The Language Myth in Western Culture

Roy Harris

• Publisher: Routledge
• Year: 2013
• Topic: Languages & Linguistics
• eBook ISBN:
  9781136751455
• Subtopic: Linguistics
• Index: Languages & Linguistics

The basic claim of this book is that for 2000 years and more the western tradition has relied on two very dubious assumptions about human communication: that each national language is a unique code and that linguistic communication consists in the utilization of such codes to transfer messages from mind to mind.

1 The Role of the Language Myth in the Western Cultural Tradition

Roy Harris

1. Introduction

The title of my paper might already seem to invoke so many question-begging terms as to call for some preliminary explanations. What is myth? What is culture? What is tradition?

As an integrationist, I may perhaps be allowed a certain measure of diffidence about definitions. I shall simply take it that we are dealing with a cultural tradition wherever a society has achieved a sufficient degree of self-awareness to discuss its own beliefs and practices by reference to authorities long since dead and to ideas which those predecessors held, or are alleged to have held. Gandhi was once asked what he thought about Western civilization and replied that he thought it would be a good idea. But I doubt that even Gandhi would have refused to acknowledge the existence of a Western cultural tradition, however civilized or uncivilized.

Nor do I suppose he would have had any more difficulty than I do with the notion that a cultural tradition may incorporate myths. But ‘What is a myth?’ is a much trickier question than ‘What is a cultural tradition?’ For my purposes a myth is a cultural fossil, a sedimented form of thinking that has gone unchallenged for so long that it has hardened into a kind of intellectual concrete.

My contention is that there is, and long has been, a language myth deeply entrenched in Western culture. The origins of this myth can be traced back over two millennia and more to the Classical period of ancient Greece. It is a myth of which we are all today – whether we regard ourselves personally as ‘Western’ or not – both beneficiaries and victims. We are beneficiaries insofar as the myth promotes certain values and practices that Western culture would be the poorer without. We are victims inasmuch as the myth inculcates certain attitudes and prejudices that prevent Western culture from realizing its full human and humane potential. For both kinds of reason, positive and negative, a fuller understanding of the way thinking about language has become fossilized in the Western tradition is worth serious attention.

2. The Language Myth

The myth I am referring to appears in various versions at different times throughout that long tradition; but, allowing for minor differences of detail, it usually takes something like the following form:

Individuals are able to exchange their thoughts by means of words because – and insofar as – they have come to understand and to adhere to a fixed public plan for doing so. The plan is based on recurrent instantiation of invariant items belonging to a set known to all members of the community. These items are the ‘sentences’ of the community’s language. They are invariant items in two respects: form and meaning. Knowing the forms of sentences enables those who know the language to express appropriately the thoughts they intend to convey. Knowing the meanings of sentences enables those who know the language to identify the thoughts thus expressed. Being invariant, sentences are context-free, and so proof against the vagaries of changing speakers, hearers and circumstances, rather as coin of the realm is valid irrespective of the honesty or dishonesty of individual transactions.

This preliminary characterization was first proposed nearly twenty years ago in a book I called The Language Myth (Harris 1981). Today I see little reason to alter that initial account in any substantive detail. The book then goes on to explain as follows how linguistic communication is supposed to work.

Suppose A has a thought that he wishes to communicate to B, for example, that gold is valuable. His task is to search among the sentences of a language known both to himself and to B, and select that sentence which has a meaning appropriate to the thought to be conveyed; for example, in English, the sentence Gold is valuable. He then encodes this sentence in its appropriate oral or written form, from which B is able to decode it, and in virtue of knowing what it means, grasp the thought which A intended to convey to him, namely that gold is valuable. (Harris 1981: 10)

This way of thinking about language seems to me to have become fossilized at a very early date in the Western tradition; fossilized in the sense not only that it is accepted by those who write authoritatively on linguistic topics, but also in that any challenge to it is regarded as an affront to common sense.

