-
This article was downloaded by: [FU Berlin]On: 24 June 2015, At:
13:22Publisher: RoutledgeInforma Ltd Registered in England and
Wales Registered Number: 1072954Registered office: Mortimer House,
37-41 Mortimer Street, London W1T3JH, UK
Click for updates
Inquiry: An InterdisciplinaryJournal of PhilosophyPublication
details, including instructions forauthors and subscription
information:http://www.tandfonline.com/loi/sinq20
Wittgenstein, Korsgaard andthe Publicity of ReasonsJoshua Gertaa
Department of Philosophy, College of William andMary, Williamsburg,
Virginia, USA.Published online: 18 Jun 2013.
To cite this article: Joshua Gert (2015) Wittgenstein, Korsgaard
and the Publicityof Reasons, Inquiry: An Interdisciplinary Journal
of Philosophy, 58:5, 439-459, DOI:10.1080/0020174X.2013.776297
To link to this article:
http://dx.doi.org/10.1080/0020174X.2013.776297
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy
of allthe information (the Content) contained in the publications
on ourplatform. However, Taylor & Francis, our agents, and our
licensorsmake no representations or warranties whatsoever as to the
accuracy,completeness, or suitability for any purpose of the
Content. Any opinionsand views expressed in this publication are
the opinions and views ofthe authors, and are not the views of or
endorsed by Taylor & Francis.The accuracy of the Content should
not be relied upon and should beindependently verified with primary
sources of information. Taylor andFrancis shall not be liable for
any losses, actions, claims, proceedings,
-
demands, costs, expenses, damages, and other liabilities
whatsoeveror howsoever caused arising directly or indirectly in
connection with, inrelation to or arising out of the use of the
Content.
This article may be used for research, teaching, and private
studypurposes. Any substantial or systematic reproduction,
redistribution,reselling, loan, sub-licensing, systematic supply,
or distribution in any formto anyone is expressly forbidden. Terms
& Conditions of access and use canbe found at
http://www.tandfonline.com/page/terms-and-conditions
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
Inquiry, 2015http://dx.doi.org/10.1080/0020174X.2013.776297
Wittgenstein, Korsgaard and thePublicity of Reasons
JOSHUA GERT
Department of Philosophy, College of William and Mary,
Williamsburg, Virginia, USA
(Received 13 December 2012)
ABSTRACT In The Sources of Normativity (Korsgaard, Christine.
The Sources ofNormativity. Cambridge: Cambridge University Press,
1996), Christine Korsgaard triedto argue against what she called
the privacy of reasons, appealing to Wittgensteinsargument against
the possibility of a private language. In recent work she
continuesto endorse Wittgensteins perspective on the normativity of
meaning, although she nowemphasizes that her own argument was only
meant to be analogous to the private lan-guage argument. The
purpose of the present paper is to show that the
Wittgensteinianperspective is not only not useful in support of
Korsgaards general project, but that itis positively inimical to
it, in two ways. First, Wittgenstein opposes views on which
prin-cipled or rule-following behavior requires that one be guided
by anything like a mentalrepresentation of a rule or principle. But
for Korsgaard, human action essentially requiresthis. Second,
Wittgenstein systematically attempts to de-emphasize the importance
ofthe first-personal perspective, and to emphasize the social
functions even of concepts thatmight seem deployed primarily from
that perspective: for example, concepts of sensa-tions and
intentions. This is the reverse of Korsgaards emphasis. The paper
also argues,however, that the private language argument does have
some implications for a theory ofrationality and reasons.
I. Introduction
Something over fifteen years ago, Christine Korsgaard tried to
argue againstwhat she called the privacy of reasons, appealing to
an interpretation ofWittgensteins argument against the possibility
of a private language.1 This
Correspondence Address: Joshua Gert, Department of Philosophy,
College of William and Mary,PO Box 8795, Williamsburg, VA
23187-8795, USA. Email: [email protected]
1Korsgaard, Sources of Normativity, 13145.
2013 Taylor & Francis
Vol. 58, No. 5, 439459,
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
2 Joshua Gert
argument of Korsgaards sparked a good deal of commentary.2 I,
amongothers, argued that she had confused Wittgensteins sense of
privacy, whichis close to incommunicability, with a very distinct
sense, which might moreclearly have been called agent-relativity.3
In particular, Wittgenstein mightbe useful in showing that if I
have a normative reason, it must be possibleto explain, in public
terms, what would count as acting against that reason.But no
argument analogous to Wittgensteins argument about the public-ity
of meaning could possibly show that the reason might not be
somethingagent-relative, such as the fact that the action will give
me, the agent, a lot ofpleasure.4
In recent work Korsgaard has expressed the view that some of her
criticswrongly took her to be suggesting that the publicity of
meaning somehowdirectly entailed the publicity of reasons.5 But
that was not, and is not, mycriticism. I took her point to be that
an argument analogous to the privatelanguage argument could be
mounted in the case of normative reasons, invirtue of the
normativity of both reasons and meanings. I still think that
thiscannot be done. But it is not the point of this paper to repeat
those argu-ments. Rather, its purpose is to show that the
Wittgensteinian perspectiveis not only not useful in support of
Korsgaards general project, but thatit is positively inimical to
that project, and in two ways. First, Wittgensteinopposes views on
which principled or rule-following behavior requires thatone be
guided by anything like a mental representation of the rule or
princi-ple one is following. But for Korsgaard, human action
essentially requires this.Second,Wittgenstein systematically
attempts to de-emphasize the importanceof the first-personal
perspective, and to emphasize the social functions even ofconcepts
that might seem deployed primarily from that perspective: for
exam-ple, concepts of sensations and intentions. This is the
reverse of Korsgaardsemphasis. Despite all this, however, the final
substantive section of this paperargues that there is a connection
between the private language argument anda correct account of
practical reasons, and offers a sketch of this connectionand of the
sort of view it supports. But first it is necessary to give a
briefpresentation of the relevant parts of Korsgaards view.
2Gert, Korsgaards Argument; Nagel, View from Nowhere; Coleman,
Public Reasons; Hurley,Kantian Rationale; LeBar, Korsgaard,
Wittgenstein; Wallace, Publicity of Reasons.3Recently, however, Jay
Wallace has helpfully made the following terminological
suggestion.Publicity can be taken to be the thesis that any
consideration that is a reason for me must also bea reason for you.
Agent-neutrality adds to publicity the further thesis that this
shared reason sup-ports our performing the same kind of action: for
example, alleviating my pain. Wallace himselffocuses on publicity,
not agent-neutrality. Wallace, Publicity of Reasons, 481.4This same
error seems to appear in Korsgaards discussion of a potential
Wittgensteinian worryabout her account of instrumental reason.
Korsgaard, Constitution of Agency, 59.5Korsgaard,
Self-Constitution, 196.
