TY - JOUR AU - Eccles, Kathryn AB - Abstract This chapter examines material published in the field of digital humanities in 2019. Key work published this year has grappled with longstanding conflicts at the heart of the field, on whether and how computational methods should be applied to humanities data, and who should validate such methodologies. The chapter begins with new work by Ted Underwood, who makes the case for hypothesis-driven methods and the modelling of humanities data. It discusses how recent work in computational literary studies had appeared to resist the trap into which much previous work had fallen, that is, work that was perceived to fall into the binaries of distant vs. close reading, computation vs. engagement, objectivity vs. subjectivity. The continued friction over the appropriateness of certain computational methodological approaches was amplified by new work that called into question the statistical methods of a number of key works in the field over past years. Nan Z. Da’s critique of computational literary studies through the lens of statistical rigour imploded the uneasy truce between computational methods and the more traditional questions and methods at the heart of literary studies. Da’s article reopens the debate about how digital humanities scholars use statistical methods, and how greater reliance on such methods may demand greater cross-disciplinary oversight to ensure that they are used in a way that is both robust and appropriate. Her contribution is examined alongside the rash of responses to it from key scholars in the field which produced an important snapshot of the fractures and fundamentals of data-driven literary studies. I then turn to new and timely work by James E. Dobson, which argues for a third way, a Critical Digital Humanities that engages critically with computational as well as humanistic scholarship. I survey important contributions on the impact of mass digitization, historicism and the archive, and how to study history in the age of digital archives and the historic web. Ian Milligan’s work provides a much-needed introduction to the potentials and pitfalls of studying recent history through the digital traces left behind. It self-consciously identifies areas in which greater cross-disciplinary scholarship and critical engagement will be needed as this area of study matures. Discussion then turns to work by Nanna Bonde Thylstrup on digital waste, which shows how connecting new media theory to waste studies can provide an important frame through which to examine issues of data toxicity and pollution. This work sets the stage for two landmark books on sex and race which implore us to take a more careful look at the toxic technologies we build and the questions we ask of them. Both Caroline Criado Perez and Ruha Benjamin examine the damage done by the reliance of data systems on the ‘default’, frequently a white male, forcing us to see anything that departs from this norm as deviant. These works make a powerful case for reinventing the systems we increasingly rely on, questioning the underlying prejudice that created them, and rethinking the modes of meaning-making ascribed to them, especially when that narrative so often assumes a benign neutrality. Finally, I examine these works alongside a new volume of essays on digital humanities and intersectionality edited by Barbara Bordalejo and Roopika Risam, which serves to amplify and contextualize the need for the approaches taken by Criado Perez and Benjamin, showing how deeply enmeshed within the field these power structures are. 1. The New Computational Literary Studies In early 2019, the field of digital humanities witnessed the rekindling of a long-standing conflict surrounding the application of quantitative methods to humanities data. The publication of a number of recent books and articles on computational literary scholarship had made a strong case for the resolution of such arguments. Beginning in late 2018 with Andrew Piper’s Enumerations: Data and Literary Study, and continuing into 2019 with Ted Underwood’s Distant Horizons: Digital Evidence and Literary Change, computational literary studies appeared to have undertaken a shift in which the ‘loaded language of objective computation versus subjective reading’ had been jettisoned in favour of a ‘pluralistic and emancipatory discourse’ (Mandal, ‘Digital Humanities’). Reviewed in last year’s YWCCT essay, Piper’s Enumerations was argued to have been at pains ‘to avoid the binaristic’ fallout that followed the publication of work by Franco Moretti and Matthew Jockers which had pitted ‘“objective” distant reading against “subjective” close reading’. Underwood’s Distant Horizons begins by questioning the notion that one critical practice has displaced another, including ‘distant reading’ challenging more traditional ‘close reading’, suggesting that something more interesting has emerged from this debate: the ‘uncovering [of] new objects of knowledge’ (p. 3). These new objects are patterns emerging from Underwood’s computational analysis. It is the scale of the discoveries, Underwood argues, that require new modes of critical analysis. Piper’s work drew attention to the concept of modelling as a means of moving away from criticism focused on the size of datasets and their relative power. In Distant Horizons, Underwood reinforces the case for modelling, stating unequivocally that ‘scholars need numbers. In particular, we need statistical models’ (p. 1), and showing throughout the subsequent chapters how statistical models