Meanings and Rules

If we're going to say something, we need to have a way of working out what it means (well enough to play the move).

Any possible algorithms for computing the meanings of self-referential paradoxical statements don't terminate - there is no way of working out what they mean - so we can never play them as moves in the game. This is guaranteed by the way they are constructed - and they can only arise in this way. "This statement is false" can never intelligibly be asserted; and as it is the possibility of its intelligible assertion which would generate chaos, we are safe from it.

Tractable algorithms are a subset of the algorithms which can be finitely described. Some algorithms which can be finitely described are intractable. They loop; or do not produce a usable expression in a finite period of time (e.g. if required to calculate the exact digital expansion of a transcendental number); or they do not produce a usable expression in a relevant period of time (e.g. soon enough for us to use it).

When we describe an algorithm in words, of course, we must have a tractable method of working out what the words mean in time to be able to use them in our description of the algorithm. This method, in its turn ...

So, of course, we have another paradox. Or another perspective on a familiar paradox.

Can we say that if we follow a method for working out what to say, then it is a method that we cannot describe? In one sense, this is obviously true: it's a practical fact that we can't explain how we speak. A computer which mimics speech cannot also necessarily produce a verbal explanation of how it does this. But is it necessarily the case that it is true?

If the computer that mimicked speech was allowed to read and decompile the code that it was running, it could state the algorithms which it was following in the language in which they were written. It might also store in its memory a plan of its own hardware, a description of the compiler etc.

This might not count as an explanation, however. It would be something similar to what a neurologist interested in speech processing would do - attempting to produce an account of the hardware and software underlying the production of speech as a phenomenon.

Would the computer (or the neurologist) be able to say why what was being said was correct? Could it produce an argument for the reliability of its statements? Presumably, since it could speak (and since this is part of having the capacity to speak), it could produce arguments for the truth of individual statements that it made. But could it produce an argument for it, itself, being a reliable implementation of a speech algorithm? Could it produce an argument that it could, in general, speak properly - that it could, in general, tell the truth?

It can't do this by examining and reproducing its own code - this would only be an account of what it did, not of why it was correct.

If we read a short piece of computer code, we might see that is was correct - the code, for instance, for doing a binary search of a sorted list is almost self-explanatory when written out in a high level language. Even here, though, there is a lot of debate about whether this 'self-explanatoriness' amounts to a proof; and the debate (I think?) virtually comes down to how we select our formalism for calculating proofs. The computer code is also a formalism of a kind, and if it is well-defined, it should produce its own proofs. Programmes written in an formally correct implementation of Lisp, for instance, will look like expressions in the lambda calculus - and if the interpreter/compiler correctly evaluates these (another question, of course ...) the results will be provably correct in that calculus.

But all we have done, even here, is to show that an expression is a theorem of a formal axiomatic system which can be encoded in a computer programme which should (barring hardware faults) correctly implement the rules of the formalism. We can show that the speaking computer is designed to correctly follow the rules that the programmer has coded into it - we cannot show that these are the correct rules unless we allow that to speak correctly is simply to utter the theorems of some FAS or another ...

And since this is something some people might once have believed, its worth saying why it can't be true: for Kripkean reasons, the 'rules' of the system can only be made intelligible within a language which already works. The idea of a 'non-linguistic' rule is incoherent. We might think of 'laws of science' here, but these are part of our descriptive grammar - a world of which it was impossible to speak would have no laws. Our reason for thinking that there are rules 'in the world' is that it is only possible to describe a world of that kind, and we know (a priori) that we can desribe the world in some way. (When we say we cannot we are saying something about the world; and if we try to say this seriously we find we can't even say that we are saying something.)

We can predict, with high reliability, which rules the computer will follow - but we cannot define the rule as 'what the computer does', however much we might, in normal circumstances, practically rely on the computer to do the right thing. Such a definition would make a computer error formally unintelligible.

If we cannot give an account of rule following and formalism without, first of all, being able to give an account of something, then our capacity to give an account cannot be accounted for in terms of our following some specific set of rules. If it could, then we would have to say that it was not possible for the underlying substrate - the hardware, if you like - to make a mistake here. We would be like perfect computers. And we couldn't even test this perfection, because there would be no other standard: There would be no intelligible enquiry that we could undertake, because the very possibility of intelligible enquiry would depend on the reliabiltiy of the hardware.

It is perfectly possible - in fact psychologists have demonstrated that this is true - that the underlying hardware is systematically faulty. People are prone to making certain kinds of cognitive errors: they make incorrect inferences, the are not good at estimating risk etc. What is the standard of correctness that the psychologists are using here? Why do we not think that this standard, as well, could just be the product of some faulty hardware?

We think it because there is an argument for the reliability of this standard - an argument which, if fully developed, would show that the standard depended on the possibility of their being any standards; of our being able to say anything true at all; of our being able to speak. Even if we never fully develop this argument, we promise it when we make a truth claim - and if we find the argument cannot be made, we withdraw the truth claim (on pain of becoming unintelligible otherwise).

It is our ability to speak that is the standard against which the rules are judged, not the rules which guarantee the reliability of what we say. We have been confused by the fact that good rules generate other good rules; so that we think this must be where all rules come from.

And also that allowing the assertion of contradiction or paradox allows the assertion of anything at all, and so of nothing. Our rules must be consistent; and we must only say things which mean something. We might also say that the rules implied by an intelligible discourse are consistent, and that it isn't possible to (properly) say something which is meaningless. If we discover that a previous statement commits us to contradiction or paradox, we reinterpret, revise, or withdraw it.

So could there be 'hidden' programming? If we can make any sense of the idea, it is only as defining a system that we can explore 'from the inside', and never hope to give a full account of.

And it would have to include everything in the world - not just us. It is the whole programme by which the world runs, and there is no machine that it runs on because that machine would, itself, have to be part of the world. It is not even a thing that we can describe, because the description would have to include our description. And a description would have to explain its own veracity - it would have to explain our capacity for truth-telling, and so would generate open questions about its own truth. Or it would have to be in a language in which an account of our truth-telling could be given; and which would (therefore) be unintelligible to us in that respect. All of these are pointers to the void - there is nothing sensible we can say here.

This is why we must both explore and calculate; and explore how to calculate, and calculate how to explore ...

A deranged speculation ...

What would it be like to be unable to talk? Can we imagine this?

I don't mean here an acquired dysphasia, or a specific neurological damage, but a profound inability - to have never learned, but to be otherwise human.

Given the plasticity of the brain, we would be neurologically different in significant ways. For the sake of the experiment, I'll assume that the language deficiency is the result of an otherwise benign event (hard as this is to imagine) so that there are not other residual neurological effects of trauma.

There have been plenty of cases of children brought up in circumstances where they failed to acquire language, usually for otherwise traumatic reasons; so this thought experiment has some concrete correlatives. They may not serve to settle the speculation I'm going to suggest, though - and perhaps there is no available evidence that could.

The speculation is that the phenomenology of such a person may be almost unrecognisable to a language user; and that certain 'obvious', and even physical categories and preconceptions may be missing.

For instance: without the self-reflection and concept construction capacities that language brings with it, could a person with this deficiency be said to have concepts of time and mortality? They might well show (through their behaviour) the capacity to predict, and to feel fear. Animals do this as well. But what would their fear be of? And could their capacity to predict encompass a picture of their whole life as a sequence of events that form part of a larger partly causal or rational network? Would they even have a recognisable perception of the passage of time?

Our earliest memories are often of events that happen about the time we learn to speak. Is this because we don't have the capacity to organise or reflect on what happens to us in a self-conscious way until we can talk about it?

And if these aspects - time, space, causality etc. - of our phenomenological frameworks depend upon language acquisition, and are absent or seriously impaired in a person who never acquires language, what should we make of questions about their material underpinnings, or their 'reality'?

If my cat doesn't think it's going to die, and doesn't think anything when it does, does it feel as though it lives forever?

[Eipcurus (?) seems to have thought so: "If death is, I am not; If I am, death is not."]

Is 'the world' driven by a 'hidden' axiomatic system?

It is possible to imagine that the 'real' world is a system which only looks unpredictable to us because we don't know enough about it. Quantum theory seems to contradict this, as it assumes that certain kinds of event are fundamentally unpredictable (e.g. the time at which an unstable nucleus will emit an alpha particle).

There are other issues, though:

Would a 'complete' system also have to predict its own predictions? This is 'algorithmic information theory' territory. The system, and especially its mechanical substrate (if these can be formally distinguished), is part of the world it is trying to predict.

Would it also have to include us? Would our language (and it's mechanical substrate ...) be part of what it could explain? If it did, would it have to include an isomorph for a theory of truth for our language? (Since it would comprise, in some sense, an ultimate 'meta-language').

This has a funny consequence, which is that we could not translate the language of this theory into our own language - because we cannot state a theory of truth for our language in our language. This means that the language of the 'ultimate theory' would be incomprehensible to us, and so, therefore, would the theory.

And if we take a clear Davidson/Quine line on this, we not only can't translate this language, we have no grounds for treating it as a language.

