A Skewed Analytic Reading of Derrida’s Early Philosophy of Language
Introduction and Apologies
I have used the adjective ‘skewed’ in my title because I could easily imagine that many of, if not all, Derrida’s loyal post-structuralist and deconstructivist fans would not in the least like or agree with my rather conventional reading of the philosopher. Also, I say that my reading is ‘analytic’, and therefore by definition ‘skewed’, because it relies more on my own knowledge of analytic philosophy than it does on the work of post-war Continental philosophy or the writings of the many American and European commentators on Derrida’s work. The majority of these people would no doubt say that I am trying to put Derrida into a box in which he simply doesn’t belong. They would say, to state a truism, that Derrida was not an analytic philosopher. Yet my positive ‘reading’ of Derrida’s philosophy does indeed owe much to analytic philosophy. This does not concern me greatly, and for two reasons. One, there isn’t much agreement on Derrida’s work in the Continental and American traditions either. Two, Derrida himself tried his best to restrict any decisive and conclusive interpretations of his work. Derrida’s attitude towards his own philosophy was partly derived from his ‘philosophy of language’ itself. For example, Derrida argued, if rather obliquely, about the ‘inscrutability of reference’ and the ‘indeterminacy of meaning and translation’ (to use, instead, Quine’s terms, see his 1969). These are the problems faced by all ‘symbol systems’. It follows, then, that if I read him in a conventional and/or analytic way, then Derrida himself should/would not have argued, if he had still been alive, against my own ‘autonomous reading’. After all, didn’t Derrida believe, along with Barthes and Foucault, that the ‘author is dead’? And if the author is truly dead, then surely too is the philosopher. Not only that, but we are all aware that even if the ‘reader’ is not in fact ‘autonomous’ vis-à-vis his reading of a work of philosophy, he or she can still get a tremendous amount from his or her own ‘misreading’. For example, many Wittgensteinians have been told, by many other Wittgensteinians, that they have, pure and simply, ‘got Wittgenstein wrong’. Yet the philosophers who get Wittgenstein wrong, even if they aren’t airtight Wittgensteinians, may still get much from their possible ‘misreadings’. If a misreading of a philosopher’s work takes us in all sorts of interesting and fruitful philosophical directions, then good on such a misreading!! Again, according to Derrida himself (at least at one point in his career and perhaps also only to a limited degree), and also according to many of his Continental and American ‘interpreters’ and fellow deconstructors, there cannot be a pure and simple misreading or correct reading of his work precisely because there cannot really be a pure and simple misreading of any work of philosophy or, for that matter, literature. I myself will remain largely noncommittal about this particular Derridean position in this paper.
When writing this paper I relied quite extensively on Christopher Norris’s book, Deconstruction: Theory and Practice (1982/91). This book was also written by a philosopher who was trained in the analytic tradition. And that, no doubt, is why I found it both easy to understand and enjoyable to read. My reading of Norris’s book was in clear distinction to my readings of many other books on Derrida. I often found these books on Derrida incomprehensible and/or simply pretentious (e.g., Marion Hobson’s Jacques Derrida: Opening Lines, 1998). Despite saying that, at one point I also took Derrida’s own works to be incomprehensible and pretentious. In fact, this was almost true by definition for someone who knew very little about post-war Continental philosophy. It took a lot of hard work and a large dollop of faith to eventually see things in a different way. However, I still believe that there is a sense in which it is the case that Derrida was indeed a game-player, a ‘charlatan’ and a ‘sophist’ (see Searle’s account, 1977). But such ‘rhetorical’ philosophical stances, along with his humour (or ‘irony’), were also derived by Derrida from his own philosophical positions on ‘Western metaphysics’. (This would no doubt be an interesting area to pursue.)
So I didn’t rely overmuch on American and Continental interpretations and elucidations of Derrida’s philosophy, but relied instead on those few works – or words - that have been written by Anglo-American analytic philosophers. (I should say here that although Norris’s book is enjoyable and, indeed, insightful, I still believe that he is a little too loyal and uncritical of Derrida’s philosophy for my taste, which is not to say that his book is never critical of him.) There is such a dearth in this area, that instead of saying that analytic philosopher So-and-So said this-and-that about Derrida, I myself have had to apply the works of analytic philosophers to Derrida’s work (see, for example, my ‘Wittgensteinian Digression’ later). In so doing I soon found that Derrida’s ‘philosophy of language’ is not so original and so distanced from analytic philosophy’s traditions as some of his poststructuralist fans would have us believe. For example, there are shades of Wittgenstein, Quine, Davidson, Feyerabend, Ryle, Popper, and even Tarski and Carnap, in certain parts of Derrida’s work. This is not to say, of course, that he has read all - or any - of these analytic philosophers. (But he most certainly did read, for example, Wittgenstein and also, of all people, J.L. Austin.) Of course there are many elements in Derrida’s philosophy that the average analytic philosopher would not dare be seen with in the Analytic Academy. But there are also elements of, say, Quine’s work that many analytic philosophers would not dare be seen with either.
All the above leads me to conclude that it is hard enough taking on board the huge world that is analytic philosophy without also attempting to do the same with Continental philosophy. (Bravo Norris!) In a sense, therefore, I was destined to come at Derrida’s ‘philosophy of language’ (do I still need the scare quotes?) from a specific ‘mode of presentation’ or a particular Husserlian ‘profile’ – i.e., the modes or profiles of certain strands in 20th century analytic philosophy. The title’s reference to Derrida’s ‘philosophy of language’ itself betrays my philosophical biases. For instance, I doubt that there is a single book called Derrida’s Philosophy of Language written by a deconstructor or by any of Derrida’s Continental and American poststructuralist followers. So I apologise again… Or, rather, I shouldn’t really apologise at all! As I’ve said, there are certain elements in Derrida’s own philosophy of language which led me to believe that my title clause, ‘skewed reading’, may not only be crudely and overly modest, but also decidedly non-Derridarean in tone. In that case, I really shouldn’t have called this paper ‘skewed’ in the first place. Despite all that, I did indeed decide to use the word ‘skewed’ after all. And I did so precisely because I am not a rigid follower of every Derridarean utterance. This means that I am not altogether happy with his philosophical positions on, for example, ‘autonomous readings’ and the ‘death of the author’. And therefore I may even be prepared to argue that there are or must be ‘correct and faithful readings’ of Derrida’s early (i.e., c.1962-c.1977) philosophical works. To argue that case, however, would be another issue entirely.
Part One: Symbol Systems, Words and Objects
Saussurian Structuralism: Le Langue
Structuralism, from Saussure onwards, clearly had a profound effect on early Derrida’s poststructuralism. It was Saussure who was the primary structuralist influence on the early works of the French philosopher. And what was Saussure’s most important theory for Derrida? The theory that ‘language is a differential network of meaning’. What does that mean?
Saussure essentially attempted to break the close link between word and its meaning/concept/object. To use Saussure’s own terms. He attempted to break the link between ‘signifier’ (the spoken or written word) and the ‘signified’ (its concept/meaning/thought). One could ask here what was traditionally deemed to be this close link between word and its meaning/concept/thought? What Norris says is that “[t]here is no self-evident or one-to-one link between ‘signifier’ and ‘signified’”(1991). What would such a link be like?
Saussure’s general linguistic thesis is essentially holistic in manner. What matters is not the link between word and its meaning/concept/thought, but between a word and other words; and therefore between a meaning/concept/thought and other meanings/concepts/thoughts. Norris gives a rather simple example of Saussure’s ‘differential network’:
Bat and cat – have the same vowels but different consonants
Bag and big – have the same consonants but different vowels
Clearly these words have some kind of phonetic relation to one another. They are parts of a ‘system of differences’. However, simple phonetic differences and relations do not seem to have a direct philosophical significance. And we will see later how Derrida used Saussurian linguistics and applied it to the ‘symbol systems’ of philosophy itself.