440
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
Wittgenstein, Korsgaard and the Publicity of Reasons 3
II. Korsgaards View
One way to explain Korsgaards view of human action is to present
it asan account of the ways in which such action differs from the
action ofnon-human animals. On her view, animals are in some sense
bundles ofembodied instincts. When they represent the world in
certain ways, that isenough to prompt them to respond in certain
corresponding ways. For a frog,characteristic shapes moving in
characteristic ways are represented as food,and they prompt
getting-and-eating behavior. The patterns of response hereare the
principles of the frogs nature. We human beings, on the other
hand,are aware of our tendencies to act on the basis of certain
perceptions, andour reflective awareness of these tendencies in
particular cases opens up whatKorsgaard calls a reflective
distance, which we need to cross before we act.Here is another way
of describing the same view. Our nature is to see certain
thingsKorsgaard calls them incentivesas potential grounds for
action.Our awareness of such factsfor example of the fact that
there is anotherserving of fried rice in the pan, and that it would
be nice to eat itis aware-ness not only of the food but also of
what Korsgaard calls an inclination toeat it. This inclination is
the object of our reflective awareness. By itself it can-not move
us to act. Rather, it presents itself as something like a proposal.
Andwhen reason says yes to this proposal, we act. These
endorsements of incli-nations do not simply fire off at random,
however: they are like judgmentsthat the action would be worth it.
And it does not really make sense for some-thing to be worth it on
one occasion, but not on another. Korsgaard doesnot regard this
requirement of consistency as mandated by a need for onesjudgments
to reflect some kind of external, objective, normative truth,
how-ever. Rather, the explanation of the requirement is that if we
are to conceiveourselves as agents, as having effects that are
distinctively our own, we have toadopt principles that will allow
this. These are principles by which we consti-tute ourselves as
agents and give ourselves a unified identity.
Contradictory,conflicting or self-defeating principles do not allow
it.How does all of this yield a view of normative reasons?
Korsgaards view
is that one does not count as acting unless one is an agent, and
that to be anagent, one needs to be unified by ones practical
principlesthe set of whichconstitute ones practical identity. That
means that a constitutive norm ofaction is that one chooses
principles that do in fact unify one as an agent.One can do this
well, or poorly, and that is the difference between good andbad
action.6 If one violates the constitutive norms too severely, one
is not evenacting at all: rather, one is being overmastered by a
fear or a compulsion. Butas long as one is coming close enough, one
is acting. And in that case, theobject of the inclination can be
cited as ones reason, since it is that object
6For present purposes we need not differentiate between morally
good and bad action andprudentially good and bad action.
441
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
4 Joshua Gert
that makes the principle of ones action seem eligible. When the
principle is infact in line with the constitutive normwhen it does
in fact unify one as anagentthen the reason is a good one.In order
to make her view clearer, Korsgaard frequently appeals to
Platos
analogy between a person and a polis. If we are to attribute an
action to thepolis, rather than to a rogue soldier or a despotic
ruler, the whole polis must beable to see itself as playing its
appropriate role in the generation of the effect.In the case of the
polis, this means that the workers propose something, therulers
decide whether it would be worth it, and if they decide it is, then
theauxiliaries carry it out. In the case of human action, it is
desires that makethe proposalswhich appear to reflective
consciousness in the form of incli-nations, which reason can
consider. If reason decides that such an inclinationis worth acting
on, spirit implements it, and weunified agentsact.
III. General Wittgensteinian Worries
In his discussion of rule-following, before the private-language
argument getsgoing, Wittgenstein is very clear that, in trying to
get clear on our capacityto follow rules, we cannot generalize from
the special case in which we havedoubts about the implications of a
rule and pause to work it out via someinterpretation. After all, in
the very act of working it out, we will be followingstill other
rules, and if we needed to work out the implications of those
rulesbefore applying them, we would never get anywhere. As a
result, he concludesthat there is a way of grasping a rule which is
not an interpretation, but whichis exhibited in what we call
obeying the rule and going against it in actualcases.7 Importantly,
the we here are those of us who may be observing therule-following
of other people. I come back to the importance of the
third-personal perspective in the next section.One of the errors
Wittgenstein is trying to help us avoid is an unwarranted
move from a contingent fact that happens, typically, to be
instantiated whenone is theorizing (e.g. because one is
introspecting or because one is con-sidering what one takes to be
somehow the most characteristic cases) to afalse theoretical claim
that that (contingent) fact is a necessary one.8 Let us
7Wittgenstein, Philosophical Investigations, 201. See also
Wittgenstein, Blue and Brown Books,34. Though they do not
mentionWittgenstein, this same point about regress and automatic
pro-cesses is echoed very clearly in Arpaly and Schroeder, with the
idea of rule-following replaced bythe idea of acting for reasons,
and the idea of interpretation replaced by the idea of
deliberation:acts of deliberation are performed for reasons (good
or bad) in virtue of something other thantheir relation to
deliberation. Arpaly and Schroeder, Deliberation, 223.8Compare the
following claim of Korsgaards, made in support of her general view
that self-consciousness requires us to unify ourselves by endorsing
the principles on which we act:The phenomenology of deliberation,
especially in hard cases, bears this out. Korsgaard,
Self-Constitution, 126 (my emphasis). But hard cases of the sort
she is considering are very rare in thelife of a human being, and
though they do highlight certain distinctive human capacities,
they
442
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
Wittgenstein, Korsgaard and the Publicity of Reasons 5
call this sort of error the philosophical fallacy, because it is
an occupa-tional hazard of philosophizing that, by engaging in that
very activity, oneoften creates this kind of distorting context.9 I
suspect that the philosophi-cal fallacy is behind the following
claim of Korsgaards: you do not think ofyourself, or experience
yourself, as being impelled into action, but rather asdeciding what
to do.10 It is true that when one pauses to observe ones action,to
see what happens when one acts, it is likely enough that one will
observea decision, or at least some reflection on whether a certain
impulse seemsworth following. But the vastly greater part of a
human lifeincluding thedistinctively human actions it contains,
such as answering a question, tryingto remember something,
following the rules of the roadare not like this.11
Korsgaards insertion of a decision into every action can be seen
as a philo-sophical response to what Wittgenstein calls an
architectural requirement.The hypothesized decision is an
ornamental coping that supports nothing.12
Its function is to allow for a picture of human actionalas, a
false onethatis more satisfying and less vertigo-inducing than one
on which our ability tothink and act relies, at bottom, on
automatic and unconscious processes overwhich we have virtually no
control.One part of Wittgensteins argument about rule-following
obviously
appeals to worries about a regress. As he puts it, at the
conclusion of his attackon a mental process conception of
rule-following:
it can be seen that there is a misunderstanding here from the
mere factthat in the course of our argument we give one
interpretation afteranother; as if each one contented us at least
for a moment, until wethought of yet another standing behind
it.13
If Korsgaards argument has the same form, we should expect to
find the samekind of problem. And we do. The endorsement of an
impulsesaying yes toan inclination and thereby adopting it into the
maxim on which one actsis itself an act and needs to be explained
by reference to some antecedentinclination and its endorsement. But
then we need some form of endorsementof the inclination to
endorse.
should not be taken as simply clearer instances of typical human
action. Typical human actionneed not involve any deliberation at
alllet alone difficult deliberation.9I discuss this fallacy in
another context in Gert, Brute Rationality, 4367.10Korsgaard,
Constitution of Agency, 109.11Cf. At the very moment at which
Harold consciously tries to call to mind competing promisesthat
would prevent him from promising to meet his son in Calgary, he is
not (if he is anythinglike us) consciously evaluating the
reasonableness of this conscious search of his memory. Arpalyand
Schroeder, Deliberation, 223.12Wittgenstein,Philosophical
Investigations, 217. Cf.Wittgenstein,Blue and Brown Books,
1718,Zettel, 31315.13Wittgenstein, Philosophical Investigations,
201.
