Putnam's Twin Earth, meaning and reference

Philosophy of Language
1

One of Hilary Putnam’s contributions to philosophy is the idea of semantic externalism, which rests on his famous Twin Earth thought experiment.

Summarized by Wikipedia:

Elsewhere in the universe there is a planet exactly like Earth in virtually all aspects, which can be referred to as “Twin Earth”. The relevant surroundings are exactly the same as for Earth; it revolves around a star that appears to be exactly like the Sun, and so on. On Twin Earth, there is a Twin equivalent of every person and thing here on Earth. The one difference between the two planets is that there is no water on Twin Earth. In its place there is a liquid that is superficially identical, but is chemically different, being composed not of H2O, but rather of some more complicated formula which can be abbreviated as “XYZ”. The Twin Earthlings who refer to their language as “English” call XYZ “water”. Finally, the date of the thought experiment is several centuries ago, when the residents of Earth and Twin Earth would have no means of knowing that the liquids they called “water” were H2O and XYZ respectively. The experience of people on Earth with water and that of those on Twin Earth with XYZ would be identical.

Now the question arises: when an Earthling (or Oscar for simplicity’s sake) and his twin on Twin Earth say ‘water’, do they mean the same thing? The twin is also called ‘Oscar’ on his own planet, and the inhabitants of that planet call their planet ‘Earth’. They are in identical psychological states, with the same thoughts, feelings, etc. Yet, at least according to Putnam, when Oscar says ‘water’, the term refers to H2O, whereas when Twin Oscar says ‘water’ it refers to XYZ. The result of this is that the contents of a person’s brain are not sufficient to determine the reference of terms they use, as observers must also examine the causal history that led to this individual acquiring the term. (Oscar, for instance, learned the word ‘water’ in a world filled with H2O, whereas Twin Oscar learned ‘water’ in a world filled with XYZ.)

I can see the point, yet this feels very dubious to me. Meaning might refer to water, yet not incorporate all the unknown qualities of water.

I met Sam last week, and say “Hey Sam”. My meaning picks out the individual, Sam. But it does not pick out Sam’s entire life history, his financial worries, his feelings about his mother, or his tattoo of a penguin on his ass. The meaning seems to be a raw designation of Sam, stopping short of his qualities.

An ancient Greek can designate water, without incorporating the fact that water is H2O into her designation. Rather, she might incorporate the fact that water is Poseidon’s piss. She picks out the same stuff, but attributes a different meaning to it.

The Oscars, it seems to me, designate different stuff, but pick it out in an identical way, and assign identical meanings to it. Suppose I walk into a coin shop, point to a presumed gold coin, and say, “I want this gold coin”. Would my meaning change if it turned out the coin was counterfeit? I don’t think so. It might make the gold attribution in my statement false. But meaning sets out truth conditions, not truth.

What do you think?

2

The Wikipedia article is imprecise. The actual article says:

Thus the extension of the term ‘water’ (and, in fact, its “meaning” in the intuitive preanalytical usage of that term) is not a function of the psychological state of the speaker by itself.

…

Thus the theory that (1) words have “intensions,” which are something like concepts associated with the words by speakers; and (2) intension determines extension — cannot be true of natural-kind words like ‘water’.

Your comments seem to be asserting that words have intensions, and that causal history is does not determine intensions, but as best as I can tell Putnam doesn’t say otherwise; he only argues that it is causal history and not intensions that determine extensions.

3

I hate the Twin Earth “thought experiment” because its conceptual impossibility renders it incoherent, at least IMO. So I’ll ignore it and focus on your real-life query.

Shouldn’t the question be about walking into a coin shop and pointing to what you believe is a gold coin? That way, it can really turn out to be counterfeit. And in that case no, your meaning would not change, since what you’d actually have meant was, “I want this coin which I believe to be gold.” That would be totally pedantic to say, of course, but it makes a difference.

4

This isn’t really my objection. It is more that the concept of extension, as he uses it, doesn’t align with how meaning works.

Extension, or reference, while it might be about the world, doesn’t exhaust all the world’s qualities. It picks out the parts relevant to discourse. H20 and XYZ are not relevant to discourse, at least before their discovery, and so they don’t belong to the extension.

5

Yes, I should have made that clear, I edited the op.

Can’t every use of a kind term be prefixed with an implicit (presumptive)?

But, this doesn’t seem to address Putnam’s argument. He would say that the extension of “presumptive gold coin” in this case is either a real or counterfeit. So the meaning of the utterance, which includes the extension, would be different in the two cases.

Why do you think it is conceptually impossible and incoherent?

6

If you ask for a good coin and are given one of fake gold, you have not been given the gold coin if there is a public use of “gold” not being complied with.

If two distinct chemical compositions of jade exist, you will properly possess a piece of jade if you have either of those gems.

The question is what these words mean, and if their meaning is associated with look over chemical composition, then that’s what it means, and not if not.

If xyz water and h2o water are indistinct to the speakers, they are both “water.” How the term Water is used is what water means. If later we learn of the difference in h20 and xyz, and such a distinction matters to us that we wish to preserve it, then what “water” means will be attached to that use and h20 will remain water and xyz won’t.

7

Forget words for the moment. If you point to water then you’re pointing to H20, and you’re pointing to H20 even if you don’t know that it’s H20. Do you at least accept this?

Now with respect to reference and extension (rather than sense and intension), the word “water” is doing exactly what pointing to water is doing.

8

I’m not sure. Certainly it can in ordinary discourse. Otherwise, as I said, we’d be unnecessarily pedantic.

Yes, that’s what Putnam would argue. The extension of the coin doesn’t change depending on whether it’s real or counterfeit. But the speaker’s meaning is not obvious; we’d have to spell it out in order to decide if the extension tracks it or not.

We can abandon Twin Earth and just ask, When two people have the same name for two very similar-appearing but in fact different things, are they referring to the same thing? We can grant that they think they are. And I would answer: So it depends on your theory of meaning. In most ordinary cases we’d probably say, No, they only think they are; they’re pointing to two things that look very much alike; but it turns out they’re different. Another reason to avoid Twin Earth: Oscar and his twin will never be able to learn this, which kind of gums up the works.

Another way of highlighting that this all depends on one’s theory of meaning – or perhaps on one’s practice when it comes to meaning.

Two different planets/universes with exact counterparts for everything except what we call “water”? I rest my case. Show me the science and I’ll entertain the possibility!

9

I accept it for physical pointing, the kind of pointing a statue can do. But I don’t necessarily accept it for intentional pointing. Unlike a statue’s outstretched arm, intentional pointing is a communicative act, and I’m doubting that the meaning of this act includes h2o when the chemical formula is not even known.

10

It’s tricky. You are asking “are they referring to the same thing” from an epistemic perspective whereby you know they are different. So it is natural for you to say they are referring to different things. That is just a fact. The real question, though, is do they mean the same thing thing when they use the word.

If they do learn the difference, then from that moment on it is clear they mean different things by “water”. In practice they might use different words like “water” and “twin water”. The real question is do they mean different things before they learn the difference.

Obviously it is scientifically impossible. But it seems logically possible. If I have an issue with it, it is that it relies on intuition in a scenario far from the situations where our intuitions developed.

11

Isn’t there some sense that the “internal meaning”, if I am using the appropriate phrase or concept, in people’s heads is actually about external things. When you define your internal meaning about water, it is because you have gone about and interacted with the world and observed properties or associations that you think of as water. And your internal meaning for water can’t be entirely private if it is shaped by a community you are interacting with, doing stuff and using the concept of water in the context of. You might have a whole bunch of associations characterising water and parts of this might be overlapping with or shared by other concepts too (e.g. water is a liquid like ketchup, or transparent like glass, both of which are talking about physical things). So what is really the difference between H2O and the other associations that characterize internal meaning of water? Some of the other associations we have regarding water might be a lot less physically precise than H2O. People may use the term “water” without knowing it is H2O, but they also may not know many other properties that is associated with what they are calling “water”, at least in so far as they have been interacting with it so far. The question is how you change your use of the word “water” when you update your web of beliefs with encountering new instances of things that may or may not have properties of “water” in different ways. Do you choose to subsume non-H2O things into your definition even if all your past instances were used in the context of “H2O”? Surely, you can do this. You might even decide on sub-categories of “H2O-water” a “non-H2O” water. But this change of word use doesn’t change the rest of the associations surrounding them, it doesn’t change the fact that before now all your interactions with water were with H2O and what you were talking about before as “water” was actually H2O. Nor would it change any other associations in your web-of meaning that you may not know about but would be able to if you did certain types of observations or experiments.

12

This is a good example because Putnam’s account can be understood as an extension of Krike’s account of the semantics of proper names, which Putnam applied to natural kinds (e.g. chemical elements or living species). In both cases, the account is meant to remedy the defects of descriptive accounts of the semantics of proper names (or of designators of natural kinds).

So, in the case of Sam, the idea is that even though you may not know of their whole life history, if you are to refer to Sam not merely as whoever stands before you and happens to have this name, but as the unique individual other people than yourself know under that name, what it is that the use of this name is meant to pick up (i.e. the reference of the name) is dependent on the history of the use of that name, including the fact that the individual so christened and the individual now standing in front of you are the same individual.

The classic example that had been used by Kripke was Kurt Gödel. Many people may only know of Gödel as the mathematician who proved the two famous incompleteness theorems. Let us suppose, Kripke suggested (from my loose recollections), that Gödel is actually a plagiarist who stole those proofs from the unpublished work of a now deceased colleague. People who only know Gödel as the (putative) author of the two proofs don’t thereby unwittingly refer to the deceased colleague but still are referring to Kurt Gödel when using the name “Kurt Gödel”. That’s because we understand proper names to properly refer to the individuals those names have initially been brought into use for tracking them irrespective of the possibly false attributes or descriptions that have thereafter been ascribed to them. (Those proper names are rigid designators, in Kripke modal semantics account.) This is something that descriptive accounts of the meaning of proper names was struggling to account for.

Putman was emphasising in his early writings on the semantics of natural kind terms that ancient Greeks already had a sophisticated enough understanding of substances, and of the distinction between pure substances (e.g. water, gold, etc.) and inhomogeneous mixtures and aggregates (e.g. wine or alloys) that they could mean to refer to as “water” (or rather as “ὕδωρ” or “hydor”) to whatever happens to be the same substance as whatever substance is habitually, or had initially, been referred to by that name in their linguistic community. This expectation, and convention, is sufficient for making their natural kind naming practices tracking their referents rigidly (i.e. in all possible worlds) in Kripke’s sense.