John Searle on Theory & the Piecemeal Approach

It's often said that analytic philosophy is essentially ‘piecemeal’ in approach and is therefore highly suspicious of not just ‘grand systems’ but of theory itself. (This was certainly the case when it comes to the late Wittgenstein.) However, over the last forty or so years many philosophers, especially metaphysicians, have been creating their own large-scale theories of philosophy, if not their own ‘grand systems’.

John Searle sees the importance of the philosophical construction of theories, if not ‘grand theories’:

"When I was in Oxford the term piecemeal was a term of praise. What we wanted were little bitty results. We were suspicious of large results. But I think that was due to the fact that philosophers who wanted large results had done such a bad job... I think Hume very much had a general theory, Locke had a general theory, even Berkeley in his way had a general theory... You want to know how it all hangs together." (191)

We can say that the ‘piecemeal’ approach was nothing but a reaction to the large-scale approaches to philosophy. Similarly, Searle’s own large-scale approach is nothing more than a reaction against the piecemeal approach. That ‘s how philosophy, and indeed politics and much else, work. Reaction and counter-reaction and so on. In thirty years’ time no doubt some philosopher, or many philosophers, will have equally disparaging things to say about Searle’s own large-scale approach and they too, in time, will create a reaction to what they've done.

In addition, Searle says that "philosophers who wanted large results had done such a bad job" (191). Who’s to say that philosophers in the future won’t also think Searle has done ‘a bad job’ with his general or theoretical philosophies? This is almost bound to be the case.

In any case, there have been a few systematisers in analytic philosophy, especially in America. Some would include, for instance, Sellars, Nelson Goodman, Dewey, and even Quine in this category. Perhaps also Davidson. It may even be the case that Dummett, here at home, is a systematiser just as his hero Frege was a kind of systematiser. Of course Russell and Collingwood were also guilty of some broad-sweeping theories or generalisation in parts of their philosophies. Indeed, because Collingwood was a historicist of sorts, it's hardly a surprise that he went for ‘large results’ when he practised his philosophy.

Perhaps a concern with the philosophical minutia of this or that says more about the types of minds or intellects of certain philosophers than it does about their antipathy towards grand systems or theory. Perhaps they are simply incapable of synthesis or of looking at things contextually or broadly. Of course we can look at this inversely. The systematisers may be incapable of dealing with the minutia for reasons of intellectual deficiency or perhaps it simply bores them rigid. The best philosophers may be those who can both synthesise and analyse and see the pros and cons of both approaches and also see when one or the other is required within a particular philosophical context.

Is it the case, anyway, that the systematisers really did ‘such a bad job’ (191) anyway? Did Kant do such a bad job? What about Hegel? Of course not everything they said was first-rate or even correct; though was everything J.L. Austin said first rate and correct? More to the point, is everything said by Searle first rate or even correct?

Being first-rate or even correct may not have anything or much to do with whether or not one is a systematiser or a practitioner of the piecemeal approach. Good and bad philosophy occurs in both approaches and isn't exclusive to either one or the other. There is good large-scale work and bad analytic work. There is good analytic work and bad large-scale work. Why generalise at all about either of these approaches, especially when bearing in mind the piecemeal protagonists’ dismissal of all generalisations and generalising tendencies?

In addition, perhaps many hard-core analytic types, even now, simply don't "want to know how it all hangs together" (191). Perhaps they would even claim that the very idea of philosophy showing us how it all hangs together doesn’t make much sense. Either that, or the very attempt to show how it all hangs together is intrinsically suspect and far too ‘Continental’ a desire for the hard-core Anglo-American analytic. Again, perhaps the large-scale just bores them just as the minutia bores many Continental philosophers.

Though perhaps ‘theory’ is nothing particularly complex or even esoteric anyway. The way Searle describes theory it would certainly not seem so:

".... the idea that you has a systematic set of propositions, logically related to each other which could account for a whole domain, that’s a wonderful idea. Euclid’s Elements is one of the classics. Aristotle’s full of theories. That’s what I want. I want a theory in the classical, Greek, Aristotelian sense." (191/2)

That way of describing theory, or a theory, seems to make just about every philosopher a theoretician – even those, like Nietzsche, Heidegger, Wittgenstein and Derrida, who are supposedly ‘anti-theoretical’. Most philosophers, both in and out of the Continent, would have thought that they had "a systematic set of propositions" (191). Again, even if they were anti-theory or anti-truth. Whether they also thought that these propositions were "logically related to each other" (191) is something else. It would depend on how they take ‘logically related’ or how serious their logical skills were. Would any philosophers want a set of propositions which were not logically – or otherwise – related to each other? Not even Derrida!

