The English-Learning and Languages Review Ć Homepage

The Mirage of Linguistics:

Flaws in Pinker's and Chomsky's accounts of language


Amorey Gethin




The importance of understanding the nature of language and thought

Language supposed to be the key to human thought and personality

Chomsky believes the form of language is determined by the form of the human mind - his theory flawed by elementary mistakes

Chomsky's argument for the existence of 'universal grammar'

Chomsky's and Pinker's mistakes regarding the forming of questions

Intonation is more basic to question meanings than structure

Diversity, not universality, characterize question forms

Identical sequences of words can be both statements and questions

Questions that are completely structure-free

Pinker's and Chomsky's double error regarding question formation

Many languages do form questions by mirror-reversal

Reversed pairs can be attached to longer sentences to form questions

Grammatical abstractions cannot exist independently of meaning

The non-existence of grammatical 'dummies'

The fallacy of the correlation of prepositions with verb-object sequences and postpositions with object-verb sequences

How do you determine whether a language is a VO language or an OV language?

Pinker has not told us the whole truth about VO and OV word orders

Failure to examine foreign languages leads Pinker into a further false assertion about innate knowledge

Demonstration that, contrary to Chomsky, grammatical analysis is impossible without first understanding meaning

Meaning must come before categorization into parts of speech

Further demonstration of Chomsky's illogicality regarding meaning and grammar

Pinker's muddled view of the connection between concepts and parts of speech

The square pegs of reality forced into the round holes of language



Understanding the nature of language and thought, or at least what they are not, is just about as important as any understanding can be. Both are at the basis of our lives; in a sense they are our lives. Is language a distinct faculty? Is it controlled by parts of the brain dedicated to language alone? Is human thought language? If it is, are we intellectual prisoners limited to thinking what language can describe, and allows us to think? Or is language a human invention? Is thought essentially independent of language, but in practice critically influenced by it? Much, politically and socially, depends indirectly on which is the correct view, and much depends on the view of linguisticians, neuroscientists and philosophers, whether they are correct or not.


The opinion of most writers on the subject seems to be that language is basic to our nature, whether it is our minds that shape language, or language that shapes our minds. Language is seen as the fascinating key to human thought and the whole human personality. The philosopher Karl Popper went so far in his reverence for language that he appeared to confuse it with reality. He believed, for instance, that small children only become aware that they are separate from others through language, at the time they begin to say "I". note 1



Noam Chomsky thinks that the form of language is determined inescapably by the form of the mind. Most of his academic colleagues seem to do little but devise or develop barren systems of linguistic analysis merely for the sake of analysis. Chomsky at least has a worthwhile ambition. He aspires to contribute to the understanding of human psychology. In Antilinguistics I have tried to both illustrate the pointlessness of most modern linguistics, and demonstrate in detail the illogicalities and frequent absurdities of Chomskyan linguistics in particular. note 2 Here I want to discuss briefly a few of the claims in Chomskyan linguistic theory and point out a number of what I think are elementary mistakes. I want to do this because Chomsky's ideas have strongly influenced people's views on the 'authority' of language in our lives, and also because discussion of those ideas indirectly raises important issues of intellectual authority, in both principle and practice.


The American philosopher John Searle explains Chomsky's argument for the existence of his well-known 'universal grammar' as follows:


"The syntax that Chomsky comes up with is extremely abstract and complicated, and that raises the question: 'How can little children learn a language when it is so complex?' You can't teach a small child axiomatic set-theory; yet Chomsky showed that English is far more complicated in structure than axiomatic set-theory. How is it, then, that little kids can learn it? His answer was that, in a sense, they already know it. It is a mistake to suppose that the mind is a blank tablet. What happens is that the form of all natural languages is programmed into a child's mind from birth." (Magee, 1982, p.170)


This circular argument is an example of the false assumptions on which the Chomskyan theory to a large extent rests. Chomsky erects a frighteningly complicated and abstract system of syntax, without evidence that it exists as a psychological reality; he then uses its very difficulty to suggest that therefore its mastery must be inborn.


