Derrida on Words and Meanings

The English analytic philosopher Hugh Mellor once said that Derrida was “wilfully obscure”, “trivial”, a “mystery monger”, “absurd” and “dangerous” (1993/1999). But why stop there? Is Derrida pretentious?  Yes.  Is Derrida difficult to read and understand?  Yes.  Is Derrida a modern-day sophist?  Yes and no.  Is Derrida a game-player?  Yes.

Yes, Jacques Derrida (1930-) was all the above.  He was also an excellent philosopher.  A genius.  An original. He was a man of contradictions.

.

The Unconceptualised Carburettor: Signified and Signifier

Here's one of Derrida’s most famous statements:

There is nothing outside the text.

This is a typical poetical locution by Derrida.  Obviously, I hope, it can’t or won’t be taken literally. (Do philosophers take the many gnomic expressions of Wittgenstein literally?) It is not an extreme statement, as some people say, it is more or less meaningless if taken literally, that is, non-poetically. Certain brands of scepticism or nihilism are extreme, but not meaningless. Think about what Derrida's statement could mean if taken literally! And yet it has been taken literally.

So how should Derrida’s sentence be taken? I think that he means that everything we say about objects, events, etc. betrays our conceptual or theoretical baggage. That is, we don’t speak the world’s own language.  The world doesn’t have a language, not even the “language of mathematics”. Derrida would have said that our comments about the world, its objects, events, etc., would be “intertextual”.  That is, what we say about the world is based upon, or a consequence of, what “texts” we have read - or “texts” other people have read - have said about the world.  (The word “texts” should be taken very broadly.  It includes books, concepts, also in books, theories, conceptual schemes, etc.)

John Searle, on the other hand, says

…suppose I have reached a deconstructionist car mechanic and he tries to explain to me that a carburettor is just text anyway, and that there is nothing to talk about except the textuality of the texts…  (From Rationality and Reason)

Although some philosophers say that ad hominem arguments should be banished from philosophy, I have to say that I simply can’t believe that Searle really believes that Derrida thought that “a carburettor is just text”. If Searle's view is largely representative of that of other analytic philosophers' views on Derrida, then it's no wonder that thy think he's a fraudster, a charlatan and all the rest. On Searle's interpretation, Derridean philosophy  would be in the same league as solipsism or some of those crazy religions one finds in the United States and, yes, extreme forms of idealism too.

Of course the carburettor isn’t “just text”.  It is a physical thing.  It isn’t made of mental items or written words.  This much Derrida would have accepted.  Searle’s very fictional car mechanic need not be a  “linguistic idealist” to say, or mean, what he said.  But he would say that everything he says about the carburettor is indeed “textually”, and therefore conceptually, mediated. The carburettor hasn't told us what to say about it. It doesn't simply have a single description:

C¹

c ® C

 
C²

The capital Cs above are the descriptions of the carburettor, and the small c is the carburettor. C, C¹ and C² are determined (in some way) by c. That is, they depend on c. But c needs to become C or C¹, etc. in order to be cognised. c doesn't change its physical reality because of C¹ or C², but it is known through  them. c as c, the "thing-in-itself", can never be known because when it is known it becomes C or C¹, etc. There is a causal determination between c and C and C¹, etc., but not a descriptive, explanatory or conceptual one.

This argument can be interpreted very simply.  Say a person has never come across a carburettor before although he speaks English as well as the car mechanic and is a rational human being.  Would he say the same things about the carburettor as the car mechanic?  Of course not.  But they would both be looking at the same thing (although N. Russell Hanson may not accept this way of putting it either – see his 1969)  The same thing would generate very different descriptions. Carburettors are human artefacts.  So the locution “carburettors as they are in themselves”, or as they are free from minds, doesn’t make much sense.  But the same is true of, say, trees. That’s not to say that the carburettor (or a tree) depends in any way on minds for its existence.  Of course it doesn’t.  However, the descriptions and the concept [carburettor] do depend on minds. So what would Derrida, or the car mechanic, say about Searle’s carburettor?  He would ask Searle: What is the relation between the concept [carburettor], or the word “carburettor”, and the physical thing we have named a “carburettor” and conceptualised about?  Derrida would argue that the concept [carburettor] couldn't

transgress the text towards a signifier outside the text whose context could take place, could have taken place outside language…There is nothing outside the text.