Even when Wittgenstein argued that ‘whole domains’ could never be accounted for (at least in the case of language or language-use), then his domain of an unaccountable domain was still his domain – even if it couldn’t be accounted for. Similarly, Gödel’s domain was still the whole of mathematics even if the whole of mathematics couldn’t be fully accounted for in terms of both its completeness and consistency (in the case, at least, of individual axiom systems).

A Wittgensteinian Take on Linguistic Rules & Meanings

Quine expresses a position on understanding expressions that is pure Wittgenstein. Christopher Hookway expresses Quine’s and Wittgenstein’s position by writing that

"I understand an expression when I know the rule which governs its use and use it in accordance with that rule". [1988]

In that case, rules take the place which abstract meanings had in much traditional philosophy. Instead of our understanding of an expression (or a statement) depending on that expression's meaning, we depend instead on its rule-of-use. More clearly, this rule will "govern its use". It tells us how we must use that expression.

Not only is this the case with understanding: the rule must make us "use the word" in accordance with it. Instead of meanings being of prime importance when it comes to understanding expressions, rules instead take that role.

Of course rules are very different things when compared to meanings.

Whereas rules tell us how to use an expression (or how they govern a word’s use), meanings don't tell us to do anything. In another sense, abstract meanings (taken in and of themselves) don't actually tell us anything. Instead, in terms of understanding an expression, what we must do is gain access to its meaning and then determine the actual nature of that meaning. A meaning, therefore, isn't normative in the way a rule is. If meanings don't determine the rules, we must use rules when using an expression. Then, in a sense, a meaning needs to be interpreted and understood; which is in opposition to the fact that rules aren't really interpreted at all (i.e., they're not abstract objects – they're linguistic).

Rules must also be understood in a different way.

If rules are always expressed linguistically and clearly, then there's a sense in which we can't go wrong in understanding and using them. Meanings, on the other hand, are either abstract objects or abstract mental entities. And if that's what they are, if they're something non-linguistic (when we first gain access to them), then there will be ways that make it the case that we could go wrong. The difference between a meaning and a rule would therefore be a little like an electrical device that comes with its instructions and one that does not. A meaning would be analogous in some way to the electrical device without instructions and the rule would correspond to the one that came with its own instructions.

Let's give an example of a linguistic rule in practice.

Take the expression 'cat'. In this case a conventional rule will tell us that this expression must always refer to a cat or to cats. We understand 'cat', then, when we accept the rule that tells us that it must be used to refer to cats. In terms of rules governing the actual use of the expression 'cat': they only apply in the situations in which we want other people to know that we're referring to (or talking about) a cat. We must only use 'cat' - and always use that word - when we want to refer to cats and also when we want to be understood by others as referring to a cat. And because this and all similar rules are both relatively strict and precise, then by definition it will rule out certain obvious - and not so obvious - mistakes in the use and understanding of 'cat'.

For example, we can't use the word to express our thought that "politicians are all liars". This is an obvious misuse of the word. However, neither can we use 'cat' to refer to only cats that are brown in colour. If that were the case, we would either require a new expression or a new rule. (In some cases we would require both.)

In terms of the misuses of the word 'cat' when it comes to understanding that word.

Clearly we can't take the rule to state that 'cat' is an expression that must always be used to express the fact that "cars are likely to crash on this road". In that case, we can say: "I don't like this road. It is a cat road." Clearly this is an extreme misunderstanding of the word 'cat' if we've already accepted the initial rules that explained to us how that word must be used and understood. However, we could misunderstand 'cat' and take it to be an expression that must always be taken as referring only to brown cats. And this would be the case according to the initial rule. It's clearly a misreading of that rule. In order for that word to be taken as always referring to brown cats, a new rule must be introduced that makes it the case that 'cat' only refers to brown cats. This, of course, would be a strange and unhelpful rule for the understanding and use of the word 'cat'. Yet it's in the nature of Wittgenstein's linguistic rules that a rule could be formulated by a particular convention or community that would indeed make it the case that within this convention or community the word 'cat' must always be understood as referring exclusively to brown cats. There's no logical reason that 'cat' couldn't be used in these counter-intuitive ways – that is, by people who were brought up on the original (our) understanding and rule for the word 'cat'!

