The cover was produced by the transcriber, using elements from the original publication, and placed in the public domain.

PREFACE.

T HIS collection of essays will show very plainly that they were not written with a view to publication in a book. As a result of this, the subject-matter is not treated consecutively, systematically, or exhaustively. Some references to momentary events at the time of writing, even, have been left unchanged.

It is hoped, however, that, even with the acknowledged imperfections, the book may be found suggestive and useful by those to whose service it is dedicated in the title-page.

Some of the chapters are slightly technical, having been originally addressed to proof-readers only; but even these are thought to be sufficiently general in their composition to be interesting and useful to authors and editors.

CONTENTS.

PROOF-READING.

CHAPTER I.
THE PROOF-ROOM .

T HOUGH commonly acknowledged theoretically, the relative importance of good proof-reading is often practically unrecognized. Doubtless few of those who employ readers will assent to this averment, and the reason for their non-assent is also the basis of the assertion. Usually the proof-room is under the authority of a general foreman or superintendent, often not a good proof-reader himself, and who must necessarily devote most of his time to other matters. If the foreman is really competent to read proof, he will manage to secure and keep a force of good readers with less trouble than those have who are not so well fitted to judge the work done.

When good work is to be done—and where is the man who avowedly does not desire good work?—accomplished workmen are required, not properly in any one department alone, but all through; and perhaps this fact is partly responsible for the notion, not uncommon, but erroneous and costly, that almost any intelligent person can read proof.

Few persons realize fully the accomplishment and acuteness of perception necessary for the best proof-reading. 8 He is the best reader who, in addition to mechanical experience and accuracy, has a comprehensive education and can apply it practically. Of course, we can not expect our reader to know absolutely everything, but he should at least know enough to suspect error when there is evident occasion for suspicion, and challenge it for the author’s attention when that is possible. He should have general information sufficient to enable him to correct absolute error when he can not refer the matter to author or editor—a contingency frequently arising in newspaper-work.

Above all, the thoroughly accomplished proof-reader will know enough not to make changes in what is written when he has no right to do so. He will often know that what is written can not be right, and yet will have sense enough not to alter it without authorization. He will also have sense enough to assume a certain amount of authority on proper occasion, as in the case of an evident slip in the copy of work that has a set form. A good example is work like the definitions of verbs in the “Century Dictionary.” In these definitions the word to is used only with the first clause. The good proof-reader will have the word omitted even if it does happen to be in the copy, notwithstanding the strictest orders to follow copy; in fact, this is so plain a case that a very good compositor even would not set the word in the wrong place. Another forcible instance comes to hand at the moment of writing, in a letter written by a New York proof-reader, who mentions Assemblyman Amos J. Cummings. Mr. Cummings never was an Assemblyman. He is a Congressman , and Chairman of one of 9 the important Congress committees; moreover, he is an old-time New York compositor. When he was an editor on a New York paper another present Congressman was reporting Brooklyn news for the same paper. Almost every Brooklyn item sent in at that time had, in the writing, parallel streets reported as crossing, or cross-streets as being parallel; and these errors were frequently corrected in the proof-room.

The proof-reader who can and does make such corrections is much better for such work than one who merely catches typographical errors, even if he sometimes allows a wrong letter to pass in reading. Certainly a New York reader, especially a union man, should know better than to write of Assemblyman Cummings; and it would be well for all proof-readers to be sufficiently up in current affairs to correct the error, though it would not be fair to insist upon such correction as part of the reader’s qualification.

The present difficulty will never cease until the money value of good proof-reading is better recognized than it ever has been. At least one union in this country has always made a maximum weekly scale, and insisted upon classing readers with all other hands, at the same wages. Employers should insist upon paying as much over the union scale as they choose, and will always find it conducive to their interest to pay liberally for proof-reading and demand first-class work.

If any one is fortunate enough to have a first-class proof-reader in his employ, he will be foolish to let that reader go, if money—within reasonable bounds—will keep him. Fifty men may try to fill the place and fail before another really competent man is found.

10 A large proof-room should have its own foreman—not merely a head reader, but one actually in authority, just as any foreman should be, and with higher pay than the other readers have, and also with the chief responsibility. The room must, of course, be subject to the general foreman with regard to many details, whether it has a separate foreman or not; but, whoever is in charge, the readers should not be too much restricted in small, formal matters. An extreme instance that will illustrate practically what is meant by this arose through strict orders not to change anything from copy, too literally obeyed. A letter was missing from a word always spelled the one way, and the reader queried its insertion. He was an ordinarily good reader, too, who certainly had not the natural habit of doing anything stupid.

Undoubtedly better work will be turned out where there is no possibility of such queries being made, for the necessity of making them, under orders, imposes upon the reader an unfair burden of useless watchfulness that inevitably rivets his attention where it is not needed, and draws it away from matters that demand the utmost care.

CHAPTER II.
SOME PRACTICAL CRITICISM FOR PROOF-READERS .

A PERIODICAL highly esteemed in literary circles, in reviewing a book, said: “The proof-reading is so bad that we infer that its author could not have seen the proofs.” The publishers of the book do their own printing, and probably think their proof-reading is as good as possible, though they may realize that it is not as good as it should be. Many employers have had trying experiences in their efforts to secure good proof-readers, and such experience may have operated in favor of poor workmen, through sheer discouragement of their employers.

An inference that “its author could not have seen the proofs,” while possibly natural, is hasty; for, while many authors examine their proofs carefully, and are reasonably quick to perceive and correct errors, most authors are not good proof-readers.

Errors in print were quite as common as they now are when “following copy” was common, as it was in New York, for instance, about thirty years ago. One of the best offices in which a man could set type was Alvord’s, flourishing at the time mentioned. In it the compositor measured for his bill absolutely everything for which a customer paid, be it a cut, a blank page, or anything else. There, likewise, he was seldom called upon to change a letter or a point except to make it 12 like his copy. Certain large offices in New York now are like Alvord’s only in the fact that their proof-reading is not good—and the authors see most of the proofs. In one important matter these offices are utterly unlike Alvord’s—no compositor can earn decent wages in them.

Employers are largely responsible for the common poorness of our proof-reading, because they have not recognized the real nature of the work, and have insisted upon classing it as mechanical. Proof-reading will never be what it should be until the proof-reader ranks with the editor both in importance and in pay. With no more pay than that of the good compositor, and sometimes with less than the first-class compositor’s pay, the proof-reader’s position will not be adequately filled. Properly qualified proof-readers seldom remain long at the reading-desk, because they can and will do better elsewhere.

Something should be done to keep the best readers as such, for they are all climbing up into other fields of labor where they find stronger inducements, both in credit and in pay. Even in the case of our large dictionaries and encyclopædias, almost every one of which is decidedly bettered by the work of some one special proof-reader, there is little acknowledgment of the fact, and so there is little encouragement for the proof-reader to remain a proof-reader.

No one is surely fit to be trusted with proof-reading on particular work without having learned by practical experience. The best proof-readers must have as a foundation a natural aptitude, and they should have at least a good common education; but even these are not 13 sufficient without practical training. One of the poorest compositors on a New York morning paper was very helpful in the proof-room occasionally, while some of the best compositors were not so good at reading. It is undeniable that printers themselves make the best proof-readers when to their technical knowledge they add scholarship.

A first-class compositor is worthy of special favor, and generally gets it. A maker-up or a stone-hand who works well and quickly, or sometimes even one who does excellent work without great speed, is a treasure. Compositor, maker-up, and stone-hand, however, all do work that must be examined and corrected by the reader; and of course that reader is best who can also do any or all of the other work. What is said of the reader’s qualifications is not altogether theoretical; it is all in line with the practical needs of every good proof-room, and every employer wants a good proof-room.

The correction of the evil, which is certainly a desideratum, may be secured eventually in one way, and that way is the one necessary for authors as well as proof-readers. We need improved methods of general education. We need more general training and development of the thinking power. Seldom indeed do even our greatest thinkers reason sufficiently. No amount of argument could prove this assertion beyond question, but some examples will serve a good purpose as an object-lesson.

One of our most prominent philologists, a man of great learning, addressed a meeting of scholars, speaking strongly in favor of what he calls “reformed” spelling—which would be re-formed indeed, but is not 14 yet proved to be entitled to the epithet “reformed.” Here is one of his assertions: “One-sixth of the letters on a common printed page are silent or misleading. Complete simplification would save one-sixth of the cost of books.” Of course, he must have meant the cost of printing. Even with one-sixth less work in printing, very nearly the old cost of binding would remain, if not all of it; and any sort of good binding is no small item in the cost of a book. But one-sixth of the space occupied by the print would seldom be saved by the omission of one-sixth of the letters. The magazine article containing the report of the address is printed with the proposed new spelling. There is not a line in it that shows omission of one-sixth of the letters now commonly used in its words. One line in a paragraph of seven lines has “batl” for “battle,” and if the two missing letters had been inserted the word “the” might have been driven over into the next line; but the total effect on the paragraph of all possible changes would have been nothing—the same number of lines would be necessary for it. Certainly the assertion that one-sixth would be saved was not sufficiently thoughtful.

A recent pretentious work on the English language and English grammar (by Samuel Ramsey) would afford an example of loose thinking from almost any of its 568 pages. A few only need be given here. As to Danish influence on early English speech, it is said that “the general effect ... was to shorten and simplify words that were long or of different utterance, and dropping or shortening grammatical forms.” It should have been easy for the author to perceive that this sentence was not well constructed; and what can be 15 worse in a book on grammar than an ungrammatical sentence? We are told that a feature of English construction due to French influence is “the placing of the adjective after the noun, or giving it a plural form— sign manual, Knights Templars .” No English adjective ever has the plural form, and Templars is rightly pluralized simply because it is a noun. “No grammar will help us to distinguish the lumbar region from the lumber region,” Mr. Ramsey says. But grammar does help us by teaching us that lumbar is an adjective and lumber a noun. In careful speech accent would indicate the difference, which should be indicated in writing by joining the elements of the second term as a compound— lumber-region . In a chapter of “Suggestions to Young Writers,” the advice is given, “Let all your words be English, sound reliable English, and nothing but English; and when you speak of a spade call it by its name, and when you mean hyperæsthesia , say so.” If a young writer “says so” by using the word instanced, will he use “nothing but English”?

Lord Tennyson is reported to have said: “I do not understand English grammar. Take sea-change . Is sea here a substantive used adjectively, or what? What is the logic of a phrase like Catholic Disabilities Annulling Bill ? Does invalid chair maker mean that the chair-maker is a sickly fellow?” But Tennyson showed plainly in his writing, by making compounds of such terms as sea-change , that he felt, at least, that sea is not used adjectively, as “adjectively” is commonly understood. He must have thought that the phrase whose logic he asked for is wholly illogical and bad English, for he never wrote one like it. His own 16 writing would never have contained the three separate words “invalid chair maker”; he would have made it “invalid chair-maker” (or chairmaker) for the sense he mentions, and “invalid-chair maker” if he meant “a maker of chairs for invalids.” Tennyson certainly used English words well enough to justify the assumption that he knew English grammar passing well.

George P. Marsh, in a lecture on the English language, said that “ redness is the name of a color,” and John Stuart Mill made a similar assertion about whiteness in his book on “Logic.” Very little thought is necessary for the decision that neither redness nor whiteness is the name of a color, though each of the words includes such a name.

It is not fashionable nowadays to conclude with a moral, but this occasion is especially enticing, and here is the moral: Every proof-reader who cares for real success in his profession should cultivate the thinking habit, and learn not to jump to a conclusion.

CHAPTER III.
THE PROOF-READER’S RESPONSIBILITY .

S TRICTLY speaking, the responsibility of a proof-reader, on any kind of work, should be very narrowly defined. In an ideal state of affairs it would never go beyond close following of copy in every detail. Even that is by no means always easy, and for a reason that should cause writers to be very lenient with proof-readers. This reason is that writers make much manuscript that is almost positively illegible, and are often careless in many details that should be closely attended to in the writing. But, since there is little ground for hoping that writers will ever generally produce copy that can be reproduced exactly, the question remains open, How much responsibility must the proof-reader assume?

A good illustration of the legal aspect of this question is found in Benjamin Drew’s book, “Pens and Types,” published in its second edition in 1889, as follows: “In an action brought against the proprietor of Lloyd’s paper, in London, for damages for not inserting a newspaper advertisement correctly, the verdict was for the defendant, by reason of the illegibility of the writing.”

“Illegibility of the writing” is a more serious stumbling-block even than most writers know it to be, 18 although many writers do know that they are great sinners in this matter. Notwithstanding the fact that it has been a subject of wide discussion, much more might profitably be said about it, and it would be a great boon to printers if somebody could devise a way of instituting a practical reform in the handwriting of authors, editors, and reporters; but the incessant necessity of deciphering what is almost undecipherable is our immediately practical concern just now. What should be the limit of the proof-reader’s responsibility here?

Some time ago a New York paper had frequent articles in a handwriting so bad that the compositors were paid double price for setting type from it. One of the compositors, in talking with a proof-reader, expressed the opinion that the readers had very easy work, and part of his reason for the assumption was the fact (as he put it) that all the copy was read for them by the compositors before the readers got it. That same evening this compositor had some of the bad manuscript mentioned, and for what the writer had intended as “June freshets” the proof-reader found in his proof “Sierra forests.” Well, the compositor read the manuscript first, but how much good did that do the proof-reader? If the latter had passed the “Sierra forests” into print, he would have deserved to be discharged, for any intelligent man should know that one of the quoted terms could not possibly be used in any connection where the other would make sense. That compositor probably knew as well as the proof-reader did that what he set did not make sense, but he also knew that the proof-reader would have to do better with it, and that, no matter how much correcting he 19 had to do, it would pay him better to do it than to lose too much time in the effort to get it right at first. Again, the compositor had practically no responsibility in the matter, though the one who shows most ability in setting his type clean from bad copy is a better workman than others, and correspondingly better assured of good employment.