443
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
6 Joshua Gert
One might try to interpret Korsgaard in such a way that
endorsement isnot itself an action, and is not chosen. For example,
one might take endorse-ment to be unreflective and automatic in the
Wittgensteinian sense in whichmost of our behavior, including the
following of rules, is unreflective and auto-matic. But this is
very far from Korsgaards intent. If we simply
respondedautomatically to the goodness of acting on a certain
inclination by acting onit, we would not be acting in a
distinctively human way, but would be likeanimals, whose behavior
is to be explained in terms of certain constitutiveprinciples that
give them their ends.14
Korsgaards appeals to Plato and to the analogy between a just
state and ajust person paint a vivid picture (the kind of picture
Wittgenstein constantlywarned us against) in which automatic
responses have no place. On such apicture, to understand how it is
that a human being can perform a certainkind of action (e.g.
applying a mental picture of a cube in order to identifysomething
as a cube), it becomes necessary to postulate that the very
samekind of action is performed (e.g. applying a picture that
represents the correctmethod of application of the picture of the
cube). In Platos case, the priorinstance of action that helps
explain the action of the state is the action of
therulersthemselves creatures with the same sort of structure as
the state. Thekind of homunculus-theoretic view of action that
Plato is offering, and thatwas offered by sense-data theorists in
the case of perception, is the result oftrying to avoid a regress
by locating the second instance of the regress-starter(an action or
a perception) in a smaller and more internal entity. But, as
criticsof the sense-data theory have successfully shown, the
regress still exists.15
Korsgaard does say some things that seem to be intended to avoid
the prob-lematic regress. They all have the following tendency: to
deny a distinctionbetween the endorsement of acting on a certain
inclination, and acting onthat inclination. Here is a pair of
representative claims.
[T]he judgment that the action is good is not a mental state
that precedesthe action and causes it. Rather, [the agents]
judgment, his practicalthinking, is embodied in the action itself.
Thats what it means to saythat the action is motivated and not
merely caused.
14Korsgaard, Constitution of Agency, 226.15Cf. Blackburn, Ruling
Passions, 2506. Interestingly, Korsgaard herself criticizes
dogmaticrationalist realists about values or reasons for having a
homuncular view. But although it is truethat the faculty of reason
on such views is, in a sense, an internal representative of an
externalstandard of reason, these views are not homuncular in the
sense of having a little internal agentwhose presence explains our
capacity to act, or a little internal perceiver whose presence
allowsus to perceive. Platos view is homuncular in this sense, as
Korsgaard presents it. And so, I think,is Korsgaards. Korsgaard,
Self-Constitution, 6.
444
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
Wittgenstein, Korsgaard and the Publicity of Reasons 7
[B]eing motivated by a reason is not a reaction to the judgment
that acertain way of acting is good. It is more like an
announcement that acertain way of acting is good.16
I think it is true that if we simply describe stopping at a red
light as anannouncement that doing so is good, or as an embodied
endorsement of theimpulse to stop, then the regress is avoided. But
the problem is then that wehave distorted the notion of an
announcement or an endorsement to such adegree that the rest of
Korsgaards theory no longer applies. On her theoryit is important
that we choose the principles on which we actin the normalsense of
choose. That is what differentiates us from the animals, who
alsohave principlesbut whose principles are given to them by
nature.17 Here is afairly clear statement of this difference:
What would have been the cause of our belief or action, had we
stillbeen operating under the control of instinctive or learned
responses,now becomes something experienced as a consideration in
favor of acertain belief or action instead, one we can endorse or
reject.18
This distinction between animal and human action simply would
not makesense if we identified the endorsement of the consideration
with the per-formance of the action that it favors. In human action
there is somethingextra: the endorsement of ones principle of
action. It is only because of thisendorsement that the principle
then yields action in something like the way inwhich an animals
action flows from the instincts and learned responses
thatconstitute its nature. One might try to defend Korsgaard by
claiming that sheis only trying to characterize a special subset of
human actions: those that areself-conscious or deliberated. The
rest are in all essentials the same as animalaction. But this
strategy for avoiding regress would require that even deliber-ated
action bottom out in animal action. And this seems far from the
spiritof Korsgaards view. Moreover, it is hard to see how we could
be responsiblefor most of our actions, if they typically resemble
the instinctive or learnedresponses of animals.19
Korsgaard might dispute my presentation of this last point. In
the samepaper from which the above passage is taken, she also
writes the following:
16Korsgaard, Constitution of Agency, 2279. See also Korsgaard,
Self-Constitution, 127.17Korsgaard, Self-Constitution,
108.18Korsgaard, Activity of Reason, 32.19This seems true even on
Korsgaards attractive view of responsibility, which takes
responsibilityfor omissions as the general model for moral
responsibility for bad action. For, unless our actionsinclude some
non-trivial reflective distance, it is hard to see how it would be
legitimate to hold usresponsible for failure to pick up the reins
and take control of [our] own movements.
Korsgaard,Self-Constitution, 175.
445
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
8 Joshua Gert
the picture I have in mind is not that there is a two-step
process: stepone, you first choose some way of identifying
yourself, and step two, youproceed to act in accordance with its
principles, like someone followinga list of rules.20
But what Korsgaard is really denying here is that action is a
three-step pro-cess: (1) choosing a way of identifying yourself;
(2) looking through the listof associated rules; and (3) acting on
the inclination that figures in the rele-vant rule. The two steps
she wants to collapse are (1) and (2). That is, as sheputs it,
adopting maxims or practical principlesis at the same time
engag-ing in the work of identity construction.21 I am happy to
collapse these twosteps. But it remains true that the act of
endorsement is in an important senseprior to the action that flows
from the endorsed principle. And it is prior toeach action; one
cannot choose once and for all, since the removal of anactual
endorsement from ones subsequent actions would close up the
reflec-tive distance that differentiates human action from the
learned behavior ofanimals.
