P r e f a c e

The Italian text of the Libro dell'Arte has been edited four times. It has been translated twice into English, twice into German, and once into French.

The First translations, Mrs. Merrifield's [1844], and Victor Mottez' [1858], were based upon the [1821] edition of Tambroni. As has been shown, Tambroni's version was incomplete and inaccurate. The manuscript which he edited was an eighteenth-century copy, and its original, our L, has since been published. L, and another independent manuscript, our R, unknown to Tambroni, formed the bases of an improved edition published [1859] by Carlo and Gaetano Milanesi. Three translations were made from this: the first, by Albert Ilg, [p. ix] formed the first volume in the Vienna series of "Quellenschriften für Kunstgeschichte und Kunsttechnik des Mittelalters und der Renaissance." The next translator, Lady Chritiana J. Herringham [1899], found Ilg's German version "a most valuable book of reference in translating difficult passages." The last translation, that of Willibrord Verkade [1913], preserves the best features of its predecessors, and adds others of its own. I acknowledge freely and gratefully my indebtedness to all. The inadequacy of Tambroni's edition largely vitiates the translations made from it; and faults in the Milanesi's edition are reflected in the translations which it underlies. Though my publication of the Italian text represents no such major improvement at the Melanesi's wrought over Tambroni's, it seems to me a sounder basis for translation than has hitherto been available.

The Milanesi edition rendered obsolete Mrs. Merrifield's translation from Tambroni; and Lady Herringham's Book of the Art, incorporating the Milanesi's improvements, became the standard English version of Cennino's work. I venture not criticize it here, not in dispraise of her translation, for it has shown its merit, but rather to justify a new one.

As an example of the disabilities inherited from the Milanesi may be cited the following passage: "Vuole essere la colla più forte di verno che di state; chè di verno il metter di oro vuole essere il tempo umido e piovoso." Lady Herringham recognized that this could not be translated as it stood, for it is nonsense; so she took liberties with it. Vuole essere la colla più forte becomes "Size is stronger"; and chè di verno, "and in winter." The resulting translation makes sense, but not the sense of the original. Turning to the source manuscripts, we find that L has: Vuole essere la cholla piu forte di verno [p. x] che di state che di verno; but di verno che was canceled, and with it, by mistake, the final di state. Judicious editing thus gives the proper reading: Vuole essere la cholla piu forte di state che di verno. That is "Size wants to be stronger in summer than in winter." And this every good craftsman knows, none better than Lady Herringham herself.

Even as a translation, of the Milanesi text, however, Lady Herringham's work is not without flaws. Sometimes through misconception based on Mrs. Merrifield's version, sometimes through carelessness, she falls into error. Thus, in Chapter 33, she translates aguzza. . . si come stanno i fusi as "sharpen. . . as if they were tin." Like Mrs. Merrifield, in Chapter 136, she confuses smeriglio with smeraldo. In describing the breathing tubes for casting from life, she translates più aperte che di sotto as "more open below"; and sieno spesso forate dal mezzo in su con busetti piccoli as "let a small hole be pierced through the middle of each." It is not necessary to extend this list, and I am too conscious of the possibility of similar errors in my own work to wish to bring a heavy indictment against my predecessor.

Some defects in the Herringham translation must be attributed to a lack of understanding, generally shared by the others. So in the case of stagno dorato, in Chapters 95-99 and elsewhere, the distinction between "golden" and "gilded" has been missed. In Chapter 62 the translation "lye or a little roche alum," for lisciva e un poco d'allume di rocca, may be due to carelessness; but the readings, in the same chapter, "rich and deep," and "deep" alone, for violante in the original, betray the translator's failure to recognize that violante is a color [xi] term and means "inclining to violet." The translation of urina as "wine," in Chapter 153, is doubtless a typographical error, as is "beat" for "heat," in Chapter 173.

Another class of error, all too easy to fall into, must be mentioned. From Chapter 47 Lady Herringham has omitted the translation of the sentence "La sua tempera non vuol d'altro che di colla"; from Chapter 107, "E tria bene queste cose insieme, come puoi sottilissimamente"; from Chapter 138, "E così a poco a poco va' brunendo un piano prima per un verso, poi"; and, farther on, "per altro verso"; while from the chapter on gilded glass she has left out the important direction "lascialo seccare sanza sole per spazio d'alcuni dì."

