Tuesday, 5 February 2019

Two Sources of Interest in Metaphysical Modality (and Kripke in Light of Them)

Source 1: We want to know the vocabulary and syntax of being, or as Rosen puts it in 'The Limits of Contingency', 'the combinatorial essence of the world'. What are the basic elements and how may they be combined?

Source 2: We want to know, as it were in advance, whether various kinds of statements that are not put in terms of the basic elements count as high-level descriptions of any possible world. 

Many of Kripke's arguments that certain kinds of statements are necessary furnish considerations which purport to show that whatever the possibilities are exactly, none of them is correctly described as one in which '...'. 

This mode of argumentation on Kripke's part can make it look like his modal notions can ultimately be explicated along conceptual or semantic lines. But, as Putnam came to appreciate in between 'The Meaning of "Meaning"' and 'Is Water Necessarily H2O?', this is not so.

In this connection it is notable that all of Kripke's distinctive modal theses are negative as regards possibility. (His claims, in the course of arguing against descriptivism about names, that well-known facts about Aristotle etc. could have been different, are an exception, admittedly - but for those arguments, I don't see that he needs these alternative possibilities to be real, metaphysical possibilities. The modality used in those arguments could be deflated to conceptual or semantic without affecting the arguments, which are after all for a semantic conclusion - that names aren't synonymous with descriptions.) As far as I know, Kripke never seriously argues that such and such really is possible, really is a way the world could have been. Rather, he just works on the assumption that there are many quite various ways things could have been, but then seeks to draw some limits in high-level vocabulary.

But these Kripkean results, if that's what they are, only satisfy interest in metaphysical modality that derives from the second source. How we might satisfy our interest that derives from the second source is largely left untackled, and this is one reason why many have found Kripke's work frustrating. He elicits epistemic hopes, someone might complain, without giving us even so much as a roadmap for how so satisfy them.

His suggestion that metaphysical possibility may coincide more closely with physical possibility than has often been supposed may however be suggestive. On the other hand, he is against physicalism, so this couldn't be the whole story from his point of view.

Thursday, 10 January 2019

Rigidity and General Terms: Two Different Analogues of the Singular Case

This post wrestles with and begins to settle on a view about the confusing issue of how Kripke's notion of rigidity may apply to general terms. 

One analogue of rigidity for general terms: how about we think of it as a rigid connection between properties (or alternatively, an additional connection between the "rigid" term and a further property).

So for example, 'water' in the first instance picks out the property of being water, which is tied rigidly to the property of being (mainly composed of) H20.

Or 'cat' is tied in the first instance to the property of being a cat, and that is tied rigidly to the property of being an animal.

But now there is another kind of thing which seems different, and comes up with sentences like 'John has the property we talked about yesterday'.

Suppose the property we talked about yesterday was the property of having the property that is discussed in Book A. And suppose Book A contains a discussion of the property of redness. Now 'has the property we talked about yesterday' rigidly designates the property of being a property we talked about yesterday, but it also non-rigidly designates the property of having the property that is discussed in Book A, as well as the property of redness.

This motivates the picture of, behind a predicate, a stack of properties, where the top one is designated rigidly, and the ones below not.

Problem: we might want to say that 'cat' rigidly designates a certain kind of animal. And I may then want to rephrase that as: 'cat' rigidly ascribes animality. 

But doesn't the 'rigid' bit here fall away? Take the phenomenological, underlying-nature-neutral counterpart of 'cat' - 'catty thing'. Now even if in our world all the catty things are cats, it doesn't sound right to say that 'catty thing' ascribes the property of being a cat at all - it's not that it ascribes that property, only non-rigidly. 

So now it is beginning to look like the distinction we are after here is between a term merely covering things with property P, vs. ascribing to them property P.

But then that seems wrong when we go back to 'John has the property we discussed yesterday', since if what we talked about yesterday was the property of redness, there is a sense in which that sentence ascribes redness to John. 

This is hell!

But this whole problem, occupying the last few paragraphs, perhaps only arises from mixing together two different analogues of rigidity that we get when we look at predicates.

It may be protested that  'John has the property ...' is not a property ascription syntactically at all, but rather a 2-place relational statement with a non rigid second term.

Be that as it may, we can still classify 'has the ...' as a predicate and can still talk about a rigid/non-rigid distinction. And so I think we need to recognise that there are at least two quite different things going on here - two different things which are a bit similar to rigidity/non-rigidity as applied to names.

One is the difference between 'has the property discussed earlier' and 'is red'. Another is the difference between 'is water' and 'is watery' (one brings being composed of H20 along with it in counterfactual scenario descriptions, and the other doesn't), or 'is a cat' and 'is a catty thing'. 

