ALGORITHMS, AI, ASTROLOGY, AND ASIMOV’S HARI SELDON

This has been one of those weeks that my email inbox has been flooded with bloggable articles. As you might imagine, many of these articles have been concerned with the encroachments of "artificial intelligence" and related technologies and techniques into every day life. A case in point is the following article discovered and shared by K.M.(with our thanks), and it prompted so many immediate thoughts and high octane speculations on my part that I have to talk about it:

SIGPSY — Special Interest Group in Psychohistory

The first thing that leapt out of that article was simply its headline, or rather, the presence of the word "psychohistory" in that title. As most regular readers here will already know, I am a big fan of the depth of thought and cultural prediction behind the late Isaac Asimov's science fiction, and particularly in his Foundation series of novels, especially in the original trilogy of novels: Foundation, Foundation and Empire, and Second Foundation.  For those who do not know, "psychohistory" is the name Asimov invented for a mathematical modeling of the major cycles and periods of history, and the characteristics and properties that distinguished each. As such, "psychohistory" also became a powerful predictive tool to forecast the general trends of the future. In the novels, this science was invented by the mathematician Hari Seldon, though given its nature, the high amount of multi-variable elements, and the sheer volume of data and analysis that would have needed to have been "crunched," one can easily imagine that such a "science" of "psychohistory" would require access to vast computing power and programming expertise that would have included what we now call "artificial intelligence."  It is worth noting that the term that Asimov invented for a fictional science in science fiction novels has now become an accepted term for a real discipline.

In outlining his conception of it in the novels, Asimov also provided a description of another attempt to describe a comprehensive mathematical science of historical-sociological and cultural modeling, an attempt made much early and based on a vast pool of data, the attempt of economist Edward R. Dewey and his Foundation for the Study of Cycles.  Again, for those who do not know the story as I recounted it in my book Babylon's Banksters, Dewey worked at the Department of Commerce in the early years of the Great Depression, and was tasked by President Herbert Hoover to discover why economies kept having boom-and-bust cycles. Dewey, accessing the vast data of the Commerce Department, came to some incredible conclusions: economic activity in all the areas of production and commodities that he studied was always cyclical, with the cycles themselves consisting of very long waves of several years (thus suggesting a profound "econophysics" was at work), and that there was nothing any government could do to reverse the overall trend line of a cycle. All governments could do was to accentuate or attenuate the trend line of the cycle.  Dewey accumulated a massive amount of data on almost every conceivable kind of cycle of economic activity imaginable, and his Foundation for the Study of Cycles still exists, and is still accumulating data and trying to derive a comprehensive "unified field theory" of cycles. I've long suspected that Asimov might have known of Dewey's work, and that his own fictional Foundation series might owe its conceptualization to Dewey.

In outlining these fictional and non-fictional efforts of Asimov and Dewey, it is to be noted that both versions are (1) highly statistical and (2) concern themselves with aggregate behavior characteristics of large human populations (3) over long periods of inter-generational time. In other words, one is looking at a modern attempt to create a "scientific" version of astrology, which, let it be noted, itself came out of thousands and thousands of observations made over thousands of years according to some ancient texts. In short, astrology was a stastistical predictive science of aggregate human behavior, not individual human behavior. That it became the latter is perhaps a reflection more of the loss of statistical forms of mathematics prior to the classical period, rather than anything else. In any case, however, the vast mathematical knowledge required to assemble and sift such a large database would have required something like our contemporary artificial intelligence even in ancient times.

And this is acknowledged in the article, for the very first paragraph describes most exactly the very "psychohistory" at the center of Asimov's Foundation novels and at the heart of Dewey's cycles:

Studying long-range historical modeling and prediction — drawing on quantitative history, complexity science, and protocol theory to develop frameworks for understanding civilizational-scale dynamics. The group maintains worldmachines.org, a collaborative platform for psychohistorical modeling.

There is even a strong suggestion that the group's work of necessity relies upon mathematics that can only be described as "higher order topologies" (topological metaphor of the medium, anyone?):

• The group planned to read P.W. Anderson's 'More is Different' paper, which explores how different levels of organization in physical systems follow distinct laws that cannot be reduced to lower levels.
• The meeting adopted a structured approach: 15-20 minutes of paper reading followed by quick comments, then transition to symposium planning discussions.
• Anderson's work on hierarchies of laws from physics to higher-order systems appears directly relevant to psychohistory methodology and understanding complexity. (Emphasis added)

Needless to say, all of this caught my eye, but what captured my attention more than even these strongly suggestive statements was the idea that history itself, as the group viewed it, was a kind of super-machine comprised of an indeterminate number of smaller machines - Dewey's individual cycles, for example - all working toward some goal that, with proper modeling, could be predicted. And thus is born the "world machine", or as I've called in in prior blogs on this subject, "world sentient programs/machines" or "world scenario computer modeling/gaming":

SIGPSY Update on Prime Radiant and World Machines

...

Aneesh gave an extended walkthrough of the software he has been building for the group: a note-taking/wiki app ("Scribe"), the "Oracle" retrieval system built on the group's corpus of books and essays, and the "Prime Radiant" — a temporal knowledge-graph system that models history as interacting "machines" and generates extrapolations/"prophecies." The second half turned to planning the upcoming symposium/special session (involving Aneesh, Sean, and Florian), which Venkatesh framed as the project's public launch. Discussion touched on how notes and essays could eventually feed automatically into the modeling system, and on how contributions and funding might scale afterward.

This intrigued me because for years - both on this website in various blogs, and in interviews (primarily with Catherine Austin Fitts) - I have maintained the hypothesis that some sort of "world sentient program" or "world modeling scenario" was being crunched out by supercomputers somewhere. My reason for maintaining such a wild and woolly hypothesis was simply because it appeared to me that the political leadership class, particularly of the west, was acting increasingly insane, and was no longer response to events in normal human ways; they seemed capable of reacting only as this or that "collective" could be expected to react. In short, they were following a computer program, and doing what that program told them to do. (For the titles of my previous blogs on this idea, search for the following titles on this website's search engine: "Confirmation: Listening to a Program Called Sentient World...", "Econophysics, Sentient World Simulations...", and Isaac Asimov's Foundation Trilogy Comes to Life."

In short, we have further confirmation that my high octane speculations from over a decade ago were over the target... and that brings us to a bit of additional high octane speculation: how might one further and more accurately intensify the "training" of artificial intelligences that rely on large language models and their implied vast databases? Why, one might release all sorts of media hysteria over artificial intelligence apocalypses, and gather response data from aggregate human behavior. In short, one might start stimulating certain kinds of behaviors by certain types of data dumps...

... and once one has said that, one has also invoked macro-observer effects and potential cosmological consequences....

In that respect, it is intriguing and necessary to note that in Asimov's original trilogy, the thing that nearly undid the whole intricate galactic psychohistory and its predictions was an outlier, an individual, with paranormal "psychic" abilities...

See you on the flip side...

(If you enjoyed today's blog, please share it with your friends.)