Lady Herringham makes no pretense of consistency in the translation of trchnical terms. Her freedom in this respect robs her translation of much weight, and often leads her astray. One or two instances must suffice as illustrations here. "Fatness" and "leanness" of pigments are qualities which are sometimes found puzzling. Without entering upon any elaborate discussion, I may say that wet clay would be called "fat," and wet sand, "lean." The corresponding adjectives in Italian are grasso and magaro. Lady Herringham translates grasso now as "unctuous," which is good; now as "full bodied," which is less good; again as "opaque," which is not good at all; and, finally, as "rich," which is, at the best, ambiguous. She translates magro quite consistently as "transparent," even in the phrase "transparent and drying," in translation of magra e asciutta, to characterize sinoper, which is as opaque as a pigment can be.

These examples of defects in Lady Herringham's rendering are advanced, let me repeat, reluctantly, with all gratitude and respect for the good qualities in her work, only to justify the publication of a new translation. It is not to be supposed that my interpretation of the text will prove final in all details, but I hope that it will be found to represent the sense of Cennino's treatise more accurately than previous translations have done. Such improvement as may be found in it is due partly to close familiarity with the text itself; partly to familiarity [p. xii] with other medieval tracts of similar nature; but the real business of translation is a "laboratory" matter, involving a knowledge of each rule in practice. This knowledge Lady Herringham possessed in some degree. If my translation reveals a better understanding, the credit belongs to others: first, and in greatest measure, to Edward Waldo Forbes, to whom these volumes are inscribed, who first expounded Cennino to me, and whose researches are embodied in every page; then to the masters under whom, through the liberality of Mr. Forbes, I carried on my study; Nicolas Lochoff, the peerless copyist of early Italian painting, and Federigo Ioni, master of archaic styles and methods; and, finally, to the students who have carried out Cennino's precepts under my direction, and translated for me--with their paintbrushes.

The Libro del'Arte was "made and composed," its author tells us, "for the use and good and profit of anyone who wants to enter the profession." I have accordingly tried in my translation to give first place wherever possible to the convenience of the practicing student and painter. It must remain for another volume to analyze Cennino's materials and methods in detail; but every effort has been made in this to translate them into the resources of modern commerce and the idiom of modern craftsmen. Thus, minio is translated as "red lead," and verzino, as "brazil," The braccio and the spanna having passed out of fashion, I give equivalents in feet or inches, rather than translate them. The 'finger," as a measurement, is too indefinite to be reduced to any fraction of an inch, but is not hard to understand; nor are those rule-of-thumb proportions, "beans," "lentils,'" etc. Measures of time are dealt with similarly: thus, xv dã, o venti becomes "two or three weeks." By an extension of the principle, azzurro della Magna appears as "azurite," rather than "German blue" or, as formerly in English, "asure of Almayne"; and verde azzurro, as "malachite": these being the names not of modern pigments, for they are not found generally in trade, but of the minerals from which, as I believe, they [p. xiii] were made, and may be made today. In the same way, campeggiare is translated as "lay in; raffermare, as "crisp up"; ritrovare, as "shape up," etc., in the belief that those expressions represent the nearest equivalents to Cennino's terms which can be found in colloquial use by modern English-speaking painters.

There are many couplets in the text, such as triare o ver macinare, tavola o vero ancona, bicchierre o ver miuolo, the terms of which have generally been regarded as synonyms. I do not think that they are quite that, though it is often difficult to define the members. Cennino is more likely to indicate a synonym by cioè than by o vero, and I have followed his practice strictly. Thus, triare is translated "work up," and macinare, "grind." They are almost interchangeable, but triare is a little more likely to be used than macinare when the case calls for grinding dry, "triturating." Tavola is translated "panel," and ancona, being a word in good standing in our fine-arts vocabulary, is kept in the italian form. Those terms are not synonymous: the "ancona" is a compound panel. "Goblet" and "glass" serve to distinguish mugliuolo and bicchiere, though perhaps the terms are really interchangeable.

In some cases, Cennino's materials have no modern commercial equivalents. This is true of his bianco San Giovanni, which I translate as "lime white." In other cases, such as his sinopia, the meaning is too general to be reduced to any single commercial term. The word "sinoper" is not in common use by painters now, but I have pressed it into service, because, like its cognate sinopia, though a generic term, it may be used in a specific sense. To translate sinopia, "Venetian red," would be to fix arbitrarily upon one of many perfectly good earth reds, all of which Cennino would unhesitatingly have called sinopia; and that to no good purpose, for there are almost as many shades of Venetian [p. xv] red in modern trade as there are colormen who sell that universal pigment. The mixed color, verdaccio, likewise, is not a definite quantity, but merely a dark, greenish or brownish tone for outlinng and shading. The Italian expression seems to fill a want in our painters' lingua technica, for it is readily adopted by students.

