PORTER GARNETT The Ideal Book

Copyright 1931 by The Limited Editions Club. Reprinted by permission of the publisher.

In adopting a prescribed title for this paper, I must begin by registering my dissent to its validity. There is no such thing nor can there be such a thing as "the ideal book." No single book, no particular style of book can be said to represent in itself an ideal below which all other books and other styles which differ from it fall. A certain book may be ideal for its purpose, but books can no more conform to a fixed ideal than can churches, cocktail-shakers, or hats. The best that one can do is to attempt to enumerate and codify those elements of good book-making that enter into what may be called the "fine" book.

It is difficult to declare oneself an advocate or exponent of fine printing or fine book-design without being misunderstood. Such a declaration, however, is not to arrogate superiority. It merely means that one believes in certain principles of craftsmanship and in upholding certain standards based upon a scrupulous and uncompromising observance of refinements and minutiæ. It is a mistake to assume that the word "fine," as applied to printing and to books, is a comparative term meaning a grade or measure of merit. Consider for a moment its true meaning: delicate, studied, subtly calculated. It represents not a grade of excellence, but a quality, a quality distinguishing those books and pieces of printing which the term properly describes from other books or pieces of printing. It may be allowed, however, that fineness is itself a comparable term; that there are, in other words, degrees of fineness. Thus a book may be fine without being of the first order of fineness. But if we are to seek for a standard of excellence equivalent to what is implied by the word "ideal," it should be obvious that only fineness of the first order can be considered. A fine book of the first order is the end-result of a sedulous effort on the part of designer, printer, and binder to bring to their artifact every care for physical and technical details, every revision in the interest of betterment, of which they are capable, to the end that the finished product shall represent the capacity of each for the fulfilment of his artistic wish, his desire for perfection. To slacken this effort, to compromise wittingly (or wilfully), to surrender to expediency, is to repudiate fineness of the first order.

It is this concern for perfection that Mr. Stanley Morison means when he says "The fine printer begins where the careful printer has left off." It is this concern with perfection that Conrad celebrated when he wrote:

"Now the moral side of an industry, productive or unproductive, the redeeming and ideal aspect of this bread-winning, is the attainment of the highest possible skill on the part of the craftsman. Such skill, the skill of technique, is more than honesty; it is something wider, embracing honesty and grace and rule in an elevated and clear sentiment, not altogether utilitarian, which may be called the honor of labor. It is made up of accumulated tradition, kept alive by individual pride, rendered exact by professional opinion, and, like the higher arts, it is spurred on and sustained by discriminating praise. This is why the attainment of proficiency, the pushing of your skill with attention to the most delicate shades of excellence, is a matter of vital concern. Efficiency of a practically flawless kind may be reached naturally in the struggle for bread. But there is something beyond—a higher point, a subtle and unmistakable touch of love and pride beyond mere skill; almost an inspiration which gives to all work that finish which is almost art, which is art."

In dealing with the constituents of the fine book I intend no disparagement of seemly, modest, and honestly-made books to which the term "fine" is not strictly applicable. Even the humblest volume, ad pauperum commoditatum, may be, by virtue of its suitability to purpose and its seemliness, wholly admirable. As for the better class of trade books, the productions of university and great commercial presses, they often display qualities of design and workmanship of a high order. Though not of the first order of fineness, they represent, with gratifying frequency, what Conrad called "efficiency of a practically flawless kind." That the best of them belong, however, to a lower stratum than the truly fine book may be, I think, quite easily demonstrated. One does not have to consider the work of the Doves Press or the Bremer Presse or the finer examples of French printing of the sixteenth or eighteenth century the ne plus ultra of book-making in order to recognize in them a quality (mark the word) which the trade edition, however charming, never does and never can attain. By reason of this quality—the quality of fineness—they are different from trade books, whether or not they are superior must remain for each of us a question of personal values. Since that is true, let us now—having cleared the ground and removed perhaps the possibility of a misapprehension with regard to the title of this paper—consider the values and the physical constituents of the fine book.

These constituents fall into three divisions: first, Dimensional (size and proportions); second, Tectonic (plan and construction); and third, Visual (appearance).

