Linguist; drawing by David Levine

Can a study suffer from an Ishmael complex? Yes, if it has special difficulties in establishing its own independence in its peculiar field and methodology. For a while it may go on in a lonely, hoity-toity way, full of airs and ires, issuing denunciations, injunctions, and excommunications, venting its hostility towards whatever it supposes to have stood in the way of its advance. In time however, as it becomes more secure (and more divided), this aggressiveness abates.

For some decades linguistics has been the leading “example” of such behavior—much to the concern of that analogue to the U. N., the would-be United Studies. For how we should think about language matters very deeply in nearly all of them. If there is a study ready to settle this for us and intent upon doing so, we should indeed take note. And if this study claims to be alone in charge of saying how language works, more is needed. For premature confidence as to that, blindly followed in the schools, could do damage hard to repair. Inter-Study trespasses have often done so in less crucial areas and on a far smaller scale.

These two books are representative of present-day British linguistics. Both stem mainly from J. R. Firth and Daniel Jones. Both are well informed on other schools of linguistics, including past and present American movements, which are themselves only recently coming to take unpolemical account of their rival developments. Both make interesting and detached attempts to relate their own theory and model to those deriving from Leonard Bloomfield, Zellig Harris, Kenneth Pike, and others. Mr. Robins’s book, moreover, thoroughly earns its title. It can give the general reader who would like to know what has been and is being done a good over-all view and within a reasonable time. And there is enough detail and exemplification to let him see how it is done and this at all levels of analysis: phonological, grammatical, semantic. What will strike him most will be the amount of concern over which theoretical models should be used in description. In comparison, the new insight being gained into language may seem disappointing. Current linguistics regards itself as a young science. It is painfully conscious of duties that its new status as a science may entail. It is also sensitive lest it be supposed to be any the less for that among the humanities. And yet a reader coming to linguistics from literary criticism or philosophy who notes what stern, self-denying ordinances the linguist lays upon himself as a scientist, will wonder about the price. Are not its rigors, its ideals of power, economy, simplicity in explanation, its refusals to use in its formulations so much that it knows safely enough in other capacities, precluding it from being as helpful as it might be to itself and to other studies? It seems likely, however, that a strong reaction to this defensive isolationism has begun.

WHAT EMERGES from both books is how the pendulum of linguistic theory has swung between devotion to form and devotion to meaning and how these oscillations have speeded up. What was awesomely dominant a little while back can now seem only awfully queer. For example, only two decades ago any reference to meaning in a linguistic description was widely treated as “mentalistic” incompetence. Today, most linguists admit that language, being a notation for samenesses and differences among situations, is inescapably dependent on meaning. Both books insist that replacement of theories and models will “continue for many years to come.” And, of course, what is learned about languages will go on increasing, however often or much the descriptive apparatus may be changed. This cumulative improvement makes the scientific part of linguistics sharply different from, say, poetry, or party doctrine, or fashions in hair-dos, philosophy or education: other activities in which oscillations may also be observed.

The two-fold title of The Linguistic Sciences and Language Teaching may lead some readers to look there for this contrast. And they will find it. The authors are authorities in linguistics. That is where their treasure is. While they stay within linguistics, what they have to say is often valuable, sometimes trenchant, and occasionally seminal. When they turn to language teaching the atmosphere changes. What they write seems frequently perfunctory, as though somebody had asked them to and they were wondering what they could find to say. Apart from common-places, the discussion is oddly remote from real problems. They do indeed remark: “We feel that somehow comprehension should enter into the criteria of literacy,” but it is evident throughout that their interest in the anatomy and physiology of comprehension, in concept-formation and in cognitive or transactional psychology, is disablingly lacking. Something hides from them the relevancy to the language-learner of what language is for. In places they assert their amateur status: “The linguist can say what is a good description of a language, and can produce such a description. But he cannot say how the language should be taught. That is a matter for teachers and those who train the teachers.” This sounds modest and reasonable too—until you realize how badly those teachers and their trainers need help and how much more help they can be given than is offered here.


What is needed is something with which a more mature linguistics—of which Roman Jakobson is an illuminating exponent—can already equip teachers: a recognition of the key role of opposition in language use. As Saussure indicated and Jakobson and Halle have formulated more fully (Fundamentals of Language, p. 60), language holds together through requirements and exclusions: in other terms, through forms seeking closure and incompatibles seeking for place. Every speech-sound, every word, every sentence…does what it does through not being others which would replace it in another situation. If here, not there; if this, not that; if now, not then; if present, neither future nor past, and so on. The name for this need to choose between alternates which must be changed as circumstances require is opposition, and it is this concept which makes possible a direct teaching of how language works, by giving the learners properly arranged opportunities to see for themselves how it does its manifold jobs. Grammar used to pretend to teach this by telling its pupils about language. That never worked. Most of the revolt against grammar among teachers has come simply from their recognition that it does not succeed. It does not enhance either passive or active command of meaning. It distracts. Here the pendulum swings in teaching fashions become evident. After swinging blindly away from all concern with grammar, the profession is now swinging blindly back—to teaching a newer grammar: In this country it is chiefly some simplification of Chomsky’s transformationism. Textbooks applying the latest things (of five years ago) compete for the market. But the misdirection of effort remains uncorrected. It was not the badness of the grammar descriptions which caused the failure, but a simpler and deeper mistake: learning how to describe a language is not at all the same thing as learning how to use it with power and discernment. In point of fact, current efforts by English teachers to use “transformational grammar” far too often result in glib manipulation of nomenclature—just as of old—and play with “tree diagrams” without bringing any improved understanding of what sentences do or how they do it.