Saussure thought it important to distinguish two components in any language:
2) the isolated speech-act or utterance (parole)
Ý
1) the general system of articulate relationships (la langue)
As will be shown, 2) is derived from 1).
The linguistic system, according to Saussure, existed before ‘any possible sequence of speech’. The speech-act depends on the overall system (la langue). More strongly, ‘meaning could be produced only in accordance with the organising ground-rules of language’ (Norris, 1991). These ‘ground-rules’ must, therefore, be a kind of grammar, or, possibly, even a kind of logical grammar.
According to Derrida, we are not in full control of our meanings precisely because a linguistic or symbol system pre-dates each individual speech-act. The “signifying system…exceeds all the bounds of individual ‘presence’ and speech”(Derrida, 1967/78). Here the word ‘presence’ means both the full presence of the speaker and the full presence of his meaning.
Remember that each speech-act or each act of writing is predated by the linguistic system in which it takes place. According to Derrida, the relation of difference is of vital importance in such systems. This means that words or ‘signifiers’ have intimate relations of difference and similarity with other words within the given symbol-system or language. More precisely, they have relations of difference because each word, etc. is different or similar to every other word within the system. This means, therefore, that a word’s meaning/concept/thought will never be fully present ‘within it’. The word in question will therefore depend on its role and position within a particular ‘system of differences’. It will follow, then, that the meanings/concepts/thoughts of words are holistic phenomena: they will dependent on their relations (of similarity and difference) with other meanings/concepts/thoughts within the linguistic system. This relativity, inscrutability and indeterminateness of words and their meanings/concepts/thoughts, as well of all their referents, has been called ‘undecidability’. Derrida argued that this undecidability within a particular linguistic or symbol-system results in a ‘free play of signs’ that is brought about because nothing within the system is fixed or determinate. Meanings/concepts/thoughts are, as it were, played with by the system’s speakers and cognisers. And, in turn, meanings/concepts/thoughts also, in a sense, do a bit of playing with their speakers and cognisers.
The Chain of Words
… each element appearing on the scene of presence, is related to something other than itself, thereby keeping within itself the mark of the past element… (Derrida, in his Of Grammatology, 1967/1977, pg. 65)
The passage above is essentially a reference to symbol-systems or ‘systems of signs’ and the basic elements that occur within each such system. We can take Derrida’s ‘elements’ to be either words or concepts (or even, in certain cases, whole sentences). However, it would be best here to think in terms of the meanings of these words or concepts and their relations to the system as a whole.
Derrida uses the term ‘presence’ here to specify that which is present or at hand, as opposed to that which is not present or not at hand. Or, more particularly, those aspects of a word that are present or at hand, and those aspects of the word that are not present or at hand. Each word is related to something other than itself. We may interpret this to mean that at first it is often thought that the word is related to the thing to which it refers, or, at the least, to the fixed and determinate meaning of the word that is present. However, what this word is related to is not (only) the thing it refers to, or even just the meaning of this word. It is, in fact, more closely related to other words, therefore to other meanings/concepts/thoughts. Each word, then, has ‘within it’, to use a spatial metaphor, things that are not immediately at hand in that word. There are aspects or parts of the word that are not immediately discernible when we read the inscription or try to comprehend its meaning/concept/thought. These other things, on the word’s periphery, as it were, are other words and, therefore, other concepts/meanings/thoughts. In Derrida’s own words, these peripheral, or seemingly peripheral, aspects or parts of the word in question leave their ‘mark’ (Derrida, 1967/78) on it. These peripherals are not there in all their fullness. But they have still, nevertheless, left their marks on the word in question. These marks are now in the ‘past’ of the present word we are using or trying to understand. Other words, and therefore other meanings/concepts/thoughts, have left their ‘marks’ on the word, and its concept/meaning/thought, under scrutiny. The word is related to these ‘past elements’. It is not, therefore, freestanding or atomic. The ‘trace’ of other words, and therefore of their meanings/concepts/thoughts, can be ‘found’, metaphorically speaking, on the word itself. It betrays the signs of belonging to other words and their meanings/concepts/thoughts.
In terms of the system of signs itself, this effectively means that the meaning of the word or sign is dispersed throughout the system - throughout its other words and their meanings/concepts/thoughts. In order to fully understand the word’s meaning, we will therefore need to go elsewhere within the system. We could also say that these dispersed elements are actually part of the word being scrutinised. Not actually distanced or dispersed from it. If we can only understand a word by metaphorically going elsewhere in the system, then these other parts of the system must in fact be somehow contained within the word, or within the word’s meaning or definition. Meanings cannot be atomic because words and concepts cannot be atomic. We can take these Derridarean arguments further, in a Quinian kind of way, by starting to talk instead about the relations between sentences, rather than between words, within a ‘web of belief’ rather than ‘a system of signs’ (see Quine’s 1969.)
We do not get a word’s full meaning just by analysing the word’s meaning or the word itself. We must, of necessity, look ‘elsewhere’ to find and clarify its meaning. We must find those aspects or parts that aren’t present or inhabiting the ‘scene of presence’ that is the word and its meaning. The word’s meaning is, therefore, to some extent open-ended and dispersed throughout its symbol system, as I have already argued. And if that is the case with a single word, then it will certainly be true of a single sentence or statement. As earlier in the case of Quine’s and his holistic ‘web of beliefs’ or ‘scientific systems’, we can take this Derridarean and Quinian holism further, as Derrida in fact did but Quine did not, by creating ever larger and larger systems. For example, we moved from words to sentences, then from individual theories to Quinian ‘scientific systems’, but we could also now talk about cultural or historical ‘totalities’, or, instead, as Derrida did, we could talk about the ‘the syntax and system’ (1967/78) that is the ‘totality’ that Derrida often called ‘Western metaphysics’. But let us get back to the humble word or signifier.
The other traditional way (up to, and beyond, J.S. Mill) of looking at the word is to say that its meaning is simply the thing it refers to – its ‘denotation’. Or the least we could say is that a word’s meaning, or, more likely, a name’s meaning, depends in an important sense on its referent or denotation. However, what if the referent or object itself belongs, in a sense, to the symbol system. That is, the way we come to know, or identify, or categorise, the referent or object, is itself dependent on the words and therefore the concepts/meanings/thoughts that are themselves contained within the system. The referent, or thing, may also be seen as being contained in the system. The thing or referent is not ‘exterior’ (Derrida, 1977) to the system. At least not the thing as it is when the cognisers and speakers of a linguistic system come to know the thing as the thing that it is to them. Alternatively, if the thing is or remains uncognised, or unseen, or unknown, then of course the thing is exterior to the system – to all symbol systems. But what is this uncognised, unknown and unseen thing as it is exterior to one and to all systems? Tell me something about it. What is it like? Of course as soon as we do answer these questions, then the thing is no longer exterior to all systems because a system’s concepts and words will be used to answer these questions. We may as well, therefore, give up on the idea of a genuinely exterior thing, just as Berkeley gave up on the nature and even the existence of unperceived objects and things. They serve no purpose. They are utterly redundant, just as the idea of matter was meaningless to Berkeley (see his 1713/1979).
