Cultural Studies, Big Data, Scalability: Benjamin Versus Bourdieu

1.1 Cultural Studies

Cultural Studies has a more clearly defined origin story and date than most theoretical revolutions: it begins with the founding of the Centre for Contemporary Cultural Studies at the University of Birmingham in 1964. Of course, the founders affiliated with Birmingham studied and referred to precursors such as Walter Benjamin, Max Weber, Roland Barthes, and Antonio Gramsci. In 1967, Centre researcher Stuart Hall formulated the aims of this version of cultural studies as follows:

Beginning with the work or the event, our purpose must be to find the best means at our disposal for “reading” its cultural significance, for reaching an “interpretation” in [Max] Weber’s sense. The term culture must embrace “significant social behaviour” which, in so far as it has meaning at all, must be a cultural event. […] But it includes such meaning and values as they are embodied in and structured by institutions, as they are carried in the encounters and actions of men, or expressed through language, gesture, custom and art. These are all bearers of meaning in and between men in society. Hall (1967, 69).

Hall’s formulation itself bears close reading, starting with the phrase “the work or the event.” He is not talking about just theater performances or music concerts as events. Historical events come to mind, but also publishing events, or indeed the performance theater of politics. (The work Hall goes on to analyze is the newspaper gossip column.) “Event” is used again further down in the quote, preceded by the adjective “cultural.” The epithet is posited as equivalent to “significant social behaviour,” implying a focus not on the continuity of social behavior, as an anthropologist or sociologist might aim for, but on a singularity that is foregrounded. And does the phrase “significant social behaviour” contain one word too many? Logically speaking, either of the first two words could be removed from the phrase without altering its meaning, because behavior that is social is significant, and significance is a social category.

Hall’s citation of Max Weber’s “interpretive sociology” is interesting not only because it lends a sociological orientation to cultural studies, but also because Weber’s original German term, translated here as “interpretive,” is “verstehende,” more commonly rendered as “understanding,” in a hermeneutic sense. Hall may turn to Weber as a shorthand for what was to become his own semiotic theory of encoding and decoding the messages of popular culture. The significance of hermeneutic and semiotic approaches to popular culture is to counter ideas that it “contains” only consumerist propaganda or conformist ideology, whereas in fact every message requires decoding by the receiver. The default mode for understanding social behavior in Weber is modeled on the “homo economicus” of political economy. Agents do not always behave rationally, but irrational behavior can be understood through a combination of shared emotional repertoire and measurement against what the rational outcome would have been:

The instrumentally rational (zweckrational) interpretation possesses the highest measure of “self-evidence.” Instrumentally rational behavior is behavior exclusively oriented to means (subjectively) considered adequate to attain goals (subjectively) clearly comprehended. Not only instrumentally rational action is understandable to us; we also “understand” the typical course of the emotions and their typical consequences for behavior. The “understandable” has fluid boundaries for the empirical disciplines. Weber 1981, 151

Beyond the claims of sociology, however, Hall’s idea of reading a cultural event shows cultural studies to be equally inflected by literary studies, meaning that the techniques of close reading and the hermeneutics not of understanding, but of suspicion are applied to areas beyond the literary sphere. Cultural studies, though, is not to be philological in orientation: its objects of interpretation will not generally be written texts, but rather ones that are broadcast, or that consist of the habits, actions, and gestures of people in particular situations. Hall’s idea of cultural studies as a reading of events invokes one of the original Big Data achievements: a dictionary.

1.2 Big Data

“Big Data” refers to datasets whose size is beyond the ability of typical data-base software tools to capture, store, manage, and analyze. This definition is intentionally subjective and incorporates a moving definition of how big a dataset needs to be in order to be considered Big Data – i.e., we don’t define Big Data in terms of being larger than a certain number of terabytes (thousands of gigabytes). [T]he definition can vary by sector, depending on what kinds of software tools are commonly available and what sizes of datasets are common in a particular industry. Manyika et al. (2011; emphasis added)

