing the need for the value of truth to be rooted in practical exigencies, Williams proves Rorty wrong by his own lights.

The paper falls into three parts, the first two setting the stage for the third. In §1, I tackle the surprisingly tricky question of what Williams’s genealogy is a genealogy of. The difficulty, I argue, stems from the fact that truthfulness plays a double role as both act and object of valuation. I argue further that Williams’s defence of the value of truth turns on the idea that while we cannot abandon the concept of truth, we can abandon our concern for it, and this is why he eschews the question of what truth is for the question of why we should cultivate truthfulness. In §2, I argue that Williams’s genealogical method is best understood as continuing the pragmatist tradition that has been called “Cambridge pragmatism”, and I delineate how the state-of-nature fiction relates to historical considerations. In §3, using three challenges raised by Colin McGinn as a foil, I show in what sense an intrinsic value can be vindicated through pragmatic genealogy. I distinguish this type of vindication from indirect utilitarianism, and I argue that Williams is concerned with a form of self-effacing functionality that genealogy is uniquely suited to dealing with. I conclude with an assessment of the wider significance of Williams’s genealogy both for his own oeuvre and for further genealogical inquiry.

1. WHAT IS WILLIAMS’S GENEALOGY A GENEALOGY OF?

It is far from straightforward to say exactly what the subject matter of Williams’s genealogy is. Some commentators present it as a genealogy of truth, and the Italian translation of the book is even titled Genealogia della verità.5 Others urge that it is a genealogy of truthfulness as opposed to truth.6 The reading I offer in this section splits the difference: it presents Truth and Truthfulness as a genealogical explanation of why we might have come to value truth intrinsically, where this means valuing the various states and activities expressive of truthfulness intrinsically. Williams’s repeated insistence that the concept of truth does not have a history (TT 61, 271) might be taken to entail that there cannot be a genealogy of truth. But does the fact that a concept lacks a history really exclude giving a genealogy of that concept? Williams’s reason for maintaining that the concept of truth lacks a history is that it is an indefinable part of a ramifying set of connected notions which play a basic role in language and thought.7 This role “is always and everywhere the same” (TT 61, 271). Truth is a formal concept that we cannot 5. Blackburn (2002) and Mordacci (2016).

6. This emphasis on truthfulness as opposed to truth is found in Honderich (2003, 140), Hacking (2004, 157), Elgin (2005, 67), and Koopman (2013).

3. Blackburn (2002) and Mordacci (2016) present it as a genealogy of truth, while Honderich (2003, 140), Hacking (2004, 140), Elgin (2005, 344), and Koopman (2013) insist that it is a genealogy of truthfulness as opposed to truth. The reading I offer splits the difference.

7. Koopman ascribes to Williams the view that “the truth itself” does “not vary with history” (2013, 69). But Williams only takes the concept of truth to be historically invariable; the truth itself will be the truth about a matter to hand, and as historically variable as the matter itself (see TT 257).

4. See especially Rorty and Engel (2007) and Rorty (1998), but also (1989, 1991).

[p. 3]

help but live by. This is not an empirical claim — Williams takes it to follow from Davidson’s (1990) hermeneutical insight that the concept of truth is essential to understanding something as a believing or saying that things are thus-and-so, and therefore to understanding any kind of variation in such believings or sayings. In recognising something as a representation of things as being thus-and-so, we are always already relying on the concept of truth in this basic role. Williams thus holds: (i) that the concept of truth has no history; and (ii) that truth is an indefinable part of a set of connected notions, such as belief and assertion, which have to be learnt together.8 Does it follow that no genealogy of the concept of truth can be given? It certainly does if one equates genealogy with history. But on Williams’s own conception of genealogy, it does not follow. For while genealogy is potentially informed by history, it remains distinct from it. A genealogy of the concept of truth can still be given, but because of (ii), we should eschew attempts to spell out what truth is in favour of asking why creatures like us might have developed the concept of truth (along with the concepts it is connected to); and because of (i), the genealogy has to be a fictional developmental model. As a model, it could start off with agents represented as lacking what historical communities always already possess, and it could represent as successively emerging what in fact has to arise together. It might thereby exhibit the point of the concept of truth and the basic human needs from which it derives. It is a mark of the power of genealogy, for Williams, that it allows us to tell a developmental story highlighting why X should have arisen even where we find it barely intelligible that there could have been a development towards X.9 There is thus nothing here to prevent us from asking such questions as “What are the concepts of truth and falsity for?” (Price 2011, 76). But when Price argues that one function of the concept of truth is to encourage us to resolve disagreements in order to reap the benefits of arguing about how things are, he is, by Williams’s lights, talking about the practical significance of truthfulness.10 Similarly, the histories of truth offered by Foucault (1994) and Shapin (1994) are best understood as being about truthfulness. It is a different enterprise again, but an equally viable one, to explain why a community which did not have a word for truth might be led to introduce one. Being able to describe things as true may function as a device of indirect reference and generalisation, enabling us to endorse or repudiate claims whose content we are unable or unwilling to specify: “I think that everything Wittgenstein said is true”; “Beware: nothing he says will be true.”11 If viewing the concept of truth as fundamental does not bar Williams from giving a genealogy in principle, it nevertheless provides him with a reason not to give one in practice, because it entails that this fundamental concept of truth is not what needs defending. What needs defending is our concern with the truth about given subject matters. For many subject matters, such as the distant past or the intricate workings of nature, this concern has datable beginnings; the corollary is that it may also have an end. It is this latter possibility which motivates Williams’s vindicatory project and its focus on truthfulness. The contrast heralded in the title Truth and Truthfulness brings into focus a distinction between a formal concept that we cannot help but live by, and a social and historical achievement that requires continual cultivation. We cannot give up on the concept of truth, least of all if we are to hold on to truthfulness (for what would we be true to?).12 But we can give up on truthfulness, in the sense of ceasing to value the finding and sharing of the truth in certain areas. Williams aims to defend this sense 10. Price (2011, 47, 139). Of course, in the practice of arguing about how things are, truthfulness is not the only way to achieve outcomes of practical value. There is something to be said for playing devil’s advocate, or for exploring and defending implausible hypotheses. There has been a lot of work recently by people like Liam Kofi Bright, Remco Heesen, Cailin O’Connor, and Kevin Zollman on whether individuals need to value the truth in order to promote collective success at pursuing the truth.

8. See Williams (1997a, 16–9; TT 45–53, 63, 84) and his reference to Campbell (1994) at 1997a, 91n8.

11. Blackburn (2013, 5) and M. Williams (2013, 135).

9. A point Williams made in a lecture cited in Fricker (1998, 165n13).

12. See TT 2–3.

[p. 4]

