Not razors, boxes!

This will pick up on certain things you say in your second-to-last post (on time and simplicity), and black-box the remarks in your last post (on Heidegger). Actually, let me start very simply: what is a "black box," for Latour? Here's what he says in Science in Action:

“The word black box is used by cyberneticians whenever a piece of machinery or a set of commands is too complex. In its place they draw a little box about which they need to know nothing but its input and output.” (Science in Action, 2-3)

So when I say I'm black-boxing what you say about Heidegger, what I mean is: I see that there's an issue about spatialized time and the differences and similarities between Latour and Heidegger, and I see that you ultimately resolve it to your satisfaction by declaring them compatible, both holding that "representation is dependent … on what reality does, rather than what it is--such that what it does is what it is." And I am content to leave it at that, and not engage you on this issue, either to argue or to agree. If I did work through it, at the cost of a great deal of time and mental energy, I would more than likely find things that I deeply disagree with Heidegger about. Or maybe not; maybe I'd become a Heideggerian (in which case I might stop caring about Latour). But in this particular case, I'm happy to just leave the box black, because I'm not a Heideggerian, and I'm satisfied that you feel the difficulties are mostly resolved. From the standpoint of dialectic, this might seem insulting or unfair, but from the standpoint of "reality" as Latour sees it it's just the way things work: not everybody needs to know everything underlying a particular state of affairs in order to accept that state of affairs as compatible with their own interests.

So I think I see what you mean when you say that "Latour is solving the huge dilemma of historicism that characterizes so much humanities research which bases itself on a scientific model." By showing us the way science actually works (as opposed to the way epistemology and the history of science claim it works), he both shows us that the dilemma is false (we don't need to try to arrive at exact, pure, perfect knowledge of history, because the scientists don't get it of nature either) and gives us a new way to work, one premised on simplicity rather than progression or accumulation. Not "Always historicize!" but "Always simplify!" (Latour — or rather, Norbert — actually says this in Aramis, on page 94. But that's next week's text.) And you're right to link this commitment to simplicity to Latour's concept of temporality, his sense of what it is possible to do with the past:

All of time becomes reversible, and there are no longer theories that get more concrete and "settled" because they are "historical" than newer or more "open" ones”

and

we slim down things by making time reversible, or stop seeing time as some process of accumulating--anything at all (here, knowledge or error).

It seems to me that the black box has great possibilities for historicist work. Because it does away, once and for all, with the “archaeological” metaphors that Foucault et al. espouse. The working assumption, for the New Historicists (and also, I think, Jameson), is always that the actors knew more — even if only unconsciously — than we do; and in order to understand them, we need to reconstruct that knowledge. So we’re always playing catch-up; and this is part of why it’s so easy to criticize, even dismiss, historicist work: well, if you just knew a bit more about X then your interpretation would be better. Your failure to take Y into account invalidates your whole argument. You don’t seem to have realized that Z is a condition of everything you say.

But we can say, with Latour: yes, the actors knew lots of things we didn’t, consciously or unconsciously; but much of what they knew took the form of black boxes which they never thought, or perhaps were unable, to open. And here’s where we have an advantage over the actors we study: we can safely open the black boxes, question their contents [the abstracting move you see, rightly, as the hallmark of “Theory”], and make connections and draw conclusions that the historical actors didn’t or couldn't.* We have nothing to lose by this activity, because it’s all already over: we don’t need the black boxes in the same way the actors did. Their projects are not ours, so we’re free to nitpick, reframe, reconsider, theorize. And I don't think Latour's against this at all. You say:

I can make an idea "open" by somehow plucking it out of time and playing with it. It is this that I think people are talking about when they confusedly call theory "ahistorical." Nevertheless, theory is only taking its cue from a notion of modernity defined by the sciences--or so I imagine Latour saying. Regardless, it is allowing the "opening" of "settled" knowledge precisely by grounding ever more thoroughly the notion that there are "open" and "settled" knowledges.