So the forms human language can take, Chomsky maintains, are biologically determined. Well, it is obvious that language is the product of the human mind. What else would it be? Chomsky, though, wants to go much further. Yet his and his supporters' argument sometimes depends on plain and simple errors. Several are evident in a much-acclaimed book by Steven Pinker, The language instinct (Pinker, 1995). Pinker argues (p.43) that:


"The universal constraints on grammatical rules also show that the basic form of language cannot be explained away as the inevitable outcome of a drive for usefulness. Many languages, widely scattered over the globe, have auxiliaries, and like English, many languages move the auxiliary to the front of the sentence to form questions and other constructions, always in a structure-dependent way. But this is not the only way one could design a question rule. One could just as effectively ... flip the first and last words, or utter the entire sentence in mirror-reversed order ... The particular ways that languages do form questions are arbitrary, species-wide conventions; we don't find them in artificial systems like computer programming languages or the notation of mathematics. The universal plan underlying languages, with auxiliaries and inversion rules, ... and so on, seems to suggest a commonality in the brains of speakers, because many other plans would have been just as useful."


But what Pinker asserts here is untrue. One cannot always "just as effectively ... flip the first and last words, or utter the entire sentence in mirror-reversed order" to form a question. If we take the statement Cats chase mice, and apply to it what is both a first and last word flip, and a mirror reversal, we get Mice chase cats, which cannot effectively be used as a question, since it is already a different statement with a meaning the reverse of Cats chase mice. So there is surely a good practical reason why people do not use first and last flip or mirror reversal for forming questions out of three-word statements like this.


It might be objected that Mice chase cats is a perfectly effective way of forming a question out of Cats chase mice because one can always use intonation to make clear one means a question. Intonation is indeed used to distinguish meanings; it is often used to distinguish between statement and question although an identical sequence of words is used for both. In Italian, for instance, Paolo aiuta Maria can mean either "Paolo helps Maria" or "Does Paolo help Maria?", according to the tone of voice. Precisely because of this, to mirror-reverse a word order to produce a question when there is no need to, and when it would only complicate matters, would be an impractical and foolish thing for people to do, and so they don't. note 3


David Bond has pointed out to me how Pinker shifts without obvious justification from talking about "many languages" to "languages" and finally to talking about "species-wide conventions" and the "universal plan underlying languages". What exactly are these "particular ways" that languages form questions? What is noticeable, actually, about the ways humans have chosen to ask questions is the diversity, not the "commonality", and a diversity, moreover, both within languages and between them. In Japanese, for instance, no inversion, nor, indeed, tinkering of any kind with the word order is involved in the formation of questions. Neko desu. (ねこ です。Cat is.) "It's a cat." The question is formed by simply adding the particle ka () Neko desu ka? (ねこ ですか。Cat is question.) "Is it a cat?"


As Bond says, many other languages, modern Greek and Italian among them, can make statements with the subject either at the beginning or the end of the sentence. The statement "Maria has telephoned" can be expressed in Greek by either Η Μαρια τηλεφωνε or Τηλεφωνε η Μαρια, or, in Italian, by Maria ha telefonato, or Ha telefonato Maria. All four of these statements could equally well serve as questions, given the appropriate intonation. Word order here does not in the least depend on whether question or statement is intended, but on the nature of the emphasis required.


In effect, in most cases such languages distinguish questions from statements solely by intonation, a device which by no stretch of the imagination could be characterized as 'structure-dependent'. This entirely pragmatic method of forming questions is frequently used even in English, and in contemporary French is arguably as common as any other. Chomsky really said that? Pinker a vraiment écrit ça? This structure-free interrogative is perhaps the only one that is truly "universal" and "species-wide". This principle finds its simplest expression in the one-word question: Cigarette?


There seem in fact to be two problems with Pinker's statements about questions. He implies that there is actually no language in the world that uses first and last word flip or mirror reversal to produce questions. He is here making the same mistake that Chomsky made already many years ago. In a discussion with Stuart Hampshire broadcast by the BBC on 17th October 1968 Chomsky maintained that


