Philosophy and Phenomenological Research
Vol. LXVIII, No. 1, January 2004
Rule-Following and Externalism*
ALEXANDER MILLER
Macquarie University
John McDowell has suggested recently that there is a route from his favoured solution to
Kripke's Wittgenstein's "sceptical paradox" about rule-following to a particular form
of cognitive externalism. In this paper, I argue that this is not the case: even granting
McDowell his solution to the rule-following paradox, his preferred version of cognitive
externalism does not follow.
1. Does the correct response to the "sceptical paradox" about rule-following
adumbrated by Kripke's Wittgenstein entail, or make more palatable, any
interesting form of cognitive externalism?' In particular, does the correct
response to the rule-following paradox entail, or make more palatable, any
form of cognitive externalism suggested by Putnam's famous Twin-Earth
argument, that the physical and social environments in which we actually
live at least partly constitute the fact that we mean what we mean by at least
some of our words?' John McDowell has recently suggested that this question
can be answered affirmatively. In this paper I want to question this suggestion.
I do not want to take issue either with McDowelľs response to
Kripke's Wittgenstein's sceptical paradox about rule-following, or the version
of Putnam's cognitive externalism which McDowell favours. I want to suggest
merely that there is no direct route from the dissolution of the rule-following
paradox to McDowelľs version of cognitive externalism.
2. I will begin with a brief recap of the Twin-Earth argument (Putnam 1975).
Imagine that Earth and Twin-Earth are very distant parts of the same possible
world. There is only one difference between Earth and Twin-Earth: whereas on
Earth, water-the clear, odourless, thirst-quenching liquid that fills rivers and
lakes etc.-has underlying molecular structure H,O, on Twin-Earth the super-
*
For comments and discussion I am very grateful to Jim Edwards, Brian Garrett, Guy
Longworth, Brian Morrison, Gregory McCulloch, Harold Noonan, Jim Stuart, and two
anonymous referees for this journal.
For the rule-following paradox, see Kripke (1982) and Miller and Wright (2002).
See Pumam (1975). The Twin-Earth style argument is restricted to words standing for
natural kinds (see n. 3 below). I won't be concerned in this paper with externalist arguments,
such as Burge's, which attempt to generalise beyond the case of natural kind
terms.
'
'
RULE-FOLLOWING AND EXTERNALISM 127
ficially indistinguishable clear, odourless, thirst-quenching liquid that fills
rivers and lakes etc. has underlying structure XYZ. Suppose that Oscar is an
Earthian speaker, while Toscar is a Twin-Earthian speaker, where Oscar and
Toscar are atom-for-atom doppelgangers. If Oscar were transported to TwinEarth,
his application of `water' to the clear, odourless liquid in the glass
would be false, since only things with the molecular structure H,O count as
water in his language; likewise if Toscar were transported to Earth, his
application of `water' to the clear, odourless liquid in the glass would be
false, since only things with the molecular structure XYZ count as water in
his lang~age.~Thus, `water'as used by Oscar has a different extension from
`water'as used by Toscar. So, the thought expressed by Oscar's utterance of
`The river Taff is full of water' is different from the thought expressed by
Toscar'sutterance of `Theriver Taff is full of water'. Since Oscar and Toscar
are atom-for-atom doppelgangers, it follows that "meanings ain't in the
head".
Since McDowell and Putnam both discuss the implications of this argument
using (broadly) Fregean notions such as `sense' and `extension', and
since-as we'll see-Kripke's Wittgenstein's sceptical argument can easily
be framed in the same terms, it will be useful to spell out the implications of
the Twin-Earth argument explicitly in terms of Sinn and Bedeutung. The
extension of `water' as used by Oscar is different from the extension of
`water' as used by Toscar. Since the Bedeutung of a predtcate is determined
by its extension, `water'as used by Oscar has a different Bedeutung from
`water'as used by Toscar. So, `The river Taff is full of water' as uttered by
Oscar expresses a different thought (Sinn)from that expressed by `The river
Taff is full of water' as uttered by Toscar. Thus, the nature of Oscar's environment
constrains the extensionof his term, the extension of the term constrains
its Bedeutung, and its Bedeutung constrains its Sinn.So the nature of
Oscar's environment constrains the Sinn of his term `water'. Likewise for
Toscar.