" ... is true"

If we introduce a new axiom into a system, it may make it inconsistent. It might also do a number of other things:

(1) It might (if it contains a lot of information) make it very complex.
(2) It might make it difficult to interpret (to give a semantic interpretation of).
(3) It might leave the issue of its consistency uncertain - either because
(a) It makes its consistency hard to calculate (complexity issue?) or
(b) It makes the system into one in which Gödel/Turing type concepts can arise. (Is this correct?)

So when I say 'X is true', and think about this as introducing a new rule or axiom, then we might accept or reject this proposal on a number of different grounds. We (my interlocutors and I) need to decide what to do with the proposal ...

What account would I give of a rejection on empirical grounds?

I might compare this with complexity: the new rule makes it very hard to play the game. People may 'explain' this by saying: "But not X!" ("But it just isn't raining!"). This doesn't help to clear the confusion. Perhaps if they said "I can't think what you might mean by insisting that X is true ..." - I don't know what you are committing yourself to; how X fits into the game. I think this would be a better answer, from someone still trying (seriously) to play. Flat contradiction "But not X!" is an end point, not a move in the game (just as, in another context "But necessary X!" would be). It's like saying "I'm not playing your game!" - I'm no longer an interlocutor on that basis.

When we experiment with an axiomatic system, we are interested in whether certain statements are theorems of the system; whether the system makes sense (is consistent?); and whether the system has an interpretation (an 'application').

Chaitin is right that a lesson of the halting problem is that mathematics requires experiments. But what kinds of experiments? Well - experiments with ways of talking. (A defining characteristic of 'real' or 'natural' languages is that they are both the fora for making these experiments and a product of their outcomes).

Serious interlocutors take responsibility for the intelligibility of their truth claims - they do not hand this over to a 'neutral' argument. If we try to make 'X is true' look something like 'X is a theorem of S', we simply move the responsibility from the 'X is true' claim to the 'the axioms of S are true' claim.

Maybe a good way of saying it is to say that when I claim that 'X is true' I also promise to make playing the X move intelligible. I'm prepared to show how it fits in, what adjustments it requires etc.; and to stand by those demonstrations and adjustments.

Within a shared game, of course, some things 'come out' as true, in the way that some statements turn out to be theorems of an axiomatic system. This is inevitable, because the shared game depends on some shared rules, and these rules have consequences. (The shared game must avoid contradictions, but only because a game which admits contradiction has no rules.)

Creating and modifying rules - introducing axioms via the 'X is true' formula - is also part of the game, although certain 'fundamental' rules (among which are the rules which render the introduction of new rules intelligible) cannot be broken. We can search for these fundamental rules (this is doing philosophy) but we cannot expect them to generate the whole system.

A new version

Version 13

The version numbers are fairly meaningless - they just represent the order I put them up in, there is no commensurate scalar improvement in the content.

I haven't left 'marked changes' between this version and no. 12, because the changes were so extensive the markups were meaningless.

I may revert for future versions, but it depends on what they look like.

Metaphysical comforts

Perhaps there is a very old confusion about what can be done with argument: we expect reassurance as well as demonstration. But the relevant 'sort' of phenomenological reassurance is not a semantic category, even though we might think we know what 'phenomenolgoical reassurance' means (beetles in boxes ...).

What would an argument for realism (or God, or Unicorns) achieve? It would show that our language game would only be intelligible if there were real things (or Gods, or Unicorns) out there; or that the real things had to be organised in a certain way; or that they would have to have certain properties. But the phenomenological reassurance we need cannot be supplied by such an argument because the argument cannot address our anxieties about the possibility of intelligiblity itself - about the possibility of producing arguments.

This is a 'nonsensical' anxiety, but only in the sense that as as soon as it is articulated it is shown to be unintelligible - because articulating it is unintelligible. But how do we privately reassure ourselves about this?

Wittgenstein is correct - the need for such a reassurance is pathological, it is a kind of mental illness.

But some of us are mentally ill - we have drifted beyond the scope of interlocutorship, to coin an ugly word. It is also clear that some 'illnesses' of this kind can be ameliorated, or abated, by 'talking cures' - counselling, psychotherapy - which clearly lead the subject back into the realm of the sane, back within the scope of our comforting shared conversation. In extreme cases, these therapies must work by showing, in the Wittgensteinian sense, since the capacity for 'reason' (the capacity to participate in the conversation) has lapsed in some way.

Whether there must always be an island of intelligibility at the centre of a therapeutic exploration, the possibility of this not being the case has to be entertained. A child learns language without first of all knowing how to speak - there is no prima facie reason why a seriously deranged person might not be in this situation, though their behaviours and interactions would not be child-like in other ways.

Do we think of learning/re-learning language as being like acquiring/re-acquiring (?) a comforting pattern of (complex) behaviour? Does playing this game soothe us?

Perhaps, because of the kind of animals that we are, it does. We can tell all sorts of stories about why it might - why, even, it makes sense that it might - but these are only going to be intelligible (in any sense) to those who can already play; the comfort comes before the intelligibility.

If we have a story that says 'And now, of course, I can explain why I should be comforted by this', we are missing the point. We have already forgotten too much to be able to understand the initial problem - we have not addressed the terror, we have just learned a way to dance around the campfire that allows us to ignore it. An argument for a fundamental ontological category, or condition, can only be a story like this.

We might think we can comfort ourselves by demonstrating the existence of God, or substance, or their co-extension, but none of these can demonstrate the reliability of the game of intelligibility within which our demonstrations work - only an unreflective, unselfconscious, probably instinctive, language user would even think that they might.

Note the odd frustration one feels attempting to persuade someone who is insane or otherwise irrational: their imperviousness to argument can seem threatening. They are doing the impossible, we think. But only because walking out of the light of the campfire is as good as impossible to us. There is no 'out-there' beyond the light of what can be discussed - we prove this to ourselves, and remark knowingly on the unsayability of the unsayable.

But once the game stops being comforting - when the models of intelligibility on offer are more terrifying than the darkness, or the familial campfire represents cognitive dissolution - then madness becomes intelligible practice, however unintelligible its utterances.

When a philosopher needs comfort, he or she should apply to a therapist, not to an argument. The argument will only show that it can work as an argument - it won't show you that you have; or will always be able; or can require someone else; to play the game.

And this problem is more than theoretically interesting, because it inspires errors. The right answer is not necessarly metaphysically comforting - unless one makes a metaphysical fundamental of argument, and this would be as much a mistake as anything else. An ontological 'rationalist' is as wrong as an ontological materialist, in this context.

If our game is only intelligible on the presumption of God's existence, or the grammatical structure of physical reality, then we may have a demonstration of these metaphysical fundamentals but at the cost of rendering them irrelevant to the psychological problem.

If you want the comfort of God's existence, an existential argument will only convince you if you already get at least as much comfort from the reliability of argument. Even to be convinced by miracles, one would need a language of miracles within which arguments from miracles could be constructed. Without this language, how would we know a miracle from anything else?

Would a cat become devout if it saw an angel? If we believe it would, it is only because we can tell an intelligible story about why it would be so, or because such a picture pre-figures any intelligible exchange we might imagine having.

We can either have comforting metaphyscial presuppositions or we can demonstrate them from the possibility of intelligible conversation. We can't do both. An argument for ontological realism cannot underpin the reliability of empirical argument because it is already an empirical argument. A scientist whose ontological realism is an indispensible heuristic can only articulate that realism 'metaphysically' (or 'nonsensically', in the Wittgensteinian sense - i.e. as a hinge, for which it is incoherent to require an argument).

When normal people talk this way - when they do things like articulating hinges - they often believe that they are talking 'philosophically'. 'Real' philosophers indulgently point out to them that philosophy is about constructing arguments, while, of course, surreptitiously seeking comfort where it cannot be found ...

Doing things with words

There is a sociological reading of Wittgenstein which makes him a kind of linguistic social constructivist; and the 'ordinary language' school of philosophy focussed, in a somewhat similar way, on actual language usage.

I think that he was more interested in what we can say than in what we do say. His examples aren't, perversely, meant to be exemplary - they are meant to be exploratory.

Not 'This makes sense' but 'Does this make sense?'.

He was showing us things, and asking whether we could see how the game was played - and how it might be developed.

Hinges and Fundamentals

The truth of a hinge is a logical consequence of the playability of the game, because denial of the hinge undermines the game. It is only in a practical (a matter of practice) sense that the game depends upon the hinge. The ground for the hinge is the playability of the game, and not vice versa.

The argument is a modus tollens, not a modus ponens: If the hinge was false, then we could not speak (this way) - we can speak (this way), therefore the hinge is true.

In OC Wittgenstein cannot explain the reliability of the hinges:

470:

"Why is there no doubt that I am called L.W.? It does not seem at all like something that one could establish at once beyond doubt. One would not think that it is one of the indubitable truths.

[Here there is still a big gap in my thinking. And I doubt whether it will be filled now.]"

Then in 474:

"This game proves its worth. That may be the cause of its being played, but it is not the ground."