may be employed. Underwood presses home the need for what he describes as perspectival modelling, a form of exploratory and experimental modelling that begins with ‘an interpretive hypothesis (a “meaning” to investigate)’. Mindful of the likely resistance to the application of statistical models in the humanities, and the clash of academic cultures implicit in this approach, Underwood reassures us that such models are ‘good at blurriness’ (p. 20) and argues persuasively for the need to set out a research question, informed by deep understanding of the works in question, and intelligent application of models that can use these works as the building blocks for further understanding. This is a significant pivot for scholars in the field of cultural analytics. In Distant Horizons, Underwood challenges us to apply this mode of analysis in order to free us from the ‘narrative of conflict and displacement [as] a rhetorical crutch’ and to give scholars a means of ‘connecting the dots’: this is an approach to longue durée literary history that does not flit between micro- and macroscopic approaches that employ different methods but, rather, employs multiple levels of ‘zoom’ to enable us to perceive patterns in more and more detail as we adjust and fine-tune each level of focus. Perhaps the most interesting of Underwood’s arguments come not in the first four excellent chapters, in which he demonstrates the versatility of his methods, but in the final chapter, where he turns from the question of whether we could to whether we should employ such methods in literary studies, taking on one of the most persistent arguments in digital humanities: the notion that quantitative methods, with their authority and implied certainty, are poised to displace other, more nuanced, critical methods. In this argument, distant reading is consistently positioned in opposition to close reading. Underwood is keen to test this argument and dispel the often alarmist ‘dark vision of the future’ (p. 146). One of the great achievements of Distant Horizons is to stress the potential connections to be made between different scales and types of analysis: ‘subjective evidence about genre, gender, and literary quality’ is central to his arguments and to his mode of interrogation, enabling him to resist ‘the notion that a firm boundary can be drawn between measurement and interpretation’ (p. 147). These arguments somewhat inevitably bring up further persistent tropes about the consequences of moving towards more social-scientific methodologies (reductionism, loss of pleasure in the endeavour) and the risks of not adopting these tools (myopia relating to the local, individual, and distinctive, blurring out the longer-term trends), reminiscent of the ‘new criticism’ and ‘two cultures’ debates of the mid-twentieth century. In order to ‘frame a fair critique of distant reading’, Underwood argues, perhaps this ‘simply moves further toward social science than most literary scholars are willing to go’ (p. 161). 2. Nan Da and the Computational Case against Pluralism In an article for Critical Inquiry entitled ‘The Computational Case against Computational Literary Studies’, this is arguably where Nan Da takes us, critically examining the technical and conceptual basis on which much of the quantitative work in computational literary studies is based. In this article, she argues that ‘the problem with computational literary analysis as it stands is that what is robust is obvious (in the empirical sense), and what is not obvious is not robust’ (p. 601). Though it is emphatically stated that the field of digital humanities itself is not the subject of her critique, her argument in this piece created a challenge that threatened to destabilize the truce of plurality sitting at the heart of much digital humanities work. Da’s contribution to the debate is to suggest that we steer clear of using ‘ideological reasoning’ to argue against computational literary studies (CLS), but to accept insteadthat it has ‘very little explanatory power that is not negated by its operations’ (p. 604). Her task, then, is to examine a series of studies which she argues have ‘conceptual fallacies’ but which she takes ‘on their own terms completely’, from sampling to testing, code to claims. Da probes the statistical basis of a series of CLS papers, which she divides into two camps, those that she claims do not produce statistically significant findings, and those that do produce results but which she claims are wrong. The work that does produce statistically rigorous findings, Da suggests, does not fully answer the questions that it poses, and the work that does appear to arrive at interesting answers to the questions posed, does so through statistical methods that are either non-robust or confused. Da’s article was a radical intervention, an attempt to amplify to the humanities the importance of the practice of ‘replication’, in which research is repeated in order to verify whether findings can be reproduced, well established as a scientific practice (though felt in many areas to be in crisis). By focusing on ‘what it is about the nature of the data and the statistical tools’ that leads to faults in both the conceptual basis of the works and the claims which rest on them, the temptation to hand-wring over whether we should use certain methodologies and approaches is replaced by the question of whether we are capable of using these tools to make the claims we want to arrive at. The article is not intended to be an exposé of the errors and oversights