Tuesday, 14 October 2014

Raisins at Dawn #1

Topics this time include: logical space and Wittgenstein, method and semantic notions in philosophy, the idea of metaphysics, Lichtenberg on selfishness, modality, semantic granularity, and F.P. Ramsey on what the vastness of the universe doesn't mean for us.

1. Apparent logical space. (A more attractive notion, in some cases, than full-blooded contingentism about something.)

In one way Wittgenstein was always emphasizing the breadth of logical space - uncharted bits of it. The road not taken. Yet in another way he is often trying to show that what may look like delimitations of bits of logical space - propositions of philosophy for instance - are actually not such. So, a bidirectional recalibration of our conceptions of logical space. The variety lies deeper, closer to the root as it were. (Alternative concepts, not opinions.)

* * *

2. The peculiar effect of semantically degenerate classifications in philosophy.

'Metaphysical modality'.

Self-fulfilling in a way; Kripke's classification has been part of what has caused subsequent accounts to seem "metaphysical".

Rehearsal of a thought I've written down before: Ross Cameron's question 'What is metaphysical about metaphysical modality?' seems to me a good one and an urgent one. Unfortunately (in my view), he comes up with a positive answer.

I would prefer to see the question as loaded - as embarrassing the proponent of the metaphysicality of metaphysical modality out of their woolly idea.

* * *

3. 'We only care for ourselves' (Lichtenberg). He says it sounds harsh, but not if understood rightly. But it's not really just a matter of getting an odd usage (as with special technical usages, or dialect usages).

Part of it may lie in granularity. But part of it has to do with fighting for associations - for limited resources. (That is something I haven't seen philosophy, especially contemporary analytic philosophy, get clear about.)

Also, the 'that already means something else' move can be used here, when Lichtenberg says 'we love not wife, child [etc.], only the sensations they produce'. We can imagine cases where that form of description applies and shows a contrast, a distinction. (Someone not normal. Not good.) And so 'We all ...' should really describe the crazy-far-out counterfactual scenario where we are all like that.

* * *

4. (16.1.14.) I have for a while been avoiding getting deep into further mysteries connected with modality so that I can get my thesis written and put my existing results in a clear light.

But really perhaps I ought to lift the moratorium now - one, since it may be now or never with some of these issues (for me, I mean), and two, I ought to get a move on even if it's not quite now or never. And three, it may enliven my thesis work by renewing my energies and enthusiasms.

Frontiers include:

- The issue of whether ascriptionally identical propositions can differ in modal status. (Relatedly, substitutivity "salva necessitate" (or whatever the proper Latin is). And: states of affairs, facts, as bearers of modal status.)

- States of affairs, facts, underlying scenarios as opposed to descriptions. (One vs. two-space-ism.)

- Varieties of modality. Machine-space modality.

I have a dim idea that in many cases where we talk about 'the possibilities', e.g. perhaps with a machine, it is in some sense indifferent between subjunctive/metaphysical and indicative/a priori modality, since they may just coincide. And in a sense it may be neither.

Consider '"p  q" is not a tautology or a contradiction; it can take the value 0 or 1' - i.e. 'it can take 0, and it can take 1'.

Now this plainly doesn't mean 'For all we can know a priori, "p  q" takes 0, and for all we can know a priori, "p  q" takes 1', i.e. 'It is not a priori that "p  q" fails to take 0, and likewise for 1'.

Nor does it mean that in some counterfactual situation, the formula takes 0, and in another, 1.

No - the 'space' in question just isn't that of scenarios we can't rule out a priori, nor that of the ways things could have been.

Statements about such spaces may indeed follow from, or imply, the original proposition, but they don't mean the same.

The relevant space here, I want to say, is the space of valuations itself. To bring in circumstances of a worldly kind where the fomula "gets" a value - from whom? - is plainly to bring in irrelevancies, extraneous matter.

We just aren't talking about world-histories, nor bits thereof.

Now there is more than one striking thing here, as it were simultaneously. More than one big difference from both subjunctive and indicative modality.

One is, as I feel like putting it, that it's all abstract. It seems like no worldly happenings, actual or otherwise, are under discussion.