We have said that one who passed into print an error like the one mentioned should be liable to discharge. This is true, because no person reasonably fitted to read proof could fail to recognize it as an error. The best proof-reader who ever lived, however, might in some similar cases fail to read what is written exactly as it was intended in the writing. Unfortunately, it is only too often the case that proper names or generally unfamiliar words are written more illegibly than common words, and names so written may easily be misprinted after the best proof-reader has done his best with them. Where it is possible, it should be the most natural thing in the world for anything hard to decipher to be submitted to its writer. Commonly this can not be done on daily newspapers, because there are so many writers who are not within reach, reporters especially being generally away in search of news; but even in the offices of newspapers, in extreme cases, and with caution in deciding when it is well to do so, the matter should be referred to an editor, for it is to the editors that final responsibility for the wording of what is printed belongs.

What has been said seems well calculated to indicate clearly the limit which the writer would place in such matters upon the proof-reader’s real responsibility. 20 Naturally and equitably that limit is the exact reproduction of what is written, as to the wording, but including proper spelling and punctuation. Even at the expense of repetition, this seems to be a good place for impressing upon writers the urgent necessity for plain manuscript, in their own interest; for that is the only sure instrument to secure beyond reasonable doubt the accuracy that is desired by all writers.

No careful author will allow his book to be printed without reading it himself in proof; but this must be mainly for the wording only, as the printer’s bill includes pay for good proof-reading. Here matters are more simple as to the responsibility for getting the right words, as even hurried work from manuscript can generally be referred to the author in cases of real doubt. Occasionally this can not be done, but these occasions are comparatively rare exceptions. Submission of reasonable doubt to the author for his decision should be an important feature of the reader’s responsibility. It hardly seems necessary to dwell upon the question with regard to book-work, such work is seldom done without time for necessary consultation. It is in newspaper and job work that the greatest practical difficulty is encountered.

One of the greatest annoyances to the newspaper-publisher and the job-printer is the fact of having to reprint gratis advertisements or jobs when some error has occurred in the first printing. Shall the proof-reader be held responsible to the extent of paying for the work? Only one answer is possible—No! Yet the proof-reader should not expect too much leniency in this respect. He must be as careful as possible. 21 There is just one possible remedy for the trouble mentioned, and that is that employers do not expect too much of such work to be done by the reader, and that the reader insist upon having reasonable time in which to do it. Nay, the employer should insist upon having a proof-reader take sufficient time, in reading advertisements or job-work, to read closely, letter by letter; and this should be had, even at the expense of hiring an additional reader whenever such work becomes more in quantity than the force already employed can handle properly.

CHAPTER IV.
STYLE AND STYLE-CARDS.

A NEW YORK composing-room was run for many years without a regular style-card, and the foreman would not allow any posting of decisions as to style. When, however, an advertisement was printed with bar rooms as two words, and the foreman happened to notice it, the proof-reader was asked sharply, “What is our style for barroom ?” It was an unwritten but established law in the office that barroom should be one word; and the foreman, in that instance, did not think of the probability that the advertiser had insisted upon his own form for the term—as, in fact, he had.

In the office where this happened the workers were as little hampered with style as any workers possibly could be, and the foreman always said he would have no style; yet there certainly was a “style of the office,” with many absurdities, such as making base ball two words and football one word, capitalizing common words of occupation before names, as Barber Smith, Coachman Brown, etc. Some of the old-time absurdities have since been corrected, baseball , for instance, now being printed as one word.

In a neighboring office the opposite extreme is exemplified, the style-card being so intricate that some good compositors have worked there many years without 23 really learning in full the “style of the office.” Some of the compositors seldom do much correcting, but the average of time lost in making really needless corrections is unquestionably greater than in the office first mentioned.

Book-offices also have their own intricacies of style, with the additional bother of having to suit the varying whims of authors and publishers. “Many men of many minds” write for the papers, but their various whims need not be humored as those of book-writers must be. Authors of books frequently insist upon having things their own way, and too often the printers have to make that way for them, in opposition to what the authors write. This is certainly something for which the authors should be made to pay. If an author is determined to have certain matters of style conform to a certain set of whims, or even of good, logical opinions, he should write accordingly or pay extra for the necessary changes.

Nothing can be more sure than the fact that every printing-office must have some working rules of the kind classed as the “style of the office,” to which the work in general must conform, even when authors’ whims sometimes interfere. At present almost every office has some style peculiar to itself, that compositors and proof-readers must learn in the beginning of their experience there, and which they must unlearn on changing their place of employment. The greatest evil in this lies in the fact that many of the peculiarities are purely whimsical. Reformation is needed, and it is within the power of a body of proof-readers to devise and inaugurate a practical reform, by choosing from 24 among the various items of style those which seem best to a majority of the readers, and requesting their general adoption by employing printers.

Benjamin Drew’s book, “Pens and Types,” has a chapter on “style” that gives valuable hints for such work of reform. We are there told that the proof-reader “at the very threshold of his duties is met by a little ‘dwarfish demon’ called ‘Style,’ who addresses him somewhat after this fashion: ‘As you see me now, so I have appeared ever since the first type was set in this office. Everything here must be done as I say. You may mark as you please, but don’t violate the commands of Style. I may seem to disappear for a time, when there is a great rush of work, and you may perhaps bring yourself to believe that Style is dead. But do not deceive yourself—Style never dies.... I am Style, and my laws are like those of the Medes and Persians.’ And Style states his true character.”

Among the numerous differences of style mentioned by Mr. Drew are some that should not be classed as style, because one of the two possible methods is logical and right, and the other is illogical and wrong. For instance, Mr. Drew says: “Here, the style requires a comma before and in ‘pounds, shillings, and pence’; there, the style is ‘pounds, shillings and pence.’” Such a point in punctuation should not be a question of style, since one way must be better than the other as a matter of principle. In this particular case there is not only disagreement, but most people seem to have fixed upon the exclusion of the comma before the conjunction in a series of three or more items, notwithstanding the fact that its exclusion is illogical and as erroneous as any 25 wrong punctuation can be. The text-books, with very few exceptions, teach that the comma should be used; and, as said above, this seems to be the only possible reasonable teaching. Each item in such an enumeration should be separated from the next by a comma, unless the last two, or any two united by a conjunction, are so coupled in sense that they jointly make only one item in the series. This curious fact of common practice directly opposed to prevalent teaching is instanced as showing how erratic style is, and how necessary it is that the “style of the office” should be fully recorded.

Nothing could be more helpful than a style-card, especially if it be made the duty of some person to add thereto each new decision affecting style, so that the type may be set with certainty that arbitrary changes will not have to be made. Conflicting corrections are continually made by different proof-readers in the same office, and even by the same reader at different times. Such things should be made as nearly as may be impossible, and nothing else will accomplish this so well as a style-card that must be followed.

CHAPTER V.
WHIM VERSUS PRINCIPLE .

C ONSCIENTIOUS proof-readers are often confronted with the perplexing problem of dealing with the whims of authors and editors. One of the most difficult phases of the problem arises in the fact that proof-readers themselves are, equally with the authors and editors, possessed of whimsical notions, and the two sets of whims clash.

What shall the conscientious proof-reader do? He can not let everything go unchallenged just as it is written; if he does, he is not conscientious in the true sense of the word, though of course writers should know what they want, and should write their matter just as it is to be printed.

The only way successfully to combat unreasonable whim is by opposing it with true principle; yet even this will not always succeed. When a clear statement of principle fails to convince a writer that he is at fault, of course the proof-reader must yield, often to his great disadvantage. All intelligent people know that printed matter passes through the hands of a proof-reader, and they naturally attribute to his carelessness or incompetency all errors in printing. Examples are not lacking.

A paragraph in a magazine says that “the poet Will Carleton has established a monthly magazine, and 27 calls it Everywhere .” This is not a true announcement of the name, as Carleton splits it into two words— Every Where —and the word is so barbarously split each time it is used in his periodical. Any one noticing this form every where in print would naturally wonder why the proof-reader did not know better. It is a matter of personal knowledge that in this case the reader did know better, but Carleton stuck to his whim, saying that he had a right to make where a noun, whether others considered it so or not.

A New York newspaper says, with reference to political action, but in words equally applicable otherwise: “There is nothing that we know of in the Constitution of the United States, nor in the Constitution of any State, nor in the United States Statutes at Large, nor in any State law, nor any municipal regulation, that hinders any American citizen, whatever his calling or his walk in life, from making an ass of himself if he feels an irresistible impulse in that direction.”

Every man has a right to refuse to conform to general practice and principle, of course; but the arbitrary whimsicality shown in writing every where , and not everywhere , must fail to find its mate in any other mind, and can be applied to suit its writer only by himself. The only way to work for such a writer is to follow copy literally always. He has not a right to expect from the proof-reader anything more than the correcting of wrong letters.

Everywhere is an adverb of peculiar origin that may itself be classed as whim; but this whim is in accord with principle, and the one that splits the word is not. Probably the word was suggested by a question, as 28 “Where are certain things done?” Answers are often made by repeating a word prominent in the question, and so it must have been in this case, “Every where.” This simulated a noun qualified by an adjective, and the two-word form was used until people realized that it was not right grammatically. Many years ago the correct single-word form was universally adopted, and it should not be dropped.

Real principle forbids the unifying in form of some words that may seem to be like everywhere but are actually of a different nature. Anyone , everyone , and oneself (the last being erroneously considered as similar to itself , etc.) are as bad as single words as every where is as two words, notwithstanding the fact that they are often so printed. Tendency to adopt such whimsicalities of form is, for some unaccountable reason, very common. It is something against which every competent proof-reader should fight, tooth and nail, because it is subversive of true principle. The utmost possible intelligent effort will not prevent common acceptance of some forms and idioms that are, in their origin at least, unreasonable; but these particular abominations are not fully established, and there is ground for belief that their use may be overcome.

Some Latin particles are used as prefixes in English, and have not the remotest potentiality of being separate English words, if the matter of making words is to be controlled by real principle. One of these is inter , meaning “between.” A paper published in Chicago is entitled the Inter Ocean , making the only possible real sense of the title something like a command to “inter 29 (bury) ocean,” as inter is not, and never can be, properly an English adjective.

Many people are now printing as separated words such mere fragments as non , quasi , counter as in counter-suit and counter-movement , vice as in vice-chairman , and a few others, though the writer has not seen ante or anti so treated. These prefixes are all of the same nature, and if one of them is treated as a separate word, every one of the others should be so.

These are things that should be combated by proof-readers who know the main principles of language form, even though they know also that human perversity is sufficiently willful at times to persist in the face of all reason.

Another sort of whim has full swing on the New York Mail and Express . That paper prints the name of its own political party capitalized, and that of the opposite party with a small initial—Republican and democrat. How the editors can suppose that this belittles the Democrats is past finding out, since it should be a matter of pride to a true United States Republican that he is a democrat. Such ignoring of language principle is silly, and belittling to those who indulge it rather than to those at whom it is aimed. It is, however, beyond the proof-reader’s province, unless the reader is sufficiently familiar with the editor to influence him by moral suasion.

Notwithstanding the certainty that authors will be more or less whimsical, it is the proof-reader’s duty to do all he can to make the matter he reads perfect in every respect. He should be able to challenge anything that does not conform to generally accepted rules of 30 grammar, and to state clearly his reasons for desiring to make changes.

A thorough practical knowledge of English grammar is indispensable to a good proof-reader, though it counts for nothing without a quick eye to detect errors. If Bullions’s English Grammar had been read by a proof-reader as well equipped in grammatical knowledge as every reader should be, that book would have been cleared of one of the most ludicrous blunders possible. After stating that abridging is cutting short, examples are given, including the following: “When the boys have finished their lessons we will play. Abridged —The boys having finished their lessons we will play.” The second sentence is one word shorter than the first, but the tense is changed, and so, of course, the sense is changed. Real abridgment, of course, would not change the time from future to present; yet this is what a noted teacher does in each of his examples of abridgment, and it is something that a thorough proof-reader would have helped him not to do.

A proof-reader can not afford to neglect study, if he desires the best kind of success. The more he studies, the better able he will be to distinguish between whim and principle, and to combat one with the other when the first is not such that he knows it can not be combated successfully. Proper study, also, of men and events, as well as of language, etc., will enable him to distinguish helpfulness from what may be considered impertinence in making queries. By its aid he will be able to give a reason with each query, in a helpful way. Many queries on authors’ proofs pass unanswered, or are merely crossed off, because their point is not apparent, 31 or because they have been made in such a manner as to give offense.

In proof-reading, as in every other pursuit, the closest student of principles and of men will ever be the most successful. Generally, as we have said elsewhere, our best proof-readers eventually pass up to an editorial chair, or into literary or other employment which is more remunerative than reading proof. No employment should be more remunerative, unless it may be some which involves the control or disposition of large sums of money. A more difficult or rarer accomplishment than that of humoring authors’ whims, while still preserving much essentially good matter from the chaotic form it would assume at the hands of unpractical writers, would be hard to name.

CHAPTER VI.
AUTHORITIES AND OPINIONS .

