Truth-Evaluable Content

For those who haven’t read the previous posts, I repeat the first two sentences of the last one: This series of posts about contextualism (or “occasion-sensitivity”) was inspired by Sanjit Chakraborty’s request that I say more about what I call the “truth-evaluable content” of sentences on occasions of utterance. Formulating what I want to say in response led me back to Davidson’s “A Nice Derangement of Epitaphs” (NDE),  a paper I discuss for its own sake as well.

(I strongly recommend you read the previous three posts, starting with June 2^nd’s.)

As I argued in the previous post, the question as to the nature of meanings shouldn’t be understood an ontological question (what sort of object is “a meaning”), it should rather be “What information should a good semantic theory provide about the words and constructions of a language?”

Davidson agreed with this throughout his career. Moreover, although I do not agree with his identification of truth-evaluable content and meaning (see previous post, again), I think that NDE does contain a reasonable answer to the similar question [in my terminology rather than Davidson’s] about “truth-evaluable content”,viz.,”What information should a good semantic description of the truth-evaluablecontent of an utterance on a particular occasion provide?”[1]

Davidson’s answer is that a good description of the truth-evaluable content of the sentences that are uttered on a particular occasion is nothing more or less than a “passing theory” (which has the form a a Tarskian truth-theory) for those sentences and the indefinitely large totality of sentences that can be derived from them by the familiar recursive devices.

MY QUESTION ABOUT DAVIDSON: The  truth-evaluable content of a sentence on a particular occasion is, then, given by its truth-condition, as specified by a passing theory that does WHAT?

One might expect Davidson to answer, “that describes tacit (implicit) knowledge of the speaker concerning his or her own words, but Davidson rejects this (as do I, but for different reasons, which I will give in the next post). Instead, he says that the passing theory in question must “model the speaker’s competence”. Davidson’s reasons are stated briefly:

“To say that an explicit theory for interpreting a speaker is a model of the interpreter’s linguistic competence is not to suggest that the interpreter knows any such theory. It is possible, of course, that most interpreters could be brought to acknowledge that they know some of the axioms of a theory of truth; for example, that a conjunction is true if and only if each of the conjuncts is true. And perhaps they also know theorems of the form ‘An utterance of the sentence ‘‘There is life on Mars’’ is true if and only if there is life on Mars at the time of the utterance.’ On the other hand, no one now has explicit knowledge of a fully satisfactory theory for interpreting the speakers of any natural language. In any case, claims about what would constitute a satisfactory theory are not, as I said, claims about the propositional knowledge of an interpreter, nor are they claims about the details of the inner workings of some part of the brain. They are rather claims about what must be said to give a satisfactory description of the competence of the interpreter.” [NDE 256]

But this doesn’t tell us what “modeling competence is”; it only tells us that it isn’t describing either knowledge or brain-machinery.

(to be continued)

---

[1] However, I would add: “over and above the information provided by a good semantic theory of the language”. NDE famously rejects  the notion of a language as something with socially share rules.

On Davidson (continued)

This series of posts about contextualism (or “occasion-sensitivity”) was inspired by Sanjit Chakraborty’s request that I say more about what I call the “truth-evaluable content” of sentences on occasions of utterance. Formulating what I want to say in response led me back to Davidson’s “A Nice Derangement of Epitaphs” (NDE),  a paper I discuss for its own sake, as well.

In a short paper[1] I published in 1974, just two years after “The Meaning of ‘Meaning’”.  I wrote that, like Sellars (not to mention Wittgenstein, Quine, and many others) I didn’t think that meanings are objects. And I added that my reason isn’t that one couldn’t identify them with objects, but that the identification would be even more arbitrary than the identification of the number one with the unit set of the null set. In sum, the question as to the nature of meanings shouldn’t be understood an ontological question (what sort of object is “a meaning”), it should rather be “What information should a good semantic theory provide about the words and constructions of a language?” Davidson agreed with this throughout his career.

