# Internalism from the Ethnographic Stance: From Self-Indulgence to Self-Expression and Corroborative Sense-Making

Author: Matthieu Queloz
Published in: *The Philosophical Quarterly* 75 (3): 1094–1120. 2025. [doi:10.1093/pq/pqae051](http://doi.org/10.1093/pq/pqae051)
DOI: [10.1093/pq/pqae051](https://doi.org/10.1093/pq/pqae051)
Canonical entry: https://www.matthieuqueloz.com/entries/internalism-from-the-ethnographic-stance-from-self-indulgence-to-self-expression-and-corroborative-sense-making/
Published PDF: https://philpapers.org/archive/QUEIFT.pdf

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

[p. 1095]

Internalism from the ethnographic stance taught as a self-contained classic. If it is contextualized at all, it is usually in relation to later interventions in the debate it helped create. Surprisingly little effort has gone into integrating Williams’ internalist thesis with the rest of his own oeuvre.

Yet it is worth asking how Williams’ internalism relates to his wider thought. His later writings contain many relevant reflections that have largely gone unnoticed in the internalism literature, such as his account of the phenomenology of moral deliberation and the need for philosophy to be accountable to a realistic psychology; his critique of ‘moral self-indulgence’; his exploration of the importance of sympathetic understanding and the possibility of ‘the ethnographic stance’; and his efforts to carve out a sense in which philosophical explanation can inform deliberation. If we can make sense of the assumption that Williams’ internalism also informs his later works, moreover, we can take them as a guide to what the view is—or became; for we need not treat his ‘Internal and External Reasons’ (1979) as the definitive articulation of a fixed background commitment. Construing Williams’ internalism dynamically, as a view that matured over the course of his career, we can try to sharpen the contours of what proved an initially rather hazy claim by integrating it, in the spirit of rational reconstruction, with Williams’ later thought. An image from Placentinus offers an evocative way into the topic: according to internalism, our reasons for action flow from our motivations like rivulets from a spring, or trickles from a fountain. When a reason for action fails to tap into any of an agent’s motivations, it is not a reason that this agent has. Less metaphorically, the internalist thesis formulates a necessary condition on the truth of statements about what reasons for action an agent has: in Williams’ preferred rendering, an agent A has a reason to φ only if there is a sound deliberative route from A’s subjective motivational set S to A’s φ-ing (2001a: 91).

Ambiguities remain, however. Is this, as T. M. Scanlon and many others have thought, meant to be a substantive normative claim, designed to inform agents’ deliberation by giving them a criterion constraining what they should count as a reason?2 Or is it a descriptive metaethical claim, detached from engaged deliberation, which proposes to elucidate in virtue of what agents have reasons? Are we being handed an actionable criterion or a remedy to philosophical puzzlement over the source of reasons? places as a response to a paper by Martin Hollis on the relation of desire to reason. These references, removed in the version reprinted in ML, help account for the focus of Williams’ essay. ‘Williams seems to be offering a substantive, normative thesis about what reasons we have’ (Scanlon 1998: 365). Proponents of this widespread construal also include FitzPatrick (2008: 165), Parfit (2011: 270), Heathwood (2011: 82), Wedgwood (2011: 292), Markovits (2014: 52), and Paakunainen (2018).

[p. 1096]

I shall argue that each of these two construals runs into problems that Williams’ later thought can help us address. If construed as a normative claim, internalism gives rise to a worry about its consistency with the rest of Williams’ thought—in particular, with his deference to the phenomenology of moral deliberation and his critique of ‘moral self-indulgence’, where agents are moved to act by their conception of themselves as possessing certain moral motivations.

If construed as a piece of detached metaethical philosophizing, however, internalism gives rise to a worry about corrosiveness—it would seem that a wholehearted commitment to morality cannot coexist, in the same agent, with a metaethical understanding of reasons as a function of our motivations.3 By connecting Williams’ internalist thesis to the rest of his thought, I show that the consistency worry can be substantiated and deployed against the normative construal. Drawing on Williams’ writings on self-expression and sympathetic understanding without identification, I develop a construal of internalism that is consistent with his later account of moral deliberation.4 Insofar as this pushes us towards a construal of internalism as a descriptive metaethical claim, however, it exposes internalism to the corrosiveness worry. I show that Williams himself came to feel the force of that worry in ELP. But I also argue that he eventually found a compelling way out. He carved out a form of philosophical sense-making capable not only of avoiding the corrosive effects of cruder forms of metaethical internalism, but of strengthening one’s confidence in reasons with an externalist phenomenology. On the resulting construal, internalism, though not normative, can nonetheless inform how we deliberate and how we address those whose motivations differ from our own. This recovers the main attraction of construing internalism as a normative thesis: the promise of making a difference in practice. II. A consistency worry about the normative construal The claim that 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 can be heard as a normative claim, proffered from and for the deliberative stance, that proposes to constrain what is to count as a reason for action. On this normative construal, it is immaterial that the internalist thesis is formulated as a third-personal See Brink (1986), Nagel (1986), Thomas (2006: ch. 6), Sagar (2014), and Smyth (2019). Though the interpretation I arrive at differs from them, I have benefited from previous interpretations of Williams’ internalism, in particular Brink (1986), Hooker (1987), Wallace (1990, 1999); McDowell (1995), Millgram (1996), Scanlon (1998, 2014); Dancy (2000), Heuer (2000, 2012); Thomas (2006), Brandhorst (2007), Skorupski (2007), Finlay (2009), Alvarez (2010), Markovits (2011, 2014); Manne (2011, 2014); Smith (2012), Sagar (2014), Sobel (2016), Ng (2017), Paakkunainen (2018), Smyth (2019), and Fricker (2020).

[p. 1097]

Internalism from the ethnographic stance constraint on third-personal reason statements. ‘A’ could be me, and I could use the internalist thesis, just as it stands, as a criterion for the identification of bona fide reasons in my own normative deliberation. I need only ask: Do I have some item D in my S such that there is a sound deliberative route from my having D to my φ-ing? On this construal, the conclusion ‘I have reason to φ’ is appropriate only if I can in principle reach it through sound deliberation from the premise ‘I have D in my S’.

For this claim to be even remotely plausible, it is crucial to realize that one’s S need not be limited to desires. Although Williams predominantly discusses subjective motivational sets in terms of ‘desires’, and accordingly uses ‘D’ as a variable for items in S, this may owe much to the fact that Williams’ original paper responded to a paper by Martin Hollis (1979) on the relation of desire to reason. As Williams clarifies, however, he takes D to cover ‘dispositions of evaluation, patterns of emotional reaction, personal loyalties, and various projects … embodying commitments of the agent’ (ML: 105). These projects need not be selfish (ML: 105). Nor need they be self-centred, in the sense in which ‘the creative projects of a Romantic artist could be considered self-centred’ (ML: 13) if it has to be him who realizes them (Wagner comes to mind). Even once the motivational basis is broadened, however, the normative construal, attractive as it may be thanks to its ability to inform deliberation, proves inconsistent with Williams’ later writings. In particular, I shall argue in the next two sections that it is overturned by two characteristics of Williams’ later thought: (i) his commitment to a realistic psychology that accommodates the externalist phenomenology of moral deliberation; and (ii) his critique of ‘moral self-indulgence’, which explicitly condemns as a self-indulgent deformation of moral deliberation just the reflexive concern with one’s own motivations that the normative construal commends.

III. Self-expression: a psychologically realistic internalism It is a guiding methodological principle for Williams that moral philosophy should be answerable to a realistic psychology:5 once the quest for timeless rational foundations is abandoned, ‘the only starting point left is ethical experience itself ’, and philosophical reflection must be answerable to the ‘phenomenology of the ethical life’ (ELP: 103). Accordingly, ‘one important constraint on theories in moral philosophy is that they should be consistent with a plausible psychology’ (1995b: 578). This requires consistency with the contents of people’s deliberations and the descriptions under which they choose See also Williams (1971: 164; PS: 182; 1995b: 568; MSH: 68–9, 76n12; WME: 216; 1999: 157; M: 22; PHD: 75; ER: 152).