It would be absurd to contend that, ideally, a book should be of a certain size. Very large books are, of course, awkward to handle and are unsuitable, let us say, for reading in bed or in a railway train. But it does not follow that, because our habits of life differ so radically from those of the more leisurely and contemplative past, the tall volume is no longer justified. The large book is not an impediment to meditative reading and, although the "handy volume" will, in most circumstances, serve every purpose, there are those who, undeluded by pragmatism and undebased by false ideas of efficiency, may still, in the seclusion of study or library, find pleasure in the leisurely perusal of, let us say, The Golden Legend, in folio, nobly enthroned upon its lectern. Again, there is nothing incongruous or unpractical about the scholar (perhaps I should say, "research-worker") making use of a huge volume, spread before him on a library table. Large volumes are, moreover, frequently justified by the fact that illustrative plates of a large size are often desirable or essential. Who will deny that reproductions of Egyptian papyri, of eighteenth century engraved portraits, of Oriental carpets, in fact, of almost all works of art other than such small objects as miniatures or jewelry, would be better in folio than in octavo or duodecimo? It can be said, I think, that the very large book should be unconditionally condemned only when its size defeats the purpose to which, by reason of its content, it would normally be put. Stateliness of form imparts dignity. It may be argued, therefore, that a great work on engraved gems, imposing in size, with plates, each showing many specimens, comports better with the character of its subject matter quite aside from any advantage it offers for comparative study, than would the same work printed as a book one might slip in one's pocket. Stateliness of form implies stateliness of content, and vice versa. Let a book be, for a generation, of such good report that it may be said to have become a classic, and a large-paper edition is justified. Let those who must cavil do so. If they cannot rise above the utilitarian ideal, they can easily obtain the work in a small format and be happy.

It may not be out of order to say at this point that, while a considerable range in the size of books is not only permissible but desirable, there are limits at both ends of the scale where practicability ceases to exist and we pass into the realm of curiosities and tours de force. Thus the miniature book, for all its charm, lies outside the confines of normal book-design. As to the maximum size that may be legitimately allowed for a book, it should never, I think, exceed the normal folio height (defined approximately by the larger moulds employed for manufacturing hand-made paper) while its bulk and weight should not preclude the possibility of holding it by the spine with one hand while turning the leaves with the other, when such a method of referring to its contents may be necessary. And now a final word as to dimensions. Large or small, the most perfect book will always be one of which the thickness bears a just and agreeable relation to its height and width. Small and slender books are delightful objects which no one could wish to abolish (one cannot say as much for the lamelliform folio, a veritable atrocity), but their inferiority to books of a meet thickness becomes apparent when, with (or, worse yet, without) their vertical, neck-twisting titles they are placed on a shelf.

We must next turn our attention to those aspects of a book which have to do with its plan and construction and which we have called tectonic.

In its physical character a book addresses itself to two of our senses, the sense of sight and the sense of touch. Because the tactile qualities of a book are relatively of less importance than its visual aspects, let us first deal with those elements which are, in part at least, evaluated through the sense of touch.

Our first impression of a book is received from its exterior, its binding. Now the qualities to be looked for in the binding of a book are: (1) the character and quality of the material, (2) suitability, (3) soundness and charm of design, (4) agreeable color (a relative term), (5) workmanship, (6) pleasantness to the touch. Granting adequacy in all of these (and no book can pretend to fineness without such adequacy), there is still another desideratum less easy to specify. It might be called (7) "the evidence of durability." A book when taken in the hand should have a feeling of compactness, almost of solidity. I do not mean by this that it should feel like a block of wood, but it should, when picked up, when opened, or when its hinges are tested, give the impression that leaves and cover are so firmly (and honestly) knit together that they constitute a unit, having in its "feel" the evidence (or the assurance) of durability.

