I am not a great fan of the fad for guestcurated exhibitions culled from a museum’s collections. How vain is the notion of the curator-as-artist that they convey! There are exceptions, of course: the “Artist’s Choice” series at MoMA, for example, was unexpectedly lacking in pretense while providing valuable information about the way artists thought.

But what if it is the very form of the exhibition, the exhibition as form, that is at stake rather than the subjectivity of the curator? The first guest-curated show to have been conceived in such reflective, typically Modernist terms was “Raid the Icebox I with Andy Warhol” at the Museum of Art of the Rhode Island School of Design in 1969 (it also traveled to Houston and New Orleans). Warhol not only pulled things out of storage, but also foregrounded the storage itself—its absurd accumulations, its dormancy in oblivion. He zoomed in on series, masses, pinnacles of randomness: a whole cabinet full of two hundred pairs of shoes, all as thoroughly documented in the catalogue as the Seurat drawing or the Cézanne painting shown alongside its neighbors on the storage racks; a chest of Navajo blankets; a row of unremarkable Windsor chairs, etc. As David Bourdon wrote in the catalogue, Warhol picked a group of “sixteenth to eighteenth century paintings” just as they were stacked in the cellar—some with “only their backs showing . . . including the miscellaneous sandbags that were strewn around on the floor.” The matter, as always with him, was choosing the nonchoice.

And what if it is a matter of showing not a particular choice but choice per se? The pair choice/nonchoice is dialectical, in fact, and the one is perceptible only against the background of the other. So just as Warhol was able to display randomness as entropic resistance to the structure of exclusion that lies at the foundation of any museum installation, any attempt at exhibiting choice as such will have to be pitted against indeterminacy: one must define a boundary between an ordered inside (where not every choice is possible) and an infinite, indiscriminate outside (where no choice is really possible because all are equally insignificant). This is exactly what art historian Hubert Damisch did in “Moves: Playing Chess and Cards with the Museum,” the show he recently curated at the Museum Boijmans Van Beuningen in Rotterdam.

How to materialize, to make tangible, the finite space of decision? Damisch takes chess as a model, “for at any time during a chess game,” as he writes in the catalogue, “the distribution of pieces on the board can be considered either the product of a given history (the succession of moves from which it results) or a ‘position’—in other words, a system—which contains all the necessary and sufficient information for the player whose turn comes next to be able to decide a move in an informed manner.” The basic idea behind this “experiment,” as Damisch calls it, is that any museum visitor plays a kind of chess game with the collection (a certain number of moves, among which he or she must choose, are at his or her disposal in a space that is organized in a manner at once linear and simultaneous), as does any curator (having a limited choice, once again, since “works of art . . . are like chessmen, each possessing its own, preinscribed movement”). But if this condition is usually ignored by both kinds of players, Damisch proposes to underscore it.

So the floor of the central portion of his exhibition was conceived as a vast black and white chessboard with some squares empty and others containing a single art object (paintings were mounted on freestanding traveling crates). The pawns, more uniform than the other “pieces,” were drawn from the Boijmans’ spectacular collection of applied art (for one “side” or “team,” silver objects; for the other, mainly glass, or china). The most striking effect was the geometrization of traffic: as one was moving through the field to examine the content of the various occupied squares, one was necessarily performing a chess move (straight, oblique, crooked). Each move was freely chosen (there was no linear path to follow, no sequential numbers given to the artworks), but none was free: choice is not free will.