IV. Wittgenstein and the First-Personal Point of View
Despite my initial criticisms of Korsgaards appeal to
Wittgenstein, and mycurrent worries about the role that reflective
endorsement could possibly playin an account of reasons, I have
come to agree with Korsgaard that there is arelation between the
private language argument and a correct view of practicalreasons. I
present my view of that relation in the next section. In this
section,however, I argue that an appreciation of one of the main
points of the privatelanguage argument undermines an important
starting-point of Korsgaardstheorizing: reflection from the
first-person point of view.One of the things the private language
argument highlights is that language
is a practice. If our use of words were not rule-governed, it
would not countas a practice, and the words themselves would not
count as meaningful. Theuse of a word being rule-governed in the
relevant sense amounts to there beinga distinction between
acceptable and unacceptable uses of that word. This inturn requires
that there be publically observable criteria that enable people
toteach the meaning of that word, and to test whether someone has
masteredthat meaning. In the case of a private language, we lack
the possibility of anindependent way to test for mastery of its
words. This means we lack a realdistinction between acceptable and
unacceptable use and have no reason toregard the words of such a
language as words at all. It is worth emphasizinghere that none of
these points makes trouble for the language of sensation or
20Korsgaard, Activity of Reason, 36.21Ibid., 367.
446
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
Wittgenstein, Korsgaard and the Publicity of Reasons 9
other psychological terms with which we describe our
experiences. There areindependent ways of testing for mastery of
such locutions as I feel a tickle onmy arm, though I know nothing
is touching it or the water seems warm tome, though I know that to
everyone else it seems cold. Although there maybe no way to test
for appropriate use in a particular case, we certainly do haveways
of knowing that someone has mastered the relevant general
techniquefor making such claims. Nor does the view of language as a
practice rule outthe possibility of a language being developed by a
solitary person on a desertisland. It only means that it must be
possible to explain, in public terms, whatit means to say that this
person has mastered this or that word in his solitarylanguage. One
way of putting this is to say that it must be in principle
possiblefor an anthropologist to come to learn this language.All of
the above points may seem relatively trivial to many
contemporary
philosophers, when they are made explicit. But there is still a
persistent ten-dency to forget them and to think of language as a
sort of vehicle for solitarythought and reflection. This is the
real point of contact between the privatelanguage argument and
criticisms of overly individualistic accounts of practi-cal
reasons. Private language can seem more plausible if one theorizes
aboutlanguage by imagining a solitary being, thinking to himself
about the worldin which he finds himself, wondering, perhaps,
whether there is an objectiveworld out there causing the patterns
in his stream of private experience:a stream he describes to
himself with words in a private language. And ifone theorizes about
practical reasons by imagining a similarly solitary being,thinking
to himself about what he ought to do, it can seem that the pri-mary
use of the concept of a practical reason is to figure in
first-personaldeliberation. And this does seem to me to be the way
Korsgaard often is the-orizing. Consider her claims that the
normative force of reasons, obligations,and values, is force that
is felt by a deliberative agent and is imperceptiblefrom outside of
the deliberative perspective and that obligations exist in
thefirst-person perspective.22 Korsgaard does not entirely
disregard the third-personal perspective, but it is clear that for
her that perspective is theoreticallyless important. As she puts
it:
It is certainly true that from a third-personal point of view,
when we callpeople vicious or irrational, we mean that they fail to
conform to cer-tain standards. But that failure is the outward
manifestation of an innercondition, and these theories do not tell
us what that inner conditionis.23
22Korsgaard, Sources of Normativity, 124, n. 39, and 257. See
also Korsgaard, Self-Constitution,104, 126. It is true that
Korsgaard emphasizes the phenomenon of our reasoning together.
Butthis is still first-personal; it is just first-person
plural.23Korsgaard, Self-Constitution, 67. Nor is she alone. Jay
Wallace, despite his doubts aboutKorsgaards appeal to Wittgenstein,
endorses what he calls the priority of deliberative judg-ment. This
thesis holds that the claims we endorse in deliberation . . .
should be taken as
447
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
10 Joshua Gert
This last remark should call to mind Wittgensteins constant
battle againstthose who take introspection to be the best source of
insight into the natureof other normative notions, like those of
meaning and understanding, and hisadmonition that An inner process
stands in need of outward criteria.24
First-personal theorizing about practical reason is even more
prevalentthan first-personal theorizing about language, perhaps
because it is obviousthat at least one important function of
language is to communicate informa-tion between people. But in both
cases, the first-person perspective restrictsour view to cases in
which one actually is acting on a reason, or actually isusing a
word. And this makes it much easier to conceive of the
normativityof the reason or the meaning as a kind of psychological
forcethough, obvi-ously, a queer one, since it must be present even
in cases in which it is totallyineffective because the agent is
irrational, or has misunderstood the mean-ing of the word.25 This
may help explain Korsgaards puzzling claim that thenormativity of
obligation is, among other things, a psychological force andher
talk about its manifestation as a psychological force.26 The
highlightedphrases indicate that normativity is more than, or that
it somehow standsbehind, certain psychological forces. This
hesitation to embrace a simpleidentity between normativity and
something psychological is understandable,since the normativity of
a reason or an obligation does not disappear, evenwhen it is
completely unrecognized and exerts no psychological force at
all.But it is difficult to see how something that is not itself
psychological canman-ifest as a psychological force. And it is
equally difficult to understand how thefact that normativity ismore
than just a psychological force can help deal withthe problems that
arise if one says that it is at least a psychological force.
authoritative characterizations of the nature of our reasons for
action. Wallace, Publicity ofReasons, 484. Of course I do not think
we should ignore these first-personal judgments. But Ido think that
if a priority thesis of relevance to proper methodology is in the
offingand that iswhat Wallaces thesis isthen third-personal
judgments are going to be more illuminating of thenature of
reasons.24Wittgenstein, Philosophical Investigations, 580. See also
Wittgenstein, Zettel, 469. Regardingthe temptation to think of
understanding as an introspectable mental process, and the idea
that theprocess of remembering is visible from the first-person
perspective, seeWittgenstein, PhilosophicalInvestigations, 1534 and
305, respectively.25Cf.: Then what sort of mistake did I make; was
it what we should like to express by saying: Ishould have thought
the picture forced a particular use on me? How could I think that?
Whatdid I think? Is there such a thing as a picture, or something
like a picture, that forces a partic-ular application on us; so
that my mistake lay in confusing one picture with another?For
wemight also be inclined to express ourselves like this: we are at
most under a psychological, nota logical, compulsion. And now it
looks quite as if we knew of two kinds of case.
Wittgenstein,Philosophical Investigations, 140; see also
1956.26Korsgaard, Self-Constitution, 34 (my emphasis). The
conflation of normativity with theparticular psychological force of
motivation has been noted before, most effectively in
Parfit,Normativity. See, for example, pp. 33643.
448
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
Wittgenstein, Korsgaard and the Publicity of Reasons 11
Here is a point that might help break the spell of the
first-personal whentheorizing about practical reason. The words
reason, rational, ought,and their everyday cognates, are words in
public languages. But all publiclanguages are social practices,
performing largely social functions, and thewords they contain are
tools or pieces that we use when participating in thatpractice.
