Brian Cox says everything we’re most intimately familiar with is emergent. But aren’t those emergent qualities our starting point in all things? Physics us unfinished anyway.
“These things aren’t reducible to particle physics” is certainly begging the question. Why isn’t “exemplifying” or “satisfying” “making an illegal move”, “throwing a segmentation fault” reducible to particle physics? We have already agreed that the operations of a computer are ultimately explained by particle physics. And a computer is exactly what is needed to “satisfy” and “exemplify” these abstract concepts. What is not needed is knowing, or defining, what a segmentation fault or an illegal move is. That belongs to the abstraction, and ultimately to the human mind using the abstraction.
And so, I do not think your dilemma holds. An instance can perfectly satisfy an abstraction, by “exemplifying the properties and relations specified by the abstraction”, while embodying none of the normative features of the abstraction. Those normative feature define the abstraction itself, not the instantiations.
You asked “why isn’t exemplifying or satisfying ‘making an illegal move’ or ‘throwing a segmentation fault’ reducible to physics”. The answer hasn’t changed: “segmentation faults” and “illegal moves” can’t be defined in terms of microphysics. Therefore, insofar as a system is really throwing a segmentation fault, it is doing something that really can’t be defined in terms of microphysics. And that’s just another way of saying that its operations are not reducible to microphysics.
You say that we previously agreed that the operations of the computer are ultimately explained by particle physics, but that’s not what we agreed. We agreed that a faithful simulation at the microphysical level would reproduce the behavior of the system. But this is just acknowledging that segmentation faults and illegal moves are realizable in microphysical substrates. But “realized by” is not “reducible to”, the latter of which usually involves something stronger like identity, translation or elimination. Furthermore, since there is nothing in the microphysics that tells us what an illegal move or a segmentation fault is, the claim that the operations of the computer are “ultimately explained” by particle physics is just incorrect on its face.
You say that a system can exemplify a segmentation fault or an illegal move without anyone knowing or defining what they are. Yes, but this is beside the point. The claim wasn’t that knowing or defining these things is necessary for their exemplification, it was that what they are is not reducible to microphysics. So again, insofar as a system exemplifies them, it is exemplifying something that isn’t reducible to microphysics.
You’ve reiterated the claim that abstractions are denizens of the human mind as if that alleviates the pressure, but it just intensifies it. You say that a concrete system can exemplify all of the properties and relations specified by the abstraction without embodying the “normative features of the abstraction”, but if “normative features of the abstraction” refers to the normative structure defined by the properties and relations themselves, then this is clearly incoherent. You can’t exemplify the properties and relations without also exemplifying their constitutive normative structure. If the system is really making a chess move, then it is really exemplifying the normative structure that makes a chess move what it is. And this is why your distinction between the “boundary” of the abstraction and the “properties and relations” of the abstraction falls apart, because the boundary is nothing more than the specification of the properties and relations themselves.
But there’s an even deeper problem lurking that we haven’t yet addressed. Since you claim that everything in the world reduces to microphysics + arrangement, then this must include minds too. But then your strategy of pushing the “normative features of abstractions” out of concrete physical systems and into “the mind” just relocates the problem rather than resolving it. This is because the mind, on your view, is itself just another concrete physical system.
And so the dilemma resurfaces: either minds can instantiate these normative features (in which case other concrete systems can too), or they can’t (in which case they can’t be relocated to the mind). One way or another, you still owe an account of how these normative features reduce to microphysics + arrangement.
Anyway, it seems like we might be reaching the point of diminishing returns in this discussion. We’ve cycled through the same set of questions and answers a few times now and I doubt that further repetition will lead to any resolution. That said, I don’t want to cut the discussion short if you feel that there is more to be gained by continuing it. Just let me know how you want to proceed. Thanks.
I’m likely on a similar page then with Brian Cox.
Those emergent qualities are our starting point in our understanding of all things, but that’s different than the starting point in all things, which sounds more like supervenience than epistemology. Only under idealism is our starting point the same as THE starting point.
True. I guess what I mean is that explanations for that “starting point” are in flux. The starting point itself, the experience of the everyday, that stands. You can’t have the explanation come in and say the experience is a lie. All the explanations can do is say “I think that’s happening because of xyz.”
I think that’s where the argument for free will ends up: it’s happening right now. Why? What about…? What does that mean?". All those questions come looking for candidates.
OK
You can’t have the explanation come in and say the experience is a lie.
Some attempted explanations (some very normal looking scientific explanations) can be shown to invalidate experience like that, which is almost the same thing. Carroll got into that, theories that look good on the surface but cannot be simultaneously believed and empirically justified.
I think that’s where the argument for free will ends up: it’s happening right now.
The argument is happening? My take is that the argument is pointless if no example of free will aiding a better decision can be found. I sound like something horrible to be avoided/selected-out if possible.
TBH I’m not satisfied.
I was not reiterating the original question. I was asking, why can’t something that satisfies or exemplifies “segmentation fault” or “illegal move”, in other words an instantiation of these concepts, be reduced. You replied in terms of definitions, which is the domain of abstractions. And you are skipping multiple “levels” between “segfault” and microphysics, thereby crossing the line between abstraction and concrete object.
Segmentation faults can be defined in terms of computer architecture. Computer architecture can be defined in terms of transistors. Transistors can be defined in terms of microphysics. And so by the transitive property, segmentation faults can be defined in terms of microphysics.
But note that in this process of definition, we have moved from the purely abstract “segmentation fault” and “illegal move” to a fully concrete physical object. When defining a “segmentation fault” in terms of computer architecture, we fix the range of possible realizers of that concept to a specific computer architecture. By defining architecture in terms of transistors, we fix the range of realizers of that architecture to a particular transistor-level implementation. And by defining every transistor microphysically, we fix the range of realizers of that implementation to an actual, specific physical computer.
