Can You Observe a Typical Universe? (script) (Patreon)

The moment you started observing reality, you hopelessly polluted any conclusions you might make about it. The anthropic principle guarantees that you are NOT seeing the universe in most typical state. But used correctly, this highly controversial idea can be extremely powerful. So, how do you correctly use the anthropic principle?

In 1543 our perspective of the cosmos was radically disrupted. Nicholaus Copernicus presented his model of the solar system which demoted the Earth from its position as the center of the universe into just one of several planets orbiting the Sun. As our astronomy improved, we realized that our sun is a typical example out of 100s of billions of stars in the Milky Way, and that the milky way is ordinary among the 100s of billions of galaxies in the observable universe. The notion that we occupy neither a central nor privileged position in the universe is called the Copernican principle after the guy who started it all. It’s extremely important - it allows us to study the distant universe confident that its laws of physics are the same as we experience on Earth. It allows us to understand the origin of the Earth and the Milky Way by studying the ancient light of distant galaxies.

The Copernican principle became such a powerful tool that it took centuries for us to realize that it’s wrong - or at least flawed. It states that we don’t occupy a central, privileged position - that’s only half true. Earth certainly isn’t central, but it IS privileged, and not at all a typical environment. Nor, perhaps, is our entire universe. Our planet and our universe must have at least one non-typical quality - they must have been able to produce us - to give rise to living creatures that can observe it. This is the anthropic principle, and it seems to contradict the  sacred Copernican principle.

In the last two episodes we talked about certain observations of the uniqueness of both our planet and our universe, and how these feed into two versions of the anthropic principle. Today we’re going to bring all these ideas together with the Copernican principle and see just how powerful - and how misleading - the anthropic principle can be.

By itself, this statement is uncontroversial - some would even say tautological. But things get interesting when you add the fact that our universe seems to have fundamental constants and initial conditions that seem extremely fine-tuned for the eventual formation of life. The anthropic principle permits a new explanation for this fine-tuning besides blind luck or design: if there are enough universes or enough regions within this universe, and their properties can vary, then even if the vast majority are lifeless, life-friendly universes can still show up. And we shouldn’t be surprised that we find ourselves in one, even though Earth and a life supporting universe might be quite atypical.

This use of the anthropic principle is highly contentious. Many scientists find it extremely unsatisfying, and lazy, and unscientific. After all, it doesn’t tell us WHY the fundamental constants take on the values that they do - or even why they may vary between universes. 

Some of the bad rap of the anthropic principle comes from a distortion of this idea. In the 1986 book the Anthropic Cosmological Principle, John Barrow and Frank Tipler misinterpret the strong anthropic principle to mean that the later evolution of observers had some causal influence on the initial formation of the universe. This commits the same flaw in reasoning that the anthropic principle may solve. The principle is NOT causal - it just tells us to account for an observer selection bias when interpreting the nature of our environment. So our finely-tuned part of the greater cosmos doesn’t have to have been custom built for us. 

The other common misuse is to assume the anthropic principle allows for any degree of extreme fine tuning of the properties of our own environment, regardless of the global distribution of properties. Here’s an example, originally laid out by Roger Penrose. The universe’s most finely-tuned parameter is its unthinkably low initial entropy. The incredible density at the Big Bang was a highly ordered state. All particles in the observable universe were packed together in a subatomic-sized dot. Everything interesting that has happened since - from the formation of stars and galaxies to the evolution of life - has been powered by the slow increase in entropy from that initial state. The universe will spend the vast majority of its perhaps-infinite life in a state of extreme disorder and high entropy - iron stars, black holes, and a mist of cold elementary particles. Not very hospitable to life.

The second law of thermodynamics tells us that entropy can only increase, which means extreme high entropy states must be the norm - in the full timeline of our universe, but probably across the multiverse, if it exists. We certainly don’t observe the universe in a typical, observer-hostile state, and so it’s tempting to use the anthropic principle here. Low-entropy regions should be vastly less common than high-entropy regions, but if they exist we’re gonna find ourselves in one. 

But guess what? That statement is wrong. The anthropic principle doesn’t work here - but to understand why we need to go all the way back to the Copernican principle. In fact we need to bring these two seemingly conflicting principles together. And in doing ƒso we’ll end up with a much more powerful version of the anthropic principle - one that will even make testable predictions.

