How Pirots Karulise Elatically

It was good to have Dale's post, "English composition" shedding some light on the history of compositionalism. The post I was referring was "Dale and Grice on meaning", I think. In any case, here is a rewrite with some expanded running commentary, then.

Carnap, as quoted by Dale, writes as early as 1928 -- the English translation is of course later.

"Let us consider the designation relation"

This is one relation that did obsess Grice. Grice's example:

Jones's dog is long-haired.
Fido is shaggy.

He wants to refer to some correlations as

R-correlate (where "R" stands for 'refer') and
D-correlate (where "D" stands for 'denote')

I should revise this, though. It's last segments of Grice WoW:lecture VI -- what I have elsewhere referred to as Grice's shaggy-dog story.

Grice writes (WoW:VI, p. 130):

Let us consider

"two species of correlation, R-correlation (referential) and D-correlation (denotational). We want to be able to speak of some particular object [item alpha] as an R-correlate of "alpha" (nominal), and of EACH member [item] of some class [set] as being a D-correlate of "beta" (adjectival)."

---- Echoes of Carnap!

---
Dale continues to quote from Carnap:

"as it holds between ... words and their meanings."

"Since [a] natural language[... -- such as English] do[es] not have general rules which allow us to deduce the meaning of a word from its form, there is no way of indicating the extension of this relation except by enumeration of all its member pairs."

One may wonder about this. It seems we have a predicate, "SHAG". This possibly should come out as marker in Katz/Fodor. In Gricean parlance, 'shag' becomes 'shaggy' -- where 'shaggy' (adjectival) is a bit like Carnap's pirots karulising elatically. It seems that if you say, as Grice wants to say:

"Fido is shaggy"

"shaggy?"

"Yes, hairy-coated. "Shaggy" means 'hairy-coated'"

---

'shaggy' cannot but BE taken as meaning what Grice takes it to mean, a sub-sentential element that is associated with hairy-coatedness. What I'm referring to is the logical form (or grammatical empty form) of something like the first line of Lewis Carroll's 'Jabberwocky'.

Or

Pirots carulise elatically.

As Carnap notes, one may NOT know the 'extension' of the crucial words here ('pirot', 'karulise', and 'elatic') but there is this abstract form:

----s ----- adj-LY

that provides a lot of 'interpretant', to echo Peirce. When it comes to 'shaggy', it's the '-y' that does the triggering. I know, NOT all adjectives in English end in -y, but it would be nice if they did.

-y is the morpheme here. It seems English is pretty transparent here. Finnish may be different 'animal' (not for Finns), and Dale's very first section to his thesis is about the ways Finns have to say, 'Snow is white', which you wouldn't recognise as a sentence (unless you are a Finn or Finn-speaker). But I should need to revise Carnap's point above, since he is not just sticking with the COPULA ('snow IS white') but trying to go deeper.

---

So let us revise the Carnap quote as quoted fully by Dale in "English composition":

Carnap:

"if a sentence consists of three words, a noun in the nominative case; a verb in the third person singular, present tense, active mood; and a noun in the accusative case, then it designates the state of affairs that the object of which the first word is the sign stands to the object of which the third word is the sign in the relation of which the verb is the sign."

Mmm.
The form seems to be
Rxy
where he is referring to "R" as being transitive. I think he is having in mind
Brutus killed Caesar.
----
And he must be thinking in Latin, where no articles are necessary:

Homo habet canem.

I.e. the man has a dog.

---

Carnap:

"if a sentence consists of three words, a noun in the nominative case [HOMO, the man]; a verb in the third person singular [HABET, has], present tense, active mood; and a noun in the accusative case [CANEM, dog-acc], ... it designates the state of affairs that the OBJECT [or R-correlate] of which the first word [HOMO] is the 'sign' stands to the object [another R-correlate, strictly] of which the third word [CANEM] is the 'sign' in the relation ['having'] of which the verb [HABET] is the 'sign'"

This relates then to this constructivist approach to 'propositional complexes' in Grice's Reply to Richards, and his clearer remarks, Jones hopes, in "Vacuous Names" (Section, Semantics for System G -- quantified formulae).