And I agree with you that BL might see much of “theory” — especially that part of it that descends from Canguilhemian epistemology, by way of Althusser, Foucault and Bourdieu — as ironically “taking its cue from a notion of modernity defined by the sciences.” But it’s important, I think, not to paint him as against theory, even the kind of theoretical maneuvers you’re talking about (which I associate, perhaps naively, with Derrida: “mak[ing] an idea "open" by somehow plucking it out of time and playing with it”). After all, he’s a theorist himself! What Latour adds, I think, is a refreshing reminder of the always provisional and local nature of theorization, and the necessity of it taking an object.

But, also, at the same time as he gives us license to theorize, he gives us license not to theorize. (Here he reminds me of Wittgenstein: "The real discovery is the one which enables me to stop doing philosophy when I want to.") Because we can always skip some steps; because we can always leave black boxes black (just not necessarily the same ones as the historical actors did). In fact, we have to do this; if we went into everything, theorized everything, we'd never get anywhere. But it's a relief, this restraint, this simplicity.

And one more thing about Latour's brand of simplicity: it's a proliferative simplicity, because it is — precisely — not a reduction. It is not the boring old medieval logic of Occam's Razor: thou shalt not multiply entities beyond necessity. Quite the contrary: it multiplies entities all over the place, but shuts up most of those entities inside black boxes, until we, the analysts, want to let them out. A way to acknowledge the (Deleuzian?) interconnectedness of everything, while still enabling us to stop doing actor network theory when we want to — what could be better?

But why is this not just warmed-over epistemology — us arrogant post- or even-more-moderns, arriving after all the games are over, playing with the views or beliefs of the ancients (in this case, the modernist ancients) from the standpoint of our superior modernity? In other words, exactly what Latour rails against in We Have Never Been Modern? Because we need to test it against the objects. And this is why, as you correctly say, “the notion rejected … is not that science works with facts, but that there are epistemological breaks, total changes in perspectives that close off not just outdated series of facts, but whole sets of ‘knowledge’-producing outlooks and techniques--all in the name of progress.” There do need to be facts, and there do need to be objects — this is what’s realist about Latour’s method — but there doesn’t need to be progress in our way of viewing them. (This is possibly why Latour's position strikes you as "pro-scientific-(but anti-science? how to describe that)-realism.") We can keep the empiricism of science without the progressivism of the history of science: the moderns’ way of looking without their certainty that they’ve revolutionized everything. Which I think is probably as good a way as any to think about the heritage of historicism, not to mention close reading.

* Some black boxes that literary historical actors have used that we can open: God; technology; race; gender; ideology; “literature”; and, yes, modernity. It’s this last that’s most exciting for my field (modernism/twentieth century poetry): because here, all the actors were so convinced that they were modern/avant-garde/progressive — and much academic criticism of their work has consisted of weighing in on whether they really were or not. (Yes, you were aesthetically modern, but you were politically reactionary! Go to the back of the class!) But if we are free to open the black boxes marked “modernity” and “modernism,” then we can really reread these authors, without being hamstrung by the framework or paradigm or epistème they themselves have brought into being.

"A slimming treatment:" time in WHNBM

At the end of my last post I brought up small groups, and said that Latour, at least in the early work, seemed to privilege them:

Don't we get the impression, reading his work, that the collectives that lend us the most to study are about the size of a lab?

This is actually a bit unfair, as it is put--and I'll try and correct things below. Nevertheless, I see We Have Never Been Modern as Latour's great effort to try and undo such a privilege.

How? In Latour's earlier work the lab was that sort of lens in which the the world is inverted. There, forces which shaped man were allowed to proliferate. This very process of proliferation then produce new shapes (if I can use that wonderful term from "Irreductions"), some of which the scientist allowed (though this implies no domination by any abstract powers) further proliferation. At a certain point, people got it in their heads that there was no scaling going on--that all these relations of force and the generation of shapes didn't exist, and what one was getting in this operation was the truth of something humans had never really approached in the first place, never manipulated over and over, that was never just the large made small.