Another is that both the naive subjunctive and the naive a priori analyses just seem wronh.

Now this may seem like just one thing, but it seems to me there may be cases which aren't all abstract, are about happnings, but also apparently not amenable to the naive analyses.

(This would perhaps suggest that their failure in the "p  q" case is as it were overdetermined or double-barrel. This then raises the question - or seems to - of whetyther there are cases which fail just in the "abstract" way, not the way common to the "p  q" case and the one I'm about to float.)

However, with the above parenthesis, I could easily be getting ahead of myself, since the sort of case I'm about to consider may perhaps very arguably be amenable to a subjunctive- or indicative-space-based analysis.

What I have in mind is ability claims. 'I can lift this weight'.

I want to break off here and separate three kinds of case for possible future reference:

(1) "p  q" can take 1 or 0.
(2) I can lift this weight.
(3) This machine can go into this and that position.

and (3') (or (4)?): I can authorize this transfer.

Also, similar to (1) but superficially like (3): 'A Turing machine such that ... can ...'.

* * *

5. What is needful now for my work is to take stock of my way of thinking as compared with highly visible analytic-philosophical ways of thinking surrounding my topics, and get clearer about what main moves and shifts are required.

So that I can give directions by which one might get to my conception.

An example: the methodology of talking about, accounting for meaning. The nature of semantic concepts.

* * *

6. Consider the real meaning of 'Yes, that's one way of describing it'.

* * *

7. It is important to emphasize the narrowness of the problem of metaphysical modality, in order to help diffuse Lewis's 'Well, I see no cheaper way' rhetoric with respect to my question [i.e. giving a non-trivial analysis of the concept of metaphysical necessity as an attribute of propositions].

Of analyzing all modality, Lewis could perhaps say that (which isn't of course to say that his analysis is good enough, or any good at all - only that nothing seems better). But with respect to my question it seems very premature indeed.

Oddly, it is far from clear that anyone has tried to solve my problem without necessarily analysing all modality. (There are of course modality-involving constructions of possible worlds, but they can be argued not to fit the bill. To leave buried stuff which my account brings out into the light.)

Current frontiers in my work:

- The difference between subjunctive/metaphysical modality and 'all non-epistemic modality' - the former being narrower.

- Its connection with the question of the reductivity of my account.


A priori possibility vs. metaphysical.
A priori possibility vs. 'can' of ability.
- 'Can' of ability applied to discrete machines vs., say, people. So different, yet a strong similarity, a strong likeness of flavour. Feels, as it were, like the same meaning (however, at a finer granularity, perhaps distinct).

(This talk of fine-grainedness, in this case, is connected with Wittgenstein's remark about language showing us 'new tricks at every turn'. Also, his remark on the subject-predicate form 'seen in this way' actually comprising many different logical forms.)

* * *

8. I need not block any part of the derivation in Kripke's puzzle. Need not take issue with any of his principles. Which isn't to say there aren't problems - for example with the principle of translation.

It is enough that I can accept both conclusions; they don't really contradict each other, operating at different granularities.

(I could say that the translation 'Londres' ---> 'London' induces a coarse granularity, though.)

* * *

9. The methodological stuff connected with my account of propositions is very important.

(There is a lot of methodological prejudice about what kinds of concepts are admissible. And about where to stop explaining.)

Also, the place of that methodological stuff in my larger system is important. It's raised especially by my notion of inherent counterfactual invariance [which I use to analyze subjunctive necessity de dicto] - Russellian propositions or sets of possible worlds, for example, don't seem like they could bear such a property.

* * *

10. Apply granularity to the verbal/genuine disagreement distinction.

* * *

11. Granularity considerations give us a way of avoiding widespread error theory about people's ordinary meaning statements (something very hard if not impossible to avoid from many analytic philosophical perspectives).

* * *

12. It would be worth articulating what is so unhelpful about classifying philosophers as holding 'direct' vs. 'mediated' reference theories. ('Mediated Reference Theory' is currently the title of a Wikipedia article.)

* * *

13. Apparently loose talk about giving different meanings to, e.g., the concept of mass (not the word 'mass') in Einstein and previous physics. Light to be shed by granularity considerations (roughly: 'concept' here indicates a coarser meaning).