The idea that there are ordered pair -- where only

"Man bites dog"

makes headlines.

----

Carnap goes on:

"If a basic language is already known, ...
this is done through a dictionary; otherwise,
the answer takes on the form, for example, of
a botanical garden, that is, a collection of
objects, each of which has its name written on it."

It is apt, I thought, that Carnap would mention botany at this point -- something which appealed both Austin and Grice ("linguistic botanising" being THE method to approach philosophy, for them). I KNOW, they mean different things -- but in a way, every time a philosopher refers to botany he means _mess_. Austin and Grice thought that it was almost impossible to provide any sort of conceptual analysis of things unless you clear the ground with 'linguistic botanising'. Carnap may be thinking of the same. Consider the statements:

Sta nevicando
It's snowing
Snow is white

and NOT Finnish for "snow is white"

but ESKIMO. It may be that Carnap is suggesting that if we are providing something like those 'procedures' (Grice would never call them 'rules') we may need to know, even in a more or less indeterminate, way, what the lexeme is to stand for. This, as linguists know, seems to be a bother or complication when it comes to the zillion words, almost, that Eskimos have for 'snow'.

Carnap goes on:

"If the meanings of the words are known, then
the answer to the correlation problem of the
designation relation for sentences can
be solved through

a general function,

which however, is usually very complicated. It is the syntax of the language in question cast in the form of a meaning rule."

----- Interesting. As I say, or hope I said, Jones is the expert here. Carnap seems to be toying with some simplistic view that reduces syntax to semantics? On the whole, logicians seem to be obsessed with providing purely SYNTACTIC accounts of, say, 'validity', -- rather than the other way round. Here Carnap seems to be saying that, if one must, one can provide a specification of the 'syntax' of System (say G) in terms of a "general function" (in Frege's sense -- cfr. his 'Kompositionalitaetprinzip' in terms of application of functions).

Note that Carnap trades in something that would give a headache to, say, M. K. Davies. "Knowledge of language". Is that supposed to be 'tacit', or what? (Davies in "Mind"). Dale himself notes this when he comments on 'in sensu diviso' issues in Lewis -- the 'sensu diviso' (NOT 'composite', as per 'resulting from composition') simplifies, rather than complicates, things for Lewis and Dale.

Interestingly, Carnap seems to be relying on some pragmatist literature, as I think to recall, but Jones may disagree, as to the boundaries of, then:

1. syntactics
2. semantics
3. pragmatics

(I think you can tell a pragmatist by his use of 'syntactics' rather than "syntax").

I.e. it seems Carnap is pretty precise, as he should, about 'syntax' there (note the title of his book, which however, does not mention 'syntax' in the German original -- but 'Aufbau').

Carnap is considering 'semantics' (a 'meaning rule' where we should check what the German for this is) -- a term which, Lady Welby, incidentally, thought ambiguous and tendentious, preferring 'signific' and 'sensific' (she also found 'semiology' and 'semiotics' equally biased or 'theory-laden', I think is the term Dale uses) Carnap writes:

"A meaning rule may (in an elementary case) have the following form."

Jones has studied this in connection with Quine and the ab-use, if that's the word Quine made of all the great teachings he got from Carnap back in Vienna (and environs). This will ring a Gricean ring later, since Grice/Strawson can then be seen as 'defending' Carnapian 'meaning postulates' or 'rules' contra Quine's alleged definitive attack.

But this is the early Carnap, and his form for a 'meaning rule':

"If a sentence S consists of three words,
a noun in the nominative case

---- "Homo"

; a verb in the third person singular, present tense,
active mood

---- "habet"

; and a noun in
the accusative case,

---- "canem"

"then it
designates the state of affairs
that the object of which the first
word is the sign stands to the object of
which the third word is the sign
in the relation of which the verb is the sign."

--- I think Carnap is also implicitly making the point that no Latin speaker would need to rely on the ORDER of things here, since

Canem homo habet

would be just interpreted (as "Canem habet homo" -- or "Habet canem homo", etc.)

Agreed. Grice played with this.