in work in this field, Da claims, as such oversights are part and parcel of data-mining work across the disciplines. Rather, Da points us to ‘the circumstances under which such errors would be permissible and which not’ (p. 605). Da’s article had an immediate impact. Recognizing the flurry of commentary and opinion pieces generated in response to it, and seeking to provide a platform on which such commentary could be preserved, the editor of the journal, Patrick Jagoda, created a shared space to foster ‘a generative discussion’ (Computational Literary Studies: A Critical Inquiry Online Forum). This discussion featured scholars whose work had been examined and criticized in Da’s article, invited to respond by the editor, but such was the furore over Da’s initial article, that the responses kept coming. As a result, each scholar’s contribution is noted as ‘Day 1’, ‘Day 2’, and ‘Day 3’, a detail that gives a strong sense of the clamour with which Da’s article was met. Responses were gathered from Mark Algee-Hewitt, who critiqued the binaries permeating Da’s article and stressed the need for the plural approach of computation hand in hand with literary criticism, an ‘augmented’ humanities where these approaches would work in tandem. Katherine Bode, in her first response (‘Day 1’) lent support to Da’s concerns about approach and agreed that statistical tests of statistical arguments are indeed crucial; she suggested that she had excluded from her article the work of other computational literary scholars who were interested in more diverse forms, and critiqued Da’s ‘constrained and contradictory framing of statistical enquiry’. Ed Finn pivoted the discussion towards rigour, arguing that Da’s plea for statistical accuracy and precision sets up computational literary studies as a field for standards of peer review across disciplinary boundaries that are unrealistic and which other disciplines would find hard to countenance. Lauren Klein, in her second response (‘Day 3’) picks up Katherine Bode’s point about exclusion, and argues for a new rigour to be led by diversity work, recognizing the structural deficiencies that Marisa Parham’s work has so richly examined and explored (Parham, ‘Ninety-Nine Problems’). Bode also returned to the forum on ‘Day 3’ to critique the responses. Having hoped for an exchange that would ‘build connections between literary and statistical ways of knowing’, she found that ‘we’re just replaying the same old arguments’ by rehashing the notion that ‘quantification and computation can only yield superficial or tautological arguments’ but in statistical terms. The final comments were left to Da herself, who argued that the fundamental premise of her article, that those who are capable of critiquing the quantitative methods underlying the field of computational literary studies ‘tend not to question how quantitative methods intersect with the distinctiveness of literary criticism’, remained steadfast. 3. A New Critical Digital Humanities? The debate surrounding the validity, relevance, and desirability of computational literary analysis summarized in Da’s article and re-enacted on the Critical Inquiry forum shows that the (broader) field of digital humanities remains highly contested and contentious. James E. Dobson’s Critical Digital Humanities: The Search for a Methodology enters the debate at a crucial moment, anticipating the key points emerging from the Da controversy, and advocating a third way, ‘equal parts critical theory and computational science’ (p. 31). Dobson urges scholars to practise critical digital humanities as well as criticizing digital humanities methods. The potential for creating a new digital humanities is released, he argues, through applying humanistic critical skills to computation. It is in this critical sphere that humanities scholars can move the discipline forward. His work builds on that of N. Katherine Hayles (How We Think) and David Berry (‘The Computational Turn’), who have previously argued that ‘computationality’ might form the proper subject of the digital humanities. Berry has signalled that this new approach could be seen as a third wave of digital humanities. Dobson suggests that we need to put in place an analytical framework in order to critique computation and computational methods, to use and reflect on computational tools within a humanities context. Critique, Dobson reflects, ‘is a tool for examining an object or concept and for locating what remains unthought and undertheorized in the imaginative and material construction, the explication and framing, and the use of this object, tool, or concept’ (p. 4). Dobson’s work anticipates and reinforces some of Da’s critiques in her polemical article. Where Da calls for more scrutiny of computational methods, and in particular scrutiny of the claims attached to the outputs of these methods, Dobson reinforces the need for mastery of these tools, suggesting that failure to do so leads to humanists withdrawing into positivism. He is also sanguine about the effort that comprehension of computational tools will require from humanists, particularly as we move towards more algorithmic and machine-learning approaches: ‘some algorithms, especially those using machine learning approaches, remain opaque even to their authors and developers, yet such understanding is essential, whether we choose to use these tools in our own work or not’ (p. 4). This is increasingly important in view of the work emerging from Safiya Noble, Kate Crawford, Caroline Criado Perez, and Ruha Benjamin that reveals systematic bias and data gaps and their impact on data systems and algorithm-based tools. Hayles has previously reflected that a critical digital humanities could be mobilized around a set of contested concepts and methodologies. Dobson’s work follows this same line, analysing and theorizing the digital humanities as a baggy field: ‘If, in the process of studying some particular phenomena, a digital humanist makes use of the exact same methods and data as a computational sociologist, what distinguishes the interpretive procedures of the digital humanities scholar from those of the social scientist? Do disciplinary boundaries matter to digital methodologies?’ (p. 10). It is clear from the reception of Da’s work in Critical Inquiry that discipline does matter to many scholars in the field, as it provides a way of integrating new methods while remaining true to the core disciplinary tenets. Da herself stresses the need for cross-disciplinary methodological scrutiny, and ensured that her work on computational literary studies was read and critiqued by an ‘out-of-field’ peer reviewer (‘Day 3’, Computational Literary Studies). Dobson finds this cross-disciplinarity both desirable and invigorating: ‘in an era dominated by social scientific and computational methods, the debates concerning methodology within the humanities demonstrate the vitality of the discipline and a very much alive and active cross-field dialogue’ (p. 11). He stresses the increasing value of ‘workflows’, a key motif through the book which he terms ‘a descriptive and discursive object’ (p. 8), drawing a persuasive line between these objects and the work of early twentieth-century critic I. A. Richards in formalizing his students’ methods of, or ‘protocols’ for, reading and critiquing poetry in Practical Criticism (1929). Dobson contests the notion that the transformation in methods and tools witnessed in the humanities, in particular through the developments in algorithms and machine learning and large-scale digital archives that call for data-driven analysis, should require a ‘shift in thinking that exceeds the other transformations’, arguing that existing critical and theoretical approaches found within the humanities have something valuable to say about these methods (p. 4). Dobson’s plea for a critical digital humanities rests on the ability of digital humanities scholars to embrace both sets of tools, foregrounding 'cultural and epistemological questions about computationalism’ (p. 6). In order to take this approach, he argues, critical digital humanities scholars will need to examine everything, asking a series of searching questions about how data or digital objects (whether born digital or converted) came about, what transformations enabled them, and what influences those processes, and the original materials underlying those processes, may have been subject to. 4. Abundance and Scale The need for a critical digital humanities that engages with the computational and computationalism is increasingly urgent, given the number of key works on web histories, web-based data archives, and data abundance in recent years. These works take us from the ‘why’ to the ‘how’ and focus largely on the methodological, but they also stress the need for humanities scholars to engage critically with the mass. Following the publication in 2018 of Niels Brügger’s The Archived Web: Doing History in the Digital Age and The SAGE Handbook of Web History edited by Brügger and Ian Milligan, Milligan’s own History in the Age of Abundance encapsulates many of the key themes, demonstrating the challenges of abundance and the theoretical and methodological shifts that historians need to undertake in order to meet them. Milligan’s text outlines the paths that future historians will increasingly need to take, through web archives, in order to access the records left by human endeavour online. These digital traces, all too easily forgotten, have largely been salvaged by the foresight of web archiving initiatives like the Internet Archive, but they are complex and fragmented and require historians wishing to use them to acquire a range of complex computational methods. Milligan’s work is an important introductory text for historians wishing to embrace web archives for historical research, richly describing the challenges of seeking answers from a collection this large (see e.g. Chapter 4, ‘Unexpected Needles in Big Haystacks’). Milligan is also excellent when thinking through the ways in which we conceptualize web and data archives, to which we frequently ascribe the characteristics of a formal archive which are largely not present. We are still grappling with these terms and values, and Milligan reminds us that we must be diligent in ‘teasing out definitional continuities and changes’ (p. 71). While historians may find it useful to use the term ‘archive’ as a ‘familiarizing point of reference’, Milligan is right to remind us that ‘Web archives are not traditional archives—not in content, form or conception’. Detailed chapters on specific archives, especially that of the now defunct GeoCities, provide the opportunity to work through some of the ethical considerations wrapped up in this type of work. As with discussions of computational methods, new approaches to modelling humanities data, and the new critical digital humanities, the ethical challenges of working with web archives require scholars to work across disciplinary boundaries in order to establish robust new modes of enquiry. Milligan draws on the work of Internet researchers