However, a significant difference between my views and those of the Davidson of NDE is that for me (1) there are such things as languages, and there is such a thing as good description of the grammar and semantics of a natural language, but it does not have the form of a Tarskian truth-theory, and it does not pretend to specify the truth-evaluable content of the sentences of a language on a particular occasion of utterence, while for NDE the meaning of a sentence must by given by a truth theory, and what we learn from a lexicon (our “prior theories”) and what we might take ourselves or someone else to mean on a particular occasion (our “passing theories”) are  both occasion-sensitive.[2]  It is because I identify the meaning of sentence (i.e. the information provided by a good lexicon) with a description of such things as the semantic markers and the stereotypes associated with the words, and not with a set of recursively specified truth conditions (as Davidson does, before and after NDE) and Davidson identifies it with the latter, that for me “truth-evaluable content” and “meaning” come apart, while for Davidson they are the same. Thus for him the occasion-sensitivity of truth-evaluable content is ipso facto the occasion-sensitivity of meaning, and hence the non-existence of socially shared meanings..

 (To be continued)

My discm pleased to join the happy chorus of those who don't think meanings are objects. Not that one couldn't identify them with ob-

jects, just that the identification will be even more arbitrary and uncon- vincing than, say, the identification of the number 1 with singleton the nullpleased to join the happy chorus of those who don't think meanings are objects. Not that one couldn't identify them with ob-

jects, just that the identification will be even more arbitrary and uncon- vincing than, say, the identification of the number 1 with singleton the null set.

---

[1] “Comment on Wilfrid Sellars”, Synthese 1974.

[2] ““As speaker and interpreter talk, their prior theories become more alike””,  quoted in my previous post, and “…knowing a passing theory is only knowing how to interpret a particular utterance on a particular occasion.”, NDE  260 (in The Essential Davidson)

Davidson and “contextualism” (aka “occasion-sensitive semantics”)

In my previous post I claimed that “A Nice Derangement of Epitaphs” (NDE)  marked a “contextialist turn” in the thought of Donald Davidson. (Charles Travis, who introduced me to what I call “contextualism” many years ago, refers to contextualism as “occasion-sensitive semantics”, hence the “aka” in the  title of this post.)  This is not to say that there aren’t any significant points on which I would disagree with NDE, and I will talk about those points later in this series of posts. Nevertheless, I will defend a reading of NDE on which it agrees with the following theses of “contextualism”, agrees, however, on particular interpretations that depend on Davidson’s famous claim that semantics should be thought of as Tarskian truth-theory – I have indicated this by means of comments in square brackets, what is outside the square brackets is from the statement of those “theses” in the previous posts:

(1) In some sense it must be true that a speaker (as we say) “knows the meaning” of each sentence that he or she is able to use prior to using it or understanding another speaker’s use of it in a new context (i.e. on a new occasion of use) and that this “knowledge of its meaning” plays an essential role in enabling the speaker to know what the sentence is being used to say on that occasion.

(2) Meanings should not be thought of as either platonic objects or as mental objects; talk of meanings is best thought of as a way of saying something about competences that speakers possess. [For Davidson, those comptences are modeled by a Tarskian semantic theory that assigns truth-conditions to a potentially infinite set of sentences. The “meanings” described in (1) – the meanings  speaker assigns to the sentences of another speaker prior to entering into a particular conversation, are called “first meanings” in NDE.]

(3) The thesis of contextualism is that in general the truth-evaluable content of sentences depends both on what they mean (what a competent speaker knows prior to encountering a particular context) and on the particular context, and not on meaning alone.”  [In NDE this becomes the claim that hearers’ theories of another speakers' first meanings (their “prior theories”) have to be replaced by different theories (“passing theories”) in the course of each conversation. The right passing theory  for a particular conversation will be one on which both speaker and hearer converge.[1] The truth conditions of sentences depend both on their first meanings and on the way those first meanings have to be modified in a particular conversation; they are not not, in general, the truth conditions assigned by the prior theory. [Moreover, there is not, according to NDE  just one conventionally fixed prior theory shared by all speakers of a language. (“As speaker and interpreter talk, their prior theories become more alike” – see n.1, below.) This leads Davidson to the famous “I conclude that there is no such thing as a language, not if a language is anything like what many philosophers and linguists have supposed.” (ibid. 265)

This is what I will be discussing.

---

[1] “The passing theory is where, accident aside, agreement is greatest. As speaker and interpreter talk, their prior theories become more alike; so do their passing theories. The asymptote of agreement and understanding is when passing theories coincide.” NDE, 260 (in The Essential Davidson).

