The exact status of Zola's has always been problematic. For Pissarro, it was livre romantique, for Edmond de Goncourt, in a no less double-edged judgment, bonne construction du roman vieux jeu. (1) Other early readers moved uncertainly between professed admiration for its emotional power and reservations about ideas expressed by l'ecrivainphilosophe in footsteps of Diderot and Balzac's Le Chef-d'oeuvre inconnu. Its critical reception, when it was first published, intuitively highlights generic admixture that a modern scholar like David Baguley has analyzed to such revealing effect. Yet Patrick Brady's compendiously titled study of 1968 accommodates its plurality in an exclusively recuperative direction that has seldom been properly challenged L'oeuvre d'Emile Zola- roman sur les arts; manifeste, autobiographie, roman a (2) And, over four decades later, even most authoritative of summations reverts to such terms of reference: S'agit-il d'un roman a clefs? C'est la lecture qui s'impose d'emblee, confirmed by fact that aujourd'hui, le debat critique continue, pour l'essentiel, de tourner autour des cles du roman. (3) Part of reason for this ubiquity of emphasis has been, of course, L'oeuvre's privileged position within attempts by art historians to cross disciplinary boundaries while, paradoxically (or unsurprisingly), losing from sight its fictional specificity. Exemplary in this respect is Kermit Champa's question-begging conclusion that Zola's novel of 1886 is, as he puts it, the most unreliable art-historical text of period (71). Wayne Anderson's apparently more judicious aside that character of Claude Lantier has been recklessly interpreted by many art historians as a portrait of (4) (82) is equally indebted to extraliterary, hors texte of real-life painter's friendship with Zola himself, brought to an end after forty years in famously ambiguous response Cezanne sent to him on receipt of a copy of L'oeuvre; (5) ensuing personal silence has provided not merely biographers with justification for treating novel as a roman a clef. The aim of this article, by contrast, is to alert its readers to properly imaginative processes of its composition. For by exploring from a very different perspective precisely those dimensions of foregrounded by Brady, we can both view this particular novel with renewed clarity and suggest wider implications of out rereading of Zola at some distance from habitual approaches to his aesthetic. Emblematic in this regard are novel's interpolated descriptions of Claude's grand tableau, some five meters by three and titled Plein air, vicissitudes of which play such an important narrative and symbolic role in text. Doubtless further encouraged, beyond approximate analogies of subject-matter and internal geometry, by hypothetical objection (voiced by character of Dubuche) to a depiction of ce monsieur, tout habille, la, au milieu de ces femmes nues (48), which had also scandalized Manet's contemporary critics, modern commentators have too often unhesitatingly identified fictional painting as a transposition of Le Dejeuner sur l'herbe. (6) It is, indeed, instantly recognizable as such, according to a respected Manet specialist. (7) But to look at description of Plein air much more closely is to be struck instead by significant differences between Claude's inventions and Manet's considerably smaller (208 x 264 cm) picture of 1863: Dans un trou de foret, aux murs epais de verdure, tombait une ondee de soleil; seule, a gauche, une allee sombre s'enfoncait, avec une tache de lumiere, tres loin. La, sur l'herbe, au milieu des vegetations de juin, une femme nue etait couchee, un bras sous la tete, enflant la gorge; et elle souriait, sans regard, les paupieres closes, dans la pluie d'or qui la baignait. Au fond, deux autres petites femmes, une brune, une blonde, egalement nues, luttaient en riant, detachaient, parmi les verts des feuilles, deux adorables notes de chair. …