Subset of Theoretical Practice

Pages
HomeBlogLinksAbout STPContact Us
Our Work
Atlas of Experimental PoliticsWorking Through Political OrganizationContribution to the Critique of Political Organization
Syndicate *
RSS /Atom

A dialogue on sensors, types and organizational composition

22 December 2022

X: How about this fella studying SENSORS categorically?

Learned of him in that "Sheaf Theory Through Examples" book.

Y: Yes! I have a book by this Ghrist guy, called "Elementary Applied Topology", which is really cool. And he has a great set of lectures on calculus on YouTube, with crazy graphics - actually he is considered one of the founders of topological data analysis.

X: Ahh i remember you mentioning him at some point!

It just seemed the perfect fit for our "organizational trinitarianism" to deal with literature on sensors categorically - and perhaps connect it to those weird articles on the categorical formalization of scientific experiments we found a while back?

Y: Yeah, good point! The unfortunate part of this area of research is that the military loves it, it's the technology that powers drones and surveillance stuff. I think there's definitely something useful for us though, if only to think the modern form of sensibility of the State.

X: I think we could be satisfied with at least a general take on what we mean by a sensor, categorically. The idea that "multiple sensors compose in such a way that some object is partially seen by them and then their perspectives can be glued together, etc"... that's something we are already investigating, for sure. But it seems that the question "what is means to sense something?" is very much the point of contact between 'types' and 'mereology', which we've been looking for.... We know the answer to that question would informally be something like "it means to be composed in such a way that interactions with the sensed thing can be typed in certain ways (as changes of this or that quality)" so thinking formally about sensors - not how to glue them, but how to define one formally, connecting composition, intelligibility and interaction - seems very much at the core of how to advance our theory of organizations as "sensible social organs".

Y: Yeah, I think that's a big question - somehow the "type" of phenomena and "type" of sensor must be compatible. In fact, I wonder if the multiplicity principle can help us here - isn't the fact that a single type can have multiple decompositions relevant for describing how sensors and phenomena can actually have totally different compositions but be able to interact?

Because we don't want to simply say that the sensor and the phenomena are the same, a political organization that seeks to eradicate poverty might not only be composed of poor people.

X: Yes. Hm.

I mean, the sort of toy situation we should be able to account for, to begin with, is, i think, some less ridiculous version of that scene i love from the movie "John dies at the end" (where a girl with phantom limb syndrome opens a door with a "phantom doorknob")... how do we formalize this fact, that a physical hand can open a physical door but a ghost door requires a phantom limb?

The interaction of "hand <--> doorhandle" is present in both, but there is this qualification of the type of hand and of doorhandle - "physical" or "ghostly" - that limits things (and which specifies something about the composition of the hand, rather than the action it performs...even if the action is limited by that substrate, etc).

Y: Yeah, the ghost door and ghost hand is the most clear (and bonkers) example.

X: Hahahah yes! I mean, we could also say "a physical person can open a physical door but only a fictional person can open a fictional door". Or that it takes an eletromagnetic field to interact with another, while a mechanical system only interacts mechanically (to keep things within more science-y level).

Even in Badiouian philosophy, this could also lead to a useful distinction between procedures: only art can change the art world just as it takes political processes to produce political effects, etc - there is a compositional condition on the type of interaction even in the truth-procedures (even if art can be used in politics, it is not definining the "type" of use of art we are making, when we have politics in mind).

Y: True, but for example an eye can see light but is not made of light, there's definitely something in common but it's not directly translatable to the type.

X: I mean, I dont know much about this, but don't photo-receptors get chemically recomposed by the interaction with photons? Something about the chemical bonds in "Retinal" in the process of phototransduction, or whatever...

And, in fact, we are made of light in some weird fundamental level anyway. "the inconceivable idea of the sun", you know..

Y: ahahaha, right, I agree, but here we had to search within the ramification of the eye and light, to look for something at a lower level of their composition, for what is common.

X: Yes, but its not the lowest possible resolution - its quite a big structure in the body, it's an organ, it's not the lowest common denominator between the eye and light - photons bombard us with radiation at more fundamental levels, but at this higher, more composed level, those "rhodes" and "cones" in the retina becomes sensitive to certain frequencies and not others, they can be stimulated within a certain range, right?

Y: I agree, so it's not the lowest common denominator, it's more like the greatest common one with regards to complexity.

Here's how the MES-ers, Ehresmann and Vanbremeersch, approach it:

WhatsApp Image 2022-12-21 at 13.59.30.jpeg

WhatsApp Image 2022-12-21 at 14.02.46.jpeg

X: Ah i see! So the core of the comparison of things of the same type is already thought in terms of behaviours of co-regulators (CR)... that makes wittgensteinian-pragmatic sense (as they say here) in that the meaning of something is its use (=its type is can be translated as a set of receipies for interacting, etc)... then the type aspect of this process is organized in the landscape and the CR is the interactive counterpart of it. The core statement there seems to be that a co-regulator "reacts in the same way to items of the same class".

In our case, i think we are looking both for an analytic principle - a robust defense of the idea that one can learn about the social environment by cultivating certain types of social relations in our organizations.. so new types of interactions (a normative description) leads to new compositions (structural description) and to making new things intelligible about the world (epistemological but also strategic gain)... but we are not just looking for an "orthopedic" claim, that you need type X of social relation to interact with the phenomena X, since we are looking for new things to emerge, so we need to "generify" this process a bit, that is, to allow for the extreme cases where our impredicative forms of connection (our impetus towards being as common as possible in ecological terms) leads to irreducibly new social forms (because we know there is a connection between certain kinds of filters and the broadening or extensions of situations).