This very general point supports the idea that the first place to
lookfor an explanation of the emergence of a particular vocabulary
in a publiclanguage is in the social tasks that vocabulary allows
us to perform. In order tosee this point more concretely, consider
the actual law: not the self-legislationthat figures in Kantian
moral theorizing, but the sort of law that is createdby politicians
and written in law books. The function of such law is
obviouslysocial. The need for this sort of lawwould not arise for a
solitary human being.This does not prevent the law from being
normative, and from giving us legalobligations.Whether or not I
happen to havemade the law, and whether or notI happen to endorse
it, I still am bound by it in an important sense. PerhapsI am not
morally bound by it. Perhaps, in many cases it would not be
irra-tional to disobey it. And perhaps, in some odd instances, it
might actually beirrational to obey it. But those are separate
issues. Rather, the important pointis that in virtue of my
membership in a community, certain laws apply to me,and if I
violate them I have acted illegally. There is no temptation to
locate thesource of legal obligation from within a first-personal
perspective. And partof the reason for this lack of temptation is
the manifestly social purposes thatthe law serves.Now, in the case
of morality, the general Wittgensteinian emphasis on the
social nature of language and on the concepts that our
linguistic practicesallow us to master suggests a fairly strong
analogy with the legal. Why, that is,do we have moral concepts:
concepts of moral obligation, moral excuse, andso on? Very
plausibly, we have these concepts for many of the same reasonsas
those that explain the emergence of the law. Indeed, morality is
easily andplausibly viewed as an informal legal system. True, it is
not written down inauthoritative form. But the difference here need
not be much greater thanthe difference between an oral tradition
and a written canon. It is true thatwe sometimes use moral concepts
from a first-personal perspective: conclud-ing that a certain
action would be morally wrong may well prevent us fromperforming
it. But of course the very same thing can be said about the
law.Whether it functions in this way depends on how much respect
the relevantfirst-person happens to have for morality, or for the
law. Some may object thatthis story cannot account for the truth of
the idea that we always have suffi-cient reason to behave as
morality requires. But in fact it will turn out thatthis charge is
false. Inasmuch as morality is a public system that we encour-age
people to follow, the account of rationality and reasons I offer
below willentail that moral requirements are always backed by
reasons that are sufficientto render them rationally
permissible.
449
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
12 Joshua Gert
It is relatively easy to maintain the Wittgensteinian
third-personalperspective on the law and on morality.27 But when it
comes to the widerdomain of practical rationality and general
practical reasons, it very easy tolose that perspective. One reason
for this is that morality is something likean enforceable system,
and this makes its social nature manifest. There is,very plausibly,
a conceptual connection between immoral behavior and liabil-ity to
punishment.28 But practical rationality is not like this. When
someonebehaves irrationally, this does not call for punishment.
Moreover, a personcould behave irrationally while alone on a desert
island, butat least onmany plausible moral viewssuch a person could
not behave immorally. Still,the Wittgensteinian perspective
encourages us to look for the social purposesserved by all kinds of
talk: talk about colors, desires, numbers, streets,
lengths,individual people, and so on. OnWittgensteins view, the
words we use to talkabout these things are best viewed a tools for
getting practical tasks done in acommunity of language speakers.
Color talk exists not because we needforour own private purposesto
categorize objects as red, green, blue andyellow.29 Rather, the
emergence of these words allows us to ask for certainthings,
respond to certain requests, issue certain warnings, and so on.What
practical tasks might be facilitated by the introduction of terms
that
mean irrational and reason into a community of language speakers
whodid not initially have them? I address this question in the next
section. Fornow I just want to stress that, in offering our
descriptions of such a com-munity before the introduction of these
terms into their vocabulary, nothingprevents us from using the same
vocabulary. Just as we can use terms such assick or male or female
to describe sheep and cowsbeings that do nothave any vocabularywe
can describe the people in our thought experimentsas rational or
irrational even prior to our imagining that they themselves
arecapable of similar descriptions.30 I hope it is clear that the
introduction of thevocabulary of rationality and reasons will not
make the members of this com-munity act significantly more or less
rationally. If someone is suffering from amental illness, or is in
a temporarily overwhelming rage, the mere availabilityof the
concepts of rationality and reasons is exceedingly unlikely to act
as acure or a balm.31 People are typically motivated by
considerations that can be
27Although Wittgenstein himself spectacularly failed to do this.
I do not know what to sayabout this except that this lecture was
delivered originally in 1929, well before his mature
phase.Wittgentstein, Lecture on Ethics.28Mill, Utilitarianism;
Gert,Morality.29Indeed, many languages do not have these particular
categories.30At least this is true if rational and irrational
action do not require the deployment of conceptsof rational and
irrational action. But if one appreciates the primarily
third-personal nature ofthese concepts, it becomes very implausible
that this could be required.31Of course in some limited cases it
might. The availability of a new vocabulary allows for newforms of
motivation: no one can be motivated by the thought this would be
irrational withoutpossessing that concept. But let us leave these
cases to the side.
450
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
Wittgenstein, Korsgaard and the Publicity of Reasons 13
conceptualized without the aid of any normative vocabulary: by
opportuni-ties to get food, affection, sleep, knowledge or relief
of discomfort. This is notto deny that human societies involve
complex and contingent social norms,and that internalization of
these norms also explains a lot of action. But anorm has been
internalized precisely when the objects of that norm begin
toprovide direct motivation, just as thoughts about food or sleep
provide directmotivation without the need for socialization. An
intermediate conclusionat this point is the following: the
introduction of the vocabulary of reasonsand rationality is not
plausibly seen as essential to the production of rationalaction.I
hope that the preceding conclusion seems plausible. Its
plausibility should
help to diminish the attraction of a widespread conception of
the natureof reasons and rationality, according to which reasons
motivate those whohave them partly in virtue of being perceived as
normative. Korsgaard herselfendorses this view, which she puts in
the following way: people are inspired todo things by the
normativity of the reasons they have for doing them, by
theirawareness that some consideration makes a claim on them.32
This thesis caneasily pass as a truism, and I think Korsgaard takes
it to be fairly straightfor-ward. But on reflection it should seem
extremely difficult to understand.Whatis this property of
normativity that a reason has, perception of which is sup-posed to
motivate us? The clearest proposal I have seenand I am
neitherendorsing nor disputing itis that a concept is normative if
its instantia-tion entails the existence of reasons.33 Obviously,
this is not much help inexplaining the normativity of reasons
themselves, which is the issue here.Korsgaard sometimes describes
the normativity of reasons in a metaphori-cal way, as that the
reason calls out to [the agent] that a certain action isto be
done.34 This description should bring to mind Wittgensteins
descrip-tion of the tempting picture of rule-following, on which a
rule forces itselfon the rule-follower in some mysterious way.
Indeed, the mysterious forcethat Wittgenstein is describing is
aptly described as the normativity of therule. And Wittgenstein is
tryingas I amto dislodge the temptation toget clear on this
normativity by viewing it from the first-personal perspec-tive.