One need not be a chess adept to make sense of the game. On a sheer iconographic level, Damisch provided easy access: Bruegel’s Tower of Babel was an obvious rook; one of Kandinsky’s Blue Rider canvases, a knight; the queen (Torso, 1936–71, a sculpture by Man Ray) and the king (Rodin’s key-bearer in the Burghers of Calais group, 1885–86) were prominently displayed. Man Ray’s Obstruction, 1920—the first suspended sculpture in twentieth-century art, a kind of readymade mobile consisting of a cascade of coat hangers—functioned as a reminder that one of the two diagrams usually given to chess in game theory is that of a treelike structure (the other being that of a grid, which is undoubtedly why Obstruction was next to a Sol LeWitt lattice). But the strategic model was also used to experiment with the possible clash of independent themes: on one “side” works reflecting on vision (a perspectival church interior by Pieter Jansz Saenredam, a blurred Chair, 1965, by Gerhard Richter); on the other, works dealing with the notion of narcissistic reflection (Rubens’ Narcissus or Magritte’s Reproduction interdite, 1937). And, to mimic the huge memory bank of past moves that is activated by a chess master during the course of a game, the walls enclosing the vast hall were adorned with several clusters of works that functioned as footnotes to those featured on the “board” (drawings and prints but also paintings, e.g., Dubuffet’s Stairs in Commemoration of Jacques Ulmann, 1967, which Damisch related to the Bruegel; a Lichtenstein Mirror, 1970, with its obvious reference to narcissism; or an interior by Emanuel de Witte, Interior with a Lady Playing at the Virginal, with its typical checkered marble floor).

These footnotes conveyed the idea of possible substitutions (once again, not entirely freely made): How would the economy of the “board” be affected if a Fragonard drawing replaced Man Ray’s tree or if the Dubuffet rather than the Bruegel were the rook? The structural operation of permutation was the central topic of the second section of the show, consisting entirely of drawings and prints. The chess metaphor was abandoned for that of the card game, which more immediately elicits the idea of a reshuffling, and Damisch took pains to theatricalize the notion of permutation (thus, to have us acknowledge both the arbitrariness and limitation of choice). If one hangs a classical study of a horse next to a drawing of two camels, or a sequence of Jan van Hugtenburgh’s seventeenth-century prints of horses indecently sprawling or lying on their back, one produces a debasement. If one displays a Richard Serra paint-stick drawing next to a Hiroshi Sugimoto photograph of a luminously blank movie screen, one induces reflections about light and dark contrasts (or is it about the void?). A grotesque figure by Goya is enough to underline the generally unnoticed foreshortening distortions in Tiepolo’s sketches. Depending on what’s in storage, one can envision meaningful substitutes for each combination (once again the curator’s choice, though statistically wide, was not infinite). The guiding principle for this particular selection was neither aesthetic excellence nor subjective taste but the quest for the maximal power of transformation that one image can exercise over a neighboring one.

The final two sections of the show were, so to speak, “site specific.” The first, through a series of photographs and architectural drawings, reflected on the history of the Boijmans, including plans for renovation (the most striking image, taken during the bombing of Rotterdam in May 1940, showed the building in profile against the city in flames). The other, alluding to the iconoclastic crisis that stirred Holland at the time of the Reformation, showed two filmic sequences on large screens facing each other: the scene in Batman in which the Joker and his crew deface the entire painting collection of an art museum, except for a canvas by Francis Bacon; and the race through the whole Louvre, lasting precisely nine minutes, forty-three seconds, in Jean-Luc Godard’s Bande à part. While neither section was essential to the exhibition, the first served as a useful reminder that the game was not entirely abstract or theoretical but was situated in time and place, and the second underscored that, if such “experiments” must be iconoclastic (assuming their goal is to cast a blow against the exclusory structure of museums), they have to be conducted according to a strategic plan, without which they end up being mere grist for the mill of the culture industry and its irrepressible transformation of museums into amusement parks. Even the Joker realized, in front of the Bacon, that a random attack would not do.

Yve-Alain Bois is Joseph Pulitzer, Jr. Professor of Modern Art at Harvard University. He was cocurator, with Rosalind E. Krauss, of “L’Informe: Mode d’emploi,” at the Centre Pompidou, the catalogue to which was recently published in English as Formless: A User’s Guide (New York: Zone Books).