There are a few materials which cannot be identified as yet with perfect certainty. Among these are giallorino and arzica, both yellows: the former a fairly bright, opaque one; the latter transparent and fugitive. Rather than risk a faulty identification, I have kept Cennino's names for them. Another question of identity arises in connection with vernice liquida; but whatever its ultimate solution, the phrase must be translated here as "liquid varnish."

In my efforts to preserve the professional character of this handbook, I have been forced into some awkwardness at times. Cennino's verdeterra has no other equivalent among English-speaking painters than the French name terre-verte. This has become, in certain quarters, so far anglicized that I am tempted to create the spelling "turvurt," to conform with the pronunciation common in trade. In the same way, the Italian gesso has been assimilated to our tongue as substantive, verb, and adjective. It has bred up a horrid trio: "gessos," "gessoed," and "gessoing," all of which will, I regret to say, be met with in these pages. If they give offense, be it remembered that the root word has still further potentiality for evil! I have consented to the extension of its grip upon us, and kept not only gesso, but two adjectives from the Italian as well: grosso and sottile. For this my justification follows.

The obvious translations, "thick" and "thin," are closed to us, for a reason which does not first appear. The modern practice of gessoing [absit injuria!], as performed commercially by frame-makers and the like [who have the decency to call it "whitening"], involves the use of 'thin white' and "thick white," which might be supposed to parallel Cennino's gesso sottile and gesso grosso, but this is not the case. Modern "thin white" corresponds to Cennino's second preliminary sizing, but contains a small amount of whiting; the 'thick white' is the gesso proper, containing much whiting, and forms the final surfacing material; whereas Cennino's gessos grosso simply serves as a [p. xv] foundation for the gesso sottile which comes after. These temptng modern similia, therefore, differ in application, and also in composition. Furthermore, gesso sottile must refer not only to a thin mixture, but to a material which is by nature "thin" or "subtle." In practice, this thinness becomes apparent; this subtlety, in comparision with the crude whiting of the modern gilder, or the heavy impasto of the gesso grosso. Painters and gilders who once work with it care not by what outlandish name it may be called; the only sufferers are those who read and do not paint; and this translation is for the painters.

In one section, that on casting, I have balked at consistency in the use of this word "gesso." Gesso there means "plaster," "plaster of Paris," and I have so translated it . [Among modern sculptors, "gesso" in English has this meaning; but painters understand by it a white priming preparation containing size, or the white filling of such preparation.] "Plaster," alas, has to do heavy duty in this little book: to serve as a noun for gesso here, for intonaco and smalto, and sometimes for pasta;and for smaltare as a verb. For we are poor, in English, in terms for palsterwork, and I suppose no tongue is richer than Italian.

I have endeavored, in the main, to avoid this sort of freedom, and to confine myself as far as possible, even at great cost to style, to a single English equivalent for every technical expression of the Italian. In some cases I have had to admit defeat. Colla, for example, just means "adhesive" in general; and no one English word will fit it always. There we are richer than Cennino, and I have been obliged to use "glue," "size," and "cement," to translate the one word colla. So also with i, which means sometimes "quill," and sometimes "pen" - "quill pen," of course. In a few cases the Italian possesses both singular and plural of a word which with us exists in only one. Thus, "charcoal," for carbone, has to become "coals," for carboni.

Some of the matters of which Cennino writes are unfamiliar to the [p. xvii] general reader, and no ingenuity in translation will make his words clear without comment. The tool called a "slice," the composition known as 'vermeil," colors in the form of "clothlets," are likely to need explanation to some readers, at least, though all these words are proper English, and the best translations to be found. In such cases the New English Dictionary is specifically cited; and it may be consulted generally as authority for definitions here.

In Cennino's occasional attempts at rhetoric, as in the first chapter, he is apt to lose himself in complications. He does not always finish his sentences, and has a disconcerting habit of changing persons and tenses and moods in the middle of an instruction. It has seemed to me pointless to preserve lapses of this sort in my translation, when his intention can be understood. In some parts, it must be confessed, a good deal of sympathy is called for; elsewhere, his little slips are easy to understand and overlook. I have tried to keep something of the flavor of his writing, but have felt as free to recast an awkward sentence as to spell corrrectly the words with which I translate the mispelled words of my original.

