XXX
Guidance for readings, 081008: Latour’s immutable mobiles and centres of calculation
Last week we expanded our concept of mind machines from the human mind and its computer models to include sociocultural cognitive systems. To get there, we read some papers on distributed cognition, which analyzes cognition in terms of activities engaged in by humans but which do not happen just inside their minds. These activities happen in the world, and by engaging in them people exercise many different skills in their performances of cognitive tasks. They also reveal a wide range of interactions with a wide variety of things, whether these be other people or material objects.
The important point about sociocultural cognitive systems is that they can have cognitive properties quite distinct from the cognitive properties of the people who perform the specific tasks which, once coordinated, comprise the system. This is a very important point, one which Hutchins emphasizes several times in the papers we read. His study of “cognition in the wild” shows that the sociocultural cognitive system of ship navigation aboard a military helicopter carrier carries out the complicated task of navigating the ship into the San Diego harbour. No single individual could do that. This feat of navigation, accomplished by the sociocultural, cognitive, navigation system can be modeled by a universal Turing machine, but that machine does not model any of the cognitive tasks performed by the humans in the system. This is an important result, and it justifies extending our concept of mind machines to include such systems.
This week, and for the following two weeks, we focus on yet another extension of the concept of mind machines. I’ve already mentioned the importance of the things involved in the tasks humans perform in distributed, sociocultural cognitive systems. We’ll now take a closer look at these things. They are all over the place, and some are so close to us, so ordinary, that we take them for granted and therefore fail to appreciate their role in our performance of cognitive tasks. Hutchins reminds us that these things are part of our made environment: we make the environment in which we exercise our cognitive powers. Because we think with these things, because we use them to think, we can extend our concept of mind machines to include them too.
Latour’s topic is how scientific knowledge is produced. His work is relevant to the course, and especially to this point in the course, because he does not think it is correct to suppose that scientific knowledge is the product of individual minds. Instead, it is the product of many different activities, distributed across a very wide terrain, but coordinated by quite specific associations between both human and non-human actors. Because it’s the whole network that produces the knowledge, the network can be analyzed as a distributed sociocultural cognitive system: a mind machine. But what’s important for this week is Latour’s focus on the very important role of specific kinds of things we find in the distributed cognitive systems of scientific knowledge production.
In his paper “Visualization and cognition: drawing things together”, Latour investigates a very common and ubiquitous kind of made object: writings, inscriptions, documents, and illustrations. He explains their cognitive functions in terms of two central concepts: their mobility and their immutability. In other words, you can move them around and they stay the same (they are stable; they don’t easily deform or degrade). They are made to be mobile and immutable. He calls them “immutable mobiles”.
Latour’s paper is rich in examples. He mobilizes a wealth of studies to support his claims about how immutable mobiles function as things that help bring about new cognitive powers and the production of new scientific knowledge. When we engage with them in our cognitive practices, they function as objects we think with. You may want to pursue some of the areas of study he presents in your own research paper. For this class, select some that interest you, that you find either compelling or questionable. Select some claims Latour makes about the cases he presents, whether for your analytical paper or to discuss in class. There’s much to choose from here: a veritable smorgasbord of cases!
Latour ends his paper by introducing another concept, but whose development he leaves for another paper: centres of calculation. We learn what these centres are in the second paper for this week (“Centres of calculation”, chapter 6 of his book Science in action: How to follow scientists and engineers through society). It explains in more detail how these immutable mobiles operate in the production of scientific knowledge.
Here’s a list of the concepts you need to understand what Latour is saying about centres of calculation. You can base your analytical paper on any one of these, or on another concept you find especially interesting.
The first is the cycle of accumulation. It’s explained by a more detailed account of the voyage of Lapérouse than we encountered in the first paper. Latour explains how, under favourable conditions, the inscriptions collected by Lapérouse become part of a network that has cognitive properties insofar as it plays a role in the production of scientific knowledge. We often think, he says, that when we read such accounts (and some are written in a fashion designed to make us think this) that when European “explorers” encountered local peoples we see a meeting between “civilized” people and “savages”. The differences between them, we sometimes think, can be expressed as differences in cognitive abilities. Latour denies this: he says the differences are actually differences in the “angle, direction and scale of the observer’s movement” (p. 217).
What does this mean? It means that if we want to know the difference between Lapérouse and the Chinese he met, we need to describe his movement relative to them in detail: Where does he come from? What is he doing there? What does he do when he gets there? How long does he stay? What does he carry back, and why? How does he help make return visits easier? To what networks are he and his movements connected, and how strong are they?