This must also mean that it can't be the case that there are absolute, necessary or un-ignorable meanings that will make it the case that 'cat' must always be taken as an expression that can only refer to cats as a whole, not only to brown cats. This certainly is engendered by Wittgenstein’s philosophical position on abstract meanings and what he takes to be their nature.

But meanings aren't rules.

So how can meaning-realists talk in terms of 'absolute', 'necessary' or 'un-ignorable' meanings which we must abide by? How do the rules-of-use come from abstract meanings when there's no way in which they can say that an expression must be understood in such-and-such a way and must also be used in such-and-such a way? Such normative demands or rules couldn’t come from meanings themselves because they never tell us what must be the case: certainly not in the case of understanding linguistic expressions.

Strangely enough, even the meaning-realist all along must have been understanding 'cat' in a certain way and using that word in a particular way. Even he would admit that, like the Wittgensteinian, he too must use 'cat' in a certain way and he must understand it in a certain way. Even if we accepted the existence, and even the nature as stated by realists, of abstract meanings, we would still need to ask the meaning-realist where, exactly, his normative musts actually come from. In addition, what about the detailed rules that determine (in precise ways) both how 'cat' must be understood and how that word must be used? For the meaning-realist there would still be detailed rules - perhaps not actually called 'rules' - or simple standards that would tell him how he must understand and use certain expressions. In other words, there would still be normative aspects to his understanding and usage of expressions.

It's not the case, therefore, that a meaning-realist denies that there are linguistic rules of some kind that specify and determine our understandings and uses of expressions. Indeed how could he? He does differ from the Wittgensteinian, however, in two ways.

Firstly, he doesn't take rules to be as essential and important as the Wittgensteinian takes them to be. Instead he takes abstract meanings to be essential and important.

Secondly, the actual nature of rules would be seen differently by the meaning-realist. Perhaps he wouldn't even use the word 'rule'.

For example, he wouldn't take rules, and therefore conventions, to be quite so autonomous, contingent, and even arbitrary as he takes the Wittgensteinian to believe them to be. He would take rules (or some other normative notion) as themselves being dependent upon and determined by abstract meanings. In that case he may think that his rules would be less contingent and less autonomous than Wittgensteinian rules. Perhaps he wouldn't take them to be contingent at all. 

Because of this possible meaning-realist's introduction of abstract meanings into this debate about linguistic rules, then many of the arguments he'd previously used about expressions and their meanings could now be applied to his position on rules. So just as abstract meanings both guaranteed and demanded universal, certain and necessary understandings and uses of expressions, now he may say similar things about his rules. He may now argue that rules themselves are universal, certain, necessary and determinate; just as the understandings and uses of expressions are of abstract meanings. Now he thinks that he can use and accept rules without - in the slightest respect -being a Wittgensteinian or a linguistic conventionalist.

We must again state what was stated earlier.

OK, I can accept the existence of abstract meanings. I can even accept the nature of meanings as you yourself have described them. For example, I can also provisionally accept that meanings are universal, certain, absolute and determinate. I can accept that such meanings also determine the linguistic rules to some extent. However, none of these provisional acceptances would - by themselves - allow me to use rules that have some kind of normative content - or even a lot of normative content. 

Again, abstract meanings can't tell me how expressions must be understood and how they must be used in specific ways. Even the universal, certain and determinate properties of meanings couldn't determine rules expressed by normative modal terms like 'must'. And if these essential and important meanings, as you see them, determine the understandings, rules and uses of expression, then they still couldn't supply us with the normative content required for linguistic rules. If I accept at all that meanings determine expressions, understandings, uses and rules, I could only accept - even only provisionally - that they determine all these things in some kind of limited way. It would be a limited determination precisely because abstract meanings alone (whatever they're taken to be) can't determine the normative nature of our understandings and uses of expressions. And therefore also the normative nature of the rules which tell us how to understand and use expressions. All the arguments I've used about this limited nature of meanings, still apply despite your descriptions of the nature of meanings.