Not everything about a word or concept is present. The ‘scene of presence’ leaves a lot out. It follows, therefore, that meaning can never be fully present in the word or the meaning of the word. It also follows that there is never any genuine ‘pure presence’. Each understanding of a word or concept, each thought, each cognition, will not be a self-contained unit. Its meaning, or its assertoric-content, will be dispersed throughout the system. Part of the meaning will not be present in any direct sense. There is no isomorphic relation between word, its meaning and the subject that understands it. This not only means that the thinker will need to look elsewhere to establish the meaning of a word or a sentence, as we have argued, he may never know precisely what it is that he means, precisely because of the dispersed nature of meaning. This is not because of any cognitive failings on the subject’s part. It may be the case that in principle a subject is not in complete control of his meanings. And because the system cannot be grasped in the time a single thought or meaning can be grasped, then the meaning of the word or sentence may remain somewhat unknown to the subject. It may even be the case that it is up to other subjects to have a go at what the original subject meant by his words. But they too, of course, will be victims of the dispersed nature of meaning. They may make other connections between the single word or utterance and all the other elements within the system they share with the initial subject.
There is never such a thing as ‘pure presence’ when we speak or understand a sentence or a single word. In fact, according to Norris, this was the ‘ruling illusion of western metaphysics’ (1991). We are not in control of our meanings and we do not fully understand the meanings that we adopt. The words and sentences and thoughts, therefore, are not ‘self-authenticating’ (Norris, 1991) – they rely on meanings, concepts and sentences outside themselves but within the system. Again, if all this is indeed correct, then no one knows exactly what he or she means when they say something. Or, more liberally, what they do say can never really stand on its own. It will need to be explained and clarified. And then the explanations and clarifications will also need explanation and clarification. This process could go on indefinitely. There will never be a time of ‘pure presence’ when a full, precise, determinate and fixed meaning offers itself up to the subject. Meaning may be fixed within the system, but not within the single subject’s act of understanding of a word and its meaning. And then, of course, there is the added problem of how other subjects will understand our words and utterances. If we do not know what we mean, then the hearers will not know entirely what we mean. And they may even interpret the utterance in a manner hugely at odds with our intentions. Those intentions too, however, will in a sense be dispersed and not fully present to the original subject.
Just as many philosophers have argued that our concepts and categories determine how and what we perceive, Derrida talks in terms of the ‘written’ nature of the ‘exterior’:
… signification is a priori written, whether inscribed or not, in one form or another, in a ‘sensible’ and ‘spatial’ element that is called ‘exterior’…The outside, ‘spatial’ and ‘objective’ exteriority which we believe we know as the most familiar thing in the world, as familiarity itself, would not appear without the grammè… (1977)
What Derrida must mean by the word ‘exterior’, in the above, is that it ‘would not appear without the grammè’. It wouldn’t appear as it does appear, not that it wouldn’t appear at all. The grammè, in Derrida’s terms, determines how we actually perceive the world or the ‘exterior’. The words and concepts that we use to ‘signify’ the world and its objects and events, have an effect on how the world is actually perceived and cognised. There isn’t one way that the world, or a part thereof, can be or is perceived and cognised. It depends on how the world is carved up or described in various prior ways. According to Derrida, we do not come into the world and then choose the means and ways of ‘signification’. These means of ‘signification’ are always already there; even though they can of course be changed. There is no signification-free world or perception of the world. Just as Kant said that the world is seen through the transcendental spectacles of, say, causation, time and space, so Derrida extends this idea and says that the world is perceived through our ‘forms of signification’, which are not, however, a priori or necessary.
Le Langue, Traces and Differance
Derrida introduced another technical term (briefly referred to earlier) in this debate: ‘traces’. It is because words or signifiers are parts of a pre-existing linguistic system, and therefore they have intimate relations with other signifiers, that these other words within the system leave a ‘trace’ on the signifier under scrutiny. The signifier is not an autonomous atom. Not only that, but these traces are so, as it were, ingrained and wide ranging that they result in the possibility that a full meaning can never be grasped by the individual speaker. Meaning, again, is indeterminate and dispersed.
Saussure, on the other hand, didn’t see this implication for speech and writing within a pre-existing linguistic system. He thought that there is a “ ‘natural bond’ between sound and sense” (Norris, 1991 ). There is some mysterious and necessary relation between speech-act and the speech-act’s meaning. The speech-act perfectly expresses the pre-existing meaning or thought. The meaning or thought and, perhaps, its reception, form a perfect union. Writing, according to many philosophers in the tradition, is supposed to sever this mysterious and intimate link between parole (the speech-act within a linguistic system) and meaning/thought/concept. (See later section.)
The term “differance” is the most famous of Derrida’s linguistic notions. It refers to words/names/signifiers and is derived from two verbs: ‘to differ’ and ‘to defer’. We’ve already talked about ‘the system of differences’. What of ‘to defer’? This again is a reference to meaning. It says that meaning is always deferred. It is never fully present at each speech-act or writing-act. In a sense, the full meaning is elsewhere in the linguistic system or, alternatively, spread out or dispersed in the linguistic system. The meaning we give a signifier can always be supplemented by the ‘play of signification’ (Derrida, 1967/78). Names/words/signs acquire new meanings because of the indeterminateness of meaning, the ‘system of differences’, and the spread out or dispersed nature of meaning. But remember that if meaning is indeterminate, then the meaning of the term ‘differance’ is also indeterminate! That word must have no settled meaning either. It is, therefore, a self-referential word: its meaning can be applied to itself.
Norris says:
Derrida deploys a whole rhetoric of similar terms as a means of preventing the conceptual closure – or reduction to an ultimate meaning – which might otherwise threaten his texts. (1991)
One may ask why Derrida took steps to prevent ‘closure’ (1967/78) when all the while he argued that meaning is intrinsically indeterminate? Whether or not he warned us by doing what he did to meaning (i.e., deconstruct), meaning, according to him, would still remain indeterminate regardless of any animadversions from the deconstructor. So why the warnings and the deconstructive strategies?
Derrida and Symbol-Systems
We can see in Karl Popper, for example, echoes of many Derridarean ideas. For instance, the idea that we cannot reach through language or words/concepts to ‘pure presence’; there is only the ‘play of the sign’. Words are defined by other words, which are themselves defined by other words. No word or concept or thing is ever captured in a ‘finite web’ (Derrida, 1967/78) of meaning:
The derivation [of a term] shifts the problem of truth back to the premises, the definition shifts the problem of meaning back to the defining terms (i.e., the terms that make up the defining formula). But these, for many reasons, are likely to be just as vague and confusing as the terms we started with, and in any case, we should have to go on to define them in turn; which leads to new terms which too must be defined. And so on, to infinity. (Popper, 1967)
And that is the main problem with definitions: they too will contain terms which themselves will need defining. A similar problem is apparent when it comes to an argument and the premises on which they are based. The premises of any argument may themselves actually depend on further premises and conclusions that may act as justifications for the validity and truth of our initial premise. Thus we have on our hands a regress of justification. But the game cannot go on forever. This is why we must, at some point, simply accept premises, arguments or definitions, if we are to get going on our particular philosophical or logical enterprise. Perhaps this is also an argument for at least a qualified kind of foundationalism. The premises that we accept, however, needn’t be ‘self-evident’ or ‘indubitable’ or anything like that. They simply need to be the starting points of reasoning if we are to avoid infinite regresses (see Wittgenstein’s 1949/1969).
These facts alone may show use that, in certain senses, we may remain trapped within language, unless, for example, we depend on starting points like ‘sense-data’ or the ‘Given’: things that are pure in nature and free from conceptual or linguistic baggage. The question is, of course, whether or not there are such things.
