From the Workshop Archives: Naming and Necessity

Lecture I
Lecture II
Speaker's Reference and Semantic Reference
Lecture III

Philosophy of Language Muxtape!

Names, indexicals, sense, what is said, truth, tense, unarticulated constituents...from a logical point of view.

The philosophy of language muxtape

Track list:

1. Biz Markie - My Name Is...
2. De La Soul - Me Myself and I (radio version)
3. Missing Persons - Words
4. ESG - You Make No Sense
5. The Cure - Speak My Language
6. Sleater-Kinney - Things You Say
7. Led Zeppelin - Communication Breakdown
8. Johnny Cash - What is Truth
9. The Germs - Lexicon Devil
10. MC5 - future/now
11. Irma Thomas - It's Raining
12. Robert Mitchum - From A Logical Point Of View

Related posts:

The Philosophy of Mind Muxtape

**UPDATE 8/14/08: In response to Aidan's suggestion, I added Led Zeppelin's "Communication Breakdown".**

Speaker's Reference and Semantic Reference

On Thursday, April 28, the workshop met to discuss Kripke's paper "Speaker Reference and Semantic Reference".

Attending the workshop were David F. (though he was feeling sick), Jason B., Jay E., Nat H., Chris F., Ben M., Will S., Aidan G. and Rachel G.

The workshop fell into roughly three different stages. I'll post them separately.

Stage I: Pre-Workshop Banter

Chris F. was reading a book of Wallace Stevens essays in preparation for Stanley Cavell's talk on Friday. Someone asked why Stevens was interesting and David F. said something about Stevens's poetry being about conceptual and non-conceptual perceptual content. David F. then told the story of Stevens's employment at an insurance agency (vice president of the Hartford Accident and Indemnity Company), and how a literary agent who was trying to track Stevens down caused a stir at the insurance agency by mentioning a book of his poems to an oblivious co-worker. According to David F., his co-workers could not believe that "Wally" Stevens was an accomplished poet.

Jason B. later showed up and also told a story about Stevens, which began, "It's interesting how it's not very interesting that Stevens was an insurance salesman". Nat H. was interested in hearing more, since David F. had just reported that fact as if it were interesting.

Jason B. said that he heard a talk (at the Franke Institute?) given by someone who had done research on early 20th century New England insurance agencies and found that it was not uncommon for them to produce in-house literary journals. So there would have been lots of insurance-salesman-poets when Stevens was writing.

"Why haven't we heard of all these other insurance poets?" someone asked.

"They weren't very good" Jason B. replied.

Discussion then moved on to a thought experiment raised in a class on perception involving a lasers, a flesh-colored bodystocking and a heliotrope sweater. There was some dispute about what color "heliotrope" was. The OED says that it's a shade of purple. Follow this link to a picture of the flowers the color is named after:

http://www.swallowtailgardenseeds.com/annuals/heliotrope.html

Stage II: More Discussion of the Causal Theory of Names

Jason B. got discussion moving by asking if we thought Kripke's methodological claim about purported counterexamples to linguistic analyses was right.

The methodological principle goes as follows: If you imagine a language in which the analysis is true, and the purported counterexamples still occur in that language, then the analysis (e.g. Russell's analysis of definite descriptions) has not been refuted by the counterexamples.

Chris F. bravely took a stab at assessing the correctness of the methodological principle. He got as far as describing Kripke's argument against Donnellan: (1) that Donnellan's phenomenon is meant as a counterexample to Russell's theory of definite descriptions, but (2) it is only a counterexample if it counts as a semantic distinction; and (3) it does not count as a semantic distinction--only a pragmatic one.

During this discussion there was some confusion over the proper pronunciation of "teetotaler", a word Kripke uses in describing the purported champagne quaffer in the corner in Donnellan's example. Chris F. preferred "teeto-taler" (said like "teeter-totter"), others demurred.

The methodological question not yet resolved, discussion then turned to the causal theory of names. Nat H. tried to link the topic of this paper with David F.'s worry from last time.