Where do rules really come from? They come from conventions and communities, according to Wittgenstein. They must therefore come from persons (or groups of persons). Now we can say that rules can't possibly come from anything that's essentially non-human or non-social. They can't come from abstract meanings, as argued. And precisely because of their normative nature, it can also be said that rules can't possibly come from anything that's non-human or non-social. The notion of normativity is applicable only to persons and, therefore, also to communities, conventions, institutions, etc. of persons. The rules too must express normative notions and display what an anti-Wittgensteinian may call anthropocentric properties.

Following on from that, the very idea that rules can somehow come from - or be determined by - abstract meanings is surely wrong. Indeed because rules are normative and normative notions are only applicable to persons and their creations, then saying that rules somehow and in some way come from meanings is as close to being a Rylian "category mistake" as anything could be.

The meaning-realist's position on meanings and rules is a very good example of Wittgenstein's claim that so many philosophers have attempted the task of transcending - or moving beyond - natural languages and the human world. In the case of the meaning-realist, he does so by stressing the essential and important nature of abstract meanings when it comes to rules and therefore to the understandings and usages of expressions. Clearly abstract meanings are neither linguistic nor personal. In that sense, the meaning-realist’s meanings transcend language and the human world. Indeed he would probably think it important that they should transcend language and the human world as a whole because such things can only offer us contingency, uncertainty, conventionality, particularity and all the other vices of the natural languages and the social world. Abstract meanings, on the other hand, offer us the exact antitheses of all these properties. And they do so precisely because they're taken to transcend language and the social.

The meaning-realist would accept that Wittgenstein is partly right in what he says about transcending or moving beyond language and the social. However, the difference would be that whereas Wittgenstein believed that such acts of transcendence are always futile precisely because they couldn't in fact be achieved; so our meaning-realist believes that acts of transcendence beyond language and the social are not only possible: they can be (or are) highly beneficial to philosophy. The meaning-realist's acts of language transcendence wouldn't be futile; as Wittgenstein argued they are.

Further Reading

Hookway, C – (1988) Quine, Polity Press
Wittgenstein, L (1953/1958) Philosophical Investigations, Oxford

An Aspect of Susan Haak’s Foundherentism

It could be said that if a foundationalist believes that “basic beliefs are about observable physical objects” (e.g., “there’s a dog”), then he's not in fact a genuine foundationalist at all. If he believes that a dog is in fact a dog, then he has already brought beliefs to bear on the dog-experience to start with.

How does he know that it is in fact a dog? Because of his prior beliefs about dogs. Perhaps he tacitly uses the conditional: If an object or animal barks and has four legs and a tail, then that object must be a dog.


“There’s a dog” can't be a foundational belief in any sense of that term. The only thing he could say is that it is a belief caused by actual interactions with dogs, whereas some of his other dog-thoughts aren't caused by causal interaction with concrete dogs. In that case, perhaps we could say that the causal interaction with a dog is foundational; though not the belief itself. In that case, as Susan Haak says, it is

“only propositions, not events [or objects], that can stand in logical relations to other propositions [or beliefs]”. (227)

The causal interaction itself is neither a belief nor a proposition. Therefore it can't “stand in logical relations to other propositions” or beliefs. The causal interaction with the dog might have been (as it were) a first cause of the belief or beliefs about the dog; though in and of itself, it is neither evidence for such beliefs nor a justification for further beliefs.


In any case, the same causal context - taken only in itself - can cause different beliefs in different people and possibly different beliefs in the same person at different times. The interpretations of our causal contacts depend on our prior beliefs and the prior concepts which we apply to our causal interactions or contacts. And even if a particular causal contact brings about the formulation of new beliefs or new concepts, these will still be dependent upon or related to prior beliefs and prior concepts. Such things do not spring up ex nihilo and neither are they “first beginnings” or “first causes” (as it were).


Even the adverbial versions of foundationalism smuggle in prior concepts and beliefs.


For example, “I am appeared to brownly” still uses the concept [brown]. And even the concept [appears] implies that the speaker is making a distinction between appearance and reality. In addition, even if that statement isn't object-involving, it's still property-involving. And where there are properties, there are also objects to which the properties can be predicated. Unless the subject thought that brown itself, or brownness itself, qua trope, appeared to him “brownly”. That isn't likely.