In a sense, it is precisely because definitions depend on their own un-defined terms that all words, or nearly all words, are inherently vague. (Compare with Quine’s ‘analytic hypotheses’ that are needed to get the game of native-translation going. In his 1969.) As Derrida would have put it: meaning is never ‘fully present’. And that is because the meaning of a single word depends upon all the meanings of all the words in the symbol system to which it belongs. We cannot have a precise definition, if the definiendum itself contains terms that are not themselves defined. But even if we do in fact define the terms in the definition, these definitions of the terms within the initial definition will also require elaboration and definition. And so on indefinitely.
The holist position on these problems will be that we must take into account the symbol-system to which the words under definition belong. But this too is problematic. How can we really take on board an entire symbol-system each time we want to define a term? Does that mean going on another kind of indefinite regress? And if not an indefinite regress, then does it mean taking on board all the other words and meanings/concepts/thought of the symbol-system? If not of the entire system, then perhaps a sub-system of the larger symbol system? In that respect, the regress will not be indefinite, but it will prove to be circular at some point. This essentially means that we would arrive back at some of the terms that we actually began with.
There are jut as many problems with this type of holism, or perhaps this type of coherentism, as there was with, say, logical atomism. Similar holistic, or coherentist (or even monist), points to these were raised by Bradley, amongst others, in the late 19th century. (And, of course, Russell and G.E. Moore, amongst others, reacted against the holistic problems that they found in the works of the German philosopher Hegel as well as the works of Bradley, Bosanquet and other British idealists.) Because of the problems with holism, Bradley concluded that no statement could be absolutely true. Similarly, we may now say that no definition is ever free from vagueness precisely because of its necessary containment within the larger symbol system to which it belongs.
Presence and Centre
We have referred to Derrida’s notion of ‘presence’ quite a few times so far. What, precisely, did Derrida mean by the term ‘presence’? In the following, from his ‘Structure, Sign, and Play’ (1967), Derrida explained what he meant, in part, by this term:
… by a process of giving it [‘structure’, ‘the latest presence’] a centre or of referring it to a point of presence, a fixed origin. The function of this centre was not only to orient, balance, and organise the structure… but above all to make sure the organising principle of the structure would limit what we might call the play of the structure. (1967/1978)
This is partly explains the philosophical desire for fixed and determinate foundations, abstract meanings, necessary rules, a priori principles, and fixed natures, all of which would orient the rest of philosophy and, indeed, the rest of human life. But all of these saviours were and are, according to Derrida, part of a totality, or part of a whole, or part of a system. They have no genuinely privileged status. This, again, is an essentially holistic position on Derrida’s part because it denies the very notion of a ‘centre’ (or, say, a ‘foundation’) to everything that supposedly goes on ‘outside’ that centre. Even if we can conceivably recognise at least some limited kind of centre or foundation, that more modest centre or foundation would still be polluted by what goes on ‘around it’. Systems will determine, to some extent at least, what we take and what we have taken to be the centre, or what we take and what we have taken to be a foundation, or what we take and what we have taken to be a fixed and determinate meaning/proposition/concept, etc. The centre is not, therefore, an axis around which we, or the rest of philosophy, spin. Everything spins. Or, to use Derrida’s own words, everything takes part within the ‘play of signs’. There is no grounding to all this fun. All we have is ‘play’.
Again, all this is a very holistic position from Derrida. We could almost say that it is almost Hegelian or even Bradlian in spirit. All ‘centres’, that is, all meanings, axioms, premises, sense-data, etc., are touched by what is perceived to be on the ‘outside’. All these hopefuls belong to the various systems that, to some extent, determine their nature. For example, concepts are always part of a system or a language in which yet other concepts ‘touch’ the concept we are initially concerned with.
According to Derrida, the centre is a fiction, for the following reason:
The centre is at the centre of the totality, and yet, since the centre does not belong to the totality (is not part of the totality), the totality has its centre elsewhere. The centre is not the centre…and which, because it can be either inside or outside, can also indifferently be called the origin or end, arché or telos… (1967/1978)
Many things have played the role of ‘centre’, according to Derrida. Take two important examples for Derrida: the Aristotelian telos and the Platonic arché. The centre is either where we, or at least Aristotelian philosophers, are heading. Or the centre is what non-Platonic philosophers have already left behind. In both cases, the end-point or beginning-point still orients us in our philosophical pursuits. We could say that arché is our Platonic birth, whereas the Aristotle’s telos is the Aristotelians, and all those who buy Aristotelianism, forthcoming heaven. Plato thought that we should try to discover the ‘centre’, that is, arché. For Plato, these starting points or ‘centres’ were the universals or forms or Ideas in his metaphysical scheme. According to Aristotle, on the other hand, we, or all Aristotelians, are heading towards the various ideal ‘centres’ or ends – the very ends for which we reason and even act.
What Derrida was saying is that once correctly analysed the centre does not exist and therefore the Platonistic or Aristotelian philosopher sets up an intolerable dichotomy between the various centres and the various ‘totalities’. The centre was always privileged. And the elements that made up the totality were seen as peripheral or even deemed to be normatively suspect. There is nothing in reality underpinning the ‘centre’. The peripherals are all that there is. The metaphysical centre was an illusion that we strived to discover but never did. For example, the ‘inner self’ or subject or ego or mind-substance was and still is deemed to be the centre that underpins what we are today and what we were not yesterday: a changing subject, or empirical self, that is part of the (empirical and contingent) totality. The ‘self’, ‘subject’ or ‘soul’ is the ‘origin’ of what comes later: the later totality of personality. Or, alternatively, we are heading towards a perfect end (telos): towards, say, ‘a rediscovery of our pure soul’, or our ‘true self’, or our ‘genuine Being’. We were, or indeed still are, heading towards the centre (telos) – Aristotle’s position. Or we have already ‘lost’ the centre or left it (arché) behind – Plato’s position. Again, both Plato and Aristotle saw the totality, as, in a sense, nothing more than the ‘flux of appearances’. The centre, on the other hand, was thought, again by both Plato and Aristotle, to be - also in a way - the ‘thing-in-itself’ (if not in the same way that Kant saw the ‘thing-in-itself’).
Philosophers, Derrida argued, have always tried to find something pure that can act as a foundation or an exemplar. Plato, of course, argued that we had foolishly left these pure entities behind. And Aristotle argued, to the contrary, that we are heading to something that is, in a sense, pure. We have the ‘pure self’, the ‘true meaning’, the ‘right position’, and so on. All our attempts to discover such things result in us denying the importance of all the peripherals. Derrida argued that there is nothing else but the peripherals. There are only, say, empirical selves, contingent expressions (with no underlying propositions), no true Justice other than our myriad attempts at justice, and so on.
The sign, or the ‘play of the sign’, is the ‘totality’, whereas the ‘signified’ was and is seen to be the ‘centre’. Derrida thinks that there is no centre. Therefore there is only the ‘play of signs’ and no ‘centre’ hidden behind or beneath the signs – there is no ‘thing-in-itself’.
Because there are no centres, but only totalities, Derrida thought that we should be happy with what he called ‘the play of signs’. We play with words that do not replicate signifieds. We play with expressions that show no respect for hidden propositions. And sentences are given their meanings by offering other sentences, not by discovering their ‘true meaning’. Also, there is no such thing as Justice, only our halting attempts to create things that can pass, with justification, for justice. And the totalities in which these words, expressions and concepts exist will, of course, colour and pollute them. They can never exist outside the totalities. More strongly, the totalities determine what has often passed for the ‘centre’. In fact, centre and periphery is in effect a false dichotomy.
The Transcendental Signified: A Wittgensteinian Digression
How does Derrida’s early philosophy of language, exactly, differ from that of, for example, many previous ‘linguistic’ philosophers? His concept of the ‘transcendental signified’ is simply his way of putting something that had already been put by various Oxbridge philosophers.