of the value of truth, expressed in the concern to get one’s beliefs right (the set of dispositions Williams labels “Accuracy”) and the concern to pass them on to others (the set of dispositions he labels “Sincerity”), which together form the “virtues of truth”. He seeks to defend the virtues of truth by exhibiting their point for us. There are two potentially confusing ambiguities in Williams’s talk of the value of truth and truthfulness which bear clarification. Williams writes that he is concerned throughout the book with “the value of truth” (TT 6). But he hastens to add that, strictly speaking, it is not truth itself that bears value: of incomprehension. The answer only begins to make sense once we clarify that we like to know what people are talking about. Similarly, Williams thinks that when we say that we value truth intrinsically, what we mean is that we value expressions of truthfulness in the broad sense in which Williams uses the term: expressions of the desire to find out the truth; persistence in the face of obstacles to inquiry; imperviousness to wishful thinking and self-deception; and care and discipline in forming one’s beliefs, in preserving the truth, and in passing it on to others. The use of metalinguistic vocabulary to refer to intrinsic values is best understood as referring to human dealings revolving around these properties. Hence, for Williams, to value the truth intrinsically is to value various states and activities associated with the truth, in particular the various states and activities expressive of truthfulness. But since to value the truth intrinsically is to be truthful, truthfulness is both act and object of valuation on Williams’s picture. This means that it would be wrong to conclude that Williams is concerned with truthfulness as opposed to truth: In a very strict sense, to speak of “the value of truth” is no doubt a category mistake: truth, as a property of propositions or sentences, is not the sort of thing that can have a value. […] The phrase “the value of truth” should be taken as shorthand for the value of various states and activities associated with the truth. Much of the discussion will be directed to the value of what I shall call the “virtues of truth,” qualities of people that are displayed in wanting to know the truth, in finding it out, and in telling it to other people. (TT 6–7) [S]ome may complain that […] it is simply these qualities [involved in being truthful] that, so far, are supposed to bear the value. People have spoken of the value of truth: is this what they had in mind? Are we right to consider only certain human attitudes toward the truth, people’s dispositions to discover it and express it? My answer is yes — it is right only to consider human attitudes. Indeed, it is part of the naturalistic outlook of this inquiry that it should be seen as an exercise in human self-understanding. (TT 60) Why can truth as a property of propositions not have value? After all, to take an example Williams uses elsewhere, when someone believes against all evidence to the contrary that her child survived the crash, the truth of her belief is of immense value to that person. But, for Williams, this is the value of a particular truth p, where the value of p’s being true just is the value of p, and “[t]his is not the value of truth, but the value of survival” (1995d, 231). And even when we ask after the value of truth in general, Williams thinks, we are not really asking after the value possessed by truth. This would be like asking after the value possessed by reference in general, or by meaning, or by other metalinguistic categories. If we were asked to list all the things we value intrinsically, answering “reference” would provoke blank stares Williams refuses to regard as an explanation a “metaphysical account which represents the objects of our knowledge and their value as in themselves entirely independent of our thoughts or attitudes”, or any other explanation “that sets truth and goodness even further above us” (TT 61). Instead, his explanation of the value of truth takes the form of

[p. 5]

self-understanding in the sense that it starts out from our dispositions to value the truth and explains these in terms of their practical value to us. Williams’s approach is thus naturalistic in a sense akin to what Huw Price calls subject naturalism:13 it approaches the value of truth via human dispositions to truthfulness, and these in turn via their point relative to basic human needs.

Even once we gain hold of the idea that the intrinsic valuing of truth manifests itself in truthfulness and involves the valuing of truthfulness, and that truthfulness therefore plays a double role as both act and object of valuation, there remains a second hazardous ambiguity in the phrase ‘the value of truthfulness’. It can be used in an appositional or a possessive sense. In the appositional sense, it picks out one among our values, namely that of truthfulness. Reference to the “value of truthfulness” is then a reference to our valuing the truth, manifested in our seeking, preserving, and telling it, i.e., in our being truthful (e.g., TT 13). By contrast, the possessive sense concerns the value that truthfulness possesses for creatures like us. It refers to the practical value of our valuing the truth, i.e., the practical value of truthfulness (e.g., TT 15).

With these clarifications in place, we can state precisely what Williams’s genealogy is a genealogy of. It is a genealogical explanation of why we might have come to value truth intrinsically — where this means valuing the various states and activities expressive of truthfulness intrinsically — which is given in terms of the practical value of valuing truth intrinsically.

It thus misses Williams’s point to say that the book’s title is “ironically misleading” in promising a treatment of truth.14 It is a treatment of truth, only one that carries with it the claim that the way to meet those who would give up on truth is not to ask what truth is, or which theory of truth we should adopt, but why we should cultivate the virtues aiming at truth, and what we have to lose if we do not. What needs defending is not our concept of truth, but our concern for it. In this context, Socratic questions about the nature of truth are better transposed into pragmatic questions about the point of truthfulness. Instead of gazing up at truth itself, we should face the deniers with our eyes set on human concern with the truth, bring out how it relates to the rest of human psychology and to various social and political issues, and show what concern with the truth does for us. This is the task shouldered by Williams’s genealogy.

2. WILLIAMS’S PRAGMATIC GENEALOGICAL METHOD

What does Williams understand by a “genealogy”? In Truth and Truthfulness, he offers the following characterisation: “A genealogy is a narrative that tries to explain a cultural phenomenon by describing a way in which it came about, or could have come about, or might be imagined to have come about” (TT 20). On this broad characterisation, genealogy is a form of explanation drawing on everything from historical references and conjectures to abstract and imaginary developmental stories in order to elucidate something via its genesis. But what is the purpose of the imaginary elements? Their centrality to Williams’s understanding of genealogy comes out in his article “Naturalism and Genealogy”, in which he characterises genealogy entirely in terms of these imaginary elements. He defines genealogy as “a fictional story which represents a new reason for action as being developed in a simplified situation as a function of motives, reactions, psychological processes which we have reason to acknowledge already” (2000, 159). On this narrower view, genealogy has four characteristics: (i) it is a developmental narrative sketching how one can get from some starting point at which there is as yet no X to an end point at which there is X; (ii) it explains the emergence of a new reason for action in terms of more primitive reasons for action which it takes as given; (iii) the transition from the starting point to the end point is rationally intelligible (as opposed to only causally intelligible); and (iv) the rational transition from starting point to end point is not just a matter of instrumental reasoning — what makes genealogies interesting is 13. See Price (2011, 189).

14. Elgin (2005, 344).

[p. 6]

precisely that the individual cannot reason his way from the starting point to the end point on his own. The end point is instrumentally inaccessible to the individual reasoner (in this case, as we shall see in §3, because it involves the creation of an intrinsic value). Williams’s genealogy of truthfulness exhibits all four characteristics of the fictional story, but it also recognises a need for philosophy to involve itself in history (TT 93) — hence the broader characterisation of genealogy. Williams’s genealogy deploys a narrative device he calls the “State of Nature” (TT 21). It differs from homonymous scenarios in political philosophy in that it already contains “a small society of human beings, sharing a common language, with no elaborate technology and no form of writing” (TT 41). But it is not, for all that, a representation of any actual society, nor even a conjectural representation of the early hominid condition. The State of Nature, Williams insists, is not the Pleistocene (TT 27). We do not read off what to put into the State of Nature from the fossil record. In fact, the State of Nature does not even have to be possible (TT 30).

I suggest that the State of Nature is most illuminatingly described not as a fiction, but as a model, which abstracts as much from past human societies as from present ones. The purpose of this model is not, in the first instance, to identify the historical origins of truthfulness, but to identify its structural origins. It serves to localise and bring out the function of the virtues of truth relative to certain contingent facts about human beings and their environment. It contributes to what Williams elsewhere calls a “structural description” (1997b, 24) of truthfulness and its roots in certain basic needs. It is only in a second step that these structural insights can then be deployed to shed light on truthfulness’s actual history. The structural insights into the necessity of a certain prototypical form of truthfulness lead us to expect that any society will always already possess an instantiation of this prototype. In light of this prototype and its basic point, moreover, functional analogies will become visible between different phenomena, so that the prototype helps us identify different instantiations of truthfulness in the historical record. The same structural insights also indicate which developments are only weakly, and which are not at all, anticipatable on the basis of generic human needs. We are thus led by these structural insights to look, as Williams does, for the historical moments around which truthfulness came to be extended to the distant past, or when it grew into a demand for authenticity.

Genealogies have to start somewhere, and in doing so they will have to take certain things for granted.15 Williams is unapologetic about this: “critical reflection should seek for as much shared understanding as it can find on any issue, and use any ethical material that, in the context of the reflective discussion, makes some sense and commands some loyalty. Of course that will take things for granted, but as serious reflection it must know it will do that” (2011, 117). Nevertheless, Williams tries to take as little for granted as possible.16 He starts out from a state-of-nature model depicting a basic and, in human terms, entirely generic epistemic predicament: human beings need information about the environment, its risks, and its opportunities (TT 58); but already the sheer fact that they are in different places at different times means that there are strong pragmatic pressures on them not just to rely on their senses in acquiring it, but to cooperate, in particular by engaging in an epistemic division of labour whereby information is pooled (TT 42); this in turn means that they need to cultivate the dispositions that make good contributors to the pool: centrally, accuracy and sincerity in some prototypical form. But since the practical value of these dispositions largely consists in their advantageousness to others, accuracy and sincerity need to come to be regarded as traits worth having for their own sake — as virtues — if the practice of effective 15. See Blackburn (2002) and Hall (2014, 556).

