This article was downloaded by: [New York University]

On: 03 May 2015, At: 08:42

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,
UK

Canadian Journal of Philosophy

Publication details, including instructions for authors
and subscription information:
http://www.tandfonline.com/loi/rcjp20

Could We be Brains in a Vat?

a
PETER SMITH
a
University College of Wales , Aberystwyth
Published online: 01 Jul 2013.

To cite this article: PETER SMITH (1984) Could We be Brains in a Vat?, Canadian Journal
of Philosophy, 14:1, 115-123, DOI: 10.1080/00455091.1984.10716372

To link to this article: http://dx.doi.org/10.1080/00455091.1984.10716372

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the
information (the “Content”) contained in the publications on our platform.
However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness,
or suitability for any purpose of the Content. Any opinions and views
expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the
Content should not be relied upon and should be independently verified with
primary sources of information. Taylor and Francis shall not be liable for any
losses, actions, claims, proceedings, demands, costs, expenses, damages,
and other liabilities whatsoever or howsoever caused arising directly or
indirectly in connection with, in relation to or arising out of the use of the
Content.

This article may be used for research, teaching, and private study purposes.
Any substantial or systematic reproduction, redistribution, reselling, loan,
sub-licensing, systematic supply, or distribution in any form to anyone is
 expressly forbidden. Terms & Conditions of access and use can be found at
http://www.tandfonline.com/page/terms-and-conditions
Downloaded by [New York University] at 08:42 03 May 2015
 CANADIAN JOURNAL OF PHILOSOPHY
Volume XIV, Number 1, March 1984
Downloaded by [New York University] at 08:42 03 May 2015

Could We be Brains in a Vat?

PETER SMITH, University College of Wales, Aberystwyth

The course of my experience is quite consistent with the hypothesis that

it is being produced by a mad scientist who is feeding into my sensory
receptors entirely delusive stimuli. Indeed, I could at this very moment
be nothing more than a brain floating in a vat of nutrients, my nerve
ends linked up to some infernal apparatus by means of which my
unknown deceiver induces in me utterly erroneous beliefs about the
world.
So begins a familiar line of thought which dramatizes an equally
familiar sceptical problem about the relation between our experiences
and the world. However, Hilary Putnam has recently offered a
marvellously ambitious argument which is intended to kill stone-dead
the philosophical fantasy that we might be deluded brains in a vat. 1 And

1 See the opening chapter of Hilary Putnam, Reason, Truth and History (Cam-
bridge: Cambridge University Press 1981). All quotations are from this chapter.

115
 Peter Smith

if there is indeed a simple semantic argument to prove that the fantasy is

necessarily false, as Putnam maintains, then the sceptic is apparently
deprived of at least one way of convincing us that his problem is a ge-
nuine one. It seems that Putnam's desired conclusion is of some impor-
tance. But shouldn't we be suspicious of all attempts to get large
philosophical conclusions from thin semantic premises? I would argue
so: and certaintly in the present case I think such initial suspicion is en-
tirely justified. In this note I try to show that Putnam's argument - in-
genious though it is - falls well short of establishing anything which will
confound the sceptic who plays with the idea that we might be brains in a
Downloaded by [New York University] at 08:42 03 May 2015

vat.
Putnam's key premise is that one's representations (whether physical
or mental)

cannot refer to certain kinds of things, e.g. trees, if one has no causal interac-
tion at all with them, or with things in terms of which they can be described.

This formulation is vague enough to capture whatever truth there is in

causal theories of reference, and for present purposes let us accept it
without further comment. It follows that brains in a vat, or at least those
which completely lack the right sort of causal interactions with the rest of
the world, may employ the representations we use but they 'cannot refer
to what we refer to.' Combine this with the claim that thinking about Xs
involves the use of representations which refer to Xs, and we can con-
clude that brains in a vat cannot even think about what we think about.
In particular, they cannot even think about such things as brains and
vats: if they employ the very sentence We are brains in a vat,' they still
cannot be meaning what we mean - i.e. they cannot use the sentence to
say or think that they are brains in a vat. So, according to Putnam, we
must infer that the brain-in-a-vat's thought 'We are brains in a vat' is
either contentless or at best means something like we are brains in a vat
in the image. But now, in the light of all this, let's consider the suggestion
that we are brains in a vat:

... part of the hypothesis that we are brains in a vat is that we aren't brains in a
vat in the image (i.e. what we are 'hallucinating' isn't that we are brains in a
vat). So, if we are brains in a vat, then the sentence 'We are brains in a vat' says
something false (if it says anything). In short, if we are brains in a vat, then We
are brains in a vat' is false. So it is (necessarily) false.

This certainly has a pleasing superficial neatness; but the surface ap-
pearance of cogency is deceptive.