Take the question: ‘What is truth?’ The questioner, by the very nature of his question, seems to think in terms of truth being some kind of thing or entity. Perhaps the questioner simply assumes that all words must have a referent. Perhaps he believes that the word ‘truth’ refers to the thing, truth. What if truth is not an entity or even a property? For example, take the Socratic question: What is beauty? We can now ask, is the entity, beauty, itself beautiful? Or, even more bizarrely, take truth again: What is truth? Is the thing, truth, also true? This now allows us to ask: if beauty, the thing, does not have the property of being beautiful, then what properties does it have? Similarly, if the thing, truth, is not true, the what properties does it have?
The other conclusion that we must draw from the Socratic attitude is the Platonic assumption that these entities, if they are entities, must exist separately from the expressions that use the words that refer to these abstract entities. Do they have an existence before becoming materialised by words or sentences? And, of course, Plato thought that they do.
On the Socratic reading, words and language itself must be peripheral to these self-subsistent entities. The Socratic philosopher must instead ‘pierce through the language to some non-linguistic reality that stands behind the words’ (Magee, 19). What kind of existence would these entities have when unexpressed? What would be their identity-conditions? Would they have any criteria of identity whatsoever? How could they have identity-conditions when in their state of existence away from all corporeality and linguistic expression?
Of course, words need not refer to objects of some kind, or, in a sense, to anything at all. We happily accept this when it comes to the logical constants ‘or’, ‘and’, ‘not’ ‘if… then’ and so on. And also with the articles ‘the’ and ‘a’. Other symbols, like ‘=’, need not refer either. What about ‘truth’ and ‘beauty’ – must they refer to entities or even to anything whatsoever? At a prima facie level, it may well seem to us that they must refer or denote in some way, otherwise how can we come to understand these words? We can’t use the methods similar to those used to understand, say, the logical constants. In the case of implicit definition we are shown what the constants ‘mean’. But we cannot say what they mean. Not only that, but the logical constants are actually used in the definitions of the logical constants. Perhaps we could say, bearing this in mind, that the entity beauty is indeed beautiful, just as, in a way, inductive principles are used to justify induction. However, perhaps the words ‘truth’ and ‘beauty’ cannot even rely on these self-referential methods of definition.
Perhaps, instead, we can say that these words are used in certain ways, but they still, nevertheless, do not refer. According to Wittgenstein (1956), they have particular roles in particular ‘language-games’. They serve a purpose within such language-games. And they can do so without referring in any way. In that case, if anything determines the meanings of the words ‘truth’ and ‘beauty’, it is the way in which these words are used. Therefore we can now say, along with Wittgenstein, that ‘meaning is use’. This is a thoroughly non-Platonic solution to the problems we have so far encountered. It is ultimately up to us, or our conventions, to determine what they mean. In the Platonic scheme, on the other hand, we find out what these words mean by matching them up, as it were, with abstract non-spatiotemporal entities. These entities will therefore be intuited by our rational minds but not be capable of being known through the senses.
Of course, we can now say that there must be more to all this than only the conventions and words themselves. There must be limits on what we can say and how we can say it. Without these Platonic entities, how can we stop ourselves talking nonsense, or saying or claiming anything we want to say or claim? Isn’t it the case that outside of words, conventions and usages, there must be things that determine our conventions, words and usages? Surely conventions, words and usages cannot be freestanding and self-contained. Such a view, surely, would inevitable lead to a form of linguistic idealism.
Certain Wittgensteinians argued that the meaning of ‘truth’ is the ‘sum total of all its possible uses’. This can’t be entirely true for the simple fact that a single subject will not be aware of ‘all [a word’s] possible uses’. The theorist of truth, however, may well be concerned with all possible uses (for theoretical reasons). It need not follow that the individual subject need be aware of all possible uses. He may simply be aware of one or a few possible uses. And, indeed, he may well decide that one particular use of the word ‘true’ is in fact the correct one, though it couldn’t be the true one (for the many reasons that Wittgenstein gave).
So what precisely do we mean by the “use of the word ‘true’ ”? One example may be that the word ‘true’ is used to recommend a sentence, or emphasise its factual claims. For example, in
The sentence ‘Snow is white’ is true.
the word ‘true’ is seen by many philosophers to simply reiterate or emphasise what has already been claimed. Or it could be a recommendation to accept the sentence that predicates the property white of the subject snow that does in fact belong to it. Although the sentence
The sentence ‘Snow is white’ is true.
and the sentence
Snow is white.
assert the same thing – that snow is white; the former example has a kind of implicit content in that it is suggesting to us that it would be a good thing to believe the proceeding statement.
What about the word ‘beautiful’ in the sentence ‘These roses are beautiful’? How is the word ‘beautiful’ being ‘used’ here? Of course it will partly depend on the context and who it is that is expressing this sentence. On one possible reading, the predicate ‘beautiful’ is being used to suggest to another person that he or she should look at and appreciate the rose. Alternatively, the word ‘beautiful’ may be used to urge a person not to deadhead the roses too soon. In that sense, the word ‘beautiful’ is used as a kind of warning-word or advice-word. By using the word ‘beautiful’ the owner of the roses is suggesting to – or warning - someone else not to destroy the roses because they are, after all, still ‘beautiful’.
Reification and Metaphor
[in] Derrida/Nietzsche – every last concept and category of western thought – including the terms ‘concept’ and ‘category’ – can be traced back to some effect of sublimated metaphor, some figural expression that philosophy needs to forget or repress. (Christopher Norris, paraphrasing Derrida, in his Deconstruction: Theory and Practice, 1991)
It is certainly the case that it is extremely hard to get oneself free from metaphor. But the danger is not to recognise that a metaphor is in fact a metaphor. This means that when we use certain concepts or categories, within philosophy or without, we are, in fact, using a metaphor that had forgotten, to speak metaphorically, that it is a metaphor. This seems to hint at the fact that we saw metaphors, things that are not entities, as entities. The philosophical tradition, argued Derrida (and indeed Nietzsche), saw metaphors as entities, or, perhaps, quasi-entities. ‘Western metaphysics’, to use Derrida own ‘totalising’ name, turned things that aren’t entities, not even abstract non-spatiotemporal entities, into entities. Past philosophers also actually personified, in various ways, many non-objectual things, just as the Greeks did in their ancient myths. Western philosophy saw and still sees, for example, ‘substances’ as entities. It saw and still thought that concepts were quasi-entities. And it was certainly thought, and, again, it still is, that essences are necessary entities ‘within’ larger contingent entities. It saw truth as some kind of entity. And, finally, it saw history, or ‘History’, as an entity. Not only that, but many philosophers - more or less - personified history too (i.e., Hegel). Why not continue. Many philosophers turned freedom, justice, right, etc., or Freedom, Justice, Right, into entities. No matter how abstract and refined all these supposed entities were seen to be, they were still nevertheless seen as entities of some kind. In many cases, it was because of this deep proclivity to reify that philosophers turned processes and contingent/conditional, and perhaps arbitrary, collections into entities. Perhaps all this reifying was a tacit attempt, by philosophers throughout the ages, to make all the phenomena that they dealt with - and reasoned about - into things that proved to be much more amenable to cognitive and philosophical manipulation. Posited concrete and abstract things would have been, and still are, easier to comprehend than mere empirical and contingent collections of phenomena or countless ill-defined and amorphous processes. We could say that just as many philosophers claimed that concepts help us carve-up the world’s unruly manifold, so too did hypothesised or posited concrete and abstract objects make the (Platonistic) philosopher’s life, if no one else’s, simpler, and they also enabled him to ostensibly solve, at least according to himself, many deep, important and perennial philosophical problems.
