THE BIG PICTURES.

I've always fetishized the "complete works" editions of my favorite writers and daydreamed about the prospect of reading any one of them from A to Z (which of course I've never done). I loathe anthologies and the arrogance of the editor who chooses for the reader what is worthy of interest. Nothing fuels my fantasy more than the illusion of having in front of me a lifework whole, with all its moments of grace and oddity there to be unearthed (even if the exhaustiveness can only be temporary, as the appearance of countless "revised and expanded" editions of books warns us, and with the criteria for completeness ever expanding). If the museum postcard rack never seems to offer the same experience, what I enjoy most while perusing the writer's complete works is that I'm the one who makes the selection of the keepsake.

This sense of empowerment and substitutive possessiveness is accentuated in the case of the catalogue raisonne (perhaps paradoxically, for the reason that what is presented for our perusal is not the works per se, as in literature, but reproductions of them: a catalogue raisonne is a kind of surrogate collection). In any event, with the catalogue raisonne the recording of a "complete corpus" dramatically underscores a dialectical effect: It is only when confronted with the "whole" oeuvre that one is able to realize in which singular way a particular oeuvre is not whole, in the sense that things don't "hang together" - and it is precisely the anomalies, the false starts, the strange offshoots that one immediately notices when leafing through a book that consists in the main of hundreds of images in sequence. (The order is what's of the essence here - that's the "raisonne" aspect, so called perhaps as a paean to Descartes: The clearer the order, the more conspicuous the perturbations.) This paradox is not always acknowledged by the authors of catalogues raisonnes. Because they become de facto experts who are responsible for attribution - and because their research has immediate consequences in the marketplace - they are often trapped, sometimes unknowingly, into consolidating the myth of the artist's consistency once he or she has attained a mature style. But the joy of the reader has much to do with the sense of being put in the loop, of suddenly becoming aware of the hitherto unknown divisions that fracture a subject, the cleavages that are indeed what make this subject particular.

Catalogues raisonnes are at most only infrequently reviewed (David Sylvester, who produced the Magritte catalogue raisonne, told me recently that not a word concerning his mammoth achievement appeared in print). This has a lot to do with the genre's dryness and its just-the-facts-ma'am posturing. But for all their common denominators, particularly what I call the ID card - entries on individual works providing title, date, medium, size, provenance, present location, exhibition history, bibliographic references - catalogues raisonnes come in all shapes and sizes (though they are always bulky). Sometimes the cataloger asserts his or her authorship as a critic, which is not without risks (think of poor Arturo Schwartz, lambasted in these very pages just over a year ago, rightly or not, for having empurpled his entire Duchamp effort with the soap opera of incest theory). Other times he or she hides behind the ethos of positivism and, rather than deciding which facts are significant, opts for guidelines that permit only the strict minimum of interpretation. My ideal is somewhere in between: Having rummaged through the archives and seen all the works included in the corpus, the cataloger arguably knows more about the topic than anyone else, and when he or she has interesting things to say, we should be all ears.

Four recent publications provide a good cross-section of the genre.

In some cases the gatherer, to my chagrin, was excessively modest. Cy Twombly: Catalogue Raisonne of the Sculptures (Schirmer/Mosel, $199) is a splendid, luxurious book in which almost every one of the 147 works covered is given a full-page color reproduction, and in some instances another, smaller view (a handful of black-and-white photographs document the sculptures that are no longer extant). It was assembled by Twombly's assistant, Nicola del Roscio, who provides a flawless JD card for each work. But del Roscio, who refers to himself here as an editor, a rarity in such publications, refrains from telling us anything about the process, formation, context, and so forth of these sculptures. The book's only commentary is Arthur C. Danto's elegant preface, in which he distinguishes Twombly's archaizing effigies from Duchamp's readymades and ponders over the relation between Twombly's sculpture and painting. We are left more or less alone in wondering, for example, about the different versions of a given work, or about the huge lag in time between the elaboration of an assemblage in "wood, wire, twine, nails, house paint and wax on fabric" (no. 10, 1953) and its casting in bronze (painted white, as is often the case with Twombly's work, and thus looking as if it had been executed in plaster) for an edition of eight (no. 11, 1989). This catalogue is indeed a gem, not only because of the quality of its printing but also because it facilitates comparison between "original" and cast, and allows one to think about the uncanny timeliness of Twombly's sculptural output (there have been no stylistic changes whatsoever from 1948 on), but we might have wished for a more authorial presence from the cataloger, whose intimacy with the work is topped only by that of the artist.

