Professional Documents
Culture Documents
Amorey Gethin - Antilinguistics - A Critical Assessment of Modern Linguistic Theory and Practice - Intellect Books (1990)
Amorey Gethin - Antilinguistics - A Critical Assessment of Modern Linguistic Theory and Practice - Intellect Books (1990)
Amorey Gethin - Antilinguistics - A Critical Assessment of Modern Linguistic Theory and Practice - Intellect Books (1990)
Antilinguistics
A Critical Assessment of Modern Linguistic Theory and Practice
Amorey Gethin
Contents
Acknowledgements iii
Introduction 1
Chapter 1
What Language is 5
Chapter 2
The Emptiness of Analysis 11
Chapter 3
Chomskyan Mistakes Made Plain 40
Chapter 4
Unapplied Linguistics 60
Chapter 5
The Fantasy of Structure 93
Chapter 6
More Fantasy of Structure 114
Chapter 7
Ungenerative Grammar 131
Chapter 8
Deep Confusion 151
Chapter 9
Being Able to Use Language 170
Chapter 10
Thinking and Language 194
Chapter 11
Language the Corrupter 220
Chapter 12
Trying to Speak the Truth 242
Chapter 13
Slavery to Authority and the Word 254
Bibliography 260
Index 263
Acknowledgements
I give my warm thanks to the former colleagues, and other friends, with whom over so many years I have spent so many
stimulating hours debating language. Among these I would particularly like to mention Michael McCarthy and David Bond,
who were unfailing in their sympathetic encouragement of my urge to question. David I thank too, very specially, for always
thinking of me whenever he came across something he thought might interest or provoke me.
As much as anybody I thank my son, Terence Gethin, for all the care he devoted to his judgements on the manuscript, and for
his objective criticism, advice, and rationality.
And without the constant support and concern of my wife, Mieko Suzuki-Gethin, I should never have completed the task.
I owe a great debt to John Lennox Cook for providing, over many years, a place of work without fashionable preconceptions
where I could think freely about language.
My gratitude also goes to Erich Müller and Irene Marti, to Agnese, Arianna, Edo, Ida, Luca and Silio Masi, to Paul Mlynek, to
Gerhard Prawda, Ingrid Freidl, Felix, and Rea, and to Giorgia, Metello and Paola Tacconi, for making life so good for us during
what were perhaps the most important stages of writing this book.
I dedicate this book to Pod and to the memory of Tora and Bobi.
AMOREY GETHIN
SASSO PISANO
JUNE 1989
Introduction
For thousands of years it was the priests of religion who were the experts not only on the nature of the universe but also on the
nature of man. So it was largely accepted that it was the priests, too, who should tell humans how they should act.
Over the greater part of the world that has changed. The new experts on the nature of humankind and on human needs are the
'social scientists' of psychology, sociology, economics, ethology, linguistics. They may disagree among themselves, just as the
priests did. But just as the priests' authority was almost universally respected whatever they said at different times and in
different places, so the 'social scientists' are widely accepted as the only proper source of understanding, their way of
discovering truth as the only possible way.
From the rule of 'infallible' science there seems to be no appeal. One can reject the old dogmas. They were based on mere
superstition and wishful thinking. But the new, those one must accept, because they are based on evidence. They are scientific.
And everything scientific is true. 'Old' religions normally only claim revelation and faith as authority. Rebels can, at worst,
claim a different revelation, a different faith. They can even appeal to reason. But that which claims to be science is an even
more unbending master than intolerant religion, for it is knowledge, the final truth, and from that there can be no appeal. So the
new religion of Marxism, for example, or the new discovery of linguistics, had to be the truth, because they are based on
scientific analysis.
Today the 'scientific' analysers are the academics. Truth comes from them, practically only from them. They are the
professionals who know what they are talking about. So if there is debate and questioning it is debate and questioning that goes
on almost entirely within that circle of academic experts. Outsiders are untrained and so unfit to have opinions worth listening to
- so practically nobody outside ever tries to interfere, for any who do will almost certainly not have any notice taken of them. In
fact, I think very few outside the universities even dream of speaking up, because they feel they would have to do so on the
academics' own terms, and so would make amateurish fools of themselves.
Yet the power of the half-secret lores of the academic world needs to be questioned. Those lores capture the minds of nearly all
those who come in contact with them, those who are considered the most intelligent members of the community, and so those
who to a large extent are given the power to design how our lives should be, now and in the future.
1
What Language Is
It is not my purpose here to propound yet another 'theory' of language. I want only to affirm what I see as the essential principle
of languages. I want to encourage 'ordinary' people, those unconditioned by philosophy or linguistics or psychology, or their
education in general, to trust in their own sense of how languages work. And I think that sense is of something much as I
propose here. So this is no exact and worked out analysis of language, no definition, no classification of parts. It is only an
invitation to recognize the practical reality of languages as humans actually experience them; and an appeal to any wavering
initiates or novices in linguistics to start again from the beginning without preconceptions.
Languages are sets of meanings.1 Those meanings are an attempt to reflect human experience of reality. That reality, of course,
includes human thoughts and feelings and fantasies as well as what 'objectively' exists and happens.
Languages are only an attempt to reflect experience. As I shall try to show later in the book, they cannot do this properly.
But as they do try to reflect life and the world, the various meanings that compose them fit together in accordance with the logic
of life, not in accordance with some abstract, formal logic. Thus the basic principle for the working of language is very simple,
and that is perhaps why almost all humans master it so easily.2 But the detailed relationships of the meanings of a language are
very complicated, because they try to reflect the even more complicated relationships of life. The only true description of a
language is the language itself. Anything else is just a game.
Those human thoughts and feelings expressed in the meanings of a language include varying human reactions to, and the
varying angles from which humans see what exists and what happens. In English, for instance,
1 I have been to London.
and
1 Some of those of a philosophical bent may complain straight away, "What does he mean by meaning?" Yet the
philosophically uninitiated will have little trouble with the term. I think they will know almost exactly what I am talking
about. And it is not by definition that the sticklers for the discipline of definition understand their own word "mean" when
they demand "What do you mean Coy meaning)?" If they did not already know they could not ask the question.