I T has been said that in certain points of style no two persons would agree in their decision. The expression is too strong, but what is really meant is certainly true. Almost every question of style finds different answers.

This has been noted as an objection to the forming of proof-readers’ associations, the objectors assuming that none of the differences of opinion can be overcome. A contrary assumption must be the basis of accomplishment, and must be proved to be true, if anything is accomplished. Discussion must be had, full and free; every opinion that finds expression must be carefully considered, and all opinions carefully compared, in order to select the best. With this object clearly agreed upon, and always kept in view, and with each member of the association pledged to support the decision of the majority, would not much good result, at least in the way of agreement in matters that are commonly left to the proof-reader’s decision?

Except for the fact that nothing can be too foolish to find a parallel in history, the assertion might be made that our proof-readers could not be foolish enough to persist in holding individual opinions obstinately in the face of real proof that they are erroneous, or even that some other opinion is really more common and 33 therefore better. An instance that happens to present itself for comparison is the tulipomania, or “craze for tulips,” in Holland early in the seventeenth century. People were so crazy then as to sell and resell tulipbulbs at ridiculously high prices, even to the extent of creating a financial panic. Human nature is the same now as then; and although the matter of choosing between variant spellings, or other variations of style, never will create a financial panic, lack of agreement in choice does cause much annoyance, and even in some cases loss of money, by stealing compositors’ time through unnecessary changing of type. The “stylomaniac” is as foolish, relatively, as were the old Dutch tulipomaniacs.

Nothing could be more advantageous to a proof-reader than a full record of forms that could be followed without change. Such a record does not exist, and probably could not be made really exhaustive. It is doubtful whether any book or periodical ever fully reproduced the spelling of any dictionary, for the simple reason that lexicographers do not recognize the practical needs of printers. Spellings, word-divisions, and capitalization have never had, in the making of a dictionary, such analogical treatment as they must have to furnish thoroughly reliable guidance for printers; yet the dictionary is and must be the principal authority.

One remarkable instance of false leading has arisen through the old-time omission of technical words in dictionaries. Indention has always been the printers’ word for the sinking in of the first line of a paragraph, yet many printers now say indentation , because it was 34 discovered that indention was not in the dictionary. The right word is given by our recent lexicographers. Drew’s “Pens and Types” protests strongly against indentation , and MacKellar’s “American Printer” uses indention , which is probably an older word than the other. Old-time printers knew too much of Latin to put any reference to saw-teeth in their name for paragraph-sinkage, and indentation is properly applicable only to something resembling saw-teeth.

Printers and proof-readers must often reason from analogy in deciding how to spell. They have not the time to look up every word, and so they often differ from their authority in spelling. Every one knows how to spell referee , and, because of the similarity of the words, many have rightly printed conferee . A letter to the editor asked why a certain paper did this, and the editor answered that he would see that it did not happen again—because Webster and Worcester had the abominable spelling conferree! Why Webster ever spelled it so is a mystery, especially as it violates his common practice. Why Worcester copied Webster in this instance is a deeper mystery, since he had been employed on the Webster dictionary and made his own as much different in spelling as he could with any show of authority. The revisers of the Webster work have corrected the misspelling, and the other new dictionaries spell the word correctly.

Word-divisions are a source of much annoyance. Here again we have the lexicographers to thank, for no one of them has given us a practical guide. There are many classes of words that should be treated alike in this respect, and not one of these classes is so treated 35 in any dictionary. Here is a short list from the “Webster’s International”:

The one thing needed here is simplification. We should be at liberty to decide, without contradiction by our highest authorities, that if conductive is divided after the t , productive should have the same division. The difference arises from a false etymological assumption. One of the words is held to be made of two English elements—a word and a suffix—and the other is treated like its Latin etymon. True science would take the Latin etymon as the source of every word ending in ive , and divide every one of them between the consonants, regardless of the fact that some such words did not exist in Latin. It is sufficient that they all follow the Latin model, as conductivus . Many other terminations are properly on the same footing, as ant , ent , or ; they are not real English formative suffixes. In every word like those mentioned ending in tive after another consonant, the division should be between the consonants. This would be truly scientific, as no real scholarly objection can be made, and it leaves the right division in each instance unmistakable, no matter how little may be known of Latin or etymology.

Simplification is the great need in all matters of form or style—the easy and scientific conclusion that in all exactly similar instances the one reasoning applies, with the one result. The men who rank as our highest authorities as to spelling, and who should 36 be best qualified to lead us, lack one necessary accomplishment—a practical knowledge of the art preservative. Their efforts now are largely devoted to what they call spelling-reform, but their kind of reform is spoiling reform. English spelling is said by them to be absurdly difficult to learn, and they say they desire to make it easy by spelling phonetically. The matter is one of large detail, the phonetic spelling has many learned advocates, and there is a true scientific basis for many radical changes; but what is proposed as our ultimate spelling will be harder to learn, as it is now indicated, than is our present spelling.

Reform is needed, but not of the kind advocated by those who now pose as reformers. Universal agreement on a choice between traveler and traveller , theatre and theater , etc., would be highly advantageous; changing have to hav , etc., is merely whimsical, especially as some of the “et cæteras” are not so simple as they claim to be—notably the arbitrary use of both c and k for the k sound.

Our philologists are not likely to do for us what we very much need to have done.

Why should not the proof-readers do it for themselves—and also for the whole English-speaking world?

CHAPTER VII.
AUTHORITATIVE STUMBLING-BLOCKS IN THE STUDY OF THE ENGLISH LANGUAGE.

W RITERS for publication ought to write just as their matter should appear in print, but often they do not. Though every educated English-speaking person is expected to know how to use his own language correctly, no one needs such knowledge more than the proof-reader does. Very commonly matters of form, as punctuation, capitalization, compounding, and almost entirely the division of words at the ends of lines, are left to the proof-reader’s decision. How shall he decide reasonably if he have not the requisite knowledge? And how shall he have knowledge without study? And how shall he succeed in his study if he use not close thought and wise discretion?

The proof-reader, like every one else, must get at least the foundation of his knowledge through the medium of books. His practical use of knowledge, his faculty for instant perception of error, and his equally useful faculty for merely challenging what an author may wish to keep unchanged—all these must be acquired or confirmed by experience; but books must furnish the groundwork. One who desires thorough equipment as a proof-reader may never cease studying.

Good books on the English language are plentiful, but even the best of them contain statements that are 38 not beyond question. It is our purpose here to note a few questionable teachings, by way of warning against acceptance of anything simply because it is found in any book, and our most prominent example is from a work really good and really authoritative.

An incident will illustrate the aim of the warning. A customer in a New York store, taking up a book treating of word-forms, asked, “Does it follow Webster?” Information that its author had not closely followed any one dictionary, but had made the work for the special purpose of selecting the best forms from all sources, caused instant and almost contemptuous dropping of the book. Evidently that person had no idea that anything in language could be right if not according to Webster. Undoubtedly there are to-day thousands who would instantly decide such a matter in just this way. Each of them has always been accustomed to refer to some one authority, and to think that what is found there must be right. Indeed, so far is this species of hero-worship carried that a critic, reviewing the book on word-forms mentioned above, could hardly find words strong enough to express his condemnation of its author, theretofore unknown to the literary world, for daring to criticise statements made by noted scholars. It is amusing to recall the fact that one of the heroes of this champion’s worship began his career in exactly the way objected to, having devoted a large part of his first book to severe condemnation of some famous grammarians for doing something that he did himself, namely, copying and preserving errors.

Even yet we have not gone back to the earliest recorded condemnation of such hero-worship. One of 39 the most famous of the grammarians scored by our preceding hero was Lindley Murray, and his stated reason for writing on grammar was identical with that of his critic—the work of his predecessors was not sufficiently accurate. Long before Murray’s time, also, “peremptory adhesion unto authority,” as Sir Thomas Brown wrote in the seventeenth century, had been “the mortallest enemy unto knowledge, and that which hath done the greatest execution upon truth.”

Where can “peremptory adhesion unto authority” be found better exemplified than in children’s persistence in believing what they are first taught? Impressions made in childhood days certainly retain a strong hold long afterward, and this should be a powerful incentive toward giving them true impressions. One of the most popular language books now in use in primary schools, if not the most popular, has conversations between teacher and pupil. Here is one: “ T. —When I say, falling leaves rustle , does falling tell what is thought of leaves? P. —No. T. —What does falling do? P. —It tells the kind of leaves you are thinking and speaking of.” Is it not simply astounding that our children must learn in school that falling leaves means a kind of leaves?

There is plenty of the same quality in books at the other extreme of schooling—the very popular university grammar, for instance, William Chauncey Fowler’s “English in its Elements and Forms,” which says: “While language has power to express the fine emotions and the subtle thoughts of the human mind with wonderful exactness, still it must be admitted that it is imperfect as a sign of thought. It is imperfect because 40 the thing signified by a term in a proposition either does not exist at all in the mind of the hearer, or because it exists under different relations from what it does in the mind of the speaker. In other words, language is imperfect because the term in a proposition, if it has any meaning in the mind of the speaker, has a different one from what it has in the mind of the hearer. Hardly any abstract term has precisely the same meaning in any two minds; when mentioned, the term calls up different associations in one mind from what it does in another.... The phrase ‘beast of burden’ might, to one mind, mean a horse ; to another, a mule ; to another, a camel .... It should be added that there is great vagueness in the common use of language, which, in practice, increases its imperfection as a medium of thought.”

Yes, there is “great vagueness,” and here, in passing, is an amusing instance of it by a well-known writer on meteorology: “All cloud which lies as a thin flat sheet must either be pure stratus or contain the word strato in combination.” Did any one ever see a cloud containing the word strato in combination? “Great vagueness” is exemplified also in the grammarian’s own writing, and in a connection that demands a full exposition of it.

We need not quarrel with the expression “thoughts of the human mind” because we do not suppose that animals have mind; but certainly mind would be sufficient, without human , in discussing language. It is another matter, though, that the next sentence shows a constructive method at variance with the rules of grammar, and of a kind which the author himself brands as 41 false syntax in his exercises. Either in the sentence is not in correct construction with the complementary or ; it would be if because it were omitted—“because the thing ... either does not exist at all, ... or exists under different relations.” In the last clause, “it exists under different relations from what it does in the mind of the speaker,” what is improperly used, since the antecedent is plural— those which should have been used instead of what ; the construction makes does a principal verb, wrongly, because it is used for does exist or exists , and even with the right verb another preposition should be inserted, thus—“from those under which it exists in the mind of the speaker.” The whole sentence would have been much better expressed in this way: “It is imperfect because sometimes a thing mentioned is either not known at all to the hearer, or presents associations to his mind different from those conceived by the speaker.”

The third sentence ludicrously transposes speaker and hearer —“because the term, ... if it has any meaning in the mind of the speaker, has a different one from what it has in the mind of the hearer.” Possibly the writer accidentally placed these words in the wrong order, and the error is one of carelessness; but error it certainly is, for of course the speaker in every instance must suppose that his words mean something, whether his hearers think so or not.

In the fourth sentence “great vagueness” is again shown. What is the meaning of “when mentioned”? As here used, it can mean only “when a term is spoken of as a term,” and that is nonsense. The sentence 42 would be complete and accurately constructed without “when mentioned.”

The fourth sentence also contains the only so-called imperfection which the grammarian mentions, “beast of burden.” Undoubtedly there are many possibilities of ambiguity, but this phrase, chosen to illustrate imperfection, is really one of the beauties of the language. It is absurd to suppose that any one would attribute to such an abstract term a concrete meaning; but even if “beast of burden” does suggest to one person a horse, to another a mule, and to another a camel, there is nothing in that circumstance to prove that language is imperfect. All that is expressed in the phrase is “some kind of beast used for carrying,” and it is not said imperfectly. The imperfection is in the mind of the writer, not in the language—unless he can give a better example. If this author had omitted this section of his work, he would have shortened his book to the extent of half a page, and he would not have afforded a text for preaching against imperfection of mental training. If a thoroughly qualified proof-reader had suggested proper corrections, in the proper way, it must be that the matter would have been bettered; and every proof-reader should know how to make such suggestions.

CHAPTER VIII.
PREPARATION OF COPY .

W HILE it is very natural, in these days of great mechanical progress, that methods and machinery should be preëminent in printers’ literature, it should not be forgotten that the “art preservative” is not entirely mechanical. Our presses are not fed with paper until after the forms are fed from paper.

How much of the brain-work should be done by the printers, and how much by writers? Mr. Theodore L. De Vinne spoke as follows concerning this important question, at the bicentennial celebration of the setting up of the first printing-press in New York by William Bradford:

“I want to ask the question, What is the writer doing for us? Is he making his copy any better? Do you get any clearer manuscript than you used to? So far as handwriting is concerned, I should say no. What we get through the typewriter is better. The copy which the author furnishes has not kept pace with the improvement in machinery. Yet at the same time the printer is asked to do his work better and quicker than before. We are asked to make bricks without the proper straw. Too much is expected of printers in regard to this matter. I have been in the printing-office for nearly fifty years, and during that time I have had occasion to handle the copy from a great many authors, and from all ranks and conditions of men, and 44 I find that the compositor and the proof-reader are expected to do more work.