What are the implications of this argument for views on the nature of
psychological states? In order to see where Putnam and McDowell diverge,
consider:
(A) The sense of `...is water' determinesits extension.
@) Grasping the sense of `... is water' is just a matter of being in a certain
psychological state.
In other words, `water'as used by Oscar and Toscar is a natural kind term. This argument
is not by itself intendedto extend to cases of predicateswhich stand for kinds which
are not natural kinds.
128 ALEXANDER MILLER
(C) Psychological states are "psychological states in the narrow sense"
("states whose attribution to a subject entails nothing about her
environment"):
(D) As far as psychological states in the narrow sense are concerned,
Oscar and Toscar are identical.
(E) Given (A), (B), (C), and (D), the extensions of `... is water', as used
by Oscar and Toscar,are identical.
But from the Twin-Earth argumentrehearsed above:
(F) The extensionsof `...is water', as used by Oscar and Toscar, differ.
(G) Contradiction,from (E)and (F).
Thus
(H> We must give up either (A), (B), (C), or (D).
At this point, Putnam and McDowell appear to disagree about how we
ought to proceed.' Putnam accepts (C), and on this basis argues that since we
cannot accept both (A) and (B), we ought to jettison (B). McDowell on the
other hand thinks we ought to reject (C), thereby opening up the possibility
of a conception of psychologicalstates which will allow us to embrace both
(A) and (B):
What is to be learned from [the Twin-Earth] reflections is not, as Putnam himself argues, that it
cannot be true both that "knowing the meaning of a term is just a matter of being in a certain
psychological state" and that "the meaning of a term determines its extension"; so that if we
retain the second of these assumptions, we must renounce the first. This presupposes that anyone
who embraces the first assumption must be restricting psychological states to "narrow"
states. Rather, the moral of Putnam's considerations is that the idea of a psychological state, as
it figures in the first assumption, cannot be the idea of a "narrow" state. That is: we should not
leave in place an idea of the mind that is shaped by the tenets of "methodological solipsism",
and conclude that meanings are not in the mind, since they are not in the head. Rather, we
should read the two assumptions in such a way that they can be true together and exploit such a
reading to force us into explicit consideration of a different conception of the mind6
McDowell (1992a), p.36.
An anonymousreferee has helpfully pointed out to me that in more recent work, Putnam
appears to accept McDowelľs critique of his earlier position. See Putnam (1994). So in
what follows, `Putnam' refers to the Putnam of (1975). If what 1 argue in the paper is
correct, the later Putnam may have been overhasty in accepting Mchwelľs critique of
his earlier self.
McDowell (1992a). p.40.
RULE-FOLLOWING AND EXTERNALISM 129
But why opt for McDowelľs reading of the Twin-Earth argument rather than
Putnam's? As we'll see, McDowell argues that the conception of mind
underpinning Putnam's reading is exploded by the correct response to
Kripke's Wittgenstein's sceptical paradox, so that once we have dissolved the
sceptical paradox, McDowelľs reading emerges as the more plausible interpretation
of the Twin-Earth argument. Before looking at this claim, some
remarks about Kripke's Wittgenstein's sceptical paradox, and McDowelľs
response to it, are in order.
3. Leťs begin by setting up Kripke's Wittgenstein's sceptical challenge
within the broadly Fregean framework we used in 52. Recall that according to
Frege, an expression like the addition sign `+' has a function as its Bedeurung.
For example, we would normally take `+' to have the addition function
as its Bedeutung, the extension of which would include the ordered triples
el, 1, 2>, <1, 2, 3>, <68, 57, 1 2 5 . (An ordered triple belongs to the
extension of a function if and only if the first two members yield the third
when presented to the function). But of course, `+' could conceivably have
some other Bedeutung. For example, it could stand for the quaddition function,
the extension of which would include <1, 1, 2>, <1, 2, 3>, and <68,
57, 5>. So we can ask: what determines which of these functions is the
Bedeutung of `+', as used by an individual or a group of speakers? And the
intuitive answer is: the meaning of `+'; how the speakers understand `+'; or
the Sinn the speakers associate with `+'. So the question now becomes: what
constitutes the fact that speakers associate one rather than another Sinn with
`+'? Given the Fregean thesis that Sinn determines Bedeutung, any fact cited
as constitutive of meaning, understanding, or Sinn, must determine Bedeutung.