By the Copernican principle, we are most likely to observe a very typical environment - this is just a statement of probability. There are more typical environments, so throw us in a random environment and chances are we’re in one of the more typical environments. But anthropic principle tells us we must account for our status as observers when we interpret our environment - including the probability of being in it. So let’s formulate a refined anthropic principle: we should find ourselves in a typical region of the cosmos that is consistent with us being observers. Any by cosmos I mean the sum total of reality, be it universe or multiverse. That allows us to be in a rare, observer-friendly environment, but tells us that we should be in the most typical such environments. Our perspective on the cosmos might be a rare one, but it should be no more rare than is necessary to explain our existence.

This refined anthropic principle totally fails to account for our extreme low entropy universe. If low-entropy regions happened just by chance, then the lower the entropy the less probability of that region forming. For example a fluctuation the size of a galaxy is insanely more likely than one the size of our observable universe. It’s much easier to produce 100s of billions of galaxy-sized fluctuations than a single fluctuation of 100s of billions of galaxies. You surely don’t need more than one galaxy to spawn a life-bearing planet - so there should be many more observers in small entropy fluctuations than in large ones.

The refined anthropic principle appears to fail here - but actually it doesn’t. We’ve just  witnessed the potential power of the principle. Under the assumption that our universe resulted from a simple random fluctuation in an otherwise high-entropy state, the anthropic principle predicts we should be in the smallest such fluctuation that could produce us. Apparently we aren’t, and so we can probably rule out a simple random entropy fluctuation as a sufficient explanation for our big bang. Or at the very least we need extra physics: for example, the right initial density fluctuations may need to evolve and expand for the right amount of time, so life is always in big universes. If we then assume that the starting conditions for our universe were typical, that can tell us something the physics of how universes are born.

For the anthropic principle to be useful and not misleading, we need to be careful. Philosopher Nick Bostrom has made valiant attempts to clean up what he calls “anthropic reasoning” in his book Anthropic Bias. In it he defines, although doesn’t invent, the self-sampling assumption, which states that “All other things equal, an observer should reason as if they are randomly selected from the set of all actually existent observers (past, present and future) in their reference class.” If there are a bajillion observers in the entire cosmos, you should consider yourself randomly selected from them. That means you’re most likely a common type of observer, and in a common environment in which observers can exist. 

The anthropic principle and the self-sampling assumption encourage Bayesian thinking - we should take into account what we already know - our “priors” - before assessing any probabilities. The prior is that we are an observer. But proper Bayesian thinking requires careful definition of priors - for example, Bostrom talks about “observers in their reference class”. How do we know what our reference class is? Carbon-based sentient life? All conscious entities? Anything capable of thinking about the anthropic principle? Even with very careful definitions, anthropic reasoning is difficult to use well. And that’s evident in the various bizarre predictions it can make - from Boltzmann brains, which we covered, to the doomsday argument - the idea that the self-sampling assumption predicts that the end of the world is nigh - which we’ll cover soon - hopefully soon enough.

The anthropic principle in its proper form is without question an important thing to take into account whenever we observe the universe. It’s one possible explanation for why our planet and our universe appear to fit us so well, even if they weren’t intentionally made for us. Douglass Adams put it best: “If you imagine a puddle waking up one morning and thinking, 'This is an interesting world I find myself in — an interesting hole I find myself in — fits me rather neatly, doesn't it? In fact it fits me staggeringly well, must have been made to have me in it!" This is such a powerful idea that as the sun rises in the sky and the air heats up and as, gradually, the puddle gets smaller and smaller, it's still frantically hanging on to the notion that everything's going to be alright, because this world was meant to have him in it, was built to have him in it; so the moment he disappears catches him rather by surprise. I think this may be something we need to be on the watch out for.”

Used well the anthropic principle gives us a deep perspective on our place in the cosmos, and can also be a powerful, albeit slippery scientific tool. But misused it can lead to unscientific conclusions - as can any tool in science. We’ll come back to some of the wilder predictions of both good and bad anthropic reasoning real soon, and talk about what you can know about your universe, given your privileged status as a typical conscious observer of space time.