NOTE: This post was prompted by a lively discussion on the PhD Design distribution list. Special thanks to Mike Zender for prompting me to flesh this out..

One of the methodological strategies I have been using for the past year and a half for my study of research through design is to construct an inventory of arguments that I find in philosophical aesthetics, literary and film criticism, and art history and criticism that assert that an artwork (literary, visual, cinematic, whatever) does inquiry, that is, it contributes to knowledge.

In this post, I sketch out one such argument, and then I apply it to research through design, and that is what I will call the Criterial Knowledge Argument for Research Through Design.

The Criterial Knowledge Argument for Literature

First, I sketch the argument as it has been developed in philosophy. Here I will sketch out the view of philosopher John Gibson, from his Fiction and the Weave of Life (Oxford UP, 2007). 

There is a background to Gibson’s argument that is worth briefly sketching, and it is the notion of ordinary language philosophy, initially proposed in late Wittgenstein’s work and further developed by subsequent philosophers, J.L. Austin and Stanley Cavell being the most relevant for Gibson. On this view, human language inscribes within it a wealth of knowledge, fine discriminations, nuances, paradigmic examples, and so forth. One philosophical method is, accordingly, the “grammatical investigation” whereby one seeks to do inquiry by studying how we ordinarily use certain key vocabulary as a means of navigating intellectual confusions. Wittgenstein is famous for demonstrating that philosophical puzzles that we get ourselves hung up on often rest on uses of language that are not ordinary, and that once one returns to the puzzle with more ordinary uses in mind, the problem often dissipates. One common methodological strategy that is asked is, “by what criteria do we recognize X as X?,” that is, in order to call this P an instance of racism (or friendship, romantic love, a mistake, or an excuse), we must have some standard criteria. These criteria precede any of us as individuals (they belong to language and the “forms of life” that uses it), and yet they are plastic enough that they can change over time.

With this background, let us turn to Gibson’s argument. It is intended to defend what he calls the “humanist intuition,” which can be characterized

as the thought–or hope–that literature presents the reader with an intimate and intellectually significant engagement with social and cultural reality. It is the idea … that literature is the textual form to which we turn when we want to read the story of our shared form of life: our moral and emotional, social and sexual–and so on for whatever aspects of life we think literature brings to view–ways of being human (Gibson, p.2, emphasis in original).

The problem is that the way that literature (and by extension, the arts) traditionally have been claimed to do that is mimetically, that is, the idea that the arts reflect or represent reality. Gibson then summarizes the wealth of philosophy and theory that problematizes this view, one of them being that literature represents characters and situations that are made up, so it can’t represent reality at all! And if it can’t even represent reality, how can it provide us any knowledge at all? But it gets even worse than that, as Gibson notes that in literature there is

a conspicuous absence of all those tools, devices, and techniques we commonly take to be essential in the search for truth, knowledge, and like forms of worldly engagement: argumentation, the offering of evidence, the setting forth of ‘the facts’, the proffering of premises, the derivation of conclusions, etc.

Are these not very similar to the skeptical arguments currently used to challenge research through design? So it seems to me that if Gibson can offer a response this skeptic, those of us defending research through design (perhaps in a future incarnation, if not its present incarnation) might also have a nice resource with which to mount our own defense.

So what is Gibson’s response?

First, a disclaimer: this is a blog post summary in a few paragraphs of an entire book of philosophy that assumes its audience is intimately familiar with ordinary language criticism and the Western literary canon. It is simply not possible for me to do justice to Gibson’s argument, so if you have an objection, I encourage you rather than to dismiss all this outright, to raise your objection constructively in the comments, and I will attempt to think through it with you. With that said…

Gibson begins with an example: the unit of measure, 1 meter. He observes that there is no metaphysical Meter that we all have common access to. Rather, it is a standard of representation, a “common cultural instrument with which we can then go on to engage in the building of representations in our world” (63). He extends this reasoning to other examples, including color. Again, there is no metaphysical Color Archive that we all have some magical access to, and yet any of us can find color charts in art stores and home improvement stores. “To make explicit the general point,” he summarizes, “when we examine our standards of representation, we are inspecting things that are obviously real though not representations of anything at all” (64). Again, the “not representations of anything at all” is cueing us that he is seeking a non-mimetic argument to justify literature as knowledge.

