I’m grateful to Barry Brown for his comments on my short Interactions piece, “After Interaction”.
Barry, as always, you’ve forced me to think more carefully about my meanderings. Indeed, my intention was to append a short reply to your comment, but your questions have demanded more and, predictably, words have got the better of me. This post, then, is my long-winded response. Thank you for giving me the chance to expand on my thoughts.
First, let me respond to your criticisms regarding the interminglings of humans and nonhumans. As I understand it, you are opposed to the idea of a kind of symmetry between the two. I concede, symmetry here raises problems, as does the implication that people and things populate the same single category. With these problems, I realise I need to make my position clearer.
Finding my inspiration in (post) ANT, feminist technoscience and, as Mol now likes to call it, a , you are right to point out that I see the human-nonhuman binary as a peculiar one. However, I see the symmetry trick to be far from, well, a trick. On the contrary, to me it feels a much more genuine and responsive starting point. Let’s consider the category problem. For starters, when would you and I imagine ourselves to share a category? Well, one rather macabre place might be on the pathologist’s bench. Another might be as one of the many millions of commuters passing through London. In both, we are — in different senses — human bodies. We can, though, imagine just as many situations in which we would be lumped into different categories – ethnicity, geography, intellectual auspices, and so on. Likewise, we could repeat this exercise with things: tables and chairs are items of furniture, but at the same time they of course can be categorised, differently, in that they reference particular styles/periods, or are made of distinct materials. Indeed, this category making could, dare we imagine it, lump things and people together: door-men and self-closing door hinges keep cold weather out, police and road humps slow traffic, etc. So the symmetry here is not one that presupposes categorical sameness, or indeed any essential categories — be they the body or mind, or people vs. things.
My obvious point here (and I apologise for belabouring it) is that categories, , are enacted. Chuck Goodwin provides us with such delightful examples of this. I particularly like his description of . So why then would we presuppose that one very particular category distinction – that between humans and nonhumans — should prevail above all others? Surely, we would want to be open minded about the ways categories are done and not to approach any phenomena insisting that one binary must be enforced?
Let me, then, take up another of your related objections and through this return to my piece in interactions and my modestly pieced together point on relationality. You ask whether it is helpful to attribute agency and normative qualities to things – productive in the short term, you suggest, yet eventually leading us into a quagmire of nonsense and confusion. Can I ask you this: in what way is human agency independent of *things*? I find it hard to think of a context in which agency resides entirely in/with humans. A more honest perspective, as I see it, is to see this more like a spectrum in which agency surfaces somewhere amongst a relational configuration of actors/agents. I slow the car down because I am moving in a car, and together we (me and the car) are obliged to respond to the material arrangement of the road, the road furniture (as it is called), and the road hump. True, remove the human and the intention to slow down is gone, but so too when the car or hump are gone. The intention is in the mixture of humans and non-humans. And your power station gives us another helpful example. Why would we want to imagine that a morality and set of accompanying activities (like investing in nuclear energy) are in someway separate to the things themselves? Without nuclear power stations, and all that ’stuff’ that goes into them, where would we find the morality. Of course, we would not. Agencies and normativities arise out of relational entanglements.
Alas, I must agree with you though. These are longstanding arguments, and I’m confident I won’t be the one who satisfactorily answers your objections. Yet I hope to have shown that the proposal is not to simply shift agential capacities from one side of the human-nonhuman binary to the other. Nor is it to blindly lump humans and nonhumans into the same category. I draw on a project that contests any such binaries or essentialist categories, and instead invites a serious examination of how worlds of humans and things are enacted.
It’s this point on relationality that brings us back to the heart of the matter, a rethinking of interactions. As a matter of fact, I’m not all that concerned with the word, as it is generally used, and I, like you, greatly admire sociologists such as Goodwin who show the interactional accomplishments that constitute ordinary scientific business. My concern is with an interaction that narrows its sights on the neatly demarcated interactions humans have with machines, and presupposes that we might easily separate and forget about the relational entanglements that run alongside and interweave with these. What Goodwin so expertly illustrates is how the assembled arrangements of human bodies (plus their talk) and things, occasion ‘worlds’ of materially configured knowing and logic. The rationale and product in science is enacted through the incessant categorising and ordering (i.e. bringing into relation) of people and things. In effect, Goodwin shows us worlds in the making — ‘world making’. As he writes:Rather than sustaining an opposition between the “mental” and the “material” such activity systems seamlessly link phenomena such as the embodied actions of participants, physical tools, language use, work relevant writing practices, etc. into the patterns of coordinated action that make up the lifeworld of a workgroup.()
So through their linking activities his archaeologists are bringing a very particular kind of ‘lifeworld’ or world into play, one enacted and sustained through materially bound activities that keep certain relations stable and others mutable. If you are advocating this kind of attention to interaction then I am all for it.