Contextualism

In my post of April 21, 2014 (to which I referred in the previous post) I explained “contextualism” thus:

“ [contextualism] does not, of course, claim that the meanings of sentences vary from context to context, or at least it does not claim that in every sense of that multiply ambiguous word “meaning,” the meaning of a sentence that one understands changes whenever one finds oneself in a new context. In some sense it must be true that a speaker (as we say) “knows the meaning” of each sentence that he or she is able to use prior to using it or understanding another speaker’s use of it in a new context and that this “knowledge of its meaning” plays an essential role in enabling the speaker to know what the sentence is being used to say in the context.

(Let me also say here that I do not think of meanings as either platonic objects or as mental objects; in my view, talk of meanings is best thought of as a way of saying something about certain world-involving competences that speakers possess. [See references in the April 21 post.] And corresponding to those competences, there are constraints on what can be done with sentences without, as we say, “violating” or at least “extending” or “altering” their meaning.)

What contextualism does deny is that the “meaning” of a sentence in this sense determines the truth-evaluable content of that sentence. The thesis of contextualism is that in general the truth-evaluable content of sentences depends both on what they mean (what a competent speaker knows prior to encountering a particular context) and on the particular context, and not on meaning alone.”

The contextualists I mentioned in previous posts were J.L.Austin, Charles Travis, and Wittgenstein. But it is a remarkable fact that in “A Nice Derangement of Epitaphs” Donald Davidson seems to (and I think does) express agreement with all of these claims. And he is aware that “These phenomena threaten standard descriptions of linguistic competence (including descriptions for which I [Davidson] am responsible).” In the coming posts I shall discuss this contextualist turn on Davidson’s part. (One thing he saw clearly is that contextualism and productivity do not conflict, which was the point of my previous post.) I will discuss this paper, not only because Davidson was an important philosopher and because “A Nice Derangement of Epitaphs” is a brilliant if baffling essay, both of which are true, but because I think he answered the question I have been promising to deal with, “what is truth-evaluable content” (though not in those words, of course), in a way one can build on and improve, but which I think is fundamentally the right way.

     (To be continued)

Context sensitivity and productivity are not incompatible

On April 21 my post began "In “Skepticism, Stroud, and the Contextuality of Knowledge” [See that post for reference], I endorsed a view of meaning that I called “contextualism” (which I credited to Charles Travis, who in turn credits Austin and Wittgenstein, although the versions of each of these three philosophers—as well as, no doubt, mine—have significant differences, as I am sure Travis would agree). A key notion that I used in that essay is “truth-evaluable content”. Sanjit Chakraborty has asked me to say more about that notion, and I shall do that in forthcoming posts."
Four days later, the Nepal earthquake occurred, and because one of my grandchildren, Lauren Chin, was trekking near the epicenter I forgot about this plan (she and her group are all fine and back home now).  But the promised "forthcoming posts" will now come. First, however, I need to dispel  a widespread confusion, namely the mistaken idea that contextualism is incompatible with the "productivity" of language.

The argument that we can understand (idealizing radically) an "endless number of sentences" always turns on recursive aspects of language. There are recursive rules that enable me to understand "If Fa then Ga". "Every F loves some G", etc., etc.,if I understand what F refers to, what G  refers to, etc.  But where context-sensitivity enters is in the fact that F, G, etc; can refer to different things in different contexts (each of which generates a further infinity of possible sentences, in which "being an F",etc., is understood in each of these ways).On one understanding of "rational" , for example, "all men are rational" might count as  true, and on another as obviously false! "Productivity" is explained by the fact that our minds do employ recursions; and the truth conditions of the "endless number of sentences" generated by any recursive scheme - say, the rules of predicate calculus  - do not depend on context sensitivity once the reference of F, G, etc. have been fixed But in practice our predicates, or the great majority of them, have different understandings in different contexts.

Sentence meaning in my philosophy is described just as it is in the traditional theory. by describing word meanings and describing the semantics of the logical words and other structural elements that we need to combine words to make sentences. The difference between my philosophy and traditional views (internalist or context insensitive theories) has to do with word meanings. not with how we get from word meanings to sentence meanings

What determines truth conditions is not word meaning, which I take to be described by "core facts" (in "Is Semantics Possible" - the "normal form" of "The Meaning of 'Meaning'")  in the case of natural kind words - and I say in the latter essay that I take it to be the task of linguistics, not philosophy to classify the different sorts of words there are in natural languages, and to devise appropriate ways of representing their meanings (but not to devise ways of representing all the possible ways those meanings could themselves be understood in different contexts). What determines truth conditions is word meaning as understood by a speaker in a context. Even a simple word like "chair" has an enormous number of ways in which it could be understood