[p. 1098]

actions, which is why Williams ‘pays attention to the psychological form in which ethical considerations have to be embodied’ (MSH: 199).6 Yet on the normative construal, internalism flouts this demand for consistency with a realistic psychology. For once we consider the phenomenology of ethical experience and the descriptions under which moral reasons figure in deliberation, it is evident that they frequently do not present themselves as reasons we have because of our motivations, but as reasons we simply have. As externalists about reasons rightly note, it is a salient datum of ethical experience that we take ourselves to have at least some reasons for action that are independent of our motivations: in deliberating about one’s obligations to others, or the expectations one is subject to, or intrinsically valuable states of affairs, one’s motivations seem to be neither here nor there. It would render deliberation oddly reflexive to insist that agents should turn their gaze back onto their own motivations in determining what they have reason to do. In place of the selfless disregard for one’s motivations characteristic of moral deliberation, the normative construal puts a self-focused concern with those motivations. This rides roughshod over the phenomenology of ethical experience.7 As a result, Williams’ internalism looks out of keeping with the rest of his thought: Why would a philosopher otherwise so attuned to the nuances of experience put forward such a self-focused view of reasons? Can it really be that Williams’ signature thesis should be the one to fall foul of his own aspiration to consistency with a realistic psychology?

It seems unlikely—not least since Williams explicitly recognized the need for an interpretation of internalism that accommodated the externalist phenomenology of practical reasoning: ‘We need a realistic account, social and psychological’, he wrote, ‘of what is going on when seemingly externalist claims, referring to a social or institutional reason, are directed at recalcitrant or unconvinced agents’ (2001a: 95).

To meet this need, Miranda Fricker (2020) has developed Williams’ suggestion that such claims can be understood by internalism as working proleptically:8 though not true at the time they are made, they are made nonetheless in the hope that they might prove self-verifying. By making A recognize what is expected of him while appealing to more general motivations in his S, such as a desire to be respected by the speaker, the statement might bring it about that A acquires the motivation that makes the statement true.9 See Williams (PS: 224; ML: 91; ELP: 11; MSH: 54; TT: 140, 262; PHD: 140). An observation echoed by Schroeder’s (2007: 103–22) ‘Too Few Reasons’ objection. See also Heuer (2004) and what Schroeder & Finlay (2017) call ‘The Central Problem’ for internalism. See Williams (1979: 26; 1995b: 570–1, 575; MSH: 40; 2001a: 95). Where necessary to concisely retain consistency with quotations from Williams, I follow him in using masculine pronouns to refer to the agent A.

[p. 1099]

Internalism from the ethnographic stance But this refinement does not suffice to resolve the consistency worry. For when Williams speaks of ‘seemingly externalist claims, referring to a social or institutional reason, [that] are directed at recalcitrant or unconvinced agents’, he runs together two distinct ways in which a reason statement can ‘seem externalist’: (a) because it presents A’s possession of that reason as independent of A’s S; (b) because it is directed at an agent whose S gives the internalist theorist too little to work with to understand the statement as true. These are two importantly different ways of seeming externalist. The former concerns the externalist phenomenology of many reason statements, including some that internalism can regard as true in virtue of the agent’s motivations. The latter concerns the fact that some reason statements seem to call for an externalist philosophical account, because the internalist account has no purchase on the agents they are directed at. Taking into account proleptic mechanisms is a reaction to (b), since it extends the ambit of internalism to encompass statements directed at agents whose S fails to render the statements true when uttered. But this does little to account for (a), the externalist phenomenology of reason statements.

As long as internalism is construed as holding that statements about my S should be capable of figuring in my reasoning as premises from which I infer that I have reason to φ, an inconsistency with the demand for a realistic psychology remains, because the phenomenology of certain reasons is such that statements about my S could not properly figure as premises in my reasoning. If, probing my reason to φ with iterated why-questions, I work my way up the chains of reasons supporting my reason to φ, those chains will sometimes come to an end in facts that are essentially not facts about my S. Much as it can be an essential feature of a language game to deny that its subject matter depends in any way on our language games, it can be an essential feature of certain normative considerations to deny that their force depends in any way on our S.

Consider a belief that Williams regards as unhintergehbar, meaning that ‘there is nothing more basic in terms of which to justify it’ (PHD: 195): (E) Every human being deserves equal consideration.

It would betray a misunderstanding of the content of (E) for an agent to try to understand the reasons to φ he has in virtue of his belief that (E) as justified, in part, by the presence of D in his S. It is precisely not a matter of what is in his S. The reasons that (E) provides are supposed to be motivation-independent reasons. It would therefore be a radically revisionary proposal to suggest that the agent should be able to justify his reason to φ through sound deliberation from (E) plus some statement (M) about his own motivations:

[p. 1100]

(E) Every human being deserves equal consideration.

(M) I have D in my S. … [sound deliberation] … Therefore, I have reason to φ.

Here, considering my own motivations, as the thought labelled (M) proposes to do, seems, in Williams’ phrase, like ‘one thought too many’.10 If internalism is to be consistent with the phenomenology of ethical experience, it must accommodate that fact.

A step in that direction is to recognize the following: for A to have reason to φ in virtue of his S does not entail that this reason should be capable of presenting itself to him in his deliberation under that description. We must distinguish the descriptions under which reasons present themselves to the agent in his deliberation from the conditions under which reason statements are true of the agent. The third-personal thought ‘A has reason to φ in virtue of his S’ can present itself in the first-personal deliberative mode, which is to say in A’s own deliberation, not as ‘I have reason to φ in virtue of my S’, but, simply, as ‘I have reason to φ’. This allows one to grant that third-personal reason statements can always be properly understood with reference to the agent’s S, as internalism insists they can, while also granting, realistically, that some reasons for action necessarily figure in the agent’s own deliberation under descriptions that make no reference to his motivational set.

Williams himself later stressed this nuance in his position: if we explain what A does in terms of his reason for doing that thing, … we cite a consideration which was effective in his coming to act because it made normative sense to him. Its making normative sense to him implies that it made normative sense in terms of his S. This does not mean that when an agent has a thought of the form ‘that is a reason for me to φ’, he really has, or should really have, the thought ‘that is a reason for me to φ in virtue of my S’. The disposition that forms part of his S just is the disposition to have thoughts of the form ‘that is a reason for me to φ’, and to act on them. (2001a: 93) Some reasons, though possessed by the agent in virtue of his S, figure in deliberation under descriptions that make no reference to his S. This reflects Williams’ conviction that ‘the moral dispositions, and indeed other loyalties and commitments, have a certain depth or thickness’ (MSH: 169), a ‘momentum’ (MSH: 199) of their own: ‘phenomenologically’, they might ‘appear as convictions that a certain behavior must or must not be performed’ (Smyth 2019: 193).

Thus, φ-ing might make normative sense to A in light of (E), a judgement that in turn makes normative sense to A, but that cannot be justified in terms of anything further, much less in terms of A’s own motivations. Yet it can be a necessary condition on that judgement making normative sense to A as a See Williams (ML: 18).

[p. 1101]

Internalism from the ethnographic stance reason for him to φ that A have D in his S. What makes normative sense to the agent as a reason can be a function of his S without implying that the agent must make normative sense of that reason in terms of his S. Its making normative sense to the agent does imply that it made normative sense in terms of his S; but this does not imply that the agent made normative sense of it in terms of his S. What makes normative sense to us in terms of his S, and what makes normative sense to the agent, can both be a function of the agent’s S without any item in his S appearing explicitly in his own sense-making. An analogy with the physiological conditions of visual experience is helpful here. A’s having eyes is a necessary condition on his having visual impressions. But that does precisely not entail that his eyes must themselves be capable of appearing in his visual field. On the contrary, the eyes that enable that visual field are the one thing that cannot appear in it. At the same time, the first thing that someone reflecting on the conditions of visual experience would note is that it requires eyes.

Analogously, A’s having certain items in his S is a necessary condition on his having certain reasons to φ. But that does not entail that these items must be capable of properly figuring in his deliberation to that conclusion. Where A’s deliberation about whether to φ because of (E) is concerned, A’s own motivations should not show up in the space of reasons bearing on that question, even though it is, according to internalism, a necessary condition on (E)’s being a reason for A that A have certain motivations. It is notably in virtue of having these motivations that A finds that (E) makes normative sense to him. But the dispositions he has in virtue of that fact do not include the disposition to make normative sense of (E) in terms of (M). He thinks, simply, that every human being deserves equal consideration—and correctly understanding that thought requires grasping that the content of his S is immaterial in this connection. As Williams later emphasized (PHD: 195–7), what it is for an agent to have a reason may be for that reason to be, for him, simply a reason. This does not mean that the agent has the thought: ‘In virtue of my S, this is simply a reason for me’. It does not even mean that he has the thought: ‘In virtue of my S, this is simply a reason’. Rather, he has the thought: ‘This is simply a reason’. That is what it is for it to be, in virtue of his S, simply a reason for him. It will be objected that, according to Williams, the agent himself must be capable of deliberating his way from his S to the conclusion that he has reason to φ. But notice that although this means that the agent’s S must express itself in the premises that figure in his deliberation—there should be ‘a sound deliberative expression of the agent’s S’ (PHD: 115)—it does not mean that the agent’s S must express itself under a description that explicitly refers to an item in his S. He need not, for instance, have the reflexive thought: ‘I desire that all human beings receive equal consideration’. He can have the outward-looking thought: ‘Every human being deserves equal consideration’. His thinking this

[p. 1102]

thought expresses his desire for equality, but without inserting that desire, or any reference to his S, into the content of that thought.