Korsgaard takes the picture of a reason calling out to an agent
tojustify the following methodological conclusion: we should
identify as rea-sons the kinds of items that first-person
deliberators take to be reasons, thekind of items that play a role
in deliberation.35 To the degree that one agrees
32Korsgaard, Constitution of Agency, 209 (my emphasis). See also
pp. 21415.33Skorupski, Irrealist Cognitivism. Korsgaard sometimes
seems to endorse this conception ofnormativity. See Korsgaard,
Constitution of Agency, 106.34Korsgaard, Constitution of Agency,
210.35Ibid. Korsgaard even seems to attribute normativity in this
sense to the representations thatanimals use in navigating the
world, explaining instinctive action as involving a sense that
a
451
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
14 Joshua Gert
with the Wittgensteinian view of language as a social practice,
one should besuspicious of Korsgaards methodology.I have been
comparing Korsgaards theorizing about normative reasons
with the interlocutors theorizing about the way that we manage
to followrules. One advantage of making this comparison is that
those who have feltthe strong intuitive pull of the interlocutors
position, but who also have man-aged to break free of it, may not
yet be immune to the temptations at workin making Korsgaards
position as intuitively plausible and attractive as it is.And I
should emphasize that it is attractive. Indeed, it is seductive.
Once oneis willing to visualize the soul as Korsgaard does (aided
by lucid interpreta-tions of Plato and Kant), all sorts of
explanations of all sorts of phenomenaseem to followparticularly
with regard to the phenomenology of reflectiveaction. So it is
important to see that the same is true of the picture of
rulesoffered by Wittgensteins interlocutor: the picture of rules as
rails, of mean-ings as entities that one can grasp, of
understanding and meaning as mentalprocesses, of the meanings of
words as parts of the meanings of sentences, andso on. The stories
these pictures suggest seem much more plausible, on theirsurface,
than the story Wittgenstein himself offers. These pictures are so
per-suasive because the metaphors they embody are so simple, clear
and resonantthat they have found a useful place in the language.
Moreover, the resonanceof these metaphors is often due to their
getting something important right, soit is no surprise that
Korsgaards books are virtually overflowing with pene-trating
insights, clearly and persuasively presented. Despite this, I think
theoverall picture she presents is seriously mistaken.Instead of
seeing useful metaphors in the language, Korsgaard sees some-
thingmore literal. She endorses Socrates view that terms such as
self-controland self-command are tracks or clues that the virtue of
temperance has leftbehind in language to show us that the soul has
both a ruled and a rulingpart.36 I would not want to deny that the
terms that appear in human lan-guages are often clueseven that they
are clues about human nature. Forexample, the existence, in most
languages, of words for the various coreemotions tells us that
human beings have typical patterns of response to cer-tain states
of affairs. But the existence of the phrase I grasped his meaning
ina flash is not a clue of this sort, and nothing but confusion can
result from thepicture of our minds reaching out with something
like a mental hand to graspthe abstract entity of someone elses
meaning. Korsgaard plausibly claimsthat Socrates interpretation of
the language of self-command is echoed inKants remarks about the
idea that we impose duties on ourselves. In Kantscase, the relevant
phrase is I owe it to myself.37 Given the problems that arise
certain response to an incentive is appropriate or called-for.
Korsgaard, Self-Constitution, 111.Cf. Shemmer, Desires As Reasons,
3356.36Korsgaard, Self-Constitution, 145.37Ibid., 146.
452
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
Wittgenstein, Korsgaard and the Publicity of Reasons 15
when one takes such metaphors too literallyone needs to be both
passive,in being bound, and active, in bindingit is no surprise
that this talk leadsKant to an antinomy, and that the resolution of
this antinomy requires anappeal to the mystery of the noumenal
self. Nor is it a surprise that bothPlato and Kant end up appealing
to metaphor and allegory in explaining thepossibility of evil
choicechoice that fails to meet the constitutive standardsfor
choiceand offer accounts of its origin that even Korsgaard regards
aselaborate and rather bewildering.38
V. A Wittgensteinian Suggestion
Here is an alternative proposal for understanding the notions of
reasons andrationality: a proposal that gives priority to the
third-personal perspective.At its core, the proposal is that the
notion of the irrational is useful to usbecause it allows us to
pick out and talk about the behavior of other people forwhich we
cannot see adequate reasons.39 The ability to pick out such a
classof action is important for a variety of reasons.40 One reason
is that the personmay actually be acting irrationally. In that case
it is important to draw atten-tion to this fact, and perhaps to try
to figure out the source of the irrationality.On the other hand, a
third persons behavior may fail to make sense to us onlybecause we
do not have all of the relevant information; we cannot see all
ofthe reasons for which the person might be acting. By expressing
our initialsuspicion that the person might be acting irrationally,
we are also inviting acorrection of that suspiciona correction that
would work by citing a cer-tain class of considerationsreasons for
actionof which we were ignorant.At the end of a discussion of the
rationality of a third persons action, wemight then have a
satisfying understanding of the considerations that aremoving her
to act. Surely the ability to express the relevant form of
puzzle-ment, and the ability to dispel that puzzlementor to be
confirmed in itareas useful as the ability to describe the colors
and shapes of plants. It is there-fore unsurprising that such a
vocabulary should have evolved and persistedin human languages. But
the usefulness here has nothing to do with first-person
deliberation, or with our own ability to act. On the present
hypothesisirrational has a function similar to ill. And the notion
of a practical reason
38Ibid., 164.39Here and elsewhere, we should be interpreted as
we human beings as a group, not each ofus, individually.40In fact,
this proposal is similar in spirit to one that Korsgaard herself
makes about the notionof the defective: we need the notion of the
defective for all sorts of purposes. Korsgaard, Self-Constitution,
34. Moreover, Korsgaard rightly links the notion of the defective
with the notionof normativity. Taken together, these points seem to
support the idea that we have the notionof normativityof something
being a reason for or against, or an actions being rational
orirrationalfor the same kinds of pragmatic social purpose for
which we need the notion of thedefective. But of course that is not
the way Korsgaard proceeds to argue.
453
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
16 Joshua Gert
is an important tool in our efforts to come to accurate beliefs
about the ratio-nal status of the actions of other people. And just
as the notion of illnesshas primarily a third-personal role, so too
do the notions of irrationality,rationality and reasons.I do not
mean to suggest that all, or even the most important, roles for
the
notions of reasons and rationality make sense only in contexts
in which weare puzzled by the behavior of a third party. I might be
wondering about yourbehavior, and you might offer me the sorts of
reasons that allow me to makesense of your behavior. But that is
consistent with the social nature of lan-guage. It is simply an
acknowledgement that the object of interest, in manyconversations,
is often one of the people in that very conversation. I do notwant
to suggest that second-personal uses of the notions of reasons and
ratio-nality are unimportant. But I do want to make this claim
about first-personalusesat least when the importance is with
respect to theoretical matters.In fact, attention to first-personal
uses is worse than unimportant; it is mis-leading. It obscures the
social function of the relevant concepts by removingthose functions
from direct view. It is true that once one has mastered thenotions
of practical rationality and practical reasons, one can turn an eye
onones own behavior. And inasmuch as those who care about us do not
wantus to act irrationally, they will teach us to avoid irrational
behavior, and toresist the short-term emotional impulses that
sometimes cause it. That is, theywill help us cultivate what one
might call a de dicto aversion to irrationalaction.41 As a result,
my own assessment that the course of action I am aboutto embark on
is irrational might, in some cases, trigger this aversion. Thatmay
be of some practical importance. But it does not help us understand
thenotions of reasons or rationality.A focus on the third-personal
perspective makes it clear that I can regard
something as a reason quite independently of its connection
either to myown motives, or to the motives of the agent whose
behavior I am assessing.Rather, the role of reasons in assessing
the behavior of third parties suggeststhat certain substantive
considerationpractical reasonswill place an actin need of
justification (by other reasons). For example, if I see someone
act-ing in a way that will obviously cause her a great deal of
pain, or that risks herlife in an avoidable way, or that will cause
her to lose certain abilities, and ifI cannot see any similar
substantive considerations on the other side, I willclassify her
behavior as irrational. These harms therefore count as
practicalreasons against the action. And it will not do anything to
help me understandher behavior to be toldeven in a convincing
waythat this person simplytends to seek pain, death and other such
evils, or that she desires them for theirown sake, or that she
endorses such desires wholeheartedly. If you manage to
41I take this use of de dicto fromMichael Smith, who argues
against the idea that the character-istic concern of a morally good
person is a de dicto concern with moral rightness and
wrongness.Smith, The Moral Problem, 802.