That is, the concept [carburettor] and the word "carburettor" are relational phenomena, they are, to use Fodor's terms, "transportable parts" () and parts of a "syntax and a system" (Derrida). That is, they bear as close a relation to other parts of a language as they do to parts of the world (perhaps parts of the world near the carburettor). That's what Derrida meant by saying that the word "carburettor" or the concept [carburettor] can't take "place outside of language". Any theory of reference would need to include other words and concepts, not just, say, a direct causal relation between word and thing named. (Perhaps if Derrida has platonised the term “text” and turned it into Text with a capital T, we would understand how broad this term is.  That is, “texts” are just as much in minds as they are on the page.)

If words and concepts can't be non-relational, or taken out of language, then neither can the things spoken of or written about. If we say such things about words and concepts, we should, almost by transitivity, say such things about things, events, etc., too. Here, though, I am talking about cognised not mind-independent things.

What is a tree or carburettor “outside of language”?  What can we say about even a tree that is not based on what we have read or other kinds of “texts”?  There may well be a “thing-in-itself”, a physical thing, which we have called in English a “tree”.  What can we say or think about a tree “outside language”?  We could use a Kantian term and call it a “noumenon”.  When we do try to say something about the tree “in-itself”, or the mind-independent tree, we find ourselves using concept-relative names, ideas and theories that clearly aren’t “outside language”.  Berkeley made similar points about the tree.  However, being an immaterialist, he claimed that the tree didn’t in fact exist without minds.  He was, therefore, far more extreme and even outré than Derrida.

Donald Davidson made a similar point about the world and therefore the tree:

[let’s give up]…on the concept of an uninterpreted reality, something outside all schemes and science.

Like Derrida, Davidson didn’t believe that there isn't an “uninterpreted reality”, or text-free world of concrete objects that we call “trees” (Berkeley did believe this), but he did believe that such a thing is, well, useless.  Our “uninterpreted reality”, which metaphysical realists love so much, is like Wittgenstein’s wheel that is not part of the mechanism. The uninterpreted tree plays no role. It is epistemically vacuous. A time waster. Indeed, an uninterpreted tree is not, well, a tree at all. The tree, or anything else, "outside all schemes and science" is a non-thing because nothing can be said about it. When we do say something, it will be said inside schemes and science, therefore not longer functional as a philosophical strategy for the metaphysical realist. Saying that the tree can have some kind of reality or existence that is fixed and determinate regardless of minds doesn't help us much if when we get to know the tree we don't know if we are faithfully mirroring its fixed and determinate reality or existence. We don't know if we are cognitively or sensorily in tune with reality. Indeed, perhaps we haven't even got a firm grip on the very concept of correctly mirroring the tree. The only true mirror or representation of the tree would be the tree itself, or at least its exact replica. The problem is that this exact replica wouldn't tell us anything about the tree. The replica wouldn't be either a description or an explanation. We would be in the ridiculous situation of having to mirror the replica that possibly did in fact faithfully mirror the original tree. But our mirroring of the replica would not be like the replica's mirroring of the original tree. In fact, they would bear virtually no resemblance to each other. The replica would be 20ft tall and made of wood, bark, leaves, atoms, molecules, etc. Our representation would be a question of concepts, words, mental items, metal images and thought processes. Our representation and the physical replication are not members of the same natural kind, as it were. Ironically, the replica would be by definition precise but useless, whereas our representations of the tree would possibly be imprecise and possibly useful. (This is partly why models are of more use than replicas in science.)