Part Two: Thought, Reason and Language
Writing and Meaning/Thought
At the heart of Derrida’s philosophy is the thesis that there is no ‘meaning’ beyond or behind the words – beyond language. Language is the beginning and the end of meaning. Derrida puts his own position in this way:
It [writing] is also to be incapable of making meaning absolutely precede writing: it is thus to lower meaning while simultaneously elevating inscription. (1967/1978)
What could it even mean to say that thought or meaning is antecedent to linguistic expression? Can we even make sense of such a belief? Sometimes people may indeed say, ‘I didn’t say what I thought correctly.’ This doesn’t mean that there was a thought waiting in the mind to get expressed in the correct manner. It simply means that the subject knows that there could have been a better way of expressing what he did actually express. This would be, therefore, a comparison not with a pre-existing thought or meaning, but a comparison with other expressed thoughts that he has previously heard (even if he can’t explicitly remember them) or imagined.
Meaning seen as an object. A proposition seen as an entity. That’s what fools many people, according to Derrida. Something linguistic must match, in some way, a meaning or a proposition or a thought. What we have here is a tacit quasi-correspondence theory of meaning. The word or utterance somehow corresponds to or matches the proposition or meaning in the way that a statement may correspond to a fact-in-the-world. This, surely, is a concretisation of meanings and propositions. Meanings are not things or entities. Indeed, on some readings, they are nothing at all (see Quine, 1953 and 1968). However, if they are something, they certainly aren’t entities or quasi-objects. If someone does say ‘I didn’t express what I thought correctly’, then that ‘thought’ too will be linguistic. It will also be sentential in some way. There is no correct abstract thought waiting to be expressed. There is only a better expression that will more accurately capture what it is the person wants to say. There is no archetype that is an abstract entity, although there may well be a perfect linguistic expression of what needs to be said. The thought or correct meaning is not, therefore, some kind of axial point around which the contingent expressions revolve. If anything needs to be matched, it is not an abstract proposition or meaning, it is the empirical things and relations that need to be linguistically expressed. And even if we allow abstract objects into our ontologies, such as numbers, logical relations, meanings and propositions, such things are still of a different order than these other abstract entities. Basically, we do not need meanings or propositions, but we may need numbers and abstract logical devices and relations (no doubt Derrida would have disagreed with this qualification too).
Believing that there is a better way of expressing something does not mean that there is a thought, meaning or proposition waiting somewhere to be expressed. That better way will also be linguistic, just as the meaning of one sentence is, in fact, another sentence expressed in a different way. The better expression is superior not because it captures the thought, meaning or proposition, but because it captures the truth about the world that it is expressing. If anything restricts and determines the correctness of the expression, it is the world itself, not thoughts, meanings or propositions, even if in that world we may well find numbers and logical relations. The thought or meaning essentially becomes the proxy for the reality of the empirical world. It is the world, not the meaning, that tells us whether or not we have expressed ourselves correctly or incorrectly. We put in place a fake intermediary between the world and ourselves. Why have a three-place relation that goes from subject, to meaning to world? Why not a two-place relation between utterance and world without a meaning as some kind of intermediary? This is not too dissimilar to Davidson’s position on sense-data or sensory stimulation. Traditional empiricists believed that we went from subject or utterance, to sense-data to world, whereas, in fact, it is a direct relation between subject and objects and events in the world. The world is not inferred from sense-data. We see the world immediately in terms of objects and events.
Meanings and propositions have been deemed necessary because they would secure certainty and absolute correctness. If we tune ourselves into the meaning and proposition, we will get things right. They help secure us from vicissitudes and uncertainties. They are the ‘centres’ (Derrida, 1967/1978) round which all contingent and empirical expressions gravitate. Without them we would be in a world of flux. It is the old Platonic worry. There must be a Justice or Truth in the sky, otherwise we will succumb to relativism and uncertainty and other such evils. But, in actual fact, there is nothing of the sort to grab hold of. All we have are the peripherals, the contingents, and the empiricals. Nothing more. Perhaps it would be nice to have ‘centres’ of various kinds. Only then could dispute and conflict be stopped. Meanings and propositions, like universals and forms, would supply us with the means to find the correct meaning of, say, ‘truth’ or ‘justice’, and propositions would give us something determinate and precise for our contingent statements to be about. Thus we could find the true and correct answers to numerous philosophical and non-philosophical problems. Or, more correctly, the Platonistic philosopher will have all the answers. And then, of course, he would supply the rest of us with the true answers to all our philosophical problems. These abstract entities would save us from ourselves. They were treated like gods in Platonic and Platonistic philosophy. They guaranteed truth, correctness and certainty. Without them, we wouldn’t have such things. We would continue to grope around in the dark without even a hint of security.
‘Writing’ does not simply transcribe things that already exists – meanings and propositions. Meaning can only exist as or in language – in the guise of language:
To write is to know that what has not yet been produced within literality has no other dwelling place, does not await us as prescription in some ‘topos ouranios’, or some divine understanding. Meaning must await being said or written in order to inhabit itself, and in order to become, by differing from itself, what it is: meaning. (Derrida, 1978)
Derrida then quotes Merleau-Ponty:
Communication in literature is not the simple appeal on the part of the writer to meanings which would be part of an a priori of the mind…The writer’s thought does not control his language from without; the writer is himself a kind of new idiom, constructing itself. “My own words take me by surprise and teach me what I think,” he said elsewhere.
He sums up:
Meaning is neither before nor after the act. It is not that which is called God, that which imprints every human course and recourse with its secondarity…
So there is not even a Platonic ‘Idea’ behind the words. The word or concept is the ‘Idea’.
Thus, the notion of an Idea or ‘interior design’ as simply anterior to a work which would supposedly be the expression of it, is a prejudice: a prejudice of the traditional criticism called idealist.
The conclusion to all this is clear. If there is nothing behind the word, and all language is a social creation, then Ideas, and the claims of western metaphysics generally, are stripped of their absolutist pretensions and shown to be merely contingent, relative and the products of history.
Reason
According to Norris, Derrida’s primary thesis was that there was a “ruling illusion of Western metaphysics”. What was that illusion?
that reason can somehow dispense with language and arrive at a pure, self-authenticating truth or method.
The very idea, if Derrida offered a correct diagnosis, of reason dispensing with language strikes many of us as very strange. How could that ever be the case? What would these thoughts be like that had somehow dispensed with language? I can easily imagine mental items, events and processes that do not in any way depend on language, but these rationalist philosophers talked in terms of, for example, ‘truth’ and ‘method’. These very concepts are thoroughly linguistic. Things indeed happen in the conscious mind that do not depend on language. However, none of these things have anything whatsoever to do with philosophical thought of any kind. If Derrida’s position is correct, then it must surely be the case that these rationalists were not only deluding us, but they may have been deluding themselves. I cannot imagine a single (philosophical) thought occurring in my mind that does not depend on language in some way. I can imagine certain things happening in my conscious mind that are not dependent on language, but none of these things would be thought, and certainly not philosophical thought. I suppose all this depends on what is meant by the word ‘thought’. For example, we can form a mental image of a lion in our minds. Not only that, but we can enlarge that mental image, etc. We can imagine many things. We can recall certain of our past experiences. We can play around quite a lot in our minds. Would any of these things be examples of thought? Of course, thoughts may accompany our mental images, our imaginary creations, or our memories, but these things are not themselves examples of thought. For example, we may form a mental image of Tony Blair and then think to ourselves that we note his Cheshire-cat smile. The mental image may have occurred before any accompanying thoughts. But the mental image without any thoughts would not itself be a thought. Again, it all depends on our definition of the concept [thought]. Perhaps the simple process of consciously enlarging a mental image is an example of a thought. It is some kind of mental process. And it is some kind of mental process that we are in control of. Perhaps that very control means that such processes are thoughts. Even if we accept this as an example of thought, it would not be a philosophical thought. It certainly wouldn’t be a thought about the nature of, say, truth or method. And it is philosophical rationalism that we are talking about here. It may even be the case that mental items and processes that are not themselves examples of thought accompany philosophical thought itself. There is no problem here. Sometimes, even all the time, when we think about Tony Blair we may conjure up a mental image of Tony Blair. But that mental image would be an accompaniment to thought, not an example of thought itself.
