But What's a Dictionary For?

BERGEN EVANS, professor of English at Northwestern University, is known to a wide television and radio audience as one of our liveliest lexicographers and literary controversialists. He takes up, in the article that follows, the defense of WEBSTER’S THIRD NEW INTERNATIONAL DICTIONARY,which was assailed by Wilson Eollett in the January ATLANTIC.

THE storm of abuse in the popular press that greeted the appearance of Websterss Third New International Dictionary is a curious phenomenon. Never has a scholarly work of this stature been attacked with such unbridled fury and contempt. An article in the Atlantic viewed it as a “disappointment,” a “shock,” a “calamity,” “a scandal and a disaster.” The New York Times, in a special editorial, felt that the work would “accelerate the deterioration” of the language and sternly accused the editors of betraying a public trust. The Journal of the American Bar Association saw the publication as “deplorable,” “a flagrant example of lexicographic irresponsibility,” “a serious blow to the cause of good English.” Life called it “a nonword deluge,” “monstrous,” “abominable,” and “a cause for dismay.” They doubted that “Lincoln could have modelled his Gettysburg Address” on it — a concept of how things get written that throws very little light on Lincoln but a great deal on Life.

What underlies all this sound and fury? Is the claim of the G. & C. Merriam Company, probably the world’s greatest dictionary maker, that the preparation of the work cost $3.5 million, that it required the efforts of three hundred scholars over a period of twenty-seven years, working on the largest collection of citations ever assembled in any language — is all this a fraud, a hoax?

So monstrous a discrepancy in evaluation requires us to examine basic principles. Just what’s a dictionary for? What does it propose to do? What does the common reader go to a dictionary to find? What has the purchaser of a dictionary a right to expect for his money?

Before we look at basic principles, it is necessary to interpose two brief statements. The first of these is that a dictionary is concerned with words. Some dictionaries give various kinds of other useful information. Some have tables of weights and measures on the flyleaves. Some list historical events, and some, home remedies. And there’s nothing wrong with their so doing. But the great increase in our vocabulary in the past three decades compels all dictionaries to make more efficient use of their space. And if something must be eliminated, it is sensible to throw out these extraneous things and stick to words.

Yet wild wails arose. The Saturday Review lamented that one can no longer find the goddess Astarte under a separate heading — though they point out that a genus of mollusks named after the goddess is included! They seemed to feel that out of sheer perversity the editors of the dictionary stooped to mollusks while ignoring goddesses and that, in some way, this typifies modern lexicography. Mr. Wilson Follett, folletizing (his mental processes demand some special designation) in the Atlantic, cried out in horror that one is not even able to learn from the Third International “that the Virgin was Mary the mother of Jesus”!

The second brief statement is that there has been even more progress in the making of dictionaries in the past thirty years than there has been in the making of automobiles. The difference, for example, between the much-touted Second International (1934) and the much-clouted Third International (1961) is not like the difference between yearly models but like the difference between the horse and buggy and the automobile. Between the appearance of these two editions a whole new science related to the making of dictionaries, the science of descriptive linguistics, has come into being.

Modern linguistics gets its charter from Leonard Bloomfield’s Language (1933). Bloomfield, for thirteen years professor of Germanic philology at the University of Chicago and for nine years professor of linguistics at Yale, was one of those inseminating scholars who can’t be relegated to any department and don’t dream of accepting established categories and procedures just because they’re established. He was as much an anthropologist as a linguist, and his concepts of language were shaped not by Strunk’s Elements of Style but by his knowledge of Cree Indian dialects.

The broad general findings of the new science are:

1. All languages are systems of human conventions, not systems of natural laws. The first — and essential — step in the study of any language is observing and setting down precisely what happens when native speakers speak it.

2. Each language is unique in its pronunciation, grammar, and vocabulary. It cannot be described in terms of logic or of some theoretical, ideal language. It cannot be described in terms of any other language, or even in terms of its own past.

3. All languages are dynamic rather than static, and hence a “rule” in any language can only be a statement of contemporary practice. Change is constant — and normal.

4. “Correctness” can rest only upon usage, for the simple reason that there is nothing else for it to rest on. And all usage is relative.

From these propositions it follows that a dictionary is good only insofar as it is a comprehensive and accurate description of current usage. And to be comprehensive it must include some indication of social and regional associations.

New dictionaries are needed because English has changed more in the past two generations than at any other time in its history. It has had to adapt to extraordinary cultural and technological changes, two world wars, unparalleled changes in transportation and communication, and unprecedented movements of populations.