This preserves a role for Williams’ necessary condition on reasons being our reasons by construing it as imposing the constraint that our motivations should express themselves in our deliberation—as Williams puts it, reasons should speak to our motivations (ELP: 269n19). But we avoid the uninviting implication that our motivations should be referred to in the fully articulated first-personal version of that deliberation. That this was Williams’ considered view comes out clearly in TT: The situation of decision cannot simply be identified with one in which an agent applies to himself the idea of what it makes sense to do. … [I]n the situation of decision, the agent does not merely consider or think about his beliefs, desires, and so on (though he is not excluded from doing so); rather, it is from the perspective of those motivational states that he considers what it makes sense for him to do. … This is a special case of its making sense to him that he will do a certain thing. What makes it special is not simply that it is a matter of the agent himself and not someone else, but the fact that in the situation of decision, the agent’s present motivational state does not primarily act as evidence or support for the conviction that it makes sense for him to act in this way: it is expressed in that conviction, just as it is expressed in the action itself. (TT: 236–37) It is therefore a mistake to identify the deliberative perspective with one in which the agent applies to himself the idea of what it makes sense to do in light of his S. From the deliberative perspective, internalism manifests itself through self-expression rather than through self-representation. It is true that Williams originally presents internalism as concerned also ‘with the agent’s rationality’ (ML: 103): ‘internal reason statements can be discovered in deliberative reasoning’ (ML: 104), he notes, and ‘what we can correctly ascribe to him in a third-personal internal reason statement is also what he can ascribe to himself as a result of deliberation’ (ML: 103). Must a construal of internalism informed by Williams’ later thought break with this aspect of his early formulation of the view?

Not at all. For notice what Williams is not saying here. He is not saying that the motivations we can ascribe to the agent in a third-personal statement are also what he can ascribe to himself, much less that the agent’s deliberation would refer to his own motivations. Williams is talking only about the result of the deliberation: the reason statement. What he says implies that the reasons we can ascribe to the agent in the third-personal form ‘A has reason to φ’ are also reasons that the agent can discover through deliberative reasoning and ascribe to himself in the first-personal form ‘I have reason to φ’. None of this contradicts the construal I advocate, which only maintains that an agent can properly deliberate from his S, in sound deliberation that is expressive of his S, and discover reason statements that are true in virtue of his S, without necessarily referring to his S in any of the premises figuring in

[p. 1103]

Internalism from the ethnographic stance the most explicit version of his deliberation. To say that the agent must be capable of soundly deliberating from his S to the conclusion that he has reason to φ is merely to say that his S must find ‘a sound deliberative expression’. What considerations become salient to an agent, and what sways him, can express underlying motivations of the agent without making reference to these motivations.

This is most obvious where motivations find unsound deliberative expression. The loftiest deliberation in terms of motivation-independent considerations might nonetheless express the agent’s desire for revenge, or longing for justice, or desire for good news. It might also express the wish to avoid uncomfortable clashes between values and projects. This is the area of motivated reasoning, and readers of Williams will have no trouble recognizing these examples as coming from his own work: one of his more Nietzschean streaks is his proclivity for exposing other philosophers’ efforts as unsound deliberative expressions of all-too-human motivations.11 What his internalist thesis reminds us of, however, is that there is also such a thing as sound motivated reasoning—indeed, that all practical deliberation, even when it makes no mention of motivations at the level of its content, should be expressive of the agent’s motivations if the reasons considered are really to be his. Internalism, Williams emphasizes, is ‘the only view that plausibly represents a statement about A’s reasons as a distinctive kind of statement about, distinctively, A’ (WME: 194). This preoccupation with the recognition and expression of individual differences is a recurring theme in Williams’ work. As he declared in one of his last interviews: ‘If there’s one theme in all my work it’s … self-expression … the idea that some things are in some real sense really you’ (Jeffries 2002).

IV. Against moral self-indulgence The distinction between deliberation expressing motivations and deliberation about motivations also plays a pivotal role in Williams’ critique of ‘moral selfindulgence’ (ML: 40–53; ELP: 11–3; WME: 212). This is the second element in his later thought that conflicts with the normative construal of internalism. Williams points out that a person who acts from generosity is precisely not motivated by the consideration that the act is generous; that would be to do something else—posing as a generous person, perhaps, or else displaying what Williams calls moral self-indulgence: being motivated primarily by one’s conception of oneself as having certain moral motivations. It is one thing to have a motivation in one’s S and to think and do things that express its presence. It is quite another to have the concept of that motivation, become conscious of that See Williams (ELP: 43, 64, 217; MSH: 73–5; WME: 216; 1999: 150; TT: 83; SP: 49).

[p. 1104]

motivation in one’s S, reason about the fact of its presence, and become motivated by that very fact. That is no longer to be moved by generosity, but by a second-order motivation referring to one’s first-order motivation. There may be special cases where such reflexivity is called for. But it would be an overgeneralization to conclude that all motivations worked like this. Williams considers it ‘one more mistaken consequence of Kantian moral theory’ (ML: 46) that the only genuine moral motivation it recognizes is a secondorder motivation that essentially involves the agent’s being conscious of himself as acting from duty. On Williams’ picture, an agent may be innocently motivated by items in his S that he has no concept of. Even if he possesses the concept of the relevant motivation, he may not apply it to himself when acting from that motivation. This would be what Williams calls intelligent innocence (ML: 46). And even if the concept is applied and the thought of his motivation is present, that is still ‘not the same as his motivation being provided by that thought’ (ML: 46). Some virtues, such as modesty, demand that one not be motivated by the thought of one’s modesty—being motivated by that thought would be to exhibit that other thing, ‘false modesty’. If someone genuinely possesses a virtue, there is generally no one concept they must deploy; rather, ‘certain ranges of fact become ethical considerations for that agent because he or she has that virtue’ (ELP: 11).

Similarly, if an agent has a particular ethical motivation, certain ranges of fact become reasons for that agent because he or she has that motivation. But this does precisely not imply that the reasons are derived from reflexive awareness of the motivation. That would be to confuse virtue with moral self-indulgence. Thus, Williams writes of his famous examples illustrating the importance of integrity that it would smack of moral self-indulgence if George and Jim were motivated ‘by thoughts about their own integrity’ (WME: 212,223n24).

The confusion is encouraged by the systematically ambiguous phrases available to external interpreters to describe agential motivations. Saying of an agent that he is ‘concerned with being generous’, for example, may mean merely that he is concerned to do the generous thing … or it may mean that he is concerned with his own generosity, where this implies that he had substituted for a thought about what is needed, a thought which focuses disproportionately upon the expression of his own disposition, and that he derives pleasure from the thought that his disposition will have been expressed—rather than deriving pleasure, as the agent who is not self-indulgent may, from the thought of how things will be if he acts in a certain way, that way being (though he need not think this) the expression of his disposition. (ML: 46, emphases added) For Williams, there is something disproportionate about allocating so much attention to one’s subjective motivations instead of focusing on the objective considerations that one’s motivations are supposed to attune one to. This is

[p. 1105]

Internalism from the ethnographic stance not just an epistemic failing, but an ethical one. It ultimately produces a misdirection not merely of attention, but of concern: it leads one to care about the wrong things (ML: 47).

So far from being Williams’ view, then, the normative construal of internalism, to the extent that it encourages a reflexive concern with our own motivations, falls prey to Williams’ critique of moral self-indulgence. Indeed, that construal might be said to be at odds with his critique of the ‘morality system’ more generally. For insofar as the popularity of the normative construal itself reflects the influence of Kantian moral theory and its embrace of the reflexive awareness of one’s own reverence for the moral law as the moral motive, the normative construal is not just a misconstrual of Williams, but an expression of the very ‘morality system’ he sought to resist.