David F.'s worry was that I might, according to the causal theory of reference, count as satisfying the conditions for referring to the island typically referred to as "Globula" with my use of the word, but have a consistent personal use of the word to refer to Will S (say I keep a diary where I record lots of observations about what I think is Globula). Should we say, with the causal theory, that I am saying and thinking a large number of things about the island that are absurdly false (such as that it has a distinctive haircut and is a snappy dresser), or that I am saying and thinking a large number of true things about Will S.? To David F. (and others) it seemed obvious what the answer would be: I am referring to Will S., not the island.

In David F.'s case, the causal chain connecting my use of "Globula" with other members of my community seems unimportant. What is important is my consistent name-using practice.

There was then some extended discussion of the "interpretationist" (Davidsonian) alternative to the causal theory of names. Roughly, the interpretationist holds that a speaker refers to a particular object with his use of a name just in case taking him to refer to that object with that name makes the best overall sense of his behavior.

The interpretationists among the workshop attendees were Kripkean in the following way: they would include reference to causal chains in the relevant features one could rely on in making sense of a person's utterances. So, for example, if I say things like "I wonder what Feinman's favorite breakfast cereal was", when I don't have any beliefs about Feinman other than that he's a famous physicist, I should be interpreted as referring to Feinman in virtue of having acquired the name from some other users of the language. I shouldn't be interpreted as believing that the description "the famous physicist" is uniquely satisfied and wondering about whoever uniquely satisfies that description. To that extent, then, the interpretationists agreed that Kripke was right to suggest the importance of causal connections to our name-using practices.

Jason B. worried that the interpretationists were also individualists--that they would deny the importance of pre-existing linguistic institutions to our name using practices. But Aidan G. and Nat H. objected--why wouldn't an interpretationist want to make use of every possible resource in making sense of a speaker, including the speaker's participation in all kinds of institutions and his causal imbeddedness in his social and physical world? David F. had our backs on this one.

At this point, Nat H. observed sotto voce to Jason B. that this kind of interpretationism that wants to accommodate all of Kripke's insights without embracing a causal theory of names is Evans's view in "The Causal Theory of Names".

David F. felt sicker. He said, "It was fun" and left.

At some point in the discussion after David F. left, Jason B. said that Naming and Necessity was perhaps unique among philosophical books in that it was 99% true.

Stage III: Return to the Methodological Question

Once we had gotten to the bottom of the causal theory of names, we returned to Jason B.'s original question: is Kripke's methodological principle in "Speaker's Reference and Semantic Reference" right?

Will S. and Chris F. had some concerns that the Strong Russell Language would not be susceptible to the Donnellan counterexamples, but as I left to use the bathroom during their discussion I can't reproduce their arguments.

Jason B. wondered what result Kripke's methodological principle would give if we introduced a language in which a particular analysis (e.g. Russell's) were true, and the purported counterexamples happed less often than in English (rather than never). Would the analysis be a little bit false?

At the end of the workshop we debated the merits of staying with the names discussion and reading either Evans's "Causal Theory of Names" or Davidson's "Nice Derangement". Nat H. argued that we had roughly reproduced the conclusion of Evans's paper, and that since we only had two sessions left we should move on to lecture III.

Which we will, in two weeks time.

Naming and Necessity, Lecture II

Last thursday the workshop met to discuss lecture II in Naming and Necessity.

In attendance were Jason B., David F., Jay E., Nat H., Zed A., Chris F., Ben M., Aidan G., Rachel G., Will S. and Justin S.

David F. bought Stella Artois, which was good, but a smaller number of beers than normal, which was bad.

Chris F. volunteered to recap what we learned last session and lead us into lecture II. In essence, he claimed that lecture I offered arguments against thesis (6) on p.71 and that the first part of lecture I consisted of arguments against theses (2-5).