Now, the lens will be a certain arrangement of this nature-culture which he calls modernity, and that process of scaling the construction of a Great Divide according to the articles of a modern Constitution. All of which makes sense: Latour is generalizing his conclusions even beyond the philosophical generalizations that, in "Irreductions," were still either products of field work or philosophical and personal insight (remember the first "pseudoautobiographical" interlude in "Irreductions" where Latour pulls over his car and sees the sky, and chants the three basic theses of that text: "Nothing can be reduced to anything else, nothing can be deduced from anything else, everything may be allied to everything else."). In other words, he is trying to articulate his viewpoint on the widest empirical grounds (something that is not philosophically possible, and so could not be done in "Irreductions"), so that they no longer have to take their authority from where they originated.

So it is surprising to see the title of the essay deny something that we basically took to be a fact: that we have been modern. Nevertheless, if one understands what Latour is getting at, particularly with respect to time, such a statement makes perfect sense.

Indeed, everything involves knowing that time, for Latour, is merely "the distant consequences of actors" ("Irreductions," 1.2.5.1; 165), or thus a result of the state of the relations of forces between actors:

"Time does not pass. Times are what are at stake between forces" (165; 1.2.5.2).

Or, as he says in We Have Never Been Modern itself:

We have never moved either forward or backward. We have always actively sorted out elements belonging to different times. We can still sort. It is the sorting that makes the times, not the times that make the sorting (76--Latour's emphasis).

It'd be a huge mistake to think that this means time is spatialized for Latour: some of the most significant commonalities between Heidegger and Latour (which we will point out soon) would be thereby erased. However, I could understand the temptation. If, as Latour says, "the connections among beings alone make time" (77), aren't these connections being made in a particular place, and thus a space? Some of Latour's statements come close to saying precisely this:

In other words, everything happens only once, and at one place ("Irreductions," 1.2.1; 162).

But what this means is only that space, like time, is dependent upon what an actant does:

We do not know where an actant is to be found. The definition of its location is a primordial struggle, during which many get lost. We can only say that some locate and others are located (1.2.3.1; 164).

Thus:

Space and time do not frame entelechies. They only become frameworks of description for those actants that have submitted, locally and provisionally, to the hegemony of another (1.2.6;165).

How, then, have we ended up thinking that something just comes before another? Time is being-shaped-by-a-force, as is clear from these quotes. Thus, a particular force, arranging things in vast amounts, would have to impose upon us the framework of the irreversibility of time, since if time is only the result of what actants do, or if "each entitity is an event" (WHNBM, 81), then (or therefore) "time becomes reversible" (73). Such a reversible time is actually quite easy to think, provided that we pay attention to what actants do and see in them the creation of what Latour calls hybrids or the proliferation of quasi-objects:

I may use an electric drill, but I also use a hammer. The former is thirty-five years old, the latter hundreds of thousands. Will you see me as a DIY expert "of contrasts" because I mix up gestures from different times? Would I be an ethnographic curiosity? On the contrary: show me an activity that is homogeneous from the point of view of the modern time. Some of my genes are 500 million years old, others 3 million, others 100,000 years, and my habits range in age from a few days to several thousand years (75).

So whence this irreversibility?

The notion of an irreversible arrow--progress or decadence--stems from an ordering of quasi-objects, whose proliferation the moderns cannot explain. Irreversibility in the course of time is itself due to the transcendence of the sciences and technologies. [...] It is a classificatory device for dissimulating the inadmissible origin of the natural and social entities from the work of mediation. [...] Just as they eliminate the ins and outs of all the hybrids, so the moderns interpret the heterogenous rearrangements as systematic totalities in which everything would hold together. Modernizing progress is thinkable only on the condition that all the elements that are contemporary according to the calendar belong to the same time. For this to be the case, these elements have to form a complete and recognizable cohort. Then, and only then, time forms a continuous and progressive flow, of which the moderns declare themselves the avant-garde and the antimoderns the rearguard while the premoderns are left on the sidline of complete stagnation (73).

In other words, modernism (or modernity) is the result of progress being imposed upon us, which is merely a result of that systematic exclusion of hybrids (here, the hybridization of reversible time, where each "instant" can be next to any other in the way that my genes can be right next to the drill in my hand) that so much of what the book details:

Modernism--like its anti-and post-modern correlaries--was only the provisional result of a selection made by a small number of agents in the name of all (76).

