In contrast to popular opinion, which says that humans are both part of nature and irrational, Rational Observer Theory assumes that nature is rational.
Although technically an “assumption” of the theory, logical/rational relations are certain, and to call them an “assumption” is misleading. I would not write “I assume that I exist.” But since the assumption of irrationality is widespread, rationality may appear to many as a questionable assumption of Rational Observer Theory.
Unfortunately I was naive to this issue in 2003 when I began learning about probability theory and developing my theory of the brain. I took it for granted that nature is rational, without ever once thinking that it could possibly be irrational. It never occurred to me that the concept of irrationality might be capable of providing insight or explaining some phenomena. Rather I implicitly assumed that everything about nature and the brain is explained by information or its absence.
I have never since doubted my “assumption.” But over the years I have increasingly recognized the importance of rationality as an issue for philosophy and science. I have also come to realize that a large majority of scientists believe, at least implicitly, that nature, and the human brain in particular, is irrational. I have named this “The Irrational Brain Hypothesis.”
Given the widespread belief in irrationality, it is not surprising that confusion and skepticism have been typical responses to my theory of the brain. Unfortunately I have never adequately addressed the issue of rationality in my publications. I have however published some of my ideas on the topic (Fiorillo, 2010; Fiorillo, 2012; Fiorillo, 2017) and I plan to publish more.
Though I do not attempt to adequately explain the relevant issues here, I provide examples below of how our notions about rationality relate to physics, probability theory, and neuroscience. I also explain why physical systems cannot be irrational, and I clarify why some behavior appears irrational.
The Reasonable Effectiveness of Mathematics in the Natural Sciences
Eugene Wigner, who won the Nobel Prize for his work on symmetries in physics, wrote a famous article titled “The Unreasonable Effectiveness of Mathematics in the Natural Sciences.” On the second page he wrote:
“The enormous usefulness of mathematics in the natural sciences is something bordering on the mysterious
and … there is no rational explanation for it.”
The effectiveness of mathematics is reasonable in my view simply because mathematics is an expression of reason, and nature is rational.
A secondary explanation is that the mathematical challenge in science is simplified by the fact that our predictions are usually conditional on only a small amount of information, like a position and velocity, and less information results in greater symmetry and simpler math. The challenge is further simplified if our only interest is accurate prediction, because there are countless mathematical rules (algorithms) that all capture the same relation between past and future, and any one of them can be used to accurately predict the future given the past (as discussed in Science’s Problem with Reality).
The Irrational Brain Hypothesis
The Irrational Brain Hypothesis is the name I have given to the nearly universal belief that people are irrational. I am hoping that the name catches on. Irrationality should not be regarded as an obvious and known fact. Rather it should be seen as a hypothesis to be debated, perhaps for one or even two minutes, before being rejected. I discuss The Irrational Brain Hypothesis in greater detail on the page of the same name.
If rationality is not inherent to all of nature, we need to explain how it arose. It is widely believed that rationality emerged through human evolution. Those who believe in the The Irrational Brain Hypothesis claim that the human brain/mind is approximately rational, and that this approximation improved over the course of evolution. Likewise, a brain is presumed to become more rational over the course of development.
That the Irrational Brain Hypothesis is true is a foundational assumption of virtually all research in the fields of Bayesian Brain Theory and behavioral economics. It is also fundamental to work on “AI” and “The Myth of Artificial Intelligence.”
Bayesian Brain Theory is particularly relevant here because my own theory is part of it. Others in this field believe that because the brain is inherently irrational, to “approximate rationality” and make predictions it must physically implement a sequence of steps to perform the math of probability theory so as to calculate probabilities (see Bayesian Brain Theory and section below on sequential logic). I argue that because nature is inherently rational, the brain has no need for performing mathematics or calculating probabilities. Formal mathematics is useful to some of us, but the vast majority of brains have no use whatsoever for it.
Probabilities Rationally Measure Evidence
Rational Observer Theory uses Bayesian probabilities to rationally quantify evidence. As summarized in The Meaure of Evidence and in Fiorillo and Kim (2021), I take the view that a probability can in principle be the uniquely correct measure of evidence, and in this sense the relation between probability and evidence/information/knowledge is rational. This is the “objective Bayesian” view of probability, developed by Laplace, Jeffreys, Cox, Jaynes, and others. Other Bayesians are subjectivists and view this ideal of objectivity/rationality as unattainable, even in principle.
What would be the value of human intellect in a world in which an objectively correct measure of evidence is literally impossible? I would quit science altogether if I believed that evidence is immeasurable in principle.
There Can Be No Physical Definition of the Irrational
The central problem with the notion that nature is irrational is that no one can even begin to define what it means. It is easy to define the rational and irrational in language and mathematics. One can even play slightly clever games with words or numbers in an attempt to make a rational argument that rationality is not rational, as in the “liar’s paradox.” But what one cannot do is to define what it means for a physical system to be irrational, or to give even a single example of an irrational relation between two or more observations.
