Abstract for Grice Abstract

We are discussing with R. E. Dale the issue of _abstractedness_ and am fascinated that in his PhD dissertation, "The theory of meaning" -- his commentary to Jones, just note, provides yet further rationale for the title! -- he used 'abstracta'.

Propositions as abstracta. I like that. Singular, abstractum of course. I think Grice does refer to 'abstract entities', about which we've discussed with Jones in our attempts to dissect the twelve betes noires (that Dale incidentally mentions in his thesis) -- and notably Intensionalism.

---- Dale makes the good point about how Fodor's "Mentalese" may be seen as a latter-day re-incarnation of Katz/Postal 'semantic markerese', and so on.

My aims were more modest. I was wondering how abstract Grice can get.

In particular, with reference to some material in Chapman, which I've discussed elsehwere (_Grice_, Palgrave) but will see if I can retrieve. Grice is providing generalisations about functions related to, say, syntactic modes. He does use "*" as a mode-indicator -- and his 'asterisk-sub-psi' is supposed to mean, the syntactic mode associated with the psychological attitude psi -- throughout WoW:V-VI --.

But here, then, just this reference to Grice's use of the simpler symbols:

α

and

β

--- in Grice WoW:VI.


When does he first introduce them? Let me check. I'm bringing this to attention vis-a-vis an interpretation (or 'popularisation', since he is clear enough) of what Grice is doing.

He has dealt with utterer's meaning, providing a 1-page long definition (WoW:V) -- that overwhelmed him so that he did follow Schiffer's suggestion, or rather what Grice found was a 'corollary' to Schiffer's moves -- "Reply to Richards" -- to the effect that it would be otiose to inquire for M-intentions in language-destitute creatures. This has a circular ring to it, but not totally vicious!

---

So, utterer-meaning more-or-less dealt with, he can focus on 'sentence meaning'. That dealt with, he can then explore subsentential items.

And here it is where

α

and

β

play a role, since they are Grice's informal symbolism for features in the utterance ('the parts' of the utterance) as it were for which he'll try to provide necessary and sufficient conditions of the form

α means ...

β means ...

I disfavour the use of 'that'-clauses here, since Grice is explicit that, strictly, oratio obliqua applies first and foremost (and last but not least, if you must!) to utterer's meaning -- only.

But the reference to

α

and

β

are a _bit_ like semantic markers. They are, though, uninterpreted. Grice, qua philosopher, is providing the 'schema' of a "meaning-specification" -- to be filled by the linguist, as it were.

Here, Dale's commentary on Schiffer's schmdog may fit in.

Recall

α

and

β

---- Dale was wondering about proto-Gricean accounts of this (was he?) -- or rather, what do we have earlier than Carnap 1928 about a 'compositional semantics' for what traditional grammar had as 'partes orationis' (parts of speech). I have read some of the mediaeval modistae and have been impressed (Erfurt, Longman Linguistics Library, Grammatica speculativa).

But Grice, for one, although his classical credentials were immune to criticism, seems to oversimplify things here?

He notes that

α

will stand for 'nominal', and

β

for 'adjectival'.

Adjectival is easy enough. He has 'shaggy' in mind (meaning 'hairy-coatedness'). The nominal is a bit of a bigger problem (though not for Grice at that point, since he is just being intentionally programatic).

When Ostertag reprinted bits of Grice's Vacuous Names in "Definite Descriptions" he rightly emphasised the section, the closing one (but one) on "names and descriptions".

And it is this type of thing that Grice has in mind when he mentions

α.

For

α

is a 'nominal' "Fido". And it's R-correlate is the denotatum of a definite description, Jones's dog.

---

It would have been nice if English grammar had allowed Grice to focus on 'dog' --

DOG SHAGGY

as it were. But no such simplicity.

"Fido" shaggy.

OK --

but

dog shaggy

comes out as pretty Tarzan-like in English (but not in Latin).

canis pulchrus.

say

BEAUTIFUL-dog.

---

(The copula can always be omitted in Latin).

---

-- So it seems Grice is satisfied, as he should, with providing a schema that mentions not particular semantic markers, like DOG, or SHAGGY (the second certainly under "β") but their abstracta. As it were.

-----

Grice then:

The introduction of

α

and

β

take place on p. 130 of WoW:

"We need," Grice writes -- not far from MIT, as he was -- "to be able to apply some such notion as a

predication

of

β

(adjectival) on

α

(nominal)."

I propose

β/α

----

Grice goes on to discuss the D-correlation with which he is specially concerned.

"W wnat to be able to speak
of some particular object"

--- obble in his more informal talks

"as an R-correlate of a

α

(nominal), and of each member of
some class as beig a D-correlate of

β

(adjectival).

---

So far he is certainly not providing the meaning of

β.

Just looking for the best way to represent a claim about, as the title of the lecture goes,

word-meaning

itself.