If the speaker can identify the property- concept [brown], then why can’t he also identify an object-concept, say, [dog]? He might not have done so that this particular time, of course; though there was nothing to stop him doing so. And if something “looks like a dog”, it still looks like a dog even if it isn't an actual dog. The prior concept [dog] is still (surreptitiously or tacitly) used by the adverbial foundationalist.

Problems With Naturalised Epistemology: Reasons & Causes

Causal conditions in and of themselves can't give us justifications for a belief. They can, however, determine the nature of belief. As far as justification is concerned, we need to make the causal conditions justify our beliefs. That is, we need to say why or how such particular causal conditions have contributed to our true or false beliefs. 

As Donald Davidson said, "causation is not itself under an aspect". It doesn't explain or justify anything. It only does so in conjunction with the epistemic practices of epistemologists who make sense of causal conditions and who also offer us the reasons why particular causal conditions - rather than others - determine the truth of our beliefs rather than their falsehood.

Reliabilists and externalists say that causal mechanisms in the brain hook up with stimuli that produce belief. Though do they hook up with stimuli that cause true belief? What makes the stimulations, casual mechanisms or whatever else cause true belief? Indeed do we even know how they cause any kind of belief - true or false?

The brain scientist can see what’s going on in the brain; though he can’t see the relation between the brain and the mind that makes sense of what goes on in the brain and what the brain gives it. Causes are precisely that - causes. And reasons are reasons. Reasons are in the domain of mind and causes are in the domain of the brain and world. Of course reasons themselves may be dependent on causes; though this wouldn't mean that reasons are nothing over and above causes.

The brain scientist can never see why we believe that Tony Blair is a politician. He can't even see the belief that P. Tony Blair being a politician or not lying is outside the brain. However, the mind - though not the brain - can both register Tony Blair being a politician or not being a politician and make sense of whether he is or is not a politician.

The brain scientist can tell us what subserves such beliefs (or even what causal mechanisms lead to this belief). However, he can't tell us why or how we come to believe that Tony Blair is a politician. Indeed we would need to tell him what the belief is and where it came from in order for him to tell us which causal mechanisms and brain states subserve such beliefs and reasoning processes. Without such prior knowledge, the brain scientist would simply be mucking about in the brain not quite knowing what it is he's trying to find or explain.

Of course causal conditions in and of themselves can't tell us everything we need to know about knowledge and knowledge-acquisition. How can they? Such conditions need to be explained and interpreted. Not only that: they need to be questioned and criticised. These causal conditions don’t just enforce themselves on the mind of the investigator. And they certainly shouldn’t enforce themselves on the minds of epistemologists. If they did, then there would be no such thing as epistemology. And there would be no such thing as empirical investigation either.

To give a simple example.

Which causal conditions are we talking about? There is an indefinite number. And if we choose certain causal conditions, the epistemologist can then ask:

i) Why have you chosen to concentrate on these and not the many others?

ii) Why is this causal condition relevant and the one you ignored irrelevant?

iii) How do you know that these causal conditions give us knowledge and the many others don’t?

iv) How do you know that you're looking at them in the right way?

Epistemology is different to science or empirical investigation. However, does even the most naturalistic epistemologist think otherwise? The naturalist simply says that we must rely on - or even defer to - empirical investigation or simply use such findings. How could it be any other way? If this weren’t the case, then epistemology would not really have a subject matter to clarify and elaborate upon. The pure Cartesian epistemologist may survive without science. Though what sort of survival would it be? It's a kind of survival that's individualistic or subjectivist. That it, the Cartesian epistemologist relies on himself and himself alone, at least in principle. So what’s to stop him making massive mistakes and barking on many false trees? 

It's the Wittgenstein argument: if there’s no one available to tell him that he has gone wrong, then how does he no that he has gone right? The world is bigger than his own mind, no matter how great and systematic his mind is. 

The Causal Relations Between Mind & World

In the Cartesian tradition, the mind’s autonomy was of paramount importance. All we need to do, essentially, is get our internal workings functioning and in good order. It doesn’t really matter about the external world. Or, more correctly, it doesn’t matter until we had got our mental ship in order.

In epistemological externalism (as well as in reliabilism), on the contrary, what happens before the formation of a belief (or before cognitive operations as a whole) is what matters. The question is simple: 

How did I acquire this particular belief? 

