In Nominalist Determinism, the scrutinizing focus is the semantic audit. By addressing not only what words mean in the vernacular but also the limits of what words can possibly mean given the constraints of the natural world, we gain a mechanical advantage: the ability to constrain our models within the known habits of matter. Combined with what we know of the physical world—the habits of matter and space, as well as biological and psychological "laws" at higher levels of organization—this audit allows us to collapse abstract concepts into their constituent material substances and processes. It identifies "ghosts" that hold no real meaning in the external world and reveals them as figments of imagination.
The mental models we possess of the external world (aka Nature) are actual physical configurations: interconnected neurons and chemicals (matter) and electromagnetic forces. However, the match of this map to the territory—in the words of Alfred Korzybski—is not perfect. Indeed, it can never be, or it would be the territory. The effectiveness of a system's actions depends on how accurately its mental models match the external world—when the map deviates too much from the territory, it fails to predict the right consequences of its actions—preventing the system from satisfying its objective function. To maintain this fidelity, the system must engage in a procedural habit of updating the map. This is the architecture of rationality.
★ Pragmatic Epistemology
Epistemology has traditionally been a search for absolute certainty—a pursuit that has proven to be a dead end. In Nominalist Determinism, we start from a position of practical utility. To build a coherent model of reality, and to understand the role of rationality, we must be surgical with our definitions:
Fact: A feature of the external world that exists independently of observation. It is a configuration of matter, a material habit in space.
Hypothesis: A proposition regarding the reality of a fact. It is a binary testable prediction of the external world.
Data: Raw, uninterpreted signals captured from the external world (the territory).
Evidence: Data that has been processed and found relevant to a specific hypothesis.
Belief: The internal state representing the probability that a hypothesis is correct.
Model (Theory): A structured framework consisting of a set of interrelated hypotheses.
Knowledge: A label for a belief held with a high degree of certainty (in science typically ≥ 95%).
In this framework, there is no functional difference between knowledge and belief; to know something simply means that we have a high confidence in a particular belief. In the vernacular, we at times say we know something even when we are not certain at all—for example that we put the keys on the table despite being aware that we may have absentmindedly left them in the jacket. It is futile to redefine knowledge to exclude cases with such uncertainty. Instead, we have to adopt the view that the word simply means that we feel confident, based on the available evidence. According to memory, we (subconsciously) estimate that the probability that we put the keys on the table is 80%, thus giving it a 20% chance of the keys being elsewhere.
★ The Four Sources of Knowledge
There are exactly four sources on which we base our pragmatic knowledge. If prompted to explain how we know something, these are the only available options in Nominalist Determinism. Sources such as revelation and feeling are voided.
Instinct (ancestral knowledge): Pre-loaded models encoded in DNA.
Experience (observation): Direct sensory data capture.
Testimony (authority): Transmitted data from others.
Inference (reason): The process of using existing data to construct hypotheses and models through deduction, induction, and abduction (per C.S. Peirce).
Deduction: Necessary conclusion based on given premises.
Induction: Very suggestive statistics.
Abduction: Fitting a model to the facts.
★ The 3 Intelligénces
While the engine and the output are commonly conflated under the general label of "intelligence," I insist on separating rationality as a distinct procedural habit. We categorize the efficiency of this loop into The 3 Intelligénces:
Rationality: The procedural fidelity of the update protocol. It is the willingness and ability to take in raw data, filter it into evidence, and update beliefs about hypotheses.
Data → Evidence → Beliefs
To successfully update, the system must learn to love being wrong: managing the negative feelings of realizing that past beliefs were faulty and embracing the process of discovery is key to being rational.
Abducing Intelligence (Power of Abduction): The specific mode of inference (the fourth source of knowledge) that constructs the most likely explanation (the model/theory) to fit the facts as they are believed to be. We must categorize it as a distinct type of intelligence, since it is conceivable that a system can be rational, accepting new beliefs, and yet be incapable of constructing theory from those facts.
Facts → Theory
In the Sherlock Holmes canon, Dr. Watson represents a system with high rationality but low abducing intelligence. He observes the same facts as Holmes and is perfectly willing to update his beliefs when proven wrong. However, he remains incapable of constructing the mental model that connects the data points. While Watson captures the experience, he does not possess the power of abduction required to synthesize those facts into a coherent theory of the crime.
In the history of science, the Danish astronomer and astrologer Tycho Brahe serves as the real-world equivalent of the rational witness. He possessed immense experience and rationality, spending decades collecting the most accurate astronomical data of his age and using it to dismantle the illusions of ancient astronomy by proving comets moved through what was previously believed to be solid celestial shells. And yet he lacked the abducing intelligence to construct the elliptical model that actually fit his own observations. It took Johannes Kepler to look at Brahe’s exhaustive facts and abduce the planetary laws that Brahe himself was theoretically blind to. Brahe held the keys to the territory, but he could not draw the map.
Model Match Intelligence (aka understanding): The resulting fidelity of the match of the constructed model (the map) to the external world (the territory). It is the static measure of how well the model corresponds to the territory.
Theory → Prediction → Action
Hark back to Nominalist Determinism and Intelligence for a more thorough analysis of Model Match Intelligence.
★ Types of Failure
Any update to the map involves friction. This is the material resistance encountered when dogma—current neural habits—must be physically overwritten by new evidence. Because the brain is a biological organ, it is subject to limbic volatility—the tendency of the ancient brain to favor existing, etched habits over the material labor of rational updating.
Irrationality: This well-known failure occurs when the pre-frontal cortex (PFC)—the evolutionarily younger part of the brain—fails to restrain the ancient limbic system: high-arousal states disrupt any of the 3 Intelligénces when emotions run amok. With this emotional perturbation, the system is unable to function intelligently, disabling the rational and/or abducing protocols, or overriding the predictions made by the models. The ancient emotional heuristic is in effect, sanity has broken down, and the system is locked by emotional anchoring.
Unrationality: This is a failure of prioritization within the PFC. As there are two distinct steps in the protocol, there are two ways that dogma—those hitherto accepted beliefs integral to a feeling of identity—can interfere: i) the system receives the data but fails to turn it into evidence, or ii) evidence is generated but not converted into updated beliefs. The rationality protocol is broken, the update fails to execute, and lacking new facts the model remains unchanged.
★ The Pillars of Rationality
Models—whether housed in biological brains, artificial neural networks, or human institutions—require continuous maintenance as they interact with the external world. High rationality is the prerequisite for a system to acquire the power of prediction necessary for success according to the system’s objective function; conversely, systems with inferior rationality are cursed with models that generate faulty predictions. To preserve predictive accuracy, a system must adhere to the two non-negotiable pillars of rationality:
Primacy of observation: Nature is the final authority. Every internal model must defer to direct sensory evidence and material facts, as no amount of logic, dogma, or consensus can override the evidence from the physical world.
Willingness to update: The model is a temporary map, not a static truth. For a system to remain functional, it must maintain the mechanical capacity to overwrite its existing neural habits the moment Nature proves them to be faulty predictors.
Bjørn Østman, Svendborg, April 2026.
