Actor Network Theory and Sleep

Sarah Foley
Aug 23, 2017 · 12 min read

Latour’s Actor Network Theory (ANT) is the preconceived interactions that happen between humans and nonhumans actors. These interactions can restrict or enable reactions, scripting how we as humans or our nonhuman counterparts act in order to move this networked system forward. These networked systems are flat and never-ending networks that outline the infrastructure of everyday life.

Latour saw systems as not only technological but also in need of human actants to allow the system to function and work seamlessly. He felt that having either a purely social or technological network was lacking and did not take into consideration the co-dependencies artifacts and humans had with on each other. Latour states “The choice is simple: either we alternate between two absurdities, or we redistribute actantional roles.” (Brey 2005,72) In ANT, both nonhuman and human actors are equal players, so much so that they are indistinguishable in the system and one is reliant on the other for the network to work. An example of this would be a sleep mobile application, the app would not work unless the person who downloaded the app interacted with and entered data about when they are going to sleep and when they woke up. The app is useless without the data. Or if you look at the example given by Latour of the gun. If you shoot a gun, it is not the gun that shoots but the shooter. The gun cannot go off without human actors. Or is the gun providing a scrip to the shooter to shoot. One actor is reliant on the other for the action to take place, but both have equal agency.

In Verbeek’s book, ‘What things do’, Verbeek builds up to ANT by looking at how human and artifacts relate to one another from different theological viewpoints, and slowly he builds the foundation for how we should look at ANT. He makes the case through looking at phenomenology and post phenomenology that things only ‘are’ because of the role of the human. In that, artifacts are given meaning by the relationship human beings have to them (Verbeek, 2005), likewise at the same time (post phenomenology) the objects are in return also shaping us, we co-constitute each other (Verbeek, 2005, 112). Verbeek talks about Merleau-Ponty’s description of artifacts as mediating “the relations between human beings and the world.” (Verbeek, 2005, 108) Or the perspective of ‘Technological intentionality’ states “Things, therefore, are not neutral “intermediaries” between humans and world, but mediators: they actively mediate this relation” (Verbeek, 2005,123).

Latour takes this mediation and says human and nonhumans (artifacts) mutually influence each other. Both are equal and constitute each other and the relationships between actors are scripts, prescriptions, programs of action and antiprograms. These dictate the next actions to be done in the network. It doesn’t matter where these come from just that they are, and kick with certain triggers.

While both human and nonhuman actants were considered co-reliant and equal parties in a system, Latour recognized the durability and reliability that nonhumans had over humans. Humans are unreliable, fickle, fallible, and need more tending to than their nonhuman counterpart. Non-humans keep going until they break, and thus according to Latour, they do a better job and are better at performing a set task. So one must outsource as many functional tasks to these nonhuman actants (i.e. the door closer)

Products interact in this system by their affordances and their scripted use or purpose. A well designed nonhuman actant understands what antiprograms are at play and uses its affordances or functionality to still have the desired effect on the human actant. (Brey 2005, 74) This agency to influence humans is only given because of the context and its place in the larger system it resides. (Brey 2005, 62)

For example, a bed is only a bed because of the person who inhabits it, otherwise, it is a decorated mattress. An unmade bed is just a mattress. Sheets outside the bed system are only large pieces of fabric. It isn’t until you make the bed, and use it, does it the system become a bed. Each artifact on the bed has a different prescription that makes the ‘bed system’ work and move forward. Each artifacts prescription tells the human how to act and calls forth programs specific to that network. For example, the prescription a pillow tells you that you should lay on the bed in this orientation. The prescription of the sheets confines the human as to call up a program of action that tells the human that it is ok to relax to the point of sleep. A blanket’s plays with a human’s body temperature which allows their body temperature to rise and then fall which reinforces the program of action and feeling of sleep. These combined scripts lull the human actor to sleep.

While ANT networks are flat and never ending, you can apply ANT down to two or three actants. A system can include a single object and script[SF1] , such as a body pillow a pregnant women wrap around themselves so they don’t sleep on their stomach, the body pillow physically restricts them rolling over.

Actor Network Theory becoming Social Practice Theory

ANT is an analysis of a stabilized system. Therefore, it does not take into account varying actions of humans. The system stabilized when the system moved forward. Human actors interact with the system in this way, therefore it is this specific interaction that makes up and is included in the network. ANT is a record of a system after the fact and doesn’t view people as subjective, just as they were. It has a one-sided view of the human, and operates as if we are all the same, with the same intentions and reactions. In analysis of the system, it is irrelevant if humans or nonhumans would ever deviate from the role they play in the system, because they have already acted in their predetermined specific way.