By contrast, connecting Williams’ internalism to his later work allows us to see how it must fit in with a wider aspiration to arrive at a philosophical outlook that is psychologically realistic: paying careful attention to the phenomenology of moral deliberation from the first-personal perspective, he draws out subtle nuances such as that between being moved by a moral motive and being moved by one’s image of oneself as moved by a moral motive; and he tries to do justice both to the respects in which our reasons seem to be ‘simply there’ and to the respects in which those reasons express our motivations. V. Sympathetic understanding and the ethnographic stance Our oscillating awareness of these ‘external’ and ‘internal’ aspects of reasons is a manifestation of a remarkable capacity that becomes increasingly prominent in Williams’ later work:12 the capacity to make sense of the reasons possessed by others whose motivations differ, sometimes radically, from our own. And there are good reasons to think that Williams took this capacity to be connected to internalism.

The connection is alluded to when he points out that internalism implies a certain stance towards the agent: the stance towards the agent that is implied by the internalist account can be usefully compared to that of an imaginative and informed advisor, who takes seriously the formula ‘If I were you …’ (2001a: 92) Unlike the deliberative stance, the stance in which one takes seriously the formula ‘If I were you’ involves considering not simply what reasons there are, but what counts as a reason for the agent into whose shoes one is imaginatively slipping.13 This requires what Williams calls ‘sympathetic understanding’ (ELP: 99–101,119, 264n11), in the Humean sense that involves feeling one’s way into See Williams (ELP: 157; 1986: 203–4; WME: 207; 1995a: 239; TT: 50–1; SP: 61; 2021: 278). Manne (2014) harnesses a superficially similar, but more Strawsonian distinction.

[p. 1106]

another perspective to the point of resonating with its concerns. Sympathetic understanding in turn enables vicarious deliberation in light of what the agent in question cares about.

This requires a far more explicit grasp of that agent’s S than one needs when deliberating on one’s own account, because one cannot rely, as one can in one’s own case, on the relevant motivations expressing themselves in how one sees the situation and what considerations carry weight with one. To grasp how the agent’s reasons differ from one’s own, one has to build up a mental model of what his distinctive motivations are. As Williams stresses, this is ‘not a kind of intuitive telepathy’ (SP: 349)—it is based on coming to know a lot about how things stand with the agent.

The same capacity is at work when we encounter people ‘who live in a culture unfamiliar to us’, Williams explains. We then need to ‘get inside’ them and to have a sense of what it is to act for the kinds of reasons that these agents have; but this does not imply that these are reasons that we ourselves have. Someone who … can imagine, improvise, enact and respond appropriately to situations … has internalised the reasons of people living in that culture; but this does not make them his reasons …. This is the capacity to take up … ‘the ethnographic stance’. (SP: 60–1) Though our exercise of this remarkable capacity is by no means confined to ethnography, it is paradigmatically exemplified by an ethnographer in the field, who ‘can come to think as the people think with whom he is living’, and ‘enact their judgements and deliberations in his own person’, even though, ‘at the end of the line, these are not his thoughts’ (SP: 348). Accordingly, Williams refers to this as the capacity to take up the ethnographic stance, which enables us to make sense of the reasons possessed by others with different motivations. (He also sometimes invokes the image of the advisor in this connection, but the ethnographic stance, in this broad sense, underlies the stance of the advisor: genuinely adopting the advisor’s stance—as opposed to preaching at someone—presupposes sympathetic understanding.) In contrast to the deliberative stance, in which one tries to figure out what one has reason to do, the ethnographic stance involves imaginatively feeling and thinking one’s way into someone else’s position, as if one were them, and taking on parts of their outlook. For Williams, this is a ‘basic way of making sense of other people’ (TT: 237). But this identification with the outlook of another is ‘temporary and, as it were, feigned’ (TT: 237). It does not car r y over into how we run our own affairs. We imaginatively internalize their reasons—but this no more makes them our reasons than an actor’s internalization of Menelaus’ rage makes it the actor’s.14 See also Williams (SP: 60–1). Among the few to discuss this capacity are Thomas (2006: 146–52; 2007: 55) and Moore (2006, 2019).

[p. 1107]

Internalism from the ethnographic stance As Williams stresses in the preface to the French edition of ELP: ‘The possibility of the ethnographic stance, of understanding an alien structure of values which one does not share, is a basic datum for moral philosophy’ (2021: 278). In his response to Simon Blackburn’s review of the book, he observes: ‘The fact that the ethnographic stance is possible seems to me very important for moral philosophy’ (1986: 204).

If Williams attaches such importance to the ethnographic stance, it is because it both prompts and sustains philosophical reflection. By making it possible to understand other ways of reasoning, it awakens us to a variety of alternatives to our own outlook. Were we permanently locked in the perspective articulated by our own reasons, we could never ascend to the vantage point from which questions about metaethics and the objectivity of reasons come into view. Using the ethnographic stance, however, we can understand that there are alternative outlooks, and that these alternatives reflect something distinctive about the dispositions and motivations of those who live by them. The ethnographic stance gives rise to forms of metaethical reflection that we would not have without it.

What is more, we can also take up the ethnographic stance towards ourselves. Suspending our emotional and rational responsiveness to some of the reason statements that present themselves to deliberation, we can consider them in a detached and disengaged manner. We can thereby grasp that our own reasons reflect something of our dispositions and motivations. This is where philosophical anxieties over subjectivism and relativism become acute. But this is also where a thesis such as internalism comes into view, particularly if read not as a normative thesis, but as a piece of detached philosophizing about reasons. The application of the ethnographic stance to ourselves is a prominent theme in ELP, if not under that heading: in the most obvious sense it is not true that all ethical value rests in the dispositions of the self, and yet, in another way, it is true. It is not true from the point of view constituted by the ethical dispositions—the internal perspective—that the only things of value are people’s dispositions; still less that only the agent’s dispositions have value. … If we take up the other perspective, however, and look at people’s dispositions from the outside, we may ask the question ‘what has to exist in the world for that ethical point of view to exist?’ The answer can only be, ‘people’s dispositions’. There is a sense in which they are the ultimate supports of ethical value. That has a practical as well as a metaphysical significance. The preservation of ethical value lies in the reproduction of ethical dispositions. (ELP: 57–8) We can rephrase this passage to make its bearing on internalism more explicit: in the most obvious sense, it is not true that all reasons rest in the motivations or dispositions of the self, and yet, in another way, it is true. It is not true from the point of view constituted by those dispositions—the internal perspective— that the only things that give one reasons are people’s dispositions; still less

[p. 1108]

that only the agent’s dispositions give him reasons. If we take up the other perspective, however, and look at people’s dispositions from the outside, we may ask the question ‘what has to exist in the world for reasons to exist?’ The answer can only be, ‘people’s dispositions’. There is a sense in which they are the ultimate supports of reasons. That has a practical as well as a metaphysical significance. The preservation of reasons lies in the reproduction of dispositions.15 Notice that Williams’ distinction between two perspectives or stances here is accompanied by a distinction between two different accounts of the truth conditions of statements about what has value—and, by extension, of what reasons an agent has (‘in the most obvious sense, it is not true … yet, in another way, it is true…’). From the deliberative stance, the truth conditions do not depend on people’s motivations or dispositions. But from the ethnographic stance, the truth conditions do depend on people’s motivations or dispositions. On the resulting dual aspect account (i.e. an account involving two perspectives or stances), the internalist contention that reasons flow from human motivations is generally correct only when situated within the ethnographic stance; but it is precisely not meant to provide a generally correct characterization of how reason statements are to be understood from the deliberative stance. This vindicates the externalist refusal to regard all reasons for action as a function of human motivations. Yet, in leaving it at that, the externalist in effect overemphasizes the deliberative perspective at the expense of the ethnographic perspective—whereas those who interpret internalism as a normative constraint make the reverse mistake of importing into the deliberative perspective an insight that has its proper place in the ethnographic perspective. By contrast, Williams, from the mid-1980s onwards, is concerned to do justice to both perspectives. His assertion, in ELP, that ethical thought ‘rests in the dispositions of the self’ (ELP: 57), but ‘can never fully manifest the fact that it rests in human dispositions’ (ELP: 222), is only intelligible in the context of a dual aspect account. Philosophical reflection may reveal that human dispositions are the ultimate support of values and reasons; but it nonetheless remains essential to many of those values and reasons that they seem to, and in one important sense do, come ‘from outside’ (ELP: 212): they make claims on us that demand to be understood not as outgrowths of our motivations, but as normative reality disclosing itself to us.

If interpreted as a normative deliberative constraint, by contrast, the internalist thesis would deny those claims any form of autonomy from subjective motivations, since, by systematically forcing to the fore a reflective awareness of one’s subjective motivations, it would impel ethical thought to fully manifest the extent to which it rests in human dispositions even from the deliberative See also Williams (PHD: 67–75).