If the “signified” (e.g. the unconceptualised thing we call a “tree” or the meaning/concept of the word "tree") drops out of the picture, then why not just talk about “signifiers” (i.e., words, concepts, theories etc. which refer to the things we call “trees”)?  Why not just talk about “texts” and deny the fact that there are true books (texts about the world rather than about other texts)? According to Derrida,  the “signifier” (i.e. the word “tree”) whose “signified” (the thing we call “tree” or  the meaning of "tree") would have no role to play “outside language”.  So it seems to follow that the “signified”  couldn’t “take place…outside of language” either.  This is not idealism, linguistic or otherwise.  Berkeley, as I said, denied that there was a mind-free world, so too, for one, did Fichte.  I don’t think that Derrida did.  I don’t think that he needed to.  He simply thought that the world is conceptually or textually mediated.

This leads, by a few inferential links, to the concept of “intertextuality”.

The Snare of Intertextuality

Julia Kristeva, in her book Semiotike (1969) wrote

every text takes shape as a mosaic of citations…every text…[transforms] other texts…

Since Kristeva’s book, this term, “intertextuality”, has become very common in continental and continentally-inspired thought. I think that has been misused (or at least overused) too, as one would expect. Without the post-structuralist  technical terms, I will attempt to show both what this concept means and its relevance.

Let’s apply the concept of “intertextuality” to philosophy itself.  What has been said (in the last section) about the relationship between language and the carburettor (or a tree or the world) can now be seen to be applicable to the relationship of philosophy itself to the world.  Or, in fact, the relationship between philosophy and… philosophy.

Francis Suarez’s Disputationes Metaphyiscae (1597) was, according to a writer who had probably never heard the term “intertextuality”, the “first systematic and comprehensive work of metaphysics written in the west that was not a commentary on Aristotle’s Metaphysics”.  And even in the 19^th century, for example, we could say that virtually all of Franz Brentano’s works tackle Aristotelian issues. 

Even though Bacon rebelled against Aristotelianism, or at least Aristotle himself, he essentially defined himself by what he was against (that is, Aristotle). It would be interesting to see how non-Aristotelian Suarez's work actually is. And even if it wasn't a "commentary on Aristotle's Metaphysics", this may not amount to much of a claim. It may still have been largely Aristotelian. Or, alternatively, Suarez, like Bacon (and around the same time) may have defined himself by what he was against and therefore have still been, in a sense, an Aristotelian in the way that Marx was essentially still a Hegelian until the end (that is, not only when he was an official Left Hegelian). And perhaps Heidegger, who also rebelled against Aristotle, was still in some ways an Aristotelian. After all, when we invert a tune or a shape (as in Cubism or 12-tone serialism), the inversion still bears a strong relation to what it inverted. Similarly with opposites. We could be in the strange position where an anti-Aristotelian or anti-Hegelian owes more to Aristotle or Hegel than an Aristotelian or a Hegelian. (Think also of Nietzsche’s reactions to “slave morality”.) Derrida, an anti-western metaphysics metaphysician, certainly read and studied more western metaphysics than many western metaphysicians and certainly more than the simply indifferent.

Let's get back to "intertextuality".

How many Berkeley scholars or students look at a tree in the same way in which Berkeley himself looked at a tree to consequently come to his immaterialist conclusions (let’s assume this happened)?  Do scholars and students only read Berkeley’s “texts” instead (in which he describes the time in which he looked at a tree and consequently came to his immaterialist conclusions)?  Or perhaps many students or scholars of Berkeley read a book that describes another book that describes another book…in which there is a description of Berkeley looking at a tree...I have found explicit examples of this phenomenon. Take this one:

A new post-graduate student on

Ü

Dummett on

Ü

Kripke on

Ü

Wittgenstein on

Ü

a private language.

And this would be to assume that Wittgenstein’s ratiocinations on the subject of a private language were untouched by other "texts". (Perhaps I’ve chosen a bad example because Wittgenstein was far from being an avid reader of other philosophers’ texts.). So there's a long regress here, but which is not, obviously, an infinite one. That is, if we see the above as a foundationalist edifice, Wittgenstein's work on a private language couldn't have been the equivalent to a "basic belief" - that is, not part of a longer chain. So as with Wittgenstein, all this is not to assume that Berkeley himself was free from the taint of intertextuality.  Even if he did indeed step out of the study to look at a tree, rather than read a book about a tree or a book about another person looking at a tree (to consequently reach his own immaterialist conclusions), the mind that looked at the tree, the mind that did the hard work, would have still been contaminated by “texts”.  For example, the “texts” of John Locke in which he makes the distinction between “secondary” and “primary qualities”. These “texts” too would have been influenced by, say, Robert Boyle’s texts or the texts of Descartes.

