What Parts Of Reality Does Science Describe, And Which Parts Are Merely Constructs?

A rose would not smell as sweet if it was called, as it likely is called in German, a sourbloodpetal. Or whatever. Naming things, getting the good names for things, brings with it many victories.

In the philosophy of science a realist—which is a name most excellent—is one who believes the objects named in physical theories are real. The objects exist, as the theories say they do, in reality. Strings, for instance, are there because theory demands it. But in reality, I say, sometimes these objects are there and sometimes they aren’t. So what do we call the position that says sometimes theories are blowing smoke, and sometimes they name real things? Can’t use realist. That’s taken.

Well, there is a name for such a view, which I’ll reveal in a moment. But I’m giving nothing away by telling you the name stinks. Smacks of academese. And that’s because the realists got there first with their name. (Incidentally, nobody beats computer scientists in naming: artificial intelligence forsooth!.)

A position that’s sort of the polar opposite of realist is empiricist. This is the view that the things named by theories don’t really matter, probably aren’t real, but that the theory is still empirically useful because it makes good predictions. The vast majority of people using machines, like cellphones, are empiricists in a way. Strings don’t have to be real for the math to be pleasant.

An intermediate position is epistemic structural realism. As in, “Hi, I’m an epistemic structural realist.” Ugh. As one source puts it:

So one way of thinking about structural realism is as an epistemological modification of scientific realism to the effect that we only believe what scientific theories tell us about the relations entered into by unobservable objects, and suspend judgement as to the nature of the latter.

Well that’s just the way you’re going to have to talk because the realists took such a commanding lead in the Name Game.

I was reminded of all this by the Bas van Fraassen article (which I learned from an Ed Feser tweet), “Science does not describe reality“, a position with which, at the current time of writing, I have much sympathy (blog, Substack). I put the condition in because of the hope that someday science will describe reality. Right now I think it does so only sometimes.

Van Fraassen (why two ‘a’s and two ‘s’s but not two ‘n’s?) reminds us that realists say that because theories make such good predictions (or when they do), it’s reason to believe the objects in the theory are real. That is a good argument, but far from convincing, as we see below. This is contrasted with the empiricist view he holds.

On an empiricist view, the aim of science is to give us empirically adequate theories. There is a distinction between being true in all respects and being true about what is observable. So to accept a theory, when there is good reason in the empirical support, need involve as belief only that the theory is empirically adequate. Acceptance need not involve a belief that the unobservable parts of the scientifically represented world are real. You may certainly add more beliefs to your acceptance, beliefs about truth and reality, if you wish! But as far as science is concerned that is just supererogatory.

A group of philosophers made comments on van Fraassenn’s article. This is my favorite of them:

John Dupre said…

As a realist and an empiricist, all I can say is that I think too much attention to physics can give philosophers a very strange take on science. I probably wouldn’t believe a lot of physics was strictly/approximately true even if persuaded that it was useful for making predictions. So much the worse of physics. But such an attitude really doesn’t make much sense for physiology, molecular biology or, for that matter, evolution.

To which I say Amen. Looking only at physics is doing philosophy of science the hard way. If you only focus on the quantum bestiary you will get eaten by the difficulties. Why not start easier? Recall we’re discussing the philosophy of science and not science itself.

Here’s something easy. In probability we have models galore that are correlational, which make excellent predictions and which we know the objects behind them are not all real, though some might be. There is no dispute about this, either.

Take the humble die throw (a nauseatingly cliche example, yes). The model, which is to say the theory, is easy: “A six-sided object with unique sides is tossed and must come up with one side uppermost; the probability of any is 1/6.” Well, if you apply this model to a real die, then you are naming actual objects in reality.

But we aren’t naming everything that exists about that object, just one or two aspects. This theory is also not a complete explanation, which we naturally hanker for. It is a partial explanation, though, because it names one of the causes: the formal cause; those six sides. It also has some information about the efficient cause: the tossing. And the material cause: the actual die you apply it to.

To get a complete explanation it to specify the substance in its entirety, and the powers that it exhibits, which is to say, we write down all the causes. And if we do, or can, then when have a theory that describes real objects in theory.

We also aren’t saying in this theory that chance is real. “Chance” isn’t a part of Reality that is causing the die to come up this rather than that. This we showed in the coin flip video (blog, Substack). Chance is good for an empirical theory, but it’s wrong at describing Reality itself: it’s not there.

Contrast dice to strings (of the subatomic kind). Are they real? How can you prove it? Can you construct a necessary and sufficient proof?

Subscribe or donate to support this site and its wholly independent host using credit card click here. Or use the paid subscription at Substack. Cash App: $WilliamMBriggs. For Zelle, use my email: matt@wmbriggs.com, and please include yours so I know who to thank.