But the little Swiss girl of Giacometti's later recollection (and Leiris's account) has nothing to do with this key example of the marvelous and objective chance. By serving as a direct, real-world model for a work of art, the little Swiss girl withdraws Invisible 0bject from the orbit of surrealism and places it in the postwar realm of Giacometti's studio, as he notoriously strained, month after month, through trial and retrial, to catch the likeness of the model posed in front of him.4 Recontextualizing the work, setting it in relation to a new group of friends and allies, like Sartre and Genet, Leiris's account draws it closer to the problematic of The Phenomenology of Perception and further from that of Les vases communicants.5
This a-chronicity is, of course, unacceptable to the historian, and thus Reinhold Hohl, the leading scholar of Giacometti's work, does not even mention the memory of the Swiss child in discussing this masterpiece of the sculptor's prewar career. But then Breton's story is, for Hohl, equally suspect. "Contrary to Breton's account," he begins, "that a mysterious object found at the flea market (it was, in fact, the prototype for an iron protection mask designed by the French Medical Corps in the First World War) had helped the artist to find his forms, Giacometti had borrowed the stylized human shapes from a Solomon Islands Seated Statue of a Deceased Woman which he had seen at the Ethnological Museum in Basel, and had combined them with other elements of Oceanic art, such as the bird-like demon of death."6
Despite the certainty of his tone, Hohl's evidence for this connection is both scant and indirect. In 1963 Giacometti had spoken to an interviewer of a reconstructed Oceanic house installed in the Basel Museum.7 Since the Solomon Islands figure had been displayed in the same gallery early in the 1930s, when it was brought back to Switzerland from the expedition that had plucked it from the South Seas, Hohl could at least assume Giacometti's knowledge of the object.8 The detail that lends the greatest credence to Hohl's claim is the schematic, railinglike support for the half-seated figure, a construction that is entirely characteristic of this type of statue and is not commonly found elsewhere.9 Since part of the power of the pose of Giacometti's sculpture comes from the enigmatic relation between the half-kneeling posture and the structural elements that seem to contain it—a flat plate against the shins in front of the figure and the peculiar scaffolding behind it—and since this construction is not "natural" to a model posed in a studio, the probability was always that its source was in another work of art. Because of the railing, because of the posture, because of the forward jut of the head and the articulation of the breasts, the Solomon Islands statue of Hohl's nomination seems a logical candidate.10
Behind Hohl's assertion of this statue as the source for Invisible Object there is a whole reservoir of knowledge about the role of primitive art in the sculptor's work in the years leading up to 1934. Primitivism had been central to Giacometti's success in freeing himself not only from the classical sculptural tradition but also from the cubist constructions that had appeared in the early 1920s as the only logical alternative. Quite precisely, Giacometti's work matured as a function of its ability to invent in very close relation to primitive sources. Just two years after leaving Bourdelle's studio he was able to execute a figure on a major scale that was "his own" by virtue of belonging, quite profoundly, to African tribal art.

"A spoon is like a woman" becomes "A woman is like a spoon"...