"Meaning" certainly has more than one meaning. But when people start arguing about the meaning of meaning there is
immediately a prime example of the inadequacy of words, of the futile slavery to language itself that I criticize in
Chapters 11 and 12.
2 But see Chapter 9.
2
The Emptiness of Analysis
I can see three important purposes in thinking about language.
1 For me by far the most important is that it may help us to understand the relationship between thought and language, how
language influences the way people think and the way they act.
2 Understanding the true nature of language, or at least what language is not, is essential for discovering the processes of the
brain in controlling language.
3 That may in turn help physicians and others to help people who have defects (as a result of birth or accident) that make it
difficult or impossible to use language normally.
On the other hand I do not think it is likely that knowing about the way the brain controls language is going to help in
understanding the effects of language and thought on each other, any more than, say, knowing how hearing works is likely to
help give us better understanding of the psychological effects of music.
There are other purposes in studying language. Language studies can be interesting or entertaining, but that does not make them
important. Some people, for instance, search for the origin of language. This is a very interesting question, but it is not
important if it does not serve any of the three purposes I have mentioned first. Some people are fascinated by quirkish details
and accidents of language, but study of these has no more right to pretensions than the collecting of match box labels. Such
study may be entertaining but it is not important.
Modern linguists, though, are obsessed with describing language and languages.
Description of a language is of course very reasonable, and useful, if it is for people who have no opportunity to go to the
country whose language they want to learn, or who want to learn something of the language before they go; or if there is a risk
of a language dying out before it is recorded in some way; and if by description one only means an account of the characteristics
of the language that are different from other languages, together with a bilingual dictionary.
But the linguists want to show themselves very scientific, and so make their descriptions sound impressive by using phrases
such as "mathematical precision" or the "rules that generate all but only the
Figure 1
But as any one sign within a sentence has both a syntagmatic and a paradigmatic role (the function of he, for instance,
being partly a matter of its relationship with the other personal pronouns, and partly its relationships with the other words
which follow it), it is perhaps wise to avoid talk of dichotomies at this point, with all their implications of mutual
elusiveness [sic]. Both kinds of relationship are necessary to carry out the complete analysis of any sentence." (David
Crystal, Linguistics, 1981, pp. 165-66)
Then, for instance, there is 'tagmemics':"
The tagmeme...tries to combine into one conceptual unit two ideas which had previously been quite disassociated, the
ideas of class and function. When you analyse a sentence in Pike's terms, you do not end up with two independent
statements, one telling you what the bits of a sentence are and the classes they belong to, the other telling you the
functions they perform (such as would be illustrated by the analysis of the sentence John kicked the ball as 'Noun phrase +
Verb + Noun phrase' and 'Subject + Predicate + Object' respectively). Rather the sentence is analysed into a sequence of
tagmemes, each of which simultaneously provides information about an item's function in a larger structure, and about the
class to which it belongs that could also fulfil that function. A metaphor that is regularly used here is to talk of a structure
as a series of 'slots' into which various types of 'filler' can go:
Figure 2
This is a very clear example of how linguists first invent a purely artificial framework and then fit language into it. They think
up an analytical pattern and impose it on language. It is an activity the opposite of looking at the nature of language and letting it
lead them to the appropriate conclusions. It is just one more exercise in analytical game playing. As the analytical system comes
before what is to be analysed it is, in principle at least, even more barren than old fashioned grammatical parsing.
This trick of inventing the system first and squeezing language into it explains how it has been possible for the linguists to
sprout so many different contraptions out of thin air and keep themselves busy.
''In order to fully understand the meaning of something, we need to know what it does not mean as well as what it does
mean. To take an example from the level of lexis, the lexical item animal has some meaning even if it occurs on its own
out of any particular context. But we do not know exactly what it means until we know whether it means 'not vegetable or
mineral' or 'not human' or 'not bird or reptile or fish'. By naming a meaning and by assigning it to a system, where it is
mutually exclusive with other named meanings, we are able to show both something of what it does include and also what
it does not
Figure 3
There are what they call 'phrase-structure rules'. The examples below are from an earlier but not basically different Chomskyan
system. They correspond to the 'Base Component' in the diagram above and they are the basis of the famous Chomskyan 'deep
structure'.1 An arrow indicates that the element to its left is to be rewritten as appears to its right. NP=noun phrase, VP=verb
phrase, sing=singular, pl=plural, T=article, N=noun, Aux=auxiliary verb, V=verb, M=modal verb. The elements in brackets in
rule (10) are 'optional'. The brackets of (3), (7), (9) and (11) mean that any of the elements contained inside them may be used.
1 See Chapter 8.
Figure 4
They apply a 'transformational rule' to the 'underlying string', thus:
with its 'associated phrase marker': (The question mark represents a little problem they had previously; they claim to have dealt
with that now.)
1 See Chapter 8.
Figure 5
After applying some more 'transformational' rules they finally arrive at a sentence - 'surface structure':
The bird may have been eaten by the cat.
And here is an example of writing which for me sums up and shows up the pointlessness of it all:
"...an important theoretical problem. The derived string produced by one transformational rule may serve as the underlying
string for the operation of a subsequent transformational rule, and will therefore need to have associated with it the
appropriate derived phrase marker. Chomsky and his followers have worked on this problem and tried to establish a set of
conventions according to which a particularkind of formal operation (e.g. deletion, permutation or substitution) is defined
to have a particular effect upon the topology of the phrase marker it transforms; and we followed these conventions when
we decided that the effect of the rule B + D + E -> E + B operating upon the underlying phrase marker shown in Figure
x was the derived phrase marker in Figure y. But this was a very simple example from the point of view of the operations
involved and the shape of the phrase marker to which they applied; and furthermore it was a purely abstract example
unaffected by any empirical considerations. The reader will appreciate that the situation is very different when it comes to
formulating a set of
where the bracket ( ) again means that the element is optional, and S=sentence, Art=article, Adj=adjective, PP=prepositional
phrase, and P=preposition. But they say no modification of this rule can indicate that 9-11 are possible grammatical sentences,
because if we allow for the optional occurrence of the quantifier Q ("all" in this case) before each of the verbs in 8, 9-11 can be
produced, but so can the ungrammatical
where X is a variable, which can stand for anything at all, or nothing. The effect on the phrase marker is shown by the dotted
arrow. (I have used a simpler example than theirs here, in order to avoid irrelevant complications.)