My basic premise in the Interactions article builds on precisely this shift from a empirical project that relies on ready-made categories and relations, to one that is genuinely about how the relations are enacted and what the enactments mean for the worlds we live in. I argue this matters for technology because in designing and building technical systems we are affording (and indeed ) certain ways of knowing and doing. We are implicated in that world making that Goodwin observes, but in our case we are giving real shape to the instruments and processes that might bring things, practices, knowings, normativities, etc. into being. We are designing vast arrays of Munsell chart-like systems and the processes that give them authority to claim things about and enforce certain orders in the world.
Finally, it’s this recognition of our inevitable participation in performing worlds, even when we valiantly try to resist a priori categories, that I claim we must show a . The narrow concern for how technical things support or mediate an exceptional human endeavour elides the shear diversity of and possibility for different and new figurings — it is to start too late, it . An openness to the unfolding relations — not just a fixation on a few but in all their varieties — invites us to be far more careful about the worlds we live in; we see that we have some part to play (even though it will always be ) and have a great deal at stake in how we want to live, and how we might make things better.
Alex – interesting stuff, and I am not without sympathy for your basic position about considering encounters and assemblies of people and material artefacts within real settings in all their richness (if I have that down correctly). I think however that you need to follow through on your analysis in a couple of places. Firstly, categories and categorisations have certain connections, that entail and restrain certain sorts of relationships. For example, if I am a woman that implies I am also human, necessarily, but don’t treat me like an object because it denies me my humanity. In your 2 examples of ‘human bodies’ it is worth considering that in the first case you have 2 corpses (or cadavers) – which does not imply necessarily that these are ex-living humans (although it is commonly the case) but rather that they had to be living (organisms). Cars and cups, computers and smart phones can only die metaphorically. In the second case I would argue that ‘humans’ or ‘citizens’ or ‘people’ is probably a more relevant category than ‘human bodies’.
But to return to first example – and the idea of ‘living organisms’ is important. If we extend agency to material objects (often created by people, which is important because the way people might be said to ‘create’ babies is rather different than creating a machine) rather than simply considering them to have effects upon us, and effects we respond to and that shape our assemblies of action and interaction what are we saying? It is a special type of agency that isn’t the same as human agency. For me agency would imply animality, and at minimum the ability to make choices and exercise those choices consciously . If there is no conscious control around things like that what sort of agency is there? (And by the way I do believe that animals have some form and degree of agency but would stop at plants because although they have propensities to thrive, and are living, I just don’t think they can be said to exercise choice in any sensible way that relates to our notions of choice). I do agree with you that the human-object binary ontology can lead to all sorts of simplifications but so can your flattening and equating move (the symmetry stuff), because there are real differences. Only humans can worry about the moral implications of nuclear power – animals and the environment can be devastated by it, cities also – but nuclear power plants don’t choose whether they will be built, decide to situate themselves close to a tsunami zone etc. Now, our inability to realise that we may not be able to control nuclear power the way we would hope may mean that Fukushima comes back at us and bites us on the backside, but it didn’t have a choice and it didn’t decide to meltdown. We need to try and clear away conceptual confusions and I’m not quite sure how attributing agency to objects does this.
David, it’s good of you to contribute to this discussion. Thank you.
I like the way you raise some of the finer points around categories and category making. In fact, I think the way that you ‘follow through’ on the two examples of bodies helps a good deal. You show that if we pay close attention to the relations, we can’t in any useful way talk about a simple, preformed category that lumps humans together. There are all sorts of ways that we humans can be categorised and these are always contingent on a sprawling web of relations.
