Could someone who knows the topic tell us (and by “us” I mean layman like me) if there is something to keep from this article? It brings so much context that after reading it I still have no idea what this “rewrite of fundamental law” is about.
There has always been something unsatisfactory about the second law of thermodynamics. Unlike many other laws in physics, which precisely state both what kinds of structures exist in reality and how they behave, the second law has a built-in lack of precision. Specifically, there are multiple statements of what the second law says, but they all suffer from referencing some concept which is inherently fuzzy at some scale. For example, they may refer to "heat" as part of the law, but the physical definition of "heat" is fuzzy at the microscale -- it only has a clear meaning for "large systems".
However, using a new physio-mathematical formalism called "constructor theory" the researchers have now found a way to state the second law of thermodynamics which is precise at all scales.
"A scientific explanation is a statement of what is there in reality, and how it behaves and how that accounts for the explicanda." (emphasis added) [1]
Deutsch argues strongly against the view of physics which says that physical explanations are just about predictions of behaviors and not about what is really out there. See also his 1997 book The Fabric of Reality.
[1] https://www.sciencedirect.com/science/article/pii/S135521981...
Sure, you can say stuff about "what is there in reality", but I don't see any way one can verify such statements.
Everything humans have access to is mediated through their perceptions. Everything that we sense "outside" of us is to us a perception.
This includes the results of all the experiments ever done, all our interaction with tools and scientific instruments, and all our observations. They're all perceptions.
Our thoughts and feelings are also perceptions. The results of our deductions and inductions are perceptions.
Humans seem to be hermetically sealed from the "outside world" (if there is one) by our perceptions, and I don't see any way out of that... not through science, not through philosophy, not through religion, not through anything.
"Mach, Einstein, and the Search for Reality" https://dash.harvard.edu/bitstream/handle/1/37902464/2002383...
In short, late in the 19th century and early 20th century, the physics philosophy of Ernst Mach held great sway. Mach's viewpoint was the very viewpoint that you describe: the only things we know about are our own perceptions and proper formulations of physical theories consists at the fundamental level of descriptions of these things.
Mach had a significant impact on many physicists' understanding of what they were doing and also had an impact on the development of logical positivism in the field of philosophy.
Einstein accepted Mach's philosophy initially. But gradually he came to see it as untenable given his own work in physics. The article linked to above describes how and why his viewpoint changed over time.
However, if I gave you the impression that I was a Machian, this is a mistake.
According to this paper, Mach equated reality with sensory experience (ie. the experiences afforded us through the five senses: seeing, hearing, touch, smell, and taste.. and perhaps some others like proprioception), and that speculation about a world "outside" and independent of such sensations was unwarranted.
This is not my position, as I include in "perception", the perception of thoughts, theories, conclusions, intuitions, feeling and everything else considered to be "internal" -- which includes everything Einstein or any other physicist could ever base their own theories on.
Also, I don't object to speculation as to an "external", "objective" world, as Mach does. Science can come up with all sorts of interesting and possibly useful things through such speculation. I'm happy to read of such conjectures, and if some people want to engage in it more power to them.
However, do I not see any way we could ever verify whether such speculation right or wrong (or, to put it another way, whether such speculation corresponded to the way the world really was), because neither mathematical simplicity or the elegant fitting in to other, larger theories (as Einstein seems to have favored) means that the speculation has anything to do with the "external" world as it is (or even that there is an external world).
Since we seem to be limited to our own experience, the belief that an "external" world exists and that we have any (direct or indirect) way to access it seems to be nothing but faith.
The scientific method is about predictions and testability of such predictions, therefore it implies behavior. Nothing more is needed for a scientific theory. To test for existence within the scientific method would require to test for non existence too, a clear paradox, which is the reason why ontological statements ("X exists") are meta-physical in nature.
How far do you take that stance?
If a biologist says a particular bacterium exists, are they doing philosophy, not science?
Edited to add another interesting case: creation "science" organizations like Answers in Genesis say that the theory of evolution is a philosophy not science since it describes non-observables in the past (events in the past are inherently not observable now). When a paleontologist says a particular animal once existed in the past, would you say they are they doing philosophy not science?
When a biologist says that a bacterium exists, they are making a descriptive statement, not an ontological one. To say that a bacterium exists is just a convenient proxy to refer to a specific aggregate as a whole. A bacterium is always reducible to its constituents, therefore has no ontological status nor existence in and of itself: in other words, if the molecules which the bacterium is made of would relinquish their structural integrity, the bacterium would cease to "exist" too.
Therefore I would agree with you that it's totally legitimate for a scientist to claim that a bacterium exists, in so far as we understand this claim as a descriptive proxy. A shorthand, essentially, to cluster together specific configurations of the underlying, supposedly more real, "stuff". And this is needed, if we want to make any progress. We have to communicate, effectively and succinctly. Science is precisely about _descriptions_ and predictions of behavior. Same goes with the paleontologist example.
It's a completely different matter instead to claim ontological status, that is, to say that something exists in and of itself, irreducible to something else, as a building block of the universe. This is a philosophical stance per-force, since the scientific method cannot follow us there. It's meta-physics by definition, and there are different rules at play there. Not Popper's falsifiability, but empirical adequacy, internal consistency, explanatory power, etc.
So when scientists claim that the ontological primitive is, e.g., the quantum field, or "energy" or anything really, they are stepping soundly and firmly into philosophy. Which is totally fine to do. We just need to be aware of this, and start to apply different rules to evaluate the validity of the claims.
Edits are typos.
However, note that in my original post I didn't claim that physical laws describe what exists at an ontologically fundamental level, just what exists in reality. And by that I mean "exists" in the non-subjective way that, say, hydrogen atoms exist. I wouldn't mean to imply that hydrogen atoms are irreducible.
Does Deutsch and his research group make claims to know what is ontologically irreducible? Not as far as I know, but I am open to correction!
[1] https://journals.aps.org/prl/pdf/10.1103/PhysRevLett.128.080...
In the scientific sense, it is not enough to show that your system agrees with experiments to say it is true, you also have to show that the alternatives do not agree with experiments. Constructor theory is metaphysical because you'd get the same analysis whether you used it, ad-hoc intuitive thinking (in practice, people do not need instruction on universal principles like cause and effect), or some other alternative. All of the non-intuitive weight in that paper is borne by the QM stuff that exists independently of constructor theory, and in contrast, if you changed that the answers would start coming out wrong.
I am not proposing that this captures the essence of positivism, but one test that you can apply to check new ideas without going through the trouble of understanding them is to see whether they introduce any new conclusions that could not have been obtained under previous frameworks. If not, they may be metaphysical re-arrangements of existing ideas.
One of the points of the paper is that constructor theory makes such statements in a way that is compatible with other known laws. This is demonstrated via the experiment. So I'm afraid I do not understand your remark that "you could take the constructor theory language out and have the same paper". Taking the constructor theory language out would undermine a major point the authors are trying to make in the paper.
Yeah, but known laws are already compatible with themselves. You might make them more comprehensible by establishing a conceptual meta-framework that offers a consistent way to phrase statements about them, but that is clearly a philosophical task.
But I'll admit I don't understand constructor theory, and maybe it can hide any interesting discrepancies by overfitting.