In effect, Kripke's Wittgenstein argues as follows. No fact is capable
of determining the Bedeutung of a linguistic expression; therefore, no fact
can be constitutive of Sinn. So, sentences such as `Jones means addition by
`+" are not truth-apt; they do not express beliefs; they do not purport to rep
resent facts; they do not have truth-conditions.
Now the precise details of Kripke's Wittgenstein's negative arguments,
their shortcomings, and the "sceptical solution" which Kripke's Wittgenstein
suggests in order to amelioratetheir "paradoxical" conclusion,needn't concern
us here.' But three points are worth noting:
Note 1: Kripke's Wittgenstein's argument is perfectly general in the sense
that it applies to all of the classical Fregean syntactic categories if it applies
to any, so that proper names, predicates, sentences, and quantifiers all fall
within its scope. So any results about how the solution to Kripke's WittgenFor
an account of the details,see Miller (1998),Chapters 5 and 6.
130 ALEXANDER MILLER
stein's paradox impacts on issues about externalism should generalise quite
straightforwardlyto all of these cases.*
Note 2: Kripke's Wittgenstein's argument is also general along a different
dimension.It appliesto all sorts of predicates: natural kind predicates such as
`golďand `water',and non natural kind predicates(such as,perhaps, `reď).
Note 3: Kripke's Wittgenstein's argument is also general along a further,
third, dimension. The argument applies to linguistic meaning, but also to
mental content.' So any results about how the solution to KW's paradox
impacts on issues about externalism should apply equally well to semantic
externalismor cognitiveexternalism.
4. How doesMcDowell respond to Kripke'sWittgenstein's sceptical paradox?
Intuitively, the fact that someone grasps a certain sense or Sinn is a fact
about that person's mind. Supposethat one accepted a conception of the mind
(call it the Master-Thesis) according to which
[The mind) is populated exclusively with items that, considered in themselves, do not sort
thingsoutsidethe mind, including specifically bits of behaviour, into those that are correct or
incorrect in the light of those items."
The Master-Thesis has it that, considered in themselves, mental states and
actsjust "stand there like a signpost"." A sign-post,considered in itself, does
not sort episodes of behaviour into those that constitute following the signpost
correctlyor incorrectly. So, if you accepted this conception of the mind,
you would have to give some account of what does sort episodes of behaviour
into those that constitute following the signpost correctly, or as acting
in accord with the meaning which is grasped. A tempting answer would be
that
What does sort behaviourinto what counts as following the sign-post and what does not is not
an inscribed board fixed to a post, consideredin itself, but such an object under a certain interpretation-such
an object interpretedas a sign-postpointing the way to a certain destination."
But accordingto Wittgenstein this idea is hopeless. Recall the passage from
Philosophical Investigations 0198:
Any interpretation still hangs in the air along with what it interprets, and cannot give it any
support. Interpretationsby themselvesdo not determine meaning.
For an excellentexplanation of why Kripke's Wittgenstein's challenge is general in the
senses specifiedin notes 1 and 2, see chapter 4 of McGinn (1984).
For the reasons why, see Miller (1998), 56.1.
lo McDowelI (1992b),p.41.
I ' See Wittgenstein(1974), $85.
McDowell (1992b),p.41,I2
RULE-FOLLOWING AND EXTERNALISM 131
Suppose that meaning such and such by an expression was a matter of putting
a certain interpretation on it. Then what is required for competence is
that the correct interpretation is put on it. But, according to the Master-Thesis,
acts of mind-such as putting an interpretation on an expression-just
"stand there like a sign-post". So, we require that the interpretation of the
expression itselfbe interpreted. And now we are off on a regress: the interpretation
of the interpretation in its turn just "stands there like a sign-post", and
so will require interpretation, and so on ad infiniturn.