And so, we cannot define “segmentation fault” in terms of microphysics because this is a category error: defining a fully abstract concept in terms of something fully particular.
This is again crossing the line between abstraction and instantiation. To say that segmentation faults and illegal moves can be realized by microphysical substrates, is to say that the concepts “segmentation fault” and “illegal move” can be satisfied by particular concrete objects that are microphysical. Then the follow up question is, are these particular objects reducible to microphysics. It is concrete objects that are reducible to physics, not abstractions.
This line is just what I have been arguing against for the past several posts. It is not the responsibility of the specific object to define what an illegal move is. This is the normative function of an abstraction.
"Normative features of the abstraction” does not refer to “the normative structure defined by the properties and relations themselves”. I don’t know what this even means. The normative features of the abstraction answer the question: what counts as the concept? What is, and isn’t, a “segmentation fault” or “illegal move”? This is what I meant by the normative boundary of the abstraction. It is this boundary that divides the world into “illegal move” and “not illegal move”, “segmentation fault” and “not segmentation fault”.
And so my point was not that “a system can exemplify a segmentation fault or an illegal move without anyone knowing or defining what they are.” Rather, a concrete system is completely disjoint from the question of what a segmentation fault or an illegal move is. The concrete reality is what it is, the abstraction is the normative structure determining whether that concrete reality does, or does not, belong to the abstract concept.
This is totally unreasonable. Placing normative features of abstractions where they belong, in minds, is a positive result whether or not the hard problem is resolved. The problem of minds is very much a special case of the problem of reducibility. If you would have your way, unless we resolve the hard problem, we have to accept that abstractions belong to concrete physical systems?
Our discussion is currently organized around a set of disagreements across multiple fronts, but I think these multiple fronts are actually manifestations of a single core disagreement. But before we take this any further I want to make sure that I am tracking your position.
As I understand it, you hold that (1) reality consists in microphysical particulars in spatial arrangement; (2) the only mind-independent manifestation of “form” is spatial arrangement itself; (3) all other apparent forms — including laws and dispositions — are abstractions generated by minds for tracking the underlying microphysical reality; and (4) minds themselves will eventually be shown to fit into this picture, though the details remain to be worked out.
Is this a fair statement of your position?
Not really.
-
Reality fundamentally consists in microphysical particulars in spatial arrangement, and natural laws. All explanations terminate here.
-
Higher level forms exist, and are mind independent. But their existence is equivalent to, and so fully explained by, and reducible to, the dynamic interplay of lower levels given their specific configuration.
-
Abstractions are mental tools that organize messy analog reality by partitioning it into conceptual buckets. The patterns they track are often real, mind independent patterns. But the partitioning of reality into entities which either do or do not fit into discrete labeled buckets is an imposition of the mind.
-
Minds are a separate question. Abstractions, and minds themselves, may or may not be ultimately reducible, independently of whether pedestrian higher level forms are. (Though it is hard to imagine a world where only minds are reducible).
If I’m understanding you correctly, you are saying that the mind must use abstractions to track mind-independent form. In other words, the mind doesn’t have any kind of “direct” access to mind-independent form that bypasses the use of abstractions. Is that right?
It is an interesting question. I would say that the mind cannot understand a form without fitting it into its existing network of abstractions.
hence not actually mind-independent. Precisely Einstein’s complaint about quantum physics.
And this would apply to the fundamental level too right? The mind’s knowledge of microphysical particulars, laws and arrangements is also mediated by abstractions?
Yes, that’s right. “Electron” is no less an abstraction than “chess computer”.
And do you agree that the abstractions that track higher-level forms are not reducible to the abstractions that track the fundamental level?
I don’t think it makes sense to say abstractions themselves can be reduced. They can be decomposed. For instance, “archipelago” can be decomposed into “island” and “cluster”. But this is not reduction.
This is conflating access and nature. As minds, of course access is mind dependent. But this says nothing about nature.
This applies to qm too. We are conscious beings, so of course our access to the world requires a consciously apprehended measurement/interaction. But does this place consciousness at the heart of reality? Or just interaction?
Obviously interpretations of quantum physics are controversial and it’s a frequent thread de-railer. Suffice to say that what quantum physics called into question was whether we can really know the fundamental particles of physics independently of the act of measurement. There are many often-quoted Copenhagen aphorisms about this, such as Heisenberg’s ‘What we observe is not nature in itself, but nature exposed to our method of questioning’ or John Wheeler’s ‘No phenomenon is a real phenomenon until it is an observed (registered) phenomenon’ (Law without Law)
And if you say it’s just interaction, the question for scientific realism is, what is the observer interacting with. And that turns out to be just what cannot be stated. That may be an epistemic rather than ontological attitude, but you can also argue that what we know exists must inevitably be conditioned by the ways we come to know it.
OK. You previously agreed that your knowledge of form at all levels is mediated by abstractions. You’ve now also said that higher level abstractions can’t be reduced to fundamental level abstractions. So with regard to your claim that higher level forms are reducible to fundamental level forms — how do you know this, given that, by your own admission, your knowledge of these forms is entirely mediated by abstractions that can’t be reduced?
We need to distinguish abstraction qua mental construct vs. abstraction qua what the abstraction represents. We use abstractions, which are not reducible, and which are necessary for thought, to determine that what a particular abstraction represents is reducible. We use X to determine that Y is Z. You will need to explain why X needs to be Z for this to obtain.