The Anthropic Principle

The anthropic principle addresses the issue of what we can infer about the universe we inhabit simply from the fact that there's somebody inhabiting it who's capable of the kinds of introspective self-analysis necessary to worry about things as philosophically convoluted as the anthropic principle. It should consequently be no surprise that quite a lot of what gets said about the anthropic principle is (to put it mildly) hogwash. At least one author I've read (whose writings I generally respect) treats the entire topic with disdain for the eminently sound reason that he's never seen any account of the matter in which he could see anything that isn't a heaving pile of bovine ordure. In the following, I endeavour to describe the little sliver of the story that's more substantial than that; either the sources he'd read missed it, or he failed to see this facet of what they were saying.

Prelude: cogito ergo sum

To get a clearer idea of the matter, step back a little way. René Descartes (1596 to 1650) is well remembered for an argument normally expressed as I think therefore I am. On the face of it, this is silly nonsense: you can't start the sentence before accepting the am-ness of an entity to be referred to as I who is to be said to be think-ing before you can infer, from that, that this I can be said to am. To understand what he was on about, you have to look at the question he was studying: is there anything of which we can be sure ?

The classical answer to this was No, justified (loosely speaking) by the observation that our senses and wits are unreliable, so anything we're ever sure of may, none the less, be hopelessly wrong and we'll subsequently realize how silly we've been (or our antecedents were) to believe it. The world is flat; when things burn they're giving off flogysten; and the distant descendants of successful conquerors and wealthy land-owners make the best rulers – obviously; and compelling reasons were found to justify each of these beliefs. What René was driving at was that you can't persuade me that I don't exist by arguing that my existence is just an illusion; because if I'm capable of being deluded into supposing I exist, there must (however you come at it) be an I that's entertaining this delusion. You might be able to persuade René to doubt his beliefs about his shape, structure, form and other overt properties, but arguing that there was no René to be subject to these delusions wasn't going to cut any ice with the René that you'd got to persuade was having those delusions.

Modern theoretical physicists, substantially following Poincaré, have a much easier time of this kind of question; frankly, we don't care if the universe of our senses is real or not, we just want to produce a nice description of the collection of observations – illusory or not – that we are pleased to call reality, and we laugh at philosophers when they tell us we shouldn't be calling it real. After all, what we call it is just a matter of sophistry. The philosophers are hung up on the need for everything to be a platonic ideal with unimpeachable true-ness to it; they aren't comfortable with a notion of reality that falls short of this ideal. Their own argument (about the fallibility of our wits and senses) persuades us that reality doesn't fit that bill, so we don't care whether they're comfortable with it. If you want to call our observations illusions, we're happy to answer that these illusions possess some interesting structure which it pleases us to analyze and characterize. It happens to make our discussions of them a whole lot easier if we entertain the delusion that these illusions are real (a term with no other meaning, as far as I can see) and that what we call observations deserve to be tagged as being of a thing it pleases us to call reality or the universe we inhabit.

All the same, there is something interesting for modern theoretical physicists to dig out of René's neurotic attempts to give philosophical credibility to the scientific revolution of his time. Hiding in all that sophisticated gobbledygook is a crucial observation: even before we begin doing all our clever experiments to discern simple patterns to help us make sense of the complex mess of our general observation of our universe, we have one interesting and curiously fascinating given – our own existence.

What Doesn't Our Existence Imply ?

What the anthropic principle says is that there's no such thing as a universe in which there are no sub-systems interesting enough to be capable of looking at that universe and noticing that it's there and they're in it. We can postulate such universes all we like, but: not only aren't they this universe (the one our illusions persuade us to believe we are observing); and not only aren't they a universe in which some weird entities we can't imagine are strapped into virtual reality systems that are fooling them into experiencing something that they understand as this universe so as to fool them/us into experiencing the illusion of observing these illusions; furthermore, by specification, such universes don't get to be conceived of as the universe by anything, because there's nothing in them to so regard them.

That said, it is worth bearing in mind that we don't necessarily have any idea of the ways universes (possibly beyond our comprehension) may contain (sub-)systems capable of noticing the given universe(s). This is the first sense in which the anthropic principle doesn't necessarily tell us much. But there's more …

Strictly, what I've just stated is the weak anthropic principle; there's also a strong form which (more-or-less) asserts that the sub-systems under discussion are us. The difference is (again: more-or-less) only that the strong form throws in a significant piece of experimental data – namely, what we know (or are willing to throw into the discourse) about ourselves. However, (for all that you may have gained the impression I treat them with contempt) the philosophers have a point: quite a lot of what we think we know about ourselves may just be a bundle of complex illusions. We really can't prove that we're not all wacky 7.39 (exactly, but that number's base eπ, and (1+√5)/2 of those dimensions are time-like) dimensional polyps hooked up to a virtual reality system that's playing merry whack with our perceptions to ensure we fail to break free of the delusion that we're … well, what we perceive ourselves to be. All the anthropic principle tells us is that (whatever we are) the actual universe of our real reality is one which does support a context which does sustain some creatures capable of sustaining the illusions we are familiar with as observed reality.

It's notable (bearing in mind the virtual-reality satire just alluded to) that we might be experiencing a set of illusions under which, in fact, it's not possible for any entity to exist which is capable of sustaining such illusions – maybe if reality really worked the way it appears (to us) to work, then we couldn't possibly exist. This is a second, and stronger, sense in which the anthropic principle doesn't tell us much.

The Reality Hypothesis

What really makes the anthropic principle interesting is when we take seriously the idea that the universe of our (possibly illusory) observations really does sustain sub-systems which really do function the way we appear to function.