What’s important to me is that you point to the ways these categories come about because of their material presence in the world, not in spite of it. In other words, categories like ‘living’ or ‘death’, or ‘citizen’, are accomplishments that come to mean something and indeed be something when humans are put into relation to things. For instance, we know well from Sudnow that dying is a socially organised affair, contingent on a variety of materially bound medical procedures and institutional arrangements.* You’re right of course to say that cars and cups, computers and smart phones don’t dye in the same way. The implication here is that it is our job (following on from the greats like Sudnow) to understand the phenomenal state of affairs that authorise things to be living or dead, not to assume such categories are in some magical way predetermined.
So let me be clear here, my claim is in no way to suggest that humans and things are, de facto, the same sort of thing (i.e., that they fall into a common stable category). It is that the worlds we encounter are, unavoidably, enacted through unfolding relations between things and people, and that it is hard to ignore that, in these worlds, things have some role to play: they are in a whole range of ways active, lively, volatile, untameable, etc. These qualities don’t permit us to class them as human, but they do give us a pretty good basis for thinking about them as agential.
Now, you raise some criteria you would need met to see agency in objects. Or it sounds like you may be forgiving here and accept there may be ‘a special type of agency’ for things. Nevertheless, for you, human/animal agency demands “animality, and at minimum the ability to make choices and exercise those choices consciously” — a “conscious control” — as well as the ability to worry. I would broadly agree, and say this is a helpful way to think about the capacities we as humans have for being in the world. But, tell me, where do these ideas of agency come from? Aren’t we doing just the kind of category work that Sudnow observes in his morgue (or that Goodwin sees amongst his archeologists)? And, as such, shouldn’t we be subject to precisely the same kind of analytical treatment? We must see here that we are busying ourselves with very particular kinds of relational achievements by separating humans from things. This is not to question whether our claims are wrong or untrue, but to see that we are making cuts that bring with them their “own worlds”. And, just as we would ask of pathology, we would want to know what other kinds of cuts/worlds might be made here. Could there be other ways in which we might separate life and death, or make the divisions between actors?
One last thing. If the Fukushima power plant “didn’t decide to meltdown”, then who did make that conscious decision? I’m not familiar with the details, but my assumption is no one did and certainly we wouldn’t be able to pin the blame on any one person. Such catastrophes are usually attributed to a chorus of events (some in parallel and others in sequence) that must get worked out. The experts and lawyers come to be the arbiters of where liability (as I see it, a kind of agency) lies, and at times (for better or worse), the finger is pointed at a thing. What are we to make of that?
I see my project to be one of thickening accounts like these. I want to be careful about clearing away and tidying, knowing that we can never stand outside of it all to, once and for all, sort things out. My choice would be let the conceptual confusions thrive, let us “stay with the trouble”, and see what possibilities arise; what other worlds we might allow for.
* I like this work for raising similar issues around people who are categorised to be in vegetative and minimally conscious states:
Sudnow, D. (1967). Dead on arrival. Society, 5(1), 36–43.
Sudnow, D. (1967). Passing on: The social organization of dying. Prentice Hall.
Hi Alex,
It’s nice to be involved in a conversation like this – and these are not easy matters to get an uncluttered view on… I needed to try and think about it all overnight!
And I do believe that we have really quite close views on this. One thing I wanted to clarify on the ‘social organisation of dying’ topic is, of course, there is a biological fact, or at least from a mundane human view a brute fact about human status. It may be difficult biologically to define exactly what life is but we don’t need to be doctors to recognise the difference between living and dead. So it is absolutely true that how dying is organised and recognised and dealt with, and what significance it should have and what ceremonies we follow, and of course centrally how we treat the dying person, is fundamentally a set of social and cultural phenomena all situated and embedded in material circumstances. Objects and artefacts are completely implicated in this. However, let’s be careful not to say death and dying are social constructions – more like our responses, activities and practices around death and dying are social constructions. Your point drawing attention to locked-in syndrome is interesting because it reminds us that there are a number of marginal cases where we are posed ontological problems both scientifically and pragmatically, and that the changes in the way we classify people and conditions can have real impacts on how we treat and interact with people, and so for their quality of life.