and the AoIR report on ethical guidelines (Ess and AoIR Ethics Working Committee, 'Ethical Decision-Making and Internet Research')), and looks at social scientists’ work on early adopters and notions of extended private space to draw conclusions about what historians should consider ethically sound when looking at large web and data archives, but it is clear that there is still a great deal of work to be done, and perhaps we should take a nod from Dobson about how to do this collaboratively, constructively, critically, and with a deeper understanding of the tools we are using. Summing up, Milligan points to these necessary changes, where historians will need to engage more actively with the collaborative construction of computational tools and more critically with the ethics of using web archives as a historical reference. 5. Theorizing Abundance: By-Products and Digital Waste One of the key images conveyed throughout Milligan’s work is the notion of the data deluge, and the title of his book gives rise to a positivist reading of the abundance of data. This is easy to understand given that the purpose of the book is to excite and equip historians for research in this sphere. Nanna Bonde Thylstrup’s work, however, gives us a different lens through which to conceptualize this deluge. Presenting an interesting theoretical intervention in ‘digital traces’, Thylstrup’s article ‘Data out of Place: Toxic Traces and the Politics of Recycling’ in Big Data and Society encourages us to change the conceptual frames around ‘digital traces’ in order to reimagine these traces as ‘digital waste’. In doing so, we are able to extend and nuance discussions around digital traces, and to connect them to social (and environmental) toxicities which include gendered and colonial structures of violence. Thylstrup’s work offers a change of perspective about the data cycle, positioning datafication—the distillation of ‘old’ information into ‘new’ and pure data points that can then offer new insights—as a process ‘premised on a logic of waste and recycling, with significant implications for how we consider datafication’s politics and ethics’ (p. 1). Understanding datafication in this way, she argues, allows us to understand the extent to which the knowledge produced in this process is marked by relations of power, and that these relations ‘constitute a system of representation’. The recycling of digital traces in this way ‘extracts and repurposes previous forms of meaning, embedding them in new representations’ (p. 2), and these representations are marred by the digital waste of the past. Building on the work of scholars of discard studies such as Jennifer Gabrys (Digital Rubbish and Program Earth) and Jussi Parikka (‘New Materialism as Media Theory’), Thylstrup extends this reading of metaphorical waste in critical data studies to theorize the toxic impact of systems of power and their representations in Big Data knowledge production. This powerful article dispels the notion that data is a ‘processed good’. By connecting new media theory, which frames digital traces as by-products, to waste theory, Thylstrup follows the whole process of what she terms the ‘waste handling process’ in datafication. ‘From their pulped state, digital traces are once again reinserted into the digital economy’, Thylstrup states, but she further argues that this process of data-recycling is political, one in which she questions whether the decision to classify something as waste or value is evenly distributed (p. 5). Thylstrup also questions the management of these ‘recycled’ data products, showing that questions need to be asked about their capacity for harm. Citing the work of Keyes, Steve, and Wernimont on facial recognition data (‘The Government Is Using the Most Vulnerable People’), she argues that data traces remain even after the data has been through its bruising journey of recycling. Despite arguing convincingly that ‘the ghostly presence of those marked by violence ends up haunting Big Data’s knowledge production processes’, the impact of Thylstrup’s article is to make tangible the cycle of data-processing and recycling by reimagining it through the language of waste. Through contesting the notions of ‘raw’ and ‘cleaned’ data, by refusing to accept the ‘washing’ of toxic datasets, we can surface and interrogate the economies of Big Data that rely on these polluted materials. It is a powerful imperative. 6. Abundance and Absence: Race, Gender, Intersectionality The remainder of this chapter will expand this focus on complicating perceived abundance, looking at a number of works which highlight the ways in which the ‘default’ data point distorts our perspective. One of the most important books to be published in 2019 was Invisible Women: Exposing Data Bias in the World Designed for Men by Caroline Criado Perez. Through a number of campaigns and interventions, Criado Perez has worked to draw attention to the lack of representation of women in public life. Invisible Women forces us to take another look at how data has historically been collected and interpreted, and brings home the complexity of unravelling data gaps and their points of origin. When campaigning in 2013 against the Bank of England’s decision to remove Elizabeth Fry from the £5 note (which would have left no women represented on the reverse of British banknotes), Criado Perez became aware of the way in which the male default could be ‘both a cause and a consequence of the gender data gap’. The Bank of England’s subjective criteria for establishing