But to give a ground is to make a move in the game. To ask whether the game has a ground is to ask, if it is to ask anything, whether the game might not be playable if there were no ground. However, the game must be playable because we cannot say that it is not. The absence of a ground would demonstrate the absence of need for a ground, if it demonstrated anything - if the game can be playable without a ground, then grounds are not required for this kind of conclusion.

And since any prospective ground could only be articulated within the game, the ground could not be more reliable than the general possibility of giving grounds. We cannot produce an argument for the reliability of argument.

So: a ground is not necessary, and no ground of the right kind can be given. We have to start from the playability of the game.

Something that can't be explained

There is (literally) no explanation for the 'fact' that I see the world from over here (from my 'perspective') and you see it from over there ...

Why should I have special access to some of the world's events and not to others? Why am I aware of any of them? Why not all of them?

We can sort of imagine (or maybe only think we can imagine) an impersonal physical universe - not LaPlacian, perhaps, but some more up to date equivalent. We might even say things like 'you are a part of this universe - a sub-system - which interacts with the rest in a specific way, and therefore is only "aware" of some events and not of others'. But this doesn't solve the problem of why I am this particular sub-system and not some other. Why am I not you?

Externally, of course, lots of kinds of explanations can be suggested. Psychologists and neurologists looking for a 'seat of consciousness' or for some systemic explanation of consciousness focus on behaviours and cognitive structures characteristic of those of us who think of our selves as conscious and of those others we attribute consciousness to.

This doesn't help very much, though, because these behaviours are only accessible to theory once they have been described, and they are - in some sense - associated with all conscious beings. What I'm interested in is my consciousness, not everyone else's. I want to know why I'm aware of the world I'm aware of. And much of this world is not only unverbalised, its unverbalisable.

So far as my participation in this conversation is concerned, of course, I could be anyone - or at least anyone like me. I wouldn't actually have to be me. If I was replaced by a cunning automaton, who would know except for myself? I might, like a Stepford wife, be ousted by a murderous doppelganger - and it might be ousted in its turn - for all this conversation was concerned.

But it would, surely, concern me. I wouldn't be here any more.

Does this make any sense? Well, not in this conversation - it is a 'hinge' of this conversation that I am the same person who wrote the previous posts in this Blogg.

Is this being a 'hinge' a solution to the problem?

In one sense yes, because it points to the public conception of a continuing person with the ability to converse. We can even understand (in this public world) why some people know some things that other people don't.

But in another sense, it just evades the question. Why am I sitting typing at this keyboard, looking at this screen. Not a public puzzle, but a question about my particular phenomenology.

Why was my world constructed with this subject? There is nothing in our shared world that can explain this to me.

Is this the same as the Zombies problem?

No. Whether or not we treat each other like zombies is a normative issue. I don't think mind theorists take the normative aspects of mind attribution seriously enough, but that's another problem. I can't treat myself as a zombie, so this question is not the same as the zombies problem.

In a sense, posing this question is a poetic move in this conversation - I am suggesting something, alluding to something, that cannot really be said; and I am doing this by appearing to say it.

(I can put the idea that it may not be possible to talk into your head by saying 'It is not possible to talk' even if this is strictly nonsensical).

Whether or not you can engage with this may have something to do with how self-conscious you are of your linguistic abilities - how much you are aware of talking as an activity depending on an inarticulable substrate of skills ...

Propositions and Rules?

Some Wittgensteinian scholars might like to distinguish between propositions and rules - or 'hinges' - but sometimes a statement functions like a propostion, and sometimes like a rule.

It's hard to say this, though because whether or not we are dealing with a proposition or a rule depends on what game we are playing - or, rather, on which part of the 'overall' game we are playing. And the game we're playing determines what we can say and what we can't say ...

When we move from one game to another, we indicate that we have done so by using explicit truth-predicates, which, I think, are always used in a rule-like way. When I say 'My cat is hungry' I mean that my cat is hungry. When I say 'It is true that my cat is hungry' I am telling you how I am going to talk, and inviting you to talk the same way. You might respond by pointing out that this rule won't work for some reason, or you might find you don't know how to play the game I seem to be proposing, but while the first statement can legitimately be described as being about the world (with it's hungry cat), the second cannot.

Thinking that we are talking about the world when we use truth predicates, rather than about how we are going to talk, is one of the most fundamental errors that philosophers have made.

Ramsey's Ladder, again

Maybe I've said this before:

We want to preserve the first rung, but not the rest. We want to leave S meaningful, or rather usefully meaningful (!) where S="X is true"; but we want to leave "S is true" vacuous.

Rules, of course, work like this. If S="It is a rule that X", this means something. But to say "It is also a rule that S" is vacuous. We don't have a 'master' rule that says we must follow the rules. To propose something as a rule is just to propose that it should be complied with (with contextual qualifications, and under agreement).

If statements of truth values are, or are importantly like, statements of rules, then statements attributing truth or falsehood to statements of truth values will always be vacuous.

The traditional liar paradox is a statment about the truth value of itself: "This statement is false" contains a statement about a truth value. Taken as a rule, it is simply incomprehensible: there is no answer to the questions 'What way does this rule require us to talk?'.

And it only appears to be consequential if we allow the rule that we should follow the rules ...

Is it true that we should tell the truth (generally, with appropriate caveats etc.)? It's vacuously true, in a playable game. Its falsehood can't be a move in a playable game.

And:

"We should always talk as though this statement is false" is incomprehensible at the first level - it simply cannot mean what, at first sight, it pretends to mean. It tells us to do something that is logically impossible. (What it appears to mean is something that would falsify an interpretive hypothesis, in the Davidsonian sense).

Cartesian Anxiety

Our central phenomenological terror is not that our perceptions might deceive us, but that our friends might:

The inexpressible hyperbolic doubt that we are always talking at cross purposes.

Pace Dante, the gates of hell bear no intelligible warning.

Interpreting behaviour

We often re-interpret or misinterpret behaviour. Kripke shows that there can always be more than one intentional description of the same behaviour.

It is natural to think, here, of there being (a) the behaviour and (b) a set of interpretations. But we shouldnt make the mistake of thinking that there must, therefore, be a way of describing the 'uninterpreted' behaviour. The only descriptions we may have might be interpretations, and the 'sameness of behaviour' might only consist in our agreement that the interpretations are interpretations of the 'same' behaviour.

We may demonstrate this 'sameness' by identifying key aspects of the situation ('navigation markers') - such as time, place, actors etc. - but these are not descriptions of behaviour, and might not include anything which told us what was actually happening.

We might imagine, also, that some mechanical descriptions of limb movements and interactions with objects might somehow 'capture' the raw elements without introducing an interpretation. Computer simulations are based on models which encode this kind of information, but they are extremely difficult to interpret as 'descriptions' - they contain a great deal more information than our normal descriptions, and it is organised in a way which is hard for a human being to decode.

While the possibility of this modelling language might have been predicted from our judgements of sameness (and the existence of the 'navigation markers') , it certainly hasn't played any important role in producing judgements or demonstrations of sameness.

Also: it isn't clear that any such language really could capture all of the things which contribute to our actual arguments for sameness (where these are produced). The navigation markers relevant to a particular demonstration may be idiosyncratic, and missing from the the encoded description. We may not refer to time and place or personae but infer these from some other data which are not generally included in the descriptions encoded in the modelling language.

So what we have is just interpretations and the judgement of sameness. And we can make this judgment reliably - it would be incoherent to suggest that we don't. We also produce intelligible arguments for sameness which do not depend on 'full descriptions', or anything like these.

Speaking Correctly

We might talk about 'speaking correctly' in terms of uttering / recording the correct sounds or symbol strings etc. When we try to say what this is we end up being very clumsy - the idea, which seems so clear to start with, evaporates as we try to articulate it. We can give examples (such as the string of characters that comprises this post), but what could we exclude?

Instinctively, we say - we could exclude meaningless utterances/recordings/artefacts (... etc.).

So does 'speaking correctly' require only that we follow the 'rules of meaningfulness'? (And if, for isntance, we allow false statements only as the contraries of true ones, and let Duns Scotus deal with the issue of trying to make them work as serious assertions ...).

This is no better, because we need to know the meanings of the rules in order to follow them correctly and also because we cannot know (Kripke) what rule someone is following simply from their behaviour: and we we would need to know this in order to know that we knew what they were saying. And we must, broadly, know what they are saying if we are having a conversation with them. (E.g. discussing meaning).

And there is another problem: we might say that two 'expressions' mean the same thing if they are correct translations of one another. This could be a practical matter - e.g. we use them interchangably in certain contexts; or a theoretical one - we say that they mean the same thing. This second option is another move in the game; and the first is identified by a move in the game - a description of the relevant practices.

In order to speak correctly we follow certain syntactical rules, but this is neither (ultimately) sufficient nor necessary. We can create new rules and we can follow the rules meaninglessly.

Is there a general limit to this?

Not one that can be stated, because if we try we will generate an open question paradox.

Are there particular limits (boulders in the torrent, to which we can cling or against which we would break)?