Back onto objects and agency. I think that I find the use of ‘agency’ in relation to material things (as opposed to e.g. animals) is awkward because I think agency necessarily implies some form of free will and sentience, so if we are to use agency for material things we say it is the ability to have impact upon humans and human practices (and of course be impacted by) without any form of intention or conscious reaction. My difficulty is that this robs agency of what I think of as necessary characteristics. It is a sort of dumb agency, but that would seem to be a contradiction in terms. I take my ontological view on this from considering how agency is used in mundane practice. I would always try to look at the continuities and discontinuities between assemblies of humans, animals, objects etc. in how they interact, work together, react and so forth, and what the role of thought – broadly speaking – might be. I agree that it is vitally important that we don’t overstate the position humans in the equation such that everything else amounts to bit players and playthings but of course the language we use to describe it is our thing so there is a necessary bias straightaway. Do animals have a concept of death? Well they don’t have the concept of death in one form (i.e. they don’t have a language with the word death) but they do seem to have an understanding – we just don’t know how it maps out and fits with ours exactly. Again, plates and apple watches don’t. I think ultimately this is about carefully describing actual material arrangements, what type of categories and distinctions are relevant and in play in those situations, and how those situations unfold. When we spend too much time at a metaphysical level we lose sight of the phenomena in question and try and map things out in too abstracted a fashion.
There is an interesting point here. When we try and deal with these ontological matters in the abstract it may actually be indeed difficult to ultimately choose between positions. We seem to have 2 positions here: (1) that there is a clearly mapped out ontological difference between humans and material objects, and that one of the things it turns on is that humans have agency and objects don’t; and (2) the position that there may be differences between humans and objects but they both have agency (but that this agency may differ in form and nature between the two). An important thing to note is that these are 2 human constructed metaphysical statements. Statement 1 may veer towards an overly human centric conception and 2 towards object animism – and I’m not sure the puzzle or choice can ever be solved in principle. Although I do think that a more extreme version of position 1 has helped get us into our current environmental difficulties — i.e. we have treated the planet more as an object than an organism. One way to try and let the fly out of the bottle is perhaps through Mike Lynch’s idea of ontography – the study of ontological matters in the wild, as they arise in everyday settings. How do ontological matters play out in practice – when and in what ways do categories, attributes and relationships apply, or get applied to people, animals, nature, and material objects and artefacts? When and for what purposes and in what circumstances are equivalences drawn, distinctions made? Are there dynamic or situation specific aspects to ontology (yes I imagine)? Where do we have trouble – like with locked-in syndrome or with advanced information technology or genetic engineering? These issues can be mapped out and given some form of resolution in concrete situations and maybe that is where we should focus our energies – I think that this could well be a point of full agreement.
Ah, constructivism! That is a thorny but important issue to bring up here. I should say straight out that I am not a proponent of social constructivism, in any of its flavours. Inspired by the likes of Haraway and Barad, to my mind constructivisim had its value, but ultimately, it is just the other side of the determinist/reductionist standpoint — it served as a counter balance to old school materialism, but in the end it simply restates the clumsy social-material, subject-object divisions we are dealing with here. Haraway deals with this so neatly in her piece from way back in ’88, which still seems so full of truths. Bellacasa (2011) and Barad (2011) give us some slightly more contemporary points of reference.
So, as for death, my interest is in precisely how we can come to be confident about death, and what it is. I would not dispute that death has some biological/organic basis; this is, indeed, a matter of fact. However, can’t we — shouldn’t we — ask ourselves how we are able to determine, use and talk about these kinds of facts? Surely, it has been a progression of instruments, organisational alignments, expertise, etc. that has given rise to the fact that death is just this sort of thing? The question is not whether these facts are true or not, but how they have come to be true — and in being true what other worlds are being subordinated, overshadowed, cast aside, etc., and what other worlds might we come to, or want to, live in?
I suppose I see these metaphysical meanderings being about getting us back to the phenomena. Yes (Lynch’s shortsightedness aside*), let us get back to how these things are done on the ground, how worlds are made, always already made. What (and who’s) purpose does it serve to keep finding ways to separate things and people? It does exactly as you say and imposes a view from outside (or as Haraway would call it ‘nowhere’). In this light, can I offer a 3rd proposition to the two you state? What if we didn’t worry about the differences between humans and objects at all, and as you say concentrate on the phenomena? It’s not, i., that we should come with some preformed ideas of both, or, ii., that we should see degrees of different sorts agencies, but that in the end we might still imagine some line of separation. Rather, it’s an openness to seeing how worlds are made and recognising it is always the bricolage of actors (of all kinds) that brings them into being. The question is always how, and we must ask that of ourselves too, along with the question how “it could be otherwise” (Woolgar and Lezaun 2013).