"universal grammar – that means that set of properties which is common to any natural language, necessarily, by biological necessity ... [has] very explicit restrictions on the kind of operations that can occur, restrictions that we can easily imagine violating. Well, er, let me give just a trivial example. If a mathematician were asked to design operations on sentences, he would think automatically of certain very elementary operations, such as, say, er, reading [sic] the sentence from back to front, or, say, permuting the third word with the tenth word, and so on and so forth ... However, such simple operations simply do not exist in natural language. For example, there is no natural language which forms questions by, er, reading a declarative sentence backwards. Now ... it's not so obvious why that should be so, because it's a very simple operation. It's a much simpler operation to state than the operation by which we formulate questions in English, let's say, as you can discover by trying to formulate that operation. Nevertheless, the principles that determine what operations may apply in a natural language preclude such simple operations as reading the sentence backwards ... all of the operations that apply to sentences are structure dependent operations ... – that's an example of a simple linguistic universal that you can't explain on the grounds of communicative efficiency, or simplicity, or anything of that sort, it must simply be a biological property of the human mind. ...
Well, I think what are interesting are the kinds of principles which are universal, but not merely by accident, that is, not merely that no language violates them, but rather that no language could violate them. They're universal in that sense. And secondly ... principles that do not have the property that you just mentioned, namely that they are somehow necessary for organisms of approximately our size and er, role in the world. I think the interesting universals are the ones which are not necessary in this sense, and there are many such. For example, I've just mentioned two, actually ... second the principle that makes it impossible to form a question, let's say, by reading the sentence back to front. Now, neither of these principles is at all necessary for communicative efficiency ... these are formal principles which one can easily imagine a language which violated these principles.[sic] ... Instead of having our complicated rule for the formation of questions, this language would have a very simple rule, so the question associated with 'John saw Bill' – you know, 'John saw Bill yesterday', would be 'Yesterday Bill saw John', or something of that sort."


Why did Chomsky change a three-word declarative sentence into a four-word one? Was it because he too saw the weakness in what he was about to say? This question needs to be answered, because so much of Chomsky's and Pinker's whole view of language seems to rest on this double point that languages do not actually use formulations that they could in fact "just as effectively" use.


On one hand it is hard to see how mirror-reversing a three-word statement would be an effective way of forming a question, and on the other it is difficult to understand what is meant by Chomsky's statement and Pinker's implication that no natural human language forms questions in this back to front way. Is that not precisely what many languages do, such as German?:


Sie Rauchen. / Rauchen Sie?

(You smoke. / Smoke you? = Do you smoke?)


So what we find is that it is very common for people to use back-to-front or mirror-reversal word orders to make questions out of two-word sequences, while the same system is very rarely, possibly never, used for three-word sequences. Don't simple basic facts like these strongly support the idea that language and languages are indeed inventions devised as practical instruments of communication? Inversion of two words to make a question is simple and clear and involves no problems, so it is one obvious method for humans to choose for making questions. Mirror-reversal of three words is not. It is surely such practicalities that determine the way humans order their words.


The reversed pair can be attached as a unit to sentences of many more words than two, without the question meaning being lost:


Smoke you all day long, or only after meals?


Such reversed pairs of course in practice often consist of more than two words, which in the equivalent English would be, for example:


Your younger brother smokes. / Smokes your younger brother?


But here we still in effect have a pair, two units or ideas, one being smokes, and the other your younger brother. At this point Chomskyans will start to talk about 'structure dependency' again and claim that humans are only able to identify the units correctly because they are genetically programmed to recognize that your younger brother (for example) is a discrete or distinct abstract grammatical structure. As we shall see later, this begs the question. What the speaker (or listener) recognizes is a distinct, separate reality in 'life', your younger brother. Nobody can recognize this as a linguistic unit until they recognize the meaning of the words, and at that point recognition of an abstract structure is irrelevant.


Here perhaps is Pinker's basic mistake, that is to say, believing that subjects, for instance, exist as grammatical abstractions independently of meanings, that your younger brother and subject are two different things. But "subject" is nothing more than a name given by grammarians to the meaning of a particular sort of relationship in real life – in this case an 'actor' relationship. Speakers and listeners don't need abstract analysis to produce or understand that meaning.


There is a clear example of what I believe to be Pinker's confusion on p.42 of The language instinct, where he writes that in a sentence like It is raining


"the it ... of course, does not refer to anything; it is a dummy element that is there only to satisfy the rules of syntax, which demand a subject."


I have discussed these postulated – and actually non-existent - dummies in some detail in Antilinguistics (pp.105-11). But what rules of syntax  are these? If Pinker means they are universal rules of syntax, he is clearly wrong. Italian, to take just one example, is constrained by no such rules.


