Book review

The Case Against Reality

How Evolution Hid the Truth from Our Eyes

by Donald D. Hoffman

The Case Against Reality by Donald D. Hoffman

There’s a lot to like in Donald Hoffman’s book. He carefully marshals a number of strands of experimental and mathematical evidence to lend weight to his claim that, ultimately, it is extremely implausible that our experience of reality in any way corresponds to how reality actually is. This is a very bold claim, and I found the thought experiment of thinking this through to be very interesting, even if ultimately I was unconvinced by the core argument.

Hoffman has spent his professional career thinking about this problem, and he’s build a very interesting view based upon it. He doesn’t deny the existence of a shared “consensual” reality – that you see what looks to you like a chair and a table, and I see the same. Instead, he denies that the chair and table exist in anything like the form we think they exist in – as three dimensional medium-sized objects. Indeed, he denies they exist at all when we’re not looking at them. Instead, our perception of them causes them to appear in our “user interface” for the world.

One of his most powerful metaphors is this “interface theory of perception”. This says that our experience of reality is rather like the graphical user interface of a computer: the rectangular object on the screen which we click on to open a file isn’t, actually, anything at all like the file really is in the computer – where it exists as a series of variations in charge or magnetic field on a solid state or spinning disk, and what looks like a file opening on the desktop is actually the cumulative action of millions of transistor state changes.

To convince us that what we think of as the world really is not at all what we think it is, Hoffman brings in a number of pieces of evidence. He spends a fair amount of time on our visual system. Here he brings together a number of visual illusions to show quite the depth of processing – with a corresponding opportunity for errors – that happens between photons striking our retina, and our perception of a chair, or a colour, or a beautiful person. This is entirely convincing to me: it’s clear that a vast amount of detail of what we might perceive is both hidden from us, and also filled in by informed guesswork by various networks in our brain.

Hoffman also points to modern physics, where quantum mechanics really does tell us some apparently strange things are happening out there. Photons from distant galaxies, millions of years in transit to us, apparently have their entire history affected by actions we take now. I feel there are other stories that we might tell about all this – from Everett’s multi-world hypothesis on upwards. Hoffman doesn’t dwell on these, but instead uses them as another explosion under our foundational belief that the world is what it looks like.

Hoffman also has some mathematics behind all of this. He has a model of consciousness – as a set of Markovian kernels (basically, things that choose amongst outcomes with different probabilities). And he also has what he calls the “Fitness Beats Truth” theorem. In this he pits mathematical creatures which prefer “fitness” against those which prefer “truth”. Given such a construction, in the framework of the game theory he uses, it’s no wonder that the creatures which prefer fitness tend to out-compete those which prefer truth: he defines fitness as the most important thing in his framework, so of course such creatures win. For me, this doesn’t really answer the question of why, in the real world, fitness wouldn’t more or less track truth. Isn’t a creature which depends on features of the landscape as they actually exist – i.e., a creature for whom fitness broadly corresponds to the state of the world – out-evolve a creature whose fitness was ultimately a fantasy? That’s why we haven’t all become addicted to the chemicals in the world which give us the most pleasure – because ultimately, fitness needs to track back to useful features of the world, rather than the ecstasy of a drug.

Despite my reservations, I did enjoy the book, and the arguments I had with it. I’d give it three and a half stars, rounded down to three for fighting too many paper tigers, rather than steel-man arguments against his position.

Leave a Reply