* I feel remiss in not taking this on, but it is a big fish/salmon to fry, so I will leave it to others far better qualified to do so. No doubt you have seen articles in the same special issue and it is also worth tracing the follow up discussion here.
Barad, K. M. (2011). Erasers and erasures: Pinch’s unfortunate ‘uncertainty principle’. Social Studies of Science, 41(3), 443–454.
Bellacasa, de la, M. P. (2011). Matters of care in technoscience: Assembling neglected things. Social Studies of Science, 41(1), 85–106.
Haraway, D. (1988). Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective. Feminist Studies, 14(3), 575–599.
Lynch, M. (2013). Ontography: Investigating the production of things, deflating ontology. Social Studies of Science, 43(3), 444–462.
Woolgar, S., & Lezaun, J. (2013). The wrong bin bag: A turn to ontology in science and technology studies? Social Studies of Science, 43(3), 321–340.
Hi Alex, just as I think we approach maybe we drift further apart. I must admit I find Lynch on ontography the most convincing and the only problem I have with him is that he doesn’t want to make his moral/political position clear, and instead he poses a few questions and makes a few points that indicate his stance without pushing through a moral critique of definitions of natural, GM modified, labelling etc. as allowed by current US law. So here is a point where a feminist reading also has it’s foothold, the critique of certain ontological arrangements that e.g. favour men in some way, as being ‘natural’.
It’s weird, I know she has many fans, but I haven’t been able to see what’s so impressive about Barad — I see some interesting things but I don’t think she is correct in her position, ontologically. I need to read Haraway (and some of the others) and get back to you.
Final point, on your third way proposal: I’m totally fine about the idea that we shouldn’t presume an a priori human/material object separation when doing our studies but I do want to point out that overwhelmingly the people we study draw various clear separations — and one of the clearest is around sentience, thinking, morality, emotion, reasoning, agency and so forth. There are borderline cases, and there are special cases when elements of these things, in limited, different capacities are found in objects (a vodoo doll for example), but that does not mean there is no difference in between people and vodoo dolls to those involved in those activities. And these are not the most common cases. In most cases there are some very clear separations.
Ah, well this makes me think we are still in the same ball park. Your ‘only problem’ with Lynch is mine as well, though for me it is not a whimsical oversight on his part, but a very big and important one. It demonstrates a refusal to recognise he is in some way inside all this mess, and very much a part of the sorting and organising of things/people that he would claim to be studying — from where I do not know. Moreover, and relatedly, it shows he fails to recognise that he himself operates in and sustains what Goodwin calls a ‘lifeworld’. He enacts, along with an assortment of tools and procedures, a certain kind of world with its epistemes and ontics, and the normativities that travel with them. It is peculiar to write this out of his otherwise deeply insightful accounts of science and ordinary life. And, I agree too, this is where feminist readings would offer a productive counterpoint. So, perhaps we have some common ground here?
I do find Barad dense in her writing (possibly overly so), but I admit to subscribing to her kind of (agential) realism and her insistence on the possibility of ‘better worlds’. It is worth reading Pinch’s review, if you haven’t, and Barad’s retort. If anything it shows some of the differences between their two sides, and how they choose to engage with the ‘troubles’. I would be very interested to hear your thoughts on Haraway’s 1988 piece as it is in line with your problem with Lynch.
And, I would agree with you about the many settings in which people comfortably reference subject-object positions (for lack of a better phrase) and human-thing binaries. There are certainly no shortage of places, activities, events, etc. in which that idea of being in the world is present. I have two brief things to say about that. One, if we were the kind of anthropologists who spent their time in far flung places, not like ours, would we accept after long enough that their ‘voodoo’ had some truth in it? Perhaps we might, but the point here is that we might be just as interested in how their truth claims (alongside and interleaved with ours) come into being — not whether they would be actually true or not, but how they constitute a kind of life (I like Helen Verran’s work in this realm). Two, what I find in my own empirical work is that people seem to work with multiple registers, for instance, talking about subject-object positions and in the same breath do something that startlingly unravels any stark distinction between the binaries. I’m not inclined to privilege either of these, but instead to wonder how we work with multiplicity and the ways this opens us up to other ways of being.
vodoo — it’s a special form of voodoo 😉
🙂
Thanks Alex and Dave for this discussion — I hadn’t quite expected all his from a Friday afternoon email!
