If… Then: The Idiom of Data Exchange
In situating the juridical in between the ritual and disciplinary as a mode of ordering both more historically contingent and more essential to imperial tensegrity I confronted an interesting problem, one which I came across in reading Durkheim. The highest achievement of the juridical would be to abolish itself, and this end is implicit in the organization of the juridical and its embeddedness in institutions. Let’s work with my definition of the juridical as placing a ceiling on the vendetta by introducing higher order arbitration at those points where the vendetta threatens to undo imperial order; while, on the other end, preserving what is ultimately a ritually enacted originary nomos establishing property rights—which really means deferring the recrudescence of the vendetta within the juridical, the ultimate endpoint of which is executing the king as a traitor. The juridical doesn’t have to eradicate the vendetta, even though that invariably seems to happen; it just needs to contain it, even while preserving “cultural” traces (young girls might still invoke the protection of their older brothers, etc.). And the juridical can introduce some reshufflings within the nomos, but not past the point where challenges to existing property threaten to exceed the possibility of adjudication.
Now, let’s say we center the juridical and focus our attention and resources on it, see it as the foundation of our narratives and theoretical discourses, study it carefully, tend to it, etc., so that we increasingly arrest violent criminals and punish them in a way satisfactory to the vast majority, refine civil courts so that even losing claimants accept the legitimacy of decisions, restore libel and defamation law so that public lying about people is too costly to be worth the monetary benefits or satisfaction, etc.; well, wouldn’t we, at a certain point, find violent crime decreasing because capture and punishment are so certain, civil court cases declining because potential claimants already know the likely settlement of their dispute, “cleaner” public discourse, etc.—leaving, that is, no work for the courts to actually do. The perfection of the juridical would then be its abolition. Could the same be true, then, of the ritual and disciplinary, or does this mark the specificity of the juridical? If we take the ritual order, we’d have to think that the purpose of ritual is commemorate the origin—the origin of the particular community, but that particular community as an instance of the origin of the human community itself. The ritual is distinguished from the profane, but does it tend towards the transcendence of this distinction? For our purposes, if ritual “properties” can be transferred and translated from ritual in the narrow sense to the hallowing of all of life as life itself is lived, so that all of reality becomes a ritual (the inspiration of mystics and romantics) then we would have the same situation: the perfection of ritual would be its abolition. It’s easy now to see that the same happens with the disciplinary, insofar as we are always, everywhere, generating, gathering and analyzing data and that we could be given progressively more access to the data our contribution flows into along with the algorithms used to order that data and derive conclusions from and that this knowledge would in turn feed back into the ways we generate data. In which case, everything is disciplinary, so, then nothing is—and everything is data, so there’s no more data?
These reflections follow from the task of designing further the notion of “data exchange” which I’ve made central without fleshing out very much. “Data” relies upon, most obviously, juridical and disciplinary orderings, on spaces and materials possessed, guaranteed by chains of possession and provenance, supervised and transferred and recorded in trustworthy ways, and measured in controlled spaces, actual or virtual. And all this depends on the originary distribution or nomos, which includes the distribution of authority. I’m positing here, as a kind of “utopian” or “messianic” horizon, the abolition of the ritual, juridical and disciplinary in what I suppose we could see as the entire social order as a total work of art or, as I would prefer, perpetual language learning, the invention and spread of idioms. This has implications for what is to count as data and how it is to be measured. In the messianic age we should all just be “giving off” to each other and to the center the information we need for our interactions, just like someone you are intimate with can deduce from a facial expression how you are to be treated that day. That means we’re working on our data exchanges with that in mind and shape our self-expressions and self-representations accordingly. And it also suggests that we anchor data exchange in what will endure, even into messianic times, which is the grammar of speech forms, the ostensive, the imperative, the interrogative, the declarative. So, back to originary grammar.
