Trust is a Law of Nature

Die Deutsche Version finden Sie hier.

The trust we put in others, and in our institutions, is a sociological feature emergent from biology. It is here that predictions first appear. This is the very definition of life and consciousness: without prediction-making and feedback-based correction, neither would exist. Trust reflects the degree to which your internal predictions have matched the external world in the past. From cells to organs to people to institutions, all these are a prediction-making pattern of matter. When predictions fail, trust collapses. Trust rightly applied is the substrate of survival, as survival requires predicting what the environment will offer, at all scales.

You are the Model

Every feature of yourself is a prediction, a model of the environment, you do not just have a model of the world; you are the model. Sorting through the possible models is needed to find a better one; this is computation. Biological evolution computes through death, also called mortal computation. Culture computes through dialogue, also called science: Models that don’t ‘fit‘ the environment disappear. The ones that do so better survive and replicate.

Your skin is a physical prediction of the sun, pathogens, and water. It is «trusts» that these elements will behave in line with what genes learned about the environment over eons. Your kidneys are another physical prediction, in this case of metabolic demand. They filter blood because they expect the blood to contain toxins. In short, you just are the best model of your lived world.

When trust is well invested, when the prediction is correct, we operate efficiently. If you can no longer «trust» your kidneys, you are forced to outsource that prediction to a dialysis machine. This is an existential tax, an increase in external complexity because the internal model has failed. The whole economy can be seen through these lenses, as every product is a crutch to our biology: be it a science book, a tooth implant, a vehicle, or a Thai meal. Trust is a bet that something we need will be provided by the expected channel.

If the bet is correct, it collects more resources than it took to create it. An existential «profit». If the bet is wrong, it collects fewer resources than the bet took to create. An existential ‘loss’.

In physics, we call this the Free Energy Principle. Anything that is alive, and continues to be so, is actively engaged in minimizing expected «surprise», also called entropy.

You are the last piece of a set of successful predictions that is 3.8 billion years long, also called evolution. As the saying goes, «You are your own existence proof.» The mere fact that you are sitting here, reading this, proves that the pattern of matter that you are has predicted the environment well enough to avoid extinction, and has been doing so for a very long time, roughly 40% of the age of the universe!

«You are the last piece of a set of successful predictions that is 3.8 billion years long»

The fine tuning of the model that you are was made through evolution, which consists of variation and selection. Variation allows testing if something different works in the environment, and the versions that match it best survive and replicate more, they are existentially selected for.

Variety is a Response to Lack of Trust

If every individual held the same model, a single «surprise» would lead to extinction. To perpetuate existence, a variety of models is created at every level, this is the sense in which diversity is needed and emerges naturally, bottom-up. One good example of this variety is seen in maturity, from the high-risk foolishness of the young to the high-precision stability of the elders. The prefrontal cortex ¹ doesn’t finish developing until the age of 25. Because there is no physical constraint inhibiting its maturation, it seems like the playful foolishness of youth is adaptive.

The «fool» explores the high-risk edges of the environment, but in so doing resolves uncertainty and expected surprise. Mathematically, an evolutionarily stable strategy requires this mix of dull-minded first-order thinkers and a few highly sophisticated people. We differ in personality, intellect, and every conceivable dimension because a populated, diverse variety is the way to remain adapted to an unpredictable world. In the areas where the world becomes reliable and trust can build, however, variety is decreased.

The Selection of the Better Models

The variety that is needed to survive in an uncertain world is not free, it requires resources, that is why it is reduced when regularities of the world are found, parallel processing is expensive, but it is done until we find a «solution». This is the increase in trust, the decrease in the need to diversify our «existential portfolio».

This selective pressure is the «chiseling» of the wrong models. In an uncertain world, we bet on diversity; in a reliable one, we bet on precision. In the words of Einstein: “keep everything as simple as possible but no simpler”. When a regularity is found—be it the sun rising or the existence of water in the atmosphere—the variety of our «existential portfolio» contracts. We stop being a thousand disparate hypotheses and collapse into a single, high-precision bet ^{2 3}. Trust, then, is the mathematical reward for predictions that have matched the world. It allows us to stop exploring the «edges» and start building on the «center.» This contraction can be dangerous: if we stop diversifying because we falsely believe the world is stable, that we found a good enough model too soon to commit to it, we become «delusional.»