So it's the required causal relation with the external world that determines whether or not a belief (or cognitive process) is justified. If it's acceptably justified, then we may have knowledge. We may not even be able to elaborate on the external causal processes that led to the formation of a belief. Though if we're in the right situation (vis-à-vis the world), then such processes which we may not be aware of will themselves somehow justify our beliefs.

Of course it's the case that we require causal relations with the external world when it comes to our perceptual knowledge. That is, in a purely empiricist manner, all we really need are reliable causal contacts with the world. Though, as Kant might have said: 

There's much more to knowledge than mere causal interaction with the world. 

Our minds and brains need to do something with all the incoming data – they need to synthesise it. That is the case even with basic perceptual beliefs.

Can we rely on perceptual information alone to make general statements about the world? That is, no single experience (or even a large group of experiences) alone will tell me that, say, that all swans are white.

And do we ever have a basic perceptual experience of someone not murdering someone else?

What about microscopic properties, objects or processes which are always beyond perceptual experience? What does pure experience tell us about these things?

What about numbers? They're supposed to be non-spatiotemporal abstract entities. How could I ever have perceptual experience of - or causal interaction with - numbers?

So not all beliefs and bits of knowledge are simply a question of reliable causal contacts with the concrete things that the beliefs and bits of knowledge are about. The traditional empiricist had no convincing answers to these epistemic problems. There are things about the mind-brain that work free of causal interactions. Or if there are causal processes in the mind-brain, then they're of such a kind that we don’t even know that they're happening. In that case, they have no epistemic relevance.

The internal world must have at least a little autonomy from the external world. Though perhaps we don’t necessarily need to think of it at all as the internal world. The mind-brain is indeed a part of the world. It is naturalistically acceptable. However, the mind-brain does things that other parts of the world don't do. It thinks about the world. It has representations and images of that world. It has intentionality – that is, directedness. It deals with meanings, concepts, truth, falsity and the rest.

Although the mind is part of the world, it's a special and unique part of it. 

We causally interact with the world, and that in turn sets of causal processes in the mind-brain: it doesn’t follow that there'll be some kind of isomorphic and infinitely repeatable set of mind-brain relations to that world. The same causal processes bring about different beliefs in different people - or even the same person at different times. Something goes on in mind-brains that can't be accounted for in the same way that things that go in the world can be accounted for. This is not a mystical conclusion. It may simply be a fact about the mind-brain’s astonishing complexity and subtlety. Indeed it has been said that the mind-brain is the most complex thing in the entire universe.

Putnam on Semantic Holism

Hilary Putnam (in this work) isn't talking about what words mean when taken separately from what he calls ‘interpretation’. He isn't even talking about the inferential or otherwise connections of words and their meanings to one another when taken separately from interpretation.

If Putnam is talking about interpretation, he's talking about how individual hearers - or even the sum of hearers or understanders - interpret the meanings of the words or sentences which they hear or read. On this account, then, a word doesn't have a meaning in glorious separation from all acts of interpretation or understanding; which is perhaps the position of Frege and so many other philosophers (perhaps Dummett too). Our interpretation of a word or a sentence will change "when we see more text" (229). Perhaps we couldn't even interpret a word or sentence unless we saw more text or heard more of what the speaker has to say. How could we interpret at all without something to help us with our interpretation? That something else, in Putnam’s words, would be ‘more text’. (It's interesting to see Putnam using Jacques Derrida's term ‘text’ rather than ‘utterance’, ‘word’ or ‘sentence’.)

Not only is the context of the text of importance when it comes to interpretation: we couldn't know what a word means unless we see that word used again and again in other contexts. Perhaps not only by the same speaker or in the same text; but also when used by other speakers and in other texts. After all, if the context of the word is the text in which it is embedded, then perhaps that text also has its context in a sum of other texts (some by the same author and some by other authors). Even if we gain access to the word or sentence’s meaning by hearing more of what the speaker says (or more of what is written in the text in which it is embedded), that greater knowledge will still be ‘finite’ and will thus it won't "infallibly show us what the word means" (229). There's also the possibility (or the likelihood) that the word will have been used "in some additional way that you haven’t taken account of" (229). In fact this is bound to be the case simply because our minds are finite in nature.

