LITERARY STUDIES IN THE DIGITAL AGE

[T]here is no method, however well adapted to a given science, that literary history can transplant and apply to its own researches. The illusion that this is possible is responsible for much poor and childish work: statistics and charts, evolution of species, and quantitative analysis are processes, methods, and hypotheses excellent in their place, but their place is not literary history.

In the last decade, and especially in the last four or five years, the insistence in this epigraph – that quantitative methods have no place in literary history – has been repeated many times. The fact that this particular passage comes from a book first published in 1922, and intended as a guide for graduate students, should demonstrate that both the application of such methods, and the resistance to them, are of considerably longer standing in debates about literary history than is generally acknowledged. Nonetheless, discussion of quantitative methods has almost certainly never been as heated or as widespread – or as apparent to the majority of literary scholars – as it is today. While there are a number of quantitative approaches to literature, the current debate focuses on Franco Moretti's work in literary history. As Priya Joshi says, literary scholars have for a long time 'regarded quantitative analysis with suspicion bordering on contempt'. But in the response to the publication in 2000 of Moretti's 'Conjectures on World Literature', and in 2005 of his book Graphs, Maps, Trees, this contempt has escalated – especially in the American humanities – to an intense stand-off.

The controversy surrounding Moretti's work is, to a significant extent, specific to it. But this debate also presents important criticisms of quantitative methods that need to be engaged with if such studies are to make a productive contribution to literary history and humanities scholarship generally. This chapter considers three closely related criticisms that have been levelled at quantitative literary research (predominantly at Moretti's 'experiments' in literary history): first, that such approaches reduce the inherent complexity and multiplicity of literature and language to uniform data; second, that quantitative methods make false claims to authoritative and objective knowledge; and finally, that such studies resonate, in problematic and complicit ways, with contemporary institutional discourses, especially neoliberal or economic rationalist managerial practices.

I am not proposing that such criticisms are never applicable to quantitative approaches; like all research practices, these can be applied in varying ways. Nor is this chapter a defence of Moretti's scholarship. Although his centrality to the debate makes an engagement with his arguments and methods unavoidable – and while I find his work well worth the engagement – some aspects of Moretti's research justify some of the criticisms that have been made. However, I will argue that reductionism, absolutism and acquiescence to neoliberalism are not intrinsic to quantitative methods. In this chapter I discuss work in book history and the digital humanities that I have found useful in developing my approach to literary historical data. Specifically, I argue that an approach based on book history's methodological pragmatism regarding the nature and use of data, and the digital humanities' method of modelling, offers a productive way of integrating empirical data with the paradigm of humanities knowledge as a critical, analytic and speculative process of inquiry. This approach maintains what Donna Haraway calls' a no-nonsense commitment to a faithful account of the "real" world', while preserving, in George Levine's words, 'a tentativeness that keeps all aspirations to knowledge from becoming aspirations to power as well'.

I Quantitative Method and its Critics

As the criticisms of quantitative approaches to literature are largely directed at Moretti's work, I will begin with a brief summary of his arguments: both against conventional approaches to literary history and for quantitative methods. For decades, Moretti has argued that a literary history based only on the texts that make up the canon offers no insight into the vast 'mass' of literature, and no basis for understanding the causes and processes of literary change. In 1983 he wrote that:

[A]t present, our knowledge of literary history closely resembles the maps of Africa of a century and a half ago: the coastal strips are familiar but an entire continent is unknown. Dazzled by the great estuaries of mythical rivers, when it comes to pinpointing the source we still trust too often to bizarre hypotheses or even to legends.

More recently, Moretti has refined this critique into a specific challenge to the reliance of literary history on detailed textual analysis or 'close reading' as the source of historical evidence. He identifies 'close reading' – where the 'representative individual' defines the 'whole', or the 'one per cent of the canon' signifies 'the lost 99 per cent of the archive' – as a form of 'topographical thinking'. The main problem with this approach, and the source of what Moretti considers as irrationality, lies in the fact that the 'rare and ... exceptional' works of the canon are by definition not representative. In taking the canon as its object, literary history fails to consider the 'banal, everyday, normal' operations of the literary field and the wider context in which literary change occurs.

For Moretti, the means of overcoming this unrepresentative focus cannot be more reading. The size of the archive renders this potential solution impossible to achieve: even 'a novel a day every day of the year would take a century or so' to cover nineteenth-century British fiction. As well as a matter of scale, close reading gives no insight into the workings of the literary system:

[A] field this large cannot be understood by stitching together separate bits of knowledge about individual cases, because it isn't a sum of individual cases: it's a collective system, that should be grasped as such, as a whole.

To this end, Moretti offers a paradigm of 'distant reading' that deliberately abstracts both the material and textual features of literary works to provide new accounts of literary history based on 'a specific form of knowledge: fewer elements, hence a sharper sense of their overall interconnection. Shapes, relations, structures. Forms. Models.' Moretti's work provides an important statement of the contribution quantitative methods can make to literary history: namely, their potential to represent historical trends and, in so doing, enable a form of analysis...