[p. 1109]

Internalism from the ethnographic stance perspective. That would be to distort ethical thought beyond recognition, engrossed by a philosophical insight dimly grasped, but delivered in the wrong place.

On the construal advocated here, the claim that A has a reason to φ only if there is a sound deliberative route from A’s S to A’s φ-ing is thus a philosophical claim, proffered from and for the ethnographic stance, that proposes to elucidate in virtue of what agents have reasons for action. This construal does not invite us to regard the thought ‘I have D in my S’ as always appropriate in deliberating to the conclusion that ‘I have reason to φ’. It is appropriate only in dealing with reasons that can be correctly understood from the deliberative stance as motivation-dependent reasons—whether I have reason to book a beach holiday in Sardinia rather than a hiking holiday in Austria may indeed depend, above all, on my motivations. But we can accept that other reasons, particularly those of a moral, social, or institutional kind, phenomenologically present themselves as reasons we have irrespective of our motivations. Thus, when Brad Hooker points out on the externalist’s behalf that at least some deliberation ‘starts not from the agent’s own subjective present motivations, but from some objective (“external”) values or requirements, fixed independently of the agent’s present motivations’ (1987: 43), he is asserting a plain truth that Williams can freely grant. My reason to give to famine relief can be grounded simply in the severity of the famine and the duty to help. But in reasoning from these premises and being struck by the force of the claim they make on me, I nonetheless express something of my S. And by understanding this connection as I reflect on it from the ethnographic stance, I demystify the force of this claim, grasping how ‘it seems to come “from outside” in the way that conclusions of practical necessity always seem to come from outside— from deeply inside’ (ELP: 212).

Hooker himself contemplates the possibility that internalism might be interpreted to allow for deliberation that essentially starts from objective considerations. But if the agent’s own motivations need not figure in his deliberation even on an internalist account of reasons, Hooker worries, internalism becomes inert or ‘impotent’ (1987: 43). In one sense, this is true: the thesis no longer rules reasons out of court simply because they purport to be grounded only in objective or external considerations—it does not require that my reason to φ be derivable from my S as a conclusion is derivable from a set of premises.

Yet, in another sense, it is not true that internalism becomes inert: internalism demystifies reason statements, giving those who had qualms about external-seeming reasons license to relax into heeding them. What it provides, in good Humean fashion, is an explanation of where reasons for action ultimately come from, an explanation which serves to render our understanding of ethical thought unmysterious and ‘naturalistic’ in Williams’ preferred sense (PHD: 73). It thereby also demystifies the authority of these reasons: they are

[p. 1110]

not simply coercive impositions from outside, but are imbued with authority by the fact that they ultimately speak to our own motivations. Significantly for a historical thinker such as Williams, internalism from the ethnographic stance also allows us to make philosophical sense of the fact that different reasons make normative sense to different people and cultures. As he notes, the internalist thesis ‘would be pointless if everyone’s values, and everyone’s S, were the same’ (PHD: 114). It is a distinctive feature of Williams’ philosophy from ELP onwards that, at least in the modern agent, a commitment to moral reasons must coexist with a reflective awareness of the fact that different reasons make normative sense to different people. Internalism answers to the need to make sense of that fact.

VI. Corrosive vs. corroborative sense-making At this point, however, a different worry arises which will not be dispelled by situating internalism within the ethnographic stance and recognizing that our motivations need not explicitly figure in our deliberations to be expressed in them. This is the worry, voiced in different forms by a number of commentators, that a wholehearted commitment to morality cannot coexist, in the same agent, with a philosophical understanding of reasons as a function of our contingent motivations.16 Accepting internalism, even if only from the ethnographic stance, threatens to have a corrosive effect on moral conviction. A look at Williams’ later work reveals that he himself came to feel the force of this corrosiveness worry. In ELP, especially, Williams acknowledges that the conviction with which we act on moral reasons cannot be isolated from our reflective philosophical understanding of the source of these reasons. Even if internalism is construed as a thesis about the rational perspective rather than from that perspective, an agent’s reflective understanding of that perspective is still the agent’s, and demands to be integrated into the agent’s life. Though distinctions such as that between reflection and action entice us to forget it, there is ultimately only one enterprise—living—and, as Williams reminds us, we have to live not only after reflection, but ‘during it as well’ (ELP: 130). What is more, Williams recognizes the requirement that ‘an outside viewpoint from which I can understand my dispositions should not alienate me from them’ (ELP: 254n18). And he acknowledges that, under conditions of modernity, understanding our moral outlook as grounded in our dispositions threatens to be corrosive. On an Aristotelian view of human nature, the virtuous agent could still draw confidence from the thought that his dispositions were the uniquely correct ones—the culmination of the human telos. But on a See Brink (1986), Nagel (1986), Thomas (2006: ch. 6), Sagar (2014), and Smyth (2019).

[p. 1111]

Internalism from the ethnographic stance modern view of human nature, the agent’s dispositions appear rationally contingent: they are only one set among many that are compatible with human nature, and there seem to be no neutrally identifiable reasons to validate that set over possible alternatives. ‘With that gap opened’, Williams grimly observes, ‘the claim I expressed by saying that agents’ dispositions are the ‘ultimate supports’ of ethical value takes on a more skeptical tone. It no longer sounds enough’ (ELP: 59).17 Yet Williams found a way out of this predicament. He did not relinquish internalism—the final chapter of his last book still unapologetically connects the notion of ‘making sense’ to internalism (TT: 237,306n6). But he sought to preserve the spirit of internalism and put it to work while avoiding the wholesale scepticism it seemed to invite in ELP. The question became the following: Is there a way of reflectively making sense of our reasons as outgrowths of our motivations while dispelling the air of rational contingency that hangs over those motivations?

The trajectory of Williams’ work after ELP can be understood as guided by a concern to answer this question. He eventually carved out a form of philosophical sense-making capable not only of avoiding the corrosive effect of cruder forms of metaethical internalism, but of affirming the conviction with which we heed certain reasons with an externalist phenomenology. Thus, his later books offer philosophical explanations of shame, responsibility, and the intrinsic value of truth that are meant to be, in Williams’ own phrase, vindicatory rather than subversive, strengthening rather than weakening our confidence (TT: 283n19).

To contrast it with metaethical sense-making that is corrosive, I shall call this kind of philosophical sense-making corroborative—in the root sense of strengthening or making more robust (from robur, ‘strength’ or ‘robustness’). What corroborative sense-making strengthens or makes more robust is our confidence in certain types of reason statements. Confidence becomes a central notion in Williams’ thought from ELP onwards.18 I take it to refer notably to our willingness to relax into heeding reason statements, i.e. to allow ourselves to be rationally and emotionally responsive to them and act on them, without hesitation or doubt.

To be corroborative rather than corrosive, an internalist account needs to present our motivations as rationally necessary rather than rationally contingent. As Williams made clear in ELP, he did not think our dispositions or motivations could be shown to be rationally necessary in anything like an Aristotelian or Kantian sense. But his later work does hold out the prospect of dispelling This is connected to the ‘cold vacancy’ felt by the ‘shaken realist’ in one of the epigraphs Williams chose for ELP; see Krishnan & Queloz (2023).

See Williams (ELP: 189; MSH: 203–9; WME: 207–8; 2000: 160; TT: 283n19; 2009: 199–210). On confidence, see Fricker (2000), Moore (2003), Hall (2014), Blackburn (2019), and Łukomska (2022). I develop a fuller account of confidence in Queloz (2024).

[p. 1112]

the contingency of at least some of our motivations by identifying reasons why we need them—reasons that are not necessarily reasons for any rational agent or even any human being, but are nonetheless reasons for us.19 To recover this particular form of rational necessity within the internalist framework, one first needs to work at the right level of generality. Instead of focusing on individual reason statements as they might figure in deliberation or making grand metaethical generalizations about all reasons for action, Williams comes to operate at a level lying subtly between these: he explores how a particular practice of treating a certain type of consideration as a reason for action relates to particular motivations of ours. Reflecting on reason statements involving obligations, for example, Williams invites us to explore ‘why we should have a special kind of ethical reason that we call an “obligation”’ by considering ‘what is likely to come about if we have a practice of recognizing obligations as a type of ethical consideration’ (PHD: 73). By putting internalism to work on a practice-by-practice basis, Williams opens up room for a contrast between different ways in which certain reasoning practices, together with the motivations immediately underpinning them, can relate to the rest of our motivational set. In light of everything else we care about, does a given reasoning practice help us to live, as Williams (2000: 161) puts it? Some practices may appear rationally contingent even in this light, because our remaining motivations give us no reason to cultivate those practices in particular; some practices may turn out to be radically in tension with the rest of our motivations, which suggests that we had better stop engaging in those practices; but at least some practices may find support in the rest of our motivational set—in the best case, they may turn out to be crucial to furthering more basic motivations that are certainly important to us. Coming to understand ‘the point and value of living a life in which obligations counted as ethical reasons’ (PHD: 73), for example, would demystify the notion of obligation and its hold over us. But it would also enable us to place the practice of deliberating in terms of obligations—both in relation to the explanatory resources we recognize as legitimate (can this practice be made sense of in naturalistic terms?) and in relation to our own motivational set (in light of our desires, projects, and commitments, does the explanation present the practice of reasoning in terms of obligations as serving a point worth serving?).