A general point here can be made as to 'meaning scepticism'. One reads of post-Griceans and neo-Griceans thinking that Grice's programme has been found to have been 'defeated'. But one reads the clear prose of Grice, and you find that he was always clear that some tasks were "Herculean". His points were 'programmatic' and 'methodological'.

If Carnap is happy with

"Homo habet canem"

Grice (WoW:VI) is well aware that one cannot expect for a compositional-semantic theory of ALL PARTS OF SPEECH, as it were!

Grice writes, as a preamble to his already long 'shaggy-dog story':

"It would be nice to give a general schema,"

-- cfr. Carnap, 'general function' --

"to show the role of word meanings (covering every
type of word) in determining (in combination) sentence
meanings (covering sentences of any syntactical structure).
But this looks like a Herculean task (in our present
state of knowledge)."

Katz/Fodor were taking notes.

Grice goes onL

"The best we can hope for is a SKETCH, for a very
restricted (but central [philosophically] range of
word types [alpha, beta] and syntactic forms, of a fragment
of what might be the kind of theory we need".

If Carnap's model is

Homo habet canem.

Grice seems to be concerned with the famous/infamous:

The cat sat on the mat.

Or

Fido is shaggy.

i.e. referring to a propositional context which will consist of the ordered pair of an item which is the referent of the subject-term in the canonical sentence, 'Jones's dog', and a second item which is the denotatum of the predicate-term, 'shaggy'.

As we have seen recently, when Jones was discussing Grice's appeal to higher order set theories in his account of 'propositional complexes', Grice knew that even a sketch involved complications. For example, when it came to quantification, proper, he found it easier to deal with

the altogether-dog

and

the one-at-a-time-dog.

for things like

"Every dog is shaggy" or "Some are". These he called 'special objects'. (This is in "Reply to Richards", specififically his commentary on Grandy's and Warner's idea that Grice is illegitimately trading in 'propositions' as gap signs -- or something.

It seems that for this seminal view of Carnap, the rule for the 'semantics' of a system, as Grice would prefer, is all about 'interpretation' and 1-correlation, as Grice would have it.

Note that thigns which look simple may not be. Strawson, for example, would bring a few headaches by considering that some formulae are best seen as not interpretable in this simple 1-correlation way. Some utterances have 'truth-value gaps'.

Grice was at times overwhelmed with the idea of 'truth', that Carnap does not consider (as perhaps he shouldn't) in the account above:

Homo-habet-canem

Instead, Grice wants to say that he needs to specify the conditions of 'factivity' for things like

"Jones's dog is shaggy".

Notably, this is in terms of inclusion. I.e. if the item which is R-correlated with the first member of the ordered pair that constitutes the 'propositional complex' is included in the class of the item which is D-correlated with the second member.

---

In other words, what makes

"Homo habet canem".

One has to look at the extension of "HAVE" (present tense, etc.) and find that it includes the ordered-pair, "Homo, canis".

One concludes: the class "HAVE" does have as correlate the pair "Homo, canis" -- the man does have the dog.

This sort of 'interpretation' seems to proceed on a case-by-case basis, which may be tiresome (to some).

Note that Grice ("on Truth", in WoW:lecture III) indeed notes that his view is for a specification of 'truth' in terms of 'factivity' -- where it is a regimented versions of this untamed concept (cfr. "The taming of the true") that he finds more maneagable. In "Aspects of Reason" he will extend the treatment of 'alethic' utterances to non-alethic modes. Like, "Jones's dog SHOULD be hairy-coated".

Carnap goes on:

"From the correlation problem, we distinguish the essence problem." What IS the 'essence' problem? Carnap writes:

"Here we do not simply ask between what objects the relation obtains, but what it is between the correlated objects, by virtue of which they are connected. The question does not ask for the constitution of the related object, but asks for the essence of the relation itself. Later on,...we shall indicate the difference between science and metaphysics..., and we shall see that the essence problems belong to metaphysics."

-- which is what the City of Eternal Truth is all about! Grice would agree. His big unfinished book he entitled, "From Genesis to Revelations", new essay in metaphysical methodology -- or something.