Figure 6
Here one sees not only how trivial but also how dangerously misleading the urge to describe is. When the Chomskyan linguists
get to the stage I have just reported, surely it is obvious that they have completely forgotten about real language and real
psychology. At this stage they find that they cannot fit all the undoubted outward facts of language into their formal system, and
now all they are aware of is the need to expand their method of expressing these facts on a piece of paper. When they have
1 I follow the linguists' convention of marking ungrammatical sentences with an *
3
Chomskyan Mistakes Made Plain
The Chomskyan explanations of language - which effectively means explanations of sentences - have several different kinds of
flaw. There is nearly always a refusal to recognize the part played by meaning, and sometimes this is taken to a perverse
extreme. Often the explanations for a given grammatical fact are just wrong, again usually because the authors take no account
of meaning but obsessively follow their usual abstract track. Sometimes even the data on which they base their arguments are
simply false too. Then there are the cases where the confusion arises mainly because the author has muddled up two different
senses of the same word. A basic mistake is where the linguists begin by talking about one meaning, and then suddenly switch
to a quite unconnected meaning, but argue as if they were still talking about the same original meaning. I shall show in more
detail in Chapter 6 how keen Chomsky himself is on this. The approach is sometimes carried even further, so that the author
begins to produce utter nonsense language in support of his theory.
Many of the Chomskyans' accounts of sentences are presented as part of their argument that there is a universal grammar that all
humans are born with a knowledge of.1 Jeremy Campbell writes in Grammatical man (1982, pp.176-77):
"Other kinds of word relations are also forbidden for no apparent reason other than that they would violate principles
which are peculiar to the human mind, and anchored in the human genes. A more recent and sophisticated example of this
phenomenon is the embedded clause in an indirect sentence, where sometimes part of the clause can be missing, and the
sentence still be correct. One can say, for example,
1 It is unclear what to do.
where the subject of the embedded clause, "what to do," is absent. If the subject were present, the sentence would read
2 It is unclear what someone does.
If the subject were simply dropped to make the shortened form, the sentence would become
3 It is unclear what does.
1 See Chapter 9.
4
Unapplied Linguistics
One of the great illusions that modern linguists appear to have about their work is that it has led to a more flexible and generous
approach to the forms of language used by others, a less narrow view of what is 'correct' in grammar, expression and accent. But
in fact the only knowledge needed as the basis of a tolerant attitude is of the most elementary kind possessed by practically
everybody: nobody of any 'class' speaks or writes as people speaking the 'same' language did three or four or more centuries ago.
One needs only a very small amount of realism and logic to see that that means language changes and so there cannot be one
fixed 'right' form, or that, for example, there is no reason why conventions of spoken language should be inferior to those of
written language. There is one simple key to greater tolerance towards other people's language, and that is greater tolerance! It is
ironical, in view of the linguists' assertions, that language snobs and dictators do and can only condemn 'deviant' usage precisely
because they know about it. For linguists to claim the credit for an improvement in attitudes is irrational vanity. The only
antidote to authoritarian and exclusive attitudes is non-authoritarian and anti-exclusive attitudes. 'Science' can have nothing to
do with it. Marxism, I think, provides an instructive parallel. Marxists claim to possess a scientific political theory. I know
personally some very gentle and very tolerant Marxists, and there are certainly many such. But for all the 'scientific' knowledge
they claim, and in many cases probably precisely because of their pride in it, Marxists have been responsible for some of the
most intolerant savageries of this century. As for a broad-minded and respectful approach to other people's 'different' language, I
suggest that many came to such an attitude long before they knew anything of modern linguistics.
PhD's do not like such ideas. They prefer to think all progress, theoretical or practical, results from research. Indeed it seems
today as if almost everything, to be accepted as true, has to be the product of 'research'. Even if it is not really research it is
better for the author of the stuff to call it that. Much of the 'research' is nothing but speculation in order to help build a system.
Much of it seems to be done and used only to confirm a theory -more often than not a theory that contradicts other theories,
themselves claimed to be founded on research. Sometimes, even, the evidence gathered is used to demonstrate the opposite of
what it does demonstrate. Research is so much respected that not only is a lot of work done that is quite
5
The Fantasy of Structure
In the sense of a single, unified system there is no such thing as structure in language.
Sadly, however, many academics and philosophers are unwilling to accept things as they seem to be. Some great hidden
principle has to be found beneath the appearance of things.
For language, the main myth that has been created is structure. The reality is simpler, but not so intellectually intoxicating, since
instead of being one profound unified system, it is a number of different rather humbler features.
As I have proposed in the earlier chapters, the basic feature of language is meaning. But the many pieces of a language
obviously have to be arranged in some order. So one aspect of language is what we can call word order. As I shall try to show
in the next chapter, however, word order certainly does not deserve the obsessive attention as a single fundamental and decisive
principle that it is given by many linguists.
Then there are other features that distinguish different languages.
For instance, some languages have types of meaning that others do not, e.g. the 'action-thing' idea that Japanese (-koto) and
English (-ing) have, and which many languages do not.
There are also combinations of meaning that one language will have, and another not. For example, many languages use
prepositions with infinitives (e.g. Spanish: "comenzó por hacer..") while English does not - it says "he began by making..", not
"...by to make.."
Again, a given combination of meanings may mean one thing in one language and something quite different in another. The
Italian equivalent of "He said he Would do it", for instance, is "Disse che lo avrebbe fatto", literally "He said he would have
done it."