Piove.Is raining.


they say in Italy. No subject 'dummy' needs to be set out. But if he means that it is just English rules of syntax he is talking about, it is hard to see how that helps his case that all children have an innate grasp of abstract structure.


Pinker also asserts (1995, pp.111-12) that


"...if a language has the verb before the object, as in English, it will also have prepositions; if it has the verb after the object, as in Japanese, it will have postpositions [= 'prepositions' after, not before, their nouns].......
"This is a remarkable discovery. It means that the super-rules suffice not only for all phrases in English but for all phrases in all languages...when children learn a particular language, they do not have to learn a long list of rules. All they have to learn is whether their particular language has the parameter value head-first, as in English, or head-last, as in Japanese...Huge chunks of grammar are then available to the child, all at once, as if the child were merely flipping a switch to one of two possible positions."


Again the whole hypothesis is based on a falsehood. Not all verb-object languages have prepositions. For example, Finnish combines the verb-object pattern with postpositions.



















                (The man put the  bottle under the table.)


Finnish, in fact, is not the only language that combines the verb-object pattern with postpositions. But it is obvious that even just one language that does not obey the Chomskyan-Pinker super-rule wrecks the entire rule, and a child can certainly not master the grammar of her language by "merely flipping a switch".


Furthermore, it is not always as easy as Pinker makes it seem to decide whether a language is a verb-object(VO) or an object-verb(OV) language. For instance, which is German? For the small children of German-speaking parents (and for anyone else, for that matter) it is impossible to decide whether sentences of the kind


Ich kann die Katze nicht finden.

I cannot the cat[O] find[V].


Ich bin glücklich, da ich die Katze habe finden können.

I am happy, as I the cat[O] have [to] find[V] be[en] able.


or of the kind


Ich suche die Katze.

I seek[V] the cat[O].


are the more common. And even if they could decide, would that determine whether German is a postposition or preposition language?


It turns out, in any case, that Pinker is not being open with us. A study has been made of a sample of languages to see precisely what the correlation is between verb-object (VO)/object-verb (OV) orders and prepositions/postpositions in those languages. It was found that of

82 VO languages 70 have Prepositions and 12 have Postpositions

and of

114 OV languages 7 have Prepositions and 107 have Postpositions

(Dryer, 1992)

So there are many exceptions to Pinker's supposed 'rule'. There does seem to be a clear trend in favour of using postpositions with object-verb and prepositions with verb-object. But that is not in any way evidence that humans are born with such grammatical rules as part of a distinct inherited language faculty. It is far more likely that the distribution of the various word orders is due to the particular minds of different groups of people, and, above all, to the particular wanderings and fortunes of those groups. To argue otherwise would be the same as to argue that humans are biologically programmed to eat bread or rice and drive about in motor cars, because that is what humans do everywhere, with the exception of a few isolated communities here and there.


One would think from the way Pinker talks about “many languages” that he is familiar with a considerable variety of them beyond English. In practice he does not seem to have bothered to examine even some of the most widely spoken. He makes much of a discovery he claims has been made by a developmental psychologist, Peter Gordon. Gordon, he says, found that three- to five-year-old children, when asked to produce compounds like mud-eater, produced compounds such as mice-eater but never compounds like rats-eater. In other words, they would use irregular plurals as the first element, but never regular plurals. They did this


“even though they had no evidence from adult speech that this is how languages work. We have another demonstration of knowledge despite ‘poverty of the input’, and it suggests that another basic aspect of grammar may be innate. …Gordon’s mice-eater experiment shows that in morphology children automatically distinguish between roots stored in the mental dictionary and inflected words created by a rule.” (Pinker, 1995, pp.146-47)


Pinker’s insistence that the children could not have learned this principle from observing somebody else’s speech makes me very sceptical about the rigour of Gordon’s experiment. For the suggestion that the children had an innate awareness that they were not allowed to use regular plurals in compounds could only be made by someone unaware that Italian parents’ genes, for instance, are apparently ignorant of the principle. In Italian there are not only compound nouns made with regular plurals as the second element, such as rompescatole (break-boxes/testicles = boxes-breaker, i.e. a bore, a pain in the neck), but also compounds with regular plurals as the first element, like fruttivendolo (fruits-seller).


