Monday, November 17, 2014

Meaning and the experts
(answer to question in previous post)
In 1976, when I delivered the John Locke Lectures at Oxford, I often spent time with Peter Strawson, and one day at lunch he made a remark I have never been able to forget. He said, "Surely half the pleasure of life is sardonic comment on the passing show".  This blog is devoted to comments, not all of them sardonic, on the passing philosophical show.
Hilary Putnam

Although the semantic externalism I defended in “The Meaning of ‘Meaning’” is relevant to speaker’s meaning as well, the explicandum of that essay was meaning in a language. I know that there are those (Noam Chomsky and, at one period, Donald Davidson) who have denied the very existence of “languages”, but  I cannot take this language skepticism seriously. If the fact that the boundaries between languages are somewhat vague and, moreover, depend on social conventions, means that languages do not exist, then, by the same token, cities do not exist either[1], but I am not a city skeptic either. More seriously, if social entities do not exist then there is no subject matter for any of the social sciences, and it is sheer chutzpah on the part of even a great linguist or philosopher of language to claim such a thing.
Not only do I believe that there are such things as languages, I believe that there are things that speakers of a language are expected to know (both in the sense of tacit “know-that” and in the sense of “know-how”) qua competent speakers of that language. “The Meaning of ‘Meaning’” enumerated a number of those things. For example, competent speakers of English are expected to know that tigers are stereotypically–not necessarily actually—striped. But, and this is where “externalism” comes in, knowing, all those things is not necessarily speaking English‑it could be speaking Twinglish, the language called “English” on Twin Earth! Meaning has another dimension, apart from the dimension of speakers’ tacit knowledge. Words like “water” and “gold” have their reference fixed partly by the actual nature of what are taken by speakers to be paradigm cases and partly by the division of linguistic labor, which allows a subset of the speakers, the relevant experts, authority over such questions as whether a liquid is really water or a metal is really gold.  But the knowledge of the experts is not knowledge it is linguistically obligatory for a speaker to have. That is why I resist saying that the experts have a better grasp of the meaning of “water” or “gold”. Nevertheless, the extension is a component of the meaning of those words, but not in the sense that speakers or even experts have to possess a description of the meaning that would single out the right liquid or substance or whatever on an arbitrary planet. (Before the development of modern chemistry, the experts might very well be using false theories, whose falsity doesn’t matter in the Earth environment but would become disastrous on Twin Earth). What speakers have to be causally linked to is the correct extension, not the correct description  of the extension. And extensions, as opposed to descriptions of extensions, aren’t things we grasp with our minds; they are out there in the world.
All this is summed up in the “normal form description for ‘water’ on page 269 of my (1975) Mind, Language and Reality.  A brief history of how I came to all this is contained my Schock Prize Lecture, “The Development of Externalist Semantics.” Theoria 79.3 (2013), 192-203.

[1] In “Explaining Language Use,” Philosophical Topics, vol. 20. no. 1, the philosophy of hilary putnam (205-232) [collected in Chomsky, New Horizons in the Study of Language and Mind (Cambridge, 2000)], Chomsky wrote, “Such terms as London are used to talk about the actual world, but there neither are nor are believed to be things in the world with the properties of the intricate modes of reference that a city name encapsulates.” (See my Reply (379-385) in the same issue of Philosophical Topics.)


  1. Gurudev, thank you very much for giving me an impressive answer to my query. Like you, I also do not believe that meanings depend on beliefs of individuals. The picture of communication is actually raised in the grasp of meaning but not in any platonic sense. But I believe that in communication system, linguistic sense of using words (to the extent meaning) is causally associated with context and the common sense of individuals. An individual can learn context sensitivity and common sense from social practices and the general beliefs of a community. I don't think that the speaker is bounded by the semantic rules. The meaning of a sentence and the content of our thoughts are dependent on a particular occasion of use that is contextually sensitive with our understanding. The idea of comprehensive grasp of meaning and the partial grasp of meaning can be signified . I firmly agree with you that meanings are public because there are similar paradigms, not only shared knowledge.
    Like you, I also agree that an expert has the comprehensive grasp of meaning ( better understanding of the extension of the term) whereas an ordinary people has a partial grasp of meaning, like water. But there are some words in society where you will not find any expertise criterion like the word 'sticky' used by Sir Dummett. I agree with you that the meaning of a natural kind term can be fixed partly by stereotype and partly by division of linguistic labour. The gradation of the criterion of expert's knowledge can be verified as it has increased from society with the change of time. Similarly, the operational definition of a word that is given by the ordinary people by its observational properties is also context sensitive. Their understanding have various syntactic structures that is associated with description of the sentnces. As far my knowledge goes, for Putnam, the nature of sentence meaning is a sort of description of the use that help a speaker to get an idea of how to use a term in a sentence. This 'knowing how' process or practice turn towards context sensitivity. If Putnam claim that meaning is totally context sensitive rather han truth condition (you may disagree), then it would be justified to claim that comprehensive and partial grasp of meaning is possible. So here we will not find any concrete idea of meaning except reference. We are well aware that the reference fixation of a term like 'water' does not depend on the comprehensive grasp of meaning of the term 'water' for an expert or a laymen. Here common belief and the succession of the belief procedures that is accepted by the society takes an important role.
    For you, 'truth evaluable content' cannot be regarded as the meaning of a sentence. It actually depends on meaning plus context of use. The disambiguate notion of truth evaluable content can help a particular occasion of use in a refered sentence. Can you accept reference of a term and the truth condition of a sentence are occasionally sensitive? If so,then it would be easy for you to accept meaning as partial and meaning as more comprehensive (context sensitive).

  2. Hilary,
    In a response to Crispin Wright you suggested that the temptation to generate a skepticism about meaning motivated by semantic externalism is incoherent. You say here:
    "What speakers have to be causally linked to is the correct extension, not the correct description of the extension. And extensions, as opposed to descriptions of extensions, aren’t things we grasp with our minds; they are out there in the world."
    I recall you saying in response to Wright that both BIVs (assuming they are referring to computer processes, or whatever) and non-BIVs (when they are speaking about what BIVs refer to) are talking about the same thing, but using different descriptions.
    And so would you say that there is a fundamental problem in attempting to talk strictly about the "extensions," rather than the descriptions? Do we avoid skepticism by showing that the skeptic must be assuming he/she can be referring to these extensions? What sense does it make to talk about what-we-refer-to-out-in-the-world?