One reason the second analogue may be counterintuitive if presented as a kind of rigidity is that in the case of singular terms, rigidity is associated with simplicity (both syntactic and semantic), but in the case of predicates (let's look at 'is gold', 'is water', 'is a cat' and put aside 'has the property...') it's the opposite. The "non-rigid" predicates just don't take any further property along with them, but the rigid ones do. I.e. 'is a catty thing' or 'is catlike' just picks out one property, but 'is a cat' is tied to the further property of being an animal.

Actually, the 'any' in 'any further property' is probably wrong! Maybe all predicates rigidly take some further properties along with them. So this second sort of "rigidity" we can talk about in connection with general terms should be thought of as relative to whatever further property is in question. (For instance, 'is a pencil' is arguably counterfactually locked to 'is a physical object'. So 'is a pencil' rigidly picks out physical objects - we might want to say something like that.)

One thing that is emerging here is that the 'has the property discussed earlier' vs. 'is red' thing is one distinction which pattern-matches with Kripke's discussion of rigidity as applied to names, but there is also another thing going on - predicates dragging further properties along with them in counterfactual scenario descriptions - which actually corresponds better with Kripke's informal applications of the notion of rigidity to general terms.

Now it looks like the general-term-"rigidity" considerations in Naming and Necessity are actually closer to the "necessity of constitution" and similar considerations than they are to the "necessity of identity" considerations.

Background Reading:



- Kripke, Saul (1980). Naming and Necessity. Harvard University Press.

- Soames, Scott (2002). Beyond Rigidity: The Unfinished Semantic Agenda of Naming and Necessity. Oxford University Press

- Salmon, Nathan (2004). Are general terms rigid? Linguistics and Philosophy 28 (1):117 - 134.

- Linsky, Bernard (2006). General Terms as Rigid Designators. Philosophical Studies 128 (3):655-667.

- Martí, Genoveva & Martínez-Fernández, José (2011). General terms, rigidity and the trivialization problem. Synthese 181 (2):277 - 293.

- Schwartz, Stephen P. (2002). Kinds, general terms, and rigidity: A reply to LaPorte. Philosophical Studies 109 (3):265 - 277.

- de Sa, Dan López (2007). Rigidity, General Terms, and Trivialization. Proceedings of the Aristotelian Society 107 (1pt1):117 - 123.

- Marti, Genoveva (2004). Rigidity and General Terms. Proceedings of the Aristotelian Society 104:131-148.

- Zouhar, Marián (2009). On the Notion of Rigidity for General Terms. Grazer Philosophische Studien 78 (1):207-229.

- Orlando, Eleonora (2014). General terms and rigidity: another solution to the trivialization problem. Manuscrito 37 (1):49-80.

- Gómez-Torrente, Mario (2004). Beyond Rigidity? Essentialist Predication and the Rigidity of General Terms. Critica 36 (108):37-54.

- Kosterec, Miloš (2018). Criteria for Nontrivial General Term Rigidity. Acta Analytica 33 (2):255-270.

Wednesday, 26 December 2018

Moral Concepts as Unsystematic on the Introduction Side

Here is a metaethical idea which I find appealing and under-explored. What if what makes metaethics and the cognitivism vs. non-cognitivism debate so difficult is that we have failed to consider a certain possibility. Namely, that expressions can count as expressing concepts if there is enough systematicity in what happens once an application is accepted, even if there is much less systematicity in what happens before an application is accepted. 

On the view I am interested in, the reason moral terms count as expressing concepts at all, and count as linguistically meaningful, is that they are fairly systematic with respect to what lies downstream from them. This is compatible with them being much less systematic with respect to what they lie downstream from - and it looks like they are much less systematic in this respect.

By analogy to natural deduction systems in logic, we might think of moral concepts as coming with elimination rules, but no definite, agreed-upon set of introduction rules. 

So, concepts like right and wrong have fairly systematic and shared relationships to, roughly, what follows from statements involving them - in terms of other statements following logically, but also more probabilistic effects, and motivations and actions flowing from their acceptance. This systematicity is what makes them concepts - and it is enough! When it comes to what licences their introduction in the first place, on the other hand, there just isn't much there in the way of tight, systematic connections. Or there may be quite a bit there, but still not enough to, for example, motivate the view that these concepts are co-extensive with any of what we'd happily call descriptive concepts. 

On this view, the famous Open Question Argument comes as no surprise. Maybe you can always, under any descriptive supposition, intelligibly still ask 'But is it right?' - and that is just because there simply are no tight systematic links of the right sort between uncontroversially descriptive concepts and moral concepts. 

(By the way, the view isn't that all such concepts - concepts with unsystematic introduction behaviour - are moral concepts. The moral concepts are a subset of these, and what distinguishes them as moral as opposed to, say, aesthetic, has to do with what lies downstream from them.)

I wonder if this sort of view has been anticipated, as it strikes me as having real potential.

Thursday, 6 December 2018

Different Attitudes about Meaningfulness in Philosophy, and Their Consequences

What makes a question or a statement meaningful as opposed to nonsensical? This is a hard question, and perhaps not a very fruitful one in philosophy since the falling out of favour of verificationism. But what if we instead ask: what difference does it make whether we think something makes sense?

I have sometimes been struck by a certain kind of self-conscious, almost defiant claim of meaningfulness that can be found in work by philosophers writing after the resurgence of metaphysics in the second half of the twentieth century. You may have noticed it yourself. Often in the course of making a point, a philosopher will say something like 'Well, the question is certainly meaningful...' or 'Whether or not it is true, there can be little doubt that this is a coherent position'. 

This liberal attitude to the question of whether something even makes sense contrasts starkly, and is often expressed in reaction to, the wary attitude of other philosophers, especially earlier in the twentieth century.

Often, the liberal attitude is supported with the idea that the wary attitude depended on a discredited view - verificationism - about what it takes to be meaningful.

Both attitudes, especially when taken stridently or dogmatically, can seem stupid. The denier of meaning can seem like a shirker. Or like someone who doesn't want to play a game and justifies it by acting superior.  The denier can seem crude - like they have this stupid hammer that they wield instead of taking the trouble to understand something (or at least admitting that they can't or won't). 

But the affirmer of meaning can also seem stupid and crude. As though they are blind, perhaps even wilfully blind, to subtle problems and difficulties. In the context of doing philosophy, this may seem like especially bad form, in that such problems and difficulties are just the thing that a philosopher should presumably try to be alive to.

It is quite clear that both tendencies have their pitfalls, and that both can be justified in particular cases. But the question of which attitude is right overall, or in a particular area, or in a particular vexed case even, seems peculiarly intractable, in something like the way moral and aesthetic issues can be intractable.  

I have become interested in thinking about this from the point of view of the consequences of the different attitudes, focusing on output rather than input. That is, rather than inquiring into when it is right to deem something meaningful, let us try to think about what might ensue given an attitude to the meaningfulness question. And that may then shed light back on the input question. 

A liberal attitude to the meaningfulness question can be refreshing, in that ideas and views get engaged with and clarified, in a way that they don't seem to if you're being skeptical about meaning. For example, someone who is very suspicious about the meaningfulness of the questions and views in what is nowadays called temporal ontology, e.g. whether the past and the things in it exist, is not, at least while being suspicious, likely to note certain (apparent) logical relationships between these views and others. For instance, the relationship to the question of whether true statements (perhaps of a certain kind) are all made true by some existing thing. Noticing these things seems to require that we, at least for a time, don't stand back and worry about meaningfulness, but come forward and handle these ideas - i.e. that we treat them as meaningful, and do the things that we do when we try to find out whether things are true or false.

But a warier attitude to meaningfulness can also be refreshing, leading us to take (in Wittgenstein's phrase) a wider look around. Fresh considerations may arise in this way, where from inside the meaning-granting perspective they would not. For example, someone suspicious about the way questions in temporal ontology are debated in philosophy seminars and publications may find themselves wondering what is really going on here, what is really animating all this discussion. They may think suspicious thoughts about the emotional associations of some of these ideas playing more of a role than the broadly scientific, "theory choice" mode of discussion of analytic metaphysics would suggest. And such thoughts may be fruitful. 

Understanding this better may help us avoid stupidities on both sides, making us more nimble and sophisticated. We don't have to pick a posture and stay in it forever. Even if a question or a view is ultimately dubious as regards meaningfulness - or needs to be transformed based on a better understanding of what is really at stake - taking the liberal attitude and working with the question or view for a stretch, i.e. treating it as meaningful enough, may yield insights which survive the transformation, or can be transferred to other places. And we may be more prepared to let this go on if we realise that it doesn't preclude us from taking the other viewpoint, i.e. getting suspicious and sniffing around more widely. And the same going the other way. One thing is that, even if the question is ultimately just really hard and not at all meaningless or in need of transformation, we may not know that, and so encouraging some suspicious minds to have a sniff around may be good. Another thing is that it may lead to a more subtle kind of improvement, where we may not radically reformulate our questions and views, but fold in new criteria and considerations that come from a wider look around.

Thursday, 4 October 2018

Notes on Modality II - "Deserving the name", the grandeur of the idea of metaphysical modality, &c.

I want to explore a 'no clear best deserver' view of the - very inchoate - role that the distinction between (metaphysically) necessary vs. contingent truths is supposed to play.

The idea is that once you drill down in the borderlands, you see that there are different ways of going.

So no clear best deserver. But also: it's not clear that anything is a good enough deserver.

The 'no clear best deserver' hypothesis raises questions: why wouldn't everyone agree? Why would this be news?

One reason might be that there really is a best desever. Or that it seemed like there was a clear best but on further reflection there are competitors and it's not clear that any one has the edge. But I think there's more to say about why this might be news. There is something very natural about thinking that there is some single distinction here. There is a strong inclination to think that there some one major, fundamental distinction between the truths that could have been otherwise and the ones that couldn't. 

So if there is a sense in which this is not so - or if we want to be circumspect about it - we could try to tell a positive story about how this comes about.

Also, it is important to remind ourselves that the strong inclination I am speaking of may not be universal, or even normal. (Think of people who encounter this philosophical discourse, feel vaguely skeptical, maybe kind of interested but not particularly compelled, don't sign up for any view in particular, or flirt with some unusual view (an undergrad fellow student of mine defended the view that all truths are necessary in a course essay), and simply move on with their lives and don't ever really think about the matter, except as something some people are concerned with.)

But, at the same time: it's not just some specific problem in some particular subliterature. Versions of it, or closely related things, come up in philosophy at many times and places. And I have the feeling that it springs from a kind of root that also gives rise to certain philosophical thoughts and frames of mind that all sorts of people have.

So, how might it come about? Perhaps: Some kind of drive towards systematic understanding run rampant. Desire to see world as mechanism with discrete states. And the point isn't that that's naive, or probably not true after all, or something like that. But one issue here may be: why should there be one clear best way of seeing the world that way? 

It is possible to get into a frame of mind where it seems like there must in some sense be a single or primary set of facts about the possible states of the world, if there really is a world at all. It is like Wittgenstein's demand for substance, for eternal objects, in the Tractatus. It can seem like if the thing you're demanding weren't so, there would be no world at all, or nothing would be true.

There's something here I want to say in connection with this 'no clear best deserver' idea, along the lines of: how little is said when the idea of metaphysical modality is introduced! How could there be enough specific information there - in phrases like 'really could have been that way' - to narrow things right down to a particular, clear distinction!  It feels completely "lay", really quite free of theory (but of course you can theorise about 'really' or formulate it in terms of 'some fundamental sense' and then theorise about that).

But on the other hand, isn't that a bit fallacious? Can't specific, detailed things come out of arrangements of ordinary, quite general concepts? And don't, for example, specific seeds produce specific trees, despite being so small?

So this thought needs to be stated carefully. There's not some general rule here that's being appealed to. Still I think there's a point here, a way to see something  - a way to break the grip of something.

The words and associated ideas we're using here are quite flexible things, on reflection. 'Things', 'really', 'could'.

You might say there's a kind of naive level of "buying into" the idea of this one major distinction between contingent and necessary truths. Just not thinking about it too much. But then if you've thought about it and seen it threatened or scoffed at or considered uninteresting, but still want to pursue it - OK, now you're after something. Now you have a dream, I feel like saying.

I think, awkward as it is to talk about, there is really no getting away from the fact that there's something fantastic about the idea, something grand. '"How things really could have been" - what could be more interesting than that?!'

* * *

Consider Kripke's suggestive remark at the end of Naming and Necessity about leaving open the extent to which metaphysical modality will turn out to be closer to physical than suspected.

It's remarkable how little investigation or facing up there has been to the issue of how such a thing might be decided. (Williamson talks about 'detailed theoretical investigation' but I can't help but feeling that this is a kind of mirage, like how Russell and the early Wittgenstein palmed off the question of what the logical atoms will turn out to be.)

Decades later, we seem to just have different viewpoints - some rationalists who are far from thinking it coincides more with physical possibility than you might think, and others who are captured by the idea of empirical insight into hard limits (due to what seems to me a kind of misinterpretation of the necessary a posteriori, but working out where the stand offs there lie would be good).

The picture I have now is this: there's a certain inchoate role, inchoate requirement or job description for this idea of the ways things really could have been (metaphysically), and then it recedes into the background when we look at cases and theories. But this needs to be scrutinised more, and this will shed light on skepticism about the notion, and on different positive views.

I don't mind if you settle on a best deserver and are happy with it, but I will want to be clear that this is what is happening, and be wary of an alternative way of thinking about the matter as something which has been investigated and a view found correct. <-- Inadequate, but there is an important feeling here that I don't want any smoke and mirrors.