16. “We must be cautious,” he writes elsewhere, “in dogmatically asserting that some given elements must be universal […]. However, even if we do not offer any rich body of what are claimed to be substantive universal necessities, there is, in the area of action and its ethical surroundings, a set of very basic ideas which, at the least, lay an extremely heavy burden of proof on anyone who claims to find a society in which these conceptions were not operative” (2006b, 67). See also Williams (2011, 170), where he advocates a naturalistic and historical conception of human nature.

[p. 7]

information pooling is to avoid succumbing to free riders (TT 59). For this to be the case, people need to be able to make sense of them as values “from the inside” (TT 91), which requires being able to articulate and connect them to other values and emotions. But while the functionalist abstraction of the State of Nature can reveal that this is so, it cannot show how these generic demands have actually been satisfied now and around here. Nor can it help us understand the “enormous degree” to which truthfulness was “changed, transformed, differently embodied, extended and so on by historical experience” (Williams 2007, 132). This is the reason why the perspicuous representation of what truthfulness does for us needs to be enriched with historical and cultural information that is invisible from the a priori standpoint of functional reflection. We need to incorporate increasingly local needs into our model of truthfulness’s development in order to account for the elaborations that our instantiations of the virtues of accuracy and sincerity have undergone.

On such a reading, Williams’s genealogy is a pragmatic genealogy: an explanation of why we came by a certain conceptual or evaluative practice in terms of what it does for us, given our needs. Historical considerations only come in to explain the various respects in which the practice we actually have differs from the generic functional model of the practice we require, given our needs. This allows us to take his genealogy as a direct response to Rorty’s question of why we should value the truth. The answer it yields is that we should value it because doing so is highly functional for us, given some of our most basic needs. If Williams offers a pragmatic genealogy, he has more in common with his avowed opponents than he admits. The immediate objection to this is that the book is explicitly directed against “the pragmatists” (TT 59) who encourage us to relinquish talk of “truth”. But pragmatism is a broad church, and much effort has recently gone into highlighting the more truth-affirming strands of pragmatism. These lead from Peirce to Ramsey, Wittgenstein, E. J. Craig, and, I submit, Williams himself.17 As Williams’s subtitle acknowledges, his genealogical method was largely inspired by Craig, whose genealogy is recognised as exemplifying a form of pragmatism (Weinberg 2006). Williams adopts it nearly wholesale to expand on the willingness of informants to be truthful, thereby falling squarely into the tradition which Cheryl Misak calls “Cambridge pragmatism” (2016).

One can distinguish truth-denying from truth-affirming pragmatists.18 Williams argues against the former (as represented by Rorty), but on grounds that place him squarely among the latter — not merely because he affirms truth, but because his approach to it is pragmatist in all but name. A truth-affirming pragmatist strives “to illuminate the concept of truth by considering its linkages with inquiry, assertion, and the acquisition of belief” (Misak 2016, 28).19 Williams, in line with the pragmatist idea that the best understanding of philosophy’s subject matter is agent-centred, elucidates truth in terms of the various “states and activities associated with the truth” (TT 7). He eschews questions such as what truth is, asking instead after the role of truth in various human dealings, such as language learning, believing, asserting, inquiring, communicating, and cooperating. Part of what licenses the application of the label to Williams, then, is that he elucidates truth in terms of what we do with truth: the human dealings expressive of truthfulness.

What also licenses the label’s application is that Williams raises a pragmatist question in the reverse direction, namely what truthfulness does for us. He argues that truthfulness “gets its point ultimately from the human interest, individual and collective, in gaining and sharing true information” (TT 126). Williams’s genealogy can thus be understood without distortion as a pragmatic genealogy — an explanation 17. Misak (2016) traces these under-appreciated strands of pragmatism from Peirce to Wittgenstein. See also Misak and Price (2017). 18. Misak (2016, ix) and M. Williams (2013, 129; 2016, 223). 19. See Wiggins (2002, 316) and Misak (2015, 264).

[p. 8]

of why we might have come to value the truth in terms of the point of valuing the truth.20 (2) The Utilitarianism Challenge: “the functional account looks like a thinly disguised form of utilitarianism, an argument to the effect that truthfulness is good because it increases the general level of human well-being.”

3. MCGINN’S THREE CHALLENGES AND SELF-EFFACING FUNCTIONALITY

Against the background of these remarks on Williams’s object and method, we can turn to the vindicatory power of pragmatic genealogy, its relation to indirect utilitarianism, and its application to self-effacing functionality. A helpful way into these questions is the critique of Williams’s genealogy advanced by Colin McGinn, which I shall use as a foil. McGinn not only highlights the resemblance of Williams’s genealogical method to just the kind of utilitarianism of which Williams was an arch-critic, but also calls into question the method of pragmatic genealogy more widely. We learn much about the contours of Williams’s genealogy by determining whether it stands up to McGinn’s critique. According to McGinn, Williams’s genealogy is vulnerable to the following three challenges: (3) The Redundancy Challenge: “once this is seen the genealogy itself becomes theoretically redundant.”21 McGinn’s first challenge, that instrumental considerations in favour of truthfulness do nothing to vindicate its intrinsic value, renders acute the question of just what Williams’s genealogy aims to achieve. One thing Williams offers is a negative vindication of truthfulness: his genealogy does not excavate anything to suggest our endorsement of truthfulness to be radically self-deceived, thus clearing truthfulness of suspicion and marking it out as stable under reflection. Another thing he offers is a naturalistic vindication of truthfulness: the genealogy enables us to make sense of truthfulness in terms of the rest of nature, in particular in terms of basic needs of cooperation and communication. Both of these vindications ensure that we can remain confident in truthfulness (1) The No Intrinsic Value Challenge: “showing the function that a virtue serves can only give it instrumental value, not intrinsic value […]. Since Williams insists, rightly, that truthfulness has an intrinsic value, […] his functional story fails, by his own standards, to capture that intrinsic value; so it does nothing, really, to vindicate the intrinsic value of truthfulness.” […] in the sense that we can understand it and at the same time respect it, support it and live within it. We can also urge it against alternative creeds whose own selfunderstandings (as divine revelations, for instance) are themselves not going to survive a genealogical inquiry. (2014, 410) Yet, as the preceding section brought out, Williams also offers a pragmatic vindication of truthfulness. His genealogy reveals the point of truthfulness, which it is shown to possess relative to needs so basic that they would be at work in anything recognisable as a human society. It is with respect to these needs that truthfulness proves its worth.

20. Belabouring the question of how to label Williams’s genealogy may seem to run counter to his own deep suspicion of labels. He repeatedly makes the point that they stand in the way of truthful inquiry, recounting how, after a lecture Ryle had given in Germany, a student had said: “I was very impressed by your lecture and would like to join your school. Unfortunately, I am a Kantian” (1995c, 186; 2007, 130). But in picking out his target by the label of pragmatism, Williams inevitably risks obscuring the pragmatism pervading his own enterprise. It is in correcting for this that the value of labelling Williams as a Cambridge pragmatist lies — it derives its value not, as one might think, from the importance we attach to labels, but from the potential disvalue of attaching importance to labels.

21. McGinn (2003, §1).

[p. 9]

And as Williams urges against the deniers, that is very much a reason not to give it up.

It is here that McGinn’s first challenge gets a grip. A pragmatic vindication may be fine as far as it goes, but it does not go far enough; showing what function something serves can give it instrumental value, but not intrinsic value; hence, the genealogy fails to vindicate the idea that truthfulness is intrinsically valuable. This is a line of argument which Williams himself acknowledges when he admits that his considerations only show that we need to treat truthfulness as though it had intrinsic value — they do not and cannot show that truthfulness does in fact have intrinsic value (TT 90).