Therefore, the golden mean of trust must be found, too much or too little trust is a pathology, this is seen in the form of the two inferential errors:

• Type I Error, illusion: Over-trusting internal noise. seeing something that isn’t there.
• Type II Error, blindness: Under-trusting real signals. not seeing something that is there.

Institutions are Extended Biology

Just like organs emerge from cells and bodies emerge from organs, institutions emerge from bodies. Coordination is needed at every level, the hormones that coordinate activity within a body are like the memes and tropes that coordinate our institutionsThe same updatability of knowledge reigns supreme here as well, the shorter the feedback loop the faster we improve our models and the more robust our trust becomes.

Switzerland systematically ranks above almost every other country in trust ratings, this is not a coincidence, direct democracy and subsidiarity shorten the institutional feedback loop.

In Switzerland, decisions are pushed to the local level, subsidiarity, allowing the “model” —the law—  to more readily and with a higher resolution synchronize with the “substrate” —the people—. Every referendum, every vote, is a real-time Bayesian update ⁴. The Swiss model is a form of governance and a self-selecting, self-evidencing, survival algorithm ⁵: it maximizes the alignment between collective expectations and behavior, minimizing “surprise” at a societal scale. Trust in the institutions is high in Switzerland precisely because the feedback loop is short and transparent. This allows the Swiss to remain mutually predictable and minimize their joint «surprise» —i.e., entropy—.

Whether bad ideas are eliminated cheaply, within, by an individual changing her mind, or expensively, without, by having their carriers die out, depends on the length of the feedback loop. The Soviet Union was the biggest experiment of long feedback loops, missing the economic signals afforded by market prices: it couldn’t update fast enough and was existentially falsified. Every hierarchy, no matter how wide and tall can fall if it gets the world wrong enough.

Collapsing Is and Ought

All values remain provisional and updatable, perhaps except this one, that values remain updatable holds the cultural position of the first law of thermodynamics. Culture—memes— emerged from biology—genes—to map the environment faster than genes could. Opposing free speech is regressing a couple million years of evolution. If we can talk about things, our knowledge updates faster than one funeral at a time. The purpose of dialogue is to let ideas die instead of us. Where open inquiry reigns, trust doesn’t need to be put in the wrong places for long, this is the scientific method in essence.

«All values remain provisional and updatable, perhaps except this one, that values remain updatable holds the cultural position of the first law of thermodynamics»

The traditional distinction between facts — what «is»— and values — what «ought» to be— is a specious philosophical boundary. The ‘naturalistic fallacy’ —that science and morality, understanding and values are orthogonal — is itself a fallacy, there is nothing unnatural about that idea, its just wrong. Where else than from the right is, the right understanding, the right explanation, the is that matches the environment, could we derive an ought from? Moral rules and ethical convictions are like all other pieces of knowledge, predictions that must be hosted by a physical «is»—a human body, a brain, or a population—. They are themselves made of other predictions.

If a population adopts an «ought» that is functionally incompatible with reality, it leads to their disappearance. If an ought leads to an existential ‘is not’ — extinction—, that belief is proven wrong in the most literal sense: there is no one left to sustain it. Virtue and vice, morality and immorality, are defined by fitness in the long term. What else could they be defined by? Norms that allow societies to persist tend to stabilize, while those that destroy their carriers are chiseled out of the world. Just like an exuberantly virulent virus runs out of hosts to infect and disappears. This is why morality across all cultures converge on a specific set of «workable rules»—such as honesty and reciprocity.

The distinction between science and morality is an illusion, a dogma-inducing regressive one. Every ethical stand is a provisional hypothesis to be falsified by further evidence. From where, if not from understanding how the world works, would we derive rules for how to behave? Some will say religion, but religion very much embodies, in an implicit way, countless iterations of this variation and selection, signal —and noise, sometimes a lot of hard-to-update vestigial noise— harnessed through the centuries.

In other words, morality and our internal models are governed by the same laws of nature: every thought, choice, and action is, in and of itself, a prediction, tested and updated against reality.

To live is to act on predictions that have been refined over billions of years. The moral universe is not above us; it is within the patterns that have survived long enough. Every breath we take, every choice we make, is both proof of what works and a wager on what can survive. To exist is to already have solved, however imperfectly, the problem of how to live by correctly allocating one’s trust.