There are the rules of logic - whose contraries immediately generate meaninglessness by generating contradictions - but whose scope of application can only be defined after we have agreed on certain interpretations, after we have agreed on what would count as conforming to or breaking the rules. ~(P&~P) is only 'obvious' if both instances of 'P' mean the same thing (and there are deeper, more inarticulable issues of interpretation as well).

We may, in other words, reduce syntax to semantics; we can never reduce semantics to syntax. We can only state syntactical rules, and we can only make unequivocal judgements about whether or not they are being followed, in the context of a meaningful conversation.

What kind of judgement is it that we are engaged in a meaningful conversaion?

It's a normative judgement on the part of the individual participants, but it is not expressed within the conversation (where it would be vacuously true, where its contradictory would be nonsense) - it is instead expressed by their participation. A doubt about it would bear on this, and a serious doubt would lead to withdrawal (not to some absurdity such as 'you are all talking nonsense', except as a rhetorical exaggeration ...).

This withdrawal might express itself in a number of ways, including an apparent, but dishonest, continuation: 'participation' with purely instrumental goals, for instance.

This is the real Cartesian anxiety - not that we are beign deceived by our senses, but that we are being deceived by one another. The first might worry philosophers, but the second could drive anyone mad - and there is no philosophical speculation which could entertain it.

Transcendental arguments ...

"Some argument is valid" (statement "V") must be a reliable move in any valid argument.

There are no reliable moves in an invalid argument (Duns Scotus).

Is it an axiom? Or is it a statement of the possibility of there being some axiomata?

What if we found that some some axiom A was (a) coextensive with V with respect to arguments or (b) was a necessary condition of 'some argument is valid' or (c) was a consequence of 'some argument is valid' but not of some other (independent) statement?

It's hard to imagine (a) being the case without this being a consequence of either (b) or (c), except that logical rules (symbol manipulation rules) look like possible candidates. Without them, it's hard to see how (b) or (c) could be demonstrated and there is some sense in which the scope of these rules is the same as the scope of valid argument (and therefore of V).

A logician might argue that V contains an undefined notion of validity, and so is 'meaningless'.

This is only the case if the definition is absent - not if it is simply incomplete. In a natural language, the concept of validity is both funtional and incomplete. Assertions of validity or invalidity may be stipulative ("This is the way we should talk."), or experimental ("Let's try talking this way.") But these stipulations and experiments may work or not work; and with them the language itself (if the breakdown is widespread) and so any possible concept of 'validty' at all.

Logic works with rules, but the language in which the logical rules are stated cannot be analysed as rule based.

Truth and honest practice

Because we have to recognise that commercial language use has legitimate operational goals, we cannot (as we might in the sciences) render 'honest practice' as 'truth-telling'.

The bomb disposal example (see "Meaning and 'transmission'") illustrates this. Honest practice here clearly excludes truth telling.

Things and Words

Because we can write down syntactical rules for parts of our language, and because we use these to demonstrate certain things, we forget that a demonstration of the validity of our arguments in general cannot be based on rules. Similarly with establishing semantic consequences - we can write down a few of these in a rule based way, but we can't completely eliminate semantic imports into the rules that we use. We can't give a general account of how language comes to have semantic content.

When I say that I can say 'The cat is hungry' - that this is a legitimate move in the language game I am playing (i.e. that it is 'true' for that game) - because of the way the world is, I am not saying that I can say 'The cat is hungry' because the cat is hungry. This is a kind of category mistake - or a mistake about the meta-level we are dealing with.

Of course it is true that if I say (seriously) 'the cat is hungry' I must also believe that the cat is hungry (or I'm not playing the game properly). For the purposes of that game, I must believe that this is the way the world is. But that is only because I am using this language to describe the world, and not some other language - e.g. the language I would use to explain how this language (the one in which the cat is hungry is a fact) works.

If I try to explain how this language works, I end up saying silly things like "'The cat is hungry' is true because the cat is hungry" (or their equivalent). Or, I say: "'The cat is hungry' is true in this language game because the world allows this language game to work". And the second statement doesn't say anything specific about the world except that we can talk about it the way we do, in fact, talk about it ...

Maybe this is 'the fact' of the slingshot - I would need to look at this again. It's where we go when we try to make general statements about what must, ultimately, be true in order for anything to be true. We end up with just this one feature of the world: that something can be true about it.

Two Dogmas

If there is something about the world that makes 'Vixens are foxes' true, it is not that Vixens are foxes, but that the world allows us to speak that way.

In fact, the only 'fundamental' thing we can say about the world is that it allows us to speak the way we do: and we can know this a priori, because we can know (in the context of speaking to one another) 'we can speak about the world' a priori.

If we try to say anything else, if we propose some other metaphysical fundamental, we are introducing a new constraint. We can do this with respect to more constrained games (for which these constraints might be hinge propositions), but in the most general cases we are caught by the 'as though' problem: that no philosophical importance can be attached to distinguishing between, for example, 'There are real physical objects in the world' and 'we can talk as though there are real physical objects in the world'. 'We can talk' and 'We can talk as though we can talk' is the only general circumstance in which the metaphysical implication is direct - the second statement is just a repetition of the first.

Syntax and Semantics

For many 'information' channels (e.g. the computer I'm using to type this, or a gas whose temperature is interesting) there is a huge disparity between the number of possible messages and the number of semantically useful messages that can be passed. There are many billions of states that my computer could be in that are entirely useless, and many many more billions of states that a cubic centimetre of air at 12 degrees Celsius can be in on a cool day in Aberdeen. Chaos and redundancy are intuitive barriers for many people coming to information theory for the first time: we instinctively think that 'information' means useful information - information that means something.

But we only get meaning through interpretation - the different states of the cubic centimetre of air are only interesting if we can distinguish them - by their consequences, or by decoding them in different ways.

This may be difficult with the gas example. We might think the computer example is easier - we can list the states of all the independent storage elements, and show where the differences lie.

But this doesn't meant that we can 'reverse engineer' all of these states and represent them as the intelligible output of a programme - even one with errors in in it.

Some states cannot arise as a result of programming errors (at any level) and can only result from hardware faults, but a single hardware fault might result in a large number of possible memory states, so that we only need to recognise them as members of a class in order to find the fault - we don't need to be able to identify and interpret each one separately.

The overall plan ....

Here is a rough synopsis:

(1) For broadly Davidsonian reasons, as well as others, we can roughly identify our 'conceptual scheme' with the conversation we are presently engaged in, and since we identify other things as conversations only if we can translate them, we end up with, in a rather ragged way, only one conceptual scheme. If we think it might have inconsistencies, we can try to find them & eliminate them. We have no intelligible way of speculating about what kind of language game this process would result in, because we would need to be able to translate it into the language we use now in order to indulge this speculation; and this is something we cannot, by definition, do.

(2) Although synthetic a priori beliefs may be elusive, synthetic a priori statements are easy to construct, as we can make general statements about the possibility of language: 'We can speak' is either true or not a statement. We need to be able to make assertions, because to deny this is to make an assertion. If we can make assertions, then 'We can speak' has the corrollary 'We can tell the truth'.

(3) Generic theories of truth (of how to speak) are blocked by the open question argument. Statements that it isn't possible to know we're telling the truth are self-refuting. So we must be able to know we're telling the truth without being able to say why, in order for language to work: and since we can assert the latter a priori, the former must be the case.

(4) The structure of fundamental arguments now becomes recursive: we can never argue 'X must be true because it is a theorem of my theory of truth', and we certainly can't argue 'X must be true because it isn't possible to tell the truth with certainty'. The only alternative is the transcendental argument: 'X must be true or it would not be possible to tell the truth'.

(5) Since 'We can speak' is an empirical statement, it is possible to construct fundamental empirical arguments transcendentally: Either Empirical(X) is true or it is not possible to make empirical statements. This is the bridge from reason to the world.

(6) While it may be possible to construct a transcendental argument (as in 5) for the reliability of the senses, this would not rescue traditional sensory empiricist epistemology - it would show that the reliability of the senses depended upon the reliability of empirical statements, and not vice versa.

Some stages obviously need elucidation (!) and there are lots of interesting consequences and further developments...

But the main elements are all here.

Another take on the Open Question Argument?

Suppose a group of interlocutors agreed on an articulated theory of truth for the language they were using: they would also, then, have to be considered to share an interpretation of the theory. And this is something they cannot articulate within the language - they can only point to the continuation of the conversation.

Another way of putting this: Sharing an articulated theory of truth is also sharing an interpretation of such a theory, and our only grounds for saying that we share an interpretation are the same grounds we have for saying we share a tacit theory of truth - that we can can continue talking to one another. And within the language game we are using to converse, we know this a priori. If the interpretations diverge, the game breaks down - and we don't even have anything with which to say that it has broken down, far less discuss our varying interpretations.

(We do, obviously, discuss varying interpretations of some linguistic moves - but only within a shared framework of truth-telling.)

Phenomena

Phenoma are also ontological constructions.

We think we 'really' share them because we can talk as though we do.

They seem fundamental because we seem to 'refer' to them introspectively. But there is no reason why any two individuals' internal states should be 'commensurable' beyond what is required for mutual intelligibility.

