The Bourgeois and Distant Reading were published in tandem last year, a coupling planned by their author and suggesting that, for all their differences, they are to be seen as belonging to a common project.footnote1 Distant Reading is a collection of previously published essays interleaved with post-factum commentary, all arguing for an expansion of the field of literary study out from the confines of the canon and the practices of close reading. If their terminus is the computer and the database, these studies proceed by way of allopatric speciation (analogously applied to the formal inventiveness of European literature), world-systems concepts, cartography and network theory. The Bourgeois is the outcome of overlapping but longer-standing interests. It unfolds on basically three levels: a literary phenomenology where historical changes, social experiences and cultural values are held and staged in the very sinews of language and ‘style’; a narrative history running through and across the literary descriptions and centred on the fate of the ‘bourgeoisie’ from the early eighteenth century to the nineteenth-century fin de siècle; and a (non-systematic and sporadic) statement and illustration of Moretti’s method, glimpses of which are provided by the essays reprinted in Distant Reading. All three levels converge on a twilight zone, an in-between space, specifically ‘between history and literature’ (the subtitle of The Bourgeois), two monumental pillars seen as porous, permeable each to the other, as both realities on the ground and as the disciplines studying those realities. The nub of the project lies with that bridging preposition, ‘between’, and any assessment of how successfully passage across it has been negotiated turns crucially on the engineering methodology used for the bridge’s construction.

One of the principal columns supporting the bridge is the category of ‘prose’, as a mode of writing and a set of literary genres—centrally the novel, the genre most adapted to the disenchanted world of modernity—but also as a historical term of experience, attitude and outlook: Goethe’s ‘Age of Prose’ or Hegel’s ‘prose of the world’, designating a whole culture, the ‘prosaic’ bourgeois order of exchange, calculation and accumulation. These are of course the terms of a tradition of historical and theoretical criticism that comes to us via Hegel, down through Lukács, and to which Moretti’s earlier The Way of the World (1987) belongs. The idea since then, and now in this book, seems to be to expand the frames of reference and the terms of analysis in the light of new directions of travel opened up since The Way of the World. The bridge is thus further secured by means of another category umbilically tied to prose: ‘style’. Prose and style. Thus might run, in echo of Moretti’s own taste for syntactic brevity, the lapidary summary of the book’s focus. This comes out as the deployment by literary texts of mappable linguistic resources of a very basic lexical and grammatical kind—nouns, verbs, adjectives. What counts is what can be counted, quantitatively (relative frequencies) but also qualitatively: junctures of lexical concentrations and so forth. In short, and echoing the title of Raymond Williams’s book, prose and style forge their alliance in the crucible of ‘keywords’. As Moretti puts it: ‘by styles [is] meant mostly two things: prose and keywords’.

This, then, is the proposed access-route to ‘the peculiarities of bourgeois culture’, understood as a ‘“mentality”’—a nod here to the Annales school—‘made of unconscious grammatical patterns and semantic associations’, the inner life of a class caught in ‘the minutiae of language’, which ‘reveal secrets that great ideas often mask’. This is a project closer to cultural anthropology than to intellectual history; and, in its foregrounding of literature, also reminiscent of Williams’s category of the ‘structure of feeling’. Moreover the account traces not just an evolving thematics of representation, but also sites and moments of stress and fracture, ‘contradictions’ to which certain kinds of literature, like ideology or as ideology, supply imaginary solutions, while other kinds, more geared to exposure and demystification, disclose the impossibility of solutions within the framework of the ‘mentality’. The lexical point of departure chooses itself, namely the key keyword of the book’s title: ‘bourgeois’. Three pages are devoted to rehearsing the semantic histories of the collocationary terms—bourgeois, bourgeoisie, Bürger, Bürgertum, bürgerlich, middle class, etc. It’s a cursory history, but that is probably a blessing, although it might have noted that Marx and Engels, in the German original of The Communist Manifesto, use both die Bourgeoisie and die bürgerliche Gesellschaft. Nevertheless, Moretti wisely does not ask what the bourgeoisie is, or embark on the generally doomed attempt to specify the necessary and sufficient conditions of a definitional meaning. Let’s not set out to define it, he healthily proposes; accept that it exists or existed, and then try to describe it, but by means of a version of thick description, attentive to detail, nuance, multi-facetedness and tension. Accordingly he sets off on an extended hunting party through the thickets of numerous (mostly European) sources with a view to assembling a collection of descriptions imposed and superimposed on the term ‘bourgeois’ across a historical time scale, the baseline terms being a series of keywords: ‘adventure’, ‘enterprise’, ‘useful’, ‘efficiency’, ‘comfort’, ‘convenience’, ‘luxury’, ‘serious’, ‘earnest’. From Defoe to Ibsen, textual thickness is the distinctive feature of the book. It is also its strongest feature, producing as it proceeds a truly dazzling collage or kaleidoscope of literary analysis.

But just how robust is the bridge supporting traffic ‘between’ literature and history? What kinds of questions can the tandem publication of the two books help us to answer, or indeed to pose? In this connection, I raise three issues, a shorthand for which might run as: discovery; sampling; explanation. First, to what extent do keyword searches tell us something we didn’t know already? Obviously the computer dumps huge amounts of data into our laps, often mined from the vast under-explored continent of the Great Unread, which would not routinely figure on our reading radar. But in any deeper sense of knowledge, what discoveries come our way? In Distant Reading, there is a chapter recording how a database was used to generate some findings in respect of 7,000 titles of British fiction published between 1740 and 1850. One of the findings concerns the relative distribution across the span of nearly a century of the normally anodyne definite and indefinite articles in the titles of, respectively, the ‘anti-Jacobin’ historical novel and the later ‘new woman’ novel; it is not the least curious feature of Moretti’s way with history and literature that the basic building blocks of his bridge are parts of speech. In the first case the typical pattern is to have the definite (The Democrat, etc); in the second case it is more the indefinite (A Hard Woman). ‘The’ looks backwards to a completed action; ‘A’ looks forwards to an open future. This certainly alerts us to the significance of a linguistic feature usually unnoticed. That is interesting, but not very; at the level that matters—understanding the genres in question—it merely tells us what we already knew: historical novels are backwards-looking; that’s why we call them historical novels.

Secondly, there is the question of the extent to which the modus operandi is that of the self-confirming circle. This of course wanders into the age-old problem of the representative status of the sample and the traps of the hermeneutic circle, to which all of us are prey. With quantitative analysis the sample pool is naturally much larger, but it is not clear that this changes much at the most basic level of interpretation: the sample as synecdoche, as part evoking or standing for a whole. How different is it, methodologically speaking, from the procedure of the greatest literary historian to have based an entire strategy on a point of departure in close reading? The explicit basis of the Auerbachian method was the so-called Ansatzpunkt, the choice of a viewing position, at once a presupposition and a perspective, that guides the movement from part to whole. Similarly, as a prior selection then used to trawl a corpus, the chosen keyword furnishes an initial platform that inevitably influences both trajectories and outcomes. It is not quite a closed system; there are surprises, but they are few. There will be many more once the data retrieval programmes can handle ‘unstructured data’—that is, information that has not been tagged, the tag of course shaping the search. What new patterns and relationships between otherwise heteroclite domains might emerge, and what new questions might then be put, is a horizon of possibility still in the process of becoming visible, let alone reached.