“There was a time when the printer was merely expected to follow copy. Now, I have no hesitation in saying that if every compositor was to follow his copy strictly, and if every proof-reader was to imitate his example, and neglect to correct errors; if books were printed as they are written, there would go up a howl of indignation on the part of authors as when the first-born of Egypt were slaughtered. I say that too much is expected of the proof-reader. He is expected to take the babe of the author and put it in a suitable dress for the public. The author should do it. Now and then you get an idea of how badly copy is prepared when out of revenge some newspaper editor prints it as the author sends it in. The reader, when he reads that copy, printed as it is written, with a misuse of italics, a violation of the rules of composition, lack of punctuation, etc., is astonished that a man of education can be so careless.”

Among other things following this, Mr. De Vinne said: “I wish to ask, on behalf of the proof-reader, a little more attention to the preparation of manuscript. The people who furnish the manuscript are not doing their share. I think it is an imposition that the proof-reader should do more than correct the errors of the compositor.”

We may well add to this plea on behalf of the proof-reader another on behalf of the compositor. Although so much type-setting is now done on time, many compositors are still at piece-work, and there is not one of them who does not suffer through the gross injustice of 45 losing time in deciphering bad manuscript. It is properly a matter of mere justice to the compositor that every letter in his copy should be unmistakable, and that every point in punctuation, every capital letter, and every peculiarity of any kind should appear on the copy just as the author wishes it to be in the printed work. Copy should be really something that can be copied exactly.

Certainly such copy is seldom produced, and there are excellent reasons for supposing that some authors—and many among the best—will never furnish plain copy in their own handwriting. One of the best reasons is indicated by this passage from a book entitled “Our English,” by Prof. A. S. Hill, of Harvard: “Every year Harvard sends out men—some of them high scholars—whose manuscripts would disgrace a boy of twelve; and yet the college can hardly be blamed, for she can not be expected to conduct an infant school for adults.”

Probably “manuscripts” refers mainly to handwriting, though it may include literary composition. The students have to take notes of lectures, and, in order to secure the largest amount of information, they write so rapidly that their manuscript can hardly be legible. Through this practice, rapid and almost formless writing becomes habitual.

Another justification for much of the bad handwriting of authors may be found in the fact that the matter is more important than the form, at least in the first making, and writers are comparatively few who can do the necessary thinking and at the same time put the thoughts on paper in perfect form. If an author can 46 write plainly and punctuate properly without losing any of his thoughts or sacrificing literary quality in any way, it is far better for his own interest, as well as for that of the printers, that he should do so; but where this is not the case it is necessary for some one to “put the babe of the author in a suitable dress for the public.”

Here is the point of the whole matter: If the work of finishing is to be done by the printers, they should be paid for doing it. There should be an extra charge for composition from poorly prepared copy, according to the extra amount of time required beyond that necessary in working from copy that can be read easily and followed literally. Nearly the full extra charge should be added to the type-setter’s pay, unless the proof-reader prepares the copy before the type is set, in which case, of course, the extra charge should be simply for his time.

Oliver Wendell Holmes, in “The Autocrat of the Breakfast-table,” says: “I am a very particular person about having all I write printed as I write it. I require to see a proof, a revise, a re-revise, and a double re-revise, or fourth-proof rectified impression of all my productions, especially verse.” A laudable desire to make his productions peculiarly his in all details must have been the incentive to all this work on proofs; but probably a close comparison of the finished work and the original manuscript would disclose many differences.

When good printers work from manuscript that can not be misread, with all details of spelling, punctuation, etc., properly attended to, and with explicit understanding that copy is to be followed literally, one proof 47 is sufficient for an author who does not have to make many changes in the wording of what has been written.

It will pay any author to make copy showing exactly what should appear in print, and to make every stroke of the writing unmistakable. If the writer can not himself produce such copy, his manuscript should be carefully revised by some one else. Any person doing such work of revision should be very cautious in order to preserve the writer’s intended expression, for often even an extra comma is disastrous. This applies also to proof-reading. The writer should be consulted, when consultation is possible, about changes from copy.

When authors have cultivated the habit of writing as they should write, or of having their copy made good for them, there will be no reasonable excuse for bad errors in printing. If Mr. De Vinne’s speech from which I have quoted, for instance, had been carefully revised by its author in the manuscript, a nonsensical misreading would probably have been avoided. One of his sentences as printed is, “We always understand how much the world is indebted to printing.” I have no doubt that he said, “We all of us,” etc.

No matter what plan is followed in its preparation, copy should certainly go to the compositor in such shape that he can read it easily and follow it absolutely. This is the only just way; and it is the surest way to secure good work.

CHAPTER IX.
COPY AND PROOF-READING .

I N a novel published some time ago, the copy contained a great deal of conversation that had to be printed in short paragraphs, each chapter being written in one long paragraph, with no quotation-marks, and almost no punctuation. The compositors had the injustice imposed upon them of breaking the matter into paragraphs, and supplying punctuation, with no recompense for doing this essential part of the author’s work. How such manuscript could secure acceptance by a publisher has never ceased to be a source of wonder, as it was not written by one whose mere name would carry it through; but a greater source of amazement is the fact that so many writers can make such abominable copy as they do make.

Certainly the writer should be the one most interested in having printed matter say what it is intended to say, and this can not be positively assured unless the written copy is accurate in form. Even the presence or absence of a comma may affect the sense in such a way that no person other than the writer can know positively whether the comma should be in or not.

Very few writers send to the printing-office such manuscript as every writer should furnish, yet they all demand accuracy in the printed matter. Let us make a bold proposition. Why should not employing printers of books combine in the determination to make an extra 49 charge for every alteration from copy, even to the insertion or removal of a comma? Why should not authors have to pay extra for the work that should be and is not done by them in the first instance? Even this, however, would not change the fact that much manuscript will not bear close reproduction in print. An author who was making many expensive alterations in proof was requested to revise his matter in manuscript, and returned it unchanged, saying that he could find nothing wrong in it.

Compositors have always labored under the injustice of being expected to punctuate the matter they set, regardless of bad punctuation in their copy. How can they know better than the author should know? This is an injustice to them mainly because they must often change the punctuation in type, thus losing time for which they are not paid. The decision is left to the proof-reader, and even the best and most intelligent compositor simply can not always be sure that he is doing what the reader will decide to be right. Other matters of style present the same difficulty.

If any particular style is to be followed, as in capitalization, punctuation, paragraphing, or any other formal matter, it is not just to demand that piece-workers shall set their type accordingly unless the copy is first carefully prepared. In other words, it is a matter of the merest justice to compositors that ordinarily they should be allowed to follow copy strictly in every detail. On some kinds of work this is not so essential, as on newspapers, for instance, where there are many writers, and matter of a certain kind is always to be set in the one way.

50 Publishers and editors of newspapers would be more just to all their workers, and probably more sure of getting what they want in style, if they could insist upon formal compliance at the hands of their writers rather than to throw the burden upon compositors and proof-readers. Responsibility for style does not rightly belong to the composing-room and proof-room; but if it must be assumed there, as commonly it must, every worker in those rooms should have an individual copy of a full and clear record of style. Those who receive work in book-offices, and who send it to the compositors, would certainly do well to question customers closely on all matters of style, especially in the case of anything other than plain reading-matter. It is well to have a distinct understanding with regard to complicated matter, and to record it when made, so that instructions may be clearly given to those who do the work.

An understanding having been had with the author or publisher, the manuscript should go first to the proof-reader and be prepared by him, so that the compositors need do nothing but follow copy closely. Of course this will not be necessary when the author furnishes good plain manuscript; but in other cases, of which there is no lack, it will surely pay.

The correction of authors’ errors is an important part of the reader’s duty, yet he should be very careful not to make “corrections” where there is a possibility that the writer wants just what he has written, even though it seems wrong to the reader. The proof-reader should not be held responsible for the grammar or diction of what he reads, except in the plainest instances, as there are many points of disagreement even among 51 professed grammarians. Plain errors in grammar or diction, as those following, the good proof-reader will correct.

A New York newspaper mentioned Frenchmen who “content themselves with sipping thimbles full of absinthe.” The reader should have known that the men do not use thimbles for the purpose of drinking, and that thimblefuls are what they sip.

When the proof-reader had a paragraph saying that “the arrivals at the hotels show a falling off of over 100 per cent.,” he should have known that this is an impossibility, since it leaves the arrivals less than none.

When another reader saw something about “the buildings comprising the old brick row,” he should have corrected it to composing . Buildings compose the row, and the row comprises buildings.

It would not be fair to expect every proof-reader to be thoroughly up in zoölogical nomenclature. No reader, though, should pass a word like depuvans unchallenged, because that is the best he can make of what is written. He should ascertain in some way that the word is dipnoans , or query it for some one else to correct. On the “Century Dictionary” the editor struck out a quotation, “The miracles which they saw, grew by their frequency familiar unto them.” His pencil happened to cross only one word in the first line, and the next proof sent to the editorial room contained the passage, “The miracles which they grew by their frequency familiar unto them.”

These are a few instances of remissness on the part of readers, the last one showing absurdity that should be impossible.

52 Some things are commonly expected of proof-readers that they can not with any reason be asked to do. When a person whose initials are J. J., for instance, writes them I. I., it is not reasonable to expect them to be printed J. J. A script I is one thing and a J is another; and no one can possibly know that the one which is written is not the right one when there is no clue, as there would be in Iohn. One lesson that writers seem bound not to learn is that proper names should be written plainly. When not written plainly they are very likely to be printed wrong.

Some kinds of changes proof-readers should not make, even if they think the writing is wrong. When a plainly written manuscript, showing care at all points, contains something about the “setting up of the first printing-press,” this should not be printed “setting-up of the first printing press”; neither should some one be changed to someone , though the barbarous someone happens to be the “style of the office.” There is no good reason for making a compound of setting up , and there is no reason for making anything but a compound of printing-press ; and someone should certainly be removed from the “style of the office” and the correct some one substituted. These two examples are selected because they were convenient, not for criticism merely, but to enforce the fact that, at least in a book or any work not containing matter from various writers, carefully written manuscript should be followed in every respect. Some authors have in this matter a just cause of complaint against printers; but it is really the result of carelessness on the part of authors in not writing as their matter should be printed and insisting upon having what they want.

CHAPTER X.
THE DICTIONARY IN THE PROOF-ROOM .

I T is said that Horace Greeley’s estimate of qualification for proof-reading called for more general knowledge than one would need in order to be a good President of the United States. By this he meant, of course, ability to read anything, from the smallest job, in the commonest language, to the most learned and most scientific writing, and to know that every thing is made right. How many proof-readers can do this? Not many. Horace Greeley knew very well that the world could not furnish such men for the proof-reader’s desk—and yet his remark was justifiable even from a practical point of view.

A recent paragraph in a trade publication said truly that “even the daily newspapers use so many foreign and technical terms as to demand a high grade of excellence among the readers.” This was said in connection with an assertion that pay for the reader’s work, and especially for the best work, is higher now than ever before. We might easily show that this is not absolutely true, for very high pay has been given for high-class work in the years that are gone, and the writer of this essay can state from personal knowledge an instance of higher pay than the highest mentioned in that paragraph; and it may be well to tell of it, because it will serve as a good introduction to our present theme. The paragraph says that its writer personally knew of two 54 men who were paid $50 a week for reading. If these men were mere proof-readers, their pay was very high; but it is not unreasonable to suppose that their work nearly approached the responsible editorial status. On a certain large work published many years ago a man was employed as proof-reader at what was then excellent pay. When that work was revised he was still known as the principal proof-reader, but his work included final editing of the copy, as well as reading the proofs, which latter he did in a critical way, making such changes in the matter as he knew were necessary. For this work he received $75 a week, and the only men known to the present writer who were paid as much as the sum first mentioned did the same kind of work.

In each of these cases the money was paid because of one qualification that stood in place of general knowledge, rather than for the actual possession of such knowledge that seems to be demanded by Horace Greeley’s estimate. Each of these readers had at hand a good reference library, and knew where to look for information on any question that arose. The special qualification was the ability to perceive or suspect error of statement, and to correct it through positive knowledge, in many cases with no need of reference, but more frequently through consulting authorities. An important complement of this qualification is the perception of correctness as well as of error, and ability to leave unchanged what is right as well as to change what is wrong.

Of course one who is really fitted to read proof must know how to spell all the common words of the 55 language, and this is not so general an accomplishment as it is naturally supposed to be. Many writers are somewhat weak in spelling, and the proof-reader must correct their errors as well as those made by compositors, for often the editors can not take time for such work, and copy is sent to the composing-room just as it is written. But few proof-readers, if any, know all the words that may rightly be classed as common. It is a matter of recent experience that one who ranks among the best of newspaper readers, in reading market reports, changed the lower-case initial of muscovado to a capital, and thought the name was a proper noun until another reader, happening to have the same matter in hand, changed the capital letter to lower-case and was called upon to give a reason for it. Recently, also, a good proof-reader allowed the term “Romance languages” to pass as “romance languages.” Romance in this use should not be unfamiliar, yet it was mistaken by compositor and reader as the common noun romance , which mistake should be impossible, as every one should know that romance is not confined to any special languages.

What such people need is a good dictionary at hand and constant use of it. Of course no busy proof-reader, especially during the rush of newspaper work, can stop every few minutes to find a word in the dictionary—much work must be dashed off at lightning speed, or as near that as possible, and no sort of interruption can be tolerated, even at the expense of printing a few typographical errors. But how much more creditable it is to the proof-reader if, even in the utmost rush, he can detect and mark all the errors, whether time can be taken to correct them in the type or not.

56 Few readers, comparatively, seem to realize the wonderful helpfulness of intimacy with some good dictionary, for very few of them use one as much as they would if they realized it. Probably most of them will continue to do just as they have always done—taking it for granted that they have no need of frequent consultation of the dictionary; but if something can be written that will impress even a few with a desire for the improvement to be attained through study of the dictionary, it is worth while to try to write it.

