Enthusiasm for humanities computing has recently renewed anxiety about whether and how humanities subjects should be represented as computable, quantifiable data. Yet such concerns are relevant to all humanists, not just the digitally-inclined. Whether we work with computational tools or with more traditional methods, we are pressured to hew to a rhetoric that partitions the qualitative and quantitative aspects of our work.

However, this is a fallacy. Rather than an immutable divide, the distinctions between qualitative and quantitative models are more akin to the wave-particle duality of light, a model of coexistence where the frame of either "wave" or "particle" is determined by the perspective and purpose of the researcher within a given instant.

We present experimental visualizations to suggest the the interleaved, fractal-like nature of quantitative assumptions and arguments built into seemingly-qualitative humanities research. No question, not even ones predicated on highly subjective interpretation such as the tracking of intellectual or artistic movements, are independent of a sense of comparative quantity that makes it possible for us to claim a trend is becoming more or less prevalent.

The double-slit diffraction experiment that demonstrates the wave-particle duality of EM radiation. (source)

The following case studies suggest how blurry the qualitative/quantitative boundary truly is in humanities research, and how an iterative process of models, measures, and data can help humanist scholars reckon with the many quantitative assumptions that are interwoven in our work It is this iterative, recursive process that we argue lies at the heart of any practice in humanistic experimental design.

Matthew Lincoln
PhD Candidate
University of Maryland, College Park

Nuria Rodríguez Ortega
Director, Departamento de Historia del Arte
Universidad de Málaga

Parting Ideas
Visual Metaphors: A Colophon

Matthew Lincoln: Measuring Tonal Paintings

Why did Dutch painters turn towards more tonal painting in the 1620s? This deceptively simple question has arisen many times in histories of seventeenth-century Dutch painting, addressed purely as a step in stylistic evolution, to a strategic reaction by artists to changing economic supply and demand.1

But what is tonal painting, anyway? Restricted palette? Thin paint application? Simple composition? What do we mean by "turn"? How many artworks, or how many artists, in how short a span of time, constitutes a significant shift? One might attempt to quantify palette through image processing or systematic pigment analysis. Yet it might be equally productive to rank hue, composition, and paint application by eye, hewing to a simple number scale. Any of these approaches may be aided or hampered by a lack of measurable evidence.

Still life with a silver tazza, Willem Claesz. Heda, 1630, oil on panel, 40 x 55.5 cm, Rijksmuseum, Amsterdam.
In this visualization, questions and concepts are clustered by theme (e.g. what is tonal painting itself, what constitutes a shift, and what the causal factors may be). By mousing over one of these elements, you can highlight the subcomponents (in red) of that element, and the parent components (in green) that, in turn, reference that element. A "starting point" is marked at the top of the diagram in capital lettering, however the user is free to navigate from there.

Rather than treating qualitative and quantitative approaches as oppositional, or even as complementary-but-separate modes of research, we would do better to drop the distinction in favor of a research model that finds value in recursive investigation, pursuing further clarity in the way we model — i.e. the way we explain and remember — our cultural history.

  1. N. R. A. Vroom, A Modest Message as Intimated by the Painters of the Monochrome Banketje, trans. Peter Gidman (Schiedam: Interbook International, 1980); John Michael Montias, “Cost And Value In Seventeenth-Century Dutch Art,” Art History 10, no. 4 (December 1987): 455; Jonathan Irvine Israel, “Adjusting to Hard Times: Dutch Art During Its Period of Crisis and Restructuring (c. 1621 - C. 1645),” Art History 20, no. 3 (September 1997): 449–76.

Nuria Rodríguez Ortega: Measuring Exhibition and Art Market Influences

Spanish art historian Juan Antonio Ramírez (1948-2009) defined the artistic culture as an ecosystem of interconnected institutions: Academy (Art History discipline), Art Critique and Theory, Art-Market, Editorial Industries and Museums.1 Other authors, using different metaphors and/or argumentations, have addressed the same idea even increasing the number of “institutions” involved, such as collecting and displaying practices.2 This approach calls our attention onto the institutional dynamics that shape the development of behaviors, practices and knowledge production processes in the field of artistic cultures, and invite us to explore which are those interconnections and how they evolve. Needless to say that it has been a topic of great interest for the art historian community, and a good number of scholarly texts have undertaken this issue.

The study case proposed in this "provocation" has been conceived as an exercise to look at this issue from other perspective, combining the qualitative questions traditionally addressed in the humanities (which are the cultural/symbolic/social/political values embedded in these interconections and which kind of meaning they produce?) with quantitative parameters based on the potentialities of data processing.

For this study case, only two institutions have been taken into consideration: exhibitions and art-market, although others –such as collecting or art critique- are implicitly involved in the model.

The model is based on the assumption that exhibitions and art market influence each other.3 Starting from this point, the following questions arise: Is that true? In which sense? How can we shape in a multidimensional way the influences between exhibitions and art-market using quantitative parameters?

In this visualization, questions and concepts are clustered by theme (e.g. data about artworks, actors, and events; component questions, and measurement methods). By mousing over one of these elements, you can highlight the subcomponents (in red) of that element, and the parent components (in green) that, in turn, reference that element. A "starting point" is marked at the top of the diagram in capital lettering, however the user is free to navigate from there.

  1. Juan Antonio Ramírez, Ecosistema y explosión de las artes, Madrid: Akal, 2004; Ramírez, El sistema del arte en España, Madrid: Cátedra, 2010.

  2. See, among others, Elizabeth Mansfield, Art History and its institutions: foundations of a discipline, New York and London: Routledge, 2002; Mansfield, Making Art History. A changing discipline and its institutions, New York and London: Routledge, 2007.

  3. "influence" is understood here as the capacity of one component/factor/fact etc. to affect other and provoke some changes.

Parting Ideas

We offer these as guiding points and questions for designing a humanistic experiment that embraces the quantitative/qualitative duality:

  • The model is an argument for how you believe the world (or one small part of it - your object of research) works - encoding your assumptions, hypotheses, and evaluative methods.
  • What are the dimensions/features/characteristics of this model, in all its multiplicity?
  • In what was can you responsibly abstract your objects of research into data?
  • How can you measure these data — even in a very subjective and contingent way — as a way to evaluate the validity of your model?
  • No one set of data, no one measurement, will give you a synthetic or objective view of all the dimensions of your model. Rather, you must experiment iteratively and recursively with multiple models, multiple measurements, and multiple datasets.
  • It is the process of this iteration — not any single result — that tends to be the most valuable scholarly output.