Why I'm not a scientific realist

Scientific realism is the view that our best scientific theories are approximately true and the entities postulated by those theories really exist. In this video I explain why I am no longer a scientific realist.

0:00 - Introduction
3:42 - Success and truth
12:50 - The selection argument
32:26 - The vulnerability criterion

[KaneB]

Those of you who have seen my series on scientific realism will probably be familiar with the points I discuss in section 1: theories can be false but successful, and approximately true but unsuccessful. For more original arguments that are not discussed in any detail in that series, go to 12:50.

[kleezer]

This is literally the greatest channel on youtube

[KnightofEkron]

I used to be a rabid Naive Realist but after reading Van Fraassen and taking a few philosophy classes as well as philosophy of science class I turned into a Constructive Empiricist like yourself. Also Kane, around the 10 minute mark when you are speaking of the Strong Nuclear Force, I would speak of instead, the assumption of the constancy of light's speed or even the speed of light as an example. Because if the speed of light was differential based on time then the implications of that would be almost universal across almost all fields.

[Trynottoblink]

This channel rocks. The only one on YouTube that’s giving us a graduate-level education in philosophy for free.

[henrikmunch]

Around 9:45, an argument is being made regarding the approximate truth of scientific theories. I think the example with the nuclear force being 2% stronger is rather misleading though.
By approximate truth of theories in physics, I'd argue that what should be meant is that the theory makes successful predictions *at some specific length/energy scale*. So Newton's equations are approximately true in this sense, even though one must use special relativity for velocities close to the speed of light.
Whatever will replace, say, quantum mechanics, still has to agree with the myriad of experiments confirming quantum mechanics predictions at the length scales we are currently testing.
The example with a world having a 2% stronger nuclear force is not approximately true in this sense, since it does not agree with any experiment at any length/energy scale we know of.

PS: fantastic video. I myself struggle with scientific realism.

[GottfriedLeibnizYT]

How do you make the observable/unobservable distinction?

[efctony]

Is the Raven wing argument written anywhere? I don't know how I would cite a YouTube video!

[waliul]

Man your channel is unjusty underrated.

[GottfriedLeibnizYT]

36:20
Carneades . org: * *HOLD MY BEER* *

[haph]

I think an unsuccessful but true model seems likely. Even if you have a "true" model, that doesn't mean you are actually capable of using it.

A completely "true" model seems to me like it would probably be so complex as to be nearly if not completely unusable. Success requires that I can actually use the model to make a prediction, truth does not.

[dionysianapollomarx]

I've been reading stuff about CS Peirce on online available resources. I think stuff from Peirce parallels most of the reasoning here. In essence, his pragmatism seems like it is both Kantian and constructive empiricist, and hence also representative realist. Also, in it, true beliefs are not themselves virtuous, since there are only representations of truth which we can "confirm, " but the method of their justification and construction should be, and he prescribes a general category of attitudes towards finding a proper method, as well as first principles thinking.

[Ansatz]

36:20 "Even if we could be radical skeptics, it would be an absurd position to take. It would render us unable to act."

People don't need beliefs or knowledge in order to act. Consider almost any gambling situation. When a person pulls the handle of a slot machine, that person does not believe that she's going to win. She'd have to be mad to have that belief given all the evidence to the contrary, yet she is able to pull the handle and even put money at stake. This lack of belief has in no way paralyzed her ability to act.

36:40 "Type I errors involve believing what is false. Type II errors involve failing to believe what is true."

It's not really an error if it is done deliberately. It's probably logically impossible to deliberately commit a type I error, since committing the error would make it impossible for the person to be aware that the belief is false. On the other hand, it's easy to deliberately commit type II errors.

For any proposition P, we know without a doubt that one of either P or not-P must be true. Therefore anyone who believes neither one must be committing a type II error, but it's not really an error if it is done by careful consideration. Perhaps neither one has a compelling case and there is clearly no way to establish which of the two is true. In such a case, no mistake is being made when a person suspends belief.

36:48 "Failing to believe what is true is bad."

Bad by what measure? Badness is an issue of ethics, and ethics is usually more concerned with actions rather than mental states. For example, consequentialism doesn't care about _why_ we do things, only the results of what we do. Deontology also couldn't care less about what beliefs motivate our actions. Perhaps what we're talking about is a kind of virtue ethics, and we're counting true beliefs as a virtue, but then what makes virtue ethics the correct meta-ethical theory, and why are true beliefs virtuous?

[InventiveHarvest]

I sometimes think about Kuhn and wonder what his motivations were. Was he seeking to refute Popper? Was he seeking to confirm Popper and was surprised by what he found? Or perhaps he was exploring his own ideas. Maybe he felt that science was more similar to a social revolution than a scientific method and wanted to confirm his ideas. Maybe he felt that science was more similar to a social revolution and sought to refute his theory and failed against the historical evidence. And, if it was the latter, would it have been then science?

[will]

My biggest problem with this is this is that I can't tell what you mean by "true". What does it mean for a theory to be 'true' or 'untrue' when they are all constructed out of words with semi-ambiguous meaning (except for any mathematical formalism built in)? The only way I can think of to determine the 'truth' of a theory is to evaluate its information content about the world and whether its predictions are correct (and this doesn't really get at any kind of Boolean 'truth'. All you can say is whether and how much some given theory encodes accurate information).
In this sense, newtonian gravity isn't 'false'. It gives accurate predictions of all sorts of phenomena in a certain domain of applicability. It encodes an enormous amount of information about the world in a verifiably correct way--but obviously it doesn't encode everything.
Outside of an information-theoretic context, I don't know what it means for a theory to be 'true'--because all language (including strings of words which constitute some 'theory') is ultimately a mechanism for transmitting information.

[conorb]

One thing that this whole debate seems to ignore is, what is truth?

[dimitrispapadimitriou]

What are "we" who are empiricists and make observations etc. We aren't outside the physical world, we are part of it

[stevenmason]

Hello, I'm wondering if you still endorse your Selection Argument or have you revised it in anyway over the years. The account of 'truth' you imply in the argument, I guess, needs to be argued for elsewhere. As you know, we can give different accounts of truth. (The account of truth you put to use in the argument is implicit in the crow example.) The argument is succesful against stronger variants of realism. (There is also the question of using Bayesian techniques for theory formation instead of Inference to the Best Explanation). Cheers

[johnnyloco]

Thanks for the video. Personally, I remain a weak scientific realist in that it just seems to be the best explanation for what we observe.

[MisterWebb]

2:20 Realism is the "default position", but it is not the "common sense view". Common sense tells us to be skeptical of authority, and our senses show us that the Earth is flat and motionless and that the Sun, stars and Moon rotate overhead. It is only through visual representations of the concept of a spherical Earth (a spinning globe in every classroom, spinning globe graphic before the TV news, etc.) that people can be persuaded to think of the world they inhabit as a ball hurtling through the vacuum of space (c.f. Stuart Brand).

[devos]

Have you read Scientific Perspectivism by Ronald N. Giere?