But what McGinn fails to see is that Williams is not trying to give truthfulness intrinsic value — he is trying to vindicate the intrinsic value it already has. It is true that the genealogy can only show that, given certain basic needs, human beings need to treat truthfulness as an intrinsic good; but it does not follow that this amounts to mere pretence. If pragmatic considerations show that regarding truthfulness as an intrinsic good is instrumentally necessary, the question is whether we can and do so regard it; if yes, we will have been vindicated in doing so. Williams takes the following two conditions to be sufficient, albeit not necessary, for X to be an intrinsic good: necessity — a matter of having strong instrumental reasons to acquire something, given the needs and purposes one already has. Consequently, we might call it the condition of Practical Exigency. As for the second condition, we might call it the condition of Conceptual and Affective Embeddedness: valuers of truthfulness must be able to make sense of truthfulness as an intrinsic value “from the inside” (TT 91), as Williams put it, which means that they must have the conceptual and emotional resources necessary for them to relate truthfulness to other things that they value, such as nobility or freedom from manipulation, and, crucially, to their emotions.23 This indicates a contextualist conception of intrinsic value that differs markedly from, for instance, G. E. Moore’s conception of intrinsic value in Principia Ethica (1993), where Moore tries to identify intrinsic goods by asking whether they are “such that, if they existed by themselves, in absolute isolation, we should yet judge their existence to be good” (236–7). On Williams’s account, by contrast, something can have intrinsic value only if the agent possesses “some materials in terms of which he can understand this value in relation to other values that he holds” (TT 92). Williams mentions the example of Ancient Greece, where truthfulness was made sense of in terms of ideas of honour and shame (1997a, 26; TT 115); at other times, truthfulness was connected with freedom, absence of manipulation, equality, and self-respect. Merely seeing the benefits of valuing something intrinsically is not enough actually to do so. But if both the Practical Exigency and the Conceptual and Affective Embeddedness conditions hold, It is necessary (or nearly necessary) for basic human purposes and needs that human beings should treat X as an intrinsic good. […] we have not simply adopted an illusion or a pretence of there being an intrinsic good. In fact, if these conditions hold, that would be a very odd thing to say, implying as it does that there is something further which would count as its really being an intrinsic good, of which these conditions offer only a surrogate or mock-up. If the They can coherently treat X as an intrinsic good.22 The first thing to note about these two conditions is that they are not necessary conditions — Williams is not committed to the claim that all intrinsic goods are goods that humans need to treat as intrinsic goods. But he takes this to be true of some intrinsic goods, and where this is the case, realising it will help explain why they are intrinsic goods. Furthermore, the necessity at issue in the first condition is practical 23. Williams (1973, ch. 13). See Goldie (2009) regarding the connection Williams draws between values and emotions, and Kusch (2009) regarding Williams’s view of values as coming in socially shared webs and systems and as making sense only in relation to each other.

22. TT 92; also Williams (2006b, ch. 8).

[p. 10]

conditions are satisfied, then we shall have constructed an intrinsic good. (TT 92) applications of this general contrast (2006a, 135). Williams proposes to explain the notion of being an intrinsic good in terms of people valuing it as an intrinsic good (for instance, by saying that an intrinsic good is what a wise or rational person would value as an intrinsic good), just as we explain being an end in terms of someone pursuing it as an end (this is not to identify intrinsic goodness with final goodness: some things, like natural beauty and works of art, are valued intrinsically without being pursued as ends [2006a, 135–6]). Hence, “something is intrinsically good if we need to value it as intrinsically good, and we can make sense of our doing so” (2006a, 136). This just is for it to fulfil the Practical Exigency and Conceptual and Affective Embeddedness conditions.

The core of the answer to McGinn’s first challenge, then, is that functionality is not supposed to give truthfulness intrinsic value, but to vindicate it as the intrinsic good it already is. Williams only has to show that we need to value truthfulness as an intrinsic good and that we can make sense of our doing so. He achieves this by deriving from more basic needs a need to value the dispositions of accuracy and sincerity for their own sake, and by tracing out the connections of truthfulness to others things we value. To do this is to construct an intrinsic good — not in the sense of creating it ex nihilo, but in the sense of drawing together and establishing connections between considerations that are, for us, already there. The book thereby pursues a wider concern — to show that, in a seeming paradox, “intrinsic values […] have their uses” (TT 127). In an atmosphere dominated by what Williams perceives as undue “scientism”,24 many are suspicious of intrinsic values, and are attracted by theories like utilitarianism partly because these promise to make sense of so much of the world in instrumental terms, which can seem like the only naturalistically intelligible form of value. Any such theory has to be grounded in some intrinsic values; but at least they are kept to a minimum. Williams, by contrast, makes us comfortable with The mistake of describing Williams as holding that we should treat truthfulness as if it were an intrinsic value when really it is only an instrumental value is the failure to notice that he need allow no sense to the term ‘intrinsic value’ except one in which it is true that truthfulness is an intrinsic value. In order better to grasp this line of thought, it helps to step back from his genealogy for a moment to understand the conception of intrinsic value in relation to which Williams develops his own, which is Christine Korsgaard’s. Korsgaard contrasts “intrinsic” with “extrinsic” on the one hand, and “final” with “instrumental” on the other (1996, 249). She maintains that these are different contrasts between different classes of things. For Williams, however, it is only the intrinsic/ extrinsic contrast which captures a difference between things, namely those which have their value in themselves and those which derive it from something else; the final/instrumental contrast distinguishes ways in which we value things, namely as means or as ends. Williams derives two ideas from Korsgaard. First, the contrast to “valuing something as an end” should be a broader category than just “valuing it as a means”. It should be something like “valuing derivatively”, of which instrumental valuing is only a species — one may value going to the concert as a way of having a good evening, but this does not mean that we value it only as a means (2006a, 122–3). Second, “intrinsic goodness” as it occurs in the intrinsic/extrinsic contrast is a matter of how we explain goodness, and in this context, an intrinsic good is one whose goodness is self-explanatory (2006a, 124). Williams urges us to “give up the unrewarding idea of intrinsic goodness being self-explanatory” (2006a, 136). Instead, intrinsic goodness should be recast in the same terms as the contrast between final and instrumental, as a matter of the way in which we value things. The fundamental contrast then becomes that between valuing intrinsically and valuing derivatively. Valuing as an end and valuing as a means are just 24. Williams (2006b).

[p. 11]

intrinsic values — partly by showing that they can be made sense of without deteriorating into pretence, and partly by showing that intrinsic valuing is instrumentally indispensable. Nevertheless, there are limitations to pragmatic genealogies’ vindicatory force. They come into view as soon as one asks: A vindication for whom? Williams’s genealogy yields pragmatic reasons, individual and collective, to cultivate the virtues of truthfulness, and shows why, beyond a certain critical mass, communication will not survive failure in that respect — there are only so many deceivers the institution of truthfulness can take. Yet realising this does not provide someone who has a strong inclination to deceive with a reason to desist. It merely constitutes an external justification which gets no internal hold on the deceiver.25 The thought that if everyone were like the deceiver, he could not exist will only get a grip, and turn the external justification into an internal one, to the extent that he otherwise shares in the ethical life, because this imagined universalisation is an essentially ethical thought.26 The genealogy offers reasons for those who already see truthfulness as intrinsically valuable to continue to do so. It is in line with the aim Williams formulates for ethical discourse: “not to control the enemies of the community or its shirkers but, by giving reason to people already disposed to hear it, to help in continually creating a community held together by that same disposition” (2011, 31). The genealogy is not an instrument of conversion. But it can promote self-understanding, and thereby strengthen the confidence of those who are, in some measure, already disposed to be truthful.27 Let us now turn to McGinn’s second challenge, that Williams’s functional story constitutes a thinly disguised form of utilitarianism. This need not be problematic in itself, but it would be a problem by Williams’s own lights, since the form of utilitarianism his vindication most closely resembles is the indirect utilitarianism he himself impugned as unstable under reflection.28 Turning Williams’s objection back on himself, we might say that the attempt to justify the disposition to value truthfulness intrinsically on purely instrumental grounds leads to a tension between the spirit being justified and the spirit justified. Under reflection, such a structure is bound to unravel. This is not for the reason we encountered above: that we need to be able to make sense of the spirit being justified from the inside. The reason it is unstable under reflection is that it tries to combine the following two incompatible thoughts: (1) Truthfulness is intrinsically valuable.