Every proof-room should possess a good dictionary. Some people think that every proof-room of any consequence does possess a good dictionary, but a little inquiry would soon convince them that this is not so. Many readers are left to do their work without even such aid in the way of reference, notwithstanding it is a fact that no certainty of good work can be had without it, and that many more works of reference are indispensable as aids to the best work. There are an amazing number of proof-rooms that are not supplied even with an old Webster’s or Worcester’s Dictionary, and a great many more than there should be that have only one or the other of those antiquated works. Once upon a time they were both good works, because they were the best yet made. But lexicography has progressed, and we now have dictionaries that surpass the old ones, in every respect, as much as our new books on any scientific subject outrank those of our forefathers.

The Century and the Funk & Wagnalls Standard dictionaries contain practically full records of our language in all details, almost sufficient to take the place of a large reference library, so far as the proof-room is 57 concerned. One or the other—or better, both—should be in every proof-room, and the proof-reader who makes the most constant studious use of one or both will soon find himself on firmer ground than he could otherwise occupy.

CHAPTER XI.
THE PROOF-ROOM LIBRARY .

H OW many proof-rooms are as well equipped with books of reference as they should be? The proprietors of some large establishments have always recognized their need and endeavored to supply it, but it is not far from the truth to say that very few employers, if any, have done all that would be profitable in this matter. A good selection of the latest reference books is seldom found in a proof-room, notwithstanding the fact that their intelligent use is one of the most important adjuncts of good proof-reading.

Reasons could easily be found for the common lack of books other than a general dictionary, or that and one or two special technical glossaries; but it will be more advantageous to give reasons why proof-readers should have and use more books than most of them do use.

Professional men have to read continually to keep up with progress in scientific knowledge. It is absolutely necessary to their success. Each of them, however, has a special demand for some particular branch of knowledge. The books these men consult are written by specialists, who choose their own subjects, and of course know the special words that must be used. A proof-reader, on the contrary, can not choose his subjects. He must undertake what is ready for him, whether it be some ordinary work, using common words only, or a scientific book filled with unfamiliar words. 59 Authors of scientific works often make abominable copy. They do not realize that the terminology so well known by them is not equally well known to the workers in printing-offices, and the most particular words are frequently written more carelessly than the common words in their manuscript. Of course these authors read their own proofs, and most of them think they are very careful in doing it; but they are not trained proof-readers, and they see the words in full rather than the individual letters, so that a wrong letter easily evades their notice. When the trained proof-reader does not know the particular words, and has no means at hand for their verification, the result is bad.

A pamphlet on ichthyological terminology will afford a good illustration. Its author wrote what was intended for “the shorter termination -pidæ is adopted rather than -podidæ .” This was printed with dashes instead of the hyphens, “termination— pidæ rather than— podidæ .” The pamphlet has Opisthrarthri and Tenthidoidea instead of Opistharthri and Teuthidoidea , and many other typographical errors in such words. Probably the proof-readers did their best to follow copy, and thought the author would be sure to correct such errors as they failed to find. If in each doubtful instance they had consulted a reasonably full list of ichthyological names, as they should have done, most of the errors might have been corrected. Proof-readers should certainly have some means of handling work intelligently, and the only way this can be done is by verification through the use of reference books.

Our general dictionaries have never attempted to give full scientific vocabularies. In fact, the two most 60 used—the old Webster and Worcester—are nearly useless in this respect, giving only the few purely scientific terms that had become familiar when they were made. Even technological terms were not freely inserted in their making. Later dictionaries, however, have increased their vocabularies very largely by adding the special terms of science. The Imperial, which is very much like a larger Webster Unabridged, contains many names of families and genera in natural history, also many special words of other science; Webster’s International has more of all kinds than the Imperial; the Century Dictionary has more than the International; but they all come far short of the full vocabulary of any science.

Forty years ago Mr. G. P. Marsh, in his “Lectures on the English language,” quoted from a scientific journal a sentence containing thirteen botanical words that have not even yet found their way into the dictionaries above mentioned, one of these words being the adjective cissoid , meaning “like ivy.” He also said, in the same lecture: “Indeed, it is surprising how slowly the commonest mechanical terms find their way into dictionaries professedly complete.” Mechanical terms, however, as well as botanical and others, have found their way into dictionaries since Mr. Marsh’s time freely, but by no means exhaustively.

Chemists and medical men string together words and word-elements almost ad nauseam , so that common dictionaries simply can not attempt to record all their combinations. Unless the proof-reader is thoroughly versed in the Greek words used by the doctors, and in the names of elements, etc., as used by the 61 chemists, his only hope rests upon special medical and chemical works. As an amusing instance of what he may have to decipher—doctors and chemists are commonly able to write illegibly, and often do so—a few words not in the general dictionaries may be cited. Chemists use words like aldehydodimethylprotocatechuic —a combination of aldehyde , dimethyl , and protocatechuic . A little thought will suffice to perceive these elements in the ugly-looking word, and in others like it; but that is not equally true in the case of such a term as androgynoarion or engastrimythismus .

Examination of any special scientific work would disclose easily the fact that the proof-reader may be called upon at any moment to read proofs of language he does not know, and can not verify without special reference books. He should not be expected to do good work without such aids.

CHAPTER XII.
THE COPY-READER .

M UCH has been written about the proof-reader and his duties and responsibilities, but comparatively little about his assistant, commonly known as the copy-holder. This name “copy-holder” is in its most frequent application a misnomer, and that is why we prefer to consider the majority of the assistants as “copy-readers,” a name, by the way, that is not new here, but has much local currency. Real copy-holders are found mainly where proof-readers work in pairs, one reading from the proof and the other following on the copy and telling when that is different from what is read. Occasionally it may be that proofs are read in this way by one regular reader and a mere holder of copy, but as a rule such work is done by a team of readers equal in standing, who alternate in the reading. Such is the common method on morning papers. On evening papers it is not unusual for the proof-reader to relieve his assistant occasionally by reading aloud from the proof, but as a rule the assistant reads from the copy, and so is a copy-reader. The distinction between “holder” and “reader” is not generally important, but is useful for the purpose of this chapter.

Until comparatively a few years ago nearly all the reading of copy was done by boys, mainly for very low 63 pay, as the real importance of the work was not yet apprehended. Now, however, we have accomplished almost a complete revolution, and copy-reading is understood to demand intelligence and quick thought of an unusual order, among young persons at least. The nearer a reader of copy comes to being truly qualified for being a proof-reader, the better for that one’s welfare, and the more fortunate the proof-reader who has that person as an assistant. That last word is just right, for a good copy-reader is truly an assistant to the proof-reader.

Some very foolish things have been said about copy-readers, and none more foolish than this one from a paper read before a society of proof-readers: “Proof-readers complain of the bad copy they have to study over. Who has to read that copy—the proof-reader or the copy-holder?” Another saying in the same paper may well be connected with this for consideration. It is: “I have known of proof-readers dozing—and even going to sleep—over proofs.” Unfortunately, the truth of the accusation can not be doubted; but it is really only one phase of something that is true of a majority of workers at anything—they do not always faithfully perform their duty. The copy-reader who takes the trouble to try to be sure that nothing is read when the proof-reader does not hear it is sure to be a dutiful and conscientious worker; yet is not even that a real duty, as well to one’s self as to one’s employer?

Again, it is the proof-reader’s duty to know that copy is read correctly—not merely to make his proof conform to what he hears, but to know that he is making it like the copy, when it should be so, which is nearly 64 always. The responsibility for getting the matter right on the proof properly belongs to the proof-reader always—never in the slightest degree to the copy-reader, with any propriety. A proof-reader has no real right, under any circumstances, to shield himself from blame by saying that “the copy-holder must have read it wrong.” Nothing could be meaner than that. But he must have some protection against such accidents, and there is a manly remedy in insisting that he shall be the judge of the copy-reader’s efficiency, or else that there shall be a distinct understanding that he must take the necessary time to verify what is read whenever he suspects it, by seeing the copy. In fact, the verification and the suspicion when necessary are very important to the proper performance of a proof-reader’s duty. This does not mean that a copy-reader has no responsibility, but only that that responsibility does not properly extend to the finished work. It is in this sense that proof-readers rightly speak of their having to study over bad copy.

Another foolish direction about copy-reading is the following, from Benjamin Drew’s book, “Pens and Types,” referring to the reading of Greek: “The method of reading will, we think, be sufficiently exemplified if we give but one line, which should be read by the copy-holder thus: Cap. K, a, grave i; t, acute u, m, b, long o subscript; k, r, long e, p, circumflex i, d, a; p, short e, r, acute i, g, r, a, ph, short e; cap. P, short e, r, s, i, k, grave short o, n; cap. smooth acute A, r, long e.” One of the best proof-readers the writer knows would not understand such mummery, because he does not know the Greek alphabet. Moreover, the 65 reader who wastes his employer’s time in having such spelling done is defrauding the employer. Such work should always be compared. The main purpose in referring to this, however, is to note the fact that both proof-reader and copy-reader are much better equipped for their work if they know the Greek alphabet than if they do not know it. And they are still better off for each additional acquirement of unusual knowledge.

A copy-reader will always find knowledge of any kind useful, and one who is ambitious and eager for advancement will be a close and ceaseless student, always acquiring new information, not only in books and periodicals, but in and from the persons and things with which one is surrounded. Particularly desirable is acquaintance with proper names of all sorts, and with important public events. So long as the world lasts, probably, reporters and editors, yea, and even authors of books, will write proper names and unusual words less legibly than they write common words. Even when reporters try to make names plain by writing each letter separately, they often form the letters, or write them without real form, so that little hope is left of absolute certainty in deciphering them. The writer has seen names in roman printing characters that would have been easier to read if written in the ordinary way with any care. Familiarity with the names likely to be written will enable a reader to master the writing with much more certainty and greater ease. In cases where no means of familiarity exist, as with initials of unknown persons, it frequently happens that the best effort of either proof-reader or copy-reader must be mere guesswork. If, as often occurs, a person’s initials are J. J., 66 and they are written I. I., and the name is not positively known, no one can tell whether they will be printed right or wrong.

The information that is most useful generally is that which gives ability to distinguish words by their meaning, and to recognize a word unmistakably through the sense of the other words of the sentence, or sometimes through a clue given in the whole context. Very few persons really know as much in this way as every one should know. A study of etymology is very useful, and the ambitious copy-reader can not afford to neglect it. Knowledge of the elements of words is one of the most helpful kinds of knowledge. So is knowledge of diction, or the right choice of words, and of syntax, or the right association of words. The writer once wrote an article in which he used “protocatechuic” as a test word, and wrote it as plain as any print, but the corrected proof sent to him had the word printed “protocatechnic,” showing plainly that the test had been too much for the reader. This probably resulted from the reader’s ignorance of the word “catechuic”; but not only every good proof-reader, but also every good copy-reader, should know that word.

Unfortunately, there are many “cranky” proof-readers who are not patient with a copy-reader who hesitates while deciphering bad manuscript. Nine times out of ten the proof-reader himself could do no better, notwithstanding that the responsibility is really his, and that special ability in such work is one of his most important qualifications. Well, such a proof-reader is simply not a gentleman, and no remedy suggests itself. As nearly as the writer can decide, the copy-reader 67 under such circumstances must either “grin and bear it” or find another situation. As in all relations in life, patience and forbearance on both sides are necessary for comfort, if not rather more so here than in most relations.

CHAPTER XIII.
PROPER ORDER OF PARTS IN A BOOK .

T HE subject of this chapter is suggested by a letter mentioning differences of opinion of various authors and publishers. Without that suggestion the chapter would never have been written, because one arrangement is so common that the writer has never thought it came short of universality. Indeed, many books have been examined since receiving the letter, and all show the same arrangement. But this, while constituting evidence of agreement among the makers of these books, is really stronger evidence of the fact that even in dealing with commonplaces it pays to be cautious in making assertions about the prevalence of any practice, and especially in asserting that anything is universal practice.

Personal experience and research fail to disclose any arrangement other than this: Frontispiece, title-page, copyright, dedication, preface, contents, list of illustrations, errata, introduction, text, index. Of course not all books have all of these features, and some books have others not here given. For instance, sometimes there is a publisher’s note, giving some explanation or announcement. Often that may appropriately occupy the copyright-page, with the copyright beneath it. Again, “Errata” are comparatively seldom given, but not 69 seldom enough. Genuinely good proof-reading would reduce the necessity to almost nothing; but genuinely good proof-reading is itself a rarity.

Now, using some of the caution that has been indicated as necessary, it must be admitted that some difference of opinion exists, and that the arrangement given here is not universal. What is the printer to do if the customer wishes some other arrangement? What is the proof-reader to do if he finds the parts arranged in an unusual manner?

Every printer who wishes to secure and keep a reputation for doing good work must attend to preservation of the proprieties as far as he can secure that. He can not, as a rule, take the matter of arrangement into his own hands, any more than he can rewrite or edit his customer’s work. Occasionally, but very exceptionally, he may be authorized to change the order or even the substance of what is to be printed, but probably no one would attempt it without distinct authorization, unless it might be one of those few who can afford to insist upon having work done in a certain way. A printer who can dictate methods or styles, with the alternative that otherwise he will not do the work, must be one who has secured sufficient permanent custom to make it unimportant whether anything more is done or not. This amounts practically to an assertion that, within reason, the customer must be allowed to have his way. But most customers are amenable to reason, and it may be suggested that it would be well to propose a change to one whose book-manuscript is wrongly arranged. Consulting a few books will show a general practice, and this, with the statement of that practice 70 already made before looking at the books, should be convincing.

