FASHION HAS BEEN with us in the Western world for 700 years, but why it should exist at all and what activates its never predictable rhythms are matters not often written about, though descriptive and prescriptive accounts are a staple of publishing. Some ingenious explanations of its driving force have been put forward: James Laver believes its is a response to the zeitgeist, Quentin Bell that it exemplifies Veblen's Theories. Anne Hollander tells us that it is an embodiment of messages delivered by art: We dress in order to look the way paintings and photographs tell us we ought to look. This theory seems to beg a question: Where does the artist get his ideas? Wherever that may be, the author of Seeing Through Clothes believes social change has little part in the process. To have fashion at all, According to her, we need an artistic tradition; but what else? It isn't want of an artistic tradition that has spared the Orient the fluxes and fevers of fashion.

If she cannot satisfactorily explain why fashion is volatile, she has nevertherless thought about the subject long and deep, and has interesting ideas about clothes as art shows them to us. For, as she says, until photography began to disclose unidealized facts in the 1840s, the best evidence for how clothes looked, how they wre worn, how they felt to the wearer, was that provided by painters and sculptors. Indeed, she goes so far as to claim that "the clothed figure looks more persuasive and comprehensible in art than it does in reality." Yet the only reality we know is contemporary. Art is the cheif witness, before photography, of what people looked like, clothed or not; but not the only one.

A lot of old clothing is still around to be seen that pre-dates photgraphy by centuries; but Hollander insists that these mute garments tell us little once the wearer has left them behind, and that as minor art forms they cannot compare to china, furniture, silver, clocks. Maybe. But Catherine the Great's silver wedding dress displayed in a museum on a headless dummy in a spotlit blaze had a thrilling presence; or-a more poignant instance-folding up the worn clothing of a dead parent is an almost insupportably intimate communication handwriting, clothes-even massproduced clothes-preserve an individual mortal presence.

Whether or not clothes can speak without their wearers, Hollander shows, with the help of bountiful illustrations, that naked human bodies, as depicted in art, have always worn the ghosts of fashionable garments. This is more than a matter of what artists and the wearers themselves see; 140 years of photographic evidence confirms that people stretch and contract and tilt their naked bodies to fit their absent costume. A 15th-century Eve, with her tiny breasts, big belly, and short legs slips into medieval dress as neatly as a paper doll; the dimpled flesh of a Rubens mirrors the bunched drapery of 17th-century costume ("Nameless anatomical bubbles and unidentifiable waves agitated the formerly quiescent adipose tissue under the mobile hides of nymphs and goddnesses as they simultaneously agitated the satin sleeves and skirts of the newly fashionable free-flowing clothes"); Goya's indressed maja's pushed-up, separated breasts and tightly crossed legs conform to Empire dress; even Blake's long-legged specters could be fashion plates of 1800. Toros are extended or contracted, thighs impossibly lengthened, buttocks thrust out, backs arched, breasts diminished of enlarge, all to confirm to fashion.It's easy enough to date a painting by costume; the idea that a naked body is just as exact an indicator as dress, while been dealt with in such scrupulous detail.

The difference between naked and nude was established by Kenneth Clark 25 years ago. Anne Hollander makes interesting use of the distinction as it regards clothing. Nudity represents and ideal: drapery is its exiguous dress. Nakedness is real: what has just been taken off or is about to be put back on is not some ennobling length of airborne cloth, but actual clothes-with buttons, cuffs, seams, and obedient to the laws of gravity. There's nothing like the presence of a discarded garment to invest the naked figure with high erotic charge, and no naked body so sexy as on distorted to conform to the fashion in dress. (This persuasive observation does not account for nudes portrayed by societies that knew no fashion. Hindu and Hellenistic sculpture, for instance whether explicity erotic or not, oozes sex.) Drapery, on the other hand, she continues, defuses sexuality, rendering nudity exemplary of some large truth, just as in workaday portraits of fully dressed people it serves as a sort of "allusion to refinement," swagged across the background, slung over a shoulder, fingering a bosom. (Again, the nice distinction is marred by evidence: The calyx of drapped cloth from which a Hellenisitic nymph or Venus emerges does nothing to subvert her earthy appeal.) Hollander writes with relish and conviction about draped cloth in its idealizing and transfiguring functions: Christ's fluttering loincloth, six yards of other-worldly linen; the surging robes of angels that buoy them up while their wings remain stiffly folded.

She does not scant male sex appeal and writes of a St. Sebastian whose "garments are shown delicately infastened and pushed downward as if by a lover's caressing hand. He gazes languishingly at the beholder; and his torso wears the elegant lozenge shape dictated by the prevailing mode, so that he seems a more deliberately desirable figure. His torturers are very near him, like prospective rapists, pretending to be archers." The clothes men wear reached their erotic climax in Regency dress: "Between the gleaming short-waisted coat...and the glittering boots was a soft, creamy expanses of tight-fitting doeskin ... which has the effect of riveting the eye on the male crotch.

On other particular she is good: On the color black and its emotive value in setting the wearer apart from the gaudy throng; its anti-fashionable use by romatic rebels and dandies; its 19th-century decline into the dress of the sober multitude of industrious men and self-effacing women; its 20th century revival in "the little black dress," its lingering assocation with danger and death (she forgets to memntion Hell's Angels, and I wish she'd say why bl ack "has obvious appropriateness for mourning").

On breasts and their changing size and shape, the significance of baring one as opposed to both; the use made by painters like Delacroix of halfbared bosoms to signify social unrest; the ups and downs of decolletage.

On underwear, which until the last century was little but an unadorned smock-like shift, with stays and corsets worn over it to alter the body's shape. But anyone unlucky enough to be spilled out of carriage would have exposed what Rowlanson's rude drawings delighted in showing-a bare bottem; for underdrawers were unheardof. By the end of the 19th century intricate frilled undergarments, including panties at last, were on absession, a shot in the arm to pornography.

On photography and the part is has played not only in showing us what people really looked like, as apposed to what artists said, but its influence on ideal looks. "It is the painted portraits of people in modern elegant clothing," she shrewdly notes, "that have become frumpy-looking and only photographs are really able to reveal the true essence of mid-twentieth-century chic-especially candid photographs." Photographs has certainly heightened self-consciousness; it has had everything to do with the way we apprehend clothes and relate to them. Photographers, models, and stars spread new styles, postures, and gestures around the world.

Of course our own clothes are the only ones we can understand inside out, and Anne Hollander is most original and perceptive when writing about the present. What people used to wear we can admire and marvel at, but it will never talk ofr language, the language we depend on clothes to speaks, conveying cryptic messages about ourselves. If we expect them to be a disguise, like dark glasses, we are deluded. They give us away. It's no paradox that when an artist wants to invest his human figures with universal meaning he does better to dress them in real clothes than in ideal drapery. She cites the mysteriously affecting paintings of Caspar David Freidrich, whose allegorical men and women wear impeccable Biedermeier costume. Maybe it's the clothing, so strange and yet so unselfconsciously worn, that gives old photographs a part of their intensely moving power.

Seeing Through Clothes should be a fascinating book. Unfortunately it is incoherent, repetitive, overloarded with evidence, and except when she's describing visual images (as in that spendid quote about St. Sebastian) her solemn prose is overwrought, self-conscious, and often unintelligible. How can a book about a subject so resonant with hidden meanings, masked anxieties, and profound satisfactions be tedious? Alas, This on often is. CAPTION: Illustration, an illustration in "Seeing Through Clothes" From "Design by Erte" (Dover Publications); Picture, "L'Odalisque," By Francois Boucher