Critique of Archaeological Reason
6. Critical reviews

Anne Burdick, Johanna Drucker, Peter Lunenfeld, Todd Pressner, Jeffrey Schnapp

Digital_Humanities (2012)

Giorgio Buccellati – July 2015

The book
Humanities: dealing with the texts
Digitality: the tools of analysis
The demise of humanism
An essential ambiguity
Digital humanism
Multilinearity
The role of design
Digital reading
Methods and techniques
Notes

The book

Comprehensive and authoritative, this book offers a good overview of the current state of “digital_Humanities.”^{Note 1}

The book is comprehensive in its coverage of the various facets of the discipline. The first chapter defines in depth the nature of the field and the second reviews the major trends, with a series of representative case studies: this is the best approach for a discipline that is so new and in such a fluid and rapid state of development. Chapter three explores ways in which the social dimension is affected, in scholarship and beyond, by the new media, while the last chapter looks at the future with its unique challenges. Slender in size, it has a particular impact becuase of its eloquent style and incisive formulations.

The book is “authoritative” in the special sense that it draws on the converging interests of the authors who come from different backgrounds and represent a wide gamut of competences: media design (Burdick and Lunefeld), bibliographical studies (Drucker), comparative literature (Presner and Schnapp). They have worked in unison (with a “consistent, if choral, voice,” p. 137) to bring about a well crafted product that serves as a demonstration of the effectiveness of “collaboration as creation” (p. 84).

As such, it opens a good window onto the established and prevalent current thinking, and it allows the reader to easily take stock of the situation of the discipline as it stands today. It is to this representative dimension of the book that I wish to draw attention, and I will do so by voicing substantial differences of opinion in the basic presuppositions, particularly with regard to the very question as to what “humanism” is (as different from the “humanities” which are seen as having taken its place). In this regard, the “humanities” approach is closer to what would have been called, at one time, computer-aided analysis of textual material.

Humanities: dealing with the texts

The centrality of the texts emerges in the book as the common denominator that unifies the field, to the point that the “humanities” seem to be synonymous with philology:

texts as the core	p. 7	text based disciplines and studies (classics, literature, philosophy, history of ideas) ... make up ... the core of ... the humanities
texts as objects	p. 32	Collection-building and curation have always defined humanistic learning: so much so that even the most ancient literary forms adopt listing, cataloging and inventorying as key features of poetic communication.
analysis	p. 37	The humanities have historically been the province of close analysis of limited data sets.

Humanistic training is thus restricted to dealing with texts, with authors and with corpora of the past, and innovation rests in developing new ways of dealing with archival goals:

being anchored to texts is the defining quality of the humanities	p. 45	By identifying elements of a system and thinking about how they relate to each other sequentially, or hierarchically, or relationally, humanists discover ways of modeling knowledge that were not part of their textual training.
data from the past were but are no longer the substance of knowledge	p. 32f	With the spread of print and the rise of modern institutions of memory ..., a new regime arises within which there exist such proliferations of historical information and cultural material that data from the past can no longer be assumed to possess a priori value. They become support for the production of knowledge, knowledge's precondition but not its substance.
corpora and archives	p. 112	Advanced work in the Digital Humanities will ... target pre-contemporary cultural corpora... Archival projects .. have begun to attract the perticipation of enthusiastic citizen scholars... Communities create and curate their own archival resources, promoting cultural awareness and a sense of citizen ownership of the cultural patrimony.
texts and authors	p. 36	With the increased fluidity of texts we see a corresponding change in authorial identity.