There had to be a beginning to this "inter-textual" regress. However, the beginning wouldn't have been necessarily another philosopher looking at a tree, or another philosopher thinking about the notion of a private language. But they may be traced back to philosophical problems or ideas that gave birth to Berkelian immaterialism or questions about the possibility of a private language. Derrida would have argued that it would simply have been example of "sign substitutions" around a given "centre". That is, the technical terms change and the subjects slightly shift their rotational patterns, but the axial point would still be in place. I don't think that this conclusion is in fact correct, although it may well be correct in many instances. Take questions about essence and essentialism. They go back thousands of years and made a comeback in the 1960s and 1970s. However, there must be some changes in philosophy that aren't just examples of  "sign substitutions". Take the areas of philosophy that have been influenced by breakthroughs in science. Perhaps they don't have - much of - a philosophical heritage. Take eliminative materialism's denial of the propositional attitudes or Quine's willingness (because of quantum physics) to question the law of excluded middle. These two examples can't really be traced back beyond the twentieth century, which is not to say that there aren't any near-equivalents in western philosophy.

We must ask a question: How original can a philosophy be? If it were utterly original (as with music, etc.), with no inferential links to anything that went before, then perhaps it wouldn't be philosophy - as we know it - at all. Is this why Derrida's work is not deemed to be philosophy by his detractors? Eliminative materialists and Quinians need to share some kind of a philosophical language with their contemporaries, otherwise their chosen language would be chaotic or even inscrutable. And, yes, so too did Derrida. Perhaps in another possible world there could be another possible philosophy that is totally alien to our own. But then we could ask: Why is it philosophy if it shares nothing with our philosophies - that is, if it is truly alien? So if Derrida was saying that there is no truly alien philosophy in this world, he was correct. His philosophy was not truly alien either.

Schopenhauer, not Derrida, asked: “When do we ever step outside books?”  He was also very critical of the parasitical nature, as saw it,  of so much thought and philosophy (i.e. the reliance on other philosophers' “texts”).  Perhaps, however, Schopenhauer’s views are aimed at the psychologies of other philosophers.  It may not have been a philosophical point about intertextuality.  If Schopenhauer had spoken in contemporary terms, he may have said that the “intertextual” web traps many of us in a way that is quite frightening (think here also of what was said about Wittgenstein earlier).  Derrida would have said, in response, that Schopenhauer was fooling himself if he really thought that he had escaped the web or snare of intertextuality.  We are all trapped.  As the car mechanic said to Searle, there is no “outside of language” (metaphorically speaking!).

What would a philosophical a priori, as it were, be like? A philosophy untouched by other philosophies (that is, other "texts")? Take Bryan Magee's account of his ex nihilo philosophising:

Until I went to university it never entered my head to associate any of these [philosophical] questions with the word 'philosophy'…I discovered that this is what they were…I had grown up a natural Kantian…I discovered…that I had been immersed in philosophical problems all my life. (1997)