More subtly, but pervasively, it has changed under the influence of mass education and the growth of democracy. As written English is used by increasing millions and for more reasons than ever before, the language has become more utilitarian and more informal. Every publication in America today includes pages that would appear, to the purist of forty years ago, unbuttoned gibberish. Not that they are; they simply show that you can’t hold the language of one generation up as a model for the next.

It’s not that you mustn’t. You can’t, For example, in the issue in which Life stated editorially that it would follow the Second International, there were over forty words, constructions, and meanings which are in the Third International but not in the Second. The issue of the New York Times which hailed the Second International as the authority to which it would adhere and the Third International as a scandal and a betrayal which it would reject used one hundred and fifty-three separate words, phrases, and constructions which are listed in the Third International but not in the Second and nineteen others which are condemned in the Second. Many of them are used many times, more than three hundred such uses in all. The Washington Post, in an editorial captioned “Keep Your Old Webster’s,” says, in the first sentence, “don’t throw it away,” and in the second, “hang on to it.” But the old Webster’s labels don’t “colloquial” and doesn’t include “hang on to,” in this sense, at all.

In short, all of these publications are written in the language that the Third International describes, even the very editorials which scorn it. And this is no coincidence, because the Third International isn’t setting up any new standards at all; it is simply describing what Life, the Washington Post, and the New York Times are doing. Much of the dictionary’s material comes from these very publications, the Times, in particular, furnishing more of its illustrative quotations than any other newspaper.

And the papers have no choice. No journal or periodical could sell a single issue today if it restricted itself to the American language of twentyeight years ago. It couldn’t discuss half the things we are interested in, and its style would seem stiff and cumbrous. If the editorials were serious, the public — and the stockholders — have reason to be grateful that the writers on these publications are more literate than the editors.

AND so back to our questions: what’s a dictionary for, and how, in 1962, can it best do what it ought to do? The demands are simple. The common reader turns to a dictionary for information about the spelling, pronunciation, meaning, and proper use of words. He wants to know what is current and respectable. But he wants — and has a right to — the truth, the full truth. And the full truth about any language, and especially about American English today, is that there are many areas in which certainty is impossible and simplification is misleading.

Even in so settled a matter as spelling, a dictionary cannot always be absolute. Theater is correct, but so is theatre. And so are traveled and travelled, plow and plough, catalog and catalogue, and scores of other variants. The reader may want a single certainty. He may have taken an unyielding position in an argument, he may have wagered in support of his conviction and may demand that the dictionary “settle” the matter. But neither his vanity nor his purse is any concern of the dictionary’s; it must record the facts. And the fact here is that there are many words in our language which may be spelled, with equal correctness, in either of two ways.

So with pronunciation. A citizen listening to his radio might notice that James B. Conant, Bernard Baruch, and Dwight D. Eisenhower pronounce economics as ECKuhnomiks, while A. Whitney Griswold, Adlai Stevenson, and Herbert Hoover pronounce it EEKuhnomiks. He turns to the dictionary to see which of the two pronunciations is “right” and finds that they are both acceptable.

Has he been betrayed? Has the dictionary abdicated its responsibility? Should it say that one must speak like the president of Harvard or like the president of Yale, like the thirty-first President of the United States or like the thirty-fourth? Surely it’s none of its business to make a choice. Not because of the distinction of these particular speakers; lexicography, like God, is no respecter of persons. But because so widespread and conspicuous a use of two pronunciations among people of this elevation shows that there are two pronunciations. Their speaking establishes the fact which the dictionary must record.

Among the “enormities” with which Life taxes the Third International is its listing of “the common mispronunciation” heighth. That it is labeled a “dialectal variant” seems, somehow, to compound the felony. But one hears the word so pronounced, and if one professes to give a full account of American English in the 1960s, one has to take some cognizance of it. All people do not possess Life’s intuitive perception that the word is so “monstrous” that even to list it as a dialect variation is to merit scorn. Among these, by the way, was John Milton, who, in one of the greatest passages in all literature, besought the Holy Spirit to raise him to the “highth” of his great argument. And even the Oxford English Dictionary is so benighted as to list it, in full boldface, right alongside of Height as a variant that has been in the language since at least 1290.

Now there are still, apparently, millions of Americans who retain, in this as in much else, some of the speech of Milton. This particular pronunciation seems to be receding, but the American Dialect Dictionary still records instances of it from almost every state on the Eastern seaboard and notes that it is heard from older people and “occasionally in educated speech,” “common with good speakers,” “general,” “widespread.”