On the subject of compounds, it is worth noting how the obsession with all-governing syntax leads to error concerning one aspect of language after another. Pinker (1995, p.386) repeats Chomsky's mistake of thinking that the position of the stress in word combinations (ráincoat as opposed to r cóat, for instance - my examples, not Pinker's) is dependent on the syntactic category. Compounds, he says, are stressed on the first element (dárkroom) while phrases are stressed on the second (dark róom). In fact stress is not dependent on syntactic categories at all, but solely on meaning. It is meaning that produces the different stresses in, for example, kítchen knife and kítchen ble. (In British English, at least, stress in this second type of combination tends to be equal on both elements. See Gethin, 1990, pp.98-104, and for an explanation of stress in word combinations. note4


Chomsky believes we are born with powers of abstract grammatical analysis, the ability to analyse sentences into their abstract 'phrase structure' quite independently of any meaning, even, indeed, if the sentences are meaningless. But this is not how either children or adults really experience language. For instance, if we consciously examine the sentence


The man the man the man knew knew knew.


it is comparatively simple to analyse it into an abstract 'phrase structure' - (x(x(xy)y)y) - but it is almost impossible to work out its meaning. This is because there are no clear images to fasten on to, to give us our bearings. It is, by Chomsky's criteria, a "well-formed" sentence, but because it is, effectively, just an abstraction, it leaves us mystified. Yet, although in formal abstract terms the following sentence is far more complicated, it is comprehensible precisely because it consists of recognizable meanings:


Did you realize that bomb a radical immigrant the finance minister that idiotic president appointed last year employs in his own private bank managed to make in the small amount of spare time the minister allows him, and put under the self-important fool's chair yesterday, was a toy?


Structure, syntax, grammar, these are a fantasy that learned people have believed in for centuries. Now, with the coming of modern linguistics, this fantasy has taken over the minds of most intellectuals.


Steven Pinker (1995, p.105) points out that words such as


destruction, way, whiteness, miles, hours, answer, fool, meeting, square root


are nouns, but not physical objects. He goes on (p.106):


"A part of speech, then, is not a kind of meaning; it is a kind of token that obeys certain formal rules, like a chess piece or a poker chip. A noun, for example, is simply a word that does nouny things; it is the kind of word that comes after an article, can have an 's stuck onto it, and so on."


Pinker here, I think, puts the cart well and truly before the horse. For how do we establish in the first place which words are going to be our nouns? We can't just decide in a vacuum, choose words at random and allocate them to the grammatical function of noun at random. The result would be meaningless chaos. There must be a criterion for deciding what sort of word is a noun, and the only possible criterion is meaning. It is simply not true that a noun is the kind of word that comes after an article or can have an 's stuck onto it, and so on. It need not have any of these things, but still be a noun - or, as I would prefer to say, still be 'thingish'. When lovers look in each other's eyes and say Darling! Tesoro! Liebling! Querido! Cherie!, the word has no grammatical environment or grammatical attachment at all and of course none is needed. What it has pleased the analysers to label as a noun has its 'thingish' meaning for the lovers independently of any so-called syntax.


Ironically, Pinker himself, in making his cart-before-horse-ish assertion, provides an excellent illustration of the point. "A noun," he says, " simply a word that does nouny things." He invents a word; he needs an adjective - or, as I would say, he needs a word that tells us 'this sort of, this quality'. He knows the meaning of -y, so he produces nouny.


So Chomsky and Pinker declare that grammar exists in the abstract, independent of meaning. A sequence of Chomsky's that has become famous in linguistic circles is

colourless green ideas sleep furiously

Chomsky says this is meaningless, but that English-speakers immediately recognise that it is grammatical. But this sequence is not meaningless. It is full of meaning, albeit very bizarre meaning. And that is precisely the only way we can tell that the sequence is 'grammatical'. We recognise the meanings of the individual words, not grammatical categories, and can see that the 'general' meanings fit each other: things can have qualities, they can do things, and do them in certain ways. If a sequence really is meaningless, we cannot analyse it. Or if a sequence could have meaning, but we don't know what that meaning is, we still can't give the words grammatical names, analysis is still impossible.

light fishes last

What's this? Has it meaning? Is it grammatical? We don't know, because we don't know what the meanings are supposed to be. This sequence could be an instruction to set fire to certain types of sea animal after everything else, or it could mean that sea animal species that are not heavy are durable, or that illumination goes fishing at the end. Or it could just be three words thrown onto the page at random. We don't know the meaning until we know the meaning, and by that stage there is no point in making any grammatical analysis. Grammar in the abstract, independent of meaning, is not possible, and is only the linguisticians' imagining. This is the basic flaw in Chomsky's argument. He presents the abstract names that grammarians have given to meanings as the controlling principle in language. He makes the names come first. He wants the tail of names to wag the dog of meaning. This cannot be.