To dispel the air of contingency attaching to a particular reasoning practice and the motivation it expresses, Williams thus seeks an explanation that reveals this bundle of reason and motivation to be rationally necessary in view of some further motivation that we are more deeply identified with and more confident in—a motivation providing a reason why we need the seemingly contingent motivation and the reasoning practice it underpins. There are thus three items This is connected to Williams’ ‘Left Wittgensteinianism’. See Queloz & Cueni (2021).

[p. 1113]

Internalism from the ethnographic stance in play: a type of reason for action; the motivation immediately underpinning sensitivity to that reason for action; and a further motivation providing a reason why we need that first motivation and its concomitant reason for action. The species of rational necessity that Williams aims to uncover is therefore instrumental or functional: apparently contingent reasons and motivations that we did not think of in functional terms turn out to serve a function in relation to more fundamental motivations and to be functionally necessary if we are going to satisfy these motivations. Thus, his account of responsibility in SN and elsewhere turns on the idea that ‘moral responsibility has a function’ (PHD: 125);20 his examination of ‘liberty’ and ‘freedom’ inquires into ‘the point of why we want these terms in the first place’ (IBD: 79); and a key step in his genealogy of the intrinsic value of truth is to raise the question of the ‘function’ of reasoning in these terms. ‘[T]hat step itself does some of the work’, Williams observes; ‘a functional account is given of something that not everyone would expect to have a functional account; and the account is given in terms of motivations that people must be granted to have anyway’ (TT: 31–3). However, Williams came to see that such functional accounts must be given in a certain way if they are not to come out as crudely reductive and ‘simply false’ (TT: 34). They must not be understood as attempts to justify particular actions; nor as attempts to spell out the importance of a certain type of consideration from the deliberative stance. Rather, they are philosophical explanations offered from something like the ethnographic stance: they are attempts to make explanatory sense of why certain reason statements make normative sense to us by placing the practice of reasoning this way within our wider motivational economy and finding a valuable role for it.

To clearly demarcate what he was trying to do from normative or justificatory reflection, Williams settled on partly imaginary genealogical narratives as a suitable format in which to offer such non-reductive functional explanations. These are narratives that start out from a fictional ‘State of Nature’ and explain the transition from an initial situation in which some reason for action R is not yet operative to a target situation in which it is operative.21 They are explanatory because they present the practice of heeding R as functional in relation to some motivation M that is taken as given in the initial situation: the practice of heeding R is shown to be conducive to furthering motivation M. As a result, the narratives present the transition to the target situation as rational: agents in the initial situation would welcome the transition as an improvement, and would, if they could, aim for the target situation (TT: 34). Yet Williams’ genealogical method is distinctive in insisting that when it represents this rational relation in terms of a process of sound deliberation, On this point, see Queloz (2021a, 2022b).

I offer a systematic account of this form of the genealogical method in Queloz (2021b); for a discussion focused on whether such genealogies must be reductive, see Queloz (2022a).

[p. 1114]

this is just a useful fiction (2000: 158; TT: 34).22 In actual fact, the transition was not, and could not have been, the result of deliberation; nor is the reasoning practice now sustained by an awareness of this functional connection. The whole point of using genealogy is to reveal an unsuspected functional connection between the practice of heeding some reason for action R and some motivation M, and it is unsuspected precisely because agents who heed R do not think of R in terms of its subservience to M. On the contrary: in the cases Williams is concerned with, such as the instrinsic value of truth, R demands to be understood in ‘external’ terms that resist functional explanation in terms of something ‘internal’ like M.23 And the particular strength of Williams’ explanation is that it accounts for this externalist phenomenology by showing how the real functional dynamics revealed by the ‘State of Nature’-fiction themselves require the value of truth to outrun its function and be understood in non-functional terms. The externalist phenomenology is thus not an accident, but something that can itself be shown to be rationally necessary in functional terms.

This retains the internalist idea that there must in principle be a sound deliberative route by which our reasons for action can be shown to tie in with our motivational set; but highlighting this deliberative route is not meant to help us figure out what to do; it is meant to help us make explanatory sense of why we came to heed certain types of reasons, and why it is right for these reasons to present themselves as motivation-independent. By representing that route through a genealogical story that situates the deliberation in the minds of expressly fictional agents in an imagined ‘State of Nature’, Williams clarifies that the place to appreciate this rational connection is from the ethnographiccum-genealogical stance rather than from the deliberative stance. Indeed, by showing that the functional connection to our motivations must, if the practice is to be functional, efface itself in favour of more external-seeming reasons, Williams explains why the two perspectives must come apart.24 What this genealogical procedure helps one achieve, according to Williams, is ‘explanation without reduction’ (TT: 90). Thus outsourced to a fictional past, the functional derivation of external-seeming reasons from motivations becomes harder to mistake for a descriptive or normative account of deliberation in the present. The reasons revealed by the genealogy are not meant to replace the reasons in terms of which we normally articulate the importance of truthfulness. On the contrary: Williams comes to realize that what enables the intrinsic value of truth to remain stable under philosophical reflection is that, unlike people who try to reason themselves into intrinsically valuing He took himself to be following Nietzsche in this; see Queloz (2021c). See Williams (TT: 33–7). On related rationales for this kind of genealogy, see also Queloz (2020).

On this notion of self-effacing functionality, see Queloz (2018).

[p. 1115]

Internalism from the ethnographic stance something based on nothing but the insight into the instrumental value of doing so, we already possess the means to make sense of its value ‘from the inside’ (TT: 91): our reasons to be truthful are supported by much more than the mere thought that our heeding them is instrumental to furthering other motivations of ours. These reasons are embedded in a wider evaluative outlook that renders normatively intelligible, from the deliberative stance, why truthfulness matters in its own right (TT: 91–2). The genealogy does not displace these reasons—it bolsters them by showing that it makes explanatory sense for them to be in place. We can thus stably see the world from the point of view of the dispositions and motivations involved in heeding these reasons without reducing the normative authority of the reasons we then heed to the instrumental value of those dispositions and motivations. The reasons still make sense to us as authoritative in their own terms.

This is not, Williams stresses, because we choose to accept these reasons as authoritative. What makes sense to us is typically not subject to our will. The Kantian ideal of autonomy whereby one critically chooses to acknowledge a reason as having authority over oneself is to that extent misleading. One does not decide that some reason for action, or some explanation for why we reason in these terms, makes sense; it comes as a discovery (TT: 261–2). As Williams puts it: ‘neither in the case of finding explanations, nor in considering what to do, can one choose what makes sense to one’ (SP: 333). But if, by discovering a functional connection between our external-seeming reasons and some of our most important motivations, we manage to make explanatory sense of the way in which certain reasons make normative sense to us, this can corroborate rather than corrode these reasons. This is unlikely to work for all our ethical ideas. But it does render intelligible how philosophy can help us ‘cherish as many as we can’ (ELP: 130)—a connection Williams explicitly draws at the end of TT (231).

Once a contrast opens up between corrosive and corroborative sensemaking, moreover, we begin to see how explanatory reflection from the ethnographic-cum-genealogical perspective can indirectly infor m nor mative deliberation. Williams resists analytic philosophy’s attempts ‘to segregate the philosophical from the normative’ (PHD: 155). He takes philosophical reflection on the connection between reasons and motivations to be capable of informing the deliberative stance by affecting our confidence in certain types of reason statements.

This emerges already from Williams’ response to Blackburn’s (1986) review of ELP. Blackburn draws a sharp distinction between explanatory philosophizing about reasons and moralizing with reasons, where the latter refers to normative deliberation. For Blackburn, these are two radically different language games, and only moralizing can properly yield justificatory considerations bearing on what to do.