Differing beliefs and values in human actors could change the course of the network, if different human actors were to take part in future circumstances. But since ANT is a retrospective of what happens, one only looks at this circumstance, the strength of the prescription and whether or not the script was followed or an antiprogram was used. If one were to consider beliefs to influence actions, the human actor’s belief is a program that was triggered by the prescription, if the prescription was strong enough it used this program to get the desired response. And the Human actor followed the script.

When I am writing this, it is unclear to me when stabilization takes place. If ANT can only be an analysis of a system ‘post hoc’ (Brey 2005,71) that means that it is just a record of what happened. But when you start comparing ANT with Social Practice Theory (SPT), stabilization seems to happen more over time. This action/reaction is the action/reaction that this actor always has when going through this networked system. For example, A mechanical door closer is always going to close the door, it will never waiver, it has stabilized. It happens in this way every time this system is run. It stabilized the first time the system ran. There is no doubt that the door will close, and since it is always the same, it has become a practice because people no longer are conscious of the action. A human actor, on the other hand, most likely will react differently, even if only slightly every time a system is run, it takes time for the system to stabilize. If it is a person closing the door instead of the mechanical hinge, the person did close the door, and is meant to close the door, but sometimes might not close the door. ANT is only concerned with one specific instance when that human actor acted, and that act stabilized the network. For SPT the human actor’s act of closing the door may not have yet stabilized enough to become a practice. ANT is not concerned with practices, but analyzing a system through an ANT lens, it is a natural extension to question when exactly is a system stabilized.

When considering what an antiprogram involves, it is a good idea to use the example given by Latour. He describes the weighted room key, explaining the action of the nonhuman key is always the same. it is weighted and burdensome, and the antiprogram or reaction the human actor has is never variable. This heavy key is supposed to be a burden and the person who is carrying this key is supposed to want to give it back to the concierge. The human actor always returns the key to the appropriate person, they never decide to just throw away the key or keep the key, as that is outside the determined system. The script is weak. There is some sort of social conditioning that states you should give the key back to the person who originally gave it to you, this in ANT is considered a ‘program of action’ triggered by the inconvenience the weighted key provides. The program is to give the key back, and the weight is just a reminder of that program. Even the social conditioning that one feels obliged to give back the key instead of tossing it is a program recalled by the trigger of having to carry around the annoying weighted room key. ANT is not about differences in a system, it assumes one human, the human that has just interacted with the system. Questions about social conditioning and differing beliefs are irrelevant in an ANT analysis. Networks according to Brey, can only be described after the fact as they need to stabilize first. (Brey 2005, 71) The human that has just interacted with the system and the result of whatever happened is the stabilized version of the system.

If you take sleep as an example, it is hard to not also consider practices, and social conditioning, as sleep is mainly a social practice. It is reductive to look at sleep purely only from the relationship between humans and nonhuman actors and the scripts that are prescribed through material affordances. A human’s actions come from programs created by your history and exposure to a specific social construction of sleep. This program is the result of social conditioning that ‘one sleeps on a bed’. Technically, you can fall asleep anywhere, but more likely than not, humans choose to sleep on a bed because of this program and a bed’s strong prescription. This is because you have slept in a bed most of your life, you have grown up sleeping in a bed, and many have been rewarded with ‘a big boy/girl bed’ when you grow out of the smaller one. It is a prize, you put meaning into the bed, and this meaning and ritual of every night preparation for bed is part of what supports and strengthens the program that allows you to sleep best when you are used to something and are within the limits of what is comfortable. Programs come from social conditioning and come from repeat practices and reinforcement. But what is comfortable is technically an emotion, which makes me ask if programs are provoked emotions?

Social Practice Theory and Sleep

SPT is the phenomena a human has when interacting in an ANT system. SPT focuses on the experience of the human more so than the nonhuman actors. The variability in our actions are of interest, as are the intricacies of whether a human slept on their side or on their back, and why they did so matters in SPT. The infrastructure is set up by ANT, but SPT is how we experience chunks of the flat network. These chunks are considered practices. Practices can have overlapping experiences. The beginning of a practice is the entry point into this flat ANT network and the end of the practice is the exit. The friction of the overlap is what sets the boundaries of the practice, when there are no boundaries, one practice is affording the other.[1] For example a bedtime ritual could be considered a practice, but it encompasses many smaller practices, completion of these smaller practices allow one’s body to increasingly allow itself to relax which leads to sleep. For example, getting ready for bed can include taking a shower, washing your face, brushing your teeth, getting into your pajamas, followed by one then getting into bed and reading until one feels like they are ready for sleep. Each smaller practice confirms the larger, and builds up to the next.