A large part of the worry is: that this process of coming to rest upon a choice of best deserver gets misrepresented as investigation into how things could have been. A failure to distinguish between getting to a concept and investigating what it applies to. But there are dangers with this hypothesis too.

'Philosophers framed the question of how things really could have been, and detailed philosophical research has led to a convergence on the view that this turns out to be a matter of ...'

In terms of what we want from the term, what gives it its life and feeling of importance - this is so easy to write off as a trifling, preliminary thing. Almost indecent to talk about. But it is very important. And it's amazing how far it reaches, so to speak. 

Friday, 10 August 2018

Notes on Modality

Some notes from my post-PhD attempts to get back to the really deep problems surrounding modality.

The truth-conditions framing of the puzzle of modality makes it vividly felt. In the case of subjunctive (counterfactual) modality, it even leads to a fruitful answer. But at base, puzzlement remains, and the seeds for solving it are not to be found in the truth-conditions frame.

Likewise, I think, for a grounding frame. However, if grounding is thought of in explanatory terms, there may be a way in.

One problem with the truth-conditions frame is that not everything can be given informative truth-conditions, including discourse which we are not inclined to problematise. Actually describing and isolating surprising features of modal discourse, e.g. with concepts like ‘intensional’, ‘hyperintensional’, ‘referentially opaque’, gives something beyond the general appeal to the question of truth-conditions. Something more to work with. This suggests that a frontier here is: giving better and more refined such descriptions.

Another frontier which is easy to neglect once you take modality philosophically seriously, but really ought not to be neglected: what is it that is bothering the skeptic? What might their view come to?

Falsity in conventionalism. Elements of truth. (Later Putnam on H20.)

‘You can problematize anything like that’ (of the truth-conditions frame). Yes, but today it is this what we have problematized. And you haven’t solved it by pointing that out.

(When does, and when doesn’t, that sort of response - ‘You can always…’ - work? (Compare the perfect island objection to the ontological argument.) Well, the problems are distinctive in the case of modality and the truth-conditions frame. The truth-conditions frame is not just a trick that does the same thing every time, so to speak.)

A couple of particularly deep problems of modality get bypassed, partly because the area is so fruitful and leads to other investigations. (E.g. subjunctive modality’s relation to the a priori. The problems of epistemic modals.)

My PhD research trajectory is an example of such a deflection.

Unity of modality - another problem. Why the same words for all this? (Think about this question and how there may be no satisfactory answer. Here it may be worth trying to imagine other possibilities.)

The unity problem can seem silly in a way, but we can’t really answer it, or say convincingly why we shouldn’t have to. What, exactly, if anything, is wrong with the question?

‘Obstacles’: just shiny problems we get attracted to, distracting us from the big game? Or real obstacles that we can’t just pass by if we want? I.e. does not clarifying the relationship of subjunctive modality to the a priori threaten to make trouble for our efforts at the deeper problems.

The problem under the truth-conditions frame is closely related to the metaphysical feeling of mystery. (The feeling of a ghost.)

Idea: modal yearning is a diffuse yearning for models, systems.

In this way the problem of modality boils down to the general problem of metaphysics.

Modal yearning as a useful heuristic, which then gets a metaphysical application. A metaphysical interpretation.

I look at things in a peculiar way and feel the problem of modality.

This modal yearning angle also explains the attraction of necessitarianism. That doctrine is a soothing balm to one whose modal yearning is bothering them. The pressure to find the system is off. (God’s mind. Copy of the world.)

‘Eliciting the problem’.

My former teacher McDermott’s odd view on metaphysical modality - that it coincides with physical - can make better sense if we think of the dynamic between ‘This is the best deserver of the name’ and ‘Nothing deserves the name’. Or: ‘I don’t know whether anything deserves the name’. ‘I can’t tell what it would take to deserve the name’.

The metaphysically wary person may yet be forced into trouble with wedge-driving argument. ‘Well, you must admit we could have sat over there instead?’, ‘You must admit that we couldn’t both have sat here and not sat here?’

I may be a skeptic myself, if that just means that no one notion deserves a name like ‘metaphysically necessary’. But I used that term as the received label for a (type of?) concept I was interested in.

Sometimes, when thinking skeptically about the puzzles of modality, a realistic impulse asserts itself. Then, it begins to seem like denying the existence of some kind of basic, metaphysical space of possibilities is tantamount to denying that there is a real world.

What does it mean to say that the world has moving parts?

‘There is no privileged system of the world, no privileged way of identifying moving parts and their ranges of movement.’

It is quite easy to feel skeptical when pondering the idea of a complete space of possibilities, or possible worlds.

