Digital Ontologies as Productive Process

From the Series: Digital Ontology

Photo by Gerd Altmann.

Something has happened to ontology; ontology is not what it used to be. One English dictionary defines ontology as the study of the fundamental foundations of being. Questioning ontologies has thus traditionally meant positing a question about the essences of things. This, in itself, is a metaphysical and political gesture. To say what something is in its essence means to define what it is not: a process of exclusion that corresponds to seizing power. Maybe that is why the generation of 1968 seemed to view questions about ontology with animosity. Developing a counterphilosophy of difference and becoming was their political project of resistance and liberation.

When I pose a question about the ontology of the digital, or the ontology of the digital subject—my specific interest—I am inspired by such a history of thought, because the ontology I am drawn to is not the ontology of truth, but one of making things: the ontologies of design, engineering, and art. Jimmy Lin (2015, 34) suggests that the criticism of correlational analysis typical of big data is based on the erroneous idea that big data can do “better science” and is, therefore, about truths. Lin instead proposes that the “end goal of big data use is to engineer computational artifacts that are more effective according to well-defined metrics.” In a sense, it is about better engineering.

To ask what digital ontology is, therefore, is to ask how a digital subject or object is made and produced in informational infrastructures. Where the digital and algorithmic is concerned, the ontological question of “what it is” becomes the question of “how it works.” This is even more important when these digital computational infrastructures become central modes of organizing, performing, enacting, cognizing, and making sense of things. What I see in the question of digital ontology is the presumption of a variety of ontologies as structures of algorithmic logic, antagonisms within computational milieus as well as practices and cultures of living with computation. A digital ontology is a pragmaticontology: the questions to ask of it are about how something is made; which models, frameworks, and figures of enactment it recruits or is forced to adopt in its becoming; and what powers it exudes.

Some of these questions may resemble those of science and technology studies (STS), which asks what a physicist’s formula is or what the scientific conclusions drawn from laboratory data are. In a representationalist paradigm, such theory or data would be understood as a representation of how the world really is (a paradigm that seems to have been overcome in performative STS). Yet this paradigm is still employed by scholars when they consider data gathered around a person’s activity in virtual and real life alike as a “data double” (Raley 2013). The concept of a data double or data shadow, arguably, implies a representationalist explanation of what and how the person really is.

Like scientific truths produced in a laboratory, such a double is not only made; it also makes. It is performative and enactive. The attention paid to the evolution of fact-evidence-document-information-data (Day 2014) in information science is instructive here, as are questions about how models are used to analyze data, predict future patterns, and enact sociocultural worlds. A data double is pulled together by forces outside of the data itself; indexicality comes from elsewhere, and its production is different from the laboratory practice of science and its regimes of objectivity (Rouvroy 2013). I suggest that taking into account these forces outside of data requires us to consider the role of such elements as ideology, hacking, and art in the production of computational ontology.

Google’s Deep Dream image interpretation. Image courtesy of Eric Wayne,

Computational materiality, which is grounded in abstraction, and the ontologies within which these abstractions are formed—that is, those of formal logic, modeling, and algorithm design—have a specificity of their own. Ontological specificity does not imply technoessentialism. These digital ontologies are antagonistic and nested, where intensive specificities of the varieties of the computational follow the unfolding of their own embryonically concentrated forces, even as they enter into a wide variety of relations. They are both specific and hybrid. Understanding how the digital is made and makes thus requires recognition of its specificity and its operational hybridity.


Day, Ronald E. 2014. Indexing It All: The Subject in the Age of Documentation, Information, and Data. Cambridge, Mass.: MIT Press.

Lin, Jimmy. 2015. “On Building Better Mousetraps and Understanding the Human Condition: Reflections on Big Data in the Social Sciences.” Annals of the American Academy of Political and Social Science 659, no. 1: 33–47.

Raley, Rita. 2013. “Dataveillance and Countervailance.” In “Raw Data” is an Oxymoron, edited by Lisa Gitelman, 121–45. Cambridge, Mass.: MIT Press.

Rouvroy, Antoinette. 2013. “The End(s) of Critique: Data Behaviourism versus Due Process.” In Privacy, Due Process, and the Computational Turn: Philosophers of Law Meet Philosophers of Technology, edited by Mireille Hildebrandt and Katja de Vries, 143–68. London: Routledge.