Digitality: the tools of analysis

Given that the humanities are based exclusively on texts, it follows that the digital dimension of the humanities should be defined by the tools used in dealing with the texts. Tools are magnified to the point where they are seen as having a power of understanding, agency and experience itself (“sentience”):

delegation of understanding	p. 105	"Tools not just tools. They are cognitive interfaces that presuppose forms of mental and physical discipline and organization. By scripting an action, they produce and transmit knowledge, and, in turn, model a world.
equality with machines	p. 105	As humans and data machines become equal partners in cultural practice, social experience and humanistic research, the humanities may no longer look like "the humanities." The scales and registers of what counts or is valued as human experience and, therefore, the objects of humanistic inquiry, will find themselves altered.
the machine as "adept partner"	p. 105f	Will we read the machine's analyses and summaries of marked texts. structured data, and natural language processing and feel we are in conversation with an adept partner, whom we will be tempted to imagine as a natural extension of our own cognitive capacities?
towards machine "sentience"	p. 105	Trust in computers' capacities for aggregation, synthesis and even selectivity is sure to grow over the coming century. Visions of machine agency and emerging sentience reek of science fiction fantasies, but unintended consequences may well be in our future.

The demise of humanism

Significantly, the term “humanism” occurs in the book only to indicate that it is a thing of the past, because one has gone from it to something else. The section title “From humanism to the humanities” (p. 5) is indeed idicative of a deeper stance vis-à-vis humanism than one might think at first. We do not have a mere temporal transition; rather, we witness the abandonment of one (humanism) in favor of the other (humanities).

Nor is the choice of terms merely a lexical matter. In the authors’ view, “humanism” is a fossil: it existed within the framework of the medieval trivium and quadrivium, and then in the Renaissance “monocular perspective” that “had human subjects at their core” (p. 5). As such, it is then superseded by the humanities, an event that can be attributed to the invention of the printing press: the print culture that followed in its wake privileged the textual tradition and thus gave rise to the humanities as we know it (ibid.).

At times the demise does not seem to be definitive yet. Statements such as the following would seem to leave open the possibility that the fossil might still be alive, after all:

core humanities concepts	p. 104	Building tools around core humanities concepts – subjectivity, ambiguity, contingency, observer-dependent variables in the production of knowledge – holds the promise of expanding current models of knowledge.
to be human	p. 82	the Digital Humanities studies and explicates what it means to be human in the networked information age
"traditional humanistic inquiry"	p. 107f	... digital humanists [who represent "traditional humanistic inquiry"] ... consider ... [the quantitative methods of the social sciences] epistemologically naive and their results generally trivial, self-evident, or flawed. They argue that the tools of the empirical sciences – statistical graphs ... – carry conviction because they assume information is observer-independent and rooted in certainty ... Such tools are ill-equipped to capture the complexities of novelistic constructions of character .
the sense of a lurking humanism	p. 108	... At stake for [the traditional humanists] in tracking this elusive universe of signs is much more than the mismatch between qualitative judgments and the quantitative strictures imposed by analytic tools or graphical expressions borrowed from the social sciences. What is at stake is the humanities' unique commitment to wrestle with uncertainty, ambiguity, and complexity; to model incommensurate temporalities and ontologies;... to attend to non-repeatable and nonstandard phenomena.
need to respect the traditional humanist	p. 92	We need to take seriously the conviction that the humanities have their own method – not based in calculation, automation, or statistical probability, but in ambiguity, interpretation, and in emodied and situated models of knowledge and knowing.

This is viewed as a “battle” raging between two camps. And the solution that is adumbrated lies is the search for better tools.

There is no adjective derived from the word “humanities,” so the person who engages in the humanities is called in the book a “humanist” – but this term has no longer a correlation to “humanism.” The authors’ “humanist” is really a person “carrying out the work of the humanities digitally” (p. 102), in other words, a philolologist who uses digital tools. Hence the solution to what are perceived as problems inherited from the old-fashioned humanism rests on the development of better text-related tools.

An essential ambiguity

In my view, the solution proposed is not bold enough and does not go to the heart of the matter. To meet the challenge of “subjectivity, ambiguity, contingency, observer-dependent variables in the production of knowledge” it is not sufficient to aim for better tools. These aim ultimately at eliminating “subjectivity” and all the rest, by breaking it all down into smaller and smaller components that are then susceptible of ever more refined quantitative analysis.