This is why there can be no 'science' - no articulated theory - of consciousness. We do not share the aspects which we would need to articulate in order to make such a science possible.

Reminiscence ....

This now seems more important to me than when I first wrote it down:

"Kripke’s paradox cannot have the consequence that we cannot be sure that someone who states it is following the rules that make its statement intelligible."

It's really the solution to the paradox, and also an explanation of how we should think about language and certainty.

Introspection

If you can participate in this conversation, then you are a language user - you will feel like a language user (if you have any relevant feelings about it at all).

If you think 'what do I feel about what it would be appropriate to say, here?', then you are likely to come up with the right answer, since knowing the right thing to say is just what competent language users can do.

But the reliability of these introspections is misleading: it can make us think that these feelings and internal processes are at the root of the reliability of the move we feel is appropriate, or even of the language game itself.

(In order to show that a particular move in the game is reliable, or correct, we must produce an argument, not a feeling. It is incoherent to ask for an argument for the reliability of the whole game: an argument for the reliability of argument would not demonstrate anything.)

And this observation extends to our empirical intuitions as well as our 'language instinct'. It is not our sensory perceptions which make us reliable theorists - it is our ability to agree with each other about them. Our theories are linguistic artifacts.

Our ability to act in a rational way - an ability any moderately sentient animal exhibits to a greater or lesser extent - is not an ability to theorise, except under interpretation. If I describe a cat as having a theory, I am describing its behaviour in a certain way - as having a certain normative content. There will always be an alternative 'cognitive', or computational (homeostatic?) interpretation which elminates this content. This isn't possible in the case of a person who articulates a theory, because we can only produce a computational account of this behaviour by reducing the linguistic performance to something mechanical. Articulating a theory is participating in a conversation - we don't automatically do this by moving our jaws in certain ways and producing certain sounds.

I might, in a very complex conversation with you, give you an account of how your physiology and neurology instantiated certain cognitive/computational process in such a way as to produce the output we recognise as 'speech'. But when you said 'Ah yes, I see', I would hardly respond 'And there's another example of it working'.

Belief, Interpretation, and Criticism

In order to believe what you say, I have to understand it. I have to interpret it.

This means that I can't just 'accept' it. There is no belief without interpretation.

And there is no interepretation without critical interpretation. If I wonder 'what sense does this make?' then I am wondering 'how can this make sense?' This is the same as trying to figure out how I can continue the conversation with you.

Just because the outcomes of these reflections can seem obvious, and the reflections themselves 'subconscious', they are not therefore 'given' in any useful sense. They are the outcomes of cognitive processes which, if we are not able to articulate them, are still our 'practice'. They are things that we do.

In this sense, I cannot believe you without critically interpreting what you say: without, in the literary sense criticising you.

Whatever private reservations we might have, and however we might (again privately) engage only provisionally in a conversation, there are only some interpretations which are consistent with it being a conversation. When we articulate a criticism, therefore, it must be consistent with the possibility of the conversation within which it is articulated.

This means that there are some criticisms which are wrong (inconsistent with the the possibility of their supporting conversation), and some which are not-wrong. A disjunctive list of the not-wrong criticisms must be a correct criticism - however qualified and inconclusive.

The consequences of this may be trivial, however. Northanger Abbey is neither a Western nor a vaccuum cleaner repair manual. And the 'disjunctive list' may be unmanageably long.

Version 12

Showing incoherence ...

The statement 'It is not possible to talk' can't mean exactly what it says and still be true. However, the following seem reasonable (at least superficially):

(1) We know what it means, because otherwise we couldn't decide that it was false.

(2) There is some sense in which we can think that it's impossible to talk:

- there is behaviour that would be consistent with this (although it would also be consistent with other things which were inconsistent with it).

- it seems possible to imagine finding out (as in some nightmare) that we have never been able to talk, and that whenever we thought we were talking we were always making some kind of mistake.

- in order to understand the falsehood of the statement, we need to have some conception of what it would be for it to be true.

(3) When I say 'It is not possible to talk', you might think 'well - can this be true?'. If you thought it was true, you would (a) not answer me (or not answer me honestly) and (b) realise that I hadn't really spoken to you. You might think 'these sounded like words that meant something, and because I was taken in by their apparent meaning they made me think - and come to a remarkable conclusion - but in fact they didn't mean anything at all'.

All this clearly has a kind of fairytale quality to it, and, I think (for reasons that are peripheral for the present, but related to the private language argument), may also be incoherent.

However - it suggests something useful:

We can show things without saying them (obviously), but we may also be able to show things by saying something, and even by seeming to say what it is that we mean to show - even though what we seem to be saying can't be true.

Imagine you have a rare kind of colour blindness, so that under indoor lighting you see red as green & green as red, but in daylight you don't. You are coming to me for evening classes in road safety, and I need to show you which colour of light means stop and which means go. What do I show you?

And what if we don't know which one of us is making the mistake? What do the words 'red' and 'green' mean here? All we know is that we don't agree about which indoor light looks like the outdoor light. We can talk, though, because we can predict what each other will say when asked.

If I wanted to show you what to look for in an outdoor red light, I would show you an indoor light that looked green to me. I might even say 'look - this is a red light'.

When I say 'it is possible to talk' must be true in any language game in which its equivalent (under interpretation) can be constructed, I may be simultaneously introducing you to the possibility of its contrary.

And: maybe if you can't imagine the contrary you can't understand the statement.

The construction of the theoretical certainty depends upon an appreciation of the phenomenological uncertainty. If you can't think the unsayable - if you can't live with Cartesian anxiety - then you will never have theoretical certainty.

My thesis depends upon showing you a possibility that cannot be articulably entertained, and showing you by pretending to entertain it - the argument is a reductio from the presumption of its own unintelligibility.

Instead of an axiomatic contradiction, we have a contradiction with the possibility of axiomatisation.

Syntax and Paradoxes

Syntactical rules may appear to set minimal standards of intelligibility, but they don't guarantee intelligibility.

We shouldn't be deceived by the 'obviousness' of some of these rules - finding them obvious is really just the condition of being able to talk. In other words, we wouldn't be able to to talk if we didn't find quite a few of them obvious.

For Kripkean reasons, we cannot render a syntactical rule mechanically: any potential mechanical isomorph could be unreliable. We cannot say 'to find the consequence of the non-contradiction rule, press this button', because we would have to use the non-contradiction rule to test the reliability of the button.

We can only state syntactical rules if we have a working semantic context: if we can speak.

So while they might represent boundaries of intelligibity they cannot exist independently of it. They cannot conceptually precede it. We can't make sense of a syntactical rule outside of a semantic context.

So: incompleteness/inconsistency theorems, which seem to be purely syntactical, can only be constructed in a semantic context within which the normative syntactical rules can be stated unequivocally. It is within this semantic context that the rules have the correct scope, and can stand as tests for any potential mechancial isomorph.

It is because we can describe a logic gate functionally that it counts as a logic gate.

(But also: we describe the world this way as well. It is only because we can describe the world functionally that we can describe it at all - Goodman's paradox is constructed the same way as Kripke's).

Godel's paradoxes, then, become semantic paradoxes because we can say whether it is true that certain syntactical rules have been observed.

?

Moore's Paradox

"We cannot talk" is, arguably, a close relative of Moore's 'paradox': "It is raining, but I do not believe it is raining".

There are lots of detailed explications of this very odd statement, but it seems to me that there is a distinction that could help us to understand (and 'solve') it - between self-contradictory statements (e.g. the Liar) and statements which are incoherent with their own status as statements.

"I cannot talk" is clearly an example of the latter. It makes perfect sense to say "X cannot talk" where X is a third party - a mute, an animal or an object; but not where we substitute the author of the statement into X. If you can't talk, you can't say so either; so if you utter these words, you either cannot be saying what they are normally taken to mean (i.e. that potential interpretaion is barred - no charitable principle can rescue it) or you are not meaning to talk seriously - perhaps you are doing a bit of contentless verbalising, or playing word games.

In the traditional formulation, "It's raining" and "I don't believe it's raining" from the same speaker are not directly contradictory, but they do challenge each other's intelligibility. Making both statements is so grossly inconsistent with the context of assertion combined what we normally mean by "believe" that they are rendered unintelligible. An interlocutor who makes these statements, as serious contributions to a conversation, is either revealing an inability to converse or seeking to undermine the game.

Meaning, validity, the Liar ...

Liar paradoxes seem unavoidable in languages that allow us to attribute validity or truth to our own judgements in an unrestricted way.

It seems likely that only languages that allow us to attribute validity or truth can also allow us to attribute meaning: can allow us to say when a gesture or signal is being appropriately ('truly') used, or - more important - can allow us to state the conditions under which a gesture would be being appropriately used.

Is this a radical suggestion: that a gesture or signal has meaning if it can be translated into a language that can be used to (at least partly) say what its meaning is? Does this link having meaning too closely to being able to attribute meaning? Some reflection on the use of 'to mean' makes me think not, but ...