The conditional “if-then,” the basis for the modern algorithm, could also be treated as a place where the speech forms converge. It is first of all, like computing language itself, on the borderline between imperative and declarative—if x then y can mean if x occurs do y just like it can mean if x is true y is also true. “If” is also, of course, a typical opening of a question, one that is raised spontaneously as one proceeds in the sequence, whether to follow a command or check a truth claim. And “if x” needs to be verified ostensively before we move on to y. If-then implies the possibility of taking all of language as endless trees of hypotheticals that can be checked at each point before proceeding to the next fork or simply marked as to be checked, perhaps by seeing the implications of proceeding as if it were true, or authoritative. We have our own “halting problem” here insofar as we do need to stop for decisions along the way, and those decisions create new trees of if-then that cluster around those halts. But we can solve this problem simply by using the split between an imperative and a declarative treatment of the if-then formula as a way of creating the fork itself. If it’s an imperative then one has to halt, and “become” the hypothesis, rather than merely considering it. It’s never completely one or the other, so we can assume infinite gradations along a continuum from 99% imperative/1% declarative to 1% imperative/99% declarative. “If I’m not here by 9 then leave without me”—pretty much imperative, but one does have the option of waiting longer and seeing where the boundary between falsifying and disobeying lies. “If there’s a storm tonight, then the roads will become impassable” seems to be pure prediction, but there’s a kind of imperative not to try and pass on the roads tonight.
Producing data starts with naming and measuring things—you name so you can measure, but new ways of measuring create new names. I’ll set aside the strict experimental sciences here, where questions of justice rarely impinge, to focus on my real interest here, human data. Measuring something basic like births is simple enough but still requires that births mostly take place in hospitals, which are public, record-keeping institutions, and that hospitals are mostly reliable record keepers and preservers—and that those health officials, ultimately from the government but also insurance companies among others, themselves are competent and don’t falsify information. To accept a piece of data is to have faith in an entire infrastructure and social order. All data is “interested”—someone has a reason for being interested in it. With humans there can be no set way to name and measure things. All data, then, already has a hypothetical, if-then character to it: if we name and measure human practices this way, this is the array we end up with. We could easily imagine, then, multiple parallel data collecting practices operating simultaneously, asking different questions requiring different names and measurements. There wouldn’t even be any need to argue about which is the best or right way, as long as the purposes of each mode are explicit; indeed, each data practice could draw upon and translate the results of the others. We could then formulate the hypothesis founding each mode of data practice in such a way as to elicit the participation, even after the fact participation, of the subjects, precisely by making explicit the boundary between imperative and declarative in the hypothesis. The hypothesis activates an imperative the population or some sector of it has been following with regard, say, to health, wellness, education, innovation, etc., and the inquirer follows the fulfilling of that imperative through its prolongation into a question (what have been the counter imperatives, scenic obstructions or facilitations, modifications of the imperative through its materializations, etc.) which the inquiry caps with a declarative.
Data exchange is always with the center, so to think in terms of converting judgments, ordinarily rendered in money or the “currency” of prison time or some other penalty has to be mediated through the center. The data gatherers would also be preparing the conditions for justice. So, let’s think in terms of an example, a hard one, say a violent criminal, whose prison time we would want to convert into a data donation program—keeping in mind, of course, that part of a prison sentence is preventing the criminal from reoffending and ensuring the safety of the community. Perhaps we can even do better than the current system does in terms of compensating the victim specifically, in his or her injury, something the current system doesn’t address at all. And, of course, all this with the ultimately “messianic” hope of doing away with punishment altogether, albeit without in the least capitulating to the kind of “as if” thinking that imagines leniency itself hastens us toward the end of the punitive dimension of social life. We would need a study of the givens of a particular crime and the history of the offender—we treat the crime as the answer to a question we present as having been posed: someone like this, in such and such conditions, is how likely to do something like this? Now, having done this thing, how likely is this person to do any one of a range of other things? The imperative is issued to the criminal to increase the likelihood that someone like that will do one thing rather than another. The criminal presents himself to the board of inquiry as a sample; of course, so do we all, in our own ways, with the boundaries between different ways of donating or withholding donations to the center being more of a spectrum than an either/or determined by the law. We have to assume the possibility of some kind of technologically enabled mobile sanctuary city condition, such as high alerts triggering rapid, if necessary automatically disabling, responses if certain boundaries are transgressed. The victim(s) of the criminal can perhaps be given privileged access to the data and be allowed to contribute to the algorithms governing the processes of naming and measuring. The more the criminal increases the likelihood that the kind of person who has done the kind of thing he has done will become the kind of person who will act differently in the future, the more conditions of surveillance and control are modified and relaxed.
