. 3
( 38 .)


either to ¬ction or accept that we cannot have what we want.”29

25 Skorupski, this volume.
26 Williams (1995c), p. 220, n. 3.
27 Nagel (1970), p. 4.
28 Williams (1995a).
29 Skorupski, this volume.

Dissatis¬ed with Williams™ own “proleptic theory of blame,” Skorupski
concludes that given that the very capacity to recognize a reason as such
is fundamental to moral agency there is a tension within this notion of
modern moral agency. It is generated by its internal drive toward crediting
everyone with “comprehensive” moral agency that, Skorupski argues, may
be false. (That drive is connected to the idea of equal respect, construed
itself as respect for those who are able to recognize moral reasons as such.)
The result, Skorupski concludes, will be a humanization of our ideals of
moral agency “ which is not to abandon them. To follow the reconstruc-
tion of Williams™ arguments that Skorupski recommends is to accept an
accurate diagnosis of a genuine tension within modern moral thought; to
retreat into one, Humean strand of Williams™ arguments however is, he
argues, “just another dogma of empiricism.” However, Skorupski convinc-
ingly demonstrates not only that the internal reasons thesis and the critique
of the “morality system” are intimately connected but also that “Hume™s
conception of practical reasons is neither the only starting point, nor the
best starting point, for Williams™ questions about morality.”30
Williams™ critique of the morality system is the explicit focus of the
paper by Kant scholar and moral philosopher Robert B. Louden.31 The
context of Williams™ presentation of the argument in Ethics and the Limits of
Philosophy made its explicit target appear to be Kant. However, Skorupski™s
discussion has already suggested that this cannot be wholly accurate and
in his scholarly examination of how much of Kant™s ethical theory could
reasonably be construed as a target of Williams™ critique Louden gives
further reason to dissent from this interpretation. Louden ¬rst identi¬es
the four basic charges that Williams leveled against the morality system in
this passage:

[M]orality should be understood as a particular development of the ethical,
one that has special signi¬cance in modern Western culture. It particularly
emphasizes certain ethical notions rather than others, developing in partic-
ular a special notion of obligation, and it has some peculiar presuppositions.
In view of these features it is also, I believe, something we should treat with
a special scepticism.32

More speci¬cally, Williams argues that the morality system mistakenly
takes obligations to be central and primary in our conception of ethical

30 Skorupski, this volume.
31 Louden, this volume; Louden™s engagement with Williams™ critique goes back to Louden
32 Williams (1985), p. 6.
10 Alan Thomas

considerations; to be committed to the thesis that obligations cannot ulti-
mately con¬‚ict; to neglect the proper role of moral emotion in the assess-
ment of moral agency; to treat obligations as automatically inescapable
and overriding; to treat all practical necessities as moral obligations; to
ignore the category of necessitated practical verdicts that are not based on
obligations; to deny the grounding of ethical considerations in an agent™s
projects; to contrast voluntariness with mere force; ¬nally, to be committed
to a philosophically ambitious notion of radical voluntariness, connected
to an ethical ideal of purity, “the ideal that human existence can be ulti-
mately just.”33 Louden assesses each of these charges in turn, speci¬cally
as leveled against Kant, described by Williams as “the philosopher who has
given the purest, deepest, and most thorough representation” of “morality,
the peculiar institution.”34
It is noteworthy that, in spite of the capacity for historical scholarship
shown in his work on ancient philosophy and in some aspects of mod-
ern philosophy, notably Descartes, Williams™ critique of Kant very rarely
engages with Kant™s actual texts and seems to aim at a broader target: Kant™s
in¬‚uence on contemporary work on ethics, as opposed to Kant™s views them-
selves.35 That leaves him open to the charge that in various respects Kant
may not turn out to be a Kantian in that sense and that Williams either
targeted a straw man, or misdescribed his real target. The materials for an
assessment of that charge are provided by Louden™s thorough examination
of the respects in which the views that Williams criticized may reasonably
be attributed to Kant.
Unsurprisingly, the verdict is mixed. On some points, such as the claim
that obligations can never ultimately con¬‚ict, Louden simply concedes that
Williams was right to criticize this aspect of Kant™s views, but also to note
that this is an instance of incommensurablity between historical outlooks.
Williams™ arguments and those of Kant do not engage with each other,
Louden implies, as Kant™s worldview simply did not acknowledge the kind
of radical pluralism that Williams takes to be central to the ethical.