Y: Great point! yes, I think this reading emphasizes the experimental-generic aspect more. It's also useful to look at existing interactions and ask WHY are they effective or ineffective, from this standpoint.

X: It fits nicely into our mediations diagram... it remains a valid question when you think "conservatively". Not all mediations will interact with all parts of a complex world.. there is already the issue of their effectivity even when they merely reproduce known structures. But you can expand this issue towards a generic extension of the world... in this case we are not just looking for the right organ to sense the right stimulus, but to devise a new organ that, composing parts that come below the void of the situation, can sense previously "un-formed" parts of the world (but the world now looks different because of it).

Y: Instead of saying we failed because we had the wrong type(s), which means our landscape is wrong, we can also say - we didn't have the right CRs in place that could treat the proper distinctions we wanted, or perhaps they distinguished TOO much when in fact they should accept that those two things are the same.

X: True - though i guess ultimately the two things you said are kind of equivalent? If your landscape implies you should act in a way, but you are not composed to act in that way (like a small group of students planing to "take over the USA") thats is both a problem of inteligibility and a problem of how the group is composed.

Thats also why i think there is this nice psychoanalytic (tektoanalytic?) feel to what we might do with other organizations: they come to us with some reports about how they work (ideally, many reports coming from different organs of the political body in question)... these reports will present mismatches between how they effectively are organized and their political language (it will be too coarse where they make practical distinctions, it will suggest points of attack they have no means to interact with, etc), and we can both help to fit the language to the organization they already are (the "therapeutics" of it) or we can also sometimes help them to see if their true vocation as an organization isnt in a completely different place than they thought (they think they are an occupation, but they are a subversive art collective that can help other occupations rather than keep a building under siege) - there might be capacities "hidden" in the mismatch between their langauge and organization, that allows them to connect in new ways to a larger ecology. That would be the "psychoanalytic" aspect, where the excess is taken to be the true emancipatory part, rather than the conscious part of what we are doing.

Of course, to be actually materialistic about it, we need to include ourselves in this picture... receiving reports, helping organizations to see this or that, etc, all of this depends on our own composition and forms of interaction... in a way, we need to become an extimate organ (a better theory of the so-called "object a"?) to other political bodies, in order to help them recompose parts of themselves in new ways.

I'm just riffing on the whole thing here, but just to show how this debate about types and composition would be totally central to this work of consultancy we are thinking about at STP.

Y : Right, an error could imply 1 or more of the 3 terms of our triad.. for example, "taking over the USA" implies an epistemological/intelligible trace of the USA (its history for example), a normative/interaction component (its relations to citizens and relations with other countries), and a compositional/structural component (its legal framework or demographics)... so what does it mean to "take over" that? Especially when the group of students has no CRs which properly can react to / filter these things - even a military may not be equipped properly, even though it is more effective in these sense of enforcing military rule, it lacks the means probably to treat the relations between citizens and the State.

X: The military could possibly take institutions, state machinery, but the people could easily say that "thats not the USA", because there is a "mode A" aspect to a Nation that is not just the State. On the other hand, a world-class athlete in an Olympics doing something terrible at the most important match - like being caught cheating or something - and tarnishing the image of the country could actually affect "the USA" more effectively, I dunno.

In the sense of affecting a set of "mode A" types of relations that people call "being american" and employ in everyday life, for example.. (and so we get back to the statement "armed militia can challenge the monopoly of violence of the state ("mode B" interaction), but it takes imaginary actors to challenge imaginary communities (nations, mode A)" which is just like our John dies at the End example).

Y: Snowden and Assange probably affected the USA more in that sense.

X: And one can also see how these things mix together.. the moment they became ineffective at the state-level (they did wish to intervene at more fundamental levels of state organization, NSA activity, etc), the mode A dimension of what they represent became the core aspect (people care nowadays more for which country Snowden is in than anything else).

X: The Assange case is quite good here, since he did release documents into a Wiki - but there are just so many people cannot read and process them all! There is no "type" to compatiblize his whistleblowing act with other political acts. So you can see "mode B" and "mode A" interacting, with the latter making up for compositional problems in the other layer.

It is a relevant problem with certain types of "cyber-activism", I believe: to think that the informational domain substitutes the organizational one... ultimately, only the part of the political process that concerned property and common acess of information is effective (they do release new information into the world, this is an effective act at the level of property relations with the State), but this does not translate into collective politics, because the appropriate structure to propagate these actions in a larger organizational ecology was never built.

Y: Right, the Wiki is a CR (CR_B?) that can produce memory (records), but this doesn't necessarily fit the mode A CR (CR_A?) of public opinion (for example social media)

X: Something like that. I think the Wikileaks project is a classical/mode B-centered CR that is incompatible with other organs and subsystems of the political ecology, so it releases documents and all our reactions were... ok what do I "do" with this? (=i dunno what type this is!)

Y: It always boils down to "go vote!" which people clearly see is not compatible with the structure of the problem.

X: ..."We just de-legitimized the State... now go use the state apparatus and trust its legitimacy".

Y: ahahaha exactly!

But to your earlier point, the upshot of all this is that the ecology of organizations needs this kind of "tektological psychoanalysis" on a large scale - and the problem of money kind of re-appears at this level, because for an organization that is concerned about making money, all its re-organizations logically follow from the same source.

Capital "fixes" the mismatches of types in an incredibly efficient way, because the money-compatible resolution can interface with all the modes and other commodity resolutions.

X: Perfect! Man, I think we could edit this conversation and turn it into a blog post!

Y: We should! It's clarified a lot of stuff for me.