So we have behaviour; then we have games - in which behaviour may or may not be within the rules of the game; then we have language games, in which we have truth and falsehood - which have some relationship with the rules of the game, but which may not just reflect them; then we have language games in which truth and falsehood can be explicitly attributed, and in which concepts like 'true' and 'meaningful' have a role.

In this third group, liar paradoxes can be formulated, open question arguments arise, and we can do philosophy: e.g. by making statements like those in this post.

Godel, Arithmetic and Validity

Godel's proofs show that any axiomatic system to which arithmetic could be reduced must be either incomplete or self-contradictory.

The proof doesn't depend on any interpretation of the symbolism that he uses, so his paradoxes are, in a sense, syntactical rather than semantic. What he shows is that any system which could encode arithmetical calculations will also be able to encode computations which can be interpreted as tests of the validity of its own theorems, and which are therefore self-referential (since the outcomes of these validity tests are also theorems).

We don't normally think of arithmetic as being intrinsically about truth-telling, or validity. Why is it impossible to do arithmetic without generating valdity tests? If Godel is right, it can't be just an accident. His method depends on being able (in prinicple) to arithmetise all logical validity testing - it seems to make logical validity testing and an essential subset of arithmetical computation isomorphic.

Numbers are truth and truth is numbers.

Version 11

Intelligibility ... I could be wandering into a bog here.

Physicalism

A physicalist would say, of course, that the condition of 'seeing a bright light' (however intentional) would have to consist in some set of physical facts - e.g. about my nervous system, about some interaction between photons and my retina, about some set of brain activities.

This has to be wrong for two reasons:

(1) Suppose a physicalist specified such a set of conditions, and then met a person who was demonstrably in the relevant physical state but denied seeing a bright light. Would the physicalist have to say they were lying and/or mistaken?

Not necessarily: they could say that their specification had failed an experimental test, and some improved specification was required. However, this process of improvement would have to come to an end, in order to save the physicalism - and it would have to be an end which did not depend on testing against reports of seeing bright lights. At some point, the physicalist account would have to be more reliable than the intentional report - it would have to provide a basis for saying that the intentional report was dishonest or mistaken.

We do, of course, often decide that people are being dishonest about intentional states - and we may also conclude that they have made mistakes in the way they express them. However, any decisions of this kind are subject to Kripkean ambiguities and Davidsonian interpretive choices. It can never be a matter of absolute fact that an intentional report is dishonest or mistaken - there will always be some interpretation under which it can be saved. But when we reject these interpretations (as we often do) we are making a normative choice.

Some religious people, for instance, claim to believe remarkable eye witness accounts of miracles in order to save the integrity of their saints. For the physicalist to insist that my brain state plus certain environmental factors indicate that I am seeing a bright light - whatever I may appear to say about this on my own behalf - they are making a choice of this kind.

From this perspective, physicalism - along with all the other metaphysical foundationalisms - becomes a position adopted through interpretive choice. I guess this is what Simon Blackburn has in mind when he talks about quasi-realism. We should talk as though we are realists. But then why not quasi-idealism? Apart from making the computations more complicated there doesn't seem to be much to object to in it ...

A really committed physicalist might deny that the examples given above are coherent - claiming that someone in the appropriate physical state would also, necessarily, say appropriate things (utter appropriate noises?). But this just skips the demonstration - in order to show that physical states and intentional states go together, they need to be separately specifiable. If they can't be, then the physicalism becomes contentless.

(2) OQ again: For the physicalist, physicalism itself (the set of beliefs comprising physicalism) would have to be reducible to some physical state of affairs. To be a physicalist is to be in an intentional state - of believing that intentional states can be reduced to physical states - that must, therefore, itself, be reducible to a physical state.

One problem with this (related to (1) above) is that it's hard to see how a physical states can be right or wrong - it's just a physical state. Also, we'd need - in principle - to be able to specify this state without begging the question (e.g. by calling it 'the state of being a physicalist'), and without being diverted by accidental equivalences (e.g. if all physicalists, and no non-physicalists, happened to be tall & have large noses, for instance). We'd also have to accomodate the possibility (as in (1)) of people claiming to be physicalists - appearing to know the arguments etc. - but being wrong, because they were in an inappropriate physical state.

A more serious problem, though, is the general OQ problem: if having a true belief about X is to be in a certain physical state, then there must be a physical test for the truth of such a belief. In particular, there must be a physical test for the truth of physicalism. If there isn't such a test - if there is some non-physical test for the truth of the belief - then physicalism is incomplete. But to insist on there being such a test is to beg the question - it's simply to insist that physicalism must be correct.

N.B. - An argument against physicalsim is not automatically an argument for some other metaphysical substrate (idealism, for instance). All these metaphysical foundationalisms are equally vulnerable to similar kinds of arguments.

Also, it is perfectly possible that there are no other fundamental objects or forces in the world beyond those discovered by physical science - in the sense that we won't find any there, no matter how hard we look. (There are no fairies at the bottom of the garden).

However, this isn't sufficient for a physicalist account of intentional states.

Imagine an artist making a representation of a face: there must be a physical medium for this, but the choice of medium is only marginally relevant to the description of the activity. No list of media and of related techniques could give an exhaustive account of this activity, because some new medium - holography is a recent example - might come along in which the intentional act of 'making a representation' could be carried out.

This problem becomes worse if you think of 'making a representation' in terms of 'making a good representation' - i.e. making a representation that is recognisable, or apt.

Inconsistency

An inconsistent system may have its own consistency as a theorem, but we could not know what this theorem meant.

Reductionism

The central ethical decision that each of us has to make is whom to treat as intelligible.

There are two aspects of intelligibility:

(1) Being able to talk - being someone with whom we can hold a conversation.
(2) Being the subject of an intelligible narrative - acting in a way which can be explained.

There are no intelligible rules which completely determine the limits of intelligible talk or action. This is the open question argument again - perhaps the central case of it. If we could state all these rules we could ask whether they defined intelligibility - whether it was intelligible to say this.

In particular, there cannot be a list of rules which relates intelligibility to factual conditions. There is a reason for this in addition to the open question problem: what counts as a fact depends on what we count as intelligible. Claims about facts are subject to intelligibility decisions.

Similarly, it is not the case that just anything can count as intelligible. We do make decisions about this, and they are not arbitrary or random. For any language game (for any 'game of intelligibility') there will be moves (e.g. allowing contradictions) which would render the game unintelligible.

Some of the conditions of intelligibility of our language game (the one we are playing now) will look universal - will look like general rules of intelligibility. We can do experiments with some of these - Quinean experiments - to see which ones we can vary or do without, and which we cannot.

There are factual statements ("we can talk") and formal statements (the non-contradiction rule) which must be true in any language game. If we found a game which rendered these false under any possible translation schema, we would have to say that it was not a language game.

But whether we had exhausted all possibilities would be an empirical matter, and subject to Kripkean limitations (had we considered all possible interpretations?). There would always be some interpretation which rendered the game trivially intelligible - though possibly too complex to play.

Phenomenological Issues

It's a mistake to look for metaphysical 'grounds' for theorising in general, since there isn't any relevant way of distinguishing metaphysical hypotheses from other kinds. Our capacity to form hypotheses - our capacity to talk about the world - isn't rendered more intelligible by the generation of further hypotheses of the same kind as those which we are trying to illuminate.

Our capacity to talk cannot, itself, by rendered hypothetical. Therefore, realism may only be a consequence of the possibility of language: it certainly can't underpin it.

Most of us think of our internal experience - our phenomenological states - as contributing to our ability to speak: we can talk about the world because we sense it, and we have a direct experience of what these sensations are.

Cases of sensory deprivation leading to specific linguistic incompetence support this: blind people cannot identify colours, nor deaf people sounds.

But we need to separate two things here:

(1) What we say about our internal states.
(2) The internal states 'in themselves'.

We can clearly talk about our internal states. However, we can only talk about them as represented in language.

I can say 'I see a bright light', but I cannot independently identify my seeing of a bright light except by making this statement (or one equivalent). The condition of seeing a bright light is irreducibly intentional.

Also, this is a public, not a private condition to be in: we can have an argument about whether I really did see a white light - or whether I made a mistake, or am telling a lie. This is true of the most primitive sensory reports - even those which look like candidates for 'sense data'.

Error and deceit are features of the reporting stage, however: I can't be mistaken or dishonest about my internal experiences in themselves. I also can't share them 'in themselves'. (There's even something metaphorical and indirect about this way of talking.)

The states 'in themselves' are beetles in boxes (PI 293). We all have them, but they only come into theoretical focus, or even appear to be shared, when we talk about them. This is why 'consciousness' in this sense can never be the subject of scientific enquiry: it stands just outside the realm of any possible articulated enquiry.

The apparent link between these phenomenological processes and being able to speak - knowing the truth about the world by being acquainted with it through sensation - is, itself, a phenomenological intuition. We might as well say about it that it is our internal experience of our capacity to speak. This is such a disturbing thought that many people find it impossible to entertain, but it becomes obvious after a little reflection: my only incontrovertable grounds for thinking you might have the same intuitions as I have are found in what you say about them.

