The Problem with Diachronic Extrapolation
23 March 2013
In Synchronic and Diachronic Approaches to Civilization and Axes of Historiography I discussed the differences between synchronic and diachronic approaches to historiographical analysis (and in much greater detail in Ecological Temporality and the Axes of Historiography). The synchronic/diachronic distinction can also be useful in futurism, and in fact we can readily distinguish between what I will call synchronic extrapolation and diachronic extrapolation.
If we understand synchrony as, “the present construed broadly enough to admit of short term historical interaction” (as I formulated it in Axes of Historiography), then synchronic extrapolation is the extrapolation of a broadly construed present across its interactions. This may not sound very enlightening, but you’ll understand immediately what I mean when I relate it to chaos and complexity. Recent interest in chaos theory and what is known as the “butterfly effect” has led some to think in terms of synchronic extrapolation since the idea of the is of a small event the interactions of which cascade to produce significant consequences.
As a form of futurism, synchronic extrapolation is not familiar (probably because it doesn’t take us very far forward into the future), but we need to keep it in mind in order to contrast it with diachronic extrapolation. Diachronic extrapolation is one of the most familiar forms of futurism today, especially as embodied in Ray Kurzweil’s love of exponential growth curves, which are usually diachronic extrapolations. One of the reasons that I remain so skeptical about the claims of Kurzweil and other singulatarians (even though I have learned a lot about them recently and have a less negative picture overall than initially) is the heavy reliance on diachronic extrapolation in their futurism. I frequently cite specific examples of failed exponential growth curves or technologies (like chemical rockets) that seem to be stuck in a technological rut (what I have called a stalled technology), experiencing little or no development (and certainly not exponential development), and I do this because readers usually find specific, particular examples persuasive.
I have discovered over the course of many conversations that most people tune out extended theoretical expositions, and only sort of wake up and pay attention when you give a concrete example. So I do this, to the best of my ability. But really, the dispute with diachronic extrapolation (and particular schools of futurist thought that employ diachronic extrapolation to the exclusion of other methods, such as the singulatarians) is theoretical, and all the examples in the world aren’t going to get to the nub of the problem, which must be given the theoretical exposition that it deserves. And the nub of the problem is simply that diachrony over significant periods of time cannot be pursued in isolation, since any diachronic extrapolation will interact with changed conditions over time, and this interaction will eventually come to constitute the consequences as must as the original trend diachronically extrapolated.
Diachronic extrapolation may be derailed by historical singularities, but it is far more frequent that nothing so discontinuous as a singularity need happen in order for a straight-forward extrapolation of present trends fail to be be realized. I specifically single out diachronic extrapolation in isolation, because the most frequent form of failed futurism is to take a trend in the present and to project it into the future, but any futurism worthy of the name must understand events in both their synchronic and diachronic context; isolation from succession in time is just as invidious as isolation from interaction across time. This simultaneous synchrony and diachrony resembles a chain reaction of ever-growing consequences from the initial point of departure.
In my two immediately previous posts — Addendum on Automation and the Human Future and Bertrand Russell as Futurist — I dealt obliquely with the problems of diachronic extrapolation. Predicting technogenic unemployment on the basis of contemporary automation, or predicting a bifurcation between annihilation or world government, is a paradigm case of diachronic extrapolation that fails to sufficiently take into account future interactions that will become as important or more important than the diachronically extrapolated trend.
This was the point that I was trying to make in Addendum on Automation and the Human Future when I wrote:
I am willing to admit without hesitation that, 250 years from now, we may well have realized a near-automated economy, and that this automation of the economy will have truly profound and far-reaching socioeconomic consequences. However, the original problem then becomes a different problem, because so many other things, unanticipated and unprecedented things, have changed in the intervening years that the problem of labor and employment is likely to look completely different at this future date.
In other words, a diachronic extrapolation of current employment trends — technogenic unemployment, new jobs created by new industries, and perennial problems of unemployment and underemployment — is helpful in so far as it goes, but it doesn’t go nearly far enough in capturing the different world that the future will be.
Similar concerns hold for Russell’s failed futurism that I reviewed in Bertrand Russell as Futurist: Russell took several trends operating at present — war, nuclear weapons, anarchic competition among nation-states — and extrapolated them into the future as though nothing else would happen in history except these closely related group of strategically significant trends.
In my post on Russell’s futurism I cited his essay “The Future of Man”, but Russell made the same point innumerable. times. In his first essay on the atomic bomb, “The Bomb and Civilization,” he wrote:
Either war or civilization must end, and if it is to be war that ends, there must be an international authority with the sole power to make the new bombs. All supplies of uranium must be placed under the control of the international authority, which shall have the right to safeguard the ore by armed forces. As soon as such an authority has been created, all existing atomic bombs, and all plants for their manufacture, must be handed over. And of course the international authority must have sufficient armed forces to protect whatever has been handed over to it. If this system were once established, the international authority would be irresistible, and wars would cease. At worst, there might be occasional brief revolts that would be easily quelled.
And in his book-length study of the same question, Has Man a Future? Russell made the same point again:
“So long as armed forces are under the command of single nations, or groups of nations, not strong enough to have unquestioned control over the whole world — so long it is almost certain that sooner or later there will be war, and, so long as scientific technique persists, war will grow more and more deadly.”
Bertrand Russell, Has Man a Future? New York: Simon & Schuster, 1962, p. 69
We have seen that armed forces continue to be under the command of individual nation-states, and in fact they continue to go to war with each other. Moreover, scientific technique has markedly improved, and while the construction of weapons of mass destruction remains today a topic of considerable political comment, the availability of improved weapons of mass destruction did not automatically or inevitably lead to global nuclear war and human extinction.
In the same book Russell went on to say:
“…it seems indubitable that scientific man cannot long survive unless all the major weapons of war, and all the means of mass destruction, are in the hands of a single authority, which, in consequence of its monopoly, would have irresistible power and, if challenged to war, could wipe out any rebellion within a few days without much damage except to the rebels.”
Bertrand Russell, Has Man a Future? New York: Simon & Schuster, 1962, p. 70
In writing these comments, we can now see in hindsight that one of the major strategic trends of the second half of the twentieth century that Russell missed was the rise in the efficacy of asymmetrical resistance to irresistible power. Russell does not seem to have recognized that authorities in possession of de facto irresistible power might choose not to annihilate a weaker power because of global opinion and the hit that such an actor would take to its soft power if it simply wiped out a rebellion. Moreover, the wide distribution of automatic weapons — not weapons of mass destruction — proved to be a disruptive force in global political affairs by providing just enough friction to the military operations of great powers that rebellions could not be wiped out within a few days.
The rise of twentieth century guerrilla resistance and rebellion was an important development in global affairs, and a development not acknowledged until it was already a fait accompli, but I don’t think that it constituted an historical singularity — as it is part of a devolution of warfare rather than a breakthrough to a new order of magnitude of war (which seems to have been what Russell feared would come about).
It has been said (by L. P. Hartley, a contemporary of Russell) that the past is a foreign country. This is true. It is also true that the future is a foreign country. (Logically, these two claims are identical; every present is the future to some past.) We ought to make no pretense to false familiarity with the future, since they do things differently there.
. . . . .
. . . . .
. . . . .