What a strange and interesting passage. Magee was not, I don't think, claiming to be "outside language", but he was claiming to have been outside philosophy. He almost appears to be claiming that we have a kind of quasi-Chomskian philosophy faculty that we are born with. If he wasn't claiming that about a universal philosophy or philosophising faculty, then he was making a claim about himself and himself alone. And that claim must have been that he was somehow genetically programmed to philosophise in the particular manner in which he did in fact philosophise. Either that, or there was some kind of mystical or spiritual explanation for Magee's childhood philosophical bent. If the first option is taken, the quasi-Chomskian philosophy faculty, then all young children throughout the world would be asking the same questions that Magee had asked when he was a young child. Well, many children do indeed ask certain philosophical questions. But what questions and problems was Magee talking about? As he put it, questions and problems that he later claimed to realise were Kantian, Schopenhauerian, Leibnitzian and Wittgensteinian and so on. If this were the case, why weren't Kantian and Leibnitzian, never mind Wittgensteinian, problems raised two thousand years before the birth of these particular philosophers? If these questions and problems are so natural (he claimed to be a "natural Kantian"), why are they certainly not raised by other cultures in our own time unless they come into contact with western philosophy? There may well be certain philosophical givens, but they certainly aren't Wittgensteinian, etc. And any givens uncovered by empirical research tend to be more theological, mystical or spiritual in aspect, rather than strictly speaking philosophical. It is of course possible that Magee was an incredible genius who not only came to Kantian questions and problems without the help of Kant, but also to Leibnitzian and Wittgensteinian problems and questions too without outside help! (He did, however, rather modestly claim that he didn't find "solutions for them".) I suppose that in the end it will be an empirical study that will determine whether Kantian, Leibnitzian, etc. problems and questions are really part of the philosophical a priori. From my own empirical knowledge and philosophical reflections, I suspect that they aren't.

So where did Kant's Kantian problems and questions come from? They came largely from other philosophers. Where did Leibniz’s Leibnitzian problems come from? Ditto. Kant wouldn't have been a Kantian, unlike Magee, without the problem of the impasse between Rationalism and Empiricism and the scepticism of Hume (amongst many other things). And Schopenhauer wouldn't have been a Schopenhauerian, again, unlike Magee, without Kant coming before him and the work of the German Idealists (amongst other things). Perhaps Magee simply felt inclined, when he got older, to squeeze his own childhood questions and problems into, say, a Kantian hole.

Schopenhauer saw himself in the way in which Magee saw himself. He saw himself as a kind of aprioristic philosopher. He didn't only take a position on the a priori within philosophy, but an a priori position towards philosophy itself. That is, he thought that the best way to do philosophy is not to read – too many? - philosophy texts. And yet in his early life he had confessed to being more or less obsessed with Kant. Schopenhauer simply took an independent position on philosophy after the fact. He was like a car driver with an extra large petrol tank who fills it up to the brim and then claims to his fellow car drivers that his car doesn't need petrol (why else would it last so long without a refill?). Of course, Schopenhauer may have effectively lived off his memories of other philosophers' texts. (Think of Mozart who memorised an entire opera in one sitting and remembered it in full for the rest of his life.)

To ram this point home.

When a student or philosopher of mind thinks about the nature of mind (or the nature of his mind), perhaps all he actually does is read and think about what, say, Jerry Fodor and Dan Dennett have said on the nature of mind.  That is, he may be caught in Fodor's or Dennett’s intertextual trap (although it’s unlikely that any philosopher of mind would rely on just two philosophers of mind).  So perhaps all his responses, reactions and commentaries on the nature of mind (or on everything else for that matter), are simply intertextual.  They are not a comment on the mind – his mind – as it is in itself.  (When we study philosophy at university, it seems that thinking about and reading texts about minds seems far more important than thinking about the nature and working’s of one’s own mind…but, again, this is a very aprioristic position!)

We would like to flatter ourselves with the view that our philosophical views occurred ex nihilo, in the way some people do about their "strong wills" or "innate sense of right". But true ex nihilo philosophical thoughts are as unlikely as ex nihilo mental volitions. As I said, there may be some givens, but whether or not they are truly philosophical is open to debate. They certainly aren't Kantian or Wittgensteinian givens, or Fodorean givens for that matter. I don't know about Derrida, but I am a poly-determinist, and I take a determinist position as much on the contents of mind as I do on anything else, therefore ex nihilo or a priori Kantianism, Wittgensteinianism, etc. seems out of the question. It is as unlikely as an ex nihilo position or idea that is later seen to square with, say, quantum mechanics or string theory (though some writers have tried their hardest to find parallels and equivalents of 20^th century science in ancient mythologies etc.).

It could be said, by Derrida, that there is no other way for things to be.  Where would the novice aprioristic philosopher, for example, get his concepts and tools from?  He wouldn’t even have the vocabulary to philosophise.  Perhaps he wouldn’t even feel the need to ask such questions in the first place without the spur of prior philosophy.  He may not think or speak Greek (see later section).