Now according to Kripke, at this point Wittgenstein accepts that there is
no such thing as a fact about what anyone means and intends, and attempts to
develop a sceptical solution to preserve meaning and intention in the light of
this conclusion:
This was our paradox: no course of action could be determined by a rule, because every
course of action can be made out to accord with the rule. The answer was: if everything can
be made out to accord with the rule, then it can also be made out to conflict with it. And so
there would be neither accord nor ~0nflict.l~
But as many commentators have pointed out, the paragraph in Philosophical
Investigations which follows this one shows that contrary to what Kripke
claims, Wittgenstein does not accept the sceptical paradox that there is no
such thing as a fact about meaning:
It can be seen that there is a misunderstandinghere from the mere fact that in the course of
our argument we give one interpretation after another; as if each one contented us at least for
a moment, until we thoughtof yet another standingbehind it. What this shows is that there is a
way of graspinga rule which is not an interpretation,but which is exhibited in what we call
"obeyingthe rule" and "going against it" in actual cases.14
l3 Wittgenstein (1974), $201.
l4
Ibid. Note that althoughmost commentators agree that Kripke misinterprets Wittgenstein
on this point, George Wilson, Alex Byrne, and David Davies are exceptions. See e.g.
Wilson (1994), Wilson (1998), Byrne (1996), Davies (1998). Wilson's interesting papers
deserve a closer examinationthan I can give them here, but I think his interpretation of
Kripke's Wittgenstein is dubious for at least the following reason. According to Wilson,
Kripke's Sceptic argues as follows:
(NS) If X means somethingby a term "A", then there is a set of properties, P,,...,P,.
that govern the correct application of "A" for X.
(G) If there is a set of properties, Pl....,P,, that govern the correct application of "A"
for X, then there are facts about X that constitute Pl,...,P, as the conditions that
govern X's use of "A".
(BSC) There are no facts about X that constituteany set of properties as conditions that
govern X's use of "A".
So, (RSC) No-one ever means anything by a term.
According to Wilson, Kripke's Wiffgensteinemphatically does not accept the "radical
sceptical conclusion"(RSC),but rather denies this, and uses this as a lever to reject the
132 ALEXANDER MILLER
McDowell suggests that what this shows is that Wittgenstein does not accept
the Master-Thesis plus the consequent need for interpretation, and then proceed
to draw the conclusion that since interpretations do not determine meaning,
there is no such thing as meaning an expression in one way rather than
another; rather, Wittgenstein blocks the route to the sceptical paradox by
refusing to accept the Master-Thesis in thefirst place.
Is Wittgenstein's refusal to accept the Master-Thesis justified? According
to McDowell, the Master-Thesis is an extremely unintuitive piece of philosophical
theory, and as such the onus is on its defenders to justify their
adherence to it. Consider a mental occurrence, like the having of the thought
that people are talking about me in the next room. Only a state of affairs in
which people are talking about me in the next room will be in accord with
this thought; if another state of affairs obtains, such as a state of affairs in
which the next room is empty, this will not be in accord with that thought.
The Master-Thesis then implies
that whatever I have in my mind on this occasion, it cannot be something to whose very identity
that normative link to the objective world is essential ...that what a person has in mind, strictly
"classical realist" conception of meaning embodied in (NS). Thus,Kripke's Wittgenstein,
according to Wilson, argues as follows:
(G), (BSC), not (RSC); so, not (NS).
But this sits ill with Kripke's remarks about the distinction between "straight" and "scepticaľ'
solutionsto the sceptical paradox about meaning. Recall that according to Kripke
"[A] proposed solution to a sceptical philosophical problem [is] a straight solution if it
shows that on closer examination the scepticism proves to be unwarranted"(l982. p.66).
That is, a straight solution is one which finds fault with the sceptic's reasoning or by
denying one of the premises in the sceptic's argument. In contrast "A sceptical solution
of a sceptical philosophical problem begins on the contrary by conceding that the sceptic's
negative assertions are unanswerable"(l982, p.66). That is, a sceptical solution
begins by accepting the negative conclusion of the sceptical argument. The problem for
Wilson's interpretation of Kripke's Wittgenstein should now be clear. According to Wilson,
Kripke's Wittgenstein blocks the inference to (RSC) by denying (NS): (NS) is false,
so although the inference from (G),(NS), and (BSC) to (RSC) is valid, the sceptic has no
cogent route to the conclusion (RSC). But this means that, by Kripke's lights, Kripke's
Wittgenstein is proposing a straight solution to the sceptical solution: one which finds fault
with the sceptic's reasoning or with one of his premises. But of course, Kripke clearly
intends his Wittgenstein to be proposing a sceptical solution to the sceptical paradox. It
seems to me, then, that Wilson's interpretation, interesting as it may be as a piece of Wittgenstein
exegesis, does not sit well with Kripke's Wittgenstein's take on these matters. I
do not claim that this point is decisive against Wilson's interpretation of Kripke's Wittgenstein,
but I will not pursue the matter further here, since the main argument in the text is
independent of these subtle exegetical questions. I hope to take up the issue more fully in
a later paper. For useful criticism of Wilson's interpretation, see Soames (1998), pp.337339.