Or, to put it in the terms of "Irreductions" (which I find clearer), the creation of power out of force (through the extension of a network, i.e. by only expanding force, but thereby representing actors). At this point, it should be clear what Latour means by his title, and why he says it: We have never been modern because the relations of force have never produced progress or irreversibility. It's only because we've allowed power to come into the picture (and thus allowed actors to represent others in more than one sense) that we think we are modern (Latour can largely be summed up by the following phrase: "There are not two problems of representation, just one," 143).

I have dwelt on this, however, not only because I want to explain the title. I also think that it is this resistance to the notion of progress that keeps asserting itself, subtly, throughout the book's attempt to account for modernity's production of and disavowal of quasi-objects by separating purification and mediation. Of course--to recap--this production and disavowal takes place because, Latour claims, moderns (or those who think themselves modern) don't see the work of purification as a particular case of mediation--the core argument of the book, I think. This means that moderns adhere to that Constitution which I have already mentioned, or see nature as distinct from culture and proceed to put acts that alter nature on one side of the "Great Divide" and acts that alter society on the other. They don't see that this separation is really only another way of linking nature and culture, or, better, creating more networks of natures-cultures, mediating or translating or passing forces (take your pick among these metaphors). In short, attributing something to a power, keeping the world divided up between culture and nature, rather than trying to look for hybrids, or those things that already are moving between nature and culture (Boyle's air-pump)--this is only allowing us to produce more quasi-objects (and quasi-subjects). If we want to more accurately alter this production, purification should be practiced as what it is---a case of this mediation.

But, back to progress and time. What Latour allows us to think is not only how various hybrids get overlooked--how we don't end up researching Boyle's pump but "classes" and "gentrification" and "extroversion" instead--but also how we confuse the history of science with history ("the history of science should not be confused with history," he says, 93). By this I mean the sort of separation that is created not just between true and false science but a very common practice that is included in this wide-ranging distinction. I mean the sort of division--and Latour acknowledges this throughout his discussion of the symmetrical asymmetries in the chapter on relativism, and, as I'm arguing, implicitly throughout the book--between something like working theories and closed theories. This, of course, goes back to his discussions of the black box in Science in Action. But now everything is shot through with very sophisticated notions of temporality, and a massive argument trying to base things precisely on all practices, not just scientific ones--and so we see how modernity itself involves this black-boxing. For at stake is not just an individual theory, but whole "paradigms," or "epistemes," or (even more precisely), "structures of feeling" (all these notions being similar). In other words, the notion rejected here is not that science works with facts, but that there are epistemological breaks, total changes in perspectives that close off not just outdated series of facts, but whole sets of "knowledge"-producing outlooks and techniques--all in the name of progress (Latour says throughout the book that the one thing modernity can't be is outdated: as we see above, it has to be avant-garde).

This gets discussed perhaps most openly in the beginning of the reconsideration of anthropology in the chapter on relativism, where Latour supplements David Bloor's principle of symmetry. Latour is recalling, in the following quote, the principle's advantages, but importantly--for us--starts by paraphrasing Canguilhem:

Only what breaks for ever with ideology is [for Canguilhem] scientific. It is difficult indeed to pursue the ins and outs of quasi-objects while following such a principle. Once they have passed into the hands of such epistemologists, they will be pulled out by the roots. Objects alone will remain, excised from the entire network that gave them meaning. [...] For such epistemologists, "Whiggish" history is not a mistake to be overcome but a duty to be carried out with the utmost rigor. The history of science should not be confused with history. The false is what makes the true stand out. What Racine did for the Sun King under the lofty name of the historian, Canguilhem does for Darwin under the equally usurped label of historian of science.

