On the limits of openness I: the digital humanities and the computational turn to data-driven scholarship

The digital humanities can be broadly understood as embracing all those scholarly activities in the humanities that involve writing about digital media and technology, and being engaged in processes of digital media production, practice and analysis. For example, developing new media theory, creating interactive electronic archives and literature, building online databases and wikis, producing virtual art galleries and museums, or exploring how various technologies reshape teaching and research.  Yet this field - or, better, constellation of fields - is neither unified nor self-identical. If anything, the digital humanities are comprised of a wide range of often conflicting attitudes, approaches and practices that are being negotiated and employed in a variety of different contexts.

In what follows my interest is not so much with the ongoing debate as to how precisely the digital humanities are to be defined and understood, but with an aspect of this emergent movement that appears to becoming increasingly dominant. So much so that for some it is rapidly coming to stand in for, or be equated with, the digital humanities as a whole.  This is the so-called ‘computational turn’ in the humanities. 

The latter phrase has been adopted to refer to the process whereby techniques and methodologies drawn from computer science and related fields, including interactive information visualisation, statistical data analysis, science visualization, image processing, network analysis, and the management, manipulation and mining of data, are being increasingly used to produce new ways of approaching and understanding texts in the humanities.  Indeed, thanks to increases in computer processing power and its affordability over the last few years, together with the sheer amount of cultural material that is now available in digital form, number-crunching software is being applied to millions of humanities texts in this way.

Before going any further I want to make it clear that it is not my intention to equate this computational turn with the digital humanities per se. Even if the latter is sometimes known as Humanities Computing - or as a transition between the so-called ‘traditional humanities’ and Humanities Computing  - what is coming to be called the digital humanities and this computational turn in the humanities are not one and the same thing as far as I am concerned.

In fact, far from equating the digital humanities with the computational turn, I want to insist on the importance of maintaining a difference between them, certainly for any understanding of what the humanities can become in an era of digital media technology. For, to date (and I acknowledge it is still relatively early days), the traffic in this computational turn has been rather one-way. As the phrase suggests, it has primarily been about exploring what direct practical uses computer science can be to the humanities in terms of performing computations on sets and flows of data that are often so large that,  in the words of the Digging Into Data Challenge, ‘they can be processed only using computing resources and computational methods’.   In the main the concern has been with either digitizing ‘born analog’ humanities texts and artifacts, or gathering together ‘born digital’ humanities texts and artifacts – videos, websites, games, photography, sound recordings, 3D data - and then taking complex and often extremely large-scale data analysis techniques from computing science and related fields and applying them to these humanities texts and artifacts.  So we have the likes of Dan Cohen and Fred Gibb’s text mining of ‘the 1,681,161 books that were published in English in the UK in the long nineteenth century’ (according to Google at least);  Lev Manovich and the Software Studies Initiative’s use of ‘software to analyze and visualize... 4535 Time magazine covers... 1074790 manga pages, and 1100+ 20th century feature films’; or Stefanie Posavec’s Literary Organism which visualizes the structure of Part One of On the Road as a tree.   

Yet just as interesting as what computer science has to offer the humanities, I believe, is the question of what the humanities - in both their digital and ‘traditional’ guises (assuming they can be distingished in this way, which is by no means certain) - have to offer computer science; and, beyond that, what the humanities themselves can bring to the understanding of computing and the shaping of the digital. Do the humanities really need to draw quite so heavily on computer science to develop a sense of what they can be in an era of digital media technology?  Along with a computational turn in the humanities, might we not also benefit from a humanities turn in our understanding of the computational and the digital?

To be sure, one of the interesting things about computer science is that, as Mark Poster pointed out some time ago, it was the first case where ‘a scientific field was established that focuses on a machine’ rather than on an aspect of nature or culture. Yet more interesting still is the way Poster was able to demonstrate that the relation to this machine in computer science is actually one of misrecognition, with the computer occupying ‘the position of the imaginary’ and being ‘inscribed with transcendent status’.  This misidentification on the part of computer science has significant implications for our response to the computational turn. It suggests computer science is not all that well equipped to understand itself and its own founding object, let alone help the humanities’ with their relation to computing and the digital.

In fact, counter-intuitive though it may seem, if what we are seeking is an appreciation of what the humanities can become in an era of digital media technology and data-driven scholarship,  we would be better advised looking elsewhere for assistance, other than primarily with computing science and engineering, science and technology, or even science in general. I almost hesitate to say it in the present political climate - although it is important to do so for precisely this reason - but we would be better off turning to the writers, poets, historians, literary critics, theorists and philosophers of the humanities right from the start.

Comments

[Paul]

Have always thought it unfair that humanists are self-training as digitalists, while computer geeks feel no compulsion to self-train as humanists. The "hybrid scholar" (in Julia Flanders' phrase) should come from more fields. Which, to follow your point, is to recommend those fields to the humanities.

But that might be wishful thinking, as your last paragraph implies. Another approach would be to all just read more Alan Liu and Matthew Kirschenbaum, scholars with exemplary attention to the transcendental and material problematics of computing and new media.

[Gary Hall]

Thanks for the comment. In subsequent parts of 'On the Limits of Openness' I want to explore some of the possible social and political reasons why this movement is so much in one direction. In particular, I want to ask whether the take up of practical techniques and approaches from computing science isn't providing some areas of the humanities with a means of defending themselves in an era of global economic crisis and severe cuts to higher education.