Agreed that categorization is a valuable and perennial concern, I worry though whether it can really be sensibly treated as a singular thing — like representation, its diversity might be too much. Categorizing fish I suspect is quite different from the final prices of the Paris Bourse.
As for the agency discussion —like Dave I was mainly concerned with the different attributes and actions that go alongside those categories. And at times poetic license may lead us to a muddle. Morgan Ames very nice paper on charismatic technology at the århus conference, was the case I was thinking of. I wasn’t convinced that attributing charisma to OLPC helps our understanding (which I thought was a straightforward category mistake), unlike Morgan’s fieldwork which surely does.
To be honest though what I found most interesting in Alex’s original article was the potential to reflecting again upon interaction as a concept. And the potential that such a thing might have for transfusing a little from STS without killing HCI. The conversation analysts have this thing where they emphasize the primacy of *the next turn* in talk for building and understanding of interaction, and I wondered if HCI’s focus on interaction had sort of had a similar affect. It limits the ability to attribute and expand discussion without limit. As famously in three mile island, the interface of the nuclear power station matters. Yet in turn as Alex points out, the interaction is part of all these different things going on around (say) the deployment of city bikes. So I wondered if the term might be a good place to work out some of our confusions.
But thanks again!
Alex — judging by your last post we are closer on these matter, most I agree with but I think you are hard on Mike mainly because he is a proponent of not letting your political affiliations produce an analysis that misrepresents the situation for your own purposes. This can lead you into the opposite problem of ironicizing without making a position clear. Mary Midgley talks about how a similar problem occurred when analytic philosophers approached moral philosophy — good on describing the landscape but disingenious in counting themselves out from having anything productive to say. Can you really not take sides in a discussion of morals, the nature of the topic kind of demands it? And thank you for contributing too Barry. Now I need to have a think about this as well, go and read some of the references, and since it’s Saturday I wish you both a good weekend!
What an incredibly thoughtful and engaged set of comments to a blog post — this flies in the face of all received wisdom about online commenting! Many thanks to all of you for trying to work through some of these complicated issues in a public forum where we may all benefit from it, and especially to Alex, for your original piece in Interactions which sparked it all.
This debate for and against what we might call ‘relational’ or ‘material semiotic’ approaches sprouts periodically in different places, but I found this iteration interesting and helpful for thinking about a couple of the key points of the debate. I should note that I am closer to Alex in thinking than to Barry or David.
The first difference that always arises is over the nature of reality — whether, as Barry argues, people and things, bodies and minds, are really different and it is a dangerous categorical mistake to behave otherwise. This one seems to be an intractable problem, never successfully resolved. To me, Science Studies has shown repeatedly that there are many ways of doing the world, of which the modern Euro-American version of matter set in time and space and a duality of mind and body, is only one, and not necessarily the most successful. Questions of truth and reality are important within this Euro-American frame, but not within many of the others, including the ones mobilised by some of the theorists Alex has been referencing. It is hard to resolve an issue (the nature of a stable reality) which for one side is crucial, and for the other is somewhat irrelevant.
Another problem seems to revolve around agency, and the patent absurdity of assigning human-like agency to objects. I agree with Barry that this has a certain shock value which can be useful in limited cases, but that otherwise if we take things as behaving in human-like ways we will mostly be lead into difficulties. It seems to me that a re-doing of the human/non-human binary requires a simultaneous re-doing of the closely related object/subject binary, and of what it is to act. The sorts of agency that we attribute to humans in a modern Euro-American frame, with their underlying implications of freedom and choice, are clearly not appropriate when the discrete human subject is abandoned as the source of action. (Abrahamsson et al’s recent piece in Environment and Planning D: Society and Space, ‘Living with omega‑3: New materialism and enduring concerns, talks about this.) A different frame needs to mobilise different concepts of how acts happen, and how such acts are distributed among entities.