It is a strange thought that reason can somehow sustain itself without recourse to anything empirical or contingent. Somehow the rationalists saw reason as a kind of blank slate or even a tabula rasa – a tabula rasa of innate faculties. It was almost as if because certain things can indeed happen within the mind that do not depend on linguistic expression, that they believed that rational thought itself could do so. It was certainly the case that Descartes’ notion of thought was very large. It included both the imagination and the having of mental images. Again, cognitive processes must surely be something both more specific and narrower than Descartes’ notion of thought.
Even if mental images and certain mental processes were indeed non-linguistic, perhaps such things would not be the way they are if it weren’t for the influence of language. Take this example. Say that we formed an image of Tony Blair smiling his Cheshire-cat smile. Perhaps our image of Tony Blair could not be the way it is if it were it not for, say, hearing people say that Tony Blair ‘has a Cheshire-cat smile’. How could we say in the first place that ‘it is a mental image of Tony Blair with a Cheshire-cat smile’ if we had never heard such a phrase, or if we had never utilised the linguistic concepts [Cheshire cat], [smile] and even the individualised concept [Tony Blair]? We could have a mental image without these necessary prerequisites, but would it be a mental image of Tony Blair or of Tony Blair’s Cheshire-cat smile? If what I have said is correct, then perhaps our language in some way determines the very non-linguistic processes or items that occur in the mind. And if that is the case, then could we call them ‘non-linguistic’ items or processes at all? There may be things that are above and beyond language in the case of these mental items and processes, but it would still be the case that they are to some extent determined and coloured by linguistic processes and our language. Of course there are things that go on in the mind that are not words and linguistic expressions, whether vocal or sub-vocal. It may still be the case that everything that does occur in the mind is determined and coloured to some extent by our language and even by our linguistically-expressed knowledge.
For example, say that someone is totally obsessed by fame. Perhaps not everything that happens in this person’s mind will be a linguistic expression of his or her thirst for fame. That thirst may touch or even determine just about everything that occurs in his or her mind. And if this were the case with something as specific as the thirst for fame, then it’s much more likely to be the case if we are taking the influence of language generally on what goes on in minds. It should not really be a great surprise that something we use each and every day of lives, even if only to ourselves, should determine and colour everything that goes on in our minds in some way. Freeing oneself from language, a la the rationalists, must be as near to impossibility as one could get. Of course those human beings that have never been brought into contact with language, if that is at all possible, would be free from language. However, these people would not be rationalist philosophers! No other type of human being is more determined and affected by language than a philosopher. And if this is the case, then it would also be true about rationalist philosophers!
So even if mental imagery, etc., are somehow distinguishable from cognitive thought and linguistic expression, it may still be the case that language and linguistic concepts still affect the nature of these non-cognitive processes and items. We are not really talking about the general ability of mental processes and items to be free from language, but the philosophical thoughts of rationalist philosophers. Generally speaking, rationalist philosophers concerned themselves with things like truth, meaning and method, not ostensibly non-linguistic items like, say, mental images, etc.
Self-Authenticating Reason and Language
We can also ask why it was seen to be the case, by rationalists, that thought that is free from language was deemed as purer, ‘self-authenticating’ (Norris, 1991), and more likely to lead to genuine truth? If language is in a sense as natural as the mind, then perhaps we shouldn’t be so sceptical about the role of language in thought. Of course the rationalists did make a distinction between mind and language. The former was seen as non-natural, whereas the latter was seen as natural. And if it is natural, as the rationalists thought, then it would be contingent, unreliable, changeable, peripheral, incapable of discovering necessities and certainties, and to generally limit what could be done when not using it.
We may therefore ask ourselves why it was the rationalist philosophers were so sceptical about the affects of language on philosophical thought. I suppose it is because language is both an empirical phenomenon and it has a contingent status. Language would therefore pollute the purity of philosophical reasoning precisely because of its empirical and contingent nature. To the rationalists, mind was pure and language was polluted. Or, more correctly, mind cleansed of all empirical vicissitudes would be in a much better position to find the truth. In the ancient tradition, mind was deemed non-natural, whereas language, of course, was a natural phenomenon. And it was the non-natural, or even the transcendental, that was deemed to secure truth from empirical vicissitudes. The rationalist vision was still essentially Platonistic in nature. Lets us remember, therefore, that Plato believed that the best state that could be achieved by a philosopher was a state in which all the data from the senses were, to use Husserl’s term, ‘bracketed’. Only then could we gain access to the pure forms and essences that essentially bear no relation at all to anything remotely empirical and contingent.
The rationalist thesis included two important aspects:
1) That language and thought/meaning/concept are not only very different phenomena, but that they can in some way exist separately.
2) That reason can gain truth and knowledge without experiential or empirical help (e.g., without the help of a public language).
Because of 2), reason is therefore ‘self-authenticating’. That means that if reason relies on language in some way, then it could not be authentic. It would be relying, after all, on something that is not itself self-authenticating. Reason is one thing, and language is another. How can reason, that is, non-empirical and non-spatiotemporal reason, use or rely on something that is empirical and spatiotemporal? Wouldn’t this be like trying to put a square shape into a round hole? It parallels, to a certain extent, the relation between mind-substance, that is non-spatiotemporal and non-extended, and objects that are examples of a substance that are spatiotemporal and extended. How can we combine these seemingly incompatible substances? More precisely, how can mind-substance form representations, ideas, images, etc., of things that are examples of material-substance? How can these two different substances have a relation or connect in any way? However, even a substance dualist like Descartes accepted that there is some kind of relation or connection. The problem was explaining it. And so we have the parallel problem of explaining the relation of pure reason, which belongs to mind-substance, to something that belongs to a different ontological order – language and its empirical expressions.
The essential problem is that the mind was deemed as non-material, non-extended, etc., so how can the mind depend or even use something natural, like a language, to gain both truth and knowledge? These two worlds were therefore seemed as (largely) incompatible. They could not, on pain of error, cooperate in that search for truth and knowledge.
If the mind’s beliefs or ideas are ‘clear and distinct’, to use Descartes’ words, then it will know if they are true and also if it has knowledge. Also, a ‘self-authenticating’ method will grow out of these ‘clear and distinct’ thought processes and ideas. This is essentially a view of Cartesian rationalism. Whether it is also applicable to all Western metaphysics (as Derrida claimed) is of course another matter.
So Descartes talked in terms of ‘clear and distinct ideas’, but what are these things? Are they linguistic in any way? In a sense, they can’t be at all linguistic because language is an empirical and natural phenomenon, which was therefore something that Descartes would have guarded himself against. And, of course, this is precisely what he did do. If they are deemed linguistic in any way, then, almost by definition, they will not be ‘clear and distinct’. Nothing natural and empirical can be ‘clear and distinct’. Only the mind’s relation to itself could be ‘clear and distinct’. Incorporating language into thoughts about ‘clear and distinct ideas’ will automatically introduce something that is empirical, uncertain, unclear, indistinct, etc. However, Descartes formulated his views about clear and distinct ideas in an empirical and natural language. So how, in fact, did he square his circle and express the non-natural, certain and necessary in a public language that was none of these things?