However, it's indeed strange that Putnam appears to be suggesting that one is required to know of every usage of a word before one can fully or accurately know that word’s meaning. Surely that can’t be right. Or, instead, we may well know its meaning; though that meaning is neither static nor determinate in that our acquiring new knowledge of other interpretations or usages of the word will have an effect on how we understand it. We don't need to know about every usage or utterance of a word in order to understand what it means. However, when we do acquire such additional knowledge, this will have an effect on how we understand the word. This, again, means that the word’s meaning is neither static in nature nor determinate. This doesn't stop us from using the word accurately or easily; nor does it stop us from understanding its meaning. It only stops us from believing that its meaning is static and determinate. Perhaps we simply don't require a word’s meaning to be static or determinate in order to communicate with ourselves and with others.

Think here of Wittgenstein’s ‘family resemblance’ argument in which he argues that different usages of the word ‘game’ don't all have an essence (as it were). Instead, each game has a family resemblance with each other game without that family resemblance needing to be grounded in an essence of games which must be shared by all games. There's no necessary and sufficient set of conditions that all games must share in order to be games or to be called a ‘game’. However, they may well share something with other games; though not with all other games. Similarly, all the uses of the word ‘game’ (or ‘liberty’ or ‘truth’ for that matter) must share something with at least some other usages of that name. Though it needn't share a determinate or static meaning with all these other usages. As long as it shares something with them - no matter how small.

This is why languages aren't static or determinate; unless they're artificial or Fregean languages! Much of what has just been said also implies that what matters is what we do with words and sentences (how we use them) – not necessarily (or only) what they mean. If that is indeed the case, then we can hardly expect a word to have the same meaning in all contexts or when used at different times and for different purposes.

This is the basic Wittgensteinian insight into words (or their meanings) and it takes us away from the Fregean position in which the sense behind words (or the Thoughts or propositions behind sentences) are both determinate and static. Perhaps only the meanings (if there are meanings at all) of the logical constants and other (logical) primitives are genuinely determinate or static. Though even here we can only define the logical constants in terms of what we can do with them (in the ‘implicit definitions’) and not in terms of their abstract meanings.

Putnam gives his own examples of this lack of determinacy or stasis when it comes to meanings. He says:

"… in American usage an armchair is a chair, but it’s not a chaise in French, it’s a fauteuil and it’s not a Stuhl in German, it’s a Sessel."

Perhaps this is more a question of linguistics or even lexicography than it is a question of semantics. However, Putnam began his career as a linguist so perhaps this connection isn't simply fortuitous. Perhaps the ‘meaning is use’ thesis entails a parallel commitment to the findings of linguistics or even to what the lexicographers say! Putnam himself says:

"One of my three majors in college was in linguistic analysis, it was the first department in the world. … it was a section of the anthropology department…" (229)

Here we don't only have a semantic holism that must incorporate linguistics: perhaps anthropology must be taken into account as well. That is, if we take our holism so far, perhaps we should take it even further into anthropology (as Wittgenstein is said to have done). And then perhaps into culture and history as a whole, as Rorty, Derrida and others have done. We can say, as the enemies of holism often do, that once we commit ourselves to holism (of whatever kind), then one's holism can't help but spread its wings farther and farther until, perhaps, we reach something like the Absolute of the 19th century Idealists. Well, that’s a thought at least.

More specifically, Putnam says that if one studies linguistics and/or anthropology, then "meaning holism is just forced on you" (229). In order to interpret language or words (or utterances) within a language, whether alien or our own, then one must interpret holistically or one can't interpret at all.

What's all this holism opposed to? According to Putnam:

"There are some accounts of meaning such as Fodor’s, according to which each word has one meaning which is fixed by its causal connection with a 'property', but that has nothing to do with the way words behave in a real language." (229)

Perhaps this is because Fodor is a Fregean at heart. Perhaps he too is committed to atomism simply because he believes, like Frege and Dummett, that atomism is the only way we can secure determinacy, the stasis of meaning, as well as its objectivity. That was, after all, precisely what Frege wanted from his semantics and thus from his ‘ideal’ (though artificial) language.

Though, as many philosophers have often said, this Fregean project essentially failed in its objectives; at least as far as the natural languages are concerned. Perhaps it didn't fail when it came to certain artificial languages (as Tarski would have acknowledged). But as Strawson and others have said, such artificial languages are woefully inadequate when it comes to the systematising of natural languages and also when it comes to everything that can be said or done within natural languages. What we are left with is a "mere skeleton of a language" and not a formalisation (or otherwise) of the language/s we use everyday of our lives.