If I speak slightingly of Cennino's rhetoric, let it not be supposed that I take no joy in his expression. In the very heart of the tangle of his first chapter lies a sentence, defining the "occupation known as painting," which seems to me as precious as anything in the book. By this definition, painting "calls for imagination, and skill of hand, in order to discover things not seen, hiding themselves under the shadow of natural objects, and to fix them with the hand, presenting to plain sight what does not actually exist." And therein lies the secret of much good work.

There are some obvious lacunae in the text, not to be supplid from manuscripts of the Libro dell'Arte now in existence. Some of these can be filled out fairly satisfactorily; and I have generally attempted to repair them in this translation, either by inserting, in pointed brackets [&lr&rr>], such words as I think necessary to complete the sense, or by indicating in footnotes the possible character of the omissions.^ Pointed brackets are used also for a few words introduced into [p. xvii] the translation to make the meaning clearer. The only editorial additions not so bracketed are the chapter headings which I have supplied from the point where the titles in the manuscripts cease.^

When Tambroni first published the Libro dell'Arte, he devised titles and numbers for the chapters which followed Cap. CXL, in his manuscript. Seventeen new titles were supplied by the Milanesi to account for chapters in R which they brought to light. These follow the chapter numbered CLX by Tambroni. Tambroni and the Milanesi made up these titles, in most cases, out of extracts from the text itself. They are consequently rather apt to be long, and sufficiently archaic in flavor to convey a false sense of authenticity. There seems to be some convenience to the reader in having the text divided into chapers and in having these chapters provided with headings of some sort. I have, therefore, made such divisions, and furnished them with titles. These titles are not bracketed, but are printed in italics, to remind the reader that they are modern inventions. In the number and wording of these headings, I have departed freely from the Tambroni-Milanesi formula.

No chapter numbers are attached to these spurious headings in this edition. As I have pointed out, neither of the existing systems of numbering is "to be considered as an accepted standard; and the two together constitute a mechanism for more than doubtful utility." Since, however, references to this work are often made by chapter number, I give in footnotes the numerals attached by Tambroni and the Milanesi to their divisions of the text. Thus, "T., CLXXI; M., CLXXXIX" stands for "the chapter numbered CLXXXIX in the Milanesi edition, the translations of Ilg, Lady Herringham, and Verkade, and the revised edition of Mottez' translation; and the same chapter, numbered CLXXI in the Tambroni edition, in Mrs. Merrifield's translation, and Mottez' as originally published." More than [p. xviii] this I cannot do: the reader must determine for himself to which series of numbers his reference belongs!

In adding these chapter headings, I have gone a step farther, and continued the division of the work into "sections," as begun by Cennino. Five of these sections are indicated in the text, and deal with the following subjects:

I. Drawing
II. Colors
III. Fresco-painting.
IV. Oil-painting and embellishments for the wall.
V. Glues, sizes, and cements.

The conclusion of the fifth section is erroneously marked in the manuscripts as the conclusion of the fourth. I have distinguished the following divisions of the remainder, not specifically indicated in the manuscripts:

VI. Panel-painting and gilding.
VII. Mordant embellishments.
VIII. A short section on varnishing.
IX. A short section on illuminating.
X. A section dealing with work on cloth.
XI. A short section on operations with glass.
XII. Part of a section dealing with mosaic.
XIII. A section dealing with miscellaneous incidental operations.
XIV. The final section, devoted to casting.

These divisions are not very significant, but may serve to emphasize the orderly charactaer of Cennino's composition.

I hope eventually to bring out a digest of the medieval writings prior to Cennino's which deal with subjects treated in the Lilbro dell'Arte. On that account I have been sparing, in my notes, of references to those sources; preferring to assemble them methodically later. I have not hesitated, however, to anticipate a little, when a quotation would serve to clear up some difficulty in the text, or confirm the accurcy of a rendering pointedly different from my predecessors'. I have likewise postponed for future treatment most of the comments [p. xix] which I have to make upon the theory and practice of the techniques here described. The purpose of the present volume is to translate the text: its full interpretation will be undertaken separately.

To this translation my teachers, students, friends, and colleagues have all contributed. Edwin Cassius Taylor, Chairman of the Department of Painting at Yale, has fostered the modern application of Cennino's methods, to my great benefit. To my colleague, Lewis E. York, I owe not only the two drawings which illustrate Cennino's casting methods, but also a great deal of helpful practical criticism. More thanks are due, I fear, than the quality of this work entitles me to render, but I cannot leave unrecorded the generous interest extended over many years by Francis Peabody Magoun, Jr., hortatu praeceptisque. [p. xx]