But this simple identification of a number of different aspects of language does not satisfy the academics. Modern linguists think
themselves greatly in advance of traditional grammar. The irony is that in the end they are just as much victims of tradition
themselves, in fact more obsessed and enslaved by the formality of analysis than any before them. They scorn old-fashioned
parsing, but actually, in principle, are parsers gone mad. They are no revolutionaries. They are the most die-hard conservatives.
They take for granted the distinction between grammar and meaning, and few, if any, of them appear to question the assumption
that structure is a fundamental part of language. It is an asumption that they
6
More Fantasy of Structure
Many linguists appear to be obsessed with the order of words. So much of what they mean when they talk about 'grammar',
'structure' or 'syntax' turns out to be only this order of words. When Chomsky draws attention to the 'grammaticalness' of
colourless green ideas sleep furiously
as opposed to
furiously sleep ideas green colourless1
he can only, of course, be referring to the word order, since everything else is exactly the same in the two sequences.
It is difficult to see the reason for singling out word order for such an important and fundamental part. I have already pointed
out how, for a start, one can only recognize any significant order of the words in a sentence through first recognizing their
meaning.
Humans can only make sounds one after the other, in a straight line so to speak, so obviously words have to be spoken in some
sort of order. This does not mean that their order is always crucial.
There are many individual words that we are certain of the meaning of even when they are used quite alone; many more
become perfectly clear when they are used with other words, even if jumbled up and without the 'proper' 'grammatical' words to
link them. In a very large proportion of cases, probably, one can understand accurately all the important meaning of a group of
words. This is because "wind" is still wind, "house" is still house, "down" is still down, and "blew'' is still blew, and because
from one's experience of the world one knows the likely connection between these different things, actions, and places. Where
'grammar' becomes useful or even necessary is when there are several possible connections between the things and actions etc.
kiss girl nose boy
is not clear. But add the 'grammar' and one only adds more meaning; 'grammar' is meaning too, it means real things in life, in
experience - things like time, and what comes before what, and who does something to who. The English convention is that the
recognizable thing that is 'said' before an
1 See p.94 above.
7
Ungenerative Grammar
1 The happy little elephant with purple ears kissed the kind-hearted mouse with quivering whiskers on the tip of his friendly
nose.
I have never heard or read that sentence before, and you almost certainly haven't either, but naturally I can make it up and you
can understand it and recognize it as 'correct'.
A very great deal is made by linguists of this so-called 'creative' ability. Chomsky is spoken of almost as if he had discovered
this great truth. Although it is apparently recognized now that it was taken for granted from the beginning in western linguistic
theory, it is constantly emphasized as something that has to be appreciated as fundamental by students of language. It is
described as "nowadays almost a truism" as if it wasn't a truism before.
Yet every person, including small children, has always known this 'truth'. It is obvious; and has no special significance. We all
know we can say what we like. Nobody, I think, believes they are limited in what they can say to sentences they have already
heard.
So why do people who are told about this obvious truth for the first time often accept it as a new truth, a discovery? For two
connected reasons, I think.
First, they are told it by their teacher-experts, who they do not believe would make a great point of something that everybody
knows about already. Second, and ironically, they probably accept it as new truth because of the part of it that is false.
Linguists talk of 'generative' grammar. They present language as creative. They appear not to see that language itself is not and
cannot be creative. Language cannot 'generate' anything. Only human beings can create and generate in language, and they do
that from their experience and with their imagination. But because linguists transfer the creative ability from people to language,
people are indeed taken in, because this is a new idea to them, that language 'generates'.
What linguists say about 'generative' grammar and a number of connected themes is a fine example of their tendency to say
what is either new and untrue; or obvious or trivial, or both.
Not only do Chomskyans claim that the grammar of a language 'generates', which is clearly untrue if that means "creates"; and
say a language can be used to say an infinite number of things, which is obviously true. They say that "The grammatical
description of" a language "may be
8
Deep Confusion
1 John is eager to please.
2 John is easy to please.
This pair of sentences is an example of Chomsky's, and has become very famous among linguists. Chomsky and the linguists
who follow him in this matter argue as follows: Both sentences have the same 'surface structure'; in traditional linguistic systems
they would get exactly the same analysis. But 'underneath' we are aware that while "John" really is the subject in 1, because he
is doing the pleasing, in 2 "John" is really the object - somebody pleases John, not the other way round. Thus, in 2, they say,
"John" is the 'underlying' or 'logical' or 'deep structure' object, although he is the 'grammatical' or 'surface structure' subject.
Similarly, often where the 'surface structures' are different, as in
3 The cat ate the mouse.
and
4 The mouse was eaten by the cat.
we know, they say, that the 'deep structures' are the same, the cat is the 'real' or 'logical' subject and the mouse is the 'real' or
'logical' object in both sentences, the relationship between cat and mouse is the same in both. So, Chomskyans maintain, we
know a grammar that does not appear directly in the sentences we actually hear and use, we know a grammar that lies beneath
the surface, and we also know rules for transforming 'deep structures' into 'surface structures' and back into 'deep structures'.
Chomsky, it is said, has drawn attention to the inner mental processes which alone can explain our competence to speak and
understand language. And a traditional analysis of the 'surface grammar', classifying the various parts of a sentence and
distributing them into various constituent groups or phrases, is completely inadequate, cannot explain the fundamental distinction
between 1 and 2 or the link between 3 and 4. A concept of 'deep grammar' has to be invoked; 'structuralist' (pre-
transformationalist) accounts of language have to be drastically revised.
Consider first some comparatively superficial, though nonetheless decisive, objections to this argument.
"John" is not the real or 'logical' subject in 2 ("John is easy to please"), they say; we are all aware he is really the object. So
suddenly they give us
Figure 1
"It will be seen that the logical subject of S1 (the matrix sentence) is Harry, and that of S2 (the embedded sentence) it is
John. Furthermore,
9
Being Able to Use Language
There can be no argument, I think, about at least some of the things that human beings need to be able to do in order to use a
language.
