There are many ways of talking about the theory of the novel, and mine will consist in posing three questions: Why are novels in prose; Why are they so often stories of adventures; and, Why was there a European, but not a Chinese rise of the novel in the course of the eighteenth century. Disparate as they may sound, the questions have a common source in the guiding idea of the collection The Novel: ‘to make the literary field longer, larger, and deeper’: historically longer, geographically larger, and morphologically deeper than those few classics of nineteenth-century Western European ‘realism’ that have dominated the recent theory of the novel (and my own work).footnote1 What the questions have in common, then, is that they all point to processes that loom large in the history of the novel, but not in its theory. Here, I will reflect on this discrepancy, and suggest a few possible alternatives.

Prose. Nowadays, so ubiquitous in novels that we tend to forget that it wasn’t inevitable: ancient novels were certainly in prose, but the Satyricon for instance has many long passages in verse; the Tale of Genji has even more (and crucially so, as hundreds of tanka poems stylize sadness and longing throughout the story); French medieval romances had a prodigious early peak in verse with Chrétien de Troyes; half of the old Arcadia is eclogues; Chinese classic novels use poetry in a variety of ways . . . Why did prose eventually prevail so thoroughly, then, and what did this mean for the form of the novel?

Let me begin from the opposite side, of verse. Verse, versus: there is a pattern that turns around and comes back: there is a symmetry, and symmetry always suggests permanence, that’s why monuments are symmetrical. But prose is not symmetrical, and this immediately creates a sense of im-permanence and irreversibility: prose, pro-vorsa: forward-looking (or front-facing, as in the Roman Dea Provorsa, goddess of easy childbirth): the text has an orientation, it leans forwards, its meaning ‘depends on what lies ahead (the end of a sentence; the next event in the plot)’, as Michal Ginsburg and Lorri Nandrea have put it.footnote2 ‘The knight was defending himself so bravely that his assailers could not prevail’; ‘Let’s withdraw a little, so that they will not recognize me’; ‘I don’t know that knight, but he is so brave that I would gladly give him my love’. I found these passages in a half page of the prose Lancelot, easily, because consecutive and final constructions—where meaning depends so much on what lies ahead that a sentence literally falls into the following one—these forward-looking arrangements are everywhere in prose, and allow it its typical acceleration of narrative rhythm. And it’s not that verse ignores the consecutive nexus while prose is nothing but that, of course; these are just their ‘lines of least resistance’, to use Jakobson’s metaphor; it is not a matter of essence, but of relative frequency—but style is always a matter of relative frequency, and consecutiveness is a good starting point for a stylistics of prose.

But there is a second possible starting point, which leads, not towards narrativity, but towards complexity. It’s a point often made by studies of dérimage, the thirteenth-century prosification of courtly romances which was one of the great moments of decision, so to speak, between verse and prose, and where one thing that kept happening, in the transfer from one into the other, was that the number of subordinate clauses—increased.footnote3 Which makes sense, a line of verse can to a certain extent stand alone, and so it encourages independent clauses; prose is continuous, it’s more of a construction, I don’t think it’s an accident that the myth of ‘inspiration’ is so seldom evoked for prose: inspiration is too instantaneous to make sense there, too much like a gift; and prose is not a gift; it’s work: ‘productivity of the spirit’, Lukács called it in the Theory of the Novel, and it’s the right expression: hypotaxis is not only laborious—it requires foresight, memory, adequation of means to ends—but truly productive: the outcome is more than the sum of its parts, because subordination establishes a hierarchy among clauses, meaning becomes articulated, aspects emerge that didn’t exist before . . . That’s how complexity comes into being.

The acceleration of narrativity; the construction of complexity. Both real: and totally at odds with each other. What did prose mean for the novel . . . it allowed it to play on two completely different tables—popular and cultivated—making it a uniquely adaptable and successful form. But, also, an extremely polarized form. The theory of the novel should have greater morphological depth, I said earlier, but depth is an understatement: what we have here are stylistic extremes that in the course of two thousand years not only drift further and further away from each other, but turn against each other: the style of complexity, with its hypothetical, concessive, and conditional clauses, making forward-looking narrative seem hopelessly simple-minded and plebeian; and popular forms, for their part, mutilating complexity wherever they find it—word, sentence, paragraph, dialogue, everywhere.