I.e. expression-meaning proper, at its minimal constituent.

----

He then goes on to consider what he calls

"P2" -- a procedure -- it is this talk of procedures (basic and resultant) that sort of irritated Chomsky in "Reflections of language" -- he found it hopelessly 'behaviouristic' talk).

P2: to utter a psi-cross-correlated ... predication of

β

on

α

if (for some addressee)
utterer wants addressee to psi-cross
a particular R-correlate of α
to be one of a particular set of
D-correlates of

β"

---- Grice will develop the idea of a 'dossier' -- as per the fragment in "Vacuous Names" that Ostertag reprints in his popular reader - which relates here: "Fido" and "Jones's dog" are two members of the same _dossier_ for the individual in question.

---

Grice is surely not trading necessity when he expands on

β:

"Suppose that, ... the following correlation... hold[s]."

'Any hairy-coated thing is a D-correlate of 'shaggy'"

---- INTERLUDE about the etymology.

From online sources:

shaggy
1590s (earlier "shagged", Old English), from "shag" (n.) + -y (2). The shaggy-dog story as a type of joke is attested from 1945.

---

The derivation being from

shag (n.)

1590s, "cloth having a velvet nap on one side".

From Old English 'sceacga', "rough matted hair or wool," cognate with Old Norse 'skegg', "beard," from Primitive Germanic (hypothetical) *skagjan, perhaps related to Old High German 'scahho', "promontory," with a connecting sense of "jutting out, projecting."

Of tobacco, "cut in fine shreds," it is recorded from 1789; of carpets, rugs, etc., from 1946. Shagbark as a type of hickory is from 1751.

---

[In case you wondered, the VERB is not cognate. 1788, probably from obs. verb sh*g (late 14c.) "to shake, waggle," which probably is connected to shake. And þe boot, amydde þe water, was shaggid. [Wyclif]].


----

- SECOND INTERLUDE. At this point, it may do to play a bit on Grice/Strawson's example of contradiction:

"My neighbour's three-year-old son is an ADULT".

Here it seems that Semantic markerese may have something to add. Or not.

(I was interested to learn via a google search on these issues that Staal, a collaborator with Grice on 'philosophico-linguistic questions', I think Grice puts it -- did criticise Katz for being unable to provide the connection between the markers, 'buy' and 'sell', say).

----

Out of that procedure above (P2) which Grice deems basic, he elaborates on a resulting one:

RP1: to utter the indicative version of a predication of

β

on

α

if U wants A to thnk U to think that a particular R-correlate of

α

to be one of a particular set of D-correlates of

β"

-- from which is clear that Grice is having in mind something like what R. B. Jones refers to when he focuses on 'universe-of-discourse' limited interpretations -- all extensionally provided for, incidentally.

A second resultant procedure, for Grice, will have the form:

RP2: to utter the indicative version of a
predication of

β

on "_Fido_" if U wants A to think Jones's dog to be one of a particular set of D-correlates of

β."

----

When it comes to the format, Grice leaves

α

and

β

behind and brings back his beloved "X"

---

"a definiens which may be adequate for
adjectival X (e.g. 'shaggy'). D7. For U, X [β -- Speranza] (adjectival) means '...'" =df "U has this [bsic] procedure: to utter a psi-cross-correlated predication of X [β -- Speranza] on α if (for some A) U wants A to psi-cross a particular R-correlate of

α

to be ..."
---

Again, to use Grice's symbols, then,

α

and

β

--- consider how he possibly was thinking of an abstract scheme. Suffice his footnote on WoW, p. 133n1:

"To have explicited correlated X with each member of a set K, not only must I have intentionally effected that a particular relation R holds between X and all those (and only those) items which belong to K, but also my purpose or end in setting up this relationship must have been to perform an act as a result of which there will be some relation or other which holds between X and all those (and only those) things which belong to K. To the definiens, then, we should add, within the scope of the initial quantifier, the following clause: "& U's purpose in effect that ∀x (......) is that (ƎR') (∀z)(R' 'shaggy' z≡z ϵ y (y is hairy-coated))." (WoW:133n)

---

-- where, if we further abstract 'shaggy' to read, generally

"β"

we get

"To have explicited correlated X [β -- Speranza]

with each member of a set K, not only must

(i) U have

intentionally effected that a particular relation

R holds between

β

and all those (and only those) items which belong to K,

but also

(ii) U's purpose or end in setting up this relationship

must have been to perform an act as a result of

which there will be some relation or other

which holds between

β

and all those (and only those) things which belong to K."

In symbols:


U's purpose in effect that ∀x (......) is that (ƎR') (∀z)(R' 'β'z ≡ z ϵ y (y is D-correlate of β)).

-- or something, as Jennings would say (*).

--- Jennings wrote "The genealogy of disjunction", so I'll trust he won't object.