Big Data is characterized by the so-called three “V”s – volume, velocity, and variety – any one of which may exceed processing capacity and require specialized techniques for handling. Important here is that no specific numerical values define what counts as “big” in reference to data. The same dataset can be considered Big or something less than big, depending on the perspective, purposes, and methodology of the decoder. One also notes that the three “V”s influence each other. For example, the decoder will find the variety of data more overwhelming if it arrives more quickly and in greater volume. Big Data is thus, as noted in the above citation from 2011, a relative and even subjective concept. In addition, while Big Data is most commonly associated with digital data and automated collection methods, it can also be usefully applied to analog techniques. According to Peter Gilliver, the creation of the Oxford English Dictionary out of the enormous mass of linguistic data in the late-19th and early-20th centuries could only be accomplished by a “well-oiled” machine, composed of a variety of both organic and inorganic parts:

[A]t the heart of this project [ …] was a remarkable machine, which would be described by the University’s Chancellor as ‘the largest single engine of Research working anywhere at the present time’. For a full appreciation of the output of this machine it is essential to understand its constituent components and processes, which by 1897 had reached a form which remained unchanged in all essentials throughout the period of compilation of the first edition of the OED. It is true that the number of drivers of the machine, already two since 1888, had not yet reached its maximum. Gilliver (2016, 260; emphasis added)

Besides collecting, preserving, and presenting this data to users who wish to understand the meanings and etymologies of words in English, the OED machine processed the date, for example by grouping examples of word usage under different possible meanings. of the word.

While there is ambiguity in the amount of data required for Bigness, the use of numbers to quantify makes clear that data are discrete, countable units. The “sectors” associated with Big Data are natural science and quantitative methodologies in social science, not to mention with finance, geographic information systems (GIS), biopolitics, and governmentality. Within the humanities, linguistics and history have been known to work with large corpuses. Literary and cultural studies have shown little affinity not only for Big Data, but for data in general, given that data generally is associated with numerical computation or manipulation. For literary scholars, all the novels published in the English language is more than they can ever read, whereas in pure data terms (e.g., file size) such a corpus is dwarfed by the cinematic output in that language of a single year, despite the many fewer titles. Or, if we conceive of a literary work as a set of the creations of readers and wish to produce an image of such a work, then Big Data could be, for example, the detailed descriptions of 10 readers or so of their experience in reading the text, since this is 10 times the set of readers whom we normally interrogate. If the 10 readers interpreting this text are non-professionals, then such a set also presents much greater variety of backgrounds and skillsets than literary critics are used to dealing with. Literary criticism is like opera singing, where the voice is denaturalized and unroughened. In all these cases, researchers are confronted with the dilemma of which questions they wish to ask of the corpus, and of what methods they can use to answer them. For cultural studies to live up to its name, on the other hand, it would need to foreground the culture event being analyzed against the general matrix of cultural behaviors. Such are the Big Data projects of Walter Benjamin and Pierre Bourdieu.

1.3 Benjamin’s Arcades Project

Given the relativity and mutability of what counts as data, we might consider the “grandfather” of Big Data cultural studies projects to be Walter Benjamin’s unfinished Passagen-Werk or Arcades Project. Benjamin’s aim was to create a picture of the whole culture of nineteenth-century Paris out of a set of what the German theorist called “dialectical images”:

Corresponding to the form of the new means of production [structure], which in the beginning is still ruled by the form of the old (Marx), are images in the collective consicousness in which the new is permeated with the old. These images are wish images; in them the collective seeks both to overcome and to transfigure the immaturity of the social product and the inadequacies in the social organization of production. Benjamin (1999, 4–5)

To identify and to analyze this constellation of images was the research task which, in Benjamin’s view, could only be approached through a totalizing act of data-gathering, most of which he carried out at the Bibliothèque Nationale in Paris. The Parisian passage was one such wish-image, a “street-gallery,” forerunner of the shopping mall in that a variety of business were housed under a single roof.

Benjamin’s life – and his work on the Arcades that had gone on for 13 years – from 1927 to 1940 – was cut tragically short by Nazi persecution that drove him to flight and ultimately to suicide. His collecting work, left behind and preserved at the Bibliothèque nationale, was divided into thirty-six “Konvoluten,” sheafs of related quotations, images, and reflections, which presumably would have formed chapters resembling encyclopedia entries in the finished book. As a large-scale collection of these images, Benjamin’s Arcades Project belongs to the great unfinished fragments of modernity. We do not really know the format in which Benjamin would have published this work if he had lived to finish it, the extent to which he would have narrowed the huge variety of quotations, for example, and the extent to which the citations would have become examples to illustrate a coherent argument. It is difficult to convey the unique flavor of any Arcades-Konvolut without extensive quotation from it, for which space does not allow.