Such reticence might have to do with the fact that the artist is alive and active, but another recent publication shows that an artist can also be looking posthumously over the editor's shoulder. Paul Klee Catalogue Raisonne: Volume I (Thames & Hudson, $225) covers the years 1883-1912 (eight more volumes are in preparation). As in the case of the Twombly, its 901 individual entries are bare-bones ID cards (unlike the former opus, however, the design of this catalogue does not shine: What is the Swiss fixation on sans-serif fonts?). Begun ten years ago, this is a collective affair (the book is unsigned). Yet, though anonymous looking, it definitely has a main author: Not the last person in charge (Christian Rumelin, who authored the short, informative introduction), but the artist himself, the first cataloger of this oeuvre.

Klee adopted a strictly chronological order, numbering his works in the course of making them, independent of medium. He started recording his works in 1911, and it is also at this time that he retrospectively compiled (signed, dated, numbered, etc.) all his production "since [his] childhood." The Swiss team was wise to follow Klee's model - and augment or correct it when there was evidence for doing so - for it reveals that historical sequencing was more important to the artist than medium, and one wonders whether the rule should not be applied to most artists of this century whenever precise dating is available. This organizational principle demonstrates how much Klee's famous experimentation with subjectiles in his mature years is grounded in a very early to-and-fro between media - and one strongly regrets that more detailed information about Klee's exceptional variety of techniques was not made available to the reader. Since the handful of oils are not isolated in a category of their own, one has a hard time singling them out. As they strongly resemble the much larger number of works on paper or cardboard, they do not jump out at you - which means that from the start Klee conceived his paintings, in some sense, as drawings.

An even more idiosyncratic aspect of the Klee catalogue is that it starts in childhood (the first recorded work dates from 1883, when the artist was three years old - and he is only sixteen at number 100!). No artist, to my knowledge, has ever included in his corpus his work as a toddler (Picasso's first recorded works date from ca. 1890, when he was nine, and he denied their infantile character). Klee's fascination for the "art of children" is well known (a substantial literature exists on the topic), but coming on the heels of the widespread display of de Kooning's late works (at the other end of the biological span, when the artist was ill with Alzheimer's disease), the publication of this dossier will force us to reconsider the nature of artistic agency. I have to admit that Klee always bored me because I found him so predictable: By its very inclusiveness, this first volume changes my mind (I hope the following ones will follow suit).

In contrast to the Twombly and Klee, Mark Rothko: The Works on Canvas: Catalogue Raisonne (Yale University Press, $150) is very much the work of an author. Given his fabulous command of the material, it is unfortunate that David Anfam's brilliant comments on individual works (covering technical, stylistic, thematic, and formal issues, not to mention aspects of conservation) are not dispersed among the entries. One can easily guess that with such a vast production (there are 834 entries), and one that becomes increasingly serial, this decision was made to avoid redundancy, but a compromise could have been found so that peculiarities would be highlighted in the entries. Rather, Anfam has written a long and beautiful essay in four parts, with abundant remarks (sometimes parked in the footnotes) referring to the works reproduced later in the book, all in color except for the few works no longer extant. The design is gorgeous but terribly impractical, as one is obliged to shift constantly between Anfam's text at the beginning of the volume (sadistically set by the designer in a tiny font and in lines that run on for several miles) and the illustrations. The more one advances in the essay, the more the muscular effort becomes unbearable (this is a heavy tome). Those who hazard the courage to brave these obstacles and treat Anfam's opus as something other than the magnificent coffee-table book it also is will be rewarded: It is far and away the best monograph ever written on Rothko.

Anfam has avoided most of the booby traps associated with the painter: the transcendentalist "spiritual" mishmash (he explains its function in the Rothko literature but keeps his distance), the "biographism" (no, the late "black paintings" are not about depression), the referential mania (no, there is no landscape there). And he provides ample new food for thought (I did not know, for example, of the two oblong predellas that Rothko at one point considered placing under some of his Harvard murals). Anfam is particularly eloquent in showing how Rothko increasingly orchestrated his solo exhibitions as color symphonies. My only major criticism (besides the fact that the book is written almost as if Barnett Newman never existed - he's mentioned only once in the text) brings us back to the issue of wholeness: Even though Anfam is cautious at first not to present the early works as premonitory ("Rothko surely cannot have prophesized his idles fixes from scratch," we are told about several late-'30s canvases that seem to prefigure his post-1950 style), he gradually loosens the reins over what seems to be a natural tendency among monograph writers toward anachronism. The sheer accumulation of technical, formal, and thematic continuities that Anfam stresses in passing end up undermining to an extent the apt periodization he proposes. Fortunately, Anfam's main, and extremely convincing, argument - that a poetics of shadows governs Rothko's work before 1936 and after 1957 - reaffirms a temporal disjunction that is essential to understanding the last decade of the artist's production.

Piet Mondrian: Catalogue Raisonne (Harry N. Abrams, $250) is a monumental enterprise, to which the two scholars involved - Robert Welsh covered the artist's production until early 1911, and Joop Joosten the years 1911-44 - have devoted a good thirty years. The first, smaller volume contains 710 entries, while the second divides between the Cubist and neoplastic works (415 entries) and the post-1911 "naturalist works" (154 entries), such as the watercolors of chrysanthemums made in the '20s, painted or drawn by Mondrian in imitation of his early style. There is no book I had awaited more eagerly and, as far as scholarship is concerned, my expectations were more than satisfied: The quantity of new data unearthed (notably the hitherto unpublished letters that are quoted in the enormous and detailed individual entries), the sheer number of previously unknown paintings discovered, the detailed, month-by-month chronology (close to a hundred pages of lilliputian script for the second volume alone, revealing a much more active artist than one imagined) - all this was bound to send a Mondrian freak like me into pure ectasy. It is far too dense, too packed for me even to attempt a proper review in the space here. Suffice it to say, the book is an absolute must for every scholar of twentieth-century art, providing the most thorough documentation ever concerning the invention (and sustained pursuit) of abstraction by one of its pioneers.

But there is an extremely depressing proviso (and the authors must have been devastated). Despite the hefty grant from the J. Paul Getty Trust for the production of the book (especially targeted to ensure the quality of the photographic reproduction), the catalogue raisonne is scandalously hideous. The design is a climax of ineptitude, the text barely legible without a magnifying glass (has anyone ever sued a publisher for eyestrain?), and I do not recall ever seeing such poor-quality reproductions (apart from the tiny selection of color plates, mostly of works that have already been reproduced many times). The 1,500 or so black-and-white photos are postage stamp-size and so badly printed that they look as though they were filmed from Xeroxes. The sketchbook pages and the late drawings are particularly ill-treated, and the myriad documentary photographs so patiently amassed by Joosten (particularly those of exhibitions during Mondrian's lifetime, which tell so much about his early reception and about early states of paintings later modified) are practically illegible. The original publisher, V & K Publishing, which subsequently sold the rights to Abrams, garbled this work in a major way. Thanks to this sloppy effort, mediocrities such as Herbin and Robert de la Fresnaye, not to mention the tons of living artists whose galleries finance their catalogues, are better served than Mondrian. My only hope is that the book will soon be out of print and that someone more responsible will undertake to reissue this fantastic store of knowledge in the form it properly deserves. Publishers beware: Every catalogue raisonne does not have to be as lavish as the Twombly volume mentioned above, but it need not be an eyesore.

Yve-Alain Bois is Joseph Pulitzer, Jr. Professor of Modern Art at Harvard University and a contributing editor of Artforum.