Humanistic Approaches to the Graphical Expression of Interpretation

The session begins with brief introductory remarks by moderator Kurt Fendt. He points out the need for new tools that will examine data in meaningful ways through aspects of interpretation and visualization. Dean Deborah Fitzgerald emphasizes the importance of support for digital humanities and visualization interpretation as supplemental to textual analysis, and the creation of new forms of scholarly and cultural communication; Peter Donaldson offers a concise welcome to participants.

It is in sharp contrast to a period of enlightenment and empirical science that a re-humanization of digital activities may now take place, says Johanna Drucker. Humanistic approaches are the motif against which she frames her assertion that “interpretation” introduces an epistemological shift—which she identifies by the rubric “from data to capta”—or data that is taken rather than data which is given.

Noting that visualization techniques originally developed for empirical sciences and quantitative analytics lack the sophistication needed by humanists, Drucker emphasizes that humanists must remember their core orientation and approach interpretation not strictly as visualization. Simply introducing more complicated statistical models doesn't solve the problem, although adding some degree of humanistic inquiry does demonstrate that data is "a composite rather than a singularity." But Drucker identifies the deeper and more serious problem as "not simply a matter of modeling humanistic statistics … but shifting the epistemological ground." She sees a serious and substantial role for the humanities "that is cultural as well as intellectual in pushing back against the dominant models of a kind of quantitative and empirical approach."

Drucker advocates a shift from “knowledge” to “knowing” by questioning certain fundamental assumptions about how we know what we know. She and her colleagues have explored this problem by studying temporality and spatiality via the history of time and its representations — timelines. After searching in vain for means to represent time as non-homogenous — probabilistic, potential-laden and discontinuous – she introduced point of view into her models as a way of expressing the subjectivity of temporality – distinctions in traditional assumptions about time.

Creating graphical representations of humanistic data – that is, data inflected with affect – is not data, it's 'capta.' "Capta," she explains, "suggests that all quantification, parameterization, representation is always about an experiential, co-dependent relationship of emergent phenomena. The phenomena don't exist outside of the cognitive perception and the perception is intervening in and influenced by the phenomena."

Ultimately, Drucker suggests, the idea of humanistic experience should be re-centered at the core of the interpretive model. "One of the things we must do," she explains, "is to replace perspective, the human scale, the point of view, the situatedness of human experience within this social and cultural order and it's representation in visual and graphical form."

Kurt Fendt and respondents Amber Frid-Jimenez and Nick Monfort pose questions of their own about the potential for the systemization of visual systems, the difficulty of approaching large, communal data sets, and distinctions between exploration and presentation in the humanities.

Questions from the audience focus on the desirability of certainty within a humanistic construct, and emerging ways in which to represent emergent phenomena through visualization. Drucker favors the dynamic modeling techniques used in meteorology, because of its potential for layered complexity.