Note that this doesn't require us to believe that the universe of our observations really is real. We might all be one (or should that be π ?) wacky polyp in some reality we can't comprehend, provided the virtual reality (or whatever) system feeding us our illusions manages to mess with that one (or however few) head(s) in clever enough ways that, in fact, we're never aware of it. In effect, that means each of us is a sub-process of one highly complex process, possibly actually operating in some reality beyond our ken. The reality property I'm suggesting is one I can best describe by analogy with a software environment: if all the primitives of the system manage to sustain, without ever giving any contrary clues, the illusion of being components built within the system, then the system is real in the sense I'm aiming at; but if a complex enough system built within that environment can determine that the primitives really aren't implemented within the system, then the system fails to sustain the illusion of being real in this sense.

We observe our universe and discover rules it follows; we can observe one another and address the question of whether a system can function as we do while appearing, within our observed universe, as we do; if we could show that the rules our universe follows preclude such behaviour from such a system, we'd have revealed a piece of unreality – an infelicity in the virtual reality system feeding delusions to our hypothetical polyp(s). If we can show how our observed universe can sustain a sub-system which functions as we do and would appear as we do, then we'll have demonstrated a piece of self-consistency to our universe. Taken in the context of such a reality property, the anthropic principle actually gets to say something – albeit something quite subtle and hard for us to examine with any care – regardless of whether we're really very different entities than we suppose ourselves to be, but experiencing such illusions as to cause that supposition.

What Does Our Reality Imply ?

So accept the anti-sophist basic assumption of all science – that what we consistently observe really is real – and consider what our existence tells us. For any universe to be real in the sense of having some denizens that so conceive it, it's got to be capable of sustaining complex sub-systems (those denizens) capable of various information processing and sensory activities sufficient to notice their universe. For the said denizens are to be us imposes even tighter constraints.

We've developed quite a sophisticated understanding of how our universe behaves; and it's clear that, if various constants that appear in our physical laws – notably the speed of light, the impedance of free space, the fermionic spin and the gravitational coupling constant (c, Z0, ℏ and G) – had even slightly different values, the differences in behaviour would be great enough to preclude existence of things like us in the resulting universe. Different values for such constants might cause the universe to be too tightly curled up on itself to support such large objects as stars; or might cause the early universe to fly apart so fast that matter never got to clump up closely enough to produce galaxies, let alone stars. We don't yet understand how the particles that make up matter come to have the particular masses and other properties they do have – it's reasonable to speculate that these properties depend on the physical constants, but we don't know the nature of that dependency – but changes to the physical constants might plausibly cause drastic changes to which nucleons and molecules are how stable.

Our observation that we exist, and are the way we are, thus limits what values the physical constants have; it is, however, somewhat philosophically tortuous to claim (as some do) that the reason the physical constants take the values they do is so as to allow us to exist. If I do an experiment to measure the speed of light, my observations limit what values the physical constant may have; but this doesn't mean the universe had to have the particular speed of light it does have, so as to allow my experiment to produce the result it did.

The requirement that the universe contain an observer of such complexity as to be able to describe the universe, or question its reality, is subtly different, and is the true heart of the anthropic principle. If the physical constants of the universe had values which precluded the existence of such an observer, any sense in which that universe exists is entirely moot, for want of any inhabitants to believe it exists. However, there are more things in heaven and earth than are dreamed of in Horatio's philosophy; and more ways for a universe to contain self-aware observers than I can hope to describe. So, while we may be able to narrow the range of values the physical constants can have (be excluding variants on the universe that are hopelessly short-lived, chaotic or dispersed, for example), I am skeptical of any expectation that we can infer their particular values from the existence of sentient observers.

Stronger Contrariness

As one way to put the anthropic principle and its consequences in perspective, suppose our culture had (and knew it had) a full account of the character of our universe. To be clear, we have a pretty full one, and a recently-discovered particle might yet be its missing Higgs boson, but it still doesn't say why we have the different fields we do have (and it needs the Higgs boson to say why the particles resulting from those fields have the masses they have); what I mean by a full account is one which even says why we have the rules we do.

Now, I conjecture that such a culture would hypothesize – as ours has hypothesized the anthropic thesis – that the universe necessarily contains, not just a sub-system capable of asking is this real, but a sub-system capable of discovering and describing a full account of its universe. It seems to me that such a culture would find this thesis natural, as modern theoretical physics finds the anthropic thesis natural.

So suppose I, in the aftermath of an evening's debate with some friends, while walking home, hit upon such a full account of the nature of our universe – but, along the way (doubtless distracted by my epiphany), I slip on the ice (I live in Norway) and break my head open upon some little fence-post around some wayside shrubbery; and either die or suffer such brain damage as to preclude my ever communicating my discovery to others. The super-anthropic principle I've just offered has been satisfied; but humanity at large never gets to know it happened.

In like manner, the anthropic thesis would, at least in principle, have been satisfied if one stray proto-lemur (in the era our culture associates with dinosaurs) had – possibly after recovering from the results of eating some partially rotten fruit – attained comprehension of the question of illusion and, hence, of what is really real. Our proto-lemur would need to have survived the ordeal long enough to think about all this a bit, but if the whole planet had been sterilized by a meteor impact shortly afterwards, still the anthropic thesis is satisfied.

Indeed, for all we know, what now forms our asteroid belt was once a planet; on which, mayhap, some precocious life-forms attained the required criterion before being blasted into oblivion by some errant celestial mechanics.

None of that actually dilutes the anthropic principle's conclusions, but I hope it will serve as a warning to those who would pull more out of it than just the minimal sober the universe necessarily contains an observer assertions that it can honestly justify.

Valid CSSValid HTML 4.01 Written by Eddy.