(There is a use of words like ‘fixed’ in philosophy which I want to call idiotic, or at least very crude. ‘So-and-so believes in fixed meanings’. ‘A fixed set of possible worlds’.)

Given this picture of the complete space of possibilities, we might ask: what is the alternative?

The picture all by itself isn’t harmful, presumably. It can be applied in a way that is not guilty of anything.

* * *

(The below section of these notes was prompted by reading a draft paper by John Burgess called 'Modal Logic in the Modal Sense of Modality'.)

My thought that the deep puzzle was there before Kripke, and that Kripke’s innovation’s main consequence with respect to it was a clearable obstacle, is there in Burgess. That the a priori bit at the core of modal epistemology is still clearly there doing its thing, and that this really just takes us back to where we were before Kripke, as far as the deep puzzle is concerned.

It is interesting that Burgess, when he wants to get to the heart of the matter surrounding the difference between views on which necessity ‘derives from us’ and those on which it doesn’t, uses theological language. He then notes that this would have been a natural way for philosophers in the 17th and 18th centuries to discuss it, but that we probably must today take it as still metaphorical. (And I think it clearly is, even from most theistic perspectives, if they be at all sophisticated.) And so arguably this theological gloss is not much better than the ‘origin’ talk. So he says he’s going to try again.

The next attempt ends up being based on the metaphors of reflection and projection. The modal facts are there, part of the world as it it in itself, vs. they aren’t, rather being a projection of our interests. Burgess, getting into profound waters, writes:

Perhaps I should give up the attempt to get rid of all metaphors, and admit that the two views I am discussing are not so much philosophical theses or doctrines as ‘metaphilosophical’ attitudes or orientations: a stance that finds the ‘reflection’ metaphor congenial, and the stance that finds the ‘projection’ metaphor congenial. But let me try a third time to describe the distinction between the two outlooks in literal terms, avoiding optics as well as theology.

Now comes the attempt:

To begin with, both sides grant that there is a correspondence or parallelism between two items. On the one hand, there are facts about the contrast between what is necessary and what is contingent, for instance, between 29 being a prime number and 29 being the number of years it takes for Saturn to orbit the sun. On the other hand, there are facts about our usage of modal auxiliary verbs such as ‘would’ and ‘might’, and these include, for instance, the fact that we have no use for questions of the form ‘Would 29 still have been a prime number if such-and-such?’ but may have use for questions of the form ‘Would 29 still have been the number of years it takes for Saturn to orbit the sun if such-and-such?’ The difference between the two sides concerns the order of explanation of the relation between the two parallel ranges of facts.

And what do I mean by that? Well, both sides grant that ‘29 is necessarily prime’, for instance, is a proper thing to say, but they differ in the explanation why it is a proper thing to say. Asked why, the first side will say that ultimately it is simply because 29 is necessarily prime. That makes the proposition that 29 is necessarily prime true, and since the sentence ‘29 is necessarily prime’ expresses that proposition, it is true also, and a proper thing to say. The second side will say instead that ’29 is necessarily prime’ is a proper thing to say because there is a rule of our language according to which it is a proper thing to say. This formulation of the difference between the two sides gets rid of metaphor, though it does put an awful lot of weight on the perhaps fragile ‘why’ and ‘because’.

Here the late Wittgenstein passage (from PI 590 - quoted below) can get in between these views. The explanation can go through language to the world, so to speak. What we have use for is not a completely arbitrary matter, if we think of the used items as not bare signs, but signs in a system of language, a body of usage.

...even supposing the pragmatist view is the right one, and the problems of the epistemology of modality are dissolved, still the pragmatist side has an important unanswered question of its own to address. The pragmatist account, as I formulated it earlier, begins by saying that we have certain reasons, connected with our various purposes in life, to use certain words, including ‘would’ and ‘might’, in certain ways, and thereby to make certain distinctions. What the pragmatist owes us is an account of what these purposes are, and how the rules of our language help us to achieve them. Addressing that issue is or ought to be at the top of the pragmatists’ to-do list.

I think this above bit needs careful thought, and that we need to be on guard against oversimplifying ideas about language having a purpose. Its multifaceted character needs to be kept in mind here. It seems to me that the reasons and the purposes aren’t really to be identified, and aren’t really the focus anyhow. We have rather to look into language itself, the working of it, and what we can and can’t ‘do something with’ - this doesn’t require that we have a settled idea of what we want to do with language. So I think there is a way which falls on the ‘pragmatist’ side of Burgess’s line, in that language (meaning) gets into the explanatory act, but where the to-do list doesn’t look like Burgess says it does or ought to.

After discussing a modal ship of Theseus paradox:

Moreover, if the question ever did prove to be, in some specific case, of practical importance, then according to the pragmatist view we could then settle it by fiat, making a new rule for ourselves, since it is rules we have made to which the answers to all questions about modality are ultimately to be traced.