None of these questions are adequately answered without paying attention to Lapérouse’s gathering of inscriptions. That activity is a large part of its point and purpose. He brings back home resources that allow people to see a distant thing and then travel back to the source and bring back more such resources. These resources are accumulated in a centre. If Lapérouse makes possible further movements that build on his, a cycle of accumulation is established. This cycle has cognitive properties. Why? Because as Latour defines it, knowledge can be described only by considering this cycle: “how to bring things back to a place for someone to see it for the first time so that others might be sent again to bring other things back” (p. 220).
But in order to understand the cycle of accumulation — hence to understand knowledge (and cognition) — one has to understand a wide variety of elements that have to be brought into coordination. This brings us to our second main concept: the conditions that allow a cycle of accumulation to take place (hence to understand knowledge and cognition). Latour explains those conditions by reference to Lapérouse’s voyage on pp. 221-222; they include designs for and construction and functioning of ships, navigation by the sun and stars, new instruments, training in their use, and handbooks on techniques, among others. Our voyage of understanding a cycle of accumulation leads us through, as Latour puts it, technoscience, which is an important element of technoculture (which gets us back to what this course is about).
Once a cycle of accumulation is established, let’s ask: what is accumulated at its centre? Is it knowledge? Power? Profit? Capital? Latour says it is none of these, but instead the resources to bring to the centre specific, unfamiliar events, people, and places. But what kind of resources are they? They are wrings, inscriptions, documents, illustrations, and objects. And here we encounter our third concept: the properties that account for the success of these resources. They must be mobile, immutable, and combinable (p. 223). Latour gives some examples of domains in which these properties succeed in “making dominance at a distance feasible”: cartography; collections; probes; observatories; enquiries (you may want to pursue some of these domains in your research paper).
The next concept for this reading guideline is a big one: the management of the immutable, combinable mobiles at the centres of calculation. It’s big because it sweeps in many other concepts. Let’s stick just to the basics, and use the term “documents” to refer to what is collected at the centres. So our new question becomes: How are all these documents managed?
Latour’s answer is that they are managed through the production of more documents! How is this possible? It’s possible because the new documents are meta-documents: they condense, summarize, reduce, transform, and collect other documents. A cascade of document levels or degrees or orders of inscriptions is therefore set up in a centre of calculation. Latour says that when you hold an nth-order inscription, you thereby hold the inscriptions at the lower orders that the nth-order inscription refers to, summarizes, etc. The higher order translates the lower order and in that way, it increases the mobility of the lower order inscriptions by pumping them up the chain and across a wider terrain.
Latour’s concept of how documents are managed in such centres materialize the concept of what it is to think abstractly, which is our next concept. For Latour, abstract thought is not something happening in the head, but something much more mundane, but very powerful: “abstract thinking” refers to the centre’s activities of handling higher-order inscriptions; thus handling nth-order inscriptions is thinking more abstractly than handling n-1 order inscriptions.
Theory is our next concept. For Latour, “theory” refers to the order of inscriptions that allows the centre “to mobilize, manipulate, combine, rewrite, and tie together all the traces obtained through the ever-extending networks” p. (241-242). A theory should not be severed from what it is a theory of, i.e. from its lower-order inscriptions; the word does not refer to some sort of abstract, mental entity. Mathematics and other formalisms should also be understood in terms of their place as very high-level inscriptions with a very high mobility, capable of travelling across multiple domains.
We’re almost done; the next concept is the application of science to the world. How is that done? Latour says it’s a matter of extending the network from the centre to the periphery. It is not a matter of connecting two disparate realms, one conceptual, or “cognitive” and another “natural” or “material”. Extending the network in this way involves very difficult, practical labour. Sometimes there are resistances and ruptures to the work of network extension; actors often refuse to be enrolled. In such cases, and Latour gives an example (pp. 248-249), the “theory” cannot be “applied” after all.
One of the ways of extending networks from the centre to the periphery is through metrology (our final concept!). “Metrology” refers to the creation of standards; standard time, standard units of weight, temperature, measurement, and so on. Latour calls this “the gigantic enterprise to make of the outside a world inside which facts and machines can survive”. Networks can be extended from the centres to the peripheries because “scientists build their enlightened networks by giving the outside the same paper form as that of their instruments inside” (p. 251). The intrusion of metrological chains into our lives is well described on p. 252, where Latour’s work reminds us of Hutchins’s notion of our made cognitive environment.
I hope these guidelines help you through these two texts. If you are writing an analytical paper on them, you have until Friday midnight (10 October) to benefit from them.