THE ROOT REASON is the difference between an intellectual apparatus designed to serve linguistic description, on the one hand, and a planned serial ordering of oppositions in sentences and situations on the other. The child’s introduction to reading and writing is all important. It can and should be the key to all the rest: the model of, the clue to, how a notation represents. Give him this and he can go forward clearsightedly: As reading becomes more established and ambitious in the mind, opposition, though still the essential principle governing meaning, ceases to need explicit exhibition. But without introduction to it the child remains all at sea as to what he is expected to look for and to do, reduced to guessing at his teacher’s wishes. The designer of this introduction should use his reflective knowledge of the language to contrive sequences of situations through which the learner can be invited to observe in detail without description) how changes in what is uttered vary systematically with what is meant. He will arrange these invitations so that they induce undistracted growth of discernment and assimilation in the learner. They can best do this by using opposition, the mainspring of comparison, linguistic and other, to engage the learner in seeing for himself the differences, for example between:

His hat is in his hands.


Her hat is on her head.

To see such things clearly is really to find out how language works. But to do so effectively the contrasting situations which give these sentences their meaning must be clearly presented to the learner: in actions or in depictions cleaned from all distraction, from all that is not strictly relevant to what the sentences are saying. (What is distractive is commonly what teachers have learned from publishers to call “attractive.”) Moreover, these invitations to observe how language works through comparing changes in sentences with changes in situations must be so arranged that what has been learned is used in the learning of what is new. And there are other such principles that have been vigorously preached and demonstrated for thirty years or more. They have been developed for the most part in the design of elementary film and TV language teaching and of programmed instruction, for which such sequencing is even more important than for the classroom teacher. Our authors complain sadly that “there exist very few statements of principles for the guidance of others” as to sequencing. But the need is not for ‘many statements but for clear and relevant ones. These have existed for some time but the authors have not been successful in finding them, perhaps because of a philosophic impediment to the thought of thought, which much of recent linguistics has inherited.


TWO IMPORTANT CONSEQUENCES concern views on observation and meaning. On observation we are offered a self-destructively limited view, due to according an inappropriate sense to observe. They assert that “concepts cannot be observed, but only postulated through the observation of particular events in particular languages” (page 153). How then do they suppose these observations of particular events to have been achieved? Every recognition, every identification, every differentiation among particulars takes place through observation of concepts: capabilities of apprehension. Every sentence in this book works only through writer and reader alike being able to observe concepts and their relations to one another: observing, for example, that two words are the same or that two senses are different or that two views are irreconcilable. And this in two of the main senses of observe: “notice” and “obey,” as in “Observe the NO SMOKING sign.” The authors, of course, know all this perfectly well. All their talk of “formal items” and of “abstraction” entails it. But they have at this point followed a fashion in giving to the word concept an irrelevant, in place of the required, meaning.

This is no quibble about a use of terms. It is a diagnosis. Here is a grave misconceiving of conceiving, an intellectual infection, which, where it is suffered, prevents linguistics from giving any but trivial advice to teachers since it cuts the needed connection between sentences and what they may be saying. When this spreads it does much damage to education—replacing thought in the classroom by parroting or by prudent conventional prattle.

To come now to meaning. It is as unnecessary to set up a conflict between form and meaning as to stage a fight between muscles and bones or between concepts and language. Language uses an incredibly elaborate system of co-operative relations among its parts at all levels in order to deal as best it can with the infinite variety of situations speakers and hearers may find themselves in. Every utterance has both a form and a meaning. The form has been developed to handle the meaning and the meaning must be explored both through the language and through examination of the situation the utterer is attempting to deal with. Both the utterance and the situation are the outcomes of systematic selections from among alternates. For the speaker, the form is his endeavor to accord in his language activity with the situation (as he sees it). For his hearers the meaning (as they in turn see it) arises from their endeavors to take account of and respond to the selections which have generated the form. Normally form and meaning vary together systematically, but naturally much goes wrong all the time, though strangely less—in some types of situation—than one would in theory expect.

Our authors evince somewhat divided minds in their remarks about meaning. (It is perhaps harder for three minds to avoid this old mind-trap than for one.) They quote Firth on Linguistics as the study of “how we use language to live” (page 151), a sensible formulation but hardly reconcilable with some of the other positions adopted. They go on to report him as emphasizing “that meaning was a property of all the types of patterning found in language; one could not describe language without describing meaning.” (Is this: “all these types of patterning mean; in describing language we should try to say how?”) They then proceed to denounce a selection of inferior attempts to describe meanings—not as inferior but for attempting to do it. (These attacks repeat with British examples a standard pattern of denunciation in recent U. S. linguistics.) Meaning as what x means and meaning as how x means are insufficiently distinguished throughout these pages: “a description must account for the internal patterns of language, and this can only be achieved if the criteria are drawn from within and not from outside language. At the same time all statements about language are statements of meaning and the task of the linguist is to work out, from observations of language in action, theories of how language works which will enable him best to make such statements” (page 151). But how is language in action to be observed unless we are ready to take into one view “the internal patterns of language” and much that is certainly “outside language”? More attention to how “the criteria” are, of necessity and in fact, “drawn”—both from “outside” and from “within” language—would start another revolution in linguistics.