The principle of symmetry, on the contrary, reestablished historicity, and--we might as well say it--elementary justice. David Bloor is Canguilhem's opposite number, just as Serres is Bachelard's. "The only pure myth is the idea of a science devoid of all myth," writes the latter as he breaks with epistemology. [...] For Serres, as for actual historians of science, Diderot, Darwin, Malthus and Spencer have to be explained according to the same principles and the same causes; if you want to account for the belief in flying saucers, make sure your explanations can be used, symmetrically, for black holes. [...] If you claim to debunk parapsychology, can you use the same factors for psychology? [...] If you analyze Pasteur's successes, do the same terms allow you to account for his failures? (93).

Now, let's keep this in our heads, and quickly recall that in my last post, I had something to say about your statement in your post on macro- and micro- actors that for Latour,

the macro-actor is only a network of "devious" micro-actors, and that they are in many ways simpler and easier to study than micro-actors.

There was a general suspicion on my part of the emphasis on the micro- and the local. I might here correcting one mistake that I made in this characterization: I didn't emphasize that this emphasis on the local is what I found odd, not the local itself, since Latour himself doesn't think that anything is either strictly local or strictly global (I forgot whole sections of WHNBM that critiques this notion: see p. 4.9-4.12). In other words, I find odd the way the word local gets used even after Latour's critique of the local/global is registered (see the use of "locally and provisionally" in "Irreductions" 1.2.6 above). Now, recalling all this (and stating it all again more accurately), let's also recall that despite all this, taking your statement in these large terms, I emphatically agreed that

it is really to his credit that Latour is the only one pointing out how, if we don't reduce things, our explanation might be simpler.

Now, given these remarks about the principle of symmetry, and the general effort to reject any sort of notion of epistemological break, with all of Latour's notions about time still in mind--I think I can elaborate on how another form of simplicity is generated by the Latourian analysis.

For Latour continues, after he asks about those errors of Pasteur that he documented so well:

Above all, the first principle of symmetry proposes a slimming treatment for the explanations of errors. It had become so easy to account for deviation! Society, beliefs, ideology, symbols, the unconscious, madness--everything was so readily available that explanations were becoming obese. But truths? When we lost our facile recourse to epistemological breaks, we soon realized, we who study the sciences, that most of our explanations were not much. Asymmetry organized them all, and simply added insult to injury. Everything changes if the staunch discipline of the principle of symmetry forces us to retain only the causes that could serve both truth and falsehood, belief and knowledge, science and parascience. Those who had weighed the winners with one scale and the losers with another, while shouting "vae victis!" (woe to the vanquished), like Brennus, made discrepancy incomprehensible until now. When the balance of symmetry is reestablished with precision, the discrepancy that allows us to understand why some win and others lose stands out all the more sharply (93-4).

Of course, Latour will go on to complicate this--he calls for another principle of symmetry to go beyond the first (and not only explain truth and falsehood in social terms, but in natural-social terms). But I'm generally saying the sort of critique of modern temporality is here in ovum, and indeed cuts through the ways that modernity is able to multiply explanations rather than simplify them.

For what happens is that all those black boxes come undone, as I said above. All of time becomes reversible, and there are no longer theories that get more concrete and "settled" because they are "historical" than newer or more "open" ones (since what is new is not always open). In short, Latour is solving the huge dillema of historicism that characterizes so much humanities research which bases itself on a scientific model. If I want to write about the 18th century (and I do), I surely can't talk about sentiment as if it still had some relevance. And if I did, it'd be awkward. Theory (in literary studies) has largely been the way out of this problem, by getting behind this awkwardness (and fetishizing it as "crisis"--cf. de Man): I can make an idea "open" by somehow plucking it out of time and playing with it. It is this that I think people are talking about when they confusedly call theory "ahistorical." Nevertheless, theory is only taking its cue from a notion of modernity defined by the sciences--or so I imagine Latour saying. Regardless, it is allowing the "opening" of "settled" knowledge precisely by grounding ever more thoroughly the notion that there are "open" and "settled" knowledges.

This we can see by the other historicists doing the same thing (as you nicely described in your "New Historicism" post), but by multiplying their explanations as to why the "settled" knowledge became settled.