Similarremarks apply to the interesting discussions of the issue in Byme (1996) and
Davies (1998).
RULE-FOLLOWING AND EXTERNALISM 133
speaking,is never, say, tlmtpeople are talking abour her in the next room but at most something
that can be interpretedas having that content, although it need not."
And McDowell finds this at worst an extraordinaryidea, and at best highly
counterintuitiveand unmotivated:
Once we realise that, the Master-Thesis should stand revealed as quite counterintuitive, not
something on which a supposed need for constructive philosophy could be convincingly
based.I6
5. For the sake of argument, leťs grant McDowell this response to the sceptical
paradox about following a rule. Our question is whether making this
concession gives us a reason to read the Twin-Earth argument in the manner
suggested by McDowell: that is, to reject (C) and hold on to (A) and (B),
rather than follow Putnam in rejecting (B) whilst retaining (A) and (C).
McDowell clearly thinks this is the case. McDowell writes
What Putnam never seemsto consideris the possibilityof a position that holds that command of
a meaning is wholly a matter of how it is with someone's mind, and combines that with the
determinationof extension by meaningso as to force a radically non-solipsistic conception of
the mind to come to explicit expression. Instead, he assumes that anyone who wants to conceive
knowledge of a meaning as wholly a matter of how it is with someone's mind must be
already committedto a theoretical conception of the mind ...which, in conjunction with Putnam's
reflections about meaning, guarantees that the wish cannot be fulfilled."
The "theoretical conception of the mind" alluded to in this passage is one
according to which the mind is the "organ of psychological activity".
McDowell writes further
Putnam, without demur, lets "mentalism"be commandeered for the view that the topic of
mental discourse can appropriately be specified as "the mindlbrain". Talk of the mindhain
embodiesthe assumption that the mind is appropriatelyconceivedas an organ, of course, with
the idea-which is in itself perfectly sensible-that if the mind is an organ, the brain is the only
organ it can sensibly be supposed to be. The assumption that the mind is an organ is one that
Putnam does not challenge...In fact much of his own thinking seems to presuppose just such a
conceptionof the mind."
And it emerges that the conception of the mind as an organ of psychological
activity is equivalent to the Master-Thesis:
htnam's governing assumption here is that a mental state or occurrence that is representational
...must in itself consist in the presence in the mind of an item with an intrinsic nature
I' McDowell (1992b),p.46.
l6 Ibid.
McDowell (1992a). pp.40-41.
McDowell (1992a), pp.42-43.
134 ALEXANDER MILLER
characterizable independently of considering what it represents. (Such a state of afairs would
be wlur an infernal arrangement in an organ of thoughtwould have to amount f0).I9
That is to say, in the terminology of McDowelľs introduced in $4 above, an
organ of thought is something whose states "just stand there like a signpost".
To conceive of the mind as an organ of thought is thus to embrace, precisely,
the Master-Thesis." So does rejecting the conception of the mind as an organ
of thought, in other words the Master-Thesis, give us a reason for rejecting
(C) rather than (B) in the argumentof $2?
6. Go back to the argument in $2, and suppose we write (C) out as follows:
(C) All psychological states are "psychological states in the narrow
sense" ("states whose attribution to a subject entails nothing about
her environment").
We've seen that McDowell wants to reject (C). But there are two ways in
which (C), thus read, can be rejected:
Weak Rejection of (C): Some psychological states-such as grasping the
sense of `water'--are not "psychological states in the narrow sense".
Strong Rejection of (C):No psychological states are "psychological states in
the narrow sense".
Given that there is a whole range of linguistic terms-those which don't
denote natural kinds-an which the argument is simply silent, it seems
likely that the most that we can establish via Twin-Earth style arguments is
Weak Rejection of (C). So for our present purposes we'll take McDowell to
be aiming at Weak Rejection of (C) (in any case, if he can't reach Weak
Rejection of (C) via rejecting the Master-Thesis, it follows trivially that he
can't reach Strong Rejection of (C) via that route either).