It seems to me that under such conditions the commission of violent crimes would become increasingly futile—if one could impose technological sanctuary conditions on individuals then one could probably construct the technological conditions making robbery, murder, rape, etc., increasingly marginal. So, we are back to the beginning, where the very thinking of the cybernetics of a concept produces the horizon of the concept’s abolition. All political agendas can be channeled this way towards, of course, the abolition of politics. In setting up Thirdness prediction markets, then, we are also setting up a kind of conceptual game of reframing confrontations so that they are channeled toward data exchange with the center and the horizon of the abolition of the conditions of the terms of the conflict itself. We need, now, to be able to set up, in these terms, an opening state, the conflict or dispute itself potentially presented for judgment, and the closing state in which justice is rendered in such a way that either/or bets on the closing state can be redeemed. A very highly trained AI model would need to be introduced into the equation, but I’ll still try to give an answer that at least allows us to imagine how one might proceed. We construct a pair of if-thens in as an imperative a form as we can make them—that is, each side is given a task the fulfillment of which would demonstrate a higher level of self-preparedness for judgment. The bet is then on which is to be deemed most likely to fulfill the imperative in the declarative judgment to be rendered by the Thirdness team—that judgment must therefore provide an explicit preferring of one position over the other. So maybe we reduce an event (as one of its forms) to a brief confrontation between some protesters and a police officer, and we construct a juridical scene in which the two sides are suing each other. The protesters might be “asked,” based on a preliminary account of the event, to provide a theory of justice and the state to justify their actions, while the police officer might be asked to propose some resource from his training and experience he might have drawn upon in controlling the protesters more effectively. The question might, then, be one of whether we can see the police officer as embodying a mode of authority, a representation of the center, that someone in that situation would have been obliged to respect and obey. Further study of the event constructs likely answers from the two sides and arrives at a judgments tipping the scale toward either side as representing, in that case, a higher mode of authority or looser adherence to the imperative from the center.
Durkheim comes to the opposite conclusion as I do here—his argument is that “crime” can never be abolished because the more well-behaved the population becomes the lower the threshold for identifying criminality—in a way this notion parallels the development of the victimary, with its continual lowering of the threshold of what is to count as “racist,” etc., except that we’d have to imagine thresholds being lowered all around. This might mean that a highly civilized order would impose the death sentence for littering, but it’s more likely that punishments would become far lighter as well because any kind of justified social disapproval would be felt with great force by the offender. This means that the tendency I’ve described above remains the same, as ever greater powers of detection (i.e., data gathering and analysis) would be needed to pick up on offenses. This makes it clear that the messianism I invoked above is always a horizon, which is to say a kind of measuring rod. So, the difference between the opening and closing state on the Thirdness prediction market is a question of the antagonists becoming better at measuring and better measurement devices. Those two poles converge in each, in their reciprocal entanglement, becoming more meaningful, which is to say more idiomatic, more simultaneously enacting the imperative and declarative side of the if… then, studying and protecting the originary distribution at the site of justice. So, in our opening state in Thirdness, we would have to know enough to generate alternative closing states in such detail that they can be bet on, while avoiding any pattern of details sufficiently recognizable so as to ensure confidence in a genuine inquiry by the Thirdness team and therefore a 50/50 bet. To beat the house you’d need to be able to, as I’ve pointed out before, have sufficient insight + computing power to identify patterns hidden even to the Thirdness team—and, in that case, the Thirdness team would want to recruit you. But maybe that, in turn, simply means that the closing state, or judgment, is an attempt to land on that mode of data extraction from and data provision to both sides as would make them both, and equally, minimally more likely of becoming capable of someday being recruited to the Thirdness team. And this, finally, means that you’d be betting on which decision the Thirdness team deems more likely to increase the likelihood, albeit asymptotic, that Thirdness will be abolished as it is taken up in all everyday practices.