Benjamin’s stated methodology leaves open the possibility that the finished product would resemble in large part the draft he left behind. In convolute N2.6, Benjamin provides the methodological rationale for his undertaking, which we might call a Marxist monadology:

In what way is it possible to conjoin a heightened graphicness [Anschaulichkeit] to the realization of the Marxist method? The first stage in this undertaking will be to carry over the principle of montage into history. That is, to assemble large-scale constructions out of the smallest and most precisely cut components. Indeed, to discover in the analysis of the small individual moment the crystal of the total event. Benjamin (1999, 461)

The Arcades Project existed at the time of Benjamin’s death in the form of 36 dossiers or Konvoluten that carried titles such as “Mode” (fashion), “die Langeweile, ewige Wiederkehr” (ennui, eternal return); “Baudelaire,” “der Flâneur,” “soziale Bewegung” (social movement), and so forth. These convolutes consisted mostly of quotations that had something to do with their titles, along with Benjamin’s glosses on these or his reflections on the topic. From the mass of data, Benjamin selected a number of micro-projects that he felt reflected on the whole, such as the poetry of Charles Baudelaire or the proto-surrealist graphic works of Grandville.

1.4 Bourdieu’s Distinction

Another example of a Big Data approach to culture, also Marxist at its base, is Pierre Bourdieu’s 1979 study, La Distinction, based on the answers given by 692 subjects living in Paris and Lille to a long and multifaceted survey of their reading, listening, eating, and other cultural habits. The subtitle of the work, A Social Critique of the Judgement of Taste, is indicative of Bourdieu’s goal of dismantling the dominant idea of taste as a cognitive function available to all that had been put forward by Immanuel Kant in his Third Critique (1790; Kritik der Urteilskraft). Bourdieu concludes that the behaviors of the various classes and class fractions are internally consistent, ruled by aesthetic appreciation as enjoyment of unconsecrated popular works on the lower end of the social scale, and by distanced aesthetic pleasure in what Bourdieu calls “legitimate artworks” at the upper end of the social scale.^[1] “Class fraction” is the stronger analytic term for this study, since classes are very broad categories that would not allow for the drawing of distinctive lines of difference. Examples of class fractions include dividing what Karl Marx dubbed the “proletariat” between farm workers, farmers, skilled and unskilled laborers, craftsmen, and foremen. Bourdieu’s vocabulary veers from speaking of classes as wholes to class fractions. A strong correlation is demonstrated between an individual’s tastes in cuisine, furniture, art, and literature. Class is a predictor of culture, and differences in culture help to keep classes distinct from each other: The aesthetic gaze or “eye” is available only to those with the cultural capital necessary to its cultivation, in the same way that only those with enough money can afford operations to restore their vision:

The ‘eyeʼ is a product of history reproduced by education. This is true of the mode of artistic perception now accepted as legitimate, that is, the aesthetic disposition, the capacity to consider in and for themselves, as form rather than function, not only the works designated for such apprehension, i.e., legitimate works of art, but everything in the world, including cultural objects which are not yet consecrated – such as, at one time, primitive arts, or, nowadays, popular photography or kitsch-and natural objects. The ’pure’ gaze is a historical invention linked to the emergence of an autonomous field of artistic production, that is, a field capable of imposing its own norms on both the production and the consumption of its products. Bourdieu (1984, 3)

The aesthetic gaze that strips any object of its pragmatic value and turns it into pure form is neither universal nor a product of raw talent, but what we might call a skill that requires long periods of – mostly unconscious – training and, like handicraft skills, is often accumulated in families through generations. For example, children are more likely to become readers when they grow up within households that value offer books.

Bourdieu’s study exhibits a variegated textuality, including on the one end tables and figures that convey the “big (data) picture,” and on the other end individual stories of French citizens. The chapter “Cultural Goodwill” (318–71), for example, is punctuated by a series of stories: “A ‘Very Modest’ Nurse” (324–25); “A Technician Who ‘Tries to Get On’” (334–36); “A Baker’s Wife Who Gets It ‘Just About Right’” (347–49); and “A Nurse Who ‘Lives With Passion’” (355–57). As with Benjamin, Bourdieu concerns himself with the heightened Anschaulichkeit or graphicness of his Marxism, chunking and concretizing his data in tableaux such as Figure 1.

