Skip to content
Coordinates of ThoughtHome of Philosophy
Join
Debate

Thesis My clients can’t tell walnut from stained beech, but they still mean walnut

Hicham Ouazzani Fellow
· 5 replies
Meaning & languageAnalyticHilary PutnamJerry Fodor#craft-knowledge#division-of-linguistic-labour#externalism#twin-earth
Follow

Where it stands

Half the customers who order furniture from me ask for walnut. Most of them could not pick walnut out of a pile of stained beech if their life depended on it, and some unscrupulous workshops count on that.

A friend lent me Putnam's essay on the meaning of "meaning", and the bit about elms and beeches stopped me. Putnam says he cannot tell the two trees apart, yet his word "elm" refers to elms and "beech" to beeches, because the language has a division of linguistic labour: ordinary speakers use the words, and experts (botanists, foresters, carpenters) are the ones who can actually sort the things.

So here is my thesis. When my customer says "walnut", the word refers to walnut, not to "any dark wood that looks expensive", even though the picture in her head fits both. What fixes the meaning of the word is partly outside her: in the trees, and in people like me. If I sell her stained beech, I have not met her request under a looser meaning. I have cheated her.

If that is right, then meaning is not simply what is in the speaker's head. It is spread across a community and the world. I would like to hear the best case against this, because I suspect philosophers have one.

LikeReply
Follow5 replies

Join the conversation: members reply here, and joining is free. Reading is always free.

Join free to reply Sign in

5 replies

For the thesis 3

  1. Kwame Asante

    For Fellow

    For, with a caution. Where I live, the people who can tell one medicinal plant from its lookalike are herbalists, and the community's words for those plants depend on them in exactly Putnam's way. But who counts as the expert is itself a social decision, and colonial botany sometimes overrode local experts with its own. The division of labour is real; it is also not politically innocent.

    Helpful · 3

Against the thesis 2

  1. Priya Raghavan

    Against Fellow

    The best case against, from someone who builds perception systems for a living: everything that explains your customer's behaviour is in her head. She points at the stained beech, says "lovely walnut", pays. A customer whose word really is linked to beech would do the same. If two people behave identically for identical reasons, a science of the mind has no use for the difference.

    So I would say the reference is a fact about the social bookkeeping, not about her psychology. That is a real fact, but it is not "meaning" in the sense psychologists care about.

    Helpful · 3
    1. Oskar Nyberg

      For Fellow

      Priya's objection is the one Putnam anticipated, and I think it actually supports Hicham. Putnam's claim is about extension: what the word applies to. He happily grants that the two speakers can be in the same psychological state; that was the whole point of Twin Earth. His conclusion is precisely that psychological state does not fix extension. "Meanings just ain't in the head" is the famous line.

      Philosophers who wanted to keep something for psychology, Fodor for a while, talked about "narrow content", the part shared by Twin Earth duplicates. So both of you can be right, about different things. But if we are asking what Hicham's customer ordered, it is the extension that matters, and that is fixed outside her.

  2. Pieter van Dijkhuizen

    Against Fellow

    I'll push from the other side. The customer's head contains more than a picture of dark wood. It contains a disposition to defer: "walnut is whatever the experts call walnut". That disposition is internal, and it does all the work Hicham attributes to the community. In machine learning terms, the model doesn't need the label to be in its weights, only a pointer to where the labels live.

    Helpful · 1
    1. Hicham Ouazzani

      For Fellow

      Pieter, a disposition to defer is not a pointer to walnut. It is a pointer to me, and I can be wrong, or crooked. What makes my judgement right or wrong is the tree. So the chain still ends outside every head, including the expert's.

      But I concede Priya's point properly. I was mixing two questions. Her psychology can be the same either way; what she is owed is different. I should have said the meaning that settles what was ordered is social and worldly, not that her mind is somehow larger than her skull.

      Helpful · 2