You might say that I have grounds in your behaviour, but these are not incontrovertible - for Kripkean (Rules and Private Language) reasons. The interpretation of behaviour is always corrigible, but you can only doubt what I say about my phenomenological states by doubting my linguistic competency or my honesty. In both cases, you are doubting whether we are, in fact, conversing in good faith - you are doubting whether we are really talking.

Aha ...

I've just spotted a difficulty, which I need to scope out:

'I can talk' looks empirical, as does 'You can talk'. For Moorean reasons, they may be slightly different.

Of course, 'You can talk' looks like 'You can walk'. But there are empirical tests which we can agree on for 'You can walk' (without which the statement would begin to lose its meaning ... ).

But while there are some similar empirical tests for 'You can talk', they aren't relevant to the fundamental case that I want to depend on. We might check whether a recovering stroke victim can talk, but we can't 'check' whether 'You can talk' is in general true: in the second sense, there is a normative factor associated with the radical translation argument. I decide (for me) that you can talk - even if only tacitly, by talking to you.

We may, of course decide what 'counts' as 'walking' as well, of course. (And we can probably completely articulate this within general tacit constraints).

But we can still talk even if we can't agree on what counts as walking ...

'Socialisation'...

We might as well call a 'socialised' individual one who can participate in and be represented within the conversation we are now having.

I'd probably have to say 'represented as a person' or 'represented as intelligible' ... which may make this a bit muddy?

Bloor (and others) on enthusiasm, private languages etc.

The private language argument is an also an anti-sceptical argument.

The issue is not whether there can be, in some representation of cognitive processing, something that looks like a private language (or, indeed, a 'language of thought') - a signalling system, for instance, that has a syntactical structure.

It is, instead, whether this thing that looks like a private language can have meanings attributed to its 'signals' that do not depend upon the meanings of language game within which it is described. Bloor thinks of a social matrix, or socialisation here. Wittgenstein is vague, but points to the need for public tests of some kinds. Locke refers to reason ...

As Bloor says, (p55, Wittgenstein, Rules and Institutions), arguments of this kind cannot 'claim absolute cogency or a decisive victory'.

But he's making the same mistake as Quine did - having found a concrete narrative (science, society) of the process which seems to exemplify or underpin public rationality, he mistakes the content for the form. It's like believing a laptop can think, or believing a word processor can speak because it can check grammar.

The normative aspects of attributing meaning, or rationality, cannot be captured in this kind of concrete narrative. Not only is there an open question problem here - about the normative status of the narrative, about it's capacity to confer validity on the processes it seeks to explain - but the very nature of the narrative itself is called into question. What is 'society' or 'socialisation'? If it becomes just 'what underpins rationality' (Spinoza? !), then this empties the narrative of content and the explanation of sense.

And , of course, the narrative depends, for its meaning, on the very categories it seeks to explain (truth, rule following, knoweldge). And, sadly, it can only claim to support them by trying to deny or marginalise the very aspect of them that we most want explained: the validity of their content.

If, on the other hand, we show that these categories can be recursively grounded in our ability to talk about things - an ability which needs no further exposition, since querying it is unintelligible - then we not only have the fundamental argument that Bloor denies is possible, but we also avoid the open question trap.

'Essential' metaphysics ...

Is it possible that being a Platonist makes it easier to do mathematics?

There's no reason why not, and plenty of pointers in the opposite direction: religious people are often better able to survive misery; people with less imagination may be less distracted by fear ...

None of these amount to arguments for the substantive positions, but if the 'heuristic postulate' was true, how should we respond?

Personally, I'd rather do philosophy. Is it impossible to be a good philosopher of mathematics and also a good mathematician?

Scientists tend to be naïve sensory empiricists. Could they do such good science if they weren't?

What delusions must a philosopher entertain in order to be good at philosophy?

Version 10
More on meaning and meaningfulness.

The dogma of any discipline is meaningless within the discipline - empirical science and religion are on equal footing in this respect. Where they are not on an equal footing is with respect to their scope:

It's quite easy to avoid language which makes religious presuppositions, but almost impossible to avoid language which makes scientific presuppositions. We can talk about the world without talking about God, but we can't talk about the world without talking about the world ...

Version 9
Working changes. Not very important.

Irrationality and games

In certain games, it can be wise to undermine your opponents' confidence in your rationality.

In a language game, we are engaged in constructing rationality: in negotiating the conditions of sense-making. In a commercial language game, the material dimensions of this negotiation can't be ignored. They partly determine what makes sense, but they also partly determine which aspects of the negotiation have to remain tacit.

In any language game - in any conversation - some of this negotiation is tacit, of course, but we think this can be ignored in 'scientific' or 'academic' language games.

But even in an ideal academic conversation ideas are products that are bought and sold ...

Commercial language

The reason this needs a closer look is because it provides a useful model of tacit negotiation over what makes sense.

In a commercial discussion, the material objectives of the parties tacitly (from the point of view of the language of the discussion) constrain the sense making opportunities. However a commercial negotiation can only continue if there is a working language within which to negotiate.

In this context, the comforting metaphors which inform the philosophical discussions of scientific truth telling look much less credible.

Even when we're standing at the door holding the key, we find that there's a particular way of turning it - a special flick of the wrist - that we need to learn.

Metalanguages ...

The one above the one we're using - in which a theory of truth for our language could be formulated - cannot be translated, and, so, cannot be recognised as a language ...

The one below, which we can give an account of, is reduced to syntax, rules, and symbols. It's meaning has been relegated to 'interpretation' and lost to truth-telling. Is it a language or a board game?

Neurath's Boat

It isn's just the planks and frames that keep the boat afloat, but also the conceptions of the sailors: their plans of the boat, their principles of naval architecture, the epistemological foundations of both ...

Which is why they can never have a complete design, never mind do a complete rebuild.

While it might seem intelligible to wonder about the foundations of any knowledge that we have, and to seek arguments for its reliability, it is a serious mistake to seek an argument for the possibility of intelligible conversation. If argument is possible, then so is intelligible conversation.

This doesn't mean that questions such as 'how do we speak?' and 'how can we speak?' don't make sense in certain contexts - e.g. the neurology of language, linguistic theory etc.. These disciplines can give us insights into the mechanical or computational processes involved in speech, or into linguistic structure. They may even show us how certain structures or processes are necessary for reliable communication, in the sense that (under interpretation) all languages incorporate these structures and processes.

What they cannot show is that some set of mechanical, cognitive, or structural criteria guarantee intelligibility - are not just necessary, but also sufficient. This is a corollary of Davidson's observations on interpretation - that there is always more than one solution to the translation problem, because we must make assumptions about the normative, epistemological, and motivational character of any potential interlocutor. The 'Principle of Charity' renders the interoluctor 'human', but not only is this a normative posture adopted by the translator, there are also quite likely ot be different renderings of this humanity.

We cannot get from grunts to sense without making normative choices, however 'obvious' these choices might seem to a routine user of a human language.

And yet sense is guaranteed, within any conversation, by the possibility of that conversation. 'This conversation is impossible' is not a possible conversational move.

Neurological, cognitive, and linguistic theories - or, indeed, epistemological theories - are all parts of some intelligible conversation or another. They cannot provide a better demonstration of the intelligibility of that conversation than their own intelligibility already guarantees.

To ask a question about that intelligibility is to ask a question about the possibility of asking questions.

Language and lying

I thought for a while that lying depended on the possibility of truth-telling. So it might, but there are other kinds of deception than lying - kinds which might depend only on the possibility of interpretation (and so on presenting behaviour designed to deceive).

There are deceptions in nature - the moth with eye patterns on its wings is not literally saying 'I am the face of a dangerous creature', although it has evolved the spots to sustain this deception.

But that's *our* explanation (interpretation), of course...

Thoughts on thoughts ...

The 'phenomenology' of thinking is, possibly, one of the things we feel most familiar with - thus Descartes' infamous computation.

But we only know what each other thinks because either (a) we have drawn conclusions from behaviour or (b) we have been told by the thinker.

(a) is always corrigible (Kripke), and (b) depends on the presumption that we understand what we are saying to each other.

Actually, behaviour interpretation isn't just 'freely' corrigible - these theories, themselves, need to be intelligible to those who discuss them.

Of course we have unarticulated thoughts. And we have thoughts which could not easily (or ever) be discovered from our behaviour. But thoughts which are forever private are also forever outside the scope of public theory - and so are outside the scope of its tests of intelligibility. Private thoughts are private.

Philosophical theorising is a public business, carried on through participation in a 'language game'. In so far as unarticulated thoughts appear in philosophical theories, they appear as intelligible - articulated - interpretations of behaviour. No private thoughts appear here in any substance (perhaps all we can say is that we may have private thoughts ...).

This isn't behaviourism: I don't deny private phenomenologies. But it does seem clear that it is only when they are articulated that they engage with theory. (Articulating them is an important and creative business, of course - but we don't always do it).

Suppose we said of all these unarticulated thoughts - whether they are disengaged from public theory, or precursors to it - that they are just the private phenomenology of language users? That we only think we share the unarticulated part because we share the articulated part through our participation in the language?