I see here an essential ambiguity in the authors’ approach (and of course in the field at large). They identify well the “core values” (subjectivity etc.), but they deal with them as if they were textual matters. Such ambiguity prevents us from exploiting digitality at its best. And this for two main reasons.

The first is that the approach is too reductive. Digitality is reduced to the level of what the printing press did a few centuries ago: just as the “print culture” favored the wider diffusion of texts, so “digital_Humanities” favors the wider diffusion of analytical tools. Which is fine for textual matters, but not beyond them.

The second problem is one of extrinsicism. By focusing on the texts as such, and exclusively so, one remains on the surface of the intellectual confrontation. The texts are an object to be treated from the outside, they are seen as a closed inner-referential system – the referentialiy to the observer and a world of values is lost.

Seen in this light, “digital_Humanities” is really understood as a branch of the social sciences.^{Note 2}

Digital humanism

My claim is that there is instead ample ground for a humanism that can remain true to its core values and be properly digital at the same time. And my further claim is that archaeology is the best place where such digital humanism can take shape.

Such notion of digital humanism is central to the argument in the CAR volume, especially in section 11.6. In particular, I make a case for the fact that archaeology is ideally suited for a constructive interaction between the social sciences and humanism in the full sense of the word. I argue that there is, on the one hand, a natively digital dimension to archaeology and, on the other, a commitment to deal with broken traditions in such a way as to recapture experience. Thus the notion of “digital humanism” is built into the very essence of archaeology, which can be seen as paradigmatic in illustrating it at its best.

Tools are indeed more than tools: I have argued this in the CAR volume and elsewhere by placing the digital revolution in the perspective of the scribal revolution. But they remain tools. They are really not equal partners.

I will not repeat here my argument but will only review some of the points made by Burdick and her co-authors where they deal with themes that are pertinent to the notion of digital humanism, even if only in an inchoate way.

Multilinearity

The authors’ description of multilinearity (though they do not use this term) is eloquent:

impact on cognition	p. 105	The cognitive horizons of digital researchers are already being deeply altered by the ability of data machines to zoom back and forth between grand sweeping views of masses of texts, data, and images and the microscopic particulars of single documents or objects.
impact on perception	p. 106	Perhaps we will become ever-more seduced by the macro and micro ends of the perceptual spectrum, by very big and very small data. How do such patterns influence the activities and perceptions of indivduals through the coercive force of normativity?
impact on conceptualization	p. 106	... the machine may provide conceptual frames and filters that provide access to, process, and shape the historical record. ... Each framework creates a different synthesis with pointers toward higher and lower levels of aggregation as well as specific documents, materials, views, models, or other evidence on which the synthesis was compiled.

This fits well with the view presented in CAR about multilinearity as a goal of digital thought. Where I differ is in the understanding of how we are to manage the impact that this new interaction with data has on the human mind.

The role of design

We presumably owe to the presence among the authors of two specialists in media design some very interesting remarks on the conceptual role of design. It is presented as “spatialized sequential narrative” (p. 14), as “intellectual method” (p. 13) as an “act of thinking” (p. 15). This aspect appears throughout the book, and the very typographical layout of its pages gives evidence of how much thought went into it. It is a matter of organization of the information, in ways that contribute to the logical construction of the argument. The “rhetoric of graphics,” as they aptly call it, is indeed something that affects the presentation of contents from the elegance of a web page to the graphic images that display statistical information:

rhetoric of graphics	p. 42	Understanding the rhetoric of graphics is an ... essential skill ... in working at a scale where individual objects are lost in the mass of processed information and data.
multimedia literacy	p. 118	Design methods inform all aspects of humanistic practice, just as rhetoric once served as both its glue and compositional technique. Comtemporary eloquence, power and persuasion merge traditional verbal and argumentative skills with the practice of multimedia literacy shaped by an understanding of the principles of design.
shaping knowledge	p. 118	Design means shaping knowledge and endowing it with form; the field of design encompasses structures of argument.
diagrammatic thinking	p. 119	The time of diagrammatic thinking is upon us. ... Authorship and display must converge in such a way that arguments become visible and can be made both graphically and spatially. Relations among visible entities, as well as verbal units of thought, become tractable in the process.