historical significance had relied upon ‘the kind of success typically achieved by men’ which meant that women were ‘far less likely to be able to fulfil any of these “objective” criteria’ (p. 16). This meticulously researched book unpicks a wide range of multilayered patterns of bias, absence, and misrepresentation that have been reproduced and replicated. In addition to being an important contribution to feminism in the twenty-first century, especially for those awakening to the problems and difficulties created by this data gap, it also serves as a crucial reminder that, in the age of data abundance, we are still only seeing part of the picture: ‘the stories we tell ourselves about our past, present, and future. They are all marked—disfigured—by a female-shaped “absent presence”’ (preface). Perhaps of most significance in this book is a rich understanding of the damage done by our reliance on the ‘default male’. Skewed or unreliable data leads to damaged systems, and it is this point that Criado Perez’s work underlines for scholars relying not only on datasets, but on the systems they have created. Through chapters on daily life, design, medicine, and public life, Criado Perez not only shows how everyday sexism is alive and well in all of these areas, but also draws attention to the historic gender data gap that misinforms the systems we rely on, systems which are increasingly forming the basis of algorithms and computational models that are said to ‘improve’ many of these services. Invisible Women alerts us to gaps in data collection that are based on faulty assumptions, which are then perpetuated by reliance on these datasets as the only available data. In piercing the flagrant data gaps relating to women’s lives, showing the vast amount of hidden labour, the additional difficulties and discomforts women experience due to these gaps, and, most worryingly, the additional dangers that women face, Criado Perez also exposes the faultiness of the concept of the ‘default male’. This concept benefits no one, and constrains society by forcing us to conform, adapt, or miss out in ways that scream, ‘Why can’t a woman be more like a man?’. But the ‘default male’ isn’t even a helpful proxy for men, let alone for women, transgender, and gender-fluid people (p. 314). In Race After Technology, Ruha Benjamin presses home a similar point, this time exposing the white American as the uncomfortable and unreliable default. Scholars, Benjamin argues, have anointed this ‘plainness’ as the ‘invisible centre against which everything is compared’, but this ‘absence’ holds significant power. The power of the ‘presumed blandness’ is immense, she argues, operating like a ‘superpower’ and conveying ‘immunity’. Both Benjamin’s and Criado Perez’s work raises the profile of the faults and defaults underlying our systems, which are reproducing not only sexism and racism but absence, blandness, and a failure to fully take account of our society. Benjamin highlights the work of scholars such as Browne (Dark Matters) on ‘racializing surveillance’, Broussard (Artificial Unintelligence) on ‘technochauvinism’, Buolamwini (‘The Algorithmic Justice League’) on the ‘coded gaze’, Eubanks (Automating Inequity) on the ‘digital poorhouse’, Noble, on ‘algorithms of oppression’, and Wachter-Boettcher (Technically Wrong) on ‘algorithmic inequity’ in describing the power of the invisible centre against which everything else is compared, and in questioning the technologies that reflect and reproduce existing inequities while appearing ‘more objective or progressive than the discriminatory systems of a previous era’ (pp. 5–6). Benjamin underlines the importance of realizing the connections between ‘biased bots, altruistic algorithms, and their many coded cousins’ (p. 7) Benjamin introduces a powerful new conceptual frame for examining these issues. Presenting the ‘New Jim Code’, a reference both to Michelle Alexander’s The New Jim Crow and indeed to the ‘old’ Jim Crow, a term that originated from the title character of an early nineteenth-century minstrel show but which rapidly became a shorthand for mocking and minimizing black people and marking exclusions and separations. Benjamin builds on the elasticity she reads into the term, which she takes to encompass a wide range of factors, from temporal to legal, geographical to institutional, and expands this to the interplay between cultural and technical coding. With the New Jim Code, she advocates for ‘race critical code studies’, informed by science and technology studies and critical race studies. Four dimensions of the New Jim Code are explored in the book: Chapter 1 examines ‘engineered inequality’, and the ways in which ‘racist robots’ are enmeshed in wider networks of power. Chapter 2 takes on ‘default discrimination’ and gives us a new understanding of the way in which structural inequity can be passed off as a ‘glitch’. A closer examination of this ‘glitch’ reveals that it functions in a similar way to deviations from Criado Perez’s ‘default male’, often seen as a momentary non-conformity. A ‘glitch’, Benjamin argues, is ‘generally considered a fleeting interruption of an otherwise benign system’, when it is more useful to think of it as ‘between fleeting and durable’ and moreover ‘not spurious but rather a kind of signal of how the system operates’ (p. 80). Chapter 3 challenges our gaze, and suggests ways in which multiple forms of ‘coded exposure’ are enacted through different technologies, from hardware to software. Benjamin’s brilliant conceptualizing of