454
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
Wittgenstein, Korsgaard and the Publicity of Reasons 17
convince me of these things, I will only be more firm in my
opinion that theagents behavior is irrational. The
desire-independence of the relevant con-siderations here is the
same as the desire-independence of what counts as anillness. The
general Wittgensteinian perspective therefore yields a conceptionof
practical reasons that includes at least a large class of objective
reasons.Moreover, sinceas a matter of contingent human natureit
makes sense tous that someone would make a sacrifice to spare some
other person one of thesubstantive harms that appear in
descriptions of these objective reasons, wecan also say that the
Wittgensteinian perspective yields a large class of
public(agent-neutral) reasons as well.Korsgaard herself says a
number of things that are surprisingly congenial
to the perspective I am offering here: a perspective from which
we explainthe emergence of certain concepts by reference to the
practical usefulness oftalk that makes use of those concepts. For
example, here is a remark of hersthat helps explain certain
features of our practices of causal attribution: Whythen do we say
that the knife, rather than the state of the world a nanosecondago,
is the cause of the cutting? That is easybecause we can use the
knifefor cutting.42 I think Korsgaard would agree that this same
remark also helpsto explain why we have such concepts as knife in
the first place. That is,it is not as if we can simply take it for
granted that we have the conceptsknife and state and world and
nanosecond and cause, and only have todecide which is the cause of
the cuttingthe knife or the state of the world ananosecond ago.
Rather, it is our own need to use things to get things donethat
helps explain why we have these concepts, and why they fit together
asthey do.43
As Korsgaard almost claims, a practical perspective on language
andconcept development can be used to explain the concept of a
person andof an action as well. That is, we have the concept of a
person not becausewithout such a concept we would not be people,
but because we have tointeract with people. And we have the concept
of an action not becausewithout such a concept we could not act,
but because the people around usbehave in various ways, and some of
them are usefully categorized togetheras actionsbecause of the
practical differences between those behaviors andother ones.
Primarily, we can systematically influence actions in certain
ways:by communicating information, making requests, threatening,
and so on.None of this has to do with how we must think of our own
action, but ofcourse it has its effect on how we do this, since we
impose the same catego-rization on our own behavior as we do on the
behavior of others. Korsgaarddoes not quite say this. Rather than
focusing on the third-personal use of thenotion of a person or an
action, she makes her knife point about the con-cepts of I and you:
Why do I say that I was the cause of the cutting, or
42Korsgaard, Self-Constitution, 89; see also 114.43For her
presentation of this sort of view in Kant, see Ibid., 389.
455
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
18 Joshua Gert
that you were? For the same reason: so that we can act and
interact.44 This isright, but it is also dangerously liable to
misinterpretation. That is, it is rightthat the concept of I is
important in social interactions. But Korsgaards wayof putting the
claim suggests that I need the concept of I simply in order
toenable me to act. As a result, she claims that The ideal of
agency is the idealof inserting yourself into the causal order, in
such a way as to make a genuinedifference in the world.45
What I think she should have said instead is very similar, but
explicitlythird-personal: it is that the idea of agency is the idea
of an agent beinginserted into the causal order, in such a way as
to make a genuine differ-ence in the world. It seems to me that
this is in fact what she does say aboutanimals. It is obvious that
no self-conception is operative in the scurrying of aroach. Still,
a roach has a form, and instincts, and therefore some principles
ofbehavior that make some of its movements into actions, and others
not. Thoseprinciples help us understand and control its behavior in
certain characteristicways, and that is why we regard the roach as
a (very primitive) agent.
VI. Conclusions
Lest my focus on Korsgaards Kantian view be taken to suggest
that I favora Humean view, let me emphasize that Humeans typically
suffer from thesame misplaced emphasis on the first-personal
perspective, and on the roleof reasons in deliberation. It is true
that Humeans recognize a third-personalrole for reasons, in
providing explanations. Indeed, this is part of what theytake to
support their view: the idea that a link to the agents desires
isrequired for reasons to play this explanatory role.46 But the
problem is thatthis explanatory function is performed even more
effectively by a distinct con-cept: the concept of a desire or
motive. Normative reasons only connect withactual behavior under
certain conditions, which may well not obtain. Forexample, the
agent may simply be ignorant of the consideration that wouldprovide
motivation if she were aware of it and were to engage in sound
delib-eration. Moreover, by emphasizing the explanatory role, the
Humean robsthe concept of a normative reason of much of its utility
in assessing behav-ior as crazy, stupid, wrong-headed, and so on.
For example, if someonesinsanely self-destructive behavior happened
to line up with her insanely self-destructive subjective
motivational set, a Humean such as Bernard Williamswould be unable
to explain the craziness of the resulting actions in terms
ofreasons against performing it.47 But surely we do use the notion
of a norma-tive reason in offering assessments in such cases, and
not merely in cases in
44Ibid., 89.45Ibid.46Williams, Internal and External Reasons,
82.47Ibid., 86.