If you present the language myth by reference to examples like Gold is valuable, then sooner or later – and sooner rather than later, in my experience – someone is bound to ask: ‘Is this a myth at all?’ Is it not, rather, a very reasonable – albeit condensed – account of what actually happens in our daily verbal transactions? On what other basis could we in practice conduct our affairs? Do we not in fact resort to saying, for example, ‘Gold is valuable’ when we want to tell someone just that; namely that gold is valuable? How else could we proceed?

There may be those who regard some components or versions of the language myth as more mythical than others (Carr, this volume). Sceptics might even feel inclined to argue that if the language myth is a myth, then a conference such as this is self-stultifying, since it both presupposes and practices successfully that very form of communication declared to be mythical. But it seems to me that even if integrationists are misguided in their characterization of the myth, the mere possibility that they might be right, or even partly right, is already enough to pose questions about Western culture that cannot just be dismissed out of hand.

My original choice of the term myth was provocative, and what I hoped to provoke was a debate among linguists concerning the theoretical basis of their discipline. It seemed to me, in the late 1970s, that not since the articulation of the Sapir-Whorf hypothesis had there been any serious challenge to the monumental complacency of mainstream linguistics. My attempt to provoke a debate was not very successful; but its failure taught me two things. By their reactions, many linguists seemed unwittingly to confirm the mythological character of their own assumptions. For instead of trying to meet the challenge head-on and defend those assumptions, what they typically did was simply deny that they believed in the myth (just as they had previously denied their linguistics was Whorfian) and then carry on exactly as before. Which struck me as rather like denying one’s belief in the Virgin Birth and immediately lighting another candle to Holy Mary. It was almost as if the denial somehow strengthened the original act of faith. And in retrospect I saw that is very characteristic of myth. Myths are not just common-or- garden illusions that can be empirically disproved and discredited. Nor, by the same token, can they be upgraded into ‘science’ overnight by dint of producing research that allegedly ‘supports’ them.

Anyone is doubly confused who supposes that findings from modern psycholinguistic experiments demonstrate conclusively the non-mythical character of the basic tenets of traditional Western thinking about language. That would be rather like concluding that there was no Oedipus myth after all, because research has lately discovered evidence of a historical person called ‘Oedipus’. Trying to prove, one by one, the ‘real existence’ of words, meanings, sentences, consonants, vowels, dialects and all the other familiar bits and pieces of the mythical linguistic apparatus would be as Quixotic a conceptual enterprise as trying to prove the real existence of square roots, odd numbers, quotients and divisors (D. Davis, this volume) in order to justify what we do with elementary arithmetic; or establish the actuality of a certain state of my brain in order to verify the truth of my statement ‘I believe that my name is Roy Harris’ (Taylor, this volume). Linguists committed to analogous validation programmes in respect of linguistic elements create their own problems, for they have taken on a task which is as pointless as it is impracticable. (‘Cantonese is a bit of a worry, you know: we haven’t yet got irrefutable evidence about whether there really is a seventh tone.’) Even more curious would be trying to prove that these really existent linguistic entities are ‘really’ subject to change over time (Love, this volume), as if showing that the ‘real’ Oedipus grew old, or eventually lost his teeth, corroborated the claim that the mythical Oedipus was not, after all, a mythical figure. The irony in the linguistic case is that these moves designed to ‘validate’ belief in a myth by denying its mythical status are typically surrogational moves: they show us the language myth in the process of trying to gain credit by offering its own terminological IOUs as security. (‘A syllable, sir? Yes, I can show you one right here on my spectrogram.’ ‘A language? I have an excellent description of one in this grammar book.’ ‘A linguistic change? How about Latin causa becoming French chose? It’s vouched for in the very latest etymological dictionaries!’) Here is the mythological term: there the alleged item it stands for. Or, failing the item itself, a ‘representation’ of it. What more is needed? What more could be required, if you already believe in a surrogational semantics? Anything lost on the representational swings can always be recouped on the representational roundabouts. What this sad state of affairs reflects is a culture whose scholars are still ‘bewitched’ (as Wittgenstein once put it) by language, not least those whose scholarship is supposedly focussed on language itself.