116
 Could We be Brains in a Vat?

Obviously, it cannot be a condition of our genuinely referring to a

given object that we are presently having some causal interaction with it.
It seems quite sufficient that we have had some appropriate interaction in
the past. So Putnam's premise about reference cannot be used to rule out
the fantasy that we are now reduced to being brains in a vat, who still
have genuine though deluded thoughts about the world in virtue of our
past causal contacts with it. 2 One might well think that conceding the
coherence of this fantasy is already enough to allow a sceptic to go about
his corrosive work (for how do I know that I haven't been a brain in a vat
subject to totally delusive experiences for the last ten minutes, ten days,
Downloaded by [New York University] at 08:42 03 May 2015

ten years? is it much comfort to be told that my experience can't have

been delusive all the way back?). But let's not dwell on this point. For
Putnam makes it quite clear that his argument is intended only to rule
out the extreme speculation that we might be permanently hallucinated
brains forever confined to a vat. Does he succeed in showing that at least
this more extreme hypothesis is necessarily false?
We can restate the first stage of Putnam's argument as follows:

(P) If there are within our world of trees and books, laboratories
and scientists, some brains in a vat which are kept permanently
hallucinated, then even if they happen to produce a mental
token of the sentence 'We are brains in a vat' they cannot
thereby be referring to brains or vats - and so they cannot be
thinking what we in fact think using those words.

This, to repeat, we are assuming to be correct; but what can we infer

from it about the supposition that we are brains in a vat? As a first move,
we might try inferring this:

(A) Suppose, counterfactually, that we had happened to be per-

manently hallucinated brains in a vat within our present world
of trees etc.; in that case, even if we had tokened the sentence
'We are brains in a vat' we would not thereby have been think-
ing what we in fact think when we use those words in the world
as it is.

2 Cf. Daniel Dennett's fantasy in his superb entertainment 'Where am I?' in his
Brainstorms (Montgomery, VT: Bradford Books 1978).

117
 Peter Smith

Now, there does appear to be a problem about the coherence of the

counterfactual supposition here - for given that we are not brains in a
vat but normal human beings, could beings permanently in that predica-
ment still be us? However, let's simply ignore that puzzle about the con-
ditions for trans-world identifications and suppose that (A) is as accep-
table as (P). For the essential point to make here is that nothing at all
follows from the counterfactual (A) about the truth or falsity of the claim
that we are indeed brains in a vat.
This is important, so let's try to get the point clear. What (A) asserts is
that, had things been different from how in fact they are, then we could
Downloaded by [New York University] at 08:42 03 May 2015

not have expressed the thought that we are brains in a vat by using the
sentence 'We are brains in a vat.' And to continue in Putnam's vein, the
most we might then have meant, by our present interpretative lights, is
something like we are brains in a vat in the image. 3 This implies that if we
had been brains in a vat, then the sentence 'We are brains in a vat' would
have been used by us to say something false (if it said anything at all). So
far, so good. But can we get from this last counterfactual to Putnam's
desired conclusion that if we are brains in a vat then 'We are brains in a
vat' now expresses something false and so (necessarily) we are not brains
in a vat?
It seems quite plain that this kind of move from a counterfactual to
the corresponding indicative conditional is quite illegitimate for it
licenses a variety of completely unacceptable conclusions. For example,
if we had lived on the waterless Twin Earth, where as is well known
'water' refers not to H 2 0 but to the superficially similar XYZ, then the
sentence 'We live on a waterless planet' would have been used by us to
express a falsehood. So (moving again from the counterfactual to the in-
dicative conditional), if we live on a waterless planet like Twin Earth,
then the sentence 'We live on a waterless planet' now expresses something
false - so (necessarily) we live on a planet with water. But this argument
is plainly absurd. On any sane view, the presence of water in the world is
just not the kind of fact that can be established by this sort of a priori
reasoning.
It is easy enough to see what goes wrong in the spurious 'water' argu-
ment. To put it schematically, a contrast is postulated to hold between
Twin Earth and our own world in virtue of which we can say that a cer-

3 To keep things simple, I will adopt Putnam's phrasing here: if you think (correct-
ly in my view) that there are further problems lurking here, then so much the
worse for Putnam's original argument.