This is dubious to me. Sometimes ‘making a game more definite’ can be protested to be really just moving to a different game. And thus there is, I think, an important sense in which the original question was inherently borderline, inherently not a clear case. Compare ‘Is this man tall?’ (where he’s sort of tall but not really). There’s something wrong with saying that we can just stipulate that the answer to this is ‘yes’. For then we have a concept (attached to ‘tall’) which is just different from the one we had where we didn’t know what to say.

This discussion of Burgess doesn’t get into the weeds of the whole space of possibility, how extensive it is, etc. Or the idea that there’s a lot of indeterminacy there, or actually different notions going by the name ‘metaphysical necessity’. Rather, he just takes a fairly standard off the shelf picture of there being some modal facts here, and contrasts two attitudes to the origin (metaphorically), optics (metaphorically), or explanation (non-metaphorically but perhaps ‘fragile’) of those facts.

‘By contrast, to those who take a pragmatist attitude there will only be the potential for a practical problem, a potential that is thus far apparently unrealized, and that if it ever were realized could always be solved by fiat.’

This I don’t think is right. Similarly I want to rebel against Burgess’s idea about what is on the to-do list of his pragmatist. To be fair, he stipulates that his pragmatist starts with the idea of ‘certain interests and purposes’, but if the core of the pragmatist attitude is just that explanations of modality go through language and meaning, then this starting with ‘certain interests and purposes’, and their foregrounding, seems to me to be an optional extra, and one which I would want to consider doing without.

This, I think, leads to a stronger position. One advocates going through meaning when explaining modality, but also is careful to be faithful to language and meaning as it really is - not a thing with ‘certain purposes’ in any strong sense, and also something which can be inherently indefinite (i.e. that certain expressions and concepts are inherently not sharp, and can’t be made definite by ‘fiat’ without changing the game).

Taking this position, it becomes clearer why we scratch our heads in some cases, and don’t just consult our purposes and lay down new rules.

* * *

One problem with the basic intuition pumping ‘We could have done otherwise today’ way of motivating the concept of metaphysical modality is that it admits of a retrospective epistemic reading which does not obviously imply anything about ways things really could have been.

Here it might be thought: for it to become clear that we mean metaphysical rather than retrospective epistemic modality, we need negative examples - necessities, i.e. claims that such and such could not have gone otherwise. The thought is: if that’s right, and it’s not right under the retrospective epistemic reading, then there must be another reading.

* * *

Trying to really follow through on a maximally realist, metaphysical viewpoint is worth doing. Call the facts objective, say they aren’t projections, talk about our intellectual access to them and let yourself have a special way of knowing. Work in that framework and then see what the questions are.

One that hasn’t been given enough attention in my view is ‘is “Necessary” vague?’. The sort of default thought is perhaps that it is not for the hard-core realist but is for the projectionist. But that’s not very compelling once you really think about it. (Tallness is real, some things really are heaps, etc.)

Idea of a system we could understand which governs the metaphysical modal space. Would this not be a kind of once and for all solution to conceptual possibility questions? (Or would you first have to find a reduction into the terms of the system?)

* * *

The 'broadly logical' terminology for metaphysical modality as a promissory note.

Note how naturally it comes to say 'It doesn't *make sense* to talk about this table having been made of ice'. 'It doesn't make sense to talk about Queen Elizabeth being born of different parents'. (*If* you're in the mood to say those things are metaphysically impossible.) 'It doesn't make sense to talk about Hesperus being distinct from Phosphorus'.

When it is put in terms of sense like this, the matter seems far less final. One thinks 'Well, maybe it could make sense in certain circumstances'. (Not so for other things. For of course we don't just mean 'these words could be given a meaning'.)

Note that these natural claims about what does and does not make sense with respect to counterfactual scenario descriptions do not parallel claims about statements about the way things actually are. It does, for instance, make sense to talk about the table actually being made of ice, or Hesperus perhaps not being Phosphorus after all.

Problem for relating modality to meaning: why, "in the abstract", would meaning be relevant? A picture: the world as a kind of machine with a space of configurations. Those are the possibilities. Our ideas or models are only a guide to the extent and nature of these insofar as they reflect the world. In any case, the order of explanation should not put meanings at the base. Our ideas or models or sentences can meaningfully assume such and such configurations *because* the world is like that.

Response: This brings out that the point of bringing in meaning isn't to block out the world. It's not that we have some end goal of cutting it out and just looking at some isolated language/thought system and explaining modal status in terms of its constitution. Rather, it's that there is a harmony there, and that we go *through* our representations to the world. Through meaning to modality. (But here I feel I'm getting too glib.) The idea that, if we don't do that, we end up stuck.