The next characteristic of a book to be noted through the sense of touch is the texture of the paper. By "texture" several things are meant: a surface agreeable to the hand, the degree of crispness, an impression of toughness (again the evidence of durability), and the degree of flexibility. Ideally, the paper in a book should satisfy all these requirements and should possess as well certain qualities of character, style, and color, pleasing to the informed eye. These will be dealt with in their proper place. The paper should be flexible, without the flimsiness characteristic of papers weak in substance. It should bend readily when the leaves are turned and should flow smoothly through the hand when all the leaves are bent at once. Stiffness in the leaves of a book (an all too common defect) is not, it should be observed, always the fault of the paper. It is often due to the choice of a paper too heavy for the size of the leaf. The same paper in a larger leaf might have the desired flexibility.

The final tactile test of a fine book (applicable, alas, to very few books indeed) resides in the character of the impression of the type on the paper. In the best printing, the surface of the page, if rubbed with the palm of the hand, shows a slight and pleasant roughness due to the sinking of the type into the paper. Such printing is rare in modern books because it is difficult of attainment with machines designed for quantity production. To attain the effect described the paper should be dampened before printing, and an ink employed that is adaptable only to the hand press. Dry paper, particularly when heavily sized, resists a deep impression. It can be heavily impressed, but there is not the same difference between the impressed and unimpressed portions, due to the impaction of the substance caused by the pressure of the type, which results when dampened paper is used. In the latter instance, the depth of impression is within the sheet, not an embossment on the reverse side. This incisiveness, without a corresponding relief on the back of the sheet, is shown when an impression without ink is made on a hand press with dampened paper and a hard packing.

In printing on dry paper it is necessary, if adequate color is to be obtained, to use such a quantity of ink, of a consistency suitable to machine-press printing, that a really deep (not merely heavy) impression cannot be imparted to paper without "spreading," which slightly modifies the sharpness of the type. The machine printer must choose therefore between a surfacy quality with sharpness and a heavy (not necessarily deep) impression with a loss of sharpness, neither of which is ideal. There are some that will question the truth of this statement, calling attention to specimens of machine printing on dry paper in which the ink has been driven into the sheet and perfect sharpness maintained. It may be said, however, in support of our contention, that, under the test of hand and eye, this perfectly printed dry sheet will be found, in the last analysis, to lack, in comparison with a sheet perfectly printed by hand on dampened paper, a certain almost-indefinable something that can perhaps be best described as a living quality. This ultimate grace arises, I think, from the fact that in competent hand-press printing the third dimension is not merely suggested but actual; we have, in other words, not merely sharpness but crispness; the effect attained is sculptural. No printing that is lifeless, or to which such terms as "slick" and "dry" may be appropriately applied can be called fine printing.

Turning now from the tactile to the visual elements of the fine book, we shall consider, first of all, that fundamental factor of all books, the text-page, upon the form or "layout" of which all other typographic elements must, to a large extent, depend. The text-page is of primary importance because by its rightness or wrongness a book must stand or fall.

The elements of the text-page that call for consideration may be grouped under three heads: first, Form (the proportions—width to height—of the type-page and the balance of the rectangle of type with the rectangle of paper); second, Space (the ratio between the areas of the type-page and the paper-page); third, Tone (the tonal value of the type mass and the relation between its tone and the white area of the margins). In the perfect text-page all these elements may be observed in nice adjustment, severally and mutually.

There are those who contend that a proper relation of margins to type-page may be arrived at by employing ratios identical with those to be found in the well-proportioned pages of the early printers. Others declare that correct margins can be created by the application of an arithmetical or a geometrical formula. It can be admitted that such procedures are, at least, safe; that is to say, the danger of malproportioned margins will be avoided. But neither the method nor the result can be ideal for the simple reason that, while providing for the factors of form and space, they fail to provide for the factor of tone. It should be obvious that a rectangle of black type, with no space (leading) between lines, and a rectangle of the same shape and size printed from light-face type and generously leaded call for different margining.

All of this may seem to be supervacaneous, but the stubborn fact remains that no one of the factors set forth above can be ignored. It is perfectly true that the accomplished book designer will compass the desired end through a sagacious application of his knowledge and taste, but we are concerned here with presenting the elements of the ideal book and it is therefore essential that all the elements, no matter how much the initiated may take some of them for granted, should be, for the benefit of the layman, categorically enumerated.

