Sabine Hossenfelder on reductionism [computers (& implementation)]

Robert Lawrence Kuhn is hosting Sabine Hossenfelder is a discussion of her new book, Existential Physics: A Scientist’s Guide to Life’s Biggest Questions. Some 20 minutes in or so Kuhn and Hossenfelder are talking about reductionism, the idea that we seem to be able to explain the properties and actions of things at some scale by invoking objects and actions at a smaller scale. Hossenfelder has expressed the idea that perhaps there are limits to this process and that, at this point, our problems aren't at the smallest scale, but rather (c. 22:44)

maybe what our problems are trying to tell us is that this ontological reductionism has reached its limits, so maybe we should not try to figure out if elementary particles are made of strings. Maybe our problem is describing how big objects come about.

Kuhn brings up the question of whether or not (c. 23:21) "there are ontologically fundamental laws that exist in large constituents that are not reduced ... to the actions of small particles." This leads him to a distinction between strong emergence (macro can't be explained in terms of the micro) and weak emergence. She rejects strong emergence (we don't have any cases) but doesn't think we need strong emergence to "have fundamental laws at large distances." She then refers to her interview with David Deutsch which eventually leads to (c. 27:23):

So, you have your microscopic description and now imagine you're able to do your derivation of the macro level, so this is weak emergence. But now you have a theory which is completely useless; it's just too difficult. And so you reformulate it in other assumptions, and those assumptions use ingredients from the macroscopic level and the no longer rely on the underlying microscopic structures. So there's no disagreement.

I think the other example that David Deutsch used is the existence of a universal computer. And that's not something which you can express in laws of the microscopic constituents, elementary particles or something. It's a property that comes out at the macro level. And it doesn't disagree with the existence of laws on the microscopic scale. And if you were really really good maybe you could start with something like string theory and then prove it would give rise to a universal computer, but it doesn't really make a lot of practical sense to try to use this kind of calculation. So you would use a theory on the macroscopic level that postulates well, universal computers exist and that's something that you can work with.

That's something I've discussed in terms of implementation, e.g. in Fecundity and Implementation in a Complex Universe, Is software a kind of mathematics?, and Physical constraints on computing, process and memory, Part 1 [LeCun].