I would tend to agree with Alex that the term ‘interaction’ is problematic, embedding as it does implications of two separate and prior entities (people and things) which then interact. But then language always embeds within it a set of metaphysical assumptions, in some cases ones we wish to embrace, and in other cases ones that we find problematic.
This led to a nice offline conv with Stuart reeves who was visiting us last week — he mentioned that bob and Wes have a nice unpublished piece which touches on many of the points here.
http://www.sharrockandanderson.co.uk/wp-content/uploads/2013/03/Ethical-Algorithms.pdf
One little last thing is to mention latour’s use of humor. From a continental perspective latour is trying to keep in play a skepticism towardst claims.
I want to come back to ‘interaction’ in the original article: I agree wholeheartedly with the point of the article but I don’t think the more radical take on agency is required to get there (maybe it is just your particular route). It is absolutely clear that a narrow focus on interaction at an interface is an incredibly simplified way of considering things. It has led to people extending the ‘interface’ beyond a screen and input device (as you note) to something that relates to wider socio-material assemblies, societal functioning, whatever. It can also be hard to locate just where an interface might be. Think about high frequency algorithmic trading –algorithms competing with other algorithms that forms the majority of trades in markets. Can we understand it by looking at a terminal screen? Nope. What can we understand about it by understanding traders interactions with technologies for showing something of its operation? I don’t have an answer to that but it’s an interesting question. What do the designers of the algorithms understand about their operation and potential impact? Another good question. There is another interesting thing that comes out of this though – the failure or success of technically instantiated and driven markets that work in this way has real impacts on peoples’ lives such that their agency (freedom of choice) may be curbed, as they suffer wage pressure, etc. These are sort of reverberating impacts rather than proximal or direct ones.
This happens in a far more mundane and direct way, when call centre worker interaction with customers is constrained by scripts embedded on technology. This is motivated ideologically – the idea is that there is a problem if agents are given too much autonomy. Too much autonomy means they are not under control. This is often positioned like it is just about ensuring standardisation, ensuring uniform customer service, protecting against agents going off-script and misrepresenting the company or giving bad advice. However, while there is some truth in this it is also about de-skilling – experienced and knowledgeable agents are expensive. Through scripts and standardisation you sell the dream that the script takes care of the work so the agent does not need to think for themselves. This reduces you labour costs, but in my experience (and I have quite a lot of it in this domain) it generally lowers the quality of service, particularly for complex and idiosyncratic inquiries. It makes call centre work low paid, high surveillance and controlled work, and so in this way technology is rather directly employed to take workers’ agency away (and their sense of self-worth etc.). But people are behind it too. And by the way it is precisely these types of move that gut and shape a service to such an extent – i.e. render it robotic and of poor quality — that the challenge for automation becomes a lot more simple, i.e. outperforming a crap service with a disengaged workforce is a lesser challenge for an automated agent. So while I don’t think technology has agency, it certainly has the potential and the actuality to provide or take away human agency. And I do strongly agree that a narrow focus on interaction prevents people from seeing things of wider importance in terms of impact, and I would even go as for to say that some of this way of thinking is designed precisely as a set of blinkers to stop people thinking about the wider (political) import of what they are doing.
On the subject of ontology – I kind of agree with both Alex and Alison in a general sense – but I think that in making this move there is a danger that you represent a Western view as monolithic when it is very far from this. If we consider religion, faith and spirituality in the broadest and most diverse sense you will see that human/material object dichotomies suddenly break and become a lot more complex, which is one of the very clear points that Winch made in (his sarcastically titled) ‘Understanding A Primitive Society’. And similarly when I read things like Eduardo Viveiros de Castro I think there may be an over-emphasis and magnification aimed at making the strange, stranger and in a way that is too neat. Back to Winch – a key point of his was of Western anthropology doing two things simultaneously – simplifying and exaggerating Western scientific rationality like it was universal in Western life while at the same time painting the tribal culture as exotic, irrational and bizarre, when it was actually pretty understandable and had its own rationality.
Therefore in my book there isn’t a satisfactory ‘ontological’ answer to these types of questions – they need careful working out in relation to particular instances. In many cases, it may well be doing a disservice to raise these endogenous definitions within everyday settings of the conceptions of differences and relations between people and things and their attributes to an ontological level. Can these really be seen as ontological questions for those involved? What would indicate that a question was ontological rather than purely practical (or maybe political)?