It is necessary at this point to allude to the dictum—voiced in high places as well as in low—that a book is primarily something to be read; that every factor which does not contribute to that end is an impertinence. The worthy champions of this faith would be on firmer ground if they heaped their condemnation upon such adjuncts of a book as actually lessen its readability. M. Paul Valéry has disposed of the matter so effectively from the aesthetic point of view in his essay, Les deux vertus d'un livre, that nothing remains to be said on that side of the question. But there are other objections to be raised to this ipse dixit of the mechanists. Their contention that what we all grant is at once the basic and paramount function of a book, its readability, is its only function would, if carried to its logical conclusion, lead to a doctrine in book design equivalent to what is known in present-day architecture as "functionalism." Since functionalism or, as it is sometimes called, the "machine and function" principle, demands that the design of a building must grow out of and be restricted by its predetermined use or purpose, it should follow that, if the sole purpose of a book is that it be something to read, there is no reason, based upon utility, for not using the whole area of the paper-page, with margins of no more, let us say, than a quarter of an inch or so. The uncomely, marginless illustrations of certain recent books represent an application of this principle. If the protagonists of the utilitarian ideal admit that margins are other than a waste of usable space, they make a concession to the aesthetic conception of a book, for the determination of margins, the mise en page, is primarily an element of design. It will be argued no doubt, in contravention of this statement, that margins make for ease of reading (utility), but that this is an untenable defensive assumption should be proved by the perfect readability of newspaper columns separated only by a light rule, or by the two-column book or magazine page with only a pica of white between the columns. Since it cannot be denied that the margins of a book, if well proportioned, promote pleasure, an aesthetic function, the true functionalist should, to be consistent, insist upon doing away with them.

We may turn now from that major fundamental of the fine book—a text-page of perfect seemliness—to a consideration of other elements. But before doing so, it may be proper to explain what some of my readers may deem an omission. I have said nothing about the choice of type. It is axiomatic that good letter is a prime essential of the good book. There are only two kinds of type, good and bad. Good types, whether based upon classical models or the quasi-original forms of contemporary type-designers, are sufficiently numerous to make a suitable selection, provided the printer knows anything whatever about the subject, quite simple. It should be pointed out, however, that, other things being equal, type cast from matrices struck from hand-cut punches is superior to machine-cut type. This superiority is a matter of real importance, chiefly in printing by means of the hand press. Only in such printing is the difference between the hand-cut and machine-cut letter fully apparent.

No type is good if some of the characters are marked by eccentricity. Unorthodox peculiarities in the forms of certain letters sometimes lend charm to a type-face, but there is a difference between a peculiarity of shape thoughtfully and discreetly arrived at and freakish variations which do not justify themselves and bespeak only a stupid desire for novelty at any cost.

Given the seemly text-page as the prime requisite of the fine book, our next consideration should be what may be called integration of the parts. Here we must again think in terms of architecture, with which art book-design has so much in common. Let the parts of a book be few or many, simple or complex, it is of the first importance that they be, one to the other and each to all, harmoniously correlated.

In the simple undecorated and unillustrated book it is not only desirable that sunken pages, if any (the first pages of chapters or sections, for example), should show an equal sinkage. But all isolated typographic elements, such as half-titles, elements of the title page, copyright notice, dedication, headings of preliminary and supplementary matter, etc., should fall on levels which, though not necessarily identical, bear a mensural, not an arbitrary, relation one to the other and to the structure of the book as a whole. This requires, perhaps, some elucidation. Suppose we give the first pages of our chapters a uniform sinkage. These pages, let us say, establish three levels for us—(1) the chapter heading, (2) chapter title, and (3) the first line of text. If we adopt the same sinkage for Contents, Illustrations, Appendices, and Index, putting the headings of this group on the same level as the chapter headings (LEVEL 1), the first text lines of this group should be on the same level as the chapter titles (LEVEL 2), or on the level of the first text lines of the chapters (LEVEL 3). If, on the other hand, we adopt a different sinkage (smaller) for the second group, we may still relate it mensurally to the first group by placing the first text lines of Group 2 on the same level as the chapter headings of Group 1. Suppose further that we place our half-titles on one of the three or four levels that have been established. We still have to deal with a copyright notice, perhaps a limit notice, a bibliographical note, and a dedication. It is not essential that all of these should fall on the same level, but it is essential that each be related to some one of the established levels. Finally, it is desirable that such major elements of the title page as a subtitle or the author's name should be placed on one of the established levels.

