Grammars of Action
I recently encountered the concept of “Grammars of Action”, introduced in Philip Agre’s 1994 paper, Surveillance and Capture. Despite its age, this paper discusses concepts that are still strikingly relevant, as emphasized in a more recent and more quickly digestible overview from 2020 (Jan Dittrich & Lisa Conrad; Fordes). Even beyond its original context, the “Grammars of Action” concept presents a powerful framework for thinking about building and using technology in the modern world.
The materials linked above extensively discuss “Capture” as a model for collaborative computerized work, i.e. “the practice of re-organizing certain activities so that they can be represented by computers in real time” (Fordes). This representation is achieved via implicitly constructed “Grammars of Action”: activities within the organization are broken down into their atomic parts, and then a “grammar” is developed to describe the way these atoms may fit together to achieve high-level goals. The linguistic metaphor here is obvious, and the parallels to programming are fitting as well, as Agre’s paper is “an analysis of the professional practice of computer people” (in his words).
Particularly among computer people, grammars of action can be subject to both the myth of “having been discovered” (vs. created), and “the mythology of transparent representation”, i.e. the belief that the grammar represents the system it describes without any ontological distortion. The Fordes authors dispel these myths:
Rather than being a transparent representation, capturing an activity actually has profound effects on the activity itself: Capture claims to represent, but it actually intervenes. Grammars of action are applied. They involve a re-organization of the activities to make them compatible with computer technologies. The redesigned activity is then encouraged by “social and technical means”, becoming ingrained in the work so that people “… out of necessity, orient their activities toward the capture machinery and its institutional consequences.”
I’ve been thinking about this as a sort of lossy compression: grammars of action can represent only a subset of outcomes within the possibility space, introducing new efficiencies, but potentially distorting the intentionality of each agent within the system. Though we begin with a firm idea of what we want to accomplish, we may be guided toward a modulated outcome. Grammars of action effect an intrinsic value system, simply by making some things more expensive than others (in time, effort, or some other resource).
In Surveillance and Capture, Agre writes with an institutional focus and a cautionary tone, but I think these are also powerfully constructive concepts for the designers of creative tools. Tooling designers must also develop grammars of action to describe the creative processes of their users, and in doing so, they have the opportunity to develop a thoughtfully opaque representation of the possibility space. Many creative professionals will speak to the value of opinionated tools that emphasize best practices, and this is exactly what I mean by an opaque representation: at each step in the creative process, a focused subset of possible next actions directs the user toward their intended outcome. A well aligned creative tool should help to clarify a user’s intentions, confidently shaping their engagement and direction.
Agre’s ideas are inspiring for both my professional and personal work, and I hope they will help me more effectively craft systems and spaces for creative output. With an intentional focus on the grammars I create, I hope to better facilitate particular aesthetics and creative approaches – a sort of proactive curation. I’m excited to keep thinking more deeply about this sort of “meta-creativity”, and to practice related design skills in my continued work on flatpickles.com, Sketchbook, and other projects.
I also want to be more aware of the grammars of action instrumented within the software I use, both for how these grammars capture my engagement, and how they may warp my intentionality. Particularly in capitalized spaces (e.g. social media), my objectives may become subtly distorted to the benefit of the controlling corporations, which might not always be in my best interest. Though these distortions aren’t always discontinuous with my expectations, it feels powerful to recognize them nonetheless.