THIS OPPOSITION of “within language” and “outside language” invites reflection. Take the somewhat important and typical contrast between he and she, for example. How far would you get with that if your criteria had to be drawn “from within” and not “from outside” language? You might get a big set of collocations (other words likely to be found in the same units of discourse). Boys’ names as against girls’ names, his as against her, trousers (once upon a time) as against skirt, pipe as against compact, etc. To pursue such collocations statistically via computers and large-scale, grant-consuming samples is, so these linguists aver, more reliable than to consider what he and she do for us in living. “The formal criterion of collocation is taken as crucial because it is more objective, accurate, and susceptible to observation than the contextual criterion of referential or conceptual similarity” (page 34). It is sad that linguistics should go on trying so hard to deprive itself of benefit from the analysis of situations—on which its avowed procedures do in fact themselves depend.

There are other strange omissions or refusals. Much praise (and space) is lavished on “Le Français Fondamental and the research work that led up to it” (1951…) as “great pioneering efforts,” in vocabulary selection. But no mention is made of the British and American labors to which the French effort was a response. “The obvious criterion for choosing one word and rejecting another seemed to be frequency of occurrence,” they write, as though this question had not been very fully thrashed out long before the French started. English Word Lists, Fries and Traver (1940), could have shown them in a few minutes how profoundly C.K. Ogden’s Basic English changed the contents of Beginning English Word Lists along with the principles of selection. “Words are not useful because they are frequent but frequent because they are useful” was his position. But to understand the importance of this change of angle, you have to think about “how language works” and about what happens when, as so often, it doesn’t. Excommunications are poor policy in scholarship. Basic English brought about a revolution in the design of elementary instruction in languages. A good number of eyebrows will be raised at the way this is here ignored.

More serious than omissions are misdirections. Languages with sufficiently different resources impose on their users importantly different frameworks of possible meanings. On this our authors seem to be in the grip of a quite non-scientific linguistic egalitarianism. They spend more than a little of their space insisting, with rebuke for other views, that “any language is as good as any other language.” Look again and you find this declaration is qualified by “essentially” (!) and that it is to be understood “in the sense that every language is equally well adapted to the uses to which the community puts it” (page 99). Maybe, but how do they know that? Have they an inventory of all the uses to which every community puts its languages? Similarly, and here again they are echoing the injunctions of American linguists, dialects are equally good. Is this science or just sympathy for the under-privileged? One is almost surprised that they do not carry on and extend this anti-snob principle to idiolects (the individual’s language resources) and so put an end to the case for education.

“THIS MISAPPREHENSION that some languages are intrinsically better than others” they are very severe on. “It is wholly false and can do a great deal of harm.” I do not know what “intrinsically” can do for them as an escape word. But certainly languages can be compared as regards what they can say and do. For some the performances possible within them are immeasurably greater than for others in the variety and extent and delicacy of the connections and distinctions they can compass. Less capable languages have not developed the necessary means. The authors are highly dogmatic on this matter, a sign perhaps that theory rather than actuality is guiding them. “All languages,” they say, “are capable of incorporating the lexical additions they require” (page 100). “Capable,” h’m; “require,” ha! Their reference is to computers.

Quite apart from this queer abstract confidence about some 3000 unknown grammars, addition of new vocabulary is in actuality by no means the comfortable matter here suggested. This is in fact a burning question, an atom bomb bonfire question. And our authors’ opinions here are fine examples of academic irresponsibility. Hardly any of the key terms of political thought in the West have as yet been assimilated into Chinese without stultifying distortions. No mutual comprehension in the most vital matters is as yet linguistically possible. And, reciprocally, the central moral terms, the names of the virtues in the Chinese tradition, are no more reproducible in English than, say sophyosyne. And for good reason. The situations these great key terms have to handle and to order and to maintain—their duties in the Western and the Chinese traditions as helping their users to live—are too different. Their concepts of “What to do,” as Confucius put it, are too deeply and organically (as with a blood type) connected with their norms of living. The gulfs between the traditions are too wide to be bridged merely by additions to vocabulary. To think they can be is to neglect values and much else in the anatomy of meanings. And alas, as Socrates showed Euthyphro, it is values that men fight for—less communicable than what they fight with. The future would look a lot brighter if the concepts a culture lives by could be grafted into another language as easily as the technicalities of the missile. And if men could more readily conceive that purposes they are opposed by may be as worthy as their own. But for that there are needed kinds of instruction in understanding which official linguistics as yet does not offer, an understanding which the aims at present preferred may easily obstruct.

This Issue

April 14, 1966