They must be aware of many different, separate parts of reality, things inside them as well as outside; they must be able to hear
and distinguish different sounds; and they must be able to understand the connection between the different sounds and the
different parts of reality - that is, they must understand that words mean things, and which words mean what. They must
understand that certain sounds made by humans have this connection while others, made by humans or not, do not (many
animals can do this a bit too). They must be able to remember all the different meanings, which means storing all the sounds
somewhere, with their connections, in a form which is not sound. They must be able to realize that if a word means something
in one situation it can be used to mean the same thing in another situation. They must be able to see patterns in the meanings of
sounds, such as that if "drop" and "look" and "reach" all just add a "t'' sound when they happened in the past, then "stop" and
"teach" ought to do the same - even if it doesn't always in fact work, and they have to notice when it doesn't work; and they
have to notice other patterns like "a potato, an apple" but "an old potato, a big apple'? They must notice the meaning of the
order in which the sounds are made. They must be able to carry out the whole process of getting the 'right' sounds, in non-sound
form, out of the memory, arranging them in the right order, and turning them into the right sounds in the mouth, and do much of
this in reverse when they listen and understand.
In their view of the ability to use language Chomskyans appear to give these facts little or no importance. They say that human
languages all have a lot in common (which is not strange), that children go through the same stages in learning whatever
language they do learn, and that this is because both the types of language they are able to learn and the form of their linguistic
development are restricted by biological inheritance. There is a 'universal grammar' known by all humans when they are born.
Language, Chomskyans say, is a separate and special faculty,
"different in kind from other cognitive systems, requiring different learning strategies and different genetic
programming...There are a number of rather obvious points that support the special-programming view of language
acquisition, and disconfirm the
1 See p.188.
10
Thinking and Language
Thought is not language. Thought is not based on language. Thought does not depend on language; language is not a condition
for thought.
There is no essential connection between language and thinking except in two senses: that language is a translating device for
the imperfect expression of thought or of the awareness of experience; and without thinking humans could not produce
language.
This does not mean that thought is not in practice influenced by language. Sadly, in most people, thinking is very much
dominated and shaped by language, even though it does not need to be. The corrupting effect of language is probably the most
important thing that needs to be shown about it.
And I am not talking about whether or not the existence of language leads to thought. It almost certainly does, since through
language people try to tell other people about their thought, and thought can arouse thought.
First, though, I must try to show how utterly different and separate language and thought are. People will never even begin to
understand either of them until they understand that; yet it is something people can work out or observe directly for themselves.
I was once chased down a Cornish lane by a duck. (We were both on foot -the duck and I, I mean. I can't remember flit was a
she or a he, but I think it regarded me either as a sexual rival or as a possible mating partner.) It was some years ago. What has
been kept inside me all that time, so that I can now tell you about it? Not words. Not the isolated word "duck", for instance,
because although that word is certainly kept inside me somewhere, it is obviously useless for keeping just this unique Cornish
occasion inside me -its whole function is to be ready for any 'duck' need that arises. Nor can it be a whole sentence, "A duck
chased me down a Cornish lane", because why not "I was chased by a duck down a lane in Cornwall" or "I ran down a Cornish
lane with a duck in pursuit" or ''I had to flee from a duck along a country road"? (I am free, when I speak to you, to use any of
these different meanings of the same event - see Chapter 8.) It is clear that it must be something else that is kept inside me,
something much closer in its nature to the original unique reality. And what 'comes up' inside me in the first place to allow me
to tell you about it? Again, it cannot be words, for if it was, there would be no reason why I should not say it was a tiger that
chased me, or
11
Language the Corrupter
Thinking, then, does not need language. But language needs thought, and words need experiences.
Some say that it is the other way round, that experiences need words.1 We couldn't have the experiences we have without the
categories made by language. Searle expresses it like this:
"I am not saying that language creates reality. Far from it. Rather, I am saying that what counts as reality - what counts as
a glass of water or a book or a table, what counts as the same glass or a different book or two tables - is a matter of the
categories that we impose on the world; and those categories are for the most part linguistic. And furthermore, when we
experience the world we experience it through linguistic categories that help to shape the experiences themselves. The
world doesn't come to us already sliced up-into objects and experiences; what counts as an object is already a function of
our system of representation, and how we perceive the world in our experiences is influenced by that system of
representation. The mistake is to suppose that the application of languages to the world consists of attaching labels to
objects that are, so to speak, self-identifying. In my view, the world divides the way we divide it, and our main way of
dividing things up is language. Our concept of reality is a matter of our linguistic categories." (Magee, 1982, p.156).
What Searle says here is true if it is a description of how things often or usually are in practice. But it is not basically true, and
cannot be true. Humans cannot apply these linguistic categories or representations, their names for things, unless and until they
are aware of the object or experience independently of language. I may not be aware of the same categories of things as our cat;
he may, for all I know, lump trees and poles together in the same category, while he no doubt has a whole world of smell that I
could never appreciate. But this is nothing to do with language. The awareness comes first, awareness of objective reality, even
if different creatures, and indeed different humans among themselves, are aware of different aspects of that objective reality.
Our cat can distinguish houses, dogs, mice, humans, and other cats for example. Free of language, he nevertheless has his
categories. His
1 I mentioned Wittgenstein's argument in Chapter 10, p.195.
12
Trying to Speak the Truth
So what is to be done about it?
The solution is not in definition and linguistic analysis of meaning -neither in principle nor in practice.
Definitions are only words about words.
Definitions do not tell one anything about things - which are, or should be, what are important in life. A word is not the reality.
And not only that. As I have tried to show, a word is from the start a distorter of the reality. A definition, saying words about a
word, is something that takes one even further from the reality, carries the distortion one level further. It does not convert the
word back to the reality, it does not enlighten in terms of reality, it does not point to the reality. It is something entirely
artificial. It does not make the connection one really needs to that reality. Most people, I suspect, believe that definitions are
about reality, about things, and that there is something solidly scientific about them. In fact they are one more barrier between
humans and straight thinking.
Nor does definition and analysis of meaning tell one anything more about what a person actually means, and even less about
what a person thinks. It is useless, one might say meaningless, to talk about what the real, proper, correct meaning of a word or
sentence is. There is no independent truth about the meaning of a word or phrase or sentence which clever or learned people
know or can find out and some others don't know. Words only mean what their speakers want them to mean; only people
express meanings.