456
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
Wittgenstein, Korsgaard and the Publicity of Reasons 19
which the agent frustrates her own ends. That is, we often
assess action ascrazy or stupid in virtue of the ends at which it
is directed, or the ends towhich it is insufficiently sensitive. It
is true, as some Humeans have empha-sized, that these latter
criticisms will not be effectively directed against theirrational
person herself if they do not hook up in some way with her
presentmotives.48 But that does nothing to undermine their utility
from the third-personal standpoint, in which you and I are
discussing the nature of someoneelses irrationality. Nor does it
mean that their use from the second-personalperspective is mere
bluff. When we tell someone she has a reason to behavein a certain
way, we need not even be pretending that this will get her to act
inthe relevant way. We may be trying to convince her that she has
an affectivedisorder and should get some treatment. Or we may
simply be trying to bringcertain rationally permissible options to
her attention.On the Wittgensteinian view, practical reasons are
not things we create as
individuals, as Korsgaard or the Humean would have it. In some
sense wemight say that they are still our creations, but the our
here is wider, andthe creation is not of the same sort. Rather,
human nature being what itis, we have developed the notions of
reasons and rationality that we have.But the same is true of the
notions of red and green as well. And given theseconcepts, it is
not up to any individual or to the human race as a whole todecide
whether blood is red or grass is green. Neither is it therefore
true thatanything we decide could make it cease to be true that the
prospect of painprovides a reason to avoid a certain action.One
worry Korsgaard expresses about her own viewand to which she
thinks she has an answeris that if all reasons are public, it
looks like thereis no room for personal projects. Any reason I have
to pursue some goal willequally be a reason for you to help me to
pursue it. And if the goal is to bringabout some state of affairs,
then you will have a reason to bring it about aswell.49 Given this,
there seems to be no room for an agent to dedicate himselfto a set
of particular goals, ignoring most of the public reasons provided
bythe goals of other people. This problem might seem to require
Korsgaardto change her view to allow at least some
privateagent-relativereasons,as she notes that Thomas Nagel did in
response to the same problem.50 Thatwould allow us to say that you
have agent-relative reasons to engage in certainprojects without
entailing that other people, even in their idle moments, haveany
reason to provide you with active support. But Korsgaard is right
to saythat our mere desires, intentions or commitments do not seem
to be whatprovides us with reasons for our particular projects.
Rather, we think of our
48Ibid., 87.49Korsgaard, Self-Constitution, 207.50Nagel, View
from Nowhere, 16474.
457
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
20 Joshua Gert
projects as being supported by the agent-neutral values of the
objects of ouraffective attitudes. Here is how Korsgaard puts the
point:
Although I may not suppose that the happiness of my loved ones
isobjectively more important than that of anyone else, I certainly
do sup-pose that their happiness is something for which there is
public reason.So the structure of reasons arising from love is
similar to the structure Ihave proposed for the case of reasons of
personal ambition. I think thatsomeone should make my darling
happy, and I want very much to bethat someone.51
But the question remains for Korsgaard how this very much
wanting ratio-nally allows one to respond to the public reasons one
prefers to respond to, tothe exclusion of otherand more
numerouspublic reasons of which one isaware. How can one spend ones
afternoon in a workshop placing the frets ona guitar when one knows
that one could be responding to public reasons thathave to do with
such things as preventable hunger, sickness and death? Surely,if
the hunger, sickness and death were ones own, one could not
rationally passthe afternoon in the workshop in this way. And if
all reasons are public, it ishard to see how the location of the
reason could make any such difference.The third-personal
perspective on reasons and rationality has no analogous
problem. On such a view it is only certain considerations that
place an actionin need of justification by reasons in the first
place: risk of pain, death, injury,and so on, to the agent. But
this does not mean that all reasons are agent-relative. For, in the
case of actions that do risk these things, we can see themas making
senseas being rationally justifiedwhen the risk is run for thesake
of something that is worth it. And while such justifying
considerationsof course include benefits such as pleasure and
knowledge for the agent, theyalso include the prevention of harms
for other people, or even to non-humananimals. The fact that these
agent-neutral considerations are worth itthatthey provide reasons
that can justify the choice of an action that we know willinvolve
our own sufferingdoes not entail that they require us to make
thosesacrifices. Nor does it entail that they require us to abandon
the pursuit of ourpersonal projects.
Acknowledgements
I thank Victoria Costa and Christopher Freiman for helpful
written com-ments on an earlier version of this paper. Thanks also
to Hector Arrese Igor,Diego Lawler, Glenda Satne and Bernardo
Ainbinder for the invitation to
51Korsgaard, Self-Constitution, 211.
458
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015
-
Wittgenstein, Korsgaard and the Publicity of Reasons 21
speak at SADAF in Buenos Aires, and to them and the other
participants forhelpful discussion.
References
Arpaly, Nomy, and Timothy Schroeder. Deliberation and Acting for
Reasons. PhilosophicalReview 121, no. 2 (2012): 20939.
Blackburn, Simon. Ruling Passions. Oxford: Clarendon,
1998.Coleman, Mary. Public Reasons and Practical Solipsism.
Southern Journal of Philosophy 43
(2005): 31736.Gert, Bernard. Morality: Its Nature and
Justification. Rev. Ed. New York: Oxford University
Press, 2005.Gert, Joshua. Brute Rationality. Nos 37, no. 3
(2003): 41746.Gert, Joshua. Korsgaards Private-Reasons Argument.
Philosophy and Phenomenological
Research 64, no. 2 (2002): 30324.Hurley, Paul. A Kantian
Rationale for Desire-Based Justification. Philosophers Imprint
1,
no. 2 (2001): 116.Korsgaard, Christine. The Activity of Reason.
APA Proceedings (November 2009): 308.Korsgaard, Christine. The
Constitution of Agency: Essays on Practical Reason and Moral
Psychology. New York: Oxford University Press, 2008.Korsgaard,
Christine. Self-Constitution: Agency, Identity, and Integrity. New
York: Oxford
University Press, 2009.Korsgaard, Christine. The Sources of
Normativity. Cambridge: Cambridge University Press,
1996.LeBar, Mark. Korsgaard, Wittgenstein, and the Mafioso.
Southern Journal of Philosophy 39,
no. 2 (2001): 26171.Mill, John Stuart. Utilitarianism. Ed.
George Sher. Indianapolis, IN: Hackett, 1979.Nagel, Thomas. The
View from Nowhere. New York: Oxford University Press, 1986.Parfit,
Derek. Normativity. In Oxford Studies in Metaethics. Vol. 1, ed.
Russ Shafer-Landau,
32580. Oxford: Clarendon, 2006.Shemmer, Yonatan. Desires As
Reasons. Philosophy and Phenomenological Research 75, no. 2
(2007): 32648.Skorupski, John. Irrealist Cognitivism. Ratio 12
(1999): 43659.Smith, Michael. The Moral Problem. Oxford: Blackwell,
1994.Wallace, R. Jay. The Publicity of Reasons. Philosophical
Perspectives 23 (2009): 47197.Williams, Bernard. Internal and
External Reasons. In Varieties of Practical Reasoning, ed.
E. Millgram, 7789. Cambridge, MA: MIT Press, 2001.Wittgenstein,
Ludwig. The Blue and Brown Books. New York: Harper and Row,
1958.Wittgenstein, Ludwig. A Lecture on Ethics. Philosophical
Review 74 (1965): 312.Wittgenstein, Ludwig. Philosophical
Investigations. Trans. G. E. M. Anscombe. New York:
Macmillan, 1953.Wittgenstein, Ludwig. Zettel. Ed. G. E. M.
Anscombe and G. H. von Wright. Trans. G. E. M.
Anscombe. Berkeley, CA: University of California Press,
1967.
459
Dow
nloa
ded
by [F
U Be
rlin]
at 13
:22 24
June
2015