[p. 1116]

Williams does not think we should distinguish that sharply between explanatory philosophizing and normative deliberation. The two ultimately have to be integrated within the consciousness of a single agent, allowing explanatory understanding to seep through into deliberative reflection. Indeed, good deliberative reflection should be guided by explanatory understanding: There is, no doubt, explanatory reflection that is not at all deliberative: but there is no thorough-going and adequate deliberative reflection that does not involve itself in explanation. Good deliberative reflection is guided by a good understanding of how things are. (1986: 207) Even on the construal advocated here, therefore, the internalist connection between reasons and motivations is still meant to inform deliberation—only less directly than the normative construal suggests.

By elucidating how the practice of reasoning in terms of a certain notion of obligation relates to our wider motivational set, for instance, reflection from the ethnographic stance could affect our confidence in that notion and the reason statements it puts within our purview. It might strengthen our confidence, and correspondingly weaken our confidence in rival conceptions that remained incurably mysterious; or it might weaken our confidence, revealing its role to be less significant than we expected, or even downright harmful. Similarly, Williams’ project in TT is to let an explanatory understanding of the intrinsic value of truth inform our attitude towards reason statements appealing to that value. This explanation is advanced in answer to a practical question raised by Richard Rorty: whether we should cease to heed reason statements claiming authority in terms of the intrinsic value of truth. Williams’ explanation informs our deliberation by strengthening our confidence in the practice of heeding this type of reason statement. It does not seek to justify any particular reason statement. But it does seek to offer us a renewed, thoroughly naturalistic, yet corroborative way of making sense of that practice. Philosophers given to what Williams calls ‘Platonic contempt for the human and the contingent’ (ER: 266) will still feel that this is not enough. But, as Williams emphasizes against Rorty, we are not ‘unencumbered intelligences selecting in principle among all possible outlooks’ (PHD: 193). We want the ways of reasoning that best suit our distinctive situation. And a vindicatory genealogy does yield reasons, for those of us who are confidently identified with the motivations that the value of truth turns out to serve, to extend our confidence to this value and the reasons it provides. This is how philosophical sense-making can dispel the apparent contingency of our concern for the truth.

VII. Conclusion According to this reconstruction of what might be termed Williams’ ‘late internalism’, genealogical explanation from the ethnographic stance can

[p. 1117]

Internalism from the ethnographic stance strengthen our confidence in certain reason statements by revealing how the practice of reasoning that way ties in with other motivations we are already more confident in. Williams is thus not a Wittgensteinian quietist who enjoins us to accept the form of life we find and eschew attempts to explain why these rather than other reasons figure in it. But neither does Williams share the sensibility of genealogical debunkers who think that our reason statements should ultimately be capable of being justified directly in terms of their genealogical explanations, and jettisoned otherwise.

Instead, the view we get from integrating Williams’ internalism with his later thought combines three ideas: that the justifications for our actions sometimes properly come to an end in reasons that are, for us, simply there; that these reasons can be understood, at the level of philosophical explanation, as being simply there for us because of who we are and what motivations we have; and that understanding how these reasons relate to our wider motivational set can tell us whether to relax into heeding them. On this construal, internalism can also be seen to tie in with wider ethical and political concerns of Williams. For it can inform our view of how we should treat others who take a different view of what their reasons are. It can affect our confidence not merely in our own reason statements, but in those of others in relation to our own. Understanding how their reasons relate to their motivational sets can tell us whether to dismiss their view as an expression of ignorance, confusion, or delusion, or whether to respect it as an expression of motivations that differ from our own, but that we can nevertheless make sense of, both in the sense that we can make explanatory sense of how they came by them, and that we can see how their preferred actions make normative sense in terms of these motivations. That need not make disagreement any less vigorous. But it tells us how to address others who take a different view of what their reasons are—paternalistically, as one might speak to children who had misunderstood something within the game they were trying to learn, or as equals, as one might speak to people standing on rational ground as solid as our own.

Internalism thereby provides a philosophical basis for respectful disagreement—disagreement that is respectful of the other party’s experience and their own sense of what their reasons are.25 In this regard, too, the proposed construal chimes with the rest of Williams’ thought. It reveals his internalism to be the philosophical underpinning of his liberalism.26 We need not assume that Williams already implicitly intended his internalism to carry all these qualifications in the 1970s. I have merely been arguing that once we try to integrate his initial thesis with his later thought, we find See Williams (2001b: 102; IBD: 13; PHD: 195). For detailed reconstructions of Williams’ arguments for the necessity of respectful disagreement, see Queloz (2023) and Cueni (2024). See Williams (PHD: 195) as well as Cueni & Queloz (2021) and Queloz & Cueni (2021).

[p. 1118]

much relevant and congenial material with which to probe and refine our construal of that thesis. The result not only renders internalism more subtle and interesting, but, fittingly, reveals how it speaks to Williams’ wider motivations.27

## References

Alvarez, M. (2010) Kinds of Reasons: An Essay in the Philosophy of Action. Oxford: OUP. Blackburn, S. (1986) ‘Making Ends Meet’, Philosophical Books, 27: 193–203. Blackburn, S. (2019) ‘Lonely in Littlemore: Confidence in Ethics and the Limits of Philosophy’, in S.-G. Chappell and M. van Ackeren (eds.) Ethics beyond the Limits: New Essays on Bernard Williams’ Ethics and the Limits of Philosophy, pp. 27–36. London: Routledge. Brandhorst, M. (2007) ‘Foundations of Practical Reason’, DPhil Thesis, University of Oxford. Brink, D. O. (1986) ‘Externalist Moral Realism’, The Southern Journal of Philosophy, 24: 23–41. Cueni, D. (2024) ‘Constructing Liberty and Equality: Political, Not Juridical’, Jurisprudence, 1–20. Cueni, D. and Queloz, M. (2021) ‘Whence the Demand for Ethical Theory?’, American Philosophical Quarterly, 58: 135–46. Dancy, J. (2000) Practical Reality. Oxford: OUP. FitzPatrick, W. J. (2008) ‘Robust Ethical Realism, Non-Naturalism, and Normativity’, Oxford Studies in Metaethics, 3: 159–205. Fricker, M. (2000) ‘Confidence and Irony’, in E. Harcourt (ed.) Morality, Reflection, and Ideology, pp. 87–112. Oxford: OUP. Fricker, M. (2020) ‘Bernard Williams as a Philosopher of Ethical Freedom’, Canadian Journal of Philosophy, 50: 919–33. Hall, E. (2014) ‘Contingency, Confidence, and Liberalism in the Political Thought of Bernard Williams’, Social Theory and Practice, 40: 545–69. Heathwood, C. (2011) ‘Desire-Based Theories of Reasons, Pleasure and Welfare’, Oxford Studies in Metaethics, 6: 79–106. Heuer, U. (2000) ‘Haben Meine Gründe Etwas mit Mir Zu Tun? Bernard Williams’ Erwiderung auf John McDowell’, in G. Meggle and J. Nida-Rümelin (eds.) Rationalität, Realismus, Revision /Rationality, Realism, Revision, pp. 606–14. Berlin, Boston: De Gruyter. Heuer, U. (2004) ‘Reasons for Actions and Desires’, Philosophical Studies, 121: 43–63. Heuer, U. (2012) ‘Thick Concepts and Internal Reasons’, in U. Heuer and G. Lang (eds.) Luck, Value and Commitment: Themes from the Ethics of Bernard Williams, pp. 219–48. Oxford: OUP. Hollis, M. (1979) ‘Rational Man and Social Science’, in R. Harrison (ed.) Rational Action: Studies in Philosophy and Social Science, pp. 1–16. Cambridge: CUP. Hooker, B. (1987) ‘‘Williams’ Argument against External Reasons’, Analysis, 47: 42–4. Jeffries, S. (2002) ‘The Quest for Truth’, The Guardian, November 30. Krishnan, N. and Queloz, M. (2023) ‘The Shaken Realist: Bernard Williams, the War, and Philosophy as Cultural Critique’, European Journal of Philosophy, 31: 226–47. Łukomska, A. (2022) ‘Confidence: On the Possibility of Ethical Knowledge’, in A. Szigeti and M. Talbert (eds.) Morality and Agency: Themes from Bernard Williams, pp. 110–31. New York: Oxford University Press. Manne, K. (2011) Not by Reasons Alone, PhD Thesis, MIT. I am grateful to Claus Beisbart and Nikhil Krishnan for detailed feedback on earlier versions. Discussions with Damian Cueni, Andreas Müller, Richard King, Markus Stepanians, Victor Braga Weber, Sophia Dandelet, and Selim Berker helped me flesh out the argument. I would also like to thank the three anonymous reviewers and the editor, Franz Berto, for an exemplary peer review process from which the paper benefited enormously. My work on this paper was supported by the Swiss National Science Foundation.

