Adam Curtis’s All Watched Over By Machines of Loving Grace (part 1, part 2, part 3) is pretty excellent. It puts forward an ambitious and interesting thesis, which I think deserves more engagement from the anti-authoritarian left than this rather defensive response at New Left Project. To try and compress Curtis’s already over compressed argument into one thesis, he identifies the idea of a self-regulating homeostasis as a widely accepted common sense of our times, and one which makes it difficult for us to think about changing the world, either about what such a change would mean or what the role of power would be in accomplishing such a change. That New Left Project response is right to point out other traditions which influence the anti-authoritarian left and have more to say about power and radical change, but this doesn’t negate what I think Curtis is trying to do. The ideological assemblage he puts together has a certain coherence, but I don’t think it’s supposed to be exhaustive, I don’t think he’s denying that there are other elements which could be assembled in other ways.
This does, though, raise a problem with the documentary, and indeed with Curtis’s work more generally. I think he’s doing this kind of Foucaldian tracing of discourses, but I’m basically guessing, because he’s not very explicit about what he is doing. There are various things about the way the program is put together that imply certain things about the epistemology, although they’re also rather contradictory. Curtis’s signature method, the construction of a documentary largely from archive footage some distantly, some closely related to the point being made, emphasizes the intellectual configuration being constructed is partial. In particular, building the program around juxtaposition tends to push against interpreting the relationships between the elements as causal, which of course is emphasized by the jumps in time throughout the program.
However the soundtrack pushes in the other direction. The ominous music that frequently plays under apparently innocuous scenes keys us to expect bad consequences, and so imbues the program with a teleology, in which the negative consequences are already present in potential form at the origin of an idea. This is what Nietzsche calls “a perverse type of genealogical hypothesis of a genuinely English style” in which everything is explained by reference to an essence lying in its origins, rather than by appealing to something “first brought in under a specific set of conditions and always as something incidental, something additional” (On the Genealogy of Morals).
It’s the punctual and incidental aspects, not the teleological, that I think make the program worth our engagement. I do, though, have one concern about the intellectual collection that Curtis assembles, as I think he may be missing some distinctions in the way various concepts change over time. Specifically, he may subsume too much under the idea of the “machine.” Isn’t there quite a difference between the mechanism of industrial machinery, the circuits of electrified machines, and the information flows of networked machines and genes? And this difference would correspond to a distinction between the cybernetic systems with which Curtis begins, and the bioinformatic ones with which he ends (I think this might understood in terms of a move from Parson’s cybernetic systems theory to Luhmann’s autopoesis, though I don’t know enough about Luhmann to be sure).
Curtis emphasizes the role of feedback in cybernetics, but he doesn’t mention that this was interpreted as making cybernetics the science of control, something which was very much of a piece with the technocratic interventionism of 1950s politics. The association of feedback systems with a form of self-regulation that eludes control comes later; at one point, Curtis briefly mentioned the move from mainframes to networked personal computers, which is a mark of this change, from self-regulating systems which are centralized and basically comprehensible, and so controllable, to systems which are self-regulating because they are so complex they elude our grasp. In this, the failure of ecological systems theories would be a further step in the development of our contemporary homeostatic “common sense,” and not, as Curtis seems to suggest, a scientific refutation of it. (“Black ecology” would then be a further development of the same theme.)
This does not mean that Curtis’s argument is wrong, just that it could be enriched by making some further distinctions. In particular, I think this move from understanding systems in terms of control to understanding them in terms of complexity helps to think about the political implications of the story Curtis is telling. The problem Curtis identifies have, I think, a lot in common with Jodi’s criticisms of left enmeshment in communicative capitalism. It’s important, then, to recognize how close those of us on the anti-hierarchical left are to some of the ideas Curtis identifies as problematic (how could we not be, as they are ideas which really do structure much of our time), as well as ways we have of interpreting and using these ideas differently.
(The title of this post is from the poem which gave Curtis his title, “All Watched Over by Machines of Loving Grace” by Richard Brautigan.)