(2) Truthfulness is only instrumentally valuable.

Indirect utilitarianism typically tries to relieve the tension by appealing to a distinction between theory and practice: we might think (2) in what Joseph Butler called the “cool hour” of reflection, yet in the thick of the action, we focus firmly on (1). But on Williams’s own account, the distinction possesses “no saving power” (1995b, 165). How is his own story any different?

One difference is that because Williams does not share the utilitarian’s commitment to there being only one really intrinsically valuable type of thing (“well-being” in McGinn’s version), Williams is free to maintain that the two thoughts being combined are really these: 25. Williams (1973, 252–3).

26. Williams’s commitment to reasons internalism shines through here: A has reason to φ only if there is a sound deliberative route from A’s subjective motivational set S to A’s φ-ing (Williams 2001, 91). He even reads Kant’s theory as a limiting case of reasons internalism (1995c, 220n3). In seeking to ground his derivation of the need for truthfulness in maximally widely shared needs, Williams’s genealogy likewise constitutes a limiting case of internalism. Thanks to A. W. Moore for the pointer.

(1) Truthfulness is intrinsically valuable.

(2′) Truthfulness is instrumentally valuable.

27. See Queloz (2018) for further discussion.

28. Williams and Smart (1973) and Williams (1995b, ch. 13; TT 90–1; 2011, ch. 6).

[p. 12]

On the indirect utilitarian account, the recognition of truthfulness’s instrumental value is achieved at the cost of its intrinsic value. But with (2′), there is no longer a contradiction, since the instrumental value ascribed is not exclusive. The recognition of instrumental value coexists harmoniously with that of intrinsic value. Another difference is that while Williams offers a two-level view that is structurally similar to indirect utilitarianism, the repartition of the justificatory weight across the two levels of the utilitarian, (1) and (2), is very different from its repartition across Williams’s two levels, (1) and (2′). The indirect utilitarian can agree that people’s motivations in being truthful should be that truthfulness is a good thing in its own right, but what really justifies thinking in this way is the consideration, offered at the more reflective second level, that this is ultimately more conducive to well-being. For Williams, by contrast, the bulk of the justificatory weight lies on the first level: the fact that truthfulness is intrinsically valuable carries more authority than the fact that things go better if we think this way. The latter fact yields in the first instance an explanation rather than a justification, even if a vindicatory one. The propensity to conceive of intrinsic and instrumental value as mutually exclusive is not specific to utilitarianism. It also manifests itself in the view that there is nothing more to be said about a good once we have recognised its intrinsic value — it is valuable, and that is all. On this view, an explanation that exhibits truthfulness’s instrumental relations to other, less refined values will appear to besmear it — to imply that truthfulness has no intrinsic value (really). Once again, however, this appears so only if we conceive of the distinction between intrinsic and instrumental as an exclusive one: either we make sense of a value on its own terms, which we treat as irreducible, or we view it purely as a means to an end and reduce its value to that of something else.29 But this sets up a false dichotomy. Nothing forces us to choose, and once one rids oneself of the exclusive conception, it becomes evident that myriad things unite both aspects.

Some have nevertheless questioned that we can be “aware at the same time of the intrinsic quality of a value and of its instrumental quality” (Hartmann and Saar 2004, 392) in the way Williams suggests.30 Yet, upon closer inspection, his genealogy avoids this question anyway: the instrumental value is not located at the same level as the intrinsic value; while what is intrinsically valuable is truthfulness itself, it is the valuing of truthfulness as intrinsically valuable which is instrumentally valuable.

This act/object distinction is also crucial to answering McGinn’s third challenge: that once we recognise the allegedly “utilitarian” nature of the genealogy, it becomes redundant: a functional account can be given directly, without the developmental story. But this is precisely what Williams is at pains to deny, and what sets his genealogy apart from more reductionist forms of functionalism. We must distinguish between: (i) a purely functional account of why we engage in our actual practice; (ii) a purely functional account of why a prototype of the practice would arise in a state-of-nature model; and (iii) a historically informed account of why we engage in our actual practice. It is not entirely clear whether McGinn means to suggest that if we do (i), then (ii) and (iii) are redundant; or that if we do (ii), then (iii) is redundant. But either way, Williams disagrees, since his position is that (i) is misguided and that both (ii) and (iii) are required. Williams takes (i), the purely functional account of why we engage in our actual practice, to be misguided. One may well be impressed by the functionality of many practices, and it is accordingly tempting to conclude that the motives upholding them are themselves functional or instrumental motives. But, for Williams, this would be a mistake: “In relation to institutions, practices, expectations, and values that actually exist, of justice, promise-keeping, truthfulness, and so on, functional accounts are simply false;” it is “just not true”, he continues, “that the 30. Hartmann and Saar take Williams to be reacting to what they perceive as a problem by integrating “an instrumental component […] into the very definition of intrinsicness”, i.e. maintaining that “‘intrinsic’ just means ‘non-egocentric’ or ‘not merely self-interested’” (2004, 393). I offer a different reading of Williams’s conception of intrinsic value above.

29. Williams (1997a, 24; TT 90).

[p. 13]

dispositions of truthfulness that we have, or that anyone else has had, can be adequately explained in functional terms” (TT 34–5). Because pure functionalism only traces out instrumental reasons for action, it misses both the fact that something’s being a form of truthfulness also constitutes a new reason for action, and that this must be so if the practice is to be functional. By focusing only on the instrumental value, an explanation of truthfulness in purely functional terms misses its intrinsic value, and it misses the functional connection between the instrumental and the intrinsic value.31 Moreover, a purely functional analysis is also blind to whatever non-functional aspects truthfulness has acquired in the course of its history. Cultural variation between groups and cultures implies that even if certain functional relations were the same across these variations, truthfulness would nevertheless also have been shaped by different contingencies in each case. What truthfulness needs to be is only a very partial guide to what it actually is.

Can the pure functionalist not shrug this off by saying that he is simply not interested in these non-functional aspects? I think not, because I take Williams’s claim to be that the non-functional aspects of the practice are essential both to its individuation and to an understanding of its functionality. First, the non-functional aspects of a practice can be completely ignored only at the cost of treating the participants’ understanding of the practice as external to it, as a mere epiphenomenon. But, for Williams, we cannot treat the understanding of the practice as epiphenomenal while retaining the practice as an object of investigation, because that understanding is integrally related to the practice itself.32 If we tried to pick out the practice without covertly having recourse to its interpretation by the practitioners, all we would be left with are sets of motions whose classification into distinguishable practices would seem arbitrary. Only when interpreted as expressions of certain reasons and motives do these movements of limbs become recognisable as practices. And when these reasons and motives make no reference to the practice’s functionality, the attempt to explain the practice in purely functional terms will appear to distort the practice, to mistake it for a pastiche version marked by single-minded instrumentalism. If there is something functional about a practice which is not understood in functional terms, we must find a place for functionality within an explanation which holds on both to the practice and to the non-functional understanding of it. Second, the state-of-nature model itself reveals that the presence of non-instrumental motivations to be truthful is one of the functional requirements on the practice of truthfulness, which means that the functionality of truthfulness cannot be accounted for in purely functional terms. Instrumental motives for being truthful cannot render the practice stable enough to fulfil its function: the instrumental value of accuracy, to some extent, but to a greater degree that of sincerity, is largely a value for others. A functionally motivated practice of truthfulness would therefore be overly vulnerable to free riders. The practice of truthfulness can be stable, and possess instrumental value, only insofar as it is intrinsically motivated. It needs to outgrow its functionality in order to be functional, which is to say that it needs to be driven by non-functional motives. And once these are effective, they may well possess more authority than the functional motives for having those non-functional motives.