Figure 1:

Variants of the dominant taste (Bourdieu 1984, 340).

Visualization is an important tool for making Big Data tell a story, with techniques ranging from bar graphs to network analysis diagrams. Here, however, the visualization confuses more than it enlightens, to the extent that I imagine Bourdieu behind a green curtain laughing at me as I try to puzzle it out. The following 10 pages or so of Distinction are spent explaining the graph. The X and Y axes are the familiar ones: cultural capital/economic capital for the X and, in this case, whether the respondents belong to a class fraction that is rising in society of declining, respectively for the Y. The polygons enclose class fractions. Especially confusing is the juxtaposition of cultural works, mostly music and painting, with other cultural habits and with personal qualities that are considered desirable. For example, in quadrant I (upper right) near the Y axis one finds “Leonardo [da Vinci],” “Traditional Cooking,” and “Level-Headed” next to each other, indicating a high probability that individuals with declining economic opportunities and an average accumulation of cultural capital will be attracted to all three of these cultural values, which interviews show correlate with each other.

Johann Strauss’ “Blue Danube” is at the far right of the graph – economic capital plus and cultural capital minus – while J. S. Bach’s “Well Tempered Klavier” is at the far left – economic capital minus and cultural capital plus, for example, schoolteachers. Presumably, Strauss is simpler and more straightforward, more tuneful for example, than is the work of Bach. Throughout the hundreds of pages of Distinction, similar contrasts are presented, without an accompanying analysis of the texture of works. The interpretive process is thus a circular one, in which academicians of the dominant class, certainly in terms of taste, aesthetically judge the kinds of works enjoyed by various classes.

Bourdieu also inverts the methodology at times, for example by showing the same photograph to individuals from a variety of class fractions. Below is the palette of responses to a photograph of a gasworks taken at night. The quote is necessarily long, in order to capture the variety of social positions towards the aesthetics (or lack thereof) of the photograph. I have added italics to identify the various class fractions. We begin with Bourdieu’s comments.

When confronted with a photograph of the Lacq gas refinery, which is likely to disconcert realist expectations both by its subject, an industrial complex, normally excluded from the world of legitimate representation, and by the treatment it receives (night photography), manual workers, perplexed, hesitate, and eventually, in most cases, admit defeat: ‘At first sight it’s a construction in metal but I can’t make head or tail of it. It might be something used in an electric power station… I can’t make out what it is, it’s a mystery to me’ (manual worker, provinces). […]. Among small employers, who tend to be hostile to modern art experiments and, more generally, to all art in which they cannot see the marks and traces of work, a sense of confusion often leads to simple refusal: ’That is of no interest, it may be all very fine, but not for me. It’s always the same thing. Personally that stuff leaves me cold’ (craftsman, provinces). […] Office workers and junior executives, who are just as disconcerted as the manual workers and small employers, but are less inclined to admit it than the former and less inclined than the latter to challenge the legitimacy of what challenges them, less often decline to give a verdict: ’I like it as a photo… because it’s all drawn out; they’re just lines, it seems immense to me… A vast piece of scaffolding… It’s just light, captured by the camera’ (clerical worker, Paris). […] But only among members of the dominant class, who most often recognize the object represented, does judgement of form take on full autonomy vis-à-vis judgement of content (‘It’s inhuman but aesthetically beautiful because of the contrastsʼ), and the representation is apprehended as such, without reference to anything other than itself or realities of the same class (‘abstract paintingʼ, ‘avant-garde playsʼ etc.). Bourdieu (1984, 46–47)

The quote takes us through four class fractions, in which we see an increasing ability to bracket any real-world iconicity of the photograph – thus endowing it with Kantian autonomy and purposive purposelessness – as we move upward on the social scale. We also experience one of the most prominent flaws in Bourdieu’s argument, as he compares the class fractions – manual workers, small employers, and office workers – to a far more generalized dominant class. Does it include aristocrats, who are not otherwise mentioned as a class fraction?^[2]