The engagement of myth with reality – or its failure to engage – is of a quite different order from that of illusion or factual error. Once you recognize an illusion as illusory, or a factual error as erroneous, you are no longer taken in by it; but myths can command credence and respect even from those who recognize, however reluctantly, their essentially mythical character. Myths are often regarded as capturing some higher or symbolic truth, which transcends their superficial lack of factuality, or makes it irrelevant. They are held to have what the Greeks called hyponoia.

Myths also have a popular appeal as explaining certain things that would otherwise be mysterious. Lévi-Strauss once described the function of myth as being ‘to provide a logical model capable of overcoming a contradiction’ (Lévi-Strauss 1963: 229). He added: ‘an impossible achievement if, as it happens, the contradiction is real’. This seems to me to apply exactly to the myth we are considering. (I shall return to the question of ‘overcoming a contradiction’ shortly.) A myth, then, unlike illusion or factual error, characteristically takes a form in which there is no straightforward way of testing its credentials. That, indeed, is one source of its power as a myth. On all these counts what I call ‘the language myth’ qualifies for inclusion. Nor do I think it can be excluded on the ground that it lacks the personifications and narratives in which myth is often clothed. For such trappings do indeed feature in some Western accounts that allegorize parts of the myth in question (e.g. the nomenclatorial exploits of Adam in the Garden of Eden, and the no less hypothetical activities of the primordial ‘Namegiver’ invoked by Socrates, not to mention those attributed to Hermes Trismegistus).

Now it is readily understandable that linguists should not take kindly to being told that what they are presenting to their students as up-to-date science is actually no more than recycled myth. Anyone bearing that message is not likely to be very popular initially, any more than Copernicus was popular among the orthodox astronomers of his day. But the question then becomes whether one prefers popularity with one’s colleagues to the advancement of serious intellectual discussion. From reactions to my book I also learned something that I had only dimly realized before. When you challenge the kind of assumptions that I had identified as the ‘language myth’, you are challenging much more than people’s beliefs about language. That is why they refuse to engage in debate, or reject the terms of debate out of hand, or say you are absurdly misrepresenting their position. For you are calling in question not simply a set of propositions of the kind that turn up in grammar books or works of linguistic theory. You are calling in question something about the way they conduct themselves not just in their professional capacity as linguists, but also as educated persons, law-abiding citizens, good neighbours, and in many other aspects of their social being. That is what offends them. You are challenging a whole cultural picture at a much more basic level than you might at first have supposed.

3. The Communication Myth

In recent years I have come to wonder whether the role of the language myth in the Western tradition was ever simply or mainly to justify certain ways of talking about words, sentences, nouns, verbs, and so on. In this paper I shall propose that it is possible to see the role of the language myth as being to interrelate and underpin certain wider assumptions concerning both the place of human beings in society and human knowledge of the natural world. From this perspective, many different facets of the intellectual development of Western culture can be viewed in terms of the adoption of, or adaptation to, the two essential components of the language myth: (i) the doctrine of telementation and (ii) the doctrine of the fixed code.

The language myth, as I have characterized it so far, can be viewed as just one version of a more general myth about communication. In its more general form, the myth would run as follows:

Certain forms of communication involve a process of transmitting messages. Individuals are able to send and/or interpret messages whenever they have come to understand and follow the relevant procedure of transmission (public or private, voluntary or involuntary, natural or artificial). This is based on recurrent instantiation of certain invariant items. These items are ‘signs’. They are invariant items in two respects: form and meaning. Knowing the form and meaning of a sign enables one to identify and interpret the message it conveys.

