"Martha fell into a trance; therefore: Martha fell" (?) (Grice, Actions and Events, 1986, p. 23)

by J. L. Speranza
--- for the Grice Club

THIS WAS MEANT as commentary to Kramer's commentary on variable polyadicity, this blog. Blogger disallowed it there, so here it goes.

Good. We SHOULD approach, seriously, what Kramer, using computer language, calls the physical-logical distinction. Or logical-physical distinction. The order of how we approach the CALLING of the distinction seems relevant. I would think it's the physical-logical distinction: first 'physical', then 'logical'.

As I understand it, it does NOT correspond to the hardware-software distinction. I think Kramer would say that BOTH the physical AND the logical belong to software, rather than to any (possible) realisation of the software into some kind of hardware. We would like to know who coined the usages of 'logical' and 'physical' in those TERMS.

Kramer, I think this blog, has referred to his son complaining (to him) that Kramer Senior treats Kramer junior as a 'physical' -- whereas Kramer Senior has repeatedly expressed to Kramer senior that Kramer Senior is only interested, or mainly interested in Kramer Junior as a "logical" -- "Kramer Junior" is a PHYSICAL (a spatio-temporal continuant); he is a "Logical" under various guises: one important, relevant, analytically noteworthy one being that that physical ("Kramer Junior") is the son of Kramer Senior.

I agree that we do play with the physical and the logical sometimes confusedly.

When it comes to words (or utterances) and even after we have accepted something like the token-type distinction, there may be a polemic as to what constitutes a physical and a logical.

It seems the physical is what Grice calls, when it comes to predicate logic, 'the predicate letters', i.e. the bricks. Logicians can be very lazy when it comes to natural language. They are just content with a few letters here: F, G, H. -- Why they start to represent predicates with the letter F onwards, rather than the more natural "A" -- or as I prefer "alpha" to represent a "feature" which is what a predicate stands for -- requires an answer in terms of jargon which I'll spare! So, Let us think that "alpha" represent a feature. Here we distinguish between "alpha" lower-case for a token and 'alpha' upper-case for a type. Since, alpha, alas, is totally boring when it comes to upper-case, I'll use the lower-case to represent a 'feature' or predicate. What interests me is the attachment of "alpha" with items which I'll symbolise by x, y, z, and so on. I am having in mind Grice's point about 'meet'.

So, suppose we want to talk about 'meet', qua English verb --. I would NOT think that there should be a different entry in a dictionary (even if it is a theoretical one, or even if it is NOT a theoretical one) accourding to the 'regimen', I think it's called in classical languages, of the 'predicate'. For one, the number of 'arguments' that "meet" takes does not really mark the point raised by Kramer re: the 'transitive' versus 'non-transitive' so let that be. "Meet" CANNOT be 'intransitive', I would think.

So we would have

α(x,y)

to read: x met with y.

The order of x and y IS important, so we need to conceive of what attaches to 'alpha' (or 'meet') as being an ordered pair. While it is perhaps logically equivalent that x met with y iff y met with x, there are presuppositional factors which depend on the topic versus comment (E.g. I follow Strawson that "The present King of France visited the exhibition" and "The exhibition was visited by the present King of France" project different implicata -- he would say 'presuppositions' -- some further would say 'entailments' -- but cfr. their negative versions, "The exhibition was NOT visited by the present king of France -- there was no such exhibition, and there was no such king").

There is the SAME type of "α" that gets realised as triadic, I would think is Grice's point, when we say,

x met with y and z

α(x, y, z)

where the structure is now "... met with ... and ...", or better, "--- met with --- and ---" to bring in the element of a slot".

When it comes to the two examples that Kramer mentions:

'sink' and 'fall', it's really 'sink' that brings the transitive issue. I cannot see how Grice took Davidson so seriously to consider "fall in a ---" to have any philosophical bearance. By doing so, Grice opens, wittingy a can of worms, because he starts to consider the relevance of a sub-clausal element in an utterance, "in a ---", which he then compares to an adverb. For he rightly points, "How that DID happen (that Mary fell)?" "Well, into a ditch, actually -- as she walked the poor pavements of Fleet Street" (Fleet street is now pretty sophisticated, in London, but it used to be a river -- called Fleet, so I assume at one point there must have been some ditches there into which one could easily fall).

This has Grice compare 'in a ditch' with 'violently'. "It was a violent fall". But what _is_ 'violent', Grice asks. Surely, 'violent' takes a different interpretation in, to use his examples (Grice, 1986, p. 23): "He sneezed violently" and "He swore violently". I added in the margins of my copy -- "Cfr. Grice on beating the wife". For he was discussing that worn-out example early in 1961 (he ommitted it in WoW -- since it's section II of "Causal Theory" -- but there is a ref. to it in "Presupposition and Conversational Implicature" before he engages in some analysis of implicata of common phrases). "He beat his wife violently". (Actually, the point here is in the 'stopped beating his wife' sort of 'joke').

Grice states his general point thus: "It ... seems to me that the deployment of the idea of specificatory modification [as in "Mary fell into a ditch" versus "into a trance"] would dispel the embarrasment which [are] noted arising from the verb- [or predicate-] dependence of certain adverbs." And the point is general, for isn't "into a ditch" a sort of adverb, logically? --. He goes on: "whether or not violence in sneezes is the same feature as violence in swearing [or beating his wife -- JLS adds -- NOT meaning anyone's wife in particular -- it's a sophisma of philosophers!], 'sneezes violently' or 'swears violently' will offer specificatory modifications, of sneezing and swearing."

I would think he IS allowing that 'violently' does not undergo a change of _Sense_ (Do not multiply senses beyond necessity: his modified Occam's razor). Similarly, 'fall' is NOT undergoing a sense in 'fell in a ditch' versus 'fell in a trance' --. The dictionary should treat both under the SAME entry. And treat the second as 'figurative'. There is A PHYSICAL metaphor involved: Mary's ego does fall, into a trance -- which IS like a physical ditch, only different.

---

It is promissory that Grice does regard the implicature theory to allow to get rid of Davidson's pseudo-problems. For it is THEN that we can provide an independently motivated account of what is philosophically important about Davidson's _question_. Grice expresses this thus in the paragraph immediately following the 'fell in a trance', when he refers to 'the problems which originally prompted Davidson's analysis' as being "on their way towards independent solution" (emphasis mine. JLS).

The problem with 'sink' may yet be a different one. True, Grice saw the "H. M. S. Rodney sank the Bismark" as not really entailing "H. M. S. Rodney sank" as a problem to a general point he had made:

"Sometimes, in one's progress through the maze of 'polyadicity', one moves between sentences which are not logically independent of one another ... this is often but not invariably the case" -- the example of 'sink'. His comment: "We need some theoretical characterisation of the occasions when such inferences ARE in order, and the idea of treating actions as ENTITIES in their own right, and SO, as subjects of attribution, may help to obtain it."

Which is back to the logical-physical or physical-logical distinction which prompted this, and to which I'll need to come back!