Simple grammatical identification ("that's the object, this is a noun, not a verb...etc.") has a practical use for those learning foreign languages. Otherwise, grammatical analysis serves no purpose whatever.


In fact Pinker goes on from the passage quoted above:

"There is a connection between concepts and part-of-speech categories, but it is a subtle and abstract one. ...when we construe some aspect of the world as an event or state involving several participants that affect one [an]other, language often allows us to express that aspect as a verb. For example, when we say The situation justified drastic measures, we are talking about justification as if it were something the situation did, although again we know that justification is not something we can watch happening at a particular time and place. Nouns are often used for names of things, and verbs for something being done, but because the human mind can construe reality in a variety of ways, nouns and verbs are not limited to those uses." (Pinker, 1995, p.106.)

So Pinker has an inkling that meaning is perhaps involved after all, but won't acknowledge it directly. Instead he talks of a subtle and abstract connection between concepts and part-of-speech categories.


Actually, much of what he says here is true, but not for the reason he intends. He is unconsciously giving an account of the unsatisfactory way humans have forced life and reality into the straitjacket of language, tried to push square pegs into round holes.




Note 1. If he meant that a child cannot say "I" till she is aware of I, he was merely stating an obvious truth that tells us nothing about language. If he meant she cannot be aware of I until she speaks it, he is patently talking nonsense. "I" and "you" cannot be spoken sensibly without understanding first. And there is not much doubt that the cat too knows he is separate from me and other cats.



Note 2 In The Cambridge encyclopedia of language (Crystal, 1987, p.103) David Crystal describes a method of "working with semantic space". (The book that apparently launched the idea of 'semantic space' is, he says, a "pioneering work".) In his example of this method people place mammal names in a diagram where the vertical dimension judges ferocity and the horizontal dimension measures size. Thus cat, for instance, is located fairly near one corner (ferocious + small), and horse towards the diametrically opposite corner (non-ferocious + large); words for any other animals can be plotted on the diagram in the appropriate places. The more similar any animals are in size and ferocity, the nearer to each other they appear in the diagram. Crystal says this is a very simple analysis; but "the general approach is illuminating, with considerable research potential".

If that is an example merely of the trivial pointlessness of much of modern linguistics, here is an example of both pointlessness and absurdity:


"Transformational grammarians believe in a rule generally called "Chomsky-adjunction": I call it "C-adjunction". This rule is supposed to be part of both the D-structure/S-structure and S-structure/LFmappings. C-adjunction moves an element, creates a new mother node for the moved element and another node, and copies the label on this other node on to the new mother. I argue that there is no such rule...The argument considers: (1) C-adjunction and the theory of transformations; (2) nodes and labels, how each are licensed and how and where they are paired and their relation to C-adjunction...(4) the proper interpretation of the Projection principle...The main conclusion is that if there is a rule of adjunction, the new mother node created by this rule is unlabelled."


This passage is from a summary of an article by Robert Chametzky (Chomsky-adjunction) in Lingua in 1994, quoted by Michael Bulley in his article A wild gene chase in the July 1997 issue of English Today. In the same article Bulley has provided a clear basic description of transformational-generative linguistics:


"The theory, then, that grammar is biologically real is tied up with the transformational-generative tradition, which holds that the words we speak, write, hear and read are the surface-structure of language, which is derived by regular transformational processes, or rules, from basic structures that have a direct relationship with our genetic make-up. So you can see how, once this idea had taken hold, an immense body of writing has come into being in the last quarter of a century to try to tie down all this surface-structure to a set of rules to account for language being as it is."