An observance of this principle makes for homogeneity of design. In reading a book so put together we are spared (without knowing that we are spared) the disturbance of our sense of balance which results, almost without our knowing it, when our eyes fall upon a page some part of which is not "tied in" architecturally with the rest of the book. The effect is similar to that of a many-paneled room with an impost cornice at a certain height in all the panels except two or three where it is either higher or lower. We have, in one case, faulty architecture; in the other, faulty book-making. In judging a book or a building, it should be borne in mind that, however charming its parts, it must be regarded as a whole. If our contemplation of it is to be attended with pleasure and comfort, its parts must be so disposed, so correlated, that they will not produce a "jumpy" effect.

It is not contended that every book in which this refinement of perfectly integrated parts has been ignored should be considered a failure because it is less than perfect. If that were true, few books would pass the test. It would indeed be hypercritical to insist that a failure to observe this principle actually spoils an otherwise well-made book. It is desirable, however, that the principle should be observed as far as the material will permit. It must be recognized, also, that sometimes the elements are so diverse—chapter headings or the internal titles of essays, short stories or poems—that a strict adherence to the principle becomes impossible. In such instances, the designer's task is still to strive for order and integration. Perfect order, symmetry, and balance may be unattainable, but this does not justify him in being haphazard. When order is observed we may not be conscious of it; when it is not observed we are aware of its absence. Movement is of the highest importance as a factor of design—such movement, for example, as is imparted to a book by this very diversity of its elements—but good design demands movement that is ordered, not arbitrary. If liberties are taken (and it is desirable, in the interest of vitality and charm, that they should be taken) they must justify themselves aesthetically; they should not only please us in themselves but as evidence of the designer's intelligence, his insight, subtlety, sensitiveness, discrimination, and tact. However diverse the elements or parts of a book may be, however they may, by reason of such diversity, render perfect order and balance impossible, their arrangement should at least possess a rationale.

This need of a fundamental balance has its basis in its pleasure-giving value. I have adverted to the analogy between book-design and architecture, let me point now to an equally pertinent analogy with the structure of poetry. "Verse," says Poe, "originates in the human enjoyment of equality, fitness. To this enjoyment, also, all the moods of verse—rhythm, meter, stanza, rhyme, alliteration, the refrain, and other analogous effects—are to be referred." Specifically, the parts of a book—half-titles, headings, etc., etc.—should be, severally and collectively, related as are such structural elements in verse as recurrent rhythms, rhymes, and refrains.

We must now consider the decorated or illustrated book. In the first place, it should be understood that no form of decoration or illustration is legitimate in the strictly fine book except such as are printed from wood or metal engraved by hand, preferably in relief which comports with type both in physical character and in the means by which the image is imparted to the paper. Process-engravings are disqualified not only because of the preponderant mechanical factor, but because mechanical engraving in relief cannot produce a line of the delicacy and purity obtained with the engraving tool. As a corollary to the principle of integration set forth above let us take first the type of decorated book to which most obviously it applies. A book carrying on various pages head-bands of varying depth and varying tone—deep, shallow, black and heavy, light and delicate—will produce a disturbing effect. Less obvious but hardly less disturbing is a succession of initial letters differing in size, in tone, or in position on the page. The book with initial letters strewn through the text, sometimes several on a single page, is a challenge to the designer. When thus arbitrarily employed it is important, in order that the initials shall not be obtrusive, that they be so selected as to size (in proportion to the page) and so integrated with the book as a whole that their "accidental" character is either disguised or lost and their recurrence actually contributes to the unity of the volume by virtue of their consistent accentual value.

An arbitrary arrangement of tailpieces is likely to produce a "jumpy" effect. Since the spaces (at the ends of chapters or sections) within which tailpieces may be placed differ in area, such decorative elements cannot always fall on the same level. This irregularity can be compensated for in a measure by adjusting the size of the decoration to the area of the space it occupies. By what may seem to be a negation of the law of balance here insisted upon, such a variation is more productive of architectural harmony than tailpieces of uniform size would be if disposed in spaces of varying area.

