PRINT February 1984


Jazz and Cinéma calendrier du coeur abstrait. Maisons

These two recently published reproductions of illustrated books provide an interesting contrast. Cinéma calendrier du coeur abstrait. Maisons, a book of poems by Tristan Tzara with woodcuts by Hans Arp, was originally published in 1920, in an edition limited to 150 copies. As illustrated books go, it was a relatively modest but handsome production, with the nicely printed poems and unassuming but lovely black and white illustrations set in gentle counterpoint to each other. A charming and witty book, its reproduction does not involve enormous difficulties and its essential character is quite successfully captured in the 1983 Thomas Press reprint. Matisse’s Jazz, originally published in 1947, is by contrast one of the most ambitious illustrated books of the century. It contains a rambling, indirectly autobiographical text by Matisse, reproduced in his own handwriting, and 20 brilliantly colored pochoir plates, carefully executed after Matisse’s original cutouts. Here the usual relationship between pictures and text is reversed. The pictures are the main subject of the book and are in effect “illustrated” by the text, which provides a graphic black and white counterpoint to their vivid colors. Because the pictures are so important and so complex, the reproduction of this book must necessarily be fraught with problems—as is painfully evident in the Braziller facsimile.

The worst thing about the Braziller reproduction of Jazz is that people who have never seen the original will think that there is virtually no difference between the two. The Braziller book is almost the same size as the original, and except for the addition of an introduction by Riva Castleman and an (awkward) English translation of the text, and the fact that it is bound instead of folded and boxed, it literally reproduces a specific copy of Matisse’s Jazz. At first glance it seems to look just like the original. But to anyone familiar with Tériade’s 1947 edition, the Braziller book is an excellent illustration of the adage that a reproduction is exactly like the original except in everything.

Feeling of color and of pictorial texture are crucial to the reproduction of Matisse’s cutouts, and it is precisely these qualities that the Braziller book lacks. This failing has been glossed over in the extensive publicity that the book has received and in the reviews and articles I have read about it. Everyone seems to take for granted the matter of fidelity to the original. Terms like “faithfully” and “exquisitely” reproduced are mixed with comments about how “authentic-looking” or “remarkably beautiful” the book is. Indeed, it is repeatedly referred to as a reprint, or republication, or new edition of Jazz, rather than as a reproduction or facsimile. Attention is repeatedly called to the care that is supposed to have gone into its manufacture and to how special paper and specially made inks were used to insure its having the look and feel of the original—which, it is also pointed out, sells for about $40,000, while the reproduction can be had for a mere $90.00.

Unfortunately, it seems that you do get only what you pay for; and I would say that the difference in price pretty well reflects the difference in quality between Braziller’s and Tériade’s Jazz. The Braziller book is woefully lacking in soul; and that is because it is regrettably lacking in material quality. The paper—specially made though it may have been—is so flimsy that you cannot avoid seeing through the page you are looking at to the text that is behind or under it. If this is irritating when you are reading the text, it is maddening when you are trying to look at the pictures. For although the plate pages have no text printed on the back of them, the text on the pages under them shows through quite distinctly and spoils the pictures. The only way to deal with this problem is to lift up the page (or pages) you are looking at so the writing underneath does not show through; but then you can never quite see the whole picture, especially if it is a double page. This fault is simply inexcusable, since it could easily have been avoided by using better paper.

The question of fidelity to Matisse’s original extends to the reproduction of its intense and powerful colors. Almost without exception, the colors in the reproduction are dull and lifeless—a travesty of the original. In her informative introduction to the Braziller book, Riva Castleman quotes a letter Matisse wrote while he was working on Jazz, in which he jokingly asked Tériade to send him a white cane and mentioned needing dark glasses in order to protect his eyes from the bright colors. And indeed, when you look at Tériade’s edition of Jazz, some of the plates actually make your eyes pop. This effect is completely lost in the reproduction, even in the most intensely colored plates.

Unlike the often excellent manuscript facsimiles that Braziller has done in the past, his edition of Jazz is not a reproduction of a unique work, the original of which can be seen in only one place in the world, and then usually only by experts. The original edition of Jazz consisted of 270 copies of the book and of another 100 portfolios of unfolded plates without the text. Thus although most people cannot afford to buy Jazz, most people who are sufficiently interested could probably gain access to a copy in a not-too-distant museum or library. But ironically, this reproduction of Jazz will in fact probably discourage viewing of the original. Because a reproduction is now available, museums will probably feel less obligation to make their fragile copies available to the public. And, paradoxically, because the reproduction is taken to be a substitute for the original, more people will know about the book, but fewer will feel impelled to go and look at the original. The Braziller book, poor as it is, will by and large probably become Matisse’s Jazz for most people.

I think that one way to judge a reproduction of a book is by how it makes you feel about the original. When you look at a facsimile of the Book of Kells or the Très Riches Heures de Jean, Duc de Berry, for example, you are quite distinctly aware that you are looking at a mechanically produced facsimile of a handmade work; the signals of the differences between the two kinds of objects are somehow built into your perception of them, and you automatically make the appropriate compensations. Moreover, looking at such a facsimile makes you want to see the original. (A similar feeling was engendered by a tiny, abridged edition of Jazz published by Piper Verlag in Munich in 1959: it made you want to go and see the real thing.)

Another kind of reaction is that engendered by a book like the Thomas Press facsimile of Cinéma calendrier du coeur abstrait. Maisons, a book meant to be read rather than looked at. Here the reprint—which includes very good translations of the poems by Mary Ann Caws—is satisfying in and of itself, despite some minor faults, such as the excessive inking of some pages. Not much is lost in the reproduction process, and it is a delight to have.

With Jazz, however we are dealing with a book that was meant primarily to be looked at, and which must therefore be regarded more as an album of fine prints than as a book. Because the original edition of Jazz was also mechanically reproduced, the difference between “reproduction” and “original” is not as immediately clear as it would be with a facsimile of, say, a manuscript. Moreover, the original Jazz was itself quite literally a kind of “reproduction,” since it was composed of pochoir reproductions of Matisse’s original paper cutouts and of gravure reproductions of his handwriting. And although Jazz is now rightly regarded as one of the greatest artist’s books of the century, Matisse was initially displeased with it, precisely because it looked like a reproduction. “It is absolutely a failure,” he wrote to his friend André Rouveyre less than three months after the book was published. “I believe that what absolutely spoils [the pictures] is the transposition [from cutout to pochoir print] which robs them of sensibility, without which what I do is nothing.” And although the judgment of history (and the later judgment of Matisse himself) affirmed the artistic success of Jazz, Matisse did not do another book in this manner.

Matisse did do a number of other cutouts expressly for printed reproduction, as book covers and posters; and with 11 of these he insisted—as he had with Jazz—that the reproduction be as accurate as possible with regard to color and texture. The Braziller reproduction of Jazz does not meet the high standards the artist set for the reproduction of his work. In the long run, it does both Matisse and Jazz a disservice.

Jack D. Flam is a professor of art history at Brooklyn College and the Graduate Center of the City University of New York.


Henri Matisse, introduction by Riva Castleman, trans. by Sophie Hawkes, Jazz (New York: George Braziller, Inc., 1983), 176 pages, 20 color plates.

Tristan Tzara, woodcuts by Hans Arp, trans. by Mary Ann Caws, Cinéma calendrier du coeur abstrait. Maisons (Cinema calendar of the abstract heart. Houses), (Ann Arbor, Michigan: Thomas Press, Inc., 1983), unpaginated, 33 poems and 19 black and white woodcuts.