In Lieu of a Syllabus:
Philosophy of Language Group (Philosophy 399, Fall 96, Spring 97):
Plans and some notes on first package of readings after Locke
Included here is the beginning of Mill's System of Logic on names; the beginning of Russell's An Inquiry into Meaning and Truth, on what is a word; and Augustine's To the Teacher, on what language is and what we are communicating when we use it.
We are not finished with Locke, so review that (book III, the first chapters) and begin the Mill, up through page 23 or so. Glance at the others if you have time. I plan to try to summarize my issues and maybe what I am interested in accomplishing. Here's a summary of the summary.
I'm interested in the question of whether there are some common presuppositions which underly basic problems in philosophy of language. I'm thinking of those problems as standard old-chestnut problems: What is meaning? What is the meaning of a word? How can language serve as a medium of communication? How can we account for novel utterances? (that is, how is it possible for someone to say something or to understand something which has not been said or understood before? What is knowing a language? What are names? How can a word refer to an indefinitely large number of things? What is it for a person to mean what she says? How are words related to the world? How are utterances used to perform actions (like promising, demanding, assuring, etc.)? What is the relation between a sentence and a proposition?
Now of course some of you know that there is something bogus about my asking this question, what are the presuppositions of these problems, because I'm really out to claim that I think there is something like a nest or set of such presuppositions, questionable presuppositions, and that if we examine those we will radically change how we do philosophy of language. Radically change means we will sweep away all those questions and start over in some dimly imagined way.
The set of presuppositions is captured in these first readings. Locke and Russell and Augustine and Mill all have parts of their story on which they are in agreement. That's the thing I'd like to clarify and then question. I think I can make a case that Kripke and Searle and Grice and Bar-On and others also have kept this bathwater while struggling mightily to get rid of babies. Locke I like because of the conciseness and eloquence and so those first three chapters of Book III can serve as a touchstone if you like. Ideas give way to meanings, to reference, to thoughts, to messages, to illocutionary act potentials or to tendencies to behave possibilities of verification or falsification or truth conditions or propositional content or something--now only dimly perceived--but the stuff that does the conveying by standing for those things, namely words and sentences, names and utterances and statements, remain importantly the same through all these different accounts.
Jeff Johnson and Mike Duffy have been going over some of the same territory during the summer, and they have pointed out that there is something troubling about the basic arguments I've been using. As I understand it, their story goes like this: jwp, you are making a case that if you think of language in a certain way, then you can make many of the problems in philosophy of language into urgent, troubling, and interesting philosophical problems; and then you show that that way of thinking about language has problems, including problems of intelligibility and paradoxical consequences. So that way of thinking about language is probably not the right way to think about language. The trouble is that these are something like psychological arguments when what is needed is philosophical arguments. You, jwp, do not show that these are the only ways to think about language, or that we cannot just start with the problems, or that there are other ways to get the problems aloft. Indeed it is hard to imagine how anyone could make a case that this is the only way to think about language, and even if you did, that does not show the philosophical problems cannot stand on their own.
So, I'm not finished, and my main reason for working on these problems is to get your help in thinking them through. I need to get the basic view out, my arguments against the view, and figure out whether there are replies to the objections raised by Jeff and Mike. Along the way, I hope we can become acquainted with some of the most important classic and current readings in philosophy of language, in particular with readings on naming (fall) and meaning (in the spring).
Addendum, beginning of spring 97
What I'd like to do here is march through some history of the philosophical concept of meaning (not that tha there is any other concept of meaning). Ideally we would not do this; rather, we would go sit at the feet of the master and get each story from the horse's mouth. If the group is interested in going back, or any of you are interested in reading and reporting back to us on any of these, that would be terrific.
Plato is often represented as claiming that the existence of names is highly important, and so naming is itself a business to be taken on with care, and only to be done by those who have insight into the essences of the things which they name. The meanings of names (and sometimes it seems as if sentences consist of nothing except names) are the forms or the reality of the things that bear those names if naming is done well. (Though at one point [Epistle VII, Stephanus 341 ff.] he says that names are not stable in the way the things they should name are stable.) Now there are grave problems with this, but they will pull us far afield. In the Cratylus, connections between names and the things they name seem to operate by way of characteristics of the thing names, and good names are ones which direct our attention to crucial characteristics, essences we might say, of the things named.
