After having dealt with blackboxing and time it is time for me to introduce a second concept in my dissertation. So, fresh of the shelf (or from the document in my computer rather), here comes a section on how blackboxes are assembled, and what such processes entail. The discussion is fueled by Harman's concept of actualism, and somewhat contradictory this takes me closer to a sort of "pure actualism" rather than a sloppy one that used to reside in my thinking. Sometimes, criticism is made so greatly, that what is criticized becomes clearer, and then perhaps even more seductive to embrace.
Concept II - assemblage
If blackboxing animated a modus operandi that sought to open up anything that superficially only appeared as inputs and outputs; taken-for-granted equipment of everyday research that upon closer inspection revealed more components, more black boxes, seeming to point towards infinity, then I need sort of a 'counter-concept' to understand how things hold together as they do. It is thus time to introduce in more detail a second concept, that of assemblages, or the verb to assemble.
First and foremost, a brief excursion to ontology. The second part of Latour's The Pasteurization of France} (Latour 1993 ), Irreductions, contains the philosohical points of departures for (at least the Latourian verison of) actor-network theory. It is written in the style of a philosophical treatise, just like Wittgenstein's Tractatus Logico-Philosophicus or Spinoza's Ethics, which perhaps makes it difficult to transpose directly to an empirical procedure. However, Latour's dedication to Science- and Technology Studies is present throughout the entire text, and I shall argue that it makes important contributions as a Vorhanden empirical equipment. The two first axioms of Irreductions are:
1.1.1 Nothing is, by itself, either reducible or irreducible to anything else.
1.1.2 There are only trials of strength, of weakness. Or more simply, there are only trials. This is my point of departure: a verb, "to try." (Latour 1993: 158)
This first passage travels along the lines of the processes of blackboxing that I described earlier. As for black boxes, they do not have essences or inherent properties, they simply lead to other black boxes, they are not reducible to their components, only associated, linked and intertwined with them. To make them work, they have to be "tried out" or put to trial. Not until then are they possible to utilize to get work done, and perhaps even to forget that they are even there.
This, however, only leads in one direction: a recursive loop, where the complexity increases the more you zoom into it. How does it hold together?
18.104.22.168 There are neither wholes nor parts. Neither is there harmony, composition, integration, or system. How something holds together is determined on the field of battle, for no one agrees who should obey and who command, who should be a part and who the whole. (Latour 1993: 164)
The Latourian ontology is completely flat, underdetermined and even, as Harman argues, actualist (Harman 2009 :16). The closure of a black box, or the holding together of an assemblage of black boxes, follows no underlying logic, no virtual or abstract principle; only what is at stake in a given empirical, concrete situation determines the contingent edges of the boxes making science possible.
This actualism in Latour is on a collision course with several other approaches, both in philosophy in general, and in Science- and Technology Studies in particular. If we begin with the latter, the sociology of science as presented in chapter 1 is disqualified with the principle of irreduction, on the grounds that the holding together of scientific knowledges can not have social causes to begin with. That would be ascribing the closure of blackboxing with causes that are of a particular kind (1), leading to an incomplete description of closure. Instead, any notion of "social", according to Latour's essay One More Turn after the Social Turn, is to be studied, not presupposed:
The same could be said of us, the so-called actor-network theorists. We extend the principle of symmetry to social sciences and we claim that they, too, are part of our problem, not of our solution. (Latour 1992b: 275).
Only when science in action and society in the making were studied simultaneously did this essential phenomenon [common production] become observable. (Latour 1992b: 282)
When studying the natural sciences, which has been the main objective for Science- and Technology Studies, the notion of "society" or "social causes" constitutes one pole in the continuum between nature and society, the two poles that make up what Latour calls "the modern constitution", separated as distinct transcendent domains of reality. This separation is however, according to Latour in We Have Never Been Modern (Latour 1993), merely the result of what he calls a "purification", a constant struggle to put on the one hand natural phenomena and natural laws on one side, and society and its social forces on the other. Purification has led modernity into a dead-lock, where politics i supposed take place only in the domain of "society", and is removed from "nature".
