Putnam's Argument

Philosopher Hilary Putnam's "elm/beech" example appears in his 1975 paper “The Meaning of 'Meaning'”, where it is the basis of one of his arguments in defence of semantic externalism. Semantic externalism is the view that the meaning of a term is determined (partly or entirely) by factors external to the person using it. This in opposition to semantic internalism, the view that a term's meaning is identical to the mental state associated with the thing to which that term refers. For Putnam's stronger and more influential externalist argument, see the Twin Earth thought experiment. 

Putnam's elm/beech argument is fairly simple and goes something like this:

  1. I cannot distinguish between an elm tree and a beech tree. They are more or less the same concept for me, or, in other words, share an intension.

  2. If semantic internalism is right and meaning is 'just in the head', then intension determines extension. My uses of "elm" and "beech", psychologically identical, must also share an extension (i.e., refer to the same thing).

  3. However, if I call a beech tree an elm, I am clearly wrong.

  4. So it is possible for terms to be affiliated with the same psychological, 'intensional' state while having different extensions.

  5. Therefore, semantic internalism is wrong — the meanings of words are not just 'in my head', but must be (at least in part) determined by external factors.

Searle's Reply

Internalist philosopher John Searle attacked this example's logical inconsistency in his 1983 Intentionality: An Essay in the Philosophy of Mind, specifically in the chapter entitled "Are Meanings in the Head?". Searle's claim is that Putnam's argument relies on two contradictory beliefs:

  1. My concept of "elm" = my concept of "beech".
  2. The extension of "elm" in my idiolect ≠ the extension of "beech" in my idiolect.

According to Searle, it is impossible to hold these two beliefs at the same time. This is because knowledge of proposition (2) relies upon also knowing that elms are not beeches and beeches are not elms, conceptual knowledge that itself must be accounted for. However, to actually account for it 'breaks' the argument, as proposition (1) cannot exist simultaneously with my conceptual knowledge that they are not equivalent terms.