33 Williams (1985), p. 195.
34 The title of Williams (1985), chapter 10; I am grateful to Tim Chappell for informing me
that the phrase “the peculiar institution” was the standard euphemism for slavery in the
antebellum South. (Also pointed out by Jenkins [2006], p. 69).
35 I once pointed this out in a conversation with Williams, particularly with regard to chapter
4 of Williams (1985), which deals in very general terms with “the Kantian project,” but
is actually more concerned with the refutation of certain arguments of Gewirth (1977).
Williams replied that chapter 4 of the book was not supposed to be about Kant, but chapter
10 was! For a similar line of concern, see Jenkins, (2006), who notes that “Williams™s critique
of Kantian moral theory appears to be almost totally disengaged from Kant™s texts,” p. 55.

On several other points, Louden follows a single strategy to defend
Kant against Williams™ criticisms: he emphasizes the transcendental status
of Kant™s view, that is, the sense in which it explains how being bound by
a moral reason is so much as possible for moral agents with the particular
and contingent psychological make up that humans in fact have.36 Given
that focus, Louden can represent the idea of placing oneself under a duty as
fundamental to Kant™s arguments, hence giving a sense to the thought that
his ethics is “personal.” It is personal in two ways: in being focused on one™s
duties to oneself and also on the development of one™s moral character in
such a way as to make oneself a reliable instrument of the moral law. This
both limits the scope of blaming others and grounds Kant™s ethics on what
has been called an “autodicy,” an ethically grounded proper self-concern,
in a way that makes Kant™s ethics nonalienating for the individual.37 Those
more sympathetic to Williams might question whether this captures the
correct sense of the “personal” in ethics.38 In particular, it might be ques-
tioned whether a derivative emphasis on moral virtue as work on oneself to
make oneself an effective instrument of morality captures the phenomeno-
logical plausibility of virtue ethics more generally and the role of moral
emotion in the assessment of a virtuous agent and his or her actions.39 Kant
is certainly concerned with the very possibility of altruistic action and with
uncovering structure within apparent contingencies of moral motivation,
but “the personal” seems to ¬gure here as the limiting case of the binding of
people in general, in view of certain very abstract features of their rational
nature.40 That, as I will discuss further later, became central to Williams™
later discussions of his own version of a “critique of morality”: he objected
in particular to its conception of the moral self as “characterless.”
Louden is very well aware that this issue of abstraction versus partic-
ularity returns, directly, to the question of whether or not all reasons are
internal reasons that Skorupski discussed. He is less charitable than Sko-
rupski in claiming that Williams™ ¬nal position on this matter is that “there
is no court of appeal outside of one™s basic desires and projects. One reasons

36 For a complementary emphasis on Kant™s work as focused not on criteria of right action
but on the constitution of the space of deontic modality itself, see Stratton-Lake (2000).
37 I take the helpful term “autodicy” from Cottingham (1981), p. 802.
38 The opposite of “personal” is not “impartial” but “impersonal,” so the question is whether
impartial ethical theories can recognize the reasons or values that we would ordinarily
describe as personal in a nonalienating way; for some arguments and further references,
see Thomas (2005b).
39 Thomas (2005b).
40 To put the point another way, one™s self-concern is the limiting case in which one substitutes
oneself as an instance of impartial moral concern for anyone.
12 Alan Thomas

instrumentally about how best to achieve them, and that is it.”41 That is the
basis for Louden™s further insistence that what Williams means by practical
necessity is, from a Kantian perspective, inevitably “heteronomous” as an
agent™s reasons can be grounded only on prior “projects.” As Louden also
notes, there is at the most abstract level a con¬‚ict between Williams™ com-
mitment to a nonreductively naturalist style of explanation of the ethical,
and Kant™s nonnaturalism, that comes out most clearly in their contrasting
treatments of freedom and voluntariness. This con¬‚ict is closely connected
to another: the con¬‚ict between the ethical ideal of purity and Williams™
view that “the idea of a value that lies beyond all luck is an illusion.”42
Louden concludes that Williams™ scepticism about the morality system, if
restricted to those aspects of that view that de¬nitely are part of the outlook
of the historical Kant, ought to be resisted.
However, it is worth noting that others more sympathetic to Williams™
position, such as Charles Taylor, might respond that Louden™s achievement
is to bracket Williams and Kant together as holding differing views of the
status of a key background evaluative commitment, namely the ideal of
purity “ one in favour, one against “ whereas the ethics of obligatory action
in contemporary ethics that was Williams™ primary concern in his critique of
the morality system simply ignores any such question.43 Feigning a localized
amnesia, those moral theories that Taylor labels “procedural” try and make
do without any appeal to those qualitative distinctions between hypergoods
that Williams tried to capture with his portmanteau term, “importance,”
where what is important to an agent is shaped by those particular evalua-
tive concerns that a given personal project makes salient to that person.44
In identifying the con¬‚ict between the historical Kant and Williams as
concerning the status of this “hypergood” of radical purity, Louden at least
identi¬es them both as addressing an important question, whereas for much
contemporary meta-ethics, Taylor argues, that question never even makes
it onto the agenda.
In the closing stages of his essay, Louden argues that there is a differ-
ent sense in which Williams fails to acknowledge an important impurity in
Kant™s ethics; given the obvious danger of equivocation, let me ¬‚ag up that
Louden is here raising a distinct issue from that concerning the extent to