Gibson’s next move is to consider other standards of representation that get him closer to literature. He considers “jealousy”:

Jealousy, to give an (overly) simple example, develops conceptually and linguistically as our culture develops institutionally. We develop institutions based on the pledge of fidelity (such as marriage) and, once we have public and so transmittable examples of people betraying those institutions, we can use the behavior of the wounded (Dido of Virgil’s Aeneas, for example) as a standard by which we can, so to speak, go on to represent the world jealously. (66)

Hopefully now the “criterial knowledge argument” is becoming clear. It basically states that literature (or more broadly art) contributes to the criteria by which our standard vocabulary and conceptual system is developed, enriched, and made useful to everyday life over time.

But there is an important caveat:

Criteria do not ‘make it the case’ that the world is really as we say it is or ‘establish the truth’ of what language calls reality. [Jeff adds: if they did, they would be mimetic, and we’ve already dismissed that theory.] They provide the conditions of mutual intelligibility and so any sort of talk at all, truth talk included.

So criteria, in the sense that Gibson is using it here, do not have truth values expressible in terms of the correspondence theory of truth (i.e., mimetic truth). (Note, in contrast, that scientific statements, including its criterial statements, for the most part do, and I think the design community’s failure to assert this distinction partly explains its struggles to respond to scientistic skepticism against research through design.) Rather, criteria ground the possibility of mutual intelligibility, of intersubjectivity, and of social forms of life that depend on shared understandings of “jealousy,” “meters,” and “colors.” And this is how Gibson establishes literature as providing knowledge contributions in a 10-page passage in his book, which I now quote from extensively:

To give an account of our language is in an important respect to give an account of ourselves, of the way in which a sign is given life by what Wittgenstein calls our ‘form of life’, a phrase that brings home the idea that language is not a mere grammatical system but is itself an expression of the our-worldly. In this respect, one of the clearest insights we can aspire to have into the way our world is is achieved by bringing to view the standards of representation, the criteria, our culture has developed in its efforts to forge a shared sense of its world…. (67, emphasis mine)

What I have done is to replace the picture the sceptic exploits with one in which language is seen not as connecting to an independent reality exclusively by representing it but at a more direct level, in its standards of representation, its criteria. (69)

And, if the story of agreement, of how we can come to represent the word commonly, requires the existence of public standards of representations, what could possibly act as our standards for terms and concepts such as these [i.e., love, suffering, exploitation, jealousy, devotion, flourishing, joy]? They are grounded not on ‘things’ or ‘objects’ at all but upon very elaborate visions of human life…. It is the institution of literature that brings to light most ‘perspicuously’, as Wittgenstein might say, how we can distill into a public object and thus make available to a culture generally something as intricate as a vision of life undone by, for example, jealousy and ambition. (71)

When we look at Dostoyevsky’s creation, we are regarding an object that is constitutive of a way in which we can see our world rather than a mimetic duplication of it…. This is what it means to claim that something–some object, some image, some narrative–is a standard of representation. It is to claim that it opens up a way of seeing the world, and all that this implies. (72, emphasis in original).

When I claim of Othello that ‘this is racism’, my ‘this’ has, I suggest, the force of registering what the text speaks to on the criterial level of what racism is. … Iago … becomes our word for racism: so complete is Iago’s expression of racism that we see exposed in his words the criteria for this feature of human reality. (78)

In short, all human social practice and mutual intelligibility rests on the shared resource of language. As human life and practices change, so too does that language to support us as living organisms. Literature is able to (a) extend and/or clarify for us, in our times, in our weave of life, what is meant by, what is at stake our own key vocabulary; (b) widely disseminate perspicuous images so that they become part of the public culture and the grounds of its social practices; (c) propose alternative ways of viewing our complex life-situations. By this means, we can easily see the difference between Proust’s madeleine cake, Virgil’s Dido, Shakespeare’s Iago (Antony, Cleopatra, Hamlet, Falstaff, Lear), Sophocles’ tragic irony on the one hand, and Monty Python’s cleaning lady on the other.