What I'm trying to show is that both procedures are done away with if we stop that sort of black-boxing which is the modern look at temporality: instead of looking for ideology (Feyerabend looks perhaps the most foolish from a Latourian point of view), or really trying to either settle or open the knowledge, we slim down things by making time reversible, or stop seeing time as some process of accumulating--anything at all (here, knowledge or error).

Another (less temporal) way of putting this is seeing only one form of representation taking place, as we hinted above. There are no representations "inside" the objects that we are looking at. When it comes to art, this gets complicated: there is no inside to the work. We can therefore do history of the book right alongside character analysis, just in the same way that we look at the scientific "facts" as scaling practices that occur in/through the lab. And it is really this that is the main thesis of We Have Never Been Modern: any reconsideration of time and history will involve conceiving of the "sorting" which produces time differently, and in such a way that there is only one sort of representation (the relations of forces). This means, then, that we have to see our networks as made of collectives, not just as nature on the one hand and society on the other.

The resulting simplification is then huge: if we admit that purification is mediation, instead of critiquing endlessly (and Latour ends the book by saying that it is in the interest of critique simply to keep going, never amending the Constitution), we will find simpler connections between actors--for they will no longer appear as powers, but in all sorts of hybrids right in front of our faces. I've tried to show, however, the less visible and perhaps more complicated way in which this also involves historical actors, supplementing your post on New Historicism, which rightly stressed the production of new historical actors--here I just try to show when they then are... which is not in any modern time.

On the issue of groups, and doing more justice to Latour on this issue--I've done a bit here to correct that but I'll come back to it more directly (via Heidegger, who I didn't get to here) in another post... I fear this general presentation of WHNBM has gone on too long!

Latour and New Historicism

I take your point that, in my first post, I ignored what is surely the most obvious thing to strike almost any reader of Guerre et Paix des microbes: that it discusses nonhuman actors (the microbes) on the same plane, and using the same analytic terms, as it does human ones (the Pasteurians, the hygienists, the physicians, et al). Partly it may be that I’m already used to this bizarre trademark move of Latour’s from his other texts, and I was in fact struck by the fact that, in G&P, he mostly does talk about human agents, with the great exception of the microbes. Leaving aside the possibility that he just hadn’t fully developed his theory of actants yet, I think Latour stays away from multiplying nonhuman actors in this book because he wants to make a strong case for how novel it was for Pasteur to introduce the microbes into political discussions in late nineteenth-century France. In other words, if he focused on how many other actants there are that somebody might have introduced, he would fail to make as strong a case for the effectiveness of the Pasteurian strategy.

(And by the way, I absolutely agree that “this perfect case of Pasteur and his microbes sometimes ends up passing over into an allegory for Latour's irreductions.” There are plenty of times when terms BL uses to discuss Pasteur could equally well apply to himself. E.g. this passage, on Pasteur’s movements: “Wherever we expect him to pursue the development of a science in which he will have some success, Pasteur chooses not to pursue this fundamental research but to step sideways in order to confront some difficult problem that interests more people than the one he has just abandoned… [But at the same time] he was faithful to a single problem, that of distinguishing the agents involved” (69); and a page later, when he speaks of “the transversal strategy that seemed to become ever more imperious as he reached the end of his course … to work on the whole of society” (70). The parallels with Latour's own course through the academic disciplines should speak for itself.)

If you don’t mind, I’d like to delay the discussion of Latour’s role vis-à-vis sociology — and also vis-à-vis Marxism which, I’d like to remind us, is historically partly coextensive with sociology: i.e. there have been many Marxian sociologists, and Marx is frequently taught in sociology departments — until we have a little more evidence; certainly by the time we get to Reassembling the Social I think his views w/r/t various different sociological schools will be clear. I promise I’m not dodging your question, which is actually very close to my own “interests” — just buying myself some time.