* * *

14. Ramsey gave a paper to the Apostles about his picture of the world, where he emphasizes that physical size isn't that important. That is, if it's a fact that we're the only life in some mind-bogglingly vast region of the universe, or even the whole universe, that doesn't really mean much for what is and isn't important. This not being hung up on size seems wise to me, and an antidote to a lot of half-baked thinking.

* * *

15. I should be quite upfront about the fact that I believe that some deep and hard-to-describe kind of change of perspective is needed in philosophy of language. A rethinking of what a solution to various problems would look and feel like.

In a sense what we need isn't new and undreamt of apparatus - actually that is nearly wrong, but I'm thinking of 2D semantics. A large part of what's needed is to recover certain (kinds of) ideas, and ways of thinking, for philosophy.

A sorting through - a revision - of scruples. Very difficult.

One manifestation: "real" semantic concepts feel unrigorous, tabooed, suspect. Their subtlety and myriad-ish nature, as well as connections with other philosophical problems, feed into this. (Granularity being a case in point of such subtlety.)

('Back to the rough ground!')

* * *

16. An important dimension along which philosophers differ (including from themselves across time or topics): the extent to which the principle of charity is applied to received - i.e. ordinary - ways of thinking. Also, the extent to which ideas are set up as conflicting.

For example, even if a theist may be led by polemics to all sorts of unwelcome-to-them stuff, the polemics may be dross. We may say in cases like this - and there are important examples not to do with religion that I need to find - 'No! Stick to your guns! Your view doesn't imply this at all!',

(Difference as to what is essential.)

Connected with this is Wittgenstein on 'making propaganda'. I could elaborate the idea as 'Regard this as essential' or, perhaps more importantly, 'Don't regard that as essential'.

Applications: Platonism, the theory of knowledge-statements, physical object statements in connection with the brain in a vat scenario.

* * *

17. Nick's palming off of the nub of Kripke's puzzle - does 'Londres' have the same meaning/sense as 'London', or in Nick's terms do they correspond to the same formal constant - as a difficult empirical issue is very important. (Importantly wrong.) And it is a move characteristic of our intellectual times.

['Nick' here refers to N.J.J. Smith, and the move in question was made in the Q&A following a talk he gave at Sydney University called 'Propositions: The Missing Ingredient'. He has not yet published any of that work, and I am not trying to pin any settled view on him here. It was just something he said one time. Below I insert some paragraphs from elsewhere which deal with this matter and contain an argument to the effect that the move in question is a mistake:

Philosophers who are, unlike Quine, serious about semantic concepts, but not given to granularity considerations (through unfamiliarity, prejudice, or a mixture of both) may fall into the trap of palming off the issues in connection with which granularity considerations come into their own as difficult empirical ones, to be solved by scientists of meaning. (This reminds me of the way Russell and the early Wittgenstein palmed off issues about their 'elements' and 'objects' as empirical ones.)

I just don't think that's going to stand up to scrutiny. How could an empirical investigation of the Pierre situation sort out once and for all whether Pierre believed, when in France, that London is pretty? If it just gives one answer or the other, how is it going to remove the dissatisfaction we will feel at (apparently) being denied what we want to say from the point of view which was not favoured by the verdict? (In fact, the very fact that the Pierre situation is underspecified, open and fictional, and yet still definitely puzzling, makes it hard to see how any empirical investigation could be relevant, even to real cases which are like this. For anything it may reveal could just be stipulated one way or the other in the Pierre story, and the puzzle would remain.)

(That said, a more technical project of 'semantics' might be distinguished from what I'm doing, and there it may be stipulated that there is to be no granularity business, but a single univocal carving up of expressions. Perhaps empirical considerations may come into this. But I'm more interested in working with and refining a more ordinary concept of meaning, of the sort used in regular thinking.)]

Meaning as role or place in the system of language. Whose system, though? (Fine structure, idiolects.) Here perhaps also lies the explanation for why it sounds funny, and can seem to philosophers to be a definite error, to talk of names as having meanings. [Two articles here follow this up.]

'Use' on the other hand - undeniable.

No comments:

Post a Comment