Here a ‘sign’ can take any physical form whatever, verbal or non-verbal. It can include signs emanating from the natural world as well as signs emanating from human agents. Let us call this broader version of the language myth ‘the communication myth’. As I have formulated it, it makes no claim to cover all forms of communication. Exactly what forms of communication it does cover can be left as an open question; and that is one of its important features. Because of this open-endedness, it was possible throughout the Western tradition to treat it as providing a general framework for theorizing about quite diverse kinds of activity. Treating as many activities and relationships as possible in terms of communication is, as I now see it, one of the most conspicuous features of that tradition.

4. The Transmission Process

The explanatory potential of any myth depends on certain basic features of its internal structure. In the case we are considering, the transmission of a message is represented as linear (in one sense of that debatable term) in that it involves a non-reversible process in time. But in the case of language this process is conceived of as linking at least two matching items. Starting from the thought that ‘Gold is valuable’ in A’s head, we proceed via the utterance of the sentence to a matching thought in B’s head. As a result of the oneway transmission, we get a second pattern that, although spatio-temporally discrete, copies the original. Thus in its simplest and most abstract form the transmission has a structure like this:

• > •

For want of any better term I will call this a ‘pattern transference’ model. The original pattern has somehow been transferred to or recreated in a new location.

The first point I want to make is that if we wish to understand the language myth we must realize that it operates with a pattern transference model; and the second point, no less important, is that pattern transference itself is not a hypothesis but a natural phenomenon. A footprint in wet sand, of the kind that Robinson Crusoe one day discovered on his island shore, leaves a pattern corresponding to the configuration of the foot that made it. The recognition of pattern transference as a natural phenomenon is of enormous value to hunter-gatherer communities. It is important to be able to recognize the difference between the tracks of a predator and the tracks of various species of game. How primitive societies conceptualized the difference between such ‘messages’ accidentally left by passing animals and the messages deliberately projected by human agents, or whether they recognized it at all, we do not know. What we do know is that some such distinction must have occurred to the precursors of Plato, who already takes it for granted that verbal messages are signs of a different kind from natural signs. But in what exactly the difference consists he is not entirely sure.

Pattern transference, it hardly needs to be stressed, is quite fundamental to traditional Western conceptions of drawing and painting. A portrait of Queen Victoria is expected (or would have been expected in her day) to ‘look like’ Queen Victoria. A ‘flower painting’ would have been expected to depict things visually recognizable as flowers. There was a time – but perhaps it is now past – when any apprentice painter would have been laughed out of the studio if incapable of producing a plausible ‘likeness’. In such cases, a pattern originating on the retina of the artist’s eye is deemed to have been transferred, via the medium of paint on canvas, to the retina of the person looking at the picture. This is analogous to the transference that, according to language-myth accounts, allegedly takes place via the medium of sound in the case of words. Mimesis, as traditionally interpreted from Plato onwards, is a semiological concept underpinned, however vaguely, by pattern transference. Just how ubiquitously this pervades Western drawing and painting is evident from the way in which it overrides any biomechanical controls on ‘likeness’. Thus, for example, it is – or was – supposedly possible for a competent painter to depict the archangel Gabriel, or the patriarch Noah, without ever having set eyes either on the heavenly messenger or ...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright
• Contents
• Contributors
• Preface
• 1. The Role of the Language Myth in the Western Cultural Tradition
• 2. The Language Myth and Historical Linguistics
• 3. The Language Myth and Standard English
• 4. The Language Myth and Linguistics Humanised
• 5. The Mythical, the Non-mythical and Representation in Linguistics
• 6. Folk Psychology and the Language Myth: What Would the Integrationist Say?
• 7. The Language Myth and the Race Myth: Evil Twins of Modern Identity Politics?
• 8. The Language Myth and Mathematical Notation as a Language of Nature
• 9. The Language Myth and the Law
• 10. The Language Myth and Western Art
• 11. The Language Myth, Schopenhauer and Music
• Index