It’s unusual for me to write about nonfiction in this newsletter but Brian Klaas’s Fluke (2024) is an unusual book. Klaas is associate professor in global politics at University College London and describes himself in the book’s introduction as “a (disillusioned) social scientist” (p. 11). His Substack is “The Garden of Forking Paths”, taking its title from a story by Borges. The book could be described as a popular science synthesis of various approaches to understanding the world, from chaos and complexity theories, through neuroscience, evolutionary biology, a bit of narrative theory, physics, economics and history. Oh, and philosophy — mustn’t forget that.

Klass is a disillusioned social scientist because he has found that the social sciences promise something they’re incapable of delivering: predictability, a straightforward, traceable relationship between an action and what follows from it, a world that, given enough data about it, we could expect to “grasp” or “comprehend”. He assures us that that’s not the kind of world we live in and, perhaps more to the point, we’re not the kind of beings who could expect to achieve a full understanding of the world, even if it were straightforwardly predictable.

We have evolved, not to perceive accurately our surroundings, but to survive (at least long enough to reproduce), amidst many threats, both known and unknown. If we’re in a situation that we feel might be dangerous, it’s more important to be able to react immediately than with a full understanding of the circumstances. In Klass’s terms, a Shortcut Creature will, on average, survive longer than a Truth Creature will. So, we haven’t evolved to be Truth Creatures. That’s to say your perceptions are distorted; you might as well accept that unavoidable fact.

Not many readers will be surprised by that insight but Klaas synthesizes it with several others which add up to a less familiar world view. Of the book’s thirteen chapters, the first eleven run through various reasons why the world is neither predictable nor susceptible to our control. Very small changes in “initial conditions” lead to enormous differences later on, or in another part of the world. We may like to believe that everything happens for a reason, but it isn’t true. Human behaviour, no less than that of locusts, is greatly changed when we “swarm” (as we do under certain conditions). Where you are in history, or your geographical location, can also make a significant difference to your life.

There’s a term I’ve avoided using so far: “linear”. The relationship or connection between those initial conditions and the eventual outcome could be described as “nonlinear”, and that would be true in the sense that a mathematical equation attempting to describe that relationship would not be a linear one. There isn’t a line — “straight line” would be a tautology: in geometry lines are straight by definition — between what is happening right now, and whatever happened in the past that caused or affected it. So “nonlinear” is not inaccurate but it could be misleading, in that it might lead some people to think that because we can’t trace the chain of cause and effect between a butterfly’s wings in Uruguay and a storm in Yunnan, that the chain doesn’t exist, or can be treated as if it didn’t.

In the twelfth chapter, Klass lays out a compelling case for determinism. While it may not be true that “everything happens for a reason”, that doesn’t contradict the proposition that everything that happens has a cause. Nothing “just happens”. The universe isn’t random, according to him.

When I first got the opportunity to mess around with a computer in 1979 — my mother, a teacher, was told by the school principal to take the new Apple ][ home and see what she could do with it — the thing I was most surprised to learn was that computers can’t generate random numbers. They can’t just pluck a number, any number, out of the air. On the other hand, they do a very convincing job of appearing to produce random numbers — the computer manual called them “pseudorandom”. I understand that things have got a lot more complicated in the meantime, but in 1979 a computer produced a pseudorandom number by shaving the next digit off a very large prime.

It didn’t happen immediately, but eventually I began to ask myself whether randomness exists anywhere. A computer can produce the convincing appearance of randomness, though not the real thing. What if the same were true of the world at large? We’re told that evolution is driven by “random mutation”, haphazard changes that occur, apparently spontaneously, in the reproductive process, which may or may not make an organism more likely to survive. But is there any requirement that the changes be truly random? Mightn’t they just as well be haphazard changes that occur, apparently spontaneously, in the reproductive process simply because of the complexity of the process itself, and the multiplicity of factors, many of them external, acting on it?

In the mid 2000s I bought a book by Nassim Nicholas Taleb, Fooled by Randomness (2001). It quickly transpired that I had been fooled by the title. I was disappointed to find that by “randomness”, Taleb meant something different from my understanding of the term. I was interested in whether things could “just happen”, without being caused by something else: could something new just pop into existence for no reason — and if it could, how could we be sure it had? Taleb (as far as I remember) was more concerned with surprises, the unforeseen, the unpredictable, irrespective of how those things came about. Twenty years later, I’m delighted to find that Fluke is the book I thought I was buying when I ordered Fooled by Randomness.

Klaas argues that that our universe is wholly deterministic and that randomness plays no part in it — except at the subatomic level, which we’ll come to. Everything that happens is caused by something else that happened before it, usually multiple things, each of which has been caused by something that preceded it. It’s just one damn thing before (and after) another, right back to the big bang, and all happening simultaneously with countless other things some of which affect our thing. It’s complicated.

