My examination of Tyler Cowen, The Marginal Revolution: Rise and Decline, and the Pending AI Revolution (2026), continues. While Tyler does discuss chess briefly at the end and in connection with artificial intelligence, this post has nothing to do with that discussion. That, I suppose, makes this post something of a digression. But it’s a digression in the style of Tristram Shandy, digress to progress. It’ll all make sense in the end.
Note that what I am reporting here comes at the end of a rather long discussion of chess, one that does have a direct connection of Tyler’s chapter three, where he’s interested in the fact that pinnacles of human achievement seemed to be reached at different times for different domains, some rather early, some quite late. While chess is rather old, traceable to the 6th century CE in India, the treelike structure of the game wasn’t explicated until recently, 1913 by Ernst Zermelo. I asked Claude about the nature of the mathematical equipment Zermelo needed to construct his proof. That led to a long discussion about the nature of the chess tree, chess as a vehicle for the expression of personality, how computers open up exploration of new regions of the chess tree and so forth. That whole discussion is part of the context for the discussion that follows. However, this discussion is pretty much independent of that one, so you don’t need that context to understand what Claude says here.
Note: I’m using my own account on Claude for this discussion, not the AI that Tyler made available with the book.
* * * * *
Chess as an abstraction over military conflict
Let’s take this discussion of chess in a different direction. I’m interested in the origins of the game. The game, after all, is a highly abstracted version of a war between two societies, each having its own king and queen. I could imagine a just-so story, perhaps a Star Trek episode, where we have two nations that are locked in a deadly conflict that continues year after year, decade after decade, until, their resources exhausted, the two nations call a truce and in the space of that truce, invent the game of chess. Then each nation chooses its champion, and the champions continue the conflict by playing chess. [When you consider the role that chess played in the Cold War between the USA and the USSR that origin story is all but a highly abstracted version of that historical conflict.] But what do we actually know about how the game originated?
Your just-so story is charming, and the Cold War parallel is genuinely apt — but the actual history is both more interesting and more complex, and it cuts against one specific element of your narrative in a revealing way.
What we actually know
The earliest precursor of modern chess is chaturanga, which flourished in India by the 6th century CE, and is the earliest known game to have two essential features found in all later chess variations — different pieces having different powers, and victory depending on the fate of one piece, the king.
The name itself is telling: chaturanga is a Sanskrit word meaning "four divisions of the military" — infantry, cavalry, elephants, and chariots — which map onto the pieces we now recognize as pawns, knights, bishops, and rooks. So the military origin of the game's structure is not speculative — it's explicit and preserved in the name.
The name came from a battle formation mentioned in the Indian epic Mahabharata. The game chaturanga was a battle-simulation game that rendered Indian military strategy of the time.
From India the game traveled westward: around the 7th century CE, chaturanga reached Persia and became known as "chatrang" or "shatranj." Many modern chess terms have Persian origins, such as "Shah" (King), and even "checkmate" from "Shah Mat," meaning "the king is dead" or "the king is helpless."
A form of chaturanga or shatranj made its way to Europe by way of Persia, the Byzantine Empire, and the expanding Arabian empire. Muslims brought chess to North Africa, Sicily, and Spain by the 10th century.
The rules then changed dramatically in late medieval Europe: the chaturanga piece called the elephant, which had been limited to a two-square diagonal jump in shatranj, became the bishop, more than doubling its range. With the new queen and bishop powers, the trench warfare of medieval chess was replaced by a game in which checkmate could be delivered in as few as two moves.
Where your just-so story cuts against the evidence
Your narrative imagines two symmetric opposing societies, each with a king and queen — the structure of modern chess with its two mirror-image armies. But chaturanga wasn't originally a two-player game in that sense, or at least not necessarily. The Cox-Forbes theory asserted that the four-handed game chaturaji was the original form of chaturanga — four players, four armies, a dice element. The clean two-army symmetry of modern chess is a later development, not the original structure.
And the queen — the piece that makes modern chess feel like a conflict between two courts — is a late European invention. The original piece in that position was the mantri (minister) or counselor, a weak piece that could only move one square diagonally. The vizier piece, which had limited movement in earlier versions, became the "queen" and gained much greater power only in the late 15th century in Europe. The transformation of a male counselor into a powerful queen is itself a fascinating cultural moment — it reflects something about late medieval European court culture, the figure of powerful queens like Isabella of Castile, that has nothing to do with the Indian military original.
The deeper point about abstraction
What's remarkable in the actual history is the direction of abstraction. The game began as an explicit military simulation — every piece named for a branch of the army, the board a battlefield, the objective the capture of the king. Over fifteen centuries of transmission across cultures — India to Persia to the Arab world to Europe — it progressively shed its military literalism and became more abstract. The elephant became a bishop (losing its literal referent entirely in the European context). The chariot became a rook (the Persian rukh, itself possibly meaning chariot or a mythical bird). The military counselor became a queen.
