Putnam’s argument is based on a couple of claims about words, pictures or mental images, and even thoughts or intentions link to things. Negatively, resemblance is not sufficient (nor is it necessary) for reference / aboutness / intentionality. Positively, one needs to do some further work and Putnam suggests standing in the right sort of causal relation.
Using this background, the argument runs as follows:
Words/thoughts can only refer to things that stand in the right sort of causal relation to them.
So ‘vats’ stand either for real vats or vats-in-the-image (ie computer generated).
If we are not brains in a vat, then the “we are brains in a vat” refers to real vats and is false.
If we are brains in a vat then “we are brains in a vat” refers to vats-in-the-image and is false.
So “we are brains in a vat” is never true, is always false.
Putnam himself summarises his argument thus:
It follows that if their ‘possible world’ is really the actual one, and we are really the brains in a vat, then what we now mean by ‘we are brains in a vat’ is that we are brains in a vat in the image or something of that kind (if we mean anything at all). But 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. [Putnam 1981]
But note that this argument stops short. What we really want is the step from
The sentence “we are brains in a vat” is (always) false.
to
It is false that we are brains in a vat; or, we are not brains in a vat.
This is normally a simple matter. It is called 'disquotation' and is exemplified in what are called instances of the T-schema (see Tarski's account of truth and Davidson's account of language for their wider use)
“snow is white” is true iff snow is white
This equivalence says that to say of the sentence mentioned on the left (the sentence: “snow is white”) that it is true, is just to say that snow is white (using, not mentioning, the contained sentence). (Strictly the connective iff only says that the two sides share a truth value. But the disquotational use within a language allows for my stronger claim.)
So if that is so simple, why does it not work for Putnam?
We can get a feel for the fact that it does not work (if not why it does not work) from the fact that we can construct a Putnam-style argument for dreaming. Assuming that dreaming is not the same as dreaming of dreaming, then by the same argument.
“we are dreaming” is never true.
That may be so, but still, sometimes we think we are awake when we are dreaming!
The reason is this. Putnam’s argument works by offering a non-disquotational account of the truth and reference of the language of those in the vat.
For brain in a vat English, ‘vat’ refers to vats in the image.
So we cannot simply disquote because we do not know which language we are speaking.
Cf this potential very quick Putnamesque argument against scepticism (from the Stanford Encyclopaedia entry):
A. If I am a brain in a vat, then my word ‘tree’ does not refer to trees.
B. My word ‘tree’ refers to trees. So,
C. I am not a brain in a vat. [(A),(B)]
This just goes to quickly. It highlights disagreement with the sceptic about B.
Note, however, the odd nature of the scepticism we end up with: we do not know either the meaning of our own words, nor the content of our own thoughts (which are true of either vats or vats-in-the-image). Weird!
(PS in response to an email:
You cannot use a Putnam style argument in the case of scepticism about the senses because - in general - one stands in the right sort of relation to pencils etc to refer to them even if one has false beliefs about them right now because, in water, they look bent etc. So that may sound good for scepticism.
But scepticism based on the fallibility of senses doesn’t seem good for a different reason (which has nothing to do with a Putnam style argument). If my senses may let me down, why don’t I simply take another look, varying the lighting or whatever? Within a dream. or as a brain in a vat, no amount of taking another look helps because everything in my experience in both of those cases is bogus. But in the real world with merely fallible senses (ie senses that sometimes let me down), taking another look can help because my senses can in general reveal the world. The sceptic will have to do more to persuade us that our senses may always be wrong. And without that we just have everyday fallibility not scepticism.
Another email: "The premiss you give for Putman's argument doesn't suggest why the computer generated images of a vat cannot be based on real vats. Thinking in terms of the Matrix, a human 'wakes up' seeing the pod he is in and all the other pods around him. But is it not entirely possible for him to have been shown this image of the pods via computer generation (prior to 'waking up'), which in itself would have been based on the real pods?"
Note that Putnam himself tries to rule this out by saying:
"To fix our ideas, let us specify that the automatic machinery is supposed to have come into existence by some kind of cosmic chance or coincidence (or, perhaps, to have always existed). In this hypothetical world, the automatic machinery itself is supposed to have no intelligent creator-designers. In fact, as we said at the beginning of this chapter, we may imagine that all sentient beings (however minimal their sentience) are inside the vat."
So the images given to the brains in a vat have no causal connection back to real trees or real vats etc. But it is an interesting question in any case and prompts two areas where it would be helpful to know more.
Just what exactly is the role of causation in Putnam’s positive gesture to how reference does work. To raise just one complexity which needs to be addressed (and is addressed by those who work on causal theories of reference in the philosophy of content). If my tree thoughts are caused by real trees then they can, indeed, be about trees. But the causal chains reach further back in time and space. The light that reflects off trees to enable me to experience them comes from the sun and is in turn caused by processes that go back to the Big Bang. But my thoughts are not about the Big Bang (but trees). Even if the images in the vat are caused by real trees what makes them the object of thought (like trees in the example) rather than lying behind thought (like the Big Bang).
In the example you describe, would the trees that are connected to the brains in the vat be made available to the brains as part of a world? Would they be like deep sea objects shown via tv cameras attached to a remote controlled submarine which thus open up the under-sea world to the remote operator in Dallas? If so then the brains in a vat are like our own brains which we think of as not in our heads but in a world. (That's obviously not quite right because we are the ones in the world not our brains but I hope the point is clear. If the brains are in causal contact in the right way with the outer world of real vats etc then their environment or their world is that rather than a kind of sceptically-powering illusion. See Dennett's great paper 'Where am I?'.)
)