How does anyone decide whether a definition or an analysis is a good one? How does one make a definition in the first place?
Only through one's previous knowledge of another kind about the meaning, a knowledge without benefit of definition, that is,
knowledge of the associations of reality (or what passes for reality) attached to the word. There is nothing absolute or universal
about those associations; they may or may not coincide more or less with other people's.
People do indeed use words variously and often inconsistently, and perhaps many deliberately deceive others by manipulating
words. But to set up a huge, elaborate, systematic analysis of the meanings given to words is as pointless as if one was aware
how dangerous generals are with their bombs and so set up a technical examination of the details of the bombs instead of
concentrating on the mentality of the generals. In the same way we should be thinking about our own mentality in connection
with language and about the general principles - psychological principles -
13
Slavery to Authority and the Word
You may feel I have made something sensational, or at least something greatly exaggerated, out of the problem of language.
And then proposed a very much less than sensational solution to that problem.
Yet I think the failure of language is perfectly plain. There has been an improvement in the life of humans in many parts of the
world. But in the main that has not been the result of ideas, which have done a great deal of harm and very little good.
Where human lives are more comfortable and pleasant it is because of the success of mathematics and technology. In
mathematics and technology if one gets it wrong things stop, fall to bits, or blow up - or never work in the first place. And one
discovers these failings rather quickly.
Where human lives are now less miserable than cruelty, indifference and selfishness used to make them, it is because of the
growing of concern, pity, tolerance and responsibility. These feelings have been supported by the technology which has helped
so many people, from the eighteenth century onwards, to know about the misery of others. For this reason and other reasons
which are unfathomable, attitudes have changed for the more tender in many communities.
What is noticeable is that where facts work directly and immediately and incontrovertibly, humans have made undoubted gains.
I do not have to face the alternatives of a rotting mouth or repeated agonies as my teeth are drawn one after the other. I can
enjoy, any day I like, by means that would have been miraculous to the contemporaries of the men who made it, large amounts
of music that enriches my life. I can visit loved people hundreds or thousands of kilometres away swiftly and in comfort, if not
in complete safety. I do not mean that the result of technology is always happiness. It is often sickening disaster. But it is human
greed, ambition or stupidity that produces those disasters. The technology itself is not to blame. And the greed, ambition and
stupidity have fed on words, and excused themselves by words.
Where there is only supposed 'knowledge', particularly where that supposed 'knowledge' claims to be about humans, human lives
have not been advanced one whit. Here there are no incontrovertible and uncontroversial facts, and so it has always been. Only
schools and abstract systems and dogmas and pretensions and propaganda and polemics. Here there are only words. This is the
realm of words and all in it has failed.
The manipulation of physical objects and of numbers has succeeded; the manipulation of words has failed, brought no good and
much agony.
Bibliography
Below I list the printed or spoken material that I have referred to in the text, together with a number of the relevant books,
articles and radio talks that have provoked or interested me most. I have not included any of the very many letters to newspapers
or non-specialist periodicals that I have often found at least as stimulating as anything else, nor, obviously, the many
conversations I have had over more than thirty years with people holding a variety of opinions on language and linguistics,
although those conversations have often been as revealing and important as any written material for my understanding of these
things.
I have deliberately also not included any of the considerable number of articles in learned journals that I have gone through. I
have not included them because although they are nearly always revealing - monotonously so - I do not want to lure readers into
suffering the same tedium and irritation I suffered. Even more important, I do not want them to think I feel it is necessary to
study such stuff in order to have valid opinions.
Artificial intelligence [talks on], BBC radio broadcasts, 28 February, 7, 14, 21, 28 March 1983.
Baby talk, BBC radio broadcasts, ca.1984
Berry, M., 1975. An introduction to systemic linguistics. 1.Structures and systems, B.T.Batsford.
Burgess, A., 1979. The metabolism of words [review of N. Smith & D.Wilson, Modern linguistics], in The Observer, 29 July
1979.
Butterworth, B., 1983. Biological twists to grammar [review of D.Lightfoot, The language lottery], in New Scientist, 20 October
1983.
Campbell, J., 1982. Grammatical man: information, entropy, language, and life, Allen Lane.
Chomsky, N., 1959. Review of B.F.Skinner, Verbal behaviour, in Language, 35.
Chomsky, N., 1976. Reflections on language, Fontana.
Chomsky, N. [talking to John Maddox], BBC radio broadcast, 21 October 1981.
Chomsky, N. & Halle, M., 1968. The sound patterns of English, Harper & Row.
Chomsky, N. & Magee, B., 1978. Noam Chomsky on the genetic gift of tongues, in The Listener, 6 April 1978.