When trying to understand what a practice is, it is defined as anything a human does daily that they do without much thought associated to the action, I.e. brushing one’s teeth, or tying your shoes. It is a learned action and can be accomplished when on auto pilot. Practices differ from ANT networks in that there are ideas around how practices ‘may be carried out and what that might mean’. (Julier, 2007, 44) It is looking at the ANT’s programs of action and asking ‘why’, and looking primarily at its meaning. Shove states that practices are beyond just what people do, and maybe beyond what people do on autopilot, she uses Schatzki’s definition that a practice is a performance, to be learned and repeated, and that a practice is shaped by meanings, knowledge, norms and materials (Shove, 2007, 13). Wakkery et al define a practice as “embodied patterns of behaviors and ways of understanding, knowing how, and desiring.” (Wakkary, 2013, 3) Basically it is the ‘why’ to scripts and programs.

Practices need to repeated over and over until actions are stabilized. Practices are also based on inertia and maintaining normalcy. The way people sleep, and structure their sleep is based on normalcy of their surrounding culture. Practices are reinforced through product ecologies as SPT can never be applied to single object. In western culture, the practice of sleep is reinforced through access to a mattress, sheets and a pillow. Sleep would be contingent on this product ecology. If someone was missing a mattress, but has sheets and a pillow it would be hard to sleep if that is different than what they are used to using. (Julier, 2007, 44) Not having a mattress deviates from the product ecology they are used to in terms of their practice of sleep.

Since practices are taught to an individual usually by another in the same culture, practices are unique to time and location, they can differ from culture to culture. For example, in Western cultures such as America, one sleeps at night, usually only in a bed, and there is a ritual around preparing for sleep. In Japan, there is a ritual similar to that the one Americans have, but sleep is not just reserved for the bed and at night. The Japanese idea of ‘Inumeri’ extends sleep into their commute to work or during the day at work. Japanese can often be seen train commuting to work. The practice is normalized in Japanese culture allowing those who want to participate to do so without shame.

Conclusion

SPT is an extension of ANT. ANT is the preconceived interactions that happen between humans and nonhumans actors. These interactions are flat, do not vary and are designed to be repeatable. While ANT’s concern is “existence precedes essence” (Verbeek, 2005, 149). SPT’s concern is more the essence, the meaning a person takes from experiencing part of the Actor network. SPT focuses on the experience of the human more so than the nonhuman actors. The variability in our actions are of interest, as are the intricacies of how a human performed a task, and why they performed the task. These are the issues that matters in SPT. SPT is concerned with how people experience chunks of the flat network called ANT.

ANT is interesting because it brings up the notion that ‘design takes away will’[2]. If an artifact is ‘well’ designed such as the weighted key chain, the human actor has no recourse but to return the key. In a way this is saying that ‘good’ design is a step away from social engineering. By using the materiality of nonhuman actors, Designers can subtlety influence human actors in the network and these human actors will be none the wiser to the intent. When explaining this to my mom however, she saw this as just understanding why people do what they do, which to me is every designers quest. To understand and provide.

SPT on the other hand tries to explore the meaning behind practices. In understanding this, I think is where the creative opportunities lie, and the value of the designers is created. In understand meaning, one could orchestrate something else that could provide just as much meaning to the user as the prior. Or Ward puts it, “practices, rather than individual desires… creates wants” (Shove 2007, 25). It is inherently consumeristic, but is also a way of understanding the human.

Bibliography

Brey, P. 2005. ‘Artifacts as Social Agents’ in Inside the Politics of Technology. Agency and Normativity in the Co-Production of Technology and Society (ed. H. Harbers), Amsterdam University Press, 61–84.

Verbeek, Peter-Paul (1970) What Things Do (trans. Crease, Robert P. 2005) The Pennsylvania State University Press University Park, Pennsylvania

Cameron Tonkenwise, class notes, September 18, 2016, October 4, 2016.

Julier, Guy (2007) Design Practice within a Theory of Practice. Design principles and practices: An international journal, Volume 1, Number 2, Leeds Metropolitan University

Shove, Elisabeth. Watson, Matthew. Hand, Martin. Ingram, Jack (2007) The Design of Everyday Life, Berg, New York

Wakkary, Ron. Desjardins, Audrey. Hauser, Sabrina. Maestri, Leah (2013) A Sustainable Design Fiction: Green Practices, Transactions on Computer-Human Interaction, ACM TOCHI

[1] Cameron Tonkenwise, Class Notes, October 4, 2016

[2] Cameron Tonkenwise, Class notes, September 16 2016

[SF1]Single object and single human

)
Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade