Hilary Putnam, Meaning and Reference, 1973



Hilary Putnam, Meaning and Reference, 1973.

( I ) Introductory Problem

Claim One. Meaning is in the Head

1. The doctrine that meanings were concepts carried with it the suggestion that meanings were psychological entities.

2. Frege changes this, and makes meanings “abstract” entities. But still grasping this entity required a psychological act.

3. So for all of these thinkers, “understanding” was a matter of being in a psychological state.

Claim Two. Meaning (intension, here) determines Extension.

1. So, what objects belong to the extension of some term P (or sentence P) is determined by the meaning-intension of P, where “meaning-intension” is the psychological state within the head of the person.

2. It must be the case, then, that if P and Q have different extensions, then P and Q must have different intensions.

Supposition: Both claims cannot be true at once.

( II ) Are Meanings in the Head?

The Twin Earth TE.

Post-1750

Assume that “water” on TE is really XYZ. So if a space ship goes to TE, then ‘water’ has the same meaning on TE and on E.

So it will be reported: “On TE, ‘water’ means XYZ” and similarly going the other way. It will be reported by the TE space traveler that “on E, ‘water’ means H20”.

Note that there is no problem about the extensions being different, because the intensions are different. In fact, we could say that we have two different words, Water-TE and Water-E. Each word has a different intension, and that intension yields a different extension.

Pre-1750

The space travelers (Oscar-1 from E, Oscar-2 from TE) do not know (and cannot figure out) that the stuff on TE and the stuff on E are different. Suppose further that the two of them do not differ in any of their beliefs (and specifically about water).

However, still the extension of water is still what it is, different in each case. But clearly if intension is a psychological state, then the intension of water in both cases must be identical. But then you’d get a falsification of claim (2) above (in section I).

The only way out is to reject (1), that meaning is psychological and in the head.

Objection: Why should we think that the extensions, pre and post 1750, are the same?

Answer: when I say that “this” stuff in my glass is water, then I intend for this stuff to stand in the same-L relation to the other stuff that people in my linguistic community intend when they say “water”.

If so, then for something to be water means it must bear the same-L relation to the stuff in my glass.

What kind of relationship is same-L? It is a theoretical relationship. It may take quite a bit of time to figure out what this stuff is (and thus what is really water and what isn’t).

Thus, the fact that Oscar-1 pre-1750 and Oscar-1 post-1750 would agree that lakes of XYZ is water does not show that the meaning of ‘water’ has changed. What changed was that pre-1750 we would have mistakenly thought that lakes of XYZ bore the same-L relation to the stuff in our glasses on earth.

Aluminum – Molybdenum

The two cannot be distinguished. On TE, “aluminum” pans are made of molybdenum.

Suppose further that the terms “aluminum” and “molybdenum” are switched. So “alum” is the name of moly, and “moly” is the name of aluminum.

So, when Oscar-1 visits, he will not suspect that the pots are not made of aluminum, especially when the TE’ers say that they are.

Further, a segment of the communities involved on TE and E can tell that one is not the other, whereas this is not true for water.

Given that we are talking about people who are not metalurgically sophisticated, the extension of ‘aluminum’ is aluminum for Oscar-1 and the extension of ‘aluminum’ is molybdenum for Oscar-2.

Isn’t it true here that they mean different things by these terms, even though their psychological states are identical?

( III ) A Sociolinguistic Hypothesis

There is a sociological division of labor when it comes to language. Use the example of gold.

Gold has a great many uses in our society, it plays a number of important sociological functions.

Think of the community as a factory: some wear gold, some sell gold, some can tell what is gold and what is not gold.

It is not efficient for everyone to take on these jobs. Wearing gold shouldn’t require that you can tell what gold is.

Point: everyone who participates in the rituals involving gold must acquire “gold” as a term in their language, but not everyone must acquire the method of ascertaining what is gold and what is not gold.

The meaning of “gold” is possessed by everyone because words have social meanings. So the fact that the true extension of ‘water’ is known (and discernable) only to the experts doesn’t mean that only they know what ‘water’ means. Rather, the other people do too, even if they can’t discern the difference.

There are two kinds of tool in the world

1. The tool used by one person.

2. The tool that comes into existence through the cooperation of a community.

( IV ) Indexicality and Rigidity

Let W1 and W2 be two possible worlds in which I exist. Now let us assume that I hold up a glass on W1 and say “this is water” and the glass is full of H20. In W2, the glass is full of XYZ. Now presume that the stuff typically called water on W1 is H20 and the stuff typically called water on W2 is XYZ.

There are now two theories of meaning that one might have concerning the meaning of water.

1. ‘water’ is world-relative but constant in meaning. “water” means the same thing in W1 and W2, but in W1 water is H20 and in W2 it is XYZ.

2. ‘water’ is H20 in all worlds. So the stuff called ‘water’ in W2 isn’t really water. So ‘water’ doesn’t have the same meaning in the two worlds.

If what has been said earlier is right, then (2) must be the right theory of meaning.

The difference between the two, written out more complexly:

(1’) (for every W) (for every x in W) (x is water iff x bears same-L relation to the stuff referred to as ‘this’ in W)

(2’) (for every W) (for every x in W) (x is water iff x bears same-L relation to the stuff referred to as ‘this’ in W1)

There is a scope difference here. The “this” differs in the two constructions; in (1’) it is relative to the world, in (2’) it is independent of this and refers to W1.

Kripke calls this a rigid designator.

If Kripke is right, then when I say ‘this is water’ then I mean (2’) and not (1’). If so, then designation is rigid.

How Rigid Designation has Consequences for Necessary Truth

1. Cross-World Relation. R is a cross-world-relation iff its extension is a set of ordered pairs of individuals not all in the same possible world. So, “taller than”. If x is 5ft in W1 and y is 4ft in W2, then (x,y) is an ordered pair in the cross-world relation. More specifically, it is (x-in-W1, y-in-W2).

2. Now understand same-L as a cross-world relation.

3. So now we can see the theory of meaning we are proposing is such that an entity x (in some W) bears the same-L to some other liquid in another W iff they share the same properties.

4. So we see that our theory of meaning for ‘water’ holds this.

So then what’s the function of intension? Clearly it does not pick out extension, so what does it do? It functions as the way of identifying “this” in the original ostension.

A result of this is: nothing is a possible world in which water is not H20.

1. It is epistemically possible that our claim that water is H20 isn’t true.

2. It is not metaphysically possible that this turn out to be false.

Putnam’s point: words are rigid designators is the same thing as saying that natural kind terms are really indexical (like “I” or “here”).

So, in conclusion,

1. Society contributes to the meaning of a term

2. The world contributes to the meaning of a term

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download