What has the proof-reader to do with this? Well, the careful proof-reader will look after all details and endeavor to get everything right. If authors wrote exactly as they should write—so that every letter and every point in their manuscript could be reproduced in print without a change—proof-readers need be nothing more than they are commonly paid for being. They would then have little to do beyond comparison of proof and copy, for the purpose of correcting compositors’ errors. Authors do not and will not prepare manuscripts as carefully as they should; indeed, they simply can not always do so, often through lack of time, and too often through inability. Many of them actually do not know how to punctuate, and they are not few who do not even know how to spell as all should know. Therefore the proof-reader must be qualified at all points for correcting not only the compositor’s work, but also that of the author.

The particular matter that we are considering is not likely to come into question before it is taken up in the composing-room, where the foreman may notice the arrangement if it is wrong, and consult some one for authority to change it. Many foremen would be likely to make it right without consultation, and then the question would arise only if the customer directed a change on the proofs. Should the foreman not notice the order—most good foremen would, though—the matter would probably come to the proof-reader unchanged, and it is as much his duty to look after this as to do anything else. Unless specifically instructed 71 beforehand, he should call attention to the error, and have it corrected if he can.

Proof-readers should be able to give a reason for everything they do or desire to do, and in this, as in all matters, there are good reasons for one method and against others. Let us take the features of the book in order as given. First, the frontispiece. Why, of course. The very name places that first, as the piece for the front or beginning. It is the picture or piece that fronts or faces the title-page. This seems hardly open to question, yet the letter mentioned above did not so place the frontispiece, and it may be just possible that the position had been disputed.

Equally unquestionable seems the position of the title-page. All writings begin with a title, so that must be the first page of reading in the book.

As the title-page necessarily is backed by a page on which no real division of the book can begin, since all beginnings are made on odd-numbered pages, it is backed by the copyright, and the dedication, as being also something not connected logically with any other part, follows next.

If there is no dedication, the preface, as merely something about the matter of the book, follows the copyright. Good reason is found for this in the fact that the preface is that which is thought necessary to say just before beginning the book proper.

Before we begin the text, however, it is thought well to state in detail what is to be found in the text, so here we place the table of contents, always properly beginning on an odd page and followed logically by the list 72 of illustrations if there is one, as that is itself really contents.

All of these features naturally lead up to the main body of the book, therefore they should all come before that. This is said before mentioning the introduction because of the logic of circumstances. An introduction, as its name implies, is that which introduces the subject of the book. It is sometimes made the first chapter of a book, which is a sufficient indication of its natural position.

Last of all should be the index, because it is a résumé, and that can not reasonably be given until we have given that upon which it is founded. It can be made only after the text is finished, therefore its natural position is after the text.

CHAPTER XIV.
THE BOOK MAKE-UP .

P RACTICAL knowledge and ability in making up book-work are acquirable only through experience. The process might be clearly described in all its details, covering the entire range from the simplest page, of a certain number of lines all of the same type, to the most complicated congeries of different-sized type and small cuts, tables, or anything else, and yet the closest student of the description would never know how to do the work properly until he had done some of it. What is meant by this may be elucidated by means of a story of personal happening, though not dealing with any attempt at written instructions, but rather with assumption from observation, and possibly some little previous experience, on the part of a compositor.

Some time ago I was foreman and proof-reader of the book-room of a large jobbing establishment in New York. Having a large pamphlet in hand, with three sizes of type, including a number of tables, and to be printed from the type, the make-up was left till the last, as a separate and special piece of work. Among the compositors were two with whom I had been associated more or less for years, so that I knew their capabilities. One of these two was first out of copy at the end of the job, so that, all things being equal, the make-up should have gone to him. All things not being considered equal, the make-up was reserved for the other 74 of the two mentioned, who was not ready for it until most of the men had been told there was no more work for them just then. My old acquaintance who had been passed by said nothing at the time, but went out and fortified himself with fire-water and came back, accompanied by one of the prominent union politicians, to “make a kick.” His argument was that, as he was out of copy first, he was entitled to the making-up work, which was admitted, with the qualification that the office was entitled to my best effort to have the work done right, and so the man thought best able to do it was the only one to whom it could be given conscientiously, notwithstanding our recognition of the union, with all that that implied. This was met with a contemptuous sneer at the idea that anything so simple as the make-up should be kept for a certain man at the expense of another. “What one man can do another can,” said the slighted one; and thereby he exposed the weakness of his position, for many men can do even the simplest work much better than many other men. “Making up!” he exclaimed; “putting in a lead, and taking out a lead, and tying a string around the page! Making up!”

Well, is making up anything more than this man said it was? Possibly not, except that there is a right way to do these things, and there are many wrong ways. Besides, the greatest objection in the case given was the man’s known inexperience of imposition. That objection would apply comparatively seldom now, as letterpress printing is done much less than it was. Still, practical knowledge of imposition is really as necessary now to the fully competent compositor as it ever 75 was, for with it he is enabled to undertake work that otherwise he can not do.

Before the making up is begun the size of the page must be determined. There is not and can not be any general rule for proportions, since commonly many circumstances must be considered of which the maker-up knows nothing, and frequently he must simply follow the directions of the foreman. One thing, however, the wise maker-up can always regulate. He should see that his page is exactly gauged to a certain number of lines of the type most used in the text, since that is the only sure guide to uniformity of length in the pages. It is not likely that any foreman will ever object to a slight change in the gauge for this purpose, if it happens that he has made or ordered one that does not conform to it.

Positive directions for determining the size of a page have been published, but I know of none that will properly apply in all cases, notwithstanding their positiveness of expression. Following is what Marshall T. Bigelow says in his “Handbook of Punctuation”: “In determining the form of a page of an oblong shape, whatever its size, a certain proportion should always be maintained. The diagonal measure of a page from the folio in the upper corner to the opposite lower corner should be just twice the width of the page. This is no arbitrary technical rule, but is in conformity to the law of proportion establishing the line of beauty; it applies equally to all objects of similar shape, and satisfies the eye completely. A long brick-shaped page or book will not look well, however nicely it may be printed. When we come to a quarto or square page, the true proportion 76 of the diagonal to the width will be found to be as 10½ : 6¼—the size of a good-shaped quarto—instead of 2 : 1, as in the oblong, or octavo. And this shape also proves as satisfactory to the eye as the former one. However large or small the page may be, these proportions should be maintained for a handsome book.” These proportions are maintained in the book from which we quote, but its pages would have been much better in shape a little narrower and a little shorter. Many handsomer books have pages that do not conform to Mr. Bigelow’s rule, though the proportions given by him are good as a general guide. A “Printers’ Grammar” published in 1808 has “a long brick-shaped page,” and is a good-looking book. It says: “Should the length of the page be left to the discretion of the compositor, he sets so many lines as he conceives a fair proportion, which is generally considered as double its width.” The page in which this is printed is not quite twice as long as its width, yet it is exceptionally long for its width, judged either by other books of its own time or by later books.

If the size of the page is not dictated by the customer—very often he will indicate it by means of some book whose size suits him—the foreman or employer will be guided by the size of the sheet and the amount of matter. Of course everybody knows this, but it is a part of the proceeding that it may be well to mention, and that may be dismissed after remarking that the length of the page should usually be such as to leave the margins nearly equal.

Practice varies somewhat as to the length of title-pages, some being sunk a little from the top, some a 77 little shorter and some a little longer than the other pages. Ordinarily they should be exactly the same as other pages in length. The usual title-page gains nothing by either shortening or lengthening. There being differences of opinion in this respect, however, compositor and proof-reader should learn what is wanted in the office where they are employed and act accordingly.

When very little matter is to occupy a page by itself, as bastard titles, copyrights, dedications, etc., the matter should stand a little above the middle of the page. Practice differs here also, some books having such pages exactly centered, and some having them placed almost two-thirds of the way up. One of the best of the old-time New York offices had a rule that a copyright, bastard title, or anything of that kind should have just twice as much blank below as there was above. All such pages in their books looked inartistic, because of such misplacing of the matter, though otherwise the taste shown was excellent. The effect generally desired is that such matter should appear at a glance to be in the center of the page, and this effect is better produced by placing the matter actually a little higher up, but only a little.

The sinkage of chapter-heading and similar pages is a matter not often treated in books, and for which there is no fixed rule. Here, again, Mr. Bigelow comes near to stating the best practice, though circumstances often necessitate differences, and tastes differ, so that it may easily happen that a customer will order a sinkage not in keeping with Mr. Bigelow’s rule, which is: “The first page of the text of a book should have about 78 two-thirds of the matter of a full page. Where chapters or other divisions occur, a uniform sinkage of the same division should be kept up through the book. In poetry this should be done as nearly as possible; but allowance may be made for the different stanzas which occur, so that they may be divided properly. A useless repetition of a half-title over the first page following should be avoided.” There are things in this that I can not understand. What does the last sentence mean? What is the exact intention of the sentence about poetry? But the prescription of uniform sinkage is good, and for the commonest sizes of pages the proportion given for the first page is about right. For a chapter-heading elsewhere in the book the same sinkage as the actual blank at the top of the first page should be used.

There are other points about the make-up of books that every compositor and proof-reader should know, but they hardly come into question, being always treated alike by all people concerned, and will be learned in the right way only through actual experience.

CHAPTER XV.
SOME QUESTIONS ANSWERED .

F OLLOWING are a few actual questions of general interest, with their answers, as they were given in the “Proofroom Notes and Queries” in The Inland Printer . In each instance the letter precedes its answer, the two being distinguished by the use of different type.

GRAMMAR AND DICTION.

Do you write “1½ inches,” or “1½ inch”?

The difficulty in deciding this question is purely logical. Two or more things must be named to justify the plural verb, says Logic, and “one and a half” is less than two. But “one and a half” is more than one, and the singular verb is grammatically restricted to one only; therefore the grammatical rule should apply, and the plural verb be used with any subject that must be read as “one and something more,” even if the something is only a fraction.

Which sentence is grammatically correct—“Ten dollars was paid,” or “Ten dollars were paid”?

Simply as a matter of grammar, with no deference to sense, the second sentence is right; but as a matter of fact, unless ten separate dollar coins or bills are paid, which seldom happens, “was paid” is much more accurate, as the real meaning is, “The amount of $10 was paid”—one thing that is named by the words that express its equivalent in smaller amounts. “Ten 80 dollars” is logically singular when it means one amount of money, and so is “ten million dollars,” although grammatically plural; therefore it is better to use the singular verb for the common intention in sense.

Is it proper to say, “Nine and six is fifteen”?

Those who insist that the rules of grammar should govern all such expressions use the plural verb in such cases, and say “Nine and six are fifteen,” because the words used express more than one thing, and that is plurality. But the logic of it is that “the sum of” the two is so much, and many scholars consequently favor the singular verb.

A correspondent incloses an advertisement containing the sentence, “Failures is the current talk now days,” and requests an opinion as to its correctness.

The sentence is clearly ungrammatical, but it is not uncommon to violate grammar rules in this way under certain circumstances, and it is to be presumed that the writer thought of such circumstances, though he may not have done so. If he thought of a number of individual failures in the plural sense, and wrote “is” to go with the clearly plural sense of the noun, he did not express his thought correctly. But he may have thought of “failures” simply as one subject of talk, and this would at least so far justify the singular verb as to leave its correctness open to discussion. We may say, “‘Failures’ is the subject of his lecture,” and reasonably expect that no one will criticise the expression. Here are three such sentences, noted within a half-hour’s reading while having our correspondent’s question in mind: “The revived Olympic games is the subject of 81 two articles.” “A thousand shares of short interest is one result of the raid.” “A few doses is sufficient.” The late Prof. William Dwight Whitney, author of “Essentials of English Grammar,” decided, while editing the Century Dictionary, that “two and two is four” is better than “two and two are four,” because the full sense is “the sum of two and two,” or something similarly unifying the idea of “two and two.” The sentence above questioned would be better if written, “Failure is the current talk,” but “now days” instead of nowadays is much more criticisable than the verb.

Which of the following sentences are correct, and by what rule? “Please state whether one or six bottles is desired.” “Please state whether one or six bottles are desired.”

In this question as written there is an erroneous use of the plural that is not at all questionable. “Which ... is correct” should have been written. Only one is contemplated, as a choice, by “which,” therefore the verb should be singular. In the sentence inquired about are is the proper verb, because the plural subject immediately precedes it, and the singular verb agreeing with “one” is understood, not expressed. Logical fullness of expression would demand something like “whether one bottle is or six bottles are”; but that is plainly undesirable. The rule is that in such cases the verb should agree with its immediate subject. Objection to the plural verb in the other sentence does not conflict with this rule, because, the pronoun “which,” meaning “which sentence,” is the direct subject, notwithstanding the intervention of other words between it and the verb.

82 I inclose two clippings from papers, which I have numbered (1) and (2). Will you kindly inform me if these two sentences are grammatically correct as printed? If not, please explain why. (1) “He made many friends, but all were in moderate circumstances, and none wanted to know any other language than their own.” (2) “This thing is so simple and so clear in my own mind that I can not see how any one can think differently; but if anybody does, I would like to hear from them.”