Under these circumstances, what is a dictionary to do? Since millions speak the word this way, the pronunciation can’t be ignored. Since it has been in use as long as we have any record of English and since it has been used by the greatest writers, it can’t be described as substandard or slang. But it is heard now only in certain localities. That makes it a dialectal pronunciation, and an honest dictionary will list it as such. What else can it do? Should it do?

THE average purchaser of a dictionary uses it most often, probably, to find out what a word “means.” As a reader, he wants to know what an author intended to convey. As a speaker or writer, he wants to know what a word will convey to his auditors. And this, too, is complex, subtle, and forever changing.

An illustration is furnished by an editorial in the Washington Post (January 17, 1962). After a ringing appeal to those who “love truth and accuracy” and the usual bombinations about “abdication of authority” and “barbarism,” the editorial charges the Third International with “pretentious and obscure verbosity” and specifically instances its definition of “so simple an object as a door.”

The definition reads:

a movable piece of firm material or a structure supported usu. along one side and swinging on pivots or hinges, sliding along a groove, rolling up and down, revolving as one of four leaves, or folding like an accordion by means of which an opening may be closed or kept open for passage into or out of a building, room, or other covered enclosure or a car, airplane, elevator, or other vehicle.

Then follows a series of special meanings, each particularly defined and, where necessary, illustrated by a quotation.

Since, aside from roaring and admonishing the “gentlemen from Springfield” that “accuracy and brevity are virtues,” the Post’s editorial fails to explain what is wrong with the definition, we can only infer from “so simple” a thing that the writer takes the plain, downright, man-in-the-street attitude that a door is a door and any damn fool knows that.

But if so, he has walked into one of lexicography’s biggest booby traps: the belief that the obvious is easy to define. Whereas the opposite is true. Anyone can give a fair description of the strange, the new, or the unique. It’s the commonplace, the habitual, that challenges definition, for its very commonness compels us to define it in uncommon terms. Dr. Johnson was ridiculed on just this score when his dictionary appeared in 1755. For two hundred years his definition of a network as “any thing reticulated or decussated, at equal distances, with interstices between the intersections” has been good for a laugh. But in the merriment one thing is always overlooked: no one has yet come up with a better definition! Subsequent dictionaries defined it as a mesh and then defined a mesh as a network. That’s simple, all right.

Anyone who attempts sincerely to state what the word door means in the United States of America today can’t take refuge in a log cabin. There has been an enormous proliferation oi closing and demarking devices and structures in the past twenty years, and anyone who tries to thread his way through the many meanings now included under door may have to sacrifice brevity to accuracy and even have to employ words that a limited vocabulary may find obscure.

Is the entrance to a tent a door, for instance? And what of the thing that seals the exit of an airplane? Is this a door? Or what of those sheets and jets of air that are now being used, in place of old-fashioned oak and hinges, to screen entrances and exits. Are they doors? And what of those accordion-like things that set off various sections of many modern apartments? The fine print in the lease takes it for granted that they are doors and that spaces demarked by them are rooms — and the rent is computed on the number of rooms.

Was I gypped by the landlord when he called the folding contraption that shuts off my kitchen a door? I go to the Second International, which the editor of the Post urges me to use in preference to the Third International. Here I find that a door is

The movable frame or barrier of boards, or other material, usually turning on hinges or pivots or sliding, by which an entranceway into a house or apartment is closed and opened; also, a similar part ol a piece ol furniture, as in a cabinet or bookcase.

This is only forty-six words, but though it includes the cellar door, it excludes the barn door and the accordion-like thing.

So I go on to the Third International. I see at once that the new definition is longer. But I’m looking for accuracy, and if I must sacrifice brevity to get it, then I must. And, sure enough, in the definition which raised the Post’s blood pressure, I find the words “folding like an accordion.” The thing is a door, and my landlord is using the word in one of its currently accepted meanings.

We don’t turn to a work of reference merely for confirmation. We all have words in our vocabularies which we have misunderstood, and to come on the true meaning of one of these words is quite a shock. All our complacency and selfesteem rise to oppose the discovery. But eventually we must accept the humiliation and laugh it off as best we can.

Some, often those who have set themselves up as authorities, stick to their error and charge the dictionary with being in a conspiracy against them. They are sure that their meaning is the only “right” one. And when the dictionary doesn’t bear them out they complain about “permissive” attitudes instead of correcting their mistake.

The New York Times and the Saturday Review both regarded as contemptibly “permissive” the fact that one meaning of one word was illustrated by a quotation from Polly Adler. But a rudimentary knowledge of the development of any language would have told them that the underworld has been a far more active force in shaping and enriching speech than all the synods that have ever convened. Their attitude is like that of the patriot who canceled his subscription to the Dictionary of American Biography when he discovered that the very first volume included Benedict Arnold!