What Williams’s genealogy shows is that while truthfulness needs to be sustained by non-instrumental motives, awareness of the instrumental or functional motives for engaging in truthfulness is not required. No instrumental valuing of truthfulness is needed to reap its instrumental value. (2′) might however be read this way, so to preempt this reading, the genealogy is best represented as issuing in the following two beliefs: 31. This is illuminatingly brought out by Craig (2007, 200). 32. See Moran (2016, 322–4).

[p. 14]

(1) Truthfulness is intrinsically valuable. has in fact outgrown the purely instrumental prototype. We can call this the phenomenon of self-effacing functionality: (2′′) The attitude expressed in (1) is instrumentally related to certain human needs if widely shared and known to be shared.

Self-Effacing Functionality: The practice is functional, but only insofar as it is sustained by motives that are autonomous, i.e. not conditional on the practice’s functionality in any given case. The practice must outrun its functionality in order to be functional. When this condition is met, the functionality of the practice will be either secondary or entirely absent from the participants’ minds as they engage in it, but for functional reasons. In this sense, the functionality is self-effacing.

It is only the fact that we value truthfulness intrinsically which is shown to have instrumental value, even if we never value truthfulness instrumentally. The genealogy relates “a value which gives us some reasons for action to other reasons for action which […] we have ‘anyway’” (Williams 2000, 160) — but while the relation between them is an instrumental relation, the reasons for action are not instrumental reasons. It is by being “bloody-minded rather than benefit-minded” (TT 59) that we reap the benefits, and the genealogy helps explain why this must be so.

This puts even more distance between Williams’s genealogy and the problem of reflective instability. The genealogical account finds in truthfulness an instrumental value which derives solely from the intrinsic valuing of it. Functionality is not part of the content of the motives, but is possessed by them.33 Functionality helps explain why we value what we value, but it is what we value which motivates and justifies what we do. This is why we should opt for genealogy in the form of (ii), the application of the purely functional apparatus to an abstract model of truthfulness, if we are neither to overemphasise nor to overlook functionality. State-of-nature genealogy is called for when dealing with a practice that needs to outgrow its own functionality in order to be functional, i.e. that needs to be sustained by motives that do not refer to its functionality. Applying the purely functional apparatus to a prototypical version of the practice and seeing its inability to account for the practice’s functionality enables us to understand why the practice More precisely, a practice is self-effacingly functional only if it meets the following conditions: (a) Functionality: the practice is functional, i.e. it makes a useful difference to the lives of those who engage in it; (b) Autonomy: the practice is sustained by motives that are not conditional on its functionality in any given case; (c) Dependence: the practice can be functional only insofar as it satisfies (b), i.e. it would be unstable, redundant, or otherwise ineffective if sustained merely by motives conditional on functionality in any given case; and (d) Explanatory Connection: the practice fulfils (b) because of (c), i.e. there is an explanatory connection between its autonomy and its dependence on autonomy.

We can call this the FADE structure: Functionality, Autonomy, Dependence, and an Explanatory connection between the latter two. Functionality will indeed fade from view in a practice which exhibits this structure, overshadowed or completely eclipsed by autonomous 33. Craig’s reading of Williams, by contrast, seems to retain the idea of “functional motivations” (2007, 200).

[p. 15]

motives carrying more authority than instrumental considerations. The FADE structure helps explain why this state of affairs is not just a matter of historical accident, but grounded in solid functional reasons. The FADE structure is not the same as the structure familiar from Critical Theory, where awareness of a practice’s functionality is radically incompatible with full-blooded engagement in it. With self-effacingly functional practices, the point is only that the functionality of the practice is not in itself enough to sustain it, and so it is a functional requirement on the practice that there be non-instrumental motives for engaging in it. This need not mean that becoming aware of the instrumental motives for engaging in the practice has a destabilising effect, as Williams’s own example of truthfulness shows. Note also that we are only dealing with self-effacing functionality when the functionality of the practice depends on there being non-functional motives, and when there is an explanatory connection between this fact and the non-functional motives sustaining it. This distinguishes self-effacing functionality from practices which fulfil conditions (a) and (b) only because the participants fail to realise the practice’s functionality and engage in it because they have been taught to do so. It also distinguishes it from practices which fulfil conditions (a) and (b) and do so for functional reasons, but which are in principle sustainable by instrumental motives alone: we eat sweet fruit not just because doing so serves some further end, such as our health, but because we like to eat sweet fruit; this intrinsic liking for sweet fruit might well turn out to have functional origins; but if we suddenly ceased to like sweet fruit, the practice would not necessarily break down; it could be sustained by instrumental concerns alone. Hence, the functionality of the practice is not truly self-effacing, but only contingently effaced. In sum: a practice is self-effacingly functional if it is functional but we do not do it for its functionality, and it is functional only because we do not do it for its functionality. As Thomas Nagel notes, this is “a common feature of ethical norms, from promise-keeping to property rights to the rule of law” (2009, 132). Genealogy is uniquely suited to describing self-effacing functionality. It shows that intrinsic values have their uses without the insight into the use leading us to lose our grip on the intrinsic value (TT 93).

What about (iii), the part of the genealogy which involves itself in history? On the interpretation offered here, it is required because a priori reflection can take us only so far: from highly general demands on human beings to a schematic understanding of what is required to meet these demands; but to cover the remaining distance between what should be the case and what is actually the case, we need to take more socio-historically local developments into account. While the model shows that truthfulness needs to be valued intrinsically, it cannot help us specify the material needed to do so. History is required to understand the ideas in terms of which we came to make sense of truthfulness. Moreover, these ideas are the products of various historical forces, and their contingent historical extensions and elaborations cannot be anticipated a priori. Generic needs cannot account for the extension of truthfulness to the distant past around the time of Thucydides (TT ch. 7), or for the fashioning of Sincerity into the ideal of personal authenticity in the course of the Romantic era (TT ch. 8). Here we need to draw on a historical understanding of the more local needs that came to bear on truthfulness. This is the point at which “philosophy, in order to do its business, must move into history” (TT 173). This is not to say that genealogy becomes “history, correctly practiced” (Nehamas 1985, 246n1).34 Williams denies that what he is offering is anything like the historian’s craft; he wants only to “mention a few of the historical divergences” and to “trace some features of the structure that give rise to the variations” (TT 95) of the prototype of truthfulness outlined in the State of Nature. Finally, the state-of-nature story gives us only a prima facie reason to continue to value truthfulness intrinsically. Certain incentive structures — for instance, in a society systematically rewarding truthfulness with material benefits — could 34. This understanding of genealogy as a species of history is endorsed by Geuss (1999, 22–3), Owen (2007, 143), Merrick (2009), and Migotti (2016). See Queloz (2017a) for a more general discussion of what, on Williams’s account, engaging with history can do for us.

[p. 16]

in principle discharge the function performed by the intrinsic valuing.35 Historical developments can both add to and subtract from our reasons to value truthfulness, both generically and in specific areas. Williams acknowledges this when he writes that there is no one reason to value truthfulness (TT 263–5), and his genealogy’s involvement in history serves to do justice to this idea. The state-of-nature considerations indicate one widely shared instrumental reason to value truthfulness intrinsically, but the later parts of the book point to other reasons, more closely tied up with our history and our own ideals, such as the ideals of liberalism. While the instrumental reasons laid out in the State of Nature do not give us a reason to prefer truthfulness about the distant past over myth, for example, liberalism does. Liberalism, on Williams’s view, both enables and encourages truthful history; but it also specially needs it. This may seem like a joke, given that truthful history tends to soak in suspicion large parts of the progressive narratives liberalism tells about itself. But, for Williams, liberalism is ultimately kept afloat by the liberalism of fear, and the fear in question is fuelled by historical consciousness of past atrocities. It is in this sense that liberals have a special reason to prefer truthful history over myth. We should be wary of reading Williams’s genealogy as superimposing, in the form of (ii) and (iii), the products of different disciplines, such as game theory and historiography. This would rightly invite concerns — voiced by Rorty (2002) and echoed by Koopman (2013, 70– 1) — about how such disparate elements could coherently be stitched together. We should see pragmatic genealogy instead as an exercise in philosophical model-building which, while informed by other disciplines, responds to concerns that are recognisably those of philosophy. A pragmatic genealogy is a dynamic model starting out from an abstract representation of a basic predicament which is then gradually de-idealised in the direction of our actual situation by tracing out how it might develop in response to generic and local needs. The purpose of such a genealogy is to render perspicuous those formative influences which allow us to answer distinctively philosophical concerns about truthfulness: What is the point of truthfulness? How does it relate to the rest of human psychology? What other ideas and values is it tied up with? And once all this becomes transparent to us, should we continue to cultivate it? These are the questions to which Williams’s genealogy forms the answer. The questions, in that combination, are clearly those of philosophy before they are those of any other discipline; and as Williams learned from Collingwood, our understanding of the questions should guide our understanding of the answers.36 Williams’s genealogy can thus be defended against McGinn’s challenges on all fronts. It vindicates truthfulness as an intrinsic value and, in the process, makes clear what this means; it is not a form of indirect utilitarianism, but differs from it in just the way required to avoid the problem of instability under reflection; and its functionalist elements do not render the genealogical elements redundant, but rather combine with them to make sense of why truthfulness must outgrow a state of mere functionality. By way of conclusion, I would like to consider the wider significance of this genealogy in the light of Williams’s conception of what philosophy should aim for.