7. So can we justify Weak Rejection of (C) by invoking the rule-following
considerations? As noted above, McDowell seems to think we can. Since
Kripke's Wittgenstein's rule-following sceptical paradox is generated by the
l9
McDowell (199h), p.43, emphasis added.
Note that McDowell sometimes gives a slightly different explanation of the idea that the
mind is an organ. For example, he writes "the cash value of this talk of organs is the idea
that states and Occurrences `in' the mind have an intrinsic nature that is independent of
how the minďs possessor is placed in the environment"(l992a, p.39). But note that on this
definition, "the mind is an organ of thought" is simply another label for the view that
there is, in the relevant sense, no such thing as wide content. And you can't make more
palatable the idea that there is such a thing as wide content simply by relabelling its opposite.
So I suggest that the definition of "organ of thought" given in the text must be the
more fundamental. This at least gives McDowell an argumentative strategy which might
possibly work (though as 1go on to suggest in what follows, it doesn'tactually do so).
RULE-FOLLOWING AND EXTERNALISM 135
Master-Thesis, we need to reject the Master-Thesis in order to avoid the paradox.
And McDowell suggests that once we have seen our way to rejecting the
Master-Thesis (the conception of the mind as an organ of thought), it should
be easier for us to see how we can reject (C) rather than follow Putnam in
rejecting (B).
Weneed to think about the precise content of the Master-Thesis. Cashed
out slightly, what McDowelľs formulation of the Master-Thesis (p.5 above)
comes to is this:
Every mental state, consideredin itself, does not sortthings outside the mind, including specifically
bits of behaviour, into those that are correct or incorrect in the light of those states.
Now this means that there are two ways of rejecting the Master-Thesis just as
there are two ways of rejecting (C) in Putnam's argument of 32:
Weak Rejection of the Master-Thesis: Some mental states (with representational
content), considered in themselves, do sort things outside the mind,
including specifically bits of behaviour, into those that are correct or incorrect
in the light of those states.
Strong Rejection of the Master-Thesis: All mental states (with representational
content), considered in themselves, do sort things outside the mind,
including specifically bits of behaviour, into those that are correct or incorrect
in the light of those states.
Now plausibly McDowell requires the Strong Rejection of the Master-Thesis
in order to convincingly undercut Kripke's Wittgenstein's sceptical challenge.
Since Kripke's Wittgenstein's sceptical challenge is perfectly general (see
notes 1-3 in 33 above), given only Weak Rejection, KW's paradox will still
be in the field for those mental states which don't, considered in themselves,
sort things outside the mind, including specifically bits of behaviour, into
those that are correct or incorrect in the light of those states. So we shall
assume that Strong Rejection of the Master-Thesis is justifiably required in
order to ward off Kripke's Wittgenstein's sceptical paradox. Then we can ask:
can we move from SRMT to the Weak Rejection of (C)?
8. So leťs grant McDowell the thought that we need to accept Strong Rejection
of the Master-Thesis (SRMT) in order to stave off the rule-following
paradox with the requisite generality. Does it follow from SRMT that we
ought to embrace the Weak Rejection of (C) as defined in 06? One thing that
is clear is that SRMT does not entail the Weak Rejection of (C). Leťs focus
on a particular example. Consider my grasp of the sense or Sinn of the predicate
`water'. The glass on the desk in front of me contains a sample of H20.
Call this sample a. According to SRMT, it could not be the case that this
136 ALEXANDER MILLER
sample a (as it actually is) fails to satisfy the predicate `water' as I understand
it and yet the state of mind which consists in my grasping the sense or Sinn
of `water' remains unaltered. If I change the extension of `water' without at
the same time changing the world, I thereby also change the state of mind
which consists in my attaching a sense or Sinn to it. Now if this is true it
seems not to entail the Weak Rejection of (C): it seems not to entail that if I
had been taught the use of `water' in an environment different from that on
Earth (i.e. if I had been taught the use of `water' on Twin-Earth), the sense or
Sinn I attach to `water' would have been different from that I actually attach
to it.