Indeed, the question of design goes beyond aesthetics. The intellectual dimension is brought out by how the very structure of the argument can be wedded to the display, even of a simple text. I have addressed this in CAR when dealing, for instance, with how segmentation can help in keeping the overall integrity of the text while allowing at the same time for the emergence and reciprocal interrelationship of the individual components.

Digital reading

In CAR I have addressed the issue of digital reading. The authors of Digital_Humanities approach the issue by contrasting close and distant reading.

close reading	p. 39	Within the humanities, close reading has been a central practice that is premised on careful attention to features contained in a text as well as its variations, history, transmission, possible meanings, and range of nuances. Close reading has its roots in the philological traditions of the humanities, but for more than a generatio has often been equated with deep hermeneutics and exegesis...
distant reading	p. 39	Distant reading is almost not reading at all, but rather engages the abilities of natural language processing to extract the gist of a whole mass of texts and summarizes them for a human reader in ways that allow researchers to detect large-scale trends, patterns and relationships that are not discernible from a single tyext or detailed analysis.
linking the two	p. 39	Rather than pitting distant reading against close reading, what we are seeing is the emergence of new conjunctions between the macro and the micro, general surface trends and deep hermeneutic inquiry, the global view from above and the local view from the ground. The digital humanist is capable of "toggling" between views of the data, zooming in and out...

Indeed, there cannot be a separation between the two. Nor has there ever been. The notion of “distant” reading reflects as rich a humanistic tradition as does “close” reading. The intense confrontation with the notion of structure, the care with which capturing the sense of the whole affects the understanding of the details, the search for deep unifying threads of large compositional frameworks – all of this is deeply rooted in humanism. And it is from the richness of this tradition that we can make sure that a digital “distant” reading does not evaporate into a sterile mass approach to aggregation for its own sake.

Methods and techniques

The “Humanities” have absorbed and refined a great number of digital techniques. With these, we have been led to “do the humanities digitally” (p. 102), which means that no really new methods have emerged. The “assumptions” that underlie the process have “become invisible,” which may “sound the death knell for Digital_Humanities as a practice” (p. 102).

This is the beginning of section 4, “Provocations.” I am arguing differently. A method has indeed emerged, but it has not been articulated in its theoretical dimension, especially not for the humanities. The core of this “method” has been the dark intimation of a digital argument. Such an argument is being applied, but its characteristics are not being identified. The nature of the argument as I have outlined it in CAR is in some ways present in practice (certainly for data bases, less so for the narrative), but one looks at it as a technique, not for what it proposes as a method.

The same goes for digital thought as the safeguard of the hermeneutic risk. In humanistic practice, the impulse remains that of an independent human judgment, one that transcends the thrust of automated processes. Statistical sensitivity has always been used (see p.107 –?), except that the underlying universe was incomparably smaller, and the structuring grammar minimal in scope and detail. Universe and grammar are changed, so that the statistical base is altogether new. But the distance of the observer remain the same: it is the distance of humanistic digital thought.

Notes

Note 1. – The linking underscore in the title (Digital_Humanities) is used to stress the tightness of the noun phrase (p. ix-x), which is used throughout as a noun in the singular (e. g., pp. 7, 122). Occasionally, the phrase is used without the underscore, but it still acts a singlura, e. g. p. 82.
Note 2. – It is interesting to note the differences in rendering the term “humanities” in Italian (“scienze umane,” “lettere”), French (“sciences humaines,” “lettres”), and German (“Geisteswissenshcaften”). Only Spanish uses, next to “letras,” a term analogous to that in English: “humanidades”.

Critique of Archaeological Reason6. Critical reviews