The “principal aim of all moral philosophy”, Williams maintains, is that “of truthfully understanding what our ethical values are and how they are related to our psychology, and making, in the light of that understanding, a valuation of those values” (1995a, 581). We can now see that Williams’s genealogical method is directly subservient to this aim. It helps us make sense of our values as responses to more basic human needs; and it facilitates a valuation of them by helping us determine the extent to which they have helped us to live. This is not the sort of valuation which yields ethical truths directly — it does not, for instance, tell us whether to be truthful in a given situation. The general insights it yields are that certain basic features of human psychology demand 35. Thanks to an anonymous reviewer for pressing me on this point.

36. Williams (2006b, ch. 24).

CONCLUSION: THE WIDER SIGNIFICANCE OF WILLIAMS’S GENEALOGY

[p. 17]

that certain dispositions or concepts be present in society in some form, while leaving it open which form this should be. This does not mean that one can then reason one’s way from the insights into the practical necessity of having those dispositions to the beliefs involved in having those dispositions. The beliefs involved in having these dispositions are not typically beliefs about dispositions, but about the social world. Yet such a genealogy can show us that there is good reason for us to live a life that involves certain conceptual and evaluative practices. It does not demonstrate the truth of statements made using these concepts; it only vindicates the disposition to accept these statements. Moreover, the vindication it offers is addressed only to those already disposed to treat truthfulness as an intrinsic value, who already possess the material necessary to making sense of it as such. Genealogy does not convert shirkers, but it can strengthen wavering confidence. The importance of strengthening our confidence in our concepts and values emerges when we consider together two of Williams’s central convictions: that we should aim for truthful self-understanding; but that our ideas are unlikely always to survive truthful scrutiny. Williams holds that ethical thought, in particular, “should stand up to reflection” and “its institutions and practices should be capable of becoming transparent” (2011, 222); yet he is also inclined to agree with Nietzsche’s diagnosis of the condition of modernity as “one in which we, at once, have a morality which is seriously unstable under genealogical explanation; are committed (by that very morality, among other things) to transparency; and find very little to hand in the way of an alternative” (2000, 160). As Williams had already put it nearly two decades earlier, in response to the assumption of much ethical theory that we have too many ethical ideas: “Our major problem now is actually that we have not too many but too few, and we need to cherish as many as we can” (2011, 130).

Part of the wider significance of Williams’s vindicatory genealogy of truthfulness thus derives from the contribution it makes to the cherishing of our ethical ideas: it shows one such idea to be stable under genealogical reflection, giving it a bill of health. Yet it derives additional significance from two further facts. One is that truthfulness is a thick concept — a concept that is both world-guided and action-guiding. Thick concepts offer both more and less stability than thin ones like good and right. They offer more stability insofar as they help stabilise the practice they are involved in by rendering judgements straightforwardly true (2011, 222). Yet thick concepts also offer less stability insofar as they are particularly liable to be unseated by reflection. Reflecting on a thick concept, we are more likely to find considerations that lead us to cease to live by that concept (1995c, 207). Truthfulness’s being stable under reflection is therefore specially significant in showing that there are thick concepts capable of surviving reflective scrutiny. This invites us to identify more such concepts through further genealogical inquiry.

Finally, the other fact from which a genealogy that takes truthfulness as its object derives special significance is that truthfulness is a driving force of genealogical reflection itself. In exhibiting truthfulness as stable under such reflection, Williams demonstrates that genealogical reflection will not peck into the dust the tree that supports it. This is a markedly different result from that of his great predecessor as a genealogist of truthfulness, Nietzsche. Nietzsche denounced the unconditional will to truth fostered by Protestant Christianity as doomed to erode its own basis.37 Truth and Truthfulness, by contrast, shows us that the roots of our sense of the value of truth reach far deeper. While the application of the virtues of truth to various areas depends on continual cultivation, these virtues are also firmly anchored in basic human needs — they derive their point from the need to gain and share information. The upshot is a combination of encouragement and admonition; encouragement because it suggests that truthfulness is capable of withstanding its own scrutiny, and thus of providing a stable basis from which to engage in genealogical reflection more widely; 37. Nietzsche (1998, III, §27). But even Nietzsche does not think we should give up on truthfulness altogether. See Queloz (2017b, manuscript) for discussions of the continuities between Nietzsche’s views and Williams’s.

[p. 18]

and admonition because it shows us what we have to lose in our sense of the value of truth.

ACKNOWLEDGMENTS

I am grateful to Markus Wild, A. W. Moore, Simon Blackburn, Rebekka Hufendiek, Paul Russell, Hubert Schnüriger, Jan Müller, Raphael van Riel, Mario Brandhorst, Deborah Mühlebach, Nico Müller, Damian Cueni, and Julia Wagner for discussions that proved valuable in writing this paper. Parts of the paper were presented at the Workshop on Conceptual Relativism at the University of Zurich and the Workshop on Social Functions at the University of Bern. I am grateful to the organisers for inviting me and to the participants for their feedback. Thanks are also due to two anonymous referees. This work was supported by the Swiss National Science Foundation [P0BSP1_162025].

BIBLIOGRAPHY

Fricker, Miranda. 1998. “Rational Authority and Social Power: Towards a Truly Social Epistemology.” Proceedings of the Aristotelian Society 98 (2): 159–177.

Geuss, Raymond. 1999. “Nietzsche and Genealogy.” In Morality, Culture and History, 1–28. Cambridge: Cambridge University Press.

Goldie, Peter. 2009. “Thick Concepts and Emotion.” In Reading Bernard Williams, edited by Daniel Callcut, 94–109. London: Routledge & Kegan Paul.

Hacking, Ian. 2004. “Critical Notice.” Canadian Journal of Philosophy 34 (1): 137–148.

Hall, Edward. 2014. “Contingency, Confidence, and Liberalism in the Political Thought of Bernard Williams.” Social Theory and Practice 40 (4): 545–569.

Hartmann, Martin, and Martin Saar. 2004. “Bernard Williams on Truth and Genealogy.” European Journal of Philosophy 12 (3): 386–398.

Honderich, Ted. 2003. “Yours Sincerely.” The Guardian, March 1.

Jenkins, Mark P. 2006. Bernard Williams. Chesham: Acumen.

Koopman, Colin. 2013. Genealogy as Critique: Foucault and the Problems of Modernity. Bloomington and Indianapolis: Indiana University Press.

Korsgaard, Christine. 1996. “Two Distinctions in Goodness.” In Creating the Kingdom of Ends, 249–274. Cambridge: Cambridge University Press.

Kusch, Martin. 2009. “Testimony and the Value of Knowledge.” In Epistemic Value, edited by Adrian Haddock, Alan Millar and Duncan Pritchard, 60–94. Oxford: Oxford University Press.