If the world is deterministic, then everything is, in effect, scripted. Determinism is the notion that change is simply a function of initial conditions (the way things are at a certain snapshot in time) and the natural laws of the universe. Everything that happens is directly and completely caused by that which came before it, a nonstop chain reaction of endless causes and effects, unfolding according to physics. (pp. 221–2)

This is fairly uncontroversial but some of the implications are startling:

The exact state of particles in the instant after the big bang, 13.7 billion years ago, determined the state of the universe in the next instant, which determined what happened in the next instant, on an on, endlessly, until the present moment. If causes and effects are fully determined in an unbroken chain of events, that means that if you brushed your teeth at 8:07 this morning of if your dog barked after seeing a squirrel in the yard, then that was fully and irrevocably determined by the initial conditions of the universe 13.7 billion years ago during the big bang. (pp. 222–3)

Yet there does seem to be an important source of randomness in the universe. Random activity, if it happens at all, does so in subatomic particles. As Klaas tells us, there are fierce disputes between scientists as to what is to be made of the bizarre behaviour — he calls it “quantum weirdness” — of those particles. While nobody knows for sure what is happening, Klaas tells us that the Copenhagen interpretation is the “dominant” one. He summarizes it like this:

the Copenhagen interpretation implies that at the tiniest levels of matter, some aspects of our world are completely random, governed not by determinism, but by probabilities. The interpretation implies that some changes at the subatomic level are unlike anything else in the known universe. They are genuinely uncaused — meaning true randomness reigns. (p. 228)

Part of me instinctively wants to insist that, if randomness is to be found at the quantum level, it must equally have some effect in the big world, in the reality that we misperceive with our senses. But that doesn’t seem to be the case. Schrödinger’s Cat is a thought experiment that demonstrates the absurdity that must result if quantum behaviour could affect the big world. You can’t have a cat that is superposed between death and life, waiting for the collapse of the probability wave, even if the particle on which its state is dependent is in that indeterminate condition.

Is it possible that the subatomic world is to some extent random, while the world at large is not? (They are of course the same world: the latter is made up of the former.) Sure, it’s possible. Stranger things have happened, maybe. It’s also possible that the Copenhagen interpretation is entirely wrong. The fact is, we don’t know. Quantum weirdness remains weird — and not understood.

The other main problem with determinism is that it seems to rule out free will. Our brains are physical things, made up of living cells, like the rest of our bodies, and subject to the same laws of physics as everything around them. Unless you believe in a “ghost in the machine” (p. 233), the current state of your brain is determined by its previous state in combination with events that influence it from outside, which in turn are determined by whatever went before, and so on, all the way back to the beginning of the universe. Where does the free will fit in?

Klaas acknowledges that there are compatibilists (p. 234) who believe that determinism and free will are compatible. (I think I might describe myself as a “naïf compatibalist”, in that I’m inclined to believe both in determinism and free will, but I’m not able to explain how they can be reconciled.) Klaas cites Sam Harris as a neuroscientist who pours scorn on compatibalism, pointing out that to make it work its adherents have to distort the idea of free will beyond recognition. Klaas himself is less doctrinaire than Harris is, but nevertheless seems close to being a “hard determinist”:

I don’t believe in free will, though I acknowledge that these questions are mind-bending, baffling and mysterious. We don’t understand consciousness, so it’s plausible that some new discovery will change how we answer this question. In discovery, never say never. But if libertarian free will is indeed what we have, then pretty much everything we know about science would have to be wrong. Compatibalist conceptions of free will aren’t at odds with science so much as they’re redefinitions of what it means to be free. (p. 237)

The reference to our not understanding consciousness is what strikes me as significant in this passage. Much earlier in the book, Klaas has acknowledged the intractability of the consciousness “problem”:

The greatest mystery of all is consciousness, and we don’t understand it. Since 1994, the thorniest challenge has been called the hard problem of consciousness, the term coined by a titan of modern philosophy, David Chalmers. Humans have long been baffled by the so-called mind-body problem, the question of whether there is something fundamentally different between what we think of as our minds and the physical, chemical structures of the brain. (p. 110)

For myself, I understand consciousness so badly that I can’t even grasp what it would mean to understand consciousness. How would one understand it? What is it that we’re trying to do?

These debates and questions bear a startling resemblance to religious controversies of the seventeenth century. Some time ago, I wrote about one of Andrew Marvell’s prose works, Remarks upon a Late Disingenuous Discourse (1678), in which he defended a nonconformist minister, John Howe, against the attacks of another, Thomas Danson. Danson was a Calvinist and Howe was a follower of Richard Baxter, who attempted to steer a “middle way” between Calvinism and Arminianism. Arminians believed in free will, which Calvinists thought was a derogation from God’s omnipotence. Howe had argued that it was clearly within the competence of an omnipotent God to create a creature endowed with free will, who was thus capable of acting against God’s commands or wishes.

It struck me that there might be an analogy to be drawn between Howe’s argument and the very different conceptions of determinism and free will that Klaas is writing about. Our universe is, I agree, wholly deterministic. But it has been around for 13.7 billion years, getting increasinly complex all the time. Evolution has been happening in it for many millions of those years, resulting in creatures, organs and behaviours of wondrous variety and ability. One of the things that evolved is that consciousness that Klaas acknowledges is such a mystery.

So, is it not possible that, in this aeons-long, deterministic, extremely complex, ghost-free universe, there evolved a brain capable of making choices, taking decisions, evaluating alternative courses, arriving at judgments, independently of its conditions? Such a brain would not, of course, be completely free of constraints, any more than would be the body it animated. Sam Harris might ridicule the idea that the will exercised by such a brain could accurately be called “free” but it’s hardly much different from what we commonly understand as free will. Perhaps we should stop calling it “free will” and come up with a more accurate term.

Is there any reason why a wholly deterministic system shouldn’t be able to accommodate a decision-making, autonomous intelligence that evolved (through deterministic processes) within that system? I’m not claiming that’s what happened, only that it’s not impossible.

Klaas accepts that most of us, in normal circumstances, feel that we’re able to act independently, to choose, to take a different course from one laid out for us. We don’t experience life as if we were helplessly following a predetermined program. Perhaps this sense of freedom is an illusion. Perhaps, but there’s a real possibility that it isn’t. That means, I think, that it makes sense for us to try to behave as if we were independent, self-acting, free agents, even if we’re not sure that that’s really the case.

I’ll be back to fiction next time.

Edition: John Murray paperback, 2025.