There was some brief debate about the importance of Kripke's emphasis on individual speakers in thesis (2). Jason B. wondered whether there might be a way of generating a social account of the cluster of properties believed to correspond to a name. Such a view would be a modified description view akin to Strawson's view discussed on p. 65, n. 27. We agreed that this proposal was slightly more plausible than the individualist view Kripke is attacking but not a good enough improvement to warrant much attention.

Discussion then concentrated on whether Kripke's causal-historical "picture" of names was an improvement on the descriptivist "picture". Zed A. quickly grew irritated with the need to constantly affix the caveats "Kripke's not offering a theory", "Kripke's only proposing an alternate 'picture'", "Kripke's not giving necessary and sufficient conditions for anything" to any criticism of Kripke's account. He proposed that we all acknowledge that all our talk for the rest of the workshop would be picture-talk.

We adopted the convention, but Zed A. remained irritated.

What about the causal-historical account of how names refer? We started out with a discussion of a thought experiment proposed by Zed A, similar to some of Gareth Evans's examples in "The Causal Theory of Names". The example was described as follows:

1. I overhear you talking about something called "Globula".

2. I intend to participate in your conversation by saying various plausible-sounding things about Globula. For example, I say "I bet Globula is one sharp dresser". Since you and the other interlocutors are referring to Globula, I count as referring as well according to Kripke's picture.

3. But what I say about Globula is not only false, it is seriously mistaken in the following way. Globula as used by you and the other interlocutors refers to a small Hawaiian island. So what I'm saying about Globula is near incomprehensible ("Globula is a sharp dresser").

4. What to make of such a situation? Do I count as referring to Globula when I say "I bet Globula is one sharp dresser?" Zed A. and Nat H. (and some others) shared the intuition that I would not be referring to Globula, so the example seemed like a problem for Kripke's picture (a problem rather than a counterexample since, presumably, a picture can't be counterexampled).

Jason B. was unimpressed. His intuitions pulled him in the opposite direction and he offered the following variation on the Globula case.

1a. Suppose I find myself in a culture where talking about Globula is taboo on certain occasions (say at dinnertime).

2a. I have overheard you using "Globula" in other contexts, and one night at dinner, I utter some sentences with the word "Globula" in them, intending to use the word to refer as you have in the past (and not knowing about the taboo).

3a. Everyone at dinner listening to me is horrified by my gross breach of etiquette. A couple of my friends pull me aside after dinner and whisper, "You fool--what were you doing all night going on and on saying all those weird things about Globula? Don't you know you're not supposed to talk about it at dinnertime?"

4a. The intuition in Jason B.'s version of the example is supposed to be: I am referring to Globula, even though I am grossly mistaken about what it is and say strange, possibly incomprehensible things about it.

So, we determined (unsurprisingly) that there can be conflicting intutions about odd cases in Kripke's picture.

David F. then offered a different example, slightly different from the previous two, to suggest that Kripke's picture fails to take account of our name using practice. His example went as follows:

1b. Like the previous cases, I don't know how others are using "Globula", but I intend to use it as they do. But unlike the previous cases, I mistakenly apply it to some other object. (This case is like Evans's Madagascar example, except that it can involve only one person).

2b. I apply "Globula" consistently not to an island, but to a person. I keep a diary where I record all sorts of information about Globula--what he's wearing, what he says to me, etc.

3b. In David F's version of the example, should we say that I am constantly referring to the Hawaiian island and saying radically false things about it (like that it was rude to me and was sharply dressed)? Or that the causal historical chain leading up to my use of the name is irrelevant and what matters is actually my coherent use of the name to refer to the person?

David F. and Aidan G. both emphatically endorsed the overriding importance of our use of the name.

Jason pointed us to a footnote (pp. 85-86 n. 36) where Kripke offers what looks like his response to this kind of case. It will involve the speaker reference / semantic reference distinction. With that in mind, we decided to read Kripke's "Speaker Reference Semantic Reference" paper for next time before going on to lecture III.

This account leaves out discussions we had of Swamp Man, the general nature of offensive language, and Terry Schiavo's ability to participate in a name-using practice.