41 For an equally trenchant criticism of Williams that comes to essentially the same conclusion,
see Millgram (1996); for some countervailing considerations see Thomas (2006), chapter
4, in addition to John Skorupski™s paper in this volume.
42 Williams (1985), p. 196.
43 Taylor (1995).
44 Thomas (2005b).

which an ethical view can represent our ethical experience as maximally free
from luck, which is the point at stake in the argument over the ethical ideal
of purity. In this different sense of “purity” versus “impurity,” the contrast is
much more analogous to that between, say, pure versus applied mathemat-
ics. Louden™s claim is that Kant™s ethics can be signi¬cantly humanized if
one acknowledges the extent to which Kant added a second, “impure” part
to his ethics in order to accommodate the ways in which moral demands
have to be tailored to the “crooked timber” of humanity out of which, he
famously remarked, “no straight thing was ever made.”45 Louden is surely
correct to balance the stereotypical representation of Kant™s ethics with an
account of his anthropological and pedagogical writings.46 However, hav-
ing identi¬ed the issue, let me note that it marks another point of contrast
between the historical Kant, even as sympathetically expounded by Louden,
and Williams speci¬cally on this contrast between identifying a “pure” and
“applied” component in ethics. As a perceptive critic of Williams, Raymond
Geuss has noted one of the key themes of Williams™ later work is that:

There are some ˜universal materials™ out of which particular human ethi-
cal conceptions are constructed (Shame and Necessity, p. 56) and Williams
believes that there is no special problem in claiming that we can know this
or what these materials are. However, he also holds that, contrary to what
Plato and Kant thought, investigation of these universal materials alone will
not throw adequate light on any particular concrete form of ethical thought
because there is no unique path from these materials to any particular human

Williams was equally sceptical, then, of the ambitious plan to identify the
pure and the applied in human ethical conceptions and to take the latter to
be the empirical “¬lling in” of a necessary and universal template identi¬ed
more abstractly; as Geuss also puts it, “the history, sociology and politics of
the case do not simply ¬ll in the details of the picture: they are the picture.”48
Both Skorupski and Louden note the importance to Williams™ critique
of the morality system of its neglect of moral emotion and the obstacles
that it places in the way of a proper understanding of guilt and shame.
Those key ethically relevant emotions are examined by Michael Stocker
in his contribution to this volume, in which Williams™ analyses of these

45 “Out of the crooked timber of humanity nothing straight can be made” was Isaiah Berlin™s
translation of a sentence in Kant (1784), sixth thesis.
46 See, for example, Louden (2000).
47 Geuss (2005), p. 232.
48 Geuss (2005), ibid.
14 Alan Thomas

emotions in Shame and Necessity are the basis for his own exploration of this
complex topic. Stocker basically agrees with Williams that, as Stocker puts
it, “shame is like moral luck in showing that the range of the ethical and
the evaluative goes well beyond guilt”; however, there are various respects
in which Stocker both builds on Williams™ account while also offering sev-
eral correctives to it. In particular, guilt experienced without shame can
only be a mistaken or pathological form of moral emotion, involving var-
ious forms of dissociation or failure of integration; Stocker claims this is
usually true of shame without guilt. There are, however, forms of shame
experienced without guilt that are nonpathological, particularly those asso-
ciated with the conjunction of the beautiful with the good in the Greek
ideal of kalokagathia, that Stocker discusses in detail.49 He adds an impor-
tant discussion of identi¬catory shame, based on those values with which
a person is identi¬ed, and shame without responsibility. Stocker further
argues that the criteria that Williams and other theorists have offered in
order to demarcate shame and guilt cannot be understood as a set of nec-
essary and suf¬cient conditions; he suggests ways of conceptualizing the
differences between shame and guilt, alongside an acknowledgment that
in most actual contexts the two moral emotions are strongly interrelated.
Stocker also argues against what he takes to be the in¬‚ated claim that shame
involves a diminution of one™s whole being, as Williams at one point claims.
Overall, however, in a penetrating discussion in¬‚uenced by psychoanalytic
and psychological as well as philosophical discussion of moral emotions,
Stocker endorses Williams™ claim that “shame can understand guilt, but
guilt cannot understand itself.”50
The issues dealt with by Stocker in ahistorical terms receive their most
extended treatment in Williams™ work in an historical approach to the moral
psychology represented in parts of the culture of ancient Greece. Shame and
Necessity, a book that originated in the Sather lectures at the University of
California, Berkeley, is comprehensively assessed in Tony Long™s paper in


. 3
( 38 .)