Index
In this index many topics will be found listed as sub-entries under the following headings: grammar; language; learning foreign
languages; learning language; linguistics; logic; meaning; meanings; pronunciation; sentences; stress; structure; teaching foreign
languages; thinking; word order; words. n = note
A
"a" 77, 81, 84-87, 203
A-over-A principle 122
abilities, see language; learning foreign languages
abstract phrases 116-19
academics and academic world (see also universities) 1-2, 37, 206-07, 255-59
monopoly 255-59
solidarity 37, 257-58
acceptability of sentences, see sentences
accidence 105
accounting for 16-17, 25-27, 29-30, 169
African languages 56-58, 191
ambiguity 25-27, 32
American girl 178-82
analysis (see also grammar; language; structure) 1, 91-92, 169
invention through 26-27
as proof of reality 38
"and" 49-50, 56-57
angels and pins 39
animal diagram 38-39
animals, relationships with 247
animals' categories 222-23
"any" 87-89
apes 139n, 192
Arabic-speakers 68
articles 77-87, 108, 135-36, 203
articulate people, unfair advantage enjoyed by 240, 249
Aspects-type grammar 21-22
atrocities 224-25, 228-31, 257-58
authority 1-2, 37, 255-58
solidarity of those in 37, 257-58
awareness 197, 248
objective 195, 220-21
B
Barwise, J. 163
"because/for" 48-50
behaviourism 4, 8n
Belgians 67
"believe", double meaning of 51-52
Berry, Margaret 19-20, 37
bicycle riding 182, 217-18
bi-lingual people 177
biology 3
brain
areas for thought and language 218
and the use of language 9-11, 138, 140, 192, 214-18
study of, through language study 34, 36, 151, 204-06, 218-19
Bresnan, Joan 101
Brazil 61
Britain 68
Burgess, Anthony 33-34
C
Campbell, Jeremy 40-41
case grammar 164-65
cases 8
categories, grammatical 77-89
categorization, danger of 214, 231-38
children (see also learning language) 10, 91, 137n, 145
enslaved by words 238
Chinese room 250
choice 20, 201-02
Chomsky, Noam 3-4, 13-14, 21, 24-26, 40-43, 51-55, 58-59, 94, 99, 114-126, 131-36, 139-42, 151, 153, 156-57, 165, 184-85,
188, 190-92, 194, 206-07, 219, 220n
absolutism of 53
classification 6, 38-39, 77-89
by words 214, 231-38
clauses
logic of 130
subordinate 48-50, 130
co-ordinate 48-50
codes and codifying 25-31, 37, 191
competence, see language, non-linguistic and linguistic factors in
complementizers 123-25
composers 216-17
compound stress rules 98-100
comprehensibility input 61
A comprehensive grammar of the English language 75-89
computer model of thought 213
computers 20-21, 90-91, 247-49
concepts dependent/independent of language 195-97
constituent analysis 17, 166
context 49-50, 53-55, 67, 84, 88, 101-03, 107, 112-13, 138-39, 155, 176n, 198, 200, 222, 226, 232
continuous, present perfect 36
cooking 216, 218-19
co-ordinate structures 56-58, 190
Coulthard 36
Crystal, David 15-17, 38-39, 132, 140, 164-68
Curtiss, Susan 182-83
D
deception through attractive language 239-40
definition 5n, 6-7, 38, 124, 233-34n, 242-51
of language, see language, principles of
degenerate evidence 173, 182-88
Descartes 197n, 210
descriptive adequacy 34
dictionaries 31, 176n
discourse analysis 36
dolphins 230-31
dummies 105-11
Dutch 67-68
E
"eager" 151-55
"easy" 151-55
echo-questions 45
effects of experts and writers (see also experts) 2-4, 195n, 232, 257-58
Einstein 200
emotions created and maintained by
F
facts, see event
faculties (see also language faculties; language faculty)
breakdown of 217
general and special 216-18
fantasies, grammatical 26-27
Fe'Fe' 57-58, 191
"few" 77
Fillmore, Charles 164-65
fingers 216
Flemish-speakers 68
"for" 42-43
"for/because" 48-50
Fodor, Jerome 197n
formal systems 11-37
French 44, 56, 104, 128, 176, 187-88, 206
Freud 59
future tense sentences 64
G
games 189
Gardner, Howard 217-19
gender 104-05
general meanings, see meanings, general
generative semantics 165
Genie 180-84
German 174, 186
Germanic languages 8, 68, 186
gerund, see "-ing"
grammar (see also language; meaning; meanings; meanings, general; sentences; structure; word order)
abstract analysis of, see analysis
acceptability of, see sentences
arbitrariness of rules of 147-48
Chomskyan, see linguistics, Chomskyan
classificatory 77-89
complicated and simple rules of 186, 190-91
explaining 34, 36, 48, 65-67, 125-27, 252
fantasies 26-27
as general meanings, see meanings, general
generative (see also grammar, transformational) 21-34, 131-50, 166-69
immediate constituent 17, 166
H
habit 105n
"habit" 227-28
Halle, Morris 99
Halliday 17
Hampshire, Stuart 115, 138
happens, thinking about what actually 182, 206-07, 214, 223, 226, 229, 232-33, 237-38, 244-45, 248-50, 253, 255
"have something done" 65-66
Hebrew 44
helping each other to understand 249-52
Holy Trinity 200-01
human nature 1
humans, as language users and cause of suffering 230-31
I
ideas distorted by (good) language 201-02, 238-40
ideologies created and maintained by words 228-31
idiomatic usage 185-86
"if" clauses 66-67
if-ishness in thought 211-12
illogical expressions and sentences 9, 55, 138, 146-47
imagination 140-41
imitation 8, 107
inborn language faculty, see language faculty
incomprehensibility 37
Indo-European languages 68
infinitives 32-33, 40-41, 105-07
information theory 20
"-ing" 25-26, 32-33, 65, 76, 93, 112, 192
inherited characteristics 180-81
inside eye 209
"instinct" 227-28
intelligence 178, 180-82, 230-32
"intelligence" 230-32
"intuition" 225-26
intuitions about language 134-35
invention through analysis 26-27
inversion 8, 67n, 147-48, 184
IQ 184
K
knowledge 1-2
elementary 61-62
failure of and danger from 254-55
L
labels, avoiding 214, 236-38
Lamb 16
language (see also grammar; learning foreign languages; learning language; logic; meaning; meanings; pronunciation; stress;
sentences; structure; thinking; word order; words)
abilities at different ages 171-82
analysis of 11-39, 77-78, 91-92, 169
and the brain 9-10, 11, 34, 36, 138, 140, 151, 192, 204-06, 214-19
and computers, alliance of 249
constraints on 40, 53, 115, 120, 122, 124, 133, 148, 184, 187, 246
conventions 8, 67n, 105, 128, 147, 184, 218
'correctness' in 60
corpus of 137
corruption by, see thinking; words
creation of by individuals 37
creativity of 14, 131-42
damage to happiness by 246-47
definition of, see language, principles of
deprivation 178-82
description of 5, 11, 29, 31-32, 34, 37
disorders 11, 218
disputes about the use of 241
distrust of 230, 249, 253
domination by 2-3, 245-47, 249
and events, distinction between 5-6, 157-58
explanations of 34, 36, 48, 65-67, 125-27, 252
faculties, general, controlling use of 138, 141, 215-18
faculty, independent, innate, programmed 40-59, 115-23, 133, 137, 146-50, 169-70, 180-93, 214-18
failure of and harm done by 3, 254-55
features of 93
good, no proof of good ideas 239-40
habits, individual 217
ideas changed and obscured by 201-02, 238-40
infinite character of 131-33
as instrument 9-10
intuitions about 134-35
as invention 217-18
M
Magee, Bryan 52, 245, 252
Marxism 1, 60, 250
mathematical analysis of language 139
mathematicians 212
mathematics 254
McCarthy, John 249
meaning (see also grammar; language; logic; meanings; sentences; structure; word order)
as basis of language, see meaning, primacy of
O
object
logical/underlying and grammatical/surface 151-53
relationship to various verbs 203-04
observation, see learning language, observation in
''one" 76-77
order of words, see word order
Orwell 14n, 251
P
paradigmatic relationship 14-15
parsing 93
passives 30-31, 158, 168-69
patterns, recognition of 67n, 147-48, 170, 182, 188-89
perceptual strategies, see language, non-linguistic ...