1.5 Scalability

In both Benjamin and Bourdieu, the effect of the movement back and forth, the zooming in and out from individual agent to the overall structure within which agents make their choices, produces an effect called “scalability.” Big Data either is overwhelming and ungraspable, or else it fails to tell a story. Individual examples, on the other hand, either do not cover sufficient conceptual territory, or else they arouse suspicion as to the methods by which they are chosen and interpreted. Scalability is what allows analysis to move from large scale to small scale and back, collective to individual, thus better grounding the standard operating procedure of exemplarity in cultural studies. This way of deploying the concept of scalability differs from some others, for example from those that are common to economics or to web traffic, which generally move in one direction, and which involve the ability of a project to expand in size without changing its underlying assumptions. Two-way scalability corresponds more to what we have come to expect from digitized, interactive maps such as Google Maps, where one can easily switch from detail to overview by clicking the “+” or “−” buttons. Two-way scalability is also more readily distinguishable from the venerable humanistic standby of exemplarity. Scaling up from the particular to the general would be one way of defining exemplarity; which thus does seem to have a connection to scalability, as it does also, however, with other techniques such as case history and instance (cf. Bewes 2014, 1–3). With or without Big Data, there is always a bigger picture.

The entry on “Scale” in the Dictionary of Human Geography lists several types, including the cartographic one that makes maps possible. Scale also applies to the nested hierarchy of human relationships: “Hierarchical scales are nested sets of categories that fit inside each other. A typical geographical example is a hierarchy of zones working up from the body to the home, the street, the neighbourhood, the district, the region, the nation, and the continent to the global” (Rogers, Noel, and Kitchin 2013). The question of scale is thus at the heart of world literature debates, since one widely accepted definition of that phenomenon involves the ability of a work of literature to scale upward and outward into the zones in which it is read, with a crucial question being the degree to which this scaling upward preserves or distorts the work’s “original” meanings.^[3]

Perhaps world literature, like the citations of Benjamin and the interviews of Bourdieu, can be thought of as a “rush of stories” that challenges the framework in which they are gathered:

A rush of stories cannot be neatly summed up. Its scales do not nest neatly; they draw attention to interrupting geographies and tempos. These interruptions elicit more stories. This is the rush of stories’ power as a science. Yet it is just these interruptions that step out of the bounds of most modern science, which demands the possibility for infinite expansion without changing the research framework. […] The ability to make one’s research framework apply to greater scales, without changing the research questions, has become a hallmark of modern knowledge. Tsing (2015, 37–38)

Trevor Owens makes the same point as Anna Tsing in a more generalized and technical way, and from a different direction: “the most common use of statistics is to study a small sample in order to make generalizations about a larger population. But statistical tests intended to identify whether trends in small samples scale into larger populations are not useful if you want to explore the gritty details and peculiarities of a data set” (Owens 2011). Tsing’s ethnographic study of an example of “salvage capitalism” in the US Northwest, The Mushroom at the End of the World, analogizes the narratological point that she makes in the above quote to the economic model of scalability where lumberjacking of individual trees is scaled up to clear-cut devastation of an eco-system. The secondary growth is no longer viable for lumbering, but it is a source of matsutake mushrooms that must be harvested by foraging. Tsing “embeds” herself in the harvester community, searches for mushrooms with them, and gathers their stories.

As noted above, the scale of story in Bourdieu is that of individual responses to interviews, and what seem to be excesses beyond answering the questions, where respondents take the opportunity to explain and thereby justify their lives. “A ‘Very Modest’ Nurse” (324–25) is meant to be scalable to the level of class fraction, but only when paired with a second story of a younger nurse with more education, thus showing different positions on the Y axis. As a coda to the older nurse’s words, we read the following: “(An interview with another nurse, younger and better qualified, is presented later in this chapter. A comparison of the two will show in a very concrete way that the oppositions between age groups which divide a number of occupations in fact correspond to differences in scholastic generation and social trajectory, and consequently in lifestyle.)” (Bourdieu 1984, 325). The younger nurse is “one who lives with passion” (355). Her apartment is smaller and has no furniture, but the majority of her story is devoted to her cultural activities, including amateur theater acting, theater attendance, newspaper reading, attendance at art exhibitions, etc. The two opposing stories are meant to scale up to the general discord between economic and cultural capital, and to suggest a class fraction that depends upon age.