To put the point another way. Consider again what McDowell says about
the thought that someone is talking about me in the next room (see the quote
on p.7 above), and leťs make it instead the thought that someone is drinking
water in the next room. According to McDowell, what the rejection of the
Master-Thesis comes to in this case is this: the thought can be conceived of
as something to whose identity it is essential that its representational content
is that someone is drinking water in the next room (my emphasis). But
what rejection of (C) comes to is this: it is essential to my thinking the
thought that someone is drinking water in the next room rhar there is (or has
been) some water in my environment (my emphasis). And there is certainly
no straightforward entailment from
(*) The psychological state which constitutes my thought that someone
is drinking water in the next room has its representational content
essentially2'
to
** Note that this is not obviously a trivial claim, amounting to something like "You think
someoneis drinkingwater in the next room only if you think someoneis drinking water in
the next room".McDowell would claim that (*) would be rejected by someone who held
to a "duplex" conception of representational states (such as that discussed in McGinn
1982);according to the duplex conception "the concept of command of a meaning ... is
the concept of somethingthat is, in itself, in the head, but conceived in terms of its relations
to what is outside the head"(1992b:37).Thus the state which constitutes the "in-thehead"
component of my thought that P could be possessed by someone else, where,
because of its sustaining some differentrelation to what is outside the head, it constitutes
an "in-the-head component of an entirely different thought. If the "in-the-head" state is
taken to be the psychologicalstate itself,then we have a violationof (*). I won't consider
here whether McDowell is fair to the "duplex" conception or whether there is a version
of the "duplex"conception which can respect (*), since my aim is only to see what follows
if we grant McDowell his solutionto the rule-following paradox and the associated
critique of the "duplex" conception. But there is no doubt that there is some tension in
McDowelľs position here; the less trivial (*) is, the less likely it is that the Master-Thesis
can be dispatchedsimply by calling it "quite counterintuitive",while the more trivial (*)
is, the less plausibleit is to claimthat it is violated by defenders of the "duplex" concep-
tion.
RULE-FOLLOWING AND EXTERNALISM 137
(**) It is essential to my thinking the thought that someone is drinking
water in the next room that there is (or has been) some water in my
environment.
Or at least, if McDowell thinks that there is such an entailment, we are still
very much in need of an argument tojustify this thought.
Or to put the point a third way. What SRMT gets you is the thought that
changing the extension of a predicate whilst keeping the state of the world
fixed necessarily goes with changing its sense or Sinn. But this claim is
simply silent on whether the sense or Sinn can stay constant if we change the
extension but also change the state of the world.
In general, then, the internal relationship between sense and extension
which is at the heart of SRMT is logically independent of the internal relationship
between environment and sense which is at the heart of Weak Rejection
of (C). But this undercuts McDowelľs appeal to the Strong Rejection of
the Master-Thesis as a way of explaining how (C) can plausibly be rejected.
The picture of the mind-the rejection of the idea of the mind as an organ or
as populated only by states allowed by the Master-Thesis-might be plausible,
but the invocation of that picture in no way helps us get a grip on how
there can be states of mind-like grasping the sense of `water'-which are
not "psychological states in the narrow sense".
This point can be pressed home by considering how matters look from the
standpoint of Putnam." Putnam runs the Twin-Earth argument and at the
final stage he rejects (B) rather than (C). McDowell rejoinds: "you've only
rejected (B) rather than (C) because you don't accept SRMT. If you accepted
SRMT you would thereby be entitled to give up (C) via accepting Weak
Rejection of (C)". But now why should Putnam not issue the following
counter-reply: "nothing you have said convinces me that accepting SRMT
entails Weak Rejection of (C). So even if I accept SRMT, I can still refuse to
accept the Weak Rejection of (C). Thus, nothing you've said yet convinces
me that I ought to give up (C) rather than (B)".
9. However, perhaps, there is a response available to McDowell here. Perhaps
McDowell can argue that what prevents Putnam from rejecting (C) rather
than (B) is his refusal to accept SRMT. That is to say, maybe the MasterThesis
entails the denial of the Weak Rejection of (C), so that unless Putnam
accepts SRMT he will not see the opportunity of reading the Twin-Earth
argument in such a way as to preserve (B) at the expense of (C). Is it plausible
to suppose that if we accept that there are no mental states of grasping
senses which considered in themselves do impose a normatively characterisable
shape on things outside the mind we must reject Weak Rejection of (C)
22
Again, I am referring here to the F'umam of (1975).