Not only does such rationalism seemingly claim to dispense with language, it also “strives to efface its textual or ‘written’ character”. Clearly Derrida is emphasising the importance of language over the importance of thought. Or, more correctly, he may have claimed that there is no cognitive thought without language.
The rationalist pursuit was an attempted escape from all that was natural, empirical and uncertain. Could that concrete wall of language ever be broken through? Could the rationalists have ever ‘dispensed with language’ to enter that pure Platonic world of pure reason? What kind of states of mind did the rationalists achieve in such pure mental states? Did they ever actually truly liberate themselves from the vicissitudes and uncertainties of language?
Part Three
Thought, Speech and Writing
It is because of much of the ideas and positions advanced in this paper so far that Derrida was thought by many to praise literature over philosophy. He thought that ‘literary texts are less deluded than the discourse of philosophy’ (Norris, 1991). Why are they less ‘deluded’? Norris offers one example. Literature ‘implicitly acknowledge[s] and exploit[s] [its] own rhetorical status’. Of course, we would need to know exactly what is meant by the word ‘rhetorical’.
Part of the up-playing of reason and thought was the up-playing of speech. There was a three-way relation here:
Meaning (reason)
ß
speech
ß
texts
Speech was deemed to be closer, as it were, to meanings/concepts/thoughts, than written texts. To use Derrida’s own term, meaning was ‘present’ in speech, but much less so in writing. With writing, according to Derrida, we have the ‘free play’ of language. Derrida thought that this is a good thing. However, the rationalist metaphysicians believed that such ‘free play’ did not occur in pure speech and pure thought. Derrida’s general idea is that it occurs all the way down the line because texts or writing infect speech and thought too.
So it was thought that there is a more perfect match between meaning and speech than there is between meaning/thought and text. If we are lucky, however, perhaps writing could perfectly match speech, which, in turn, can perfectly match its meaning. Writing could be a ‘more or less faithful transcription of the elements of speech’. This was the hope in traditional metaphysics, according to Derrida.
This is Derrida himself on the subject, commenting on the importance of speech over writing:
This logocentrism, this epoch of the full speech, has always placed in parenthesis, suspended, and suppressed for essential reasons, all free reflection on the origin and status of writing. (Derrida, 1977a, )
Therefore speech, within metaphysics, had always denied its origin in writing.
Derrida stressed the notion of ‘meaning as presence’. Derrida’s talked in terms of the ‘degradation of writing’ and the purported importance of achieving a pure, writing-free meaning:
reason looks for a ground or authenticating method immune to the snares of textuality. (Norris, 1991)
And what did the rationalist metaphysicians believe pure reason could offer? This:
If meaning could only attain to a state of self-sufficient intelligibility, language would no longer present any problem but serve as an obedient vehicle of thought. (Norris, 1991)
This ‘self-sufficient intelligibility’ would be similar to the ‘clear and distinct’ ideas of Descartes: it would be transparent and true. And, we might add, transparent and true because God, or Reason, is not a ‘deceiver’. With pure reason, speech, unlike writing, would be ‘obedient’ to thought, meaning and concept.
Again, why was this importance given to speech? It was because it was thought that the “speaker’s ‘presence’ [is] behind the words”. This was not so much the case, if at all, in the case of texts or writing generally. Thus, speech was given a ‘privileged status’ (i.e., parole). Even a structuralist like Barthes stressed the importance of speech over writing. Here’s a quote from the French structural linguist:
… a language is possible only starting from speech; historically, speech phenomena always precede language phenomena (it is speech which makes language evolve)… (Barthes, 1967)
Speech or “[v]oice becomes a metaphor of truth and authenticity” (Norris, 1991). It is authentic because the speaker is ‘self-present’ and therefore the meaning must also be ‘self-present’ (at least to some degree). Writing, to the contrary, is ‘lifeless’. It is lifeless because it is one further removed from meaning/concept/thought than is speech. When we speak, this is what was supposed to happen:
one is able to experience…an intimate link between sound [i.e. speech] and sense [i.e. meaning], an inward and immediate realization of meaning which yields itself up without reserve to perfect, transparent understanding. (Norris, 1991)
The voice is closer to its meaning/concept/thought. It can fully grasp its meaning/concept/thought. Therefore meaning/concept/thought will be transparent to the speaker. Or, to use Descartes’ terms, meanings/thoughts/concepts are ‘clear and distinct’ to the speaker. The case with writing was deemed to be very different. Writing
destroys this ideal of pure self-presence. It obtrudes an alien, depersonalised medium, between utterance and understanding. (Norris, 1991)
Not only that, but “[i]t occupies a promiscuous public realm where authority is sacrificed to the vagaries and whims of textual ‘dissemination’ ”.
Not only is writing once removed from speech, it can be also once removed from the author of that writing because of public ‘dissemination’ in, for example, books, libraries, etc. This means that writing severs the close link between utterer and his meaning/concept/thought (i.e. ‘self-presence’). It becomes corrupted and further and further removed from the truth:
… The reading must always aim at a certain relationship, unperceived by the writer, between what he commands and what he does not command of the patterns of the language that he uses… (Derrida, 1977a)
This undecidability or indeterminateness was traditionally thought to be absent during speech-acts or utterances. The speaker is master of his concepts or his meanings. He fully determines what means what. Writing, on the other hand, endlessly displaces meaning, to paraphrase Norris. More than that, writing ‘governs language and places it for ever beyond the reach of a stable, self-authenticating knowledge’ (Norris, 1991).
However, according to Derrida, this writing-speaking dualism is essentially bogus. Oral language is also and always infected with writing anyway. And, we should say, writing is also always infected with oral language. Derrida called oral language ‘generalized writing’. The belief that meaning is fully present he called the ‘metaphysics of presence’.
References and Further Reading
Austin. J. L. (1963) How To Do Things With Words
Barthes, R. (1967) Elements of Semiology, trans. A. Lavers and C. Smith
Berkeley, G. (1713) Three Dialogues Between Hylas and Philonous, various editions
Davidson, D. (1989) ‘A Coherence Theory of Truth and Knowledge’ in Truth and Interpretation:
Perspectives on the Philosophy of Donald Davidson, ed. E. LePore
Derrida, J. (1972a) La Dissemination
(1972b) Marges de la philosophie
(1973) Speech and Phenomena, and Other Essays on Husserl’s Theory of Signs, trans. D.
Allison
(1977) Of Grammatology, trans. G.C. Spivak
(1978) Writing and Difference, trans. A. Bass
Descartes, R. (1637/1641) A Discourse on Method and Meditations and Principles, various editions
Hawkes, T. (1977) Structuralism and Semiotics
Heidegger, M. (1927) Being and Time, various editions
Husserl, E. (1970) The Crisis of the European Sciences and Transcendental Phenomenology, trans. D.
Carr
Kant, I. (1787) A Critique of Pure Reason, various editions
Norris, C. (1991) Deconstruction: Theory and Practice (revised edition)
Popper, K. (1963) Conjectures and Refutations
(1945) The Open Society and its Enemies: 2 Vols. I. Plato, II. Hegel and Marx. (Volume 1 is particularly relevant to Derrida’s own position on Plato and his notion of a metaphysical ‘centre’. However, Popper’s take on Hegel, in Vol. II, should also be of particular interest to, say, the Continental Derridarean, if not to the Anglo-American analytic philosopher.)
Quine, W.V.O. (1953) From a Logical Point of View
(1969) Ontological Relativity and Other Essays
Rorty, R. (1977) ‘Derrida on Language, Being and Abnormal Philosophy’ in the Jo