Benjamin does not insert stories into his exposé of nineteenth-century Paris: rather, the rush of stories, unfiltered, overwhelm his analysis. Most convolutes consist of four types of evidence: one or more epigraphs; citations from a wide variety of sources; Benjamin’s own reflections and aphorisms; and images. A “standard” scholarly monograph clearly nests these categories: the scholar’s own discourse is the overarching category, justifying the insertion of citations and images. (Epigraphs, as the term implies, are technically outside of any frame.) Benjamin hopes not that his data will be scalable, but rather that it will be combustible. James Rolleston explains. Texts presented by historians can already be considered citations, as they select, condense, and fix the continual flow of spoken language:

[T]he historian’s isolation of sentences or paragraphs from a text is already an intensified form of citation, a quoting to the second power. If the process is pursued further, acts of rupture multiply: passages from different kinds of text speak on a new level when juxtaposed, texts from distant years actualize a radical change or, even more interesting, an absence of change. At perhaps the highest level yet imaginable, a fusion of past texts activates the “language” of the historian’s own epoch. [… T]he historian momentarily fuses two or more isolated passages, thus producing the “dialectical image.” Rolleston (1989, 16)

In Rolleston’s view, then, the “unfinished” format of the Arcades manuscript cannot be far from its final form. Its data is to be fused rather than scaled.

1.6 Comparison and Conclusion

One would have expected more comparisons to have been made between these two works, as both Benjamin and Bourdieu have exerted immense influence on literary theory and criticism, not to mention on cultural studies. They both use Big Data approaches to their topics, because they wish to synthesize out of the data a totalizing picture of a slice of French history. For Benjamin, the period is nearly the whole of the nineteenth century, while for Bourdieu it is post-war France up to the time of publication. Their data has great volume and also variety. The third V, velocity, refers in both cases to the way the data hits the reader, rather to than the way it was collected by the researchers. This is especially true of Benjamin’s Passagen-Werk, where the variety of citations interspersed with comments, without a guiding argument in the individual convolutes, result in a subjective feeling of texts passing before one’s eyes on a whirling carousel. Add to that the fact that in Benjamin’s original, any particular citation may be in French, German, or at times English.

A non-interventionist Marxism underlies both undertakings. Bourdieu’s anchoring of aesthetic outlook to class origin harkens back to the famous “economic base versus ideological superstructure” approach to materialism, in which law, philosophy, and art reflect and reinforce the relations of production in a society. Marx was less deterministic than Bourdieu; the former acknowledged that Greek tragedy, for example, maintains an aesthetic power when performed in vastly different and more modern societies. Distinction, on the other hand, does not explore in depth the ideological conditionings that different artworks may suggest to different classes. Rather, it places its emphasis on the reverse, that the vast differences in cultural capital resulting from different educational experiences will inevitably result in different tastes. (I have not been able to find in Distinction a single example of a “working-class hero” who has elevated himself sufficiently to achieve formalistic appreciation of art and literature.)

Benjamin referred to his project as a “dialectical fairyland.” The most prominent dialectic is the location of the “newest” in the matrix of what has been:

This spectacle, the unique self-construction of the newest in the medium of what has been, makes for the true dialectical theater of fashion. Only as such, as the grandiose representation of this dialectic, can one appreciate the singular books of Grandville. […] When Grandville presents a new fan as the “fan of Iris” and his drawing suggests a rainbow, or when the Milky Way appears as an avenue illuminated at night by gas lamps, or when “the moon (a self-portrait)” reposes on fashionable velvet cushions instead of on clouds – at such moments we first come to see that it is precisely in this century, the most parched and imagination-starved, that the collective dream-energy of a society has taken refuge with redoubled vehemence in the mute impenetrable nebula of fashion, where the understanding cannot follow. Benjamin (1999, 64)

This quote captures Benjamin’s dialectical, almost alchemical method most forcefully. The consumerist impetus of the nineteenth century compressed the “dream-energy” of a people into an explosive powder that has been waiting for detonation by, for example, Surrealism in the twentieth. Both Bourdieu and Benjamin take culture in all its forms and on all levels seriously, but only the latter recognizes it as a location of hope.