138 ALEXANDER MILLER
i.e. we must conclude that "all psychological states are psychological states
in the narrow sense". Take any such state: my grasping the sense of `water'
for example. Then it would follow that I could be in that very same state of
mind even though the world hadn't changed but the extension of the term had.
Does it follow from this that I could have been in that very same state of
mind even if I had been educated in the use of `water' in a Twin-Earth style
environment? It seems not: this is, at best, extremely unclear. Again, if this
is the point McDowell intends to press, we are badly in need of an argument
to back it up?3
10. My conclusion is thus that even if we grant McDowell his favoured way
of undercutting Kripke's Wittgenstein's sceptical paradox about rule-following,
it is at best very unclear how that is supposed to legitimate a reading of
Putnamian Twin-Earth style arguments which conclude, inter alia, that at
least some states of mind-such as grasping the sense of a natural kind term
like `water'-are not "psychological states in the narrow sense". I therefore
end with a challenge to those philosophers who think that there is a route
from the rule-following considerations to McDowell-style cognitive externalism:
explain how we can move from SRMT, either with or without auxilliary
premises, to the Weak Rejection of (C),or find a better route from rulefollowing
to e~ternalism.2~
~
23
And even if McDowell had such an argument, iťs not clear that it would be enough. That
which prevents Putnam from rejecting (B) rather than (C) would have been removed. But
this hardly amounts to a reason to reject (C) rather than (B), which is surely what Putnam
will rightly demand of McDowell.
The avoidance of metaphor is important here: it is perhaps easy to slide from the thought
that a state of mind does not "just stand there like a signpost" to the thought that that state
of mind is, in the familiar sense, "wide". Once we think seriously about what underlies
the "signpost" metaphor, it is less easy to make the slide. It seems to me that a similar
problem vitiates much of McDowelľs (1986). Just as in the papers under consideration
above he confuses SRMT with the Weak Rejection of (C), in his (1986) he appears to
confuse something like SRMT with allowing the possibility of object-dependent thoughts.
This confusion seems to me to be no better than the one I criticise in the text, but I cannot
justify this suspicion here.
24
RULE-FOLLOWING AND EXTERNALISM 139
REFERENCES
Byme, A (1996) "On Misinterpreting Kripke's Wittgenstein", Philosophy
and Phenomenological Research 56, pp.339-343.
Davies, D (1998) "How Sceptical is Kripke's `Sceptical Solution'?', Philosophia
26, pp.119-140.
Kripke, S (1982) Wittgenstein on Rules and Private Language (Oxford: Oxford
UniversityPress).
McDowell, J (1986) "Singular Thought and the Extent of Inner Space", in J.
McDowell and P. Pettit (eds.) Subject, Thought, and Context (Oxford:
Clarendon Press), pp.137-168.
McDowell, J (1992a) "Putnam on Mind and Meaning", Philosophical Topics
McDowell, J (1992b) "Meaning and Intentionality in Wittgenstein's Later
McGinn, C (1982) "The Structure of Content", in A. Woodfield (ed)Thought
McGinn, C (1984) Wittgenstein on Meaning (Oxford: Basil Blackwell).
Miller, A (1998) Philosophy of Language (Montreal: McGill-Queens University
Press).
Miller, A and Wright, C (eds.) (2002) Rule-Following and Meaning (London:
Acumen).
Putnam, H (1975) "The Meaning of Meaning", in his Mind, Language, and
Reality (Cambridge: Cambridge University Press), pp.215-271.
Putnam, H (1994) "Sense, Nonsense, and the Senses: An Inquiry into the
Powers of the Human Mind", Journal ofPhilosophy 91,445-517.
Soames, S (1998) "Facts, Truth-Conditions, and the Skeptical Solution to
the Rule-Following Paradox", Philosophical Perspectives 12, pp.313-
348.
Wilson, G (1994) "Kripke on Wittgenstein and Normativity", Midwest Studies
in Philosophy XIX, pp.366-390. Reprinted in Miller and Wright
(2002).
Wilson, G (1998) "Semantic Realism and Kripke's Wittgenstein", Philosophy
and Phenomenological Research 58, pp.99-122.
Wittgenstein, L (1974) Philosophical Investigations (Oxford: Basil Black-
well).
V01.20 No.1, pp.35-48.
Philosophy", Midwest Studies in Philosophy XVII, pp. 40-52.
and Object (Oxford: ClarendonPress).
140 ALEXANDER MILLER