Look here at particular cases.

When there's an impossibility, one might say, it is only because our conceptions or expressions possess degrees of freedom not possessed by the world.

But something seems to have gone wrong already with that thought. Think about how poorly ideas about, say, the necessity of origin and constitution fit with this picture. (As though there were this thing, the table (or the Queen), which could be manipulated some ways but not others. But of course the table and the Queen are not 'elements' of the world, so to speak, so we feel that this picture is inappropriate. This relates to Quine's idea of the "museum myth".)

The issue of whether there is a single, main concept of metaphysical modality is important. As too is the issue of vagueness. (Kripke's results are negative - whatever may be possible, this isn't.)

Drive towards thinking there is a single concept may be bound up with a yearning to find the system - a generalised, metaphysicalised natural thinking impulse. (Necessitarianism as a balm.)

The one other bit, I think, although it may be implicit in part of the above, is this: the feeling that what our ideas or talk permits may just reflect ignorance or prejudice. Now, there is something to this, but it easily gets out of control or totalised. (This pattern-matches with the philosophical point that we couldn't be wrong all of the time in, say, arithmetic.) (Note that we can, in one fell swoop so to speak, perform radical "manipulations" with our pictures - p-zombies, e.g., or a world with just one atom perhaps, or a non-physical world. But also note the uneasiness which comes with it, the head-scratching aspect, which we don't normally have with clear examples of alternative possibilities.)

Finally, something I think is emerging only dimly now but is important, perhaps very important: the 'leaving it ragged' stuff, the 'making room for head-scratching' stuff. I think some kind of statement of the potential importance of that would be good to have.

* * *

You might ask: what does the credential matter? What follows from it? This is a hard, but I think suggestive, question.

One point: naturalness of ‘doesn’t make sense’ locutions in explanations of Kripkean necessities. Also, making sense in the relevant sense does have to do with the world: not all configurations permitted by the system of language looked at in isolation have the right kind of usability - lack a special status. And the world plays a role in that. (PI 590 passage.)

(The passage referred to above is from section 590 of Wittgenstein's Investigations and I mention it again below. It runs:
[...] So does it depend wholly on our grammar what will be called (logically) possible and what not,—i.e. what that grammar permits?”—But surely that is arbitrary!—Is it arbitrary?—It is not every sentence-like formation that we know how to do something with, not every technique has an application in our life [...].)

One tries to put meaning in the picture by saying: this proposition is one of those whose necessity is guaranteed by what it means (not that they all are - the C. Anthony Anderson point about disjunctive cases): no statement with that meaning could have been contingent. But if one has an extensionalist view of meaning - what is picked out - then this will automatically be true of all necessities. So that doesn’t distinguish the view of the philosopher who wants to have meaning in the picture, in some sense, in their philosophical understanding of modality.

If you have an internal aspect of meaning, then you can perhaps still put the point in terms of necessitation. Still say something that distinguishes from an out-in-the-worlder. Otherwise, looks like you need something like grounding or explanation.

A bare configuration - one we can sort of do something with - one we can do something with.

When you make the world part of the explanation for why a construction doesn’t have the credential - doesn’t have modal genuineness - the meaning view may seem to collapse into an ‘out there in the (nonsemantic) world’ view.

What is so important about having meaning in the picture? What is wrong with the out there in the world view? One problem is that when we try it, and try to talk explanatorily about modal facts, we shoot past the question of meaning - we try to talk in the object language about modally genuine and non-modally-genuine situations. We as it were beg the question in favour of possibility. ‘The world state where Hesperus, that very object, is distinct from Phosphorus, that very object, is an impossible one’. One wants to object that in a sense “the world state” isn’t there.

You may end up representing modal reality in a way which invites the thought that it could have been different. (Stark with modal realism.) Or, you end up with a mysterious ontology of things which you say could not have been different - this gets represented as a kind of super-structure. Now, I think the person who wants meaning in the picture is sometimes feeling here that there’s a kind of inappropriate bedrock - that there’s helpful stuff which is left unsaid if we adhere strictly to an ‘out there in the (nonsemantic) world’ view. (Here recall how it’s quite natural, when explaining the metaphysical necessity of identity, to say something like ‘Given that Hesperus is Phosphorus, it doesn’t make sense to talk about Hesperus, that very thing, not being Phosphorus, that very thing. You’d as it were be saying that this one thing is distinct from itself.’ The movement of thought here is something like: don’t look for, as it were, a counterfactual possibility that is being ruled out as impossible, rather see how there is no such thing answering to that name.)