118
 Could We be Brains in a Vat?

tain sentence ('We live on a waterless planet') that would have been true
if the Twin Earthers had used it with the references we give the terms in
the sentence, is in fact used by them to say something false. Obviously, if
we go on to consider the possibility that our world is Twin Earth, we
cannot coherently retain the postulate of a contrast which can be used to
show that the sentence is false as used by Twin Earthers; so we cannot in-
fer that we (if we are Twin Earthers) use it to say something false.
Similarly, a contrast is postulated in (A) to hold between brains in a vat
and ourselves as we in fact are, in virtue of which we can say that a cer-
tain sentence (We are brains in a vat') that would have been true if the
Downloaded by [New York University] at 08:42 03 May 2015

brains in a vat had used it with the references we give the terms in the
sentence, is in fact used by them to say something false. Again it is ob-
vious that if we go on to consider the possibility that we are brains in a
vat, we cannot coherently retain the postulate of a contrast which can be
used to show that the sentence is false as used by brains in a vat; so we
cannot infer that we (if we are brains in a vat) use it to say something
false.
In short, then, the counterfactual claim (A) cannot be directly used to
show that we are not brains in a vat. But can we move from Putnam's
starting point (P) to his desired conclusion without going via (A)?
If we are going to use (P) in Putnam's manner to show that we can't be
brains in a vat, then we need to put ourselves (so to speak) in the place of
the brains in a vat mentioned in (P) and derive a contradiction. We have
just seen that we don't get far by inserting ourselves into the story in an
explicitly counterfactual way. And the same reasoning also shows that
we can hardly identify ourselves straight out (in a non-counterfactual
mode) with the brains in a vat mentioned in (P). For (P) is initially
asserted from a point of view external to the brains in a vat and presup-
posing a contrast between ourselves and the poor deluded brains. How
could we coherently go on to reinsert ourselves into the described scene
as the brains in a vat? To try to do so while still appealing to (P) would
simply be to adopt a contradictory view as to where we are supposed to
be in the story; and so the fact that we could go on to derive further con-
tradictions would be of no significance at all.
To reinforce this point, let's pause very briefly to examine what hap-
pens if we do bulldoze ahead and just identify ourselves with the brains
in a vat mentioned in (P). We get the following:

(B) If we are brains in a vat kept permanently deluded within our

world of trees and books, laboratories and scientists, then even
if we happen to produce a mental token of the sentence We are

119
 Peter Smith

brains in a vat' we cannot thereby be referring to brains or vats

- and so we cannot be thinking what we in fact think using
these words.

This certainly seems to end up in absurdity: so can we deduce that its in-
itial supposition must be false? Well, note that the most that this would
show is that we are not permanently deluded brains in a vat placed
within the given everyday world of trees, books and so on. And that, of
course, was never in question. As Putnam himself emphasizes, whatever
the content of the speculation that we might be brains in a vat, it is surely
Downloaded by [New York University] at 08:42 03 May 2015

not intended to suggest that we might be a brain in a vat 'in the image.'
On the contrary, we can all agree that, while our given world might con-
tain deluded brains in a vat, we are certainly not there among them. The
issue that the brain-in-a-vat speculation is supposed to raise is whether
our given world as a whole is delusory: it is not intended to advance a
bizarre hypothosis about where we are within the given world.
So, identifying ourselves with the brains in a vat in (P) - whether
counterfactually or otherwise - cannot produce Putnam's anti-sceptical
conclusions. So what is going on? Putnam's argument which looked so
brisk and neat now appears to be a confused failure.
As we noted before, the difficulty of applying (P) to the supposition
that we are brains in a vat is that this would involve us in adopting an
unstable view as to where we are in the story. Still, perhaps we can
reconstruct something like the intended argument for Putnam's conclu-
sion, using the ingredients which he provides, while sticking to a single
unshifting perspective on our place in the world. Exercising a fair
measure of interpretative charity, let's briefly explore this possibility.
Suppose that (for whatever reason) you have become convinced that
you are intermittently being deceived about the world by the machina-
tions of the proverbial mad scientist who interferes with your brain -
call him The Evil Demon.' In particular, you believe that all your seem-
ing encounters with cats past and present have been due entirely to his
deceptions and not to causal interactions with real feline beasts. Then
assuming a minimal causal theory of reference, you will conclude that
the term 'cats' does not really refer to cats: or rather less misleadingly,
you will hold that the term simply lacks genuine reference. As a conse-
quence, you will withdraw assent from the claims about cats which you
previously accepted. All this is quite coherent. Moreover, you can ex-
plain to yourself why your new convictions are coherent. By rehearsing
your beliefs about the machinations of the Evil Demon, you can see how
your delusions came about, and give sense to your revised views.