Returning for a moment to the undecorated book, it may be remarked in passing that verse, particularly a collection of short lyrics, does not lend itself to good book-design. It should be enough to point out that the disproportion between type mass and white paper caused by short measure and the frequently meager letter-press deprive books of verse of the book's basic structural factor, the rectangle of type. How decoration can be employed to overcome this deficiency is perfectly exemplified in the original edition of Dorat's Les Baisers.

With the principles of balance and unity still in mind, it will hardly, I think, admit of contradiction that the scattering of odd-sized illustrations through the text is incompatible with both of these principles. Such illustrations, particularly those of irregular shape bounded on two or three sides by type, are as destructive of balance and unity as is poor fenestration in a building. It is not enough that something like a balance is effected on facing pages (an elementary principle in layout); the lack of a complete integration of the pictures with the book and the disturbance created by distorting the letter-press into odd shapes preclude the possibility of such a book being regarded as well-planned, much less ideal, however charming it may be in detail.

We have seen, while considering the major aspects of book-design, in what wise paper must be judged with regard to those first or immediate impressions gained from seeing and feeling it. I must now carry the consideration of paper a little farther. Since style and character are essential qualities of the fine book, we must insist upon these qualities in every element of its substance. Now style and character at their utmost are peculiar (for reasons that have to do with the methods of manufacture) to hand-made paper only, laid or wove, and, it may be further insisted, to only the best hand-made paper. Desirable as wove paper is for certain purposes, it cannot be denied that it has less character than the laid sheet. It is also true that no feature of fine laid paper gives more character to a sheet than the so-called "antique" factor, a slight thickening of the pulp and greater opacity along the chain-lines. By an "improved" method of mould-making, introduced by Baskerville, this thickening was eliminated, but, whatever mechanical superiority its absence may represent, there can be no question but that it represents a loss of character. All book papers produced by machinery (particularly the laids in which the effect of laid lines is mechanically faked) are as much imitations of and substitutes for hand-made paper as machine-made lace is a substitute for hand-made lace, and the disparity in quality is as great. We speak of "imitation lace" and "real lace," meaning machine-made and hand-made; we might, with equal propriety, speak of "imitation paper" and "real paper." Ideally, then, the fine book, in the fullest and strictest sense of the term, can be printed on no other than paper that is hand-made and of the best quality.

As to the color of paper for fine books, the whole question may be considerably clarified at once by the statement that everything suggestive of artificiality should be avoided. A paper that is chalky white or bluish white tells us at once that the rags which went into its manufacture were chemically (that is to say, artificially) bleached. A great many toned papers, described as "cream" or "india," are artificially colored and show it. The most desirable tone for fine book paper is the "natural" tone of unbleached (and sorted as such) linen rags. Its slight creamish color is at once pleasant to the eye and holds the promise of that agreeable mellowness which comes, very slowly, with age. A number of very pleasant books have been printed in recent years on gray, blue, green, and brownish papers (the last usually a deliberate simulation of ancient paper), but, despite their charm, they are, I think, open to the charge of affectation, against which, if true, there is of course no defense. If not actually "arty," they come perilously near to it.

It has not been my purpose in this paper to lay down the rules for making a fine book, for, after all, rules are of no use whatever (in an art or in a craft) except to be broken—wisely. Neither has any attempt been made, since this is not a technical treatise, to outline the methods by which the results described may be produced. I have tried merely to set forth the various criteria by which fine books should be judged and the principles (quite different from rules) that underlie them. If the "specifications" seem over-exacting, if they are to be dismissed as trop raffinés, I must ask the caviler if that which purports to be "fine" can be "too refined"? Let those who wish to compromise (with popular taste, with outlay and returns, with honesty, with self-respect, or with machinery) do so, but unless the thing they produce represents, with eloquence and beauty, the full and unconditional employment of every realizable aid to betterment, physical and technical, it is something other than a fine book of the first order. We must discourage ourselves in order that we may be strong.