The ultimate of “permissiveness,” singled out by almost every critic for special scorn, was the inclusion in the Third International of finalize. It was this, more than any other one thing, that was given as the reason for sticking to the good old Second International — that “peerless authority on American English.” as the Times called it. But if it was such an authority, why didn’t they look into it? They would have found finalize if they had.

And why shouldn’t it be there? It exists. It’s been recorded for two generations. Millions employ it every day. Two Presidents ot the United States — men of widely differing cultural backgrounds — have used it in formal statements. And so has the Secretary-General of the United Nations, a man of unusual linguistic attainments. It isn’t permitting the word but omitting it that would break faith with the reader. Because it is exactly the sort of word we want information about.

To list it as substandard would be to imply that it is used solely by the ignorant and the illiterate. But this would be a misrepresentation: President Kennedy and U Thant are highly educated men, and both are articulate and literate. It isn’t even a freak form. On the contrary, it is a classic example of a regular process of development in English, a process which has given us such thoroughly accepted words as generalize, minimize, formalize, and verbalize. Nor can it be dismissed on logical grounds or on the ground that it is a mere duplication of complete. It says something that complete doesn’t say and says it in a way that is significant in the modern bureaucratic world: one usually completes something which he has initiated but finalizes the work of others.

One is free to dislike the word. I don’t like it. But the editor of a dictionary has to examine the evidence for a word’s existence and seek it in context to get, as clearly and closely as he can, the exact meaning that it conveys to those who use it. And if it is widely used by well-educated, literate, reputable people, he must list it as a standard word. He is not compiling a volume of his own prejudices.

AN INDIVIDUAL’S use of his native tongue is the surest index to his position within his community. And those who turn to a dictionary expect from it some statement of the current status of a word or a grammatical construction. And it is with the failure to assume this function that modern lexicography has been most fiercely charged. The charge is based on a naïve assumption that simple labels can be attached in all instances. But they can’t. Some words are standard in some constructions and not in others. There may be as many shades of status as of meaning, and modern lexicography instead of abdicating this function has fulfilled it to a degree utterly unknown to earlier dictionaries.

Consider the word fetch, meaning to “go get and bring to.” Until recently a standard word of full dignity (“Fetch me, I pray thee, a little water in a vessel” — I Kings 17:10), it has become slightly tainted. Perhaps the command latent in it is resented as undemocratic. Or maybe its use in training dogs to retrieve has made some people feel that it is an undignified word to apply to human beings. But, whatever the reason, there is a growing uncertainty about its status, and hence it is the sort of word that conscientious people look up in a dictionary.

Will they find it labeled “good” or “bad”? Neither, of course, because either applied indiscriminately would be untrue. The Third International lists nineteen different meanings of the verb to fetch. Of these some are labeled “dialectal,” some “chiefly dialectal,” some “obsolete,” one “chiefly Scottish,” and two “not in formal use.” The primary meaning — “to go after and bring back” — is not labeled and hence can be accepted as standard, accepted with the more assurance because the many shades of labeling show us that the word’s status has been carefully considered.

On grammatical questions the Third International tries to be equally exact and thorough. Sometimes a construction is listed without comment, meaning that in the opinion of the editors it is unquestionably respectable. Sometimes a construction carries the comment “used by speakers and writers on all educational levels though disapproved by some grammarians.”Or the comment may be “used in substandard speech and formerly also by reputable writers.” Or “less often in standard than in substandard speech.”Or simply “dial.”

And this very accurate reporting is based on evidence which is presented lor our examination. One may feel that the evidence is inadequate or that the evaluation of it is erroneous. But surely, in the face of classification so much more elaborate and careful than any known heretofore, one cannot fly into a rage and insist that the dictionary is “out to destroy . . . every vestige of linguistic punctilio . . . every criterion for distinguishing between better usages and worse.”

Words, as we have said, are continually shifting their meanings and connotations and henee their status. A word which has dignity, say, in the vocabulary of an older person may go down in other people’s estimation. Like fetch. The older speaker is not likely to be aware of this and will probably be inclined to ascribe the snickers of the young at his speech to that degeneration of manners which every generation has deplored in its juniors. But a word which is coming up in the scale — like jazz, say, or, more recently, crap — will strike his ear at once. We are much more aware of offenses given us than of those we give. And if he turns to a dictionary and finds the offending word listed as standard — or even listed, apparently — his response is likely to be an outburst of indignation.