Note 3 Another objection might be that there is often ambiguity in language anyway and that therefore the drawback of ambiguity in the case of mirror-reversing a three-word statement to form a question is insignificant. But it is hard to see how that justifies calling such a stark ambiguity a "just as" effective way of doing so. Language consists of words in single file, one after another, and that is the great restriction on how it is organized, not biologically imposed restrictions in the human mind. As a result of this basic restriction, when people start producing longer or more complex sentences, ambiguities are bound to arise. But that is no reason for them to say "Oh, well, there are going to be ambiguities anyway, so let's make it even more difficult for ourselves, even at the most basic level."


Pinker does not in fact raise the points of either intonation or ambiguity, perhaps because he too realizes they would be invalid. Or perhaps because, even if valid, they would hardly strengthen his case for the decisive nature of an innate awareness of abstract structure. Yet it is strange all the same that he does not attempt some explanation of how a first and last word flip or a mirror-reversal would be a practical way of forming questions, because as it stands his claim is so clearly false. (Pinker says nothing about this issue in his latest book, Words and Rules. For a brief comment on this work, see Postscript 1 below.)



Note 4 Because in two-element combinations there are effectively only two alternative stress patterns, it is not always easy to decide.






Postscript 1

Michael Bulley makes the following brief comment apropos Pinker's latest book, Words and rules:


"So Steven Pinker thinks that, if a 3-year old child says "I bringed you a cup of tea", that sentence belies a "mental task of almost bewildering sophistication" I think not. This is not a case of the complex workings of genetically embedded grammatical structures; for there are none. The child simply wants to communicate the idea "I’ve brought you a cup of tea", but is not yet at the stage of getting it right in conventional English. That he might say "bringed", or even "brung", is hardly to be marvelled at. It is exactly the same as if an adult were to say "I laid in bed" for "I lay in bed". It is just a mistake caused from lack of knowledge, and that is all there is to it."


Bulley writes with admirable clarity and logic on various language topics. If you are interested in exposures of the irrationality and frequent excesses of much of contemporary linguistics, read his articles in, among other journals, English Today. Some are listed in the references below.


Postscript 2

I have argued above that Chomsky's linguistic theories have basic weaknesses and should be seriously questioned.


On the other hand, it is sad that his political work is little known among the general public, and, where known, is largely ignored. Noam Chomsky has spent almost a lifetime, not seeking power, but supporting the cause of true freedom and justice, and speaking out in defence of the oppressed. The world might be a better place if it paid less attention to his linguistic message and more to his political one.




Bulley, Michael, 1997. A wild gene chase, English Today 51, July 97

Bulley, Michael, 1998. Not to be found in the brain, English Today 55, July 98.

Bulley, Michael, 1999. There ain't no grammaticality here, English Today 59, July 99.

Chomsky, Noam, 1968, in conversation with Stuart Hampshire. Broadcast by the BBC.

Crystal, David, 1987. The Cambridge encyclopedia of language, Cambridge University Press.

Dryer, M.S., 1992. The Greenbergian word order correlations. Language, 68, 81-138.

Gethin, Amorey, 1990. Antilinguistics, Intellect.

Magee, Bryan et al., 1982. Men of ideas. Some creators of contemporary philosophy, Oxford University Press.

Pinker, Steven, 1995. The language instinct. The new science of language and mind, Penguin Books.

Pinker, Steven, 1999. Words and rules. The ingredients of language, Weidenfeld & Nicolson.


The Editor welcomes your comments or contributions to discussion of this article.


It consists of extracts partly from the author's Language and thought: a rational enquiry into their nature and relationship, published by Intellect (, and partly from an article in The Raven 34, together with recently added material.


One of the most important aims of The English-Learning and Languages Review is to stimulate debate about the fundamentals of the global English-teaching industry. However, it has many other purposes. It gives advice on passing language exams. It presents information on English grammar, and in the future, it is hoped, on the grammar of other languages, often approaching problems from a new angle. It debates linguistic theory, and discusses language-learning principles. It invites contributions on all those subjects, and from those who find their own particular delight in the infinite variety of the languages of the world. It also welcomes material in languages other than English.

The English-Learning and Languages Review and its individual contributors assert their Copyright  © on all the material published in it. Nevertheless, the Review gives permission for unlimited reproduction of the piece above, The Mirage of Linguistics, on condition that:

  • acknowledgement is made of the source
  • no changes are made except with the Review's permission
  • no restraint is imposed on further reproduction of the material

·         copies are not sold


The English-Learning and Languages Review Ć Homepage