The Myth of the Museum: Quine and Derrida on Meaning

It will be a sufficient condition for James's having the concept of so-and-so that he should have mastered the intelligent use (including the use in made-up sentences) of a word for so-and-so in some language. (Geach, Mental Acts)

I am in interested in the nature of our thoughts, not with their expression in natural language. (Joseph Levine, note 14. of ‘On Leaving Out What It’s Like’)

There is the familiar experience of not being able to find the words to express one’s ideas…Attributions of intentions and beliefs to dogs smack of anthropomorphism…thinking is really ‘talking to oneself’ – silent speech. (Davidson, from ‘Thought and Talk’)

The gist of Derrida's philosophy is the thesis that there is no meaning beyond the words - beyond language. Language is the beginning and the end of meaning:

It [writing] is also to be incapable of making meaning absolutely precede writing: it is thus to lose meaning while simultaneously elevating inscription.

Derrida rejected the dualism of word and meaning. They are not truly separate. Meanings are thoroughly linguistic. We can accept that a certain proposition lies behind its different sentential expressions.  But does it follow that the proposition is separate from and prior to all sentential expressions because it is not specific to one of them? The same, perhaps, is true of statements and their truth-conditions. We can, as with propositions, accept the distinction. Of course truth-conditions aren't sentences. However, what would truth-conditions be without their sentential expressions? It is minds, through sentences, that individuate and specify truth-conditions. We do not find or observe truth-conditions (in the manner in which Davidson said that we do not experience sensory data) - we stipulate or make them. And the same is true of meanings, on whatever theory of meaning we adopt. Meanings are linguistically individuated. Again, this is not to say that meanings are sentences or words. However, they can't exist without sentences or words and therefore minds. This isn't "linguistic idealism" in that there is no reason, at first, to reject propositions, truth-conditions and meanings of whatever kind. They may exist, but only in linguistic/mental clothing.  However, the "clothing" metaphor is not precise enough because people can, after all, exist without clothing.  Meanings, on the other hand, can't exist without linguistic clothing. The things that may constitute or make-up the meaning of S may well do so, but such things aren't the meaning. The constituents are necessary but not sufficient for meaning.

Despite all this, I don't see why Derrida's "inscriptions" should be, to use Derrida’s word, "elevated". Inscriptions are no more or less important than (possibly) mind-independent phenomena. Why go from the metaphysical realist's emphasis on the mind-independent world to the glorification of writing? Isn't this too a "binary opposition" (Derrida's technical term)? If it is, why renounce Tweedle Dee only to opt for Tweedle Dum? There is a causal relation between the world and our "inscriptions". As mitigated empiricists, we should say that sentences, words, inscriptions, languages, etc., would be nothing without the ever-so-concrete world. I suppose the elevation of inscription does indeed come close to the dreaded "linguistic idealism", but only because the Derridean indulges in the "dualism of scheme and content" (Davidson) by emphasising the "scheme" and - almost - forgetting the "content". However, Derrida has said that such an inversion of writing and meaning would be wrong: it would simply constitute another “binary opposition”. Whether or not Derrida stuck to his pronouncement on such inversions is, of course, another matter. (One suspects that he liked to hint at his “linguistic idealism”.)

It is the mistake of concluding that because of the fact that not all the constituents of meanings, propositions, truth-conditions, etc., are mind-dependent, that meanings, etc. themselves aren't mind-dependent (or dependent on particular minds). We are hooked by causal forces to the mind-independent world. However, causal forces aren't themselves meanings (or facts, truth-conditions, etc.).