The modern constitution short-circuits the way we think about science, since the purification process only leaves us with two distinct alternatives: Either we explain society with nature, or the other way around; we explain nature with society (Latour 1992b: 277). The third position, which I mentioned already in chapter 1, to explain society with society, follows along the similar lines as explaining nature with society. It is however neglected by Latour, since his main interest is the natural sciences throughout his works (2). My goal is here to show that black boxes of a hybrid character also are present in the social sciences, and that they contribute to the purification of a "society" pole in the modern constitution as much as the natural sciences do. There is however one big difference. If the natural sciences create and purify objects that contains "no society" in the modern constitution; a microbe, an atom or a planet observed by a telescope which are "pure" from politics, values and society; the social sciences on the contrary stand before a positive task; to make the society pole full of social objects, to populate it with human existence.
To study the social sciences, the amodern philosophy of Latour's then needs to be taken into serious account. The black boxes that I keep opening up; response rates, questionnaires, public opinions, measurement scales; need to be treated as hybrid objects (or hybrid networks, Latour 1993: 11), that both function in the work of translation (convincing, negotiating, mobilizing, transporting) and in the service of purification (constructing "society" as a transcendent pole from where "social phenomena" can be explained).
The recognition of hybrids suggests a very different methodological approach to scientific statements. To clarify this difference, I will introduce a straw man analyst: The statement "45 per cent of the population want to lower migration quotas", as exemplified earlier, is a fact that circulates in the news media, and is produced by the SOM-institute. Now, a critical analyst could take this statement, and several like just like it, and argue that this presents a 'shallow picture of society and human desires', 'fails to recognize the deeper meanings of xenophobia', or 'introduces a mechanistic quantitative description of a complex society'. Even if such criticism would be justified, it does nothing more than questioning the style of reasoning which is already there. To challenge ready-made science in such a way fails to recognize the workins of hybrid networks.
Instead, following the trajectories of assembled black boxes, puts me on a wholly different vector. Departing in a press conference, then moving to a media report, I go further to a SOM-report and in this report I find a much more complex assembly of new boxes, each of them containing even more boxes, some of them stable enough only to be mentioned briefly (such as the "simple random sample"), and some of them described in detail with the "hood left open" for the reader to scrutinize and evaluate the content (as with the series of post-cards, telephone reminders and questionnaires).
Stalking black boxes makes the account of what the SOM-institute are doing much richer than measuring the ready-made results up against a pre-defined framework of reference. As these components are taken out of the box, they are no longer confined to the specific event of the survey. They are still however related, in the sense that they can progressively be made to share the same goal.
But how do they hold together? In Pandora's Hope Latour introduces one level above the black box, punctualization, which designates "routine use", where the inner components of a black box are not even visible (Latour 1999:184). The more we open up the boxes, however, the larger the crowd becomes, and the more difficult it turns out to confine each component to one particular use. After all, we can use the same questionnaire for another survey, at another time. We may use the same private contractor to order a different type of field work, the simple random sample is so mobile that it can be used even outside the social sciences.
The same researchers that are included in the SOM-survey for a few years, may pursue another career move and leave. What needs to be done is to study how these components are translated, negotiated, combined, or, as I will call it: how they are assembled. Social scientific assemblages have looked very differently throughout history, with specific problems and possibilities. To get a more thorough picture of why these assemblages became important in modern societies, why this tedious practice of making surveys was worthwhile, I need to go into some previous research on the social sciences. (Which will be the topic for the next post, where I will prepare for a defense of actualism)
(1) This debate is often simplified for rhetorical reasons. As far as I can tell, no sociologist of science has ever claimed that social causes are the only causes that bring about scientific knowledge. Rather the opposite, key figures in the social studies of science have repeatedly stressed that social factors are composite factors, not determinate ones. As David Bloor programmatically claims "Naturally there will be other types of causes apart from social ones which will co-operate in bringing about belief." (Bloor 1976:4-5).
(2) This is true for his major works in the 1980s and 1990s. Recently, however, Latour has shown a great interest in the sociologist Gabriel de Tarde. See for example Latour & Lépinay (2009).