The first sentence can not rightly be utterly condemned, although “none” is simply “no one,” and so is primarily singular. It is not uncommon to use the word with a plural pronoun or verb, as including more than one, and it is not wrong to do so. It would undoubtedly be right, however, to say “none wanted other than his own.” The second sentence is positively and unqualifiedly bad, notwithstanding the fact that the error is a very common one. “I would like to hear from him” would be right. In cases like both of these (supposing that one prefers the singular pronoun in the first) it is preferable to use the masculine singular, despite the inclusion of women among those meant by the other words, because it agrees in number, and while it means a man and not a woman, “man” is inclusive of women, though it is essentially a masculine word.

Will you kindly inform me whether the subjoined sentence is wrong? “The events in Field’s life—his birth at St. Louis in 1850; his education at Williams, Knox, Amherst, and Missouri State Universities; his connection with the St. Louis Journal , Kansas City Times , Denver Tribune , and Chicago News ; and his rise in journalism—were sufficiently commented upon at the time of his unfortunate death a little over a year ago to require special mention now.” It is claimed by a literary friend that the word “not” should be inserted after “ago,” making the phrase read “not to require special mention now.” 83 I maintain that the clause beginning with “to” is a clause of result. For substitute the word “enough” for “sufficiently”—which means the same—and see how it reads: “The events in F’s life ... were enough commented upon at the time of his ... death ... to require special mention now.”

The sentence is incomplete without “not” after “ago,” or a corresponding change, as “to require no special mention.” Its intention is that no mention is now required, and why not say so? Substitution of “enough” for “sufficiently” makes no difference, and I must confess that I do not know what “a clause of result” is, as I never heard of one before, at least with any meaning that is at all fitting for anything that can be intended here.

Many authors, especially those who dabble with statistics, use the words “native language.” On consulting the Century Dictionary, under the head “Native,” I find the following definitions: “3. Of or pertaining to one by birth, or the place or circumstance of one’s birth; as, native land, native language. 4. Of indigenous origin or growth; not exotic or of foreign origin or production.” Now, will you kindly explain the native language of a person born in Switzerland, where it is stated that in one canton the language used is Italian, in another German, and in still another French? Likewise of Alsace-Lorraine, which at one time is a part of France and at another time is an integral portion of Germany? Then, let us take Brazil. A person born in that country is called a Brazilian, yet speaks the Portuguese tongue. Colonization, also, leads to a strange condition of affairs. When this country was settled there were several languages, yet English became the predominant one. Still, if I am not mistaken, English is not of indigenous origin or growth here. While I am well aware that the words have been used by some of the best writers, I am still of the opinion that it is not strictly correct, and that some other expression might be used. As an example, I will state 84 that I saw recently a case where it was printed that a child was born in Canada of Italian parents and that he could read and write his native language. What is his native language?

One’s native language is that to which he is born—that is, it is the one he acquires most naturally, being, of course, his parents’ native speech, wherever he may be born. Dictionaries can not multiply definitions for every possible mutation of human affairs. The definitions quoted are absolutely right, even if various languages are spoken in one country. An Italian Swiss’s native language is Italian; in Alsace-Lorraine the native language of some of the people is German, and that of others is French; in Brazil the native language of natives is Portuguese. The second definition quoted has no connection with languages, except that of the kind shown in saying that “the native languages of America are the Indian languages”; it is not intended for the case in question. Our native language is English, not primarily through the place of our birth, but because of the circumstance that we are born to that language, born of parents who use it and from whom we instinctively acquire it. In the last case noted—the child born in Canada—the native language is Italian. No reasonable objection to the expression seems possible.

Would you say, “About one person in ten doesn’t know that their neighbors are saving money,” or do you think “his neighbors” better?

“His” is decidedly better. It is never right to use a singular noun and a plural pronoun, or any other disagreement in number. It seems advisable in a case like that of the question here to say “About one man in 85 ten,” etc., because it is a business matter, and presumably men are principally concerned. However, if generalizing by the noun “person” is preferred, that need not lead to the real grammatical error of using a plural pronoun. Of course a person may not be masculine, and that is why so many people make the error in number—to avoid supposed conflict in gender. But “man” is sufficiently generic to include all mankind, and the fact of its being masculine in gender, and demanding a masculine pronoun, need not be considered an insuperable objection to its use in the inclusive sense. All readers would know that the mere matter of general expression did not exclude women and children from business dealings. Changing “man” to “person,” though, still leaves the masculine pronoun good, for grammar demands agreement in number, and it has been custom from time immemorial to use in such cases the word that denotes the supposedly stronger sex. Thus we should say, “The animal draws his load better under certain conditions,” in a general sense by no means precluding the female animal from consideration; and why not “the person” also? We are the more willing to discuss this matter now because of a recent revival of the silliness that would have us use the ridiculous word “thon,” meaning “that one,” in such cases. Here is the latest outcropping of this nonsense: “We are prone to prefer the new words to the old, and many men and women find a pleasure in introducing a word not familiar to the average individual. Such a word is ‘thon,’ a contraction of ‘that one,’ proposed in 1858 by Charles Crozat Converse, of Erie, Pennsylvania, as a substitute for the clumsy combinations ‘he 86 or she,’ ‘him or her,’ etc., as in the sentence, ‘The child must be taught to study thon’s lesson.’ The word is so convenient that it is a wonder that it remains new to most people. The want of it caused the United States Supreme Court once upon a time to render a decision that ‘his’ in a law should be construed ‘his or her,’ so that women might be as amenable to the law as the male lawmakers themselves. This ruling allows writers of laws to avoid the use of ‘his or her,’ etc., every time a personal pronoun has to be used. But in every-day use the ruling of the courts does not count, and we need to use ‘thon’ every day of our lives.” It was not the want of any such abominable formation as “thon” that led to the court decision, but that decision merely fixed in law what had always been a real principle in language. With correct understanding of language facts, no one ever need say “his or her,” for “his” alone is really sufficient. The abomination “thon” remains new to most people because there is absolutely no need of it.

FORM OF WORDS.

Is it possible to construct the following sentence so as to give three distinct and separate meanings without changing the wording? The sentence is, “Twenty two dollar bills weigh as much as a silver dollar.”

Yes. Twenty-two dollar bills, twenty two-dollar bills, and twenty-two-dollar bills (though there is no bill issued for $22).

Please explain the correct manner of compounding the following adjectives: “Life-insurance company,” “fire insurance company,” “tornado insurance company.” I am under the 87 impression that they should be used as written above, for this simple reason, namely: In the first instance it is possible to place an insurance upon your life, and therefore the two adjectives adhere and become compound. In the latter two cases it is different—you do not place insurance upon fire or tornado, but you insure against them, and you do not insure against life; therefore, in the last two instances, the two adjectives do not adhere directly and should not be used as compound adjectives. I would also like to inquire further, if either of the above is incorporated in the full name of an organization, should they in any such case be compounded?

If compounding occurs in any of the terms, it should in all, as they are exactly alike grammatically. Difference of meaning in the understood prepositions should not affect the forms. No compounding is really necessary, although the terms are compounds etymologically. If we tried to compound every term that could be reasonably joined in form no dividing line would ever be reached. Usage, especially in the names of corporations, is against compounding in these cases.

A large book is now in press (about 150 pages having been electrotyped). Throughout these pages the apostrophe and additional s were used in names ending with s , viz., Lewis’s, Parsons’s, Adams’s, etc. Proofs are now returned with final s deled, which fact leads the Autocrat of the Composing-room (the Chairman) to arise and assert that “while the practice may be correct, it is behind the times,” “all good enough fifty years ago,” “won’t go in good offices nowadays,” “never used in first-class work,” closing with the remark that he doesn’t see why it is not used in griffins’ [griffins’s] heads (!), Orphans’ [Orphans’s] Home (!), calmly ignoring the fact that in the first instance a common noun, plural, is used, and in the latter a proper noun, same number. The reader contends that the apostrophe and additional s as marked are correct, and refers to the Harper publications, Scribner’s , the Century , and 88 the work of any good printing house. Who is right, or which is right (all questions of “style” aside)?

That Chairman evidently does not know the difference between singular and plural, or at least does not know the grammatical distinction of the forms, that has been just what it now is for more than fifty years. “Adams’s,” etc., are the right forms, beyond any possible reasonable objection; the only difficulty is that some people will not use the right forms, and have been so thoroughly drilled in the use of wrong forms that they insist that the wrong ones are right.

Please tell me what kind of mark (if any) should be placed after 4th, 21st, and like words used in a sentence where if the word were spelled out there would be no mark; as, “On the 21st of September.” My opinion is that the form is not an abbreviation. It certainly is a contraction, but nothing seems left out.

No mark should be used. The opinion that the form is not an abbreviation is a good opinion, because there is no abbreviating. Abbreviating is done by leaving off a part of the word, and it is commonly shown by using a period at the end of the short form; but some short forms, while they really are abbreviations, are not technically known as such, because they are quite properly included in another category, that of nicknames or merely short names. In this latter class are “Ed,” “Fred,” “Will,” etc. In the ordinal words of our question there is no cutting off from the end, but only substitution of a figure for the numeral part of the word, with the same ordinal termination that is used in the word when spelled out. How can anything “certainly” be a contraction when nothing seems left out? A contraction 89 is a form made by leaving out a part from between the ends and drawing the ends together, commonly with an apostrophe in place of the omitted part, as in “dep’t” for “department”; but some real contractions are known as abbreviations by printers, because they are printed in the form of abbreviations, as “dept.,” which is often used instead of the other form. The dates with figures certainly are not contractions, as there is no omission, but mere substitution of a figure for the corresponding letters. Possibly the doubt arose from the fact that the Germans do make abbreviations of ordinal words by using a figure and a period, omitting the termination, as “21. September,” which shows plainly why the point is used.

In reading the proofs of a bicycle catalogue recently the writer compounded the words handle-bar, tool-bag, seat-post, etc., on the ground that they were all technical terms in this connection and were therefore properly compounded. For this action he was criticised, his critic claiming that handle-bar is the only proper compound of the three words mentioned, inasmuch as neither the bar nor the handle is complete alone, while in the other cases named the parts are complete by themselves. Will you kindly give your opinion on this matter?

The words mentioned are compounds, though they are more frequently printed in the wrongly separated form than in their proper form. Mere technicality, however, is not a good reason for compounding any words. It is the fact that “handle” and “bar” are two nouns joined to make a new noun that makes them become one word instead of two. “Handle-bar” is no more technical than “spinal column,” for instance, is anatomical (another kind of technicality), yet the first term 90 is one word and the other is two. In the latter term the first word is an adjective, fulfilling the regular adjective office of qualifying. The other name has no qualifying element, being a mere name, representing the phrase “bar used as a handle.” How any one can imagine such a difference as that neither the bar nor the handle is complete alone, while in the other cases named the parts are complete by themselves, passes understanding. The circumstances are identical—two nouns in each case joined to make a new noun representing such phrases as “bag used to hold tools,” “post to support a seat,” etc. Even the accent as heard in the first part of each name truly indicates compounding. The principle is exactly the same as that which made the Greeks and Latins join two nouns in one, through which we have “geography,” which is no more truly one word than is its literal English translation, “earth-writing.”

One of our printers, in setting up a job, came across the words “large tobacco firm.” He felt sure a hyphen should be used after the word “tobacco,” so it would not be understood as a large-tobacco firm. To please him, I told him to put it in, but told him its absence showed that the tobacco firm was large, and not the tobacco. What do you do with such words as “honey crop”? I compound it when it means the first stomach of the bee, but not when the word “crop” means harvest.

Certainly, if any hyphening is done in the first words instanced, it must be that which is mentioned; but none is necessary, and probably few persons would ever think of it. Our correspondent seems to have given a hasty answer to the question, as in fact it is not strictly true 91 that the separated words show that the firm is large, and not the tobacco. It would seem more accurate to say that no one (speaking generally) would misunderstand the separated words, because the natural conclusion is that the firm does a large business. On the contrary, if the actual intention should be that the firm dealt in large tobacco, that fact would be fixed beyond question by making a compound adjective “large-tobacco.” The distinction between “honey crop” and “honey-crop” is excellent. A principle is illustrated by it that would be worth a great deal to everybody, if only it could be established and widely understood and applied. It is difficult to state it clearly, although the two kinds of meaning seem to show a very plain difference, that might easily be less apparent in a sentence containing only one of them. We can not say that “honey” is a true adjective in the separate use, but it comes much nearer to the true adjective force in one use than it does in the other. “Honey-crop” for the stomach, as “the crop (stomach) in which honey is stored,” is simply one noun made by joining two nouns. “Honey-bag” is the word given in dictionaries for this. All the grammarians who ever wrote about this subject say that in our language two nouns so used together simply to name one thing become one word (meaning merely that they cease to be two words in such use). Of course there is much disagreement, and it does not seem probable that everybody will ever write all such terms alike; but it is absolutely certain that some compound words of such make are as fully established as if their elements were not usable separately, and it seems impossible to distinguish in any reasonable way between one such name and 92 any other. In other words, if “honey-bag” is a compound—and it is, no matter how many or what persons write it as two words—“mail-bag,” “meal-bag,” and every similar name of a bag is a compound; and if names of bags, then likewise every similar name of anything else is a compound.