So, as Derrida put it, meaning doesn't "precede writing" (). However, many Derrideans mistakenly "elevates inscription" (as I’ve argued). If meanings preceded writing or mind then perhaps the concrete golden mountain would have come before the word or concept "golden mountain", infinity before "infinity" and even nearly all swans being white before the sentence "nearly all swans are white". Perhaps it is easier to accept the sentence "Nearly all swans are white" than the words "infinity" and “Golden Mountain”. Swans existed before that sentence. And, perhaps, the whiteness of swans. And so did the fact that nearly all swans are white. Even if we accepted the old theory of meaning that stated that meanings were the denotations of words (but what about "round square"?), denotations would still not be the meanings. White swans are not meanings. They, in the denotational theory of meaning, may be part of meaning, but they would not be sufficient for meaning. It is minds, through sentences, that individuate the meaning of "Nearly all swans are white". It is a mistake to think of meaning preceding writing.  Again, the things needed for meaning (white swans, even abstract objects, etc.) may well exist before linguistic expressions, but they do not, on their own, entirely constitute meaning. The meaning of "Golden Mountain” couldn't have existed before minds because it is an invention of minds. So, in this case, how can the meaning of the word "golden mountain" absolutely precede "writing"? What would it mean to say that the meaning of the word "infinity" came before "writing"? Take Quine's position on this:

[I refuse] to admit meanings, for I do not thereby deny that words and statements are meaningful…even though McX construes meaningfulness as the having…of some abstract entity which he calls a meaning… (1948)

And elsewhere:

…the myth of a museum in which the exhibits are meanings and the words are labels. To switch languages is to change the labels…The primary objection persists even if we take the labelled exhibits not as mental ideas but as Platonic ideas or even as the denoted concrete objects. (1969)

Labels or words aren't peripheral to meaning - they are necessary. To use Quine's terms, there would be no exhibit without labels. So this simile of Quine's doesn't seem to work at first because exhibits could exist without labels…But could they? The physical reality of exhibits does not depend on labels, but they would not be exhibits without their labels. We wouldn't know what they were exhibiting or what is exhibited without labels. Therefore they would not be exhibits simpliciter. In Derrida's terms, something would exist before "writing", that is, whatever is needed for meaning, but meaning itself wouldn't exist before "writing". Simply because Quine's exhibits can exist mind-independently, or without labels, this doesn't mean that exhibits are analogous, in Quine's terms, to "Platonic ideas" or whatever.

Quine, being a semantic behaviourist, said that meanings are determined by our "dispositions to overt behaviour" (45). Basically, Quine meant here by "overt behaviour" things being said or things being written. Therefore this will include the disposition to say something or write something. If meanings were not a "disposition to overt behaviour", or something "written" or "said", then they could float around in the mind, as it were, without sentential clothes. In a sense, therefore, meaning would be like elements in Fodor's Language of Thought that must come before public expression in a contingent and non-universal language. The comparison is not exact however, because the LOT is, obviously, language-like in the first place, whereas meanings, on a non-Quinian or non-Derridean reading, are non-linguistic and not even proto-linguistic (e.g., "Platonic ideas", "denoted concrete objects", abstract objects, truth-conditions, propositions, Fregean Thoughts, etc.). Derrida, on the other hand, said that meaning "must await being said or written" (10, above). Of course, "being said" is "overt behaviour". Again, Quine wouldn't have denied the distinction made between exhibits (meanings) and labels (words). It's just the case that the exhibit, in this case, can't exist without the label (word). Or, to stretch Quine's simile again, exhibits in a museum wouldn't be of much use without labels. How would we know what we were looking at? "What's being exhibited?" Similarly with Derrida. He can make a conceptual distinction between "literality" and meaning without accepting that they can be separated.

Again, there is a conceptual separation to be made between sentences and meanings or propositions. For example, different sentences can express the same meaning (as I said earlier about propositions). For example:

"The British Prime Minister is a liar"

and

"The current incumbent at Number 10 tells lies"

both could be said to have the same meaning. But is there a meaning of "The current incumbent at Number 10 tells lies", or, should I say, the truth-conditions of the current incumbent of Number 10 telling lies, that could escape all sentential expressions? We've already said that the denoted concrete object (say Tony Blair) or the denoted concrete actions (say, Tony Blair telling lies) can't be the meaning of the various sentential expressions. Tony Blair lying clearly needs to be part of the story, but it's not all of it. Even if there's something abstract which, as it were, refines the meaning, mind, through its sentences, would still need to do the individuating (Cf. with the individuation of facts, objects and events). After all, before Tony Blair lied on one particular occasion, he didn't lie. And after he lied on one particular occasion, he didn't lie. And even when he was lying, he was also, say, speaking to journalists or standing on a podium. So the truth-conditions of Tony Blair lying are, as it were, atomised. That is, individuated from the surrounding manifold. This is why meanings can't be denoted concrete objects or acts/events, etc. And this is why even more refined abstract objects like "Platonic ideas", propositions, truth-conditions can't be meanings on their own. The meaning can't escape all sentential expressions. 

