I keep coming across statements about the anthropic principle leaving
its mark on physics that strike me as ill-informed, most recently in a
book I am presently reading “The Edge of Physics” by Anil Ananthaswamy:
“The anthropic principle – the idea that our universe has the properties
it does because we are here to say so and that if it were any
different, we wouldn’t be around commenting on it – infuriates many
physicists, including [Marc Davis from UC Berkeley]. It smacks of
defeatism, as if we were acknowledging that we could not explain the
universe from first principles. It also appears unscientific. For how do
you verify the multiverse? Moreover, the anthropic principle is a
tautology. “I think this explanation is ridiculous. Anthropic principle…
bah,” said Davis. “I’m hoping they are wrong [about the multiverse] and that there is a better explanation.””
The anthropic principle has been employed in physics as a proposed
explanation for the values of parameters in our theories. I’m no fan of
the anthropic principle because I don’t think it will lead to big
insights. But it’s neither useless nor a tautology nor does it
acknowledge that the universe can’t be explained from first principles.
- The anthropic principle doesn’t necessarily have something to do with the multiverse.
The anthropic principle is true regardless of whether there is a
multiverse or not and regardless of what fundamentally is the correct
explanation for the values of parameters in our theories. The reason it
is often mentioned in combination with the multiverse is that proponents
of the multiverse argue it is the only explanation, and no further explanation is needed or necessary to look for.
- The anthropic principle most likely cannot explain the values of all parameters in our theories.
There are a lot of arguments floating around that go like this: If the value of parameter x
was just a little larger or smaller we’d be fucked. The problem with
these arguments is that small variations around one out of two dozen
parameters leave out most possible combinations of parameters. You’d
really have to consider modifications of all parameters together to be
able to conclude there is only one supportive of life, which is however
not a presently feasible calculation. And though this calculation is not
feasible, the claim that there is really only one combination of
parameters that will create a universe hospitable to life is on shaky
ground already because this paper put forward a universe that seems capable of creating life and yet is entirely different from our own. And Don Page had something to say about this too.
The anthropic principle might however still work for some parameters if
their effect is almost independent on what the other parameters do.
- The anthropic principle is trivial, but that doesn’t mean it’s useless.
Mathematical theorems, lemmas and corollaries are results of derivations
following from assumptions and definitions. They essentially are
the assumptions, just expressed differently, always true and sometimes
trivial. But often, they are surprising and far from obvious, though
that is inevitably a subjective statement. Complaining that something is
trivial is like saying “It’s just sound waves” and referring to
everything from engine noise to Mozart.
And so, while the anthropic principle might strike you as somewhat silly
and trivially true, it can be useful for example to rule out values of
certain parameters of our theories can have. The most prominent example
is probably the cosmological constant which, if it was too large,
wouldn’t allow the formation of structures large enough to support life.
This is not an empty conclusion. It’s akin to me seeing you drive to
work by car every morning and concluding you must be old enough to have a
driver’s license. (You might just be stubbornly disobeying laws, but
the universe can’t do that.) Though, this probably doesn’t work for all
parameters, see 2.
- The anthropic principle does not imply a causal relation.
Though “because” suggests so there’s no causation in the anthropic
principle. An everyday example for “because” not implying an actual
cause: I know you’re sick because you’ve got a cough and a runny nose.
This doesn’t mean the runny nose caused you to be sick. Instead, it was
probably some virus. Alas, you can carry a virus without showing
symptoms so it’s not like the virus is the actual “cause” of my knowing.
Likewise, that there is somebody here to observe the universe did not
cause a life-friendly universe into existence. (And the return, that a
life-friendly universe caused our existence isn’t the case because
life-friendly doesn’t mean interested in science, see 3. Besides this,
it’s not like the life-friendly universe sat somewhere out there and
then decided to come into existence to produce some humans.)
- The applications of the anthropic principle in physics have actually nothing to do with life.
As Lee Smolin likes to point out, the mentioning of “life” in the
anthropic principle is entirely superfluous verbal baggage (my words,
not his). Physicists don’t usually have a lot of business with the
science of self-aware conscious beings. They talk about formation of
large scale structures or atoms. Don’t even expect large molecules.
However, talking about “life” is arguably catchier.
- The anthropic principle is not a tautology in the rhetorical sense.
It does not use different words to say the same thing: A universe might
be hospitable to life and yet life might not feel like coming to the
party, or none of that life might ever ask a why-question. In other
words, getting the parameters right is a necessary but not a sufficient
condition for the evolution of intelligent life. The rhetorically
tautological version would be “Since you are here asking why the
universe is hospitable to life, life must have evolved in that universe
that now asks why the universe is hospitable to life.” Which you can
easily identify as rhetorical tautology because now it sounds entirely
stupid.
- It’s not a new or unique application.
Anthropic-type arguments, based on the observation that there exists
somebody in this universe capable of making an observation, are not only
used to explain free parameters in our theories. They sometimes appear
as “physical” requirements. For example: we assume there are no negative
energies because otherwise the vacuum would be unstable and we wouldn’t
be here to worry about it. And requirements like locality, separation
of scales, and well-defined initial value problems are essentially based
on the observation that otherwise we wouldn’t be able to do any
science, if there was anybody to do anything at all.