And science, like all other forms of life, depends on criterial knowledge. Imagine a marital counselor who had no access to criterial representations of romantic love, friendship, trust, betrayal, open communication, jealousy, “the other woman,” “the seven year itch,” the powerful attraction of sexual taboo, and on and on. This is not to say that these supersede the technical theory of psychology; that is absurd. It is to say that that technical theory of psychology couldn’t exist in the first place if it didn’t rest on these criterial understandings, and that it is only by doing so that our hypothetical counselor can perceive the relevance of her or his own theory to the verbal and non-verbal accounts produced by the couple on the couch; for this is what they share, how they become mutually intelligible.

The Criterial Knowledge Argument for Research Through Design

It’s hopefully obvious now what I’m going to say next. One argument in support of research through design as a form of inquiry is that it can, in potential, also develop and reveal criterial knowledge in much the same ways that literature does. It would do so by offering perspicuous images of a designed world, which are powerful enough to help the public think otherwise, or to reveal a new way of seeing things to designers, to offer in a single image a means of perceiving and understanding an issue of interest to our everyday life, or to offer an image of the future that motivates people to change what they are doing in order to bring that future into being.

In Bardzell et al. (2015), Shaowen Bardzell, Lone Koefoed Hansen, and I argue that Hiromi Ozaki’s Menstruation Machine does this kind of work. The domain of interest is biotechnology and posthumanism, and the perspicuous image, which is a design, is of a fashionable, functional device that allows its user to experience, at least on a physiological level, but we would argue inevitably more than that, menstruation. It does so by dripping simulated blood and using electrodes to induce cramping.

And this image, this possibility, raises questions about what menstruation has come to mean, about who would want to wear such a thing and why, about how we ought to regard menstruation (helping us see, for example, that dismissing menstruation’s design significance with an “eww” and refusing to apply design thinking to it is a form of patriarchal austerity?), about how designs can reify but also clarify avant-garde philosophical concepts such as Judith Butler’s notion of gender as performance, Haraway’s cyberfeminist notion of the cyborg, cultural discourses on transgender, and feminist HCI, among others.

Any contribution that substantially influences criterial understandings of a community seems to me, prima facie, to be “generalized knowledge.” It is knowledge, because through it we know how to use our language to have and to communicate very sophisticated and subtle thoughts, as we need to in everyday life. It is general, because many people acquire this knowledge and use it in many situations.

Virgil’s Dido has helped Western societies comprehend a life undone by jealousy for over two thousand years. Proust’s madeleine cake was not merely a pleasingly decorative bunch of words; more than that, it proposed, in an elegant and parsimonious vision, a highly relatable theory of human memory and perception, which has had reverberations across literature, philosophy, and neuroscience. Milton’s Satan was written not to provide aesthetic pleasures, but rather to help 17th-century Christians navigate the seductive appeals of rebellion and sin by helping them understand the mechanisms and consequences of that seduction; in our more secular world, Milton’s Satan has since then continued to stand as the paradigm example of the Rebel: “Better to reign in Hell, then serve in Heav’n.” (That Milton’s Satan was the model for Mary Shelley’s Frankenstein, and that the latter was the inspiration for Scooby Doo merely underscores my point: even a 7-year-old’s criterial understandings are, however developmentally primitive, informed by Milton’s paradigmatic work.)

It is easy to see how this is distinct from how natural scientists use the notion of “generalized knowledge.” But again, I am inclined to ask why anyone would take for granted that research through design is epistemically close enough to a natural science that it would inherit its standards, logics, and rhetorics (?!).