This would shed a different light on the motivation to talk of 'languages of thought'. Actually, what we have is 'thoughts of language'. And, of course, these thoughts appear to have a grammar ...

The rest remain private.

Truth telling ...

The rules for truth telling are whatever rules (tacit and explicit) that we follow to make conversation possible.

Sometimes, maybe generally, tacit rules can only be made explicit by changing the rules - even to the extent of changing what we mean by 'explicit'.

The open question argument guarantees that we can never make all the rules we are using explicit (in any intelligibly translatable sense).

A language for which the rules have been made fully explicit can be treated entirely syntactically - as s signalling system, or not a language at all.

Meaning and 'transmission'

Suppose I believe that it isn't possible to talk to people. I might still know that if I utter the string "It isn't possible to talk to people" in your hearing, then you will entertain this same thought.

This is credible without any mention of meaning, because I can make you think things without 'saying' anything at all. I can make you think there is someone behind you by glancing over your shoulder.

But despite the causal process being credible, I cannot mean by my utterance that it isn't possible to talk. If I did, then I would also mean that that utterance didn't mean anything - including that it isn't possible to talk.

Suppose you were disabling a bomb, and I was instructing you. You have to choose one of two wires to cut: a red one and a blue one. If you cut the blue wire, you will be blown to bits, and if you cut the red wire the bomb will be disabled.

You don't know which wire to cut, but I do. However, you believe that I want to blow you up, and that I will tell you to cut the wrong wire. Unknown to you, I know of this belief - however, I do not want to blow you up.

I therefore tell you to cut the blue wire, knowing that you will, as a result of your belief, cut the red one.

I have transmitted a signal to you that has led you to cut the right wire, but this does not render 'Cut the red wire' as a valid translation of what I said, which was 'Cut the blue wire'.

Meaning Commitments

We need to be committed to something *intelligible* by what we say, and there's a 'public' aspect to intelligibility.

And there's also a normative aspect. We partly 'decide' to treat some moves as intelligible. We certainly decide how much effort to put into rescuing intelligibility...

(More 'private language' issues ...)

The Private Language Argument

The problem with trying to construct a private language is that it can't count as a language - it can't be translated.

Only public languages can be translated.

It isn't so much that we wouldn't be able to check our rule following, as that intelligibility is conferred by public descriptions and explanations.

If I was the last human being, and was 'writing' this, I would really be saying nothing (whatever my delusions). At best, I would be acting out my own (unintelligible) nature.

Or I might be making notes for myself to read later ...

Whatever that would mean?

In fact, there's maybe an even better way of putting it:

If I talk of a group (Group S) having a 'social norm', or of sharing a judgement (N), I am saying to the participants in my conversation (Group E) that this is the case. If the shared norm or judgement of Group S is expressed in the language of group S, then the norm must also be intelligible in the language of Group E. Otherwise we would have to doubt whether we had a correct translation (Davidson and Quine).

What I'm saying here is that if we understand the language of Group S, then we understand N.

Here's a slightly contentious step: If we understand N, then we understand how to adhere to N, at least in principle. This means that, even if we believe N to be a 'bad' norm for us, we understand how Group S has come by it. For a shared conversation to work (one that included S and E), we would need to resolve this dissonance - either S would come to reject N, or we would come to accept N, or we would both adhere to a norm N' which gave a contingent account of N that we could both accept (S would see that there were circumstances in which N didn't hold, and we would see that there were circumstances in which it did).

If, despite our best efforts, we cannot achieve a shared conversation between E and S because we cannot reach an accommodation over N, then we have, again, to be sceptical about our understanding of N, and, therefore, of our judgement that we can attribute this norm to S.

Kripke's presentation of his paradox is deceptive because 'quadding' is intelligible to us, as are a great many (maybe infinitely many) other rules that a subject might be following.

But each one of them can receive the 'N' treatment, or it doesn't count as a rule. In other words, the paradox relies on a dissonance which can always be resolved.

If we could never guess, nor check, which rule a subject was following, why would we imagine they were following a rule at all? Can it make sense to say they are definitely following a rule, but not one of the ones we have been able to check? (This is beginning to sound like Gregory Chaitin territory ...).

If we try to recast Kripke, in line with this suggestion, we find ourselves saying that we can never guarantee that our interpretation of someone's behaviour as intelligible is correct. This doesn't seem so radical.

This approach also solves a problem for 'group judgements': there is nothing especially privileged about a group over an individual with respect to the private language argument. Each may make exactly the same mistake - judging that they are following a rule when they are not. This is perfectly clear from real examples of groups amending judgements about social norms under economic or political pressure.

What is not possible is to speculate that the rules of the conversation within which we are speculating may vary in a way which fundamentally changes the nature of the activity of speculation.

Is the solution I've outlined for Kripke the same as his solution?

Maybe.

I don't like 'collective brute facts', but that may just be an unfortunate coining. He definitely relates the solution to a 'form of life'.

He gets into a muddle about groups - talking as though 'we all roughly agree' can mean anything, and doesn't just beg the question. This is similar to a kind of mistake Wittgenstein could be accused of - treating the 'form of life' as something that could be described, or even pointed to, within the language game. I know he could be read as denying this in a few places, but he talks as though it can be done.

Do I? Only metaphorically, or heuristically. I have a better way of saying the same thing: through the unintelligibility of denying we can talk, and constructing a more conventional analytic argument from that starting point.

If the solution to the paradox is tied to the possibility of language, it has the neat result that we can only state the paradox if it is either soluble or harmless (like the semantic paradoxes).

Also: I think my solution reduces the attractiveness of 'internalist' accounts (like Chomsky's?). Or at least, it renders these scientific rather than normative or epistemological. We may discover all sorts of brain processes which underpin, or 'explain', language use in our case, but this is just functional sub-strate, and the normative/truth-telling/meaning aspects could be implemented in other ways - e.g. by a computer programme, or an alien.

We can rule out these possibilities, of course, but that , itself, would be a normative judgement. Computer's are beginning to 'talk' quite well. Whether or not we finally include them, or their descendants, in our community of conversation will be a social issue rather than a cognitive one.

Just as whether we, now, would regard that development as acceptable is.

Version 8

I've changed my mind about 'as if' - it's not that easy ...

And I have made some necessary changes the bit about Kripke.

Aesthetics

Here's an odd thing: there doesn't seem to be an open question argument against a general aesthetic theory. Asking whether a theory of beauty is beautiful seems to be perfectly intelligible.

Maybe it's just a mix of the other four - knowledge, truth, goodness and meaning. Maybe Keats was partly right ...

Something rings true about this, though: a work of art can guide us to something new by way of meaning, truth, knowledge or goodness. It does this without saying - but by showing. We can't 'translate' it, but we can tell stories about it.

Kripke and Goodman

Kripke's paradox depends on attributing a single intentional stance on the basis of a specific description of behaviour.

This is probably incoherent.

We don't describe a table as wanting to remain stationary, regardless of the "behavioural" evidence.

To attribute a single intentional stance is also to attribute the capacity to have intentional stances - and, therefore, to attribute other intentional stances. It seems unlikely that it could ever be intelligible to attribute a completely incoherent set of intentional stances to someone. In fact, for Davidsonian reasons, attributing intentional stances is probably like attributing rationality, or language: there is a normative aspect to it, and it's likely to be holistic. We're attributing a form of life.

In particular we're not depending on a restricted set of evidence, nor on a purely 'computational' argument - and most of the peculiar possibilities allowed by Kripke's account would be eliminated by holistic and 'principle of charity' types of consideration.

A similar type of argument can probably deal with Goodman's paradox. A language game needs to be playable - this is a significant constraint on 'grue' type semantics.

Version 7

Another update, with some new sections started ...

The famous objection to logical positivism - that its fundamental discriminatory principle was neither analytically true nor synthetically testable - can be extended to any general empiricist epistemology (any theory about how we obtain reliable knowledge about the world).

This is because any theory of this kind must also be a theory about the world - a theory that the world is the kind of place about which knowledge can be acquired in this way.

Our epistemological theories are theories about the world, and so must either be self-validating or incomplete.

Perhaps this is just another way of saying that they are subject to the open question objection, and so require recursive exposition.

The typology constrains the scope of "membership" in a specific way. Also, the construction of the typology depends upon the way this constraint works.

The rule "A set cannot be a member of itself" sounds simple but isn't. When it's modified to avoid 'overkill' it rapidly becomes complex - only becoming intelligible again when it approaches very closely to "A set cannot be a member of itself if this would result in contradiction" - which, of course, is exactly what we're trying to avoid by other means.

This pattern is beginning to sound familiar ... Which means, I suppose, that I may have made it up.

The reason that meta-language solutions to the semantic paradoxes can't work is because to characterise the concept which generates the paradox (e.g. 'truth'), a language which is not in the hierarchy is required.

I would suppose that the theory of types is similarly vulnerable to a diagonalisation paradox based on the intensional definition of a set which depended on a feature of the hierarchy in such a way that the set became both typeless and typed?

I'm sure this must have been done, and I'll have a look.