However, I would say that concepts are not necessarily linguistic (see opening dictum by Geach), but they are necessarily mental. I think that, say, dogs can have concepts. We can have concepts partly or wholly determined by mental images (Tye, 1990). I agree with Quine when he denied that meanings are abstract objects. But he also argued against "pernicious mentalism" (1969) vis-à-vis meaning. Such a "mentalism" regards "a man's semantics as somehow determinate in his mind". I certainly don't think that a meaning is a mental image or mental "entity", for reasons already given. But I do think meanings are essentially mental. Again, meanings may include mental images or entities as constituents, but it is sentences or "overt behaviour" which individuate the meanings. The mental sentence turns the mental image or entity or concrete objects or truth-conditions, whatever you favour, into meanings. The meanings themselves aren't concrete or abstract objects, mental entities, truth-conditions, or anything else. To put this another way. Rather than meanings being dependent on sentences or "over behaviour", they are dependent on cognition (which may well be "over behaviour" if vocally expressed). So Quine and Derrida were correct to stress the importance of "overt behaviour" or "writing" when it comes to meaning, but wrong to argue that it is all primarily a question of sentential expression. The mental is primary to meaning, but language is not altogether necessary. However, the non-linguistic aspects of meaning could also be expressed in "overt behaviour", either in speech or non-linguistic behaviour.

So according to Quine and Derrida, meanings are not the "a priori" (11) of writing, words, sentences and vocalisations ("over behaviour"). They don't float around in a platonic realm or, for that matter, stand still in a concrete world waiting to be given sentential clothes. They don't exist as meanings until they put on their sentential clothes. The things that are necessary for meaning may well exist mind-independently in a platonic realm or in the concrete world, but they are not themselves meanings.

It may seem too obvious to state, but those that stress language when it comes to meaning (or propositions, truth-conditions, facts) do not thereby deny the other constituents of meaning. Derrida, however, comes dangerously close to doing so when he said that "there is nothing outside the text". However, let's be grown-up about this. This statement was just a piece of "rhetorical flourish" (Shalkoswi, 2000). It was, if you like, a bit of philosophical titillation. And it was also a way of getting the point across via the medium of prose. After all, Quine also said that "to be is to be the value of a variable". This wasn't meant to be seen as expressing the view that the value of a variable suddenly sprang into existence when it was written down on the page as part of a quantificational schema. We can't escape metaphors in language, just as we can't escape metaphor or models in the case of science.

"Writing" does not simply transcribe something that already exists - meaning. Meaning can only exist as language - in the guise of language.  Derrida quoted Merleau-Ponty who articulated a quite  well-known phenomenon within psychology. He said that his "'own works take [him] by surprise and teach [him] what [he thinks]'" (pg. 11). I have heard many people say, in a roundabout way, that they didn't know they knew so much until they started writing it down. The words, even the sentences, gave them meaning. The meanings had "no dwelling place" and "do not await us" until we expresses ourselves. Meaning "becomes…what it is" when in sentential clothing. We could express this more strongly. We can't even think without language. Or, if that's too Derridean, we could say that we can't think about many things in certain ways without language. Words and sentences aren't an afterthought. And Derrida, with a typical Derridean pun, classes the view that meaning is "anterior" to words and sentences as "idealistic". So there is not even an "Idea" behind the words, the words are the "Idea".

The conclusion to all this is fairly clear. If there is nothing behind the word, and all language is a social creation, then ideas and metaphysics are stripped of their "absolutist" pretensions and shown to be contingent, relative and the products of history. However, these rather broad assertions will be elaborated upon elsewhere in this paper