The Channel Changed the River

In the AI Story

The standard model of technology adoption treats tools as external to their users: the carpenter builds the hammer, uses it, and remains unchanged except for the calluses on his hands. The hammer extends the hand's capacity without altering the hand's structure (except superficially). Deacon's insight: this standard model fails catastrophically for cognitive technologies. Writing did not merely extend memory; it restructured the cognitive habits of literate populations, producing forms of thought (systematic reasoning, cross-cultural synthesis, scientific method) that oral cultures could not support. The technology reached into the user's cognitive architecture and reorganized it.

Language represents the deepest instance of this dynamic because it operated at biological as well as cultural timescales. The cultural reorganization (the emergence of symbolic communication practices) created biological selection pressures that reorganized neural architecture across evolutionary time. The brain's plasticity—the capacity to rewire itself in response to experience—provided the mechanism at the individual level; genetic variation in the ease of symbolic learning provided the mechanism at the population level. The spiral was self-reinforcing: better brains enabled more complex language, which selected for better brains, which enabled yet more complex language.

The AI parallel: cultural timescales are orders of magnitude faster than biological. The neural architecture will not change genetically in any timeframe that matters—hundreds of thousands of years minimum, probably millions. But cognitive habits change in months. Attentional norms change in years. Educational practices change in a generation. The cognitive environment in which the next cohort develops can be fundamentally different from the environment in which the current cohort developed, and that environmental difference will shape the cognitive capacities the next generation builds—not genetically but developmentally, through the neural plasticity that allows every human brain to reorganize itself in response to the cognitive demands it encounters.

The dams, in this framework, are the deliberate structures—educational, institutional, cultural, personal—that shape the cognitive environment to preserve the capacities that matter most. If the AI-saturated cognitive environment systematically rewards symbolic fluency over indexical grounding, rapid production over embodied struggle, the generation developing in that environment will build different cognitive habits than prior generations. The habits will be adaptive to the AI-mediated environment—and potentially maladaptive to the forms of cognition (deep understanding, embodied judgment, purposive orientation) that the environment does not select for.

Origin

The reversal emerged from Deacon's sustained examination of the neuroanatomical evidence that the standard story of language origins could not explain. If the brain evolved first and then invented language, the brain should show general expansions (bigger everywhere) or random variations. It shows neither: it shows targeted reorganizations in exactly the regions symbolic processing requires. The only explanation: language was already present, in proto-form, creating the selection pressures that produced the reorganizations.

The metaphor of the channel changing the river is not Deacon's (it is deployed in this simulation to connect his framework with Segal's river-of-intelligence metaphor from The Orange Pill), but the substance is pure Deacon: the recognition that the most important causal relationship is not linear (brain → language) but reciprocal (brain ↔ language), and that reciprocal causation over sufficient time produces entities that cannot be understood independently.

Key Ideas

Tools restructure toolmakers. Cognitive technologies do not merely extend existing capacities—they reorganize the cognitive substrate itself, changing what is easy, what is hard, what is conceivable.

Reciprocal causation over time. The brain shaped language; language shaped the brain; neither is explanatorily prior; the co-evolutionary spiral is the phenomenon requiring explanation.

Targeted neural reorganization. The specific regions enlarged in the human brain correspond to symbolic language's demands—working memory, inhibition, vocal precision—not to general intelligence.

Cultural co-evolution at compressed timescales. AI may reshape cognitive habits, attentional norms, skill distributions on timescales of years rather than millennia—cultural evolution, not genetic, but co-evolutionary nonetheless.

Dams shape the developmental environment. Institutional, educational, and cultural practices determine the selection pressures on cognitive development in AI-saturated environments—directing the co-evolution or allowing it to proceed blindly.