120
 Could We be Brains in a Vat?

Now, let's ask what sort of limits there are on your thus 'thinking
away' the familiar external world. In particular, could you suppose that
the Evil Demon had been deceiving you about the existence of brains?
Suppose you attempt to think this. Then you will withdraw assent from
the claims about brains which you previously accepted. But could you
now make sense of your new convictions? In this case, you could hardly
rehearse your old thoughts about how The Evil Demon is tampering with
your brain, for these thoughts are now to be rejected along with all other
claims about brains. So it seems that here we run up against a limit on
what we can intelligibly think to ourselves about our predicament as
Downloaded by [New York University] at 08:42 03 May 2015

brains (whether situated in a vat or otherwise) which are subjected to

delusory sensory inputs by a mad scientist.
In short, then, it seems that making sense of the fantasy that we are
permanently hallucinated brains in a vat requires us to preserve the
thought that we really are brains. But surely if we regard ourselves as
permanently hallucinated, we should withdraw assent from all proposi-
tions about the delusory world, including propositions about brains and
vats. So it seems that the supposition that we are brains in a vat both re-
quires assent and dissent from the thought that we are brains. Now, does
this reasoning at last give Putnam what he wants?
Surely not; for if this is Putnam's underlying reasoning, then it seems
that the conclusion it supports is of very limited interest after all. Let's
grant that if we think we are quite radically deceived then we will hold
that we are deceived in particular about brains and vats - so the sup-
position that we might be brains in a vat of a familiar this-worldly kind
will be undermined along with all our other suppositions about this-
worldly things. So let's allow that, on a rather literal-minded understand-
ing, the claim 'We are permanently hallucinated brains in a vat' has to
be rejected as self-defeating, and the speculations of my opening
paragraph above have to be abandoned. But surely, it would be a
peculiarly unimaginative sceptic who got himself completely stuck in this
sort of tangle by simultaneously asserting and throwing doubt on the ex-
istence of brains in a vat. Surely - to repeat a point that's already been
made - the speculation that might be epitomised by the slogan 'We
could be brains in a vat' is not really intended to commit us to the
thought that we are perhaps in a vat of a this-worldly kind. The specula-
tion is rather that our relation to the real world could be that which is ex-
emplified in the image by what we call hallucinated brains in a vat on the
one side, and the world in the image on the other side. Just as fictional
characters in a novel can exemplify the possibility of a certain kind of
relationship, so, within the image, brains in a vat and the rest of the

121
 Peter Smith

world can exemplify the possibility of a certain kind of person/world

relation which we might call the delusive relation. The claim dramatical-
ly phrased as 'We could, for all we know, be hallucinated brains in a vat'
is more soberly put: our relation to the world could be the delusive one.
To make sense of the sober version, we need brains and vats in the image
(or, as Putnam loosely puts it, 'things in terms of which they can be
described'), and we need some formal concepts - for example, the idea
of a relation between our thoughts and the world. And it seems that the
preconditions of giving sense to the sober version can be fulfilled without
thereby making it false. 4
Downloaded by [New York University] at 08:42 03 May 2015

Now this appearance may, of course, be entirely deceptive. But my

claim is that nothing in Putnam's present argument shows that the sober
reading of the claim 'We could be hallucinated brains in a vat' must be
false. For this key premise, to repeat, is that for thoughts involving terms
like 'cat' or 'brain' or 'vat' to be true of the world, our use of the terms
must be causally linked in an appropriate way with the world. Plainly
one can accept this claim about the causal preconditions of reference for
terms without thereby committing oneself to any view about the precon-
ditions of use for formal concepts such as that of a person/world rela-
tion.
In short, as far as Putnam's considerations concerning reference go,
our relationship to what there really is could be that exemplified (in the
given world as we seem to experience it) between hallucinated brains in a
vat and the rest. No doubt, Putnam would protest that the notion of
'what there really is' deployed here smacks of the unacceptable sort of
transcendental realism against which he has argued so vigorously
elsewhere. 5 Perhaps so: but rejecting this notion will require entirely dif-

4 Those with whom I have discussed Putnam's argument seem equally divided into
two camps - one holding that Putnam is obviously right, the other that he is ob-
viously wrong. I can't agree that the matter is obvious either way, for (as I have
suggested above) it is not clear how Putnam's argument is properly to be
understood. But I can now offer a diagnosis of the dissent: Those who side with
Putnam have seen that he points to reasons for thinking that the speculation that
we are radically deluded brains in this-worldly vats is self-defeating. Those who
side against Putnam have seen that the role of fantasies about brains-in-vats is to
provide a model of an epistemic predicament, and it cannot be ruled out that we
are in such a predicament merely by showing that we are not actually brains-in-
vats of a this-worldly sort.
5 See e.g. Hilary Putnam 'Realism and Reason' in his Meaning and the Moral
Sciences (London: Routledge and Kegan Paul1978).

122
 Could We be Brains in a Vat?

ferent arguments to those we sketched at the outset. Certainly, the

brains in a vat' argument does not supply independent support for there-
jection of realism. Putnam's premises about reference leave it open that
our relation to the world could be the delusive one - or, in a phrase, we
could be brains in a vat.

August 1982
Downloaded by [New York University] at 08:42 03 May 2015

123