performance principles, see language, non-linguistic ...
Perry, J. 163
Persian-speakers 68
PhD's 60
philosophers 4, 9, 163, 206, 208, 214, 243-44, 252-53
philosophy 3
false problems of 243
linguistic 243-45, 250
phonetics and phonology, see pronunciation
phrase markers 23-24, 29
phrase-structure 22-25, 28-29, 166
phrases, abstract 116-19
Pike 15-16
politics 3, 257-58
popular works 258
pragmatic factors in language, see language, non-linguistic ...
preconceived models 19
present participle, see "-ing"
priests 1
"probable" 145
professors 4
pronouns 203
pronunciation (see also stress)
conventions of 67n
and meaning 44-48, 99-104
of English "oo", Norwegian and Swedish 68-69
of foreign languages 174, 176
by different races 61
and structure 28, 45, 110-19
study of phonetics 89-90
propositions 161
Q
question forming 52-55, 115-25,
questions, indirect 125-26
Quirk, Randolph 75-89, 248-49
Quirk et al.'s grammar of English 75-89
R
rationality and reasoning 1-2, 252, 255
as alternative to research 61-62
Ravel 218-19
reactions to events included in language 5-6, 157-68
reality, see language and reality; meaning and recognition of reality
recognition of reality and meaning, see meaning and recognition of reality
of the thought of others 249-52
reference words unused in thought 203
relative clauses in various languages 43-44
relative pronouns 203
religion 1
religious fanaticism 258
repetition, so-called 'recursive structures' 33
research and researchers 2, 39, 60-62, 255
rigidness and rule-boundness 13-14, 148-49
rules, see grammar, rules of; language and situations; word order
rulers 307
Russian students 230-31, 250
S
Sacks 36
Sakharov 200
Saussure 14, 17
scale and category grammar 16-17
Scandinavian passives 169n
schools, language 72-75
science and scientists 1, 144
scientific theory, generative grammar as 142
Searle, John 190, 197n, 206-07, 222, 249-50
semantic space 38
semantics, see meaning; meanings
Semitic languages 68
sense, making 53, 57-58, 193
sensitivities of expression 76
sentences
acceptability of 13-14, 44-59, 76, 94-98, 131-50
deriving 23-24, 124, 166-68
'impossible' 13, 28-30, 40-59, 65, 77, 94-95, 117, 120-26, 130, 142, 146
operations on 115-24
subjective judgement of 53, 134-39, 144
related 25-26, 30, 124, 158-59, 166-69
restrictions on, see language, constraints on; logic of sentences
Shebalin 219
"since" 64
T
tagmemics 15-16
Tanner, Tony 248
taxonomy 14
teaching foreign languages (see also learning foreign languages) 30-31, 61-90
course books 70-71
practicality and psychology 31
principles and qualities needed 72-75
techniques and approaches 69-72
training courses 73-75
U
ultimate constituents 105
unacceptable sentences, see sentences
unclear cases in generative grammar 142-45
underlying strings 23-26, 32
ungrammaticality, see sentences
United States 61
universal grammar, see grammar, universal
universals, relative 190-91
universities (see also academics) 1
linguistics departments 72-75
A university grammar of English 83-84, 86
university students, conditioning of 255-56
unnecessary words 105
W
war games 27-28
"weather" 77
Webster, Richard 248
Weizenbaum, Joseph 251
Welsh 169n
wh-movement 44-46, 51, 124
whales 232-33
"who...?" 51-55
Williams, Bernard 246, 252-53
Wilson, Deidre 28-34, 36, 43-46, 48-49, 56-58, 98-103, 132, 137n, 140, 143-46, 148, 153, 161-63, 172-73, 180-82, 185, 191-
94, 206n
Wittengenstein 197-98, 210
word forms 169
word order (see also grammar; language; sentences; structure) 8, 31, 44-45, 48-50, 56-57, 94, 114-29, 178-79
conventions of 8, 67n, 128, 147, 184
inversion 8, 67n, 147, 184
and logic of life 127
as form of meaning 8, 31, 129
necessary 127-29
obsession with 93, 114, 179
restrictions on 53, 115-28
rules 56-58
words (see also language; thinking) anticipating and distorting thought and feelings 233
associations, false, created and transferred by 231-35
attitudes determined and modified by 232-36
basic evil of 238
classificatory, as substitutes for thought 214, 234-38
combinations of that prevent thought 223-24
emotions created and maintained by 224, 228-31
freedom from 249-51
as easy and false generalizations 232-37
ideas changed and obscured by 201-02, 238-40
ideologies created and maintained by 228-31
inflexibility of 231-33
new and false, invented through combinations of experiences 227-28
'physical' 211, 250
problems created by existence of 234
as relacements for thought 243-44
responsibility to 250-51
seeing behind 250-52
single, corruption by 225-26
would-ishness 211-12
writing
good, no proof of good ideas 239-40
standards 91