Fabio Gironi recently interviewed me at length for an issue of Speculations, a “Journal of Speculative Realism.” The subject was science and philosophy, which I’ve been known to opine about at some length. But here we’re talking great length indeed. The interview isn’t available separately, but you can download the pdf of the whole issue here (or buy it as a bound copy). My bit starts on page 313. (The rest of the issue is also worth checking out.)
I’m a big believer that academic disciplines should engage in messy interactions, not keep demurely separate from each other. But it’s a tricky business. Just because I’m (purportedly) an expert in one thing doesn’t make me an expert in everything else; on the other hand, it is possible that one area has something to offer another one. So I am in favor of dabbling, but with humility. It’s good for people to have thoughts and opinions about issues outside their immediate expertise, and to offer them in good faith, but it’s bad if they become convinced that experts in other areas are all idiots. So when you find yourself disagreeing with the consensus of expertise in some well-established field, it might very well be because of your superior insight and training, or maybe you’re just missing something. Hopefully in an exchange like this I have something to offer without making too many blunders that would make real experts cringe.
Here’s a sample of the interview.
SC: I would be extremely suspicious of any attempts to judge that the world must ‘necessarily’ be some way rather than any other. I can imagine different worlds—or at least I think I can—so I don’t believe that this is the only possible world. That would also go for any particular feature of the laws this world follows, including their stability. Maybe the laws are constant through time, maybe they are not. (Maybe time is a fundamental concept, maybe it isn’t). We don’t yet know, but it seems clear to me that these are empirical questions, not a priori ones. Because we want to understand the world in terms that are as simple as possible, the idea that the underlying laws are stable is an obvious first guess, but one that must then be tested against the data. Said in a slightly different language: any metaphysical considerations concerning what qualities the world should properly have can be taken seriously and incorporated into Bayesian priors for evaluating theories, but ultimately those theories are judged against experiment. We should listen to the world, not decide ahead of time what it must be.
Here’s an example of an a priori, necessarily true, judgment about the world: in order for us to know anything about the world, it must present itself to us in ways that we can apprehend it.
Sean, I completely agree with your main point (as usual!). That said, now for some nitpicking. I could imagine _defining_ “The Laws of Nature” as a set of rules that describes the world and is valid everywhere and at all times.
Differently put, if you found a law that seemed varied across space and time (e.g. the metric tensor in GR), wouldn’t you go looking for some deeper law that was able to describe the variation (in the example, the Einstein field equations)? Maybe the metric tensor could have been just be random with no underlying law to be found, but that would be hard to imagine (which brings us back to your point that we cannot make a priori statements about the properties of the world, I suppose).
Or maybe I am just lacking the imagination to think of the kind of inhomogeneity you had in mind.
I don’t know. I strongly suspect that if we are talking about truly fundamental laws, we can arbitrarily define them as invariant. That is to say, if you give me some set of laws which varies in time or space, I can write an equivalent set of laws that describes those variations in time or space while my laws remain invariant.
Maybe there is the possibility of getting around this by proposing some set of laws whose variations cannot, even in principle, be described by an invariant theory. But I am strongly skeptical that that is possible.
Now, if we were talking about the effective laws of physics as opposed to fundamental ones, then that is a more interesting question. But I think that at present, it is more or less solved by the existence of spontaneous symmetry breaking in the standard model of particle physics. With the existence of spontaneous symmetry breaking, it is no longer a question of whether the effective laws of physics vary, but instead a question of how much.
Jason (or anyone else),
Okay – if the laws of our universe vary in time or space, you can develop an equivalent set of invariant laws that describe those variations.
SO – does there exist *any* set of observations which can’t be described mathematically?
If not, then what is really so unreasonable about the effectiveness of mathematics in the natural sciences (in reference to Eugene Wigner’s article with that title)?
Mathematics describes the universe that we observe…but it also describes an infinite number of universes that we don’t observe. Mathematics is just a descriptive language par excellence…yes?
Maybe?
Thoughts?
Can you imagine a world where nothing it in can be numbered or counted? Can you imagine a world where events take place but they have no causes? Can you imagine a world outside of time?
Any world we imagine must meet certain conditions for us to know it, and by “must meet” I mean they are a priori and necessary. To follow your analogy, we cannot listen to the world unless it makes sounds.
Very nice interview, Sean. It’s refreshing to see a physicist who is sufficiently knowledgeable about philosophy to give sensible answers to a philosopher’s questions.
There was one question where you kind of punted, and I wished you had had more to say. Gironi asked you (p. 320) to describe what it was that non-mathematically trained people were missing when they only read the popularized versions of scientific theories/discoveries. As a fellow popularizer of physics, I think this is a very interesting question, and one that is hard to answer. (One is tempted to respond, “Well, just learn the math and then you’ll know what you’re missing.”) It’s particularly important for philosophers, if they are going to try to take ideas from modern physics and do philosophy with them.
So here’s a nudge to get you to address this issue more fully in some future blog post.
Really good advice, and like so much really good advice, rarely followed by those who give it.
Max Thomas @ 5:
Kind of, yes, kind of. This argument has a few major flaws: 1. the problem of other minds suggests that even if you (or I) can’t imagine something that it might still be imaginable to someone (or some thing). 2. What we can imagine has little bearing on what is actually true (I almost said “no bearing,” but the fact that we can imagine in the first place probably entails a few truths). 3. It’s an argument from “lack of imagination” which creationists use all the time but which aren’t at all normative (largely due to (2)).
At any rate, not being able to imagine something doesn’t provide grounds for ruling out the possibility. I have trouble imagining a universe where things cannot be counted (a difficulty that hunter gatherer tribes probably don’t share, incidentally) but that doesn’t mean it’s impossible.
Again, “imagine” makes the argument problematic — the universe is not bounded by our imaginations. But there are other problems here. You argue that because the universe is knowable there must be a priori and necessary truths. But how do you know the universe is knowable? Again, following the same analogy, we can only know about those parts of the universe that make sounds (and it’s not necessary that every part of the universe should make sounds) and we have to trust that those sounds aren’t lies.
And I have yet to hear an example of an “a priori truth” that wasn’t either a mathematical statement (the truth of which depends on the axioms of some motivating mathematical theory) or “bachelors are unmarried” (which is true by virtue of a contingent definition rather than the structure of the universe itself). While I wouldn’t rule out a priori and necessary truths any more than I would rule out an uncountable universe, I’ve certainly never heard any convincing examples of either. Compared to the successes of empiricism and scientific reasoning, “a priori” and “necessary” are just about useless concepts for determining anything about the universe, at least as far as I’ve seen.