No one can do this because it would require asserting that ‘X is true and X is false’ is a true statement. For example, ‘the earth is both flat and not flat at the the same time.’ Some scientists have a lot of nerve, but apparently not that much. We cannot see the earth as both flat and round at the same time, and even if we could, we would not believe it. As another example, consider that scientists measure the distances between three bodies, and then they find that their numbers are inconsistent with the triangle inequality. Of course they would not even consider the possibility that the inequality is untrue or that nature is irrational. They know that something went wrong in their measurement.
When reason appears to clash with observation, reason always wins. Nonetheless, the Irrational Brain Hypothesis has been very well funded for decades, in both public and private sectors, with numerous scientists and engineers trying to figure out how to modify a physical system to make it less irrational.
The Local Nature of Evidence Explains the Appearance of Irrationality
Irrationality is used to explain a wide variety of behavior, from that of children to dictators. More precisely, it is used to explain away behavior. To label a behavior “irrational” indicates nothing but our failure to comprehend it. What we interpret as irrationality in nature is in fact ignorance (lack of evidence). Different observers have different evidence (which I equate with beliefs), and one observer is ignorant about what another observer knows. If I interpret what you say or do as irrational, my judgement is based entirely on my evidence and my corresponding ignorance of you. I label you “irrational” because your actions are not consistent with my evidence (including my beliefs about what you know).
Children tend to do stupid things from the perspective of adults, and we often label them “irrational.” But they are entirely rational, and behavior we recognize as inappropriate is explained by their relative ignorance. They acquire evidence about the world as they age, but they do not acquire reason.
Evidence is local in space and time. The world is full of contradictory evidence, and so is my brain, but the evidence and contradictions are distributed. Local evidence is never self-contradictory.
One can recognize this through introspection. What I consciously know here and now is just a tiny fraction of the evidence in my brain. I am this tiny fraction of my brain’s evidence. My current evidence (I) may contradict the evidence elsewhere in my present brain, or in my past brain, or in your brain, or elsewhere in this sentence (which grows over time as I write it), but I never consciously contradict my present self in the present. I always exist in the present, and I am always rational.
Every belief is based on evidence, or more precisely, every belief is evidence. I may consciously believe that the earth is round at the same time that I tell you it is flat. But this is not irrational, and it is not necessarily a lie. The neurons that were consciously thinking it is round are not the same as the neurons that caused my muscles to say it is flat. Some neurons believe one thing and some believe another, just like people. This simply reflects the fact that evidence is distributed across space.
Likewise, evidence is distributed over time, and therefore the beliefs of an observer change. There is nothing irrational about believing that X is true, and later believing X is false, and perhaps never even noticing the contradiction. I contradict myself at least once for every sentence I write, even before I type it and revise it again and again. This indicates my limited knowledge and intelligence, but it is not irrational, because my beliefs and actions are based on evidence, not proof, and contradictory evidence is not the same as logical inconsistency.
When contradictory evidence comes together in space and time in a neuron, it cancels, as one would expect. This is nicely illustrated in visual neurons of the lateral geniculate nucleus. If such a neuron receives synaptic excitation from retina that is evidence for light in a particular region of space, it also receives “opponent inhibition” that is evidence against light in that same region of space. Thus the neuron sums and “weighs” contradictory evidence.
The Unattainable Ideal of Sequential Logic
Much of what we think it means “to be rational” corresponds to what I call “sequential logic.” This is what we attempt to do when we think, and when we do math. We try to follow a sequence of rational relations, and if we do it properly our thoughts, or mathematical expressions, are logically consistent with one another over time and over each step of the calculation. But because evidence is temporally local, and thus changes over time, sequential logic across time is an unattainable ideal.
Humans are the only things we know to be capable of performing math (our computers merely do something we interpret as performing math). However, it is an understatement to say that we do not do it well. Even our best mathematicians make one mistake after another (though usually never written, and very rarely published). This can result in formal statements or equations that are illogical. But it does not mean that the mathematicians, which are physical things, are illogical. When mathematicians do calculations, they remain animals trying to make sense of the external and future world given their limited evidence and corresponding uncertainty. The fact that their evidence at one time contradicts their evidence at another is not a logical inconsistency within the mathematicians at any given time. Our serious mathematicians effectively deal with the problem of illogical equations by repeating a calculation over and over until they have a great deal of evidence that it is correct.
The mathematics of humans is an aberration of nature. It is a fortuitous aberration, since it allows us to characterize the rational relations inherent throughout nature. The effectiveness of our mathematics is not unreasonable, but it is certainly remarkable.
Illogical/irrational relations are false and do not exist. Existence can only consist of those propositions that are in fact true now or might be true in the future (regardless of what a local observer might believe about them). Within Rational Observer Theory, the relation between evidence and propositions, and between the evidence of one observer and that of another, is rational.