Instead, I’ll pick up on your fruitful parenthesis about Latour’s use of narratives — even fictional ones — and the New Historicism. You’re quite right that Latour is “trying to admit new sorts of groups” into sociology, although one could equally say that he’s trying to disqualify others, or at least ignore them (he’s noticeably uninterested in classes, for instance). You say “it's only a narrative that will recover these other ‘groups’” and, a little later (and even better) “we narrativize to reactivate certain groups.” This is exactly right, and what’s going on in Pasteurization does seem very close at times to the New Historicism (or, more accurately probably, to Foucault, whose work lies behind all New Historicist practice). You said that “these new groups are people like the hygienists, who disappear from the accounts of power, the narratives of conflict.” First, a slight caveat: according to Latour, the hygienists weren’t missing from “the accounts of power,” they were recorded as consequences of Pasteur’s discovery: Pasteur discovers microbes, and society is changed by the gradual repercussions of this epochal discovery (what Latour elsewhere criticizes as the “diffusion” model). So Latour is not adding anything to the record, the way New Historicists do, he’s just adjusting relations between objects, granting more autonomy to the hygienists than some previous historians have: “Where would the hygienist movement have gone without Pasteur and his followers? In its own direction. Without the microbe, without vaccine, even without the doctrine of contagion or the variation in virulence, everything that was done could have been done: cleaning up the towns; digging drains; demanding running water, light, air, and heat” (23).

But even if Latour were talking about some group whose role in the Pasteur story had been totally overlooked, I would say he still operates differently from the New Historicists. If the New Historicists (and not only them) would say Let’s recover forgotten actors, then Latour could say, Let’s invent or discover new actors/actants [is there a difference between these? I’m still not sure I’ve got a handle on the distinction] and see if we can extend the networks that we already master by doing so. Even if the narrativizing practices are basically the same, it seems to me that the gesture is still different. Latour is at once less “political” (less interested in politics, or ideology) than the NHs, and more political, in that he is actively constructing new polities of things whereas the NHs just look for more political explanations of what we already know existed. Again, whereas the NHs' supposedly radical recoveries show how things were more firmly fixed than we thought, how literature or culture has been determined from the outset by ideology or power relations, Latour's speculations about networks actually change things a lot more, but they're also more fragile and contingent: easier to argue with. (One of the most effective, but also most frustrating, aspects of grand Foucauldian concepts like epistème, dispositif, etc., is how difficult they are to alter or contradict.)

This difference is totally consonant with what I wanted to say about Latour’s swerve away from traditional sociology: the New Historicists are in fact more “sociological” than Latour in their desire to dig up new “social facts” that previous archaeologists had somehow missed, to make the historical record fuller and more factual than it used to be. Fuller, but less interested (and often, I'd venture to say, less interesting). To me, it seems like giving interests to more actors is precisely what historians and sociologists don’t do: rather, they believe there are just a few interests (economic, neurological, or cultural) and that these determine how all the smaller actors act. So when you say “we see that the sort of move away from ‘interest’ that you describe so well is done precisely to move away from a position that would have to give ‘interests’ to things,” I think we may actually be interpreting differently: I read BL as saying that things, like people, do have interests, in that they start from positions of inequivalence and are translated by other things (such as, but not only, humans). Again, I’m not sure I’m with you when you say that

perhaps we can see the move away from "interest" that you describe as a consequence, not of some opposition to interests held in common, but as an opposition to giving microbes "interests" rather than what they really need — reality.

Because “interests,” for Latour, are “reality” (or, at least, are real): they’re what lie between actors at punctiform instants, making them want to move. So I don’t see a “move away from ‘interest’” at all: rather, I see an extension of it, to things as well as people. The important feature, though, is that, as I quoted last time, “common interests are in the long term necessarily divergent” (65), and all the more so, I would think, when you extend the notions of “community” and “politics” to encompass nature and technology as well as human beings.

I feel like I might confuse the issue by going on, but I just want to make clear that I don't think Latour, at least in this early work, does away with the sociological understanding of the world as "interests." On the contrary, it actually seems to be one of the few sociological concepts he wants to keep (see his disparaging remarks, passim, on "legitimacy," "prestige," "power," etc.). What he doesn't like about conflict sociology isn't the emphasis on interests, but the insistence that these interests are only of a few kinds, and that actors always seek to destroy, rather than enlist, those with different interests. That's my sense of things right now, at any rate: I'm sure this will all become much clearer as we read on.