The appended clipping is from a proof of a college publication, and is part of a class history. It appears as it came from the compositor’s hands. The editor of the annual in which it will appear submitted the first of my questions (indicated below) to the president of his college, and though the latter enjoys considerable local prominence as an educator and a Greek scholar, yet was he unable to enlighten us upon this point. “In oratory we have shown our powers, and look forward to the time when the Demosthenes of ’Ninety-eight will sway senates and our Ciceros the political world.” What is the plural form of “Demosthenes”? The plural is clearly the form the author had in mind while writing it, but I am ignorant of either rule or authority governing such cases. Would you prefer reconstructing the sentence? To cover our ignorance somewhat, I suggested the following: “In oratory we have shown our powers, and now look forward to the time when ’Ninety-eight’s disciples of Demosthenes will sway senates, and its Ciceros the political world.” In the word “Reinoehl” (a proper noun), should the diphthong be used? I stated that it should not be used, and was contradicted by the editor of this same publication, who said that the president of the college maintained that the diphthong was correct. Though I could quote no authority, yet I believe I am right. The word is a German one, as you will have noticed. The words Schaeffer, Saeger, and Steinhaeuser appear without the diphthong on the same page with the word Reinoehl, yet they passed unchallenged by the editor. Would they not come under the same head as the one mentioned first?

The quotation does not seem to show positively that 93 a plural was intended. As there was only one Demosthenes sufficiently famous for the comparison, so the writer might mean only the one best oratorical student. It is not an unnatural inference, though, that the plural was intended. The plural form of “Demosthenes” is “Demostheneses.” Why hesitate over that any more than over “Ciceros”? A regular English plural is as good for one as for the other. Greek common nouns with the termination es form the plural by substituting æ for that ending, as “hoplites, hoplitæ; hermes, hermæ.” Our second example is originally a proper name, but was and is used as a common noun, meaning a bust that may or may not represent the god Hermes; but this is not a good argument in favor of a Greek plural of “Demosthenes.” The change suggested is not good, because “disciples” is not meant, the intention being merely to note a similarity, and not a studied imitation: In the German name separate letters should be used, as they represent umlaut interchangeably with a double-dotted vowel without the e ; thus, either “Reinoehl” or “Reinöhl” is right, but “Reinœhl” is wrong. The college president must have had the umlaut character (ö) in mind, not the ligature (æ), in answering the question. All the names mentioned are amenable to the same decision; what is right in one is right in all.

An advertisement writer brought to the office, a few days since, copy for an advertisement for a certain complexion soap in which the word which is underlined occurred: “Combined with the emollience of cucumber juice.” The proof-reader queried the word to the author, informing him that it could not be found in the dictionary (International, 1891); his response 94 was that the word expressed the idea intended to be conveyed better than any other that he knew of, and therefore he should use it, regardless of the dictionary. I have since examined the Century Dictionary and fail to find the word. The question arising in my mind is, Should the proof-reader endeavor, when the author is present, as he was in this case, to induce him to use a word for which authority can be produced, or should the author be allowed, without a word of protest, to coin words at his own sweet will? It seems to me that the proof-reader should not be required to blindly follow an author in a case of this kind after he has satisfied himself that there is no warrant, except the whim of the author, for the use of such words.

Not long since, in reading a catalogue of road machinery I noticed “barrow-pit.” Being somewhat in doubt whether it should be compounded, as already written, or two words, I consulted the International, and also the Century Dictionary, but failed to find the word in either, finally concluding to use the hyphen. Which is correct—barrow-pit, or barrow pit, or barrowpit? My preference is for the use of the hyphen.

The writer was perfectly justifiable. If no word not in a dictionary could be used, the language could not grow, and there would be many ideas left inexpressible, for want of words. Johnson’s dictionary contained many more words than any preceding work, and each new dictionary since issued has increased the record. This could not have been done if people had not used new words. Although “emollience” is not in any dictionary, there is sufficient authorization in the fact that -ence is used in forming nouns from adjectives in -ent, something that any one may do at any time, just as one may add -less to any noun, as “cigarless,” having no cigar. Emollience is the only possible single word for “character of being emollient (softening).” This is not properly a case of “whim.” The only proper restriction against such neologism is that it should not be indulged 95 unnecessarily, as when there is already existent a good word for the sense to be expressed.

“Barrow-pit” is the only form that principle and commonest usage will justify for this word—but the same principle gives also “advertisement-writer,” “complexion-soap,” “cucumber-juice,” and “road-machinery,” each of which you write as two words. Your decision to use the hyphen in “barrow-pit” is in accordance with all text-book teaching on the subject, and unless such teaching is applicable in all strictly similar cases it is all bad . It can hardly be necessary to reach any such pessimistic conclusion as that expressed in a letter from a country superintendent of schools—“I do not know anything about it, and I do not believe any one else does.” Our grammarians are not all idiots. What possible principle could justify such a difference as “advertisement writer” and “proof-reader” (for “one who writes advertisements” and “one who reads proof”)? If one of them is one word, the other also is one, the only difference being that some such familiar short words are written without a hyphen.

You in a recent edition, speaking of Roman type, used lower-case r . We write to ask what, if any, warrant you have among grammarians or lexicographers for the lower-case initial letter in an adjective of this class. Would it by the same authority be proper to use a lower-case in the word “Parisian,” “Chicago” used as an adjective, etc.?

No rule as to capitalizing has wider acceptance or better basis in principle than that an adjective derived from a proper noun should be capitalized, and “Roman” is such an adjective. However, in the connection this 96 word has in the matter with which we are dealing, the lower-case letter is not wrong, though “parisian,” “chicago” in any use, or any other such use of a lower-case initial letter would be wrong. Reasons will be given after some authorities are cited. The “Century Dictionary” says: “Roman, a. ... [ l. c. or cap. ] Noting a form of letter or type of which the text of this book is an example”; also, “Roman, n. ... [ l. c. ] A roman letter or type, in distinction from an italic .” The “Standard,” under the noun, “[R- or r-] A style of ceriphed type. ... also, a black gothic letter, etc.” The “Imperial,” the standard Scotch dictionary, says of the adjective, “applied to the common, upright letter in printing, as distinguished from italic ,” and of the noun, “A roman letter or type.” Benjamin Drew, in “Pens and Types,” page 199, in speaking of specimens of old-style type given in his book, says: “The next is a Fac-simile of four roman and three italic Lines.” He says on page 57, in introducing two lists of foreign words: “The roman list is destined to be continually lengthening, while the italic, save as it receives new accretions from foreign sources, must be correspondingly diminishing.” Webster and Worcester missed the point of distinction in usage that was discerned by the other lexicographers, and they capitalize “Roman” and “Italic.” The questioner does not say anything about “italics,” used in the same paragraph with “roman,” yet evidently the two words should be treated alike. In fact, neither word in this use has its literal sense, nor conveys a thought of Italy or Rome. When this literal sense is expressed the words should be capitalized, just as “Parisian” and 97 “Chicago” should be. Webster actually says that “Roman” means “upright, erect,” which is plainly not a meaning showing connection with a proper noun, and, in fact, is not a true definition for the word with which it is given. The word has no real sense other than its literal one, but the literal allusion is so far removed from conscious apprehension in the printing use that it is proper and prevalent usage to write it as a common noun or adjective, just as such form has become prevalent in many other cases, as—

Have our correspondents ever noticed these words in books? The writer of this answer has no hesitation in asserting that “italics” and “italicize,” which have far more literary use than “roman,” will be found with a lower-case initial much more frequently than otherwise; and the same is true of “roman” in printers’ use, which must be looked for mainly in printers’ books. What is here said, however, should not be applied too strictly; the word in question should be capitalized in special work such as that of our correspondents, where probably all similar words have capitals, as Gothic, Doric, Ionic, etc.

SPELLING AND DICTIONARIES.

Kindly permit me to make a few comments. As to “honour, fervour, ardour,” etc., you say that “undoubtedly the American way ( i. e. , honor, etc.) is better than the other, historically as well as economically.” I suppose that “economically” means the saving of one letter; that I do not consider as worthy of note at all. As to the historical point, the words in 98 Latin are all “honor, ardor, fervor, labor, color,” etc.; but then in French, through which they came into English, they are “honneur, couleur,” etc., so that it seems to me that the u is historically defensible.

“Sceptical” or “skeptical”—a matter of indifference; the hard c represents the Greek kappa in any case. I suppose you spell “speculator,” yet the Greek is σπεκουλάτωρ; so “sceptre” is the Greek σκήπτρον. So we might write “spektakle” if we cared to do so; indeed, many Greek scholars do use k where ordinary people would use c , as “Asklepiad, Korkyra,” etc.

“Ascendant, ascendancy”—the usual plan is to take the letter found in the supine of the Latin verb; thus, “dependent,” from Latin “dependens,” “intermittent,” from Latin “intermittens,” “dominant,” from Latin “dominans,” and so on. On this plan “ascendent” and “ascendency” would be right, as “scando” and “ascendo” make “scandens” and “ascendens.”

You say, “Each of the large dictionaries is worthy of acceptance as final authority in every instance.” Not by everybody, by any manner of means. There are many better scholars than the dictionary-makers. Would you expect Mr. Gladstone, John Ruskin, Andrew Lang, Archbishop Temple, Bishops Lightfoot and Westcott, Dean Farrar, and many others to accept the dictum of a dictionary man in every instance? Why, I do not do it myself. Indeed, though I possess Greek, Latin, and French dictionaries, I have never possessed an English one, and do not much regard them or the people who think them infallible. Educated people in England have no such opinion about dictionaries; in fact, they consider themselves the source of authority in matters of usage and pronunciation. Oxford and Cambridge men and members of the educated classes in England are the sole arbiters in such matters; there is no appeal against them. Richard Grant White thoroughly grasped this and expressed it very well. Just as all classical scholars try to write Attic Greek, i. e. , the Greek of the inhabitants of one Greek city, and entirely disregard the millions of other Greeks (even though so eminent as Homer and Herodotus), so all English-speaking people should model their language on that of the educated classes of Great Britain.

99 “Economically,” as used in the article criticised above, meant the saving of one letter, and as many scholars, both English and American, are noting such economy, and making it very important, it may be concluded that it is worthy of note. Certainly the spellings “honour,” etc., are defensible historically—but no assertion has been made that they were not; the saying was merely that the other way is better historically. The words came into English through French, but the Latin spelling is preferable for more reasons than one. If we are to preserve the u because it is in the French words, is not the reasoning equally applicable to the whole syllable in which the letter is used? Would it not be equally reasonable to preserve the other u in the first syllable of “couleur”? The French themselves once spelled these words—or most of them— or . They changed them probably to represent better the natural French sound of such syllables. Because Englishmen first learned such words from Frenchmen does not seem a valid reason why the former may not revert to the historical original, which is more in keeping with English analogy, and better represents the English sound.

As to “sceptical” and “skeptical,” one who knows the need of a vast majority of English-speaking people of an authoritative choice between the two forms can never admit that the spelling is “a matter of indifference,” even if it could be reasonably admitted on any ground. Our correspondent is unfortunate in his selection of an example here, for σπεκουλάτωρ seems to be not a true Greek word, but only a transliteration of Latin “speculator,” the true etymon of the English word, 100 which does not come from Greek. We might have written “spektakle” if we had cared to do so, as it is spelled with k s in some Teutonic languages; but in the close connection here there is a strong suggestion that this word might also be Greek, which it is not. The reason for preferring “skeptical” is that there is not another English word in which c in the combination sce is hard, and so “sceptical” is a very bad spelling, even if it is prevalent in Great Britain.

On the plan mentioned in the letter “ascendent” and “ascendency” are right; but the other spellings are copied from the French, so potent with our correspondent in the other case, and are prevalent in present usage. “Ascendant” and “ascendancy” are preferable for this reason, and because the use of these spellings removes one of the puzzling differences which most people can not understand or explain. The plan mentioned would also give “descendent,” which has no currency as a noun, though it has been used as an adjective, and “descendant” and “ascendant” are so much alike in their nature that it is better not to make them different in form.

“Each of the large dictionaries is worthy of acceptance as final authority in every instance” was intended only as an assurance that those who desired such an authority—and there are many such—might reasonably accept the one chosen, without trying to make exceptions. There could be no intention of dictating that scholars should “accept the dictum of a dictionary man” in every instance, for that would be “putting the cart before the horse” with a vengeance. One need feel no hesitation in saying, however, that the English-speaking 101 educated man does not live, and never will live, who can afford to ignore utterly dictionaries of English. No dictionary is made as our correspondent seems to assume that all are made, though probably every one of them has provided employment for some men not so thoroughly educated as men can be. Educated people, in America as well as in England, make the scholarly part of the language, though it contains much that is made by the common people and that finds just as thorough establishment as that made by the scholars. Dictionary-makers never pose as language-makers. They are recorders of what is already made, which is so great in quantity that no scholar can hope to master the fiftieth part of it so thoroughly as to need no record of it. Even supposing that Oxford and Cambridge men and members of the educated classes in England are the sole arbiters in such matters—it is not supposable, though—how is the rest of the world to know their decisions if they are not recorded? Any record of them will constitute a dictionary, for that is exactly what a dictionary is—namely, a record of the accepted details of diction. As a matter of fact, also, our actual dictionary-makers, those who are vested with authoritative decision, are selected from among the very men for whom independence of dictionary men’s dicta is claimed. Noah Webster, Dr. Worcester, Professor Goodrich, Professor Whitney, Dr. March, President Porter, Dr. C. P. G. Scott, and Dr. J. A. H. Murray—not to mention the many other English scholars who have been dictionary-makers—rank with the men named in the letter, if some of these do not outrank some of those in scholarship, and they are the ones who 102 choose where there is a choice in making the record. Dictionaries contain errors, and scholars are independently above acceptance of the errors; but we may repeat the saying that when once a large dictionary is chosen as authority it is better, as to matters of spelling, to accept it in full.