But the dictionary can neither snicker nor fulminate. It records. It will offend many, no doubt, to find the expression wise up, meaning to inform or to become informed, listed in the Third International with no restricting label. To my aging ears it still sounds like slang. But the evidence— quotations from the kiplunger Washington Letter and the Wall Street Journal — convinces me that it is I who am out of step, lagging behind. If such publications have taken to using wise up in serious contexts, with no punctuational indication of irregularity, then it is obviously respectable. And finding it so listed and supported, I can only say that it’s nice to be informed and sigh to realize that I am becoming an old fogy. But, of course, I don’t have to use it (and I’ll be damned if I will! “Let them smile, as I do now, At the old forsaken bough Where I cling”).

In part, the trouble is due to the fact that there is no standard for standard. Ideas of what is proper to use in serious, dignified speech and writing are changing — and with breathtaking rapidity. This is one of the major facts of contemporary American English. But it is no more the dictionary’s business to oppose this process than to speed it up.

Even in our standard speech some words are more dignified and some more informal than others, and dictionaries have tried to guide us through these uncertainties by marking certain words and constructions as “colloquial,” meaning “inappropriate in a formal situation.” But this distinction, in the opinion of most scholars, has done more harm than good. It has created the notion that these particular words are inferior, when actually they might be the best possible words in an informal statement. And so — to the rage of many reviewers — the Third International has dropped this label. Not all labels, as angrily charged, but only this one out of a score. And the doing so may have been an error, but it certainly didn’t constitute “betrayal” or “abandoning of all distinctions.” It was intended to end a certain confusion.

In all the finer shades of meaning, of which the status of a word is only one. the user is on his own, whether he likes it or not. Despite Life’s artless assumption about the Gettysburg Address, nothing worth writing is written from a dictionary. The dictionary, rather, comes along afterwards and describes what has been written.

Words in themselves are not dignified, or silly, or wise, or malicious. But they can be used in dignified, silly, wise, or malicious ways by dignified, silly, wise, or malicious people. Egghead. for example, is a perfectly legitimate word, as legitimate as highbrow or long-haired. But there is something very wrong and very undignified, by civilized standards, in a belligerent dislike for intelligence and education. Yak is an amusing word for persistent chatter. Anyone could say, “We were just yakking over a cup of coffee.”with no harm to his dignity. Hut to call a Supreme Court decision yakking is to be vulgarly insulting and so, undignified. Again, there’s nothing wrong with confab when it’s appropriate. But when the work of a great research project, employing hundreds of distinguished scholars over several decades and involving the honor of one of the greatest publishing houses in the world, is described as confabbing (as the New York Times editorially described the preparation of the Third International), the use of this particular word asserts that the lexicographers had merely sat around and talked idly. And the statement becomes undignified — if not, indeed, slanderous.

The lack of dignity in such statements is not in the words, nor in the dictionaries that list them, but in the hostility that deliberately seeks this tone of expression. And in expressing itself the hostility frequently shows that those who are expressing it don’t know how to use a dictionary. Most of the reviewers seem unable to read the Third International and unwilling to read the Second.

The American Bar Association Journal, for instance, in a typical outburst (“a deplorable abdication of responsibility”), picked out for special scorn the inclusion in the Third International of the word irregardless. “As far as the new Webster’s is concerned,”said the journal, “this meaningless verbal bastard is just as legitimate as any other word in the dictionary.”Thirty seconds spent in examining the book they were so roundly condemning would have shown them that in it irregardless is labeled “nonstand” — which means “nonstandard,” which means “not conforming to the usage generally characteristic of educated native speakers of the language.”Is that “just as legitimate as any other word in the dictionary"?

The most disturbing fact of all is that the editors of a dozen of the most influential publications in America today are under the impression that authoritative must mean authoritarian. Even the “permissive” Third International doesn’t recognize this identification — editors’ attitudes being not yet, fortunately, those of the American people. But the Fourth International may have to.

The new dictionary may have many faults. Nothing that tries to meet an ever-changing situation over a terrain as vast as contemporary English can hope to be free of them. And much in it is open to honest, and informed, disagreement. There can be linguistic objection to the eradication of proper names. The removal of guides to pronunciation from the foot of every page may not have been worth the valuable space it saved. The new method of defining words of many meanings has disadvantages as well as advantages. And of the half million or more definitions, hundreds, possibly thousands, may seem inadequate or imprecise. To some (of whom I am one) the omission of the label “colloquial” will seem meritorious; to others it will seem a loss.

But one thing is certain: anyone who solemnly announces in the year 1962 that he will be guided in matters of English usage by a dictionary published in 1934 is talking ignorant and pretentious nonsense.