In Plato is the beginning then of the notion that the meanings of words are to be thought of on the model of names and things named, with some attention drawn to the possibility of intermediaries between those two. Usually these kinds of accounts are called theories of reference, but they are often acknowledged as theories of meaning as well or are called referential theories of meaning. Names refer to the things they name or to some intermediate like for instance characteristics or properties of the things named; words in general might then be thought of as naming their meanings, to which they refer.
What is often called the ideational theory of meaning is found in Locke. The meaning of a word is the idea which it stands for or produces in the minds of speakers and hearers. For this, Locke is the very eloquent horse's mouth. On this account, then, philosophy of language is explicitly a subdiscipline of philosophy of mind.
We looked at objections to the ideational theory, but here I'll just wave my arms at a couple of those. Descartes pointed out that if we think of ideas as like mental images we are hard pressed to account for the difference between the meaning of “thousand-faceted-solid” and “eleven-hundred-faceted-solid” since the images of the two will be insignificantly different even if the meaning seems clearly different. Berkeley points out that though the word triangle surely has a meaning, we can conceive of no such general idea as a triangle--what we are capable of having in our mind is always more specific in being obtuse or acute, right or oblique, even though the word has to mean triangle-in-general which is not obtuse or oblique. Wittgenstein (after worrying about images of leaves in the way Berkeley worries about triangles) points out that the existence of ideas need not be present for words to do their work--whether I have a particular idea or no idea at all, if I have conversations my conversations remain the same, and even if we were to discover some invariably-present idea that would not be the determiner of the roles which the words play in conversations in the way philosophers who think those ideas are the meanings require.
Wittgenstein is often wrongly characterized as replacing the ideational theory of meaning with a theory which claims that meaning is use. Though it is not his child it still wears his name and is influential. On this account one determines the meaning of an expression by looking to see how the word is used in the language. How a word, for instance, is used in the language, what its role is in linguistic activities--these are the meaning of the word. William Alston, influenced by J.L. Austin, works to amend this story as follows: the idea of use in Wittgenstein is a technical notion; use can be thought of as the way a word contrubutes to the performance of certain speech acts or actions which require words. Austin calls one of the main families of speech acts by the technical label of illocutionary acts, and Alston then speaks of the possible contribution of a word to one of those illocutionary acts as illocutionary act potential, which is his analysis of meaning. So two words could have the same illocutionary act potential or the same meaning if they could be interchanged in an utterance without changing what was accomplished by that utterance.
The logical positivists, to whom we owe the philosophical respectability of the idea of nonsense mostly because of their awesome facility with logic and their self-righteousness, address the question of what it is for an expression to be meaningful rather than meaningless in a way which winds up having implications for an analysis of the meaing of an expression. On their view, (with exceptions for logical tautologies and logical contradictions) sentences are meaningful only if there is in principle a public way to verify them or falsify them. The meaning then of utterances comes to be identified with the method of verification (or falsification) or the conditions to be satisfied which would provide that verification or falsification. Note that this trades heavily on the notion that the main business of communication is to say truths, and conditions of verification wind up later, in paler and quieter versions (in e.g. Davidson talking about sentences rather than words), popping up out of their graves to plague us as truth conditions.
The ideational theory implied a close relation or an identity between expression meaning and speaker meaning. Paul Grice reexamines this and finds an important distinction between the two which has come to be generally accepted, especially in linguistics. We need to slow down a moment. Grice's question is whether what an expression means can be reduced to what a person means by that expression. His genius is probably in raising the question, rather than in the answer he provides, but here's that answer. He claims that expression meaning is a particular kind of nonnatural meaning “meaning-sub-NN”, which though is parasitic or grounded in a more natural kind of meaning, that which we see in cases like someone remarking that “those spots mean measles.” The other contribution is to relate the notion of meaning to the notion of intention; what a person means to do Grice thinks of as within the realm of natural meaning, and it provides part of the basis for nonnatural meanings. What a person means to do is roughly what that person intends, and so the meaning of expressions is to be analyzed in terms of the intentions of persons who use the expression. “Sam meant[sub-nn] something by the utterance u,” gets treated by Grice as equivalent to “Sam intended his utterance u to produce a certain effect in his audience by their recognition of his intention.”