Again, perhaps both Dummett and Fodor have simply not accepted the lessons given to us by the late Wittgenstein when he stressed the ‘meaning is use’ thesis, language-games and the anthropological realities of the natural languages. Indeed Dummett, for one, has often stressed his antipathy towards the late Wittgenstein; especially when it came to the ‘sceptical’ results of the latter’s theories about meaning.

Popper on Causal-Logical Necessity

 
It's often said that causal connections aren't logically necessary – not even necessary causal connections. This is the central gist of Hume’s position and from which he derived so many of his arguments about causation as a whole. Popper, on the other, did think that causal connections are logically necessary... but not so quick! They're only logically necessary

‘in the sense that they follow deductively once we assert the appropriate natural law’. (11)

If we don't assert the appropriate natural law, or any natural law, then we can't say that they are logically necessary. This is a classic case of the principle that no matter if a logical axiom or premise is true (that is, even if it is false), what follows from it will still follow deductively and validly from it (providing one’s inferences are valid).

Did Popper only mean ‘logically necessary’ in this amended sense? Could he have used the words ‘logically necessary’ in any other way? That is, any logical necessity that causal connections have are simply inherited from the natural laws from which they are derived or deductively inferred. We can't expect anything more about causal logical necessity than this.

Now we can express an example of a scientific law that is expressed in causal terms:

An event of type C has occurred. But whenever an event of type C occurs, an event of type E later occurs.

Again, this will only happen of necessity if we also assert the appropriate natural law relating cause C to effect E. Without that natural law (or its assertion/assumption), an event of type C could be followed by an event of type F, or by anything for that matter!

However, the Humean can still express his problems with Popper’s position. Dale Jacquette writes:

"A defender of Hume on the contingency of causal connections might nevertheless object that although the inference is deductively valid, and to that extent carries necessity from assumptions to conclusion, the conclusion itself is not necessary unless the assumptions are also logically necessary, and that no scientific laws correlating causes to effects are logically necessary." (11)

This means that it doesn't matter if the move from the assumptions to the conclusion is deductively valid if the assumptions themselves aren't logically necessary (they may not even be true to bring about deductively validity). After all, causal matters are about the world. They have nothing to do with deductively validity or deductive inference.

A Humean would argue that although the inference from the assumptions to the conclusion is logically necessary (in that if the assumptions are taken to be true this truth is passed on to the conclusion), the assumptions themselves aren't logically necessary.

Does this mean that the passage from event type C to event type E isn't logically necessary? Or that event type C, taken by itself, isn't logically necessary?

If event type C isn't logically necessary, then how can it pass on logical necessity to event type E?

So now we can say that not even the move from event type C to event type E is logically necessary if event type C isn't itself logically necessary.

Again, we aren't talking about formal or subject-less deductively validity or inference here, but an aspect of the world – causation!

To put the conclusion at its most basic. Hume argued that no scientific laws correlating causes to effects are logically necessary. That is the gist of Hume’s argument. Not that there is no such thing as causation or causal connection. Not even that there are no such things as causal regularities. Of course there are! No. His point is simple. There are no scientific laws correlating causes to effects. A scientific law is required to be both universal and exceptionless. On Hume’s empiricist grounds, we have no way of observing the universal or truly knowing that something is indeed exceptionless. It follows that no causal relation - say between event type C and event type E - can be deemed to be universal and exceptionless. Thus it can't instantiate or fall under a genuine natural scientific law. That is Hume’s point against scientific and rationalist views of necessary causal relations.

Jacquette puts much of the above in the following way:

"Again, it might be questioned whether the logical necessity obtaining between the assumptions and conclusion of a valid inference about real world events necessarily qualifies or attaches to the events themselves… " (11)

What we have here are two things:

i) The logical necessity obtaining between the assumptions and conclusion of a valid inference about real world events.

and

ii) The logical necessity obtaining between events in the real world or between event type C and event type E.

Clearly they aren't the same thing.

We can even say that i) is analogous to a de dicto modality, whereas ii) is analogous to a de re modality. One kind of necessity is about the form of an inference from assumptions to a conclusion. The other kind of necessity is said to obtain between events in the actual world. There is a world of difference between the two. The latter would be metaphysical or ontological necessity, whereas the former would be strict logical necessity. It appears, then, that Popper attempted to fuse logical necessity with metaphysical necessity.