McGinn, Colin. 2003. “Isn’t It the Truth?” The New York Review of Books L (6).

Merrick, Allison M. 2009. “On Nietzsche’s Genealogical Mode of Inquiry.” Doctor of Philosophy, School of Humanities, University of Southampton.

Migotti, Mark. 2016. “History, Genealogy, Nietzsche: Comments on Jesse Prinz, Genealogies of Morals: Nietzsche’s Method Compared.” The Journal of Nietzsche Studies 47 (2): 212–227.

Blackburn, Simon. 2002. “Where the Tiger Went.” The Times Literary Supplement, October 18, 6–7.

―. 2013. “Pragmatism in Philosophy: The Hidden Alternative.” Philosophic Exchange 41 (1): 2–13.

Campbell, John. 1994. Past, Space, and Self. Cambridge, Mass., and London, England: MIT Press.

Craig, Edward. 2007. “Genealogies and the State of Nature.” In Bernard Williams, edited by Alan Thomas, 181–200. Cambridge: Cambridge University Press.

Davidson, Donald. 1990. “The Structure and Content of Truth.” The Journal of Philosophy 87 (6): 279–328.

Elgin, Catherine Z. 2005. “Williams on Truthfulness.” The Philosophical Quarterly 55 (219): 343–352.

Fleischacker, Samuel. 2004. “Review of Bernard Williams, Truth and Truthfulness: An Essay in Genealogy.” Ethics 114 (2): 380–385.

Foucault, Michel. 1994. “La vérité et les formes juridiques.” In Dits et Ecrits 1954–1988, 538–646. Paris: Gallimard.

[p. 19]

Misak, Cheryl. 2015. “Pragmatism and the Function of Truth.” In Meaning Without Representation: Essays on Truth, Expression, Normativity, and Naturalism, edited by Steven Gross, Nicholas Tebben and Michael Williams, 262–278. Oxford: Oxford University Press.

―. 2016. Cambridge Pragmatism: From Peirce and James to Ramsey and Wittgenstein. Oxford: Oxford University Press.

Misak, Cheryl, and Huw Price, eds. 2017. The Practical Turn: Pragmatism in Britain in the Long Twentieth Century. Oxford: Oxford University Press.

Moore, G. E. 1993. Principia Ethica. Edited by Thomas Baldwin. Cambridge: Cambridge University Press.

Moran, Richard. 2016. “2015 Mark Sacks Lecture: Williams, History, and ‘the Impurity of Philosophy’.” European Journal of Philosophy 24 (2): 315–330.

Mordacci, Roberto. 2016. “From Analysis to Genealogy. Bernard Williams and the End of the Analytic-Continental Dichotomy.” Philosophical Inquiries 4 (1): 71–84.

Nagel, Thomas. 2009. “Williams: The Value of Truth.” In Secular Philosophy and the Religious Temperament: Essays 2002–2008, 131–138. Oxford and New York: Oxford University Press.

Nehamas, Alexander. 1985. Nietzsche: Life as Literature. Cambridge, Mass.: Harvard University Press.

Nietzsche, Friedrich. 1998. On the Genealogy of Morality. Translated by Maudemarie Clark and Alan J. Swensen. Indianapolis and Cambridge: Hackett.

Owen, David. 2007. Nietzsche’s Genealogy of Morality. Stocksfield: Acumen.

Price, Huw. 2011. Naturalism Without Mirrors. Oxford: Oxford University Press.

Queloz, Matthieu. 2017a. “Does Philosophy Have a Vindicatory History? Bernard Williams on the History of Philosophy.” Studia Philosophica 76: 137–52.

―. 2017b. “Nietzsche’s Pragmatic Genealogy of Justice.” British Journal for the History of Philosophy 25 (4): 727–749.

―. 2018. “How Genealogies Can Affect the Space of Reasons.” Synthese. 10.1007/s11229-018-1777-9.

―. Manuscript. “Nietzsche’s Vindicatory English Genealogy of Truthfulness.”

Rorty, Richard. 1989. Contingency, Irony, and Solidarity. Cambridge: Cambridge University Press.

―. 1991. Objectivity, Relativism and Truth. Vol. 1, Philosophical Papers. Cambridge: Cambridge University Press.

―. 1998. Truth and Progress. Vol. 3, Philosophical Papers. Cambridge: Cambridge University Press.

―. 2002. “To the Sunlit Uplands.” London Review of Books 24 (21): 13–15.

Rorty, Richard, and Pascal Engel. 2007. What’s the Use of Truth? Translated by William McCuaig. New York: Columbia University Press.

Shapin, Steven. 1994. A Social History of Truth: Civility and Science in Seventeenth-Century England. Chicago: University of Chicago Press.

Weinberg, Jonathan M. 2006. “What’s Epistemology For? The Case for Neopragmatism in Normative Metaepistemology.” In Epistemology Futures, edited by Stephen Hetherington, 26–47. Oxford: Oxford University Press.

Wiggins, David. 2002. “An Indefinibilist cum Normative View of Truth and the Marks of Truth.” In What is Truth?, vol. 1, Current Issues in Theoretical Philosophy, edited by R. Shantz, 316–332. Berlin: de Gruyter.

Williams, Bernard. 1973. Problems of the Self: Philosophical Papers 1956– 1972. Cambridge: Cambridge University Press.

―. 1995a. “Ethics.” In Philosophy 1: A Guide Through the Subject, edited by A. C. Grayling, 545–582. New York: Oxford University Press.

―. 1995b. “Making Sense of Humanity.” In Making Sense of Humanity: And Other Philosophical Papers, 79–89. Cambridge: Cambridge University Press.

―. 1995c. “Replies.” In World, Mind, and Ethics: Essays on the Ethical Philosophy of Bernard Williams, edited by J. E. J. Altham and Ross Harrison, 185–224. Cambridge: Cambridge University Press.

―. 1995d. “Truth in Ethics.” Ratio 8 (3): 227–238.

[p. 20]

―. 1997a. Der Wert der Wahrheit. Wien: Passagen.

―. 1997b. “Shame, Guilt, and the Structure of Punishment.” In Scham und Schuld: Festschrift aus Anlass der Verleihung der Dr. Margrit Egnér-Preise 1997 zum Thema Scham und Schuld, 21–30. Zürich: Margrit Egnér-Stiftung.

―. 2000. “Naturalism and Genealogy.” In Morality, Reflection, and Ideology, edited by Edward Harcourt, 148–161. Oxford: Oxford University Press.

―. 2001. “Postscript: Some Further Notes on Internal and External Reasons.” In Varieties of Practical Reasoning, edited by Elijah Millgram, 91–97. Cambridge, Mass.: MIT Press.

―. 2002. Truth and Truthfulness: An Essay in Genealogy. Princeton: Princeton University Press.

―. 2006a. The Sense of the Past: Essays in the History of Philosophy. Edited by Myles Burnyeat. Princeton and Oxford: Princeton University Press.

―. 2006b. “Philosophy as a Humanistic Discipline.” In Philosophy as a Humanistic Discipline, edited by A. W. Moore, 180–199. Princeton and Oxford: Princeton University Press.

―. 2007. “Truth and Truthfulness.” In What More Philosophers Think, edited by Julian Baggini and Jeremy Stangroom, 130–146. London: Continuum.

―. 2011. Ethics and the Limits of Philosophy, Routledge Classics. Oxford: Taylor & Francis.

―. 2014. “Why Philosophy Needs History.” In Essays and Reviews: 1959– 2002, edited by Michael Wood, 405–412. Princeton: Princeton University Press.

Williams, Bernard, and J. J. C. Smart. 1973. Utilitarianism: For and Against. Cambridge: Cambridge University Press.

Williams, Michael. 2013. “How Pragmatists Can Be Local Expressivists.” In Expressivism, Pragmatism and Representationalism, edited by Huw Price, 128–144. Cambridge: Cambridge University Press.

―. 2016. “Pragmatism, Sellars, and Truth.” In Sellars and His Legacy, edited by James R. O’Shea, 223–260. Oxford: Oxford University Press.