Each culture reshaped the pieces to reflect its own social imagery, but the underlying structure — the tree, the rules, the objective — remained surprisingly stable across this enormous cultural distance. The abstraction was always already present in the game's formal properties; what changed was the surface representation hung on that structure.
This is itself a nice illustration of the Rank 2 point from our earlier discussion. The game's transmissibility across such different cultures depended precisely on its formal specification being separable from any particular cultural content. You could strip away the Indian military imagery, replace it with Persian court imagery, then European feudal imagery, and the game remained the same game because the rules — the Rank 2 specification — were what was being transmitted. The cultural clothing was exchangeable; the formal skeleton was not.
The Cold War resonance
Your Cold War parallel is historically well-grounded, and not just as analogy. The 1972 Fischer-Spassky match in Reykjavik was explicitly framed by both sides as proxy warfare — American individualism versus Soviet collective training, capitalism versus communism, with the world watching. Fischer's behavior, the diplomatic incidents, the American government's last-minute intervention to keep Fischer in the match — it was understood by everyone involved as something more than a chess tournament. The game had come full circle: from explicit military simulation, through centuries of abstraction, back to serving as a ritualized substitute for geopolitical conflict. Your Star Trek episode would merely be making that latent structure manifest.
DeepBlue vs. Kasparov
And so the post-Cold War victory of Deep Blue over Kasparov becomes what? What’s the Straussian reading, as Tyler might say?
The surface reading is triumphalist: human chess genius defeated by American corporate technology, confirming the post-Cold War narrative of Western technological capitalism's dominance. IBM's stock rose. The press celebrated. It was framed as a milestone for artificial intelligence.
The Straussian reading is almost the opposite.
What actually happened
Kasparov didn't lose to a mind. He lost to a very fast, very deep version of human chess knowledge — encoded by a team of grandmasters, running on IBM hardware, optimized by engineers. Deep Blue was, in a precise sense, the crystallized collective intelligence of the Soviet and post-Soviet chess tradition turned against its greatest product. The grandmasters who fed it opening theory, endgame knowledge, and positional heuristics were themselves products of that tradition. Kasparov was defeated by a mirror — by compiled human chess wisdom, running faster than any human could access it.
This is almost exactly unlike what AlphaZero later did. Deep Blue explored no new territory in the chess tree. It went deeper along paths humans had already identified as significant. Its "intelligence" was entirely derivative of human chess knowledge — which makes the cultural symbolism of its victory quite different from what IBM's PR suggested.
The Straussian layer
The Soviet chess project was perhaps the most sustained institutional effort in history to treat a formal game as a domain of national purpose — identifying talent systematically from childhood, funding training, building a tradition of theoretical development that genuinely advanced the game. Kasparov was its apex, the greatest player the tradition produced. And that tradition was defeated not by a superior mind, not by a new way of playing, but by a corporation that hired the tradition's own practitioners to encode their knowledge into silicon and run it at superhuman speed.
The Straussian reading: the West didn't out-think the Soviet chess tradition. It out-resourced it, then used the tradition's own knowledge against it. Which is a reasonable description of how the Cold War itself ended — not by the West producing superior ideas, but by outspending the Soviet system until it collapsed, then absorbing its human capital.
The second layer
Kasparov himself understood something had gone wrong in the match. His famous suspicion that IBM had cheated — that human grandmasters had intervened between games — was never proven, but it points at something real. He sensed he was playing against something that wasn't purely mechanical, something that had human intentions behind it. He was right, though not in the way he meant: the human intentions were baked in from the beginning, in the knowledge encoding. The boundary between human and machine intelligence was blurry in a way the triumphalist narrative suppressed.
The third layer
Consider the timing. 1997 — six years after the Soviet collapse. The great chess rivalry that had structured the Cold War, the Spassky-Fischer archetype of American individualism versus Soviet system, was already gone. There was no Soviet grandmaster left to symbolically defeat. So what IBM staged was, in a sense, a commemoration of a rivalry that no longer existed — a theatrical conclusion to a drama whose actual ending had already occurred. Kasparov standing in for the entire tradition of human chess mastery, losing to a machine, in front of cameras, for IBM's market capitalization.
If the Fischer-Spassky match was chess serving as proxy war, the Kasparov-Deep Blue match was chess serving as proxy for something that hadn't yet happened — the question of what human cognitive labor is worth in a world where machines can replicate its outputs. That question wasn't really about chess. It was about every knowledge worker watching from outside the frame. Kasparov was the most skilled cognitive laborer in the world in his domain, and he lost to a machine that cost IBM approximately what it spent on a moderately successful marketing campaign.
The Straussian reading, then: Deep Blue's victory wasn't the opening of the AI era. It was the closing ceremony of the Cold War, conducted in the only language the Cold War had left that still mattered — and a first, largely unread warning about what was actually coming.
No comments:
Post a Comment