[p. 1119]

Internalism from the ethnographic stance Manne, K. (2014) ‘Internalism about Reasons: Sad but True?’, Philosophical Studies, 167: 89–117. Markovits, J. (2011) ‘Internal Reasons and the Motivating Intuition’, in M. Brady (ed.) New Waves in Metaethics, pp. 141–65. London: Palgrave. Markovits, J. (2014) Moral Reason. Oxford: OUP. McDowell, J. (1995) ‘Might There Be External Reasons?’, in J. E. J. Altham and R. Harrison (eds.) World, Mind, and Ethics: Essays on the Ethical Philosophy of Bernard Williams, pp. 68–85. Cambridge: CUP. Millgram, E. (1996) ‘‘Williams’ Argument against External Reasons’, Noûs, 30: 197–220. Moore, A. W. (2003) ‘Williams on Ethics, Knowledge and Reflection’, Philosophy, 78: 337–54. Moore, A. W. (2006) ‘Maxims and Thick Ethical Concepts’, Ratio, 19: 129–47. Moore, A. W. (2019) ‘Realism and the Absolute Conception’, Language, World, and Limits: Essays in the Philosophy of Language and Metaphysics, pp. 165–81. Oxford: OUP. Nagel, T. (1986) The View from Nowhere. New York: Oxford University Press. Ng, G. (2017) ‘Bernard Williams and the Materials of Internalism’, PhD Thesis, University of Reading. Paakkunainen, H. (2018) ‘Internalism and Externalism about Reasons’, in D. Star (ed.) The Oxford Handbook of Reasons and Normativity, pp. 143–70. Oxford: OUP. Parfit, D. (2011) On What Matters. Oxford: OUP. Queloz, M. (2020) ‘From Paradigm-Based Explanation to Pragmatic Genealogy’, Mind, 129: 683–714. Queloz, M. (2021a) ‘The Self-Effacing Functionality of Blame’, Philosophical Studies, 178: 1361–79. Queloz, M. (2021b) The Practical Origins of Ideas: Genealogy as Conceptual Reverse-Engineering. Oxford: OUP. Queloz, M. (2021c) ‘Choosing Values? Williams contra Nietzsche’, The Philosophical Quarterly, 71: 286–307. Queloz, M. (2022a) ‘Genealogy, Evaluation, and Engineering’, The Monist, 105: 435–51. Queloz, M. (2022b) ‘The Essential Superficiality of the Voluntary and the Moralization of Psychology’, Philosophical Studies, 179: 1591–620. Queloz, M. (2023) ‘The Dworkin–Williams Debate: Liberty, Conceptual Integrity, and Tragic Conflict in Politics’, Philosophy and Phenomenological Research, 1–27. Queloz, M. (2024) The Ethics of Conceptualization: A Needs-Based Approach. Oxford: OUP. Queloz, M. and Cueni, D. (2021) ‘Left Wittgensteinianism’, European Journal of Philosophy, 29: 758–77. Sagar, P. (2014) ‘Minding the Gap: Bernard Williams and David Hume on Living an Ethical Life’, Journal of Moral Philosophy, 11: 615–38. Scanlon, T. M. (1998) What We Owe to Each Other. Cambridge, MA: Harvard University Press. Scanlon, T. M. (2014) Being Realistic about Reasons. Oxford: OUP. Schroeder, M. (2007) Slaves of the Passions. Oxford: OUP. Schroeder, M. and Finlay, S. (2017) ‘Reasons for Action: Internal vs. External’, in E. N. Zalta (ed.) The Stanford Encyclopedia of Philosophy. Skorupski, J. (2007) ‘Internal Reasons and the Scope of Blame’, in A. Thomas (ed.) Bernard Williams, pp. 73–103. Cambridge: CUP. Smith, M. (2012) ‘A Puzzle about Internal Reasons’, in U. Heuer and G. Lang (eds.) Luck, Value and Commitment: Themes from the Ethics of Bernard Williams, pp. 195–218. Oxford: OUP. Smyth, N. (2019) ‘The Inevitability of Inauthenticity: Bernard Williams and Practical Alienation’, in S.-G. Chappell and M. van Ackeren (eds.) Ethics beyond the Limits: New Essays on Bernard Williams’ Ethics and the Limits of Philosophy, pp. 188–208. London: Routledge. Sobel, D. (2016) From Valuing to Value: A Defense of Subjectivism. Oxford: OUP. Thomas, A. (2006) Value and Context: The Nature of Moral and Political Knowledge. Oxford: Clarendon Press. Thomas, A. (2007) ‘The Nonobjectivist Critique of Moral Knowledge’, in A. Thomas (ed.) Bernard Williams, pp. 47–72. Cambridge: CUP. Wallace, R. J. (1990) ‘How to Argue about Practical Reason’, Mind, 99: 355–85. Wallace, R. J. (1999) ‘Three Conceptions of Rational Agency’, Ethical Theory and Moral Practice, 2: 217–42.

[p. 1120]

Wedgwood, R. (2011) ‘Instrumental Rationality’, Oxford Studies in Metaethics, 6: 280–309. Williams, B. (1971) ‘Conversation with Bernard Williams: Philosophy and Morals’, in B. Magee (ed.) Modern British Philosophy, pp. 150–65. New York: St. Martin’s Press. Williams, B. (1973) Problems of the Self: Philosophical Papers 1956–1972. Cambridge: CUP. [PS] Williams, B. (1979) ‘Internal and External Reasons’, in R. Harrison (ed.) Rational Action: Studies in Philosophy and Social Science, pp. 17–28. Cambridge: CUP. Williams, B. (1981) Moral Luck. Cambridge: CUP. [ML] Williams, B. (1985) Ethics and the Limits of Philosophy. London: Routledge. [ELP] Williams, B. (1986) ‘Reply to Simon Blackburn’, Philosophical Books, 27: 203–8. Williams, B. (1995) ‘‘Replies’, in J.E.J. Altham and R. Harrison (eds.) World, Mind, and Ethics: Essays on the Ethical Philosophy of Bernard Williams, pp. 185–224. Cambridge: CUP. [WME] Williams, B. (1995) Making Sense of Humanity and Other Philosophical Papers 1982–1993. Cambridge: CUP. [MSH] Williams, B. (1995a) ‘Truth in Ethics’, Ratio, 8: 227–42. Williams, B. (1995b) ‘Ethics’, in A.C. Grayling (ed.) Philosophy 1: A Guide through the Subject, pp. 545–82. Oxford: OUP. Williams, B. (1999) ‘Bernard Williams’, in A. Pyle (ed.) Key Philosophers in Conversation: The Cogito Interviews, pp. 142–63. London: Routledge. Williams, B. (2000) ‘Naturalism and Genealogy’, in E. Harcourt (ed.) Morality, Reflection, and Ideology, 148–61. Oxford: OUP Williams, B. (2001) Morality: An Introduction to Ethics. Cambridge: CUP. [M] Williams, B. (2001a) ‘Postscript: some Further Notes on Internal and External Reasons’, in E. Millgram (ed.) Varieties of Practical Reasoning, pp. 91–7. Cambridge, MA: MIT Press. Williams, B. (2001b) ‘Liberalism and Loss’, in M. Lilla, R. Dworkin and R. Silvers (eds.) The Legacy of Isaiah Berlin, pp. 91–103. New York: New York Review of Books. Williams, B. (2002) Truth and Truthfulness: An Essay in Genealogy. Princeton: Princeton University Press. [TT] Williams, B. (2005) In the Beginning Was the Deed: Realism and Moralism in Political Argument. Princeton: Princeton University Press. [IBD] Williams, B. (2006) Philosophy as a Humanistic Discipline. Princeton: Princeton University Press. [PHD] Williams, B. (2006) The Sense of the Past: Essays in the History of Philosophy. Princeton: Princeton University Press. [SP] Williams, B. (2009) ‘A Mistrustful Animal’, in A. Voorhoeve (ed.) Conversations on Ethics, pp. 195– 214. Oxford: OUP. Williams, B. (2014) Essays and Reviews 1959–2002. Princeton: Princeton University Press. [ER] Williams, B. (2021) ‘Ethics, a Matter of Style? Introduction to the French Edition’, Philosophical Inquiries, 9: 269–84. © The Author(s) 2024. Published by Oxford University Press on behalf of The Scots Philosophical Association and the University of St Andrews. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs licence (https://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial reproduction and distribution of the work, in any medium, provided the original work is not altered or transformed in any way, and that the work is properly cited. For commercial re-use, please contact journals.permissions@oup.com