the notion of ‘exposure’ to account for light and shade, for the extent to which these are revealed and developed through technology, and for the risks involved in submitting (or being submitted) to this process draws together multiple types and layers of oppression in one of the most affecting chapters in this excellent book. A chapter on ‘technological beneficence’ follows, which demonstrates how limited and faulty definitions and operationalizations of ‘fairness’ can result in further discriminatory practice. The book concludes by mapping the range and scale of efforts to resist the New Jim Code, producing a resounding imperative to critically engage with the progressive narratives surrounding new technologies in order to advocate for new ‘justice-orientated design practices’. Race After Technology, like Invisible Women, is an extremely important work, examining what happens when cultural coding becomes embedded in the technical coding of software programs. Both texts reveal the numerous technological advances that are frequently sold as morally superior because they purport to rise above human bias, and the reality that they could not exist without data produced through histories of exclusion and discrimination. As Benjamin reminds us, ‘Codes are both reflective and predictive. They have a past and a future’ (p. 31), but it cannot be an unquestioned future when seen through the framing of ‘default discrimination’ presented by Benjamin and Criado Perez. Encouraging us to think and act ‘beyond the dominant genre’ allows us to imagine the possibility of ‘telling different stories about the past, the present, and the future’ (Benjamin, p. 31). This is a plea we also see in works on intersectionality. Last year saw the publication of Bodies of Information: Intersectional Feminism and Digital Humanities, edited by Elizabeth Losh and Jacqueline Wernimont, a hugely important body of work which showed that digital humanities scholarship could enter into a rich and productive dialogue with scholars of human–computer interaction, science and technology studies, and media studies in order to ‘expand our notions of text and context, archive and canon, and code and program’ (p. xii). Intersectionality in Digital Humanities, edited by Barbara Bordalejo and Roopika Risam, brings together a similarly wide range of scholarship that resists crude categorization and broadens our thinking about how race, class, gender, sexuality, nation, and other dimensions have shaped our data, archives, and practices. Many of the themes raised in works reviewed in this chapter reverberate around this volume, despite the fact that some of the work reproduced in it originates from many decades ago. For instance, the volume includes Moya Bailey’s work ‘All the Digital Humanists Are White, All the Nerds Are Men, but Some of Us Are Brave’, originally published in the Journal of Digital Humanities in 2011, which sought to draw attention to and to reject what she termed the ‘add and stir model of diversity’ which she likened to a ‘practice of sprinkling in more women, people of colour, disabled folks’. Bailey exposed the short-sightedness of such an approach, foregrounding the work of Criado Perez and Benjamin discussed above. In their introduction to the volume, Bordalejo and Risam rail against assumptions about the lack of bias in technologies, bias that Criado Perez and Benjamin explicitly situate at the heart of their work, and yet, as Bordalejo and Risam show us, much of this bias continues in scholarship, including the digital humanities. These biases are ‘often hidden, deeply embedded within the methods subtending scholarly practices, encoded by human actors who have failed to explore how their own biases are translating to the technologies they are designing’ (p. 3). Dorothy Kim’s essay in this volume, on how the replication of dominant and harmful structures of power in digital humanities archives can produce and perpetuate an ethics of harm for people of colour, echoes Benjamin’s work on ‘coded exposure’. Intersectionality in Digital Humanities also includes Roopika Risam’s work on the ‘fictional dichotomy’ of ‘hack vs. yack’ within the digital humanities, making a case for the coming together of theory and practice through cultural critique that is echoed in James Dobson’s vision in Critical Digital Humanities. Like Dobson, Risam argues for an expansive definition of critical digital humanities to include intersectional lenses, which she argues are key to realizing the potential of the discipline. Much of this early work therefore feels as vital today as it did then, and richly deserves the amplification that inclusion in such a collection produces. Overall the volume pushes for a paradigm shift that would prompt the much-needed ‘difficult conversations about equity, justice, and the influence of intersectionality on digital humanities practices’ (p. 5); given the level of synergy with other emerging work in the digital humanities in 2019, it appears that the time is right for such a shift. Indeed, it may already be under way. Books Reviewed Benjamin Ruha , Race After Technology ( Cambridge : Polity , 2019 ). ISBN 9 7815 0952 6406. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bordalejo B. , Risam R. , eds, Intersectionality in Digital Humanities ( Leeds : Arc University Press , 2019 ). ISBN 9 7816 4189 0519. Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Criado Perez Caroline , Invisible Women: Exposing Data Bias in a World Designed for Men ( London : Chatto & Windus , 2019 ). ISBN 9 7817 8474 1723. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Dobson James E. , Critical Digital Humanities: The Search for a Methodology ( Urbana : University of Illinois Press , 2019 ). ISBN 9 7802 5208 4041. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Milligan Ian , History in the Age of Abundance ( Montreal : McGill-Queen’s University Press , 2019 ). ISBN 9 7807 7355 6973. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Underwood Ted , Distant Horizons: Digital Evidence and Literary Change ( Chicago : Chicago University Press , 2019 ). ISBN 9 7802 2661 2836. Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Journal Articles Reviewed Da Nan Z. , ‘The Computational Case against Computational Literary Studies’ , Critical Inquiry , 45 . 3 (Spring 2019 ), 601 – 39 ; [accessed 22 January 2020]. Google Scholar Crossref Search ADS WorldCat Thylstrup Nanna Bonde , ‘Data out of Place: Data Waste and the Politics of Data Recycling’ , Big Data and Society , 6 . 2 (September 2019 ), [accessed 6 June 2020]. Google Scholar OpenURL Placeholder Text WorldCat References Alexander Michelle , The New Jim Crow: Mass Incarceration in the Age of Colorblind Racism ( New York : New Press , 2012 ). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Berry David , ‘The Computational Turn: Thinking about the Digital Humanities , Culture Machine , 12 ( 2011 ). Google Scholar OpenURL Placeholder Text WorldCat Broussard Meredith , Artificial Unintelligence: How Computers Misunderstand the World ( Cambridge : MIT Press , 2018 ). Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Browne Simone , Dark Matters: On the Surveillance of Blackness ( Durham, NC : Duke University Press , 2015 ). Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Brügger Niels , The Archived Web: Doing History in the Digital Age ( Cambridge, MA : MIT Press , 2018 ). Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Brügger N. , Milligan I. , eds, The SAGE Handbook of Web History ( London : SAGE , 2018 ). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Buolamwini Joy , ‘The Algorithmic Justice League’ , Medium (December, 2016 ), [accessed 14 June 2020]. Google Scholar OpenURL Placeholder Text WorldCat Computation Literary Studies: A Critical Inquiry Online Forum, [accessed 30 January 2020 ]. Ess, Charles, and AoIR Ethics Working Committee, “Ethical Decision-Making and Internet Research: Recommendations from the AoIR Ethics Working Committee”, approved 27 November 2002, [accessed 7 June 2020]. Eubanks Virginia , Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor ( New York : St Martin’s Press , 2018 ). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gabrys J. Digital Rubbish: A Natural History of Electronics ( Ann Arbor : University of Michigan Press , 2010 ). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gabrys J. , Program Earth: Environmental Sensing Technology and the Making of a Computational Planet ( Minneapolis : University of Minnesota Press , 2016 ) Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Hayles N. Katherine , How We Think: Digital Media and Contemporary Technogenesis ( Chicago : Chicago University Press , 2012 ). Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Keyes O. , Steve N. , Wernimont J. , ‘The Government Is Using the Most Vulnerable People to Test Facial Recognition Software’ , Slate Magazine (17 March 2019 ). [accessed 14 June 2020]. Google Scholar OpenURL Placeholder Text WorldCat Losh Elizabeth , Wernimont Jacqueline , eds, Bodies of Information: Intersectional Feminism and Digital Humanities ( Minneapolis : University of Minnesota Press , 2018 ). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Mandal A. , ‘Digital Humanities’ , The Year’s Work in Critical and Cultural Theory , 27 . 1 ( 2019 ), 364 – 86 ; . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Noble Safiya Umoja , Algorithms of Oppression: How Search Engines Reinforce Racism ( New York : NYU Press , 2018 ). Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Parham Marisa , ‘Ninety-Nine Problems: Assessment, Inclusion, and Other Old-New Problems’ , American Quarterly , 70 . 3 ( 2018 ), 677 – 84 . Google Scholar Crossref Search ADS WorldCat Parikka J. , ‘New Materialism as Media Theory: Media Natures and Dirty Matter’ , Communication and Critical/Cultural Studies , 9 . 1 ( 2011 ), 95 – 100 . Google Scholar Crossref Search ADS WorldCat Andrew Piper. Enumerations: Data and Literary Study . (Chicago . University of Chicago Press, 2018) . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Richards I. A. , Practical Criticism: A Study of Literary Judgement (1929; New York : Routledge , 2011 ). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Risam Roopika , New Digital Worlds: Postcolonial Digital Humanities in Theory, Praxis, and Pedagogy ( Evanston, IL : Northwestern University Press , 2019 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Wachter-Boettcher Sara , Technically Wrong: Sexist Apps, Biased Algorithms, and Other Threats of Toxic Tech ( New York : Norton , 2017 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © The English Association (2021) All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) © The English Association (2021) All rights reserved. For permissions, please email: journals.permissions@oup.com TI - 5Digital Humanities JF - The Year's Work in Critical and Cultural Theory DO - 10.1093/ywcct/mbaa014 DA - 2021-03-28 UR - https://www.deepdyve.com/lp/oxford-university-press/5digital-humanities-FEpAX6UbRv SP - 86 EP - 101 VL - 28 IS - 1 DP - DeepDyve ER -