The SEP article on scientific reductionism notes that the etymology of the word “reduction” is “to bring back” something to something else. So in a methodological sense, reduction is bringing one theory or ontology back to a simpler or more fundamental theory or ontology. The Wikipedia entry on reductionism identifies different kinds: ontological, methodological, and theory reductionism. I think the ontological one is the most interesting here, the proposition that all of reality consists of a small number of building blocks.
Most reductions aren’t particularly controversial, at least not in science. There aren’t many arguments that chemistry doesn’t reduce to physics, or geology to both those sciences. Today it’s not controversial that biology reduces to them as well, although this is a relatively recent development.
As late at the early 1900s there were people arguing that life was somehow different, that it was distinguished by a vital force, an ancient idea. Few talk about vital forces today. Biologists learned about evolution through natural selection, genetic inheritance, proteins, DNA, RNA, and overall organic chemistry. Life is now seen as largely a molecular chemical enterprise, albeit a hideously complex one.
This raises an important point. Most reductions are conservative, retaining the reduced concept, but not all. Sometimes it’s eliminative, as in the case of a vital force, or other things like phlogiston or a luminiferous ether. It seems to depend on whether the reduced concept remains useful.
Today there remain at least two areas where people tend to resist reductionist accounts: consciousness and quantum measurement.
The consciousness one goes back to Rene Descartes’ famous distinction between mental and physical substances. Descartes saw no issue with a mechanistic understanding of reality, except for the mind, which he could not conceive of being reducible to mechanisms. He was far from alone. Gottfried Leibniz presented his mill thought experiment, that if the mind were a mill which we entered, we wouldn’t find anything there that explained perception. The mind, he agreed with Descartes, had to be a different kind of thing entirely.
Although a lot of what these guys saw as irreducible has been reduced. Today, psychological concepts like memory and cognition are understood to be neural processes, albeit with still many unanswered questions. But contemporary philosophy of mind often draws a new line at perceived characteristics, typically called qualities or qualia. Because these characteristics are introspectively opaque, they seem irreducible. And studying some of them has proven hard, therefore many assume they’re fundamentally inaccessible to anyone but the subject.
The question is whether the notion of fundamental qualia really explains anything. Does it convey meaningful information? Certainly qualities understood as just perceived characteristics seem useful enough. But regarding them as fundamental seems to obscure rather than convey information.
As a reductionist, I think of qualities as categorizing conclusions. (If that seems radical, consider that the etymology of the Latin root phrase “qualis” is “of what kind.”) Our nervous system qualifies a stimulus for a category when a particular range of neural firing patterns trigger a galaxy of associations, some innate, but many learned, which collectively add to the richness of the experience of that perceived characteristic (redness, sweetness, pain, etc).
Am I completely confident this is the answer? No, but as an explanation, it seems like a more fruitful place to explore. I suspect future scientific studies will validate some aspects of it, but not others. But even if it’s completely wrong, these kinds of theories seem to spur more experimental work than simply assuming qualities are fundamental and inaccessible.
In the case of quantum mechanics, it’s observation that’s often taken to be fundamental. In its strongest forms, this ends up pairing with the idea of consciousness being fundamental. Although the more cautious variants see just measurement as fundamental (or interaction). This can be the idea that quantum states don’t really exist, that measurement itself creates reality, or that quantum states do exist but physically collapse in a measurement, a fundamental change in reality.
In the early years of quantum theory, something like these views seemed inescapable, and most of the physics community closed ranks around them. But there were holdouts, including Albert Einstein and Erwin Schrὅdinger, who kept digging, discovering the phenomenon of entanglement, which would later be used by David Bohm and Hugh Everett to posit mechanistic explanations for the disappearance of quantum effects. But it was the work of H. Dieter Zeh and Wojciech H. Zurek in the 1970s and 80s that really fleshed out the detailed explanation we now call decoherence.
Today, few question whether entanglement and decoherence happen, although many do continue to argue that they’re only useful mathematical tools. Even if they are real physical processes, whether they serve as a full explanation of what’s happening in measurement depends on your preferred interpretation of quantum mechanics. But the key thing is it’s an explanation that wasn’t found by those who were satisfied with measurement being fundamental.
Which gets to why I’m a reductionist. I can’t prove that ontological reductionism is true. Maybe there are unique aspects of reality that aren’t built on a few common building blocks. But there seems to be a lot of history showing that assuming it’s true is far more fruitful than assuming complex concepts are fundamental. From Thales positing that water was the fundamental substance to later Greeks assuming there were four fundamental elements, the history of assuming anything is fundamental seems cautionary at best.
Which is why when I hear “X is fundamental,” I’m reflexively skeptical. We can’t even confidently say that about “elemental” particles, quantum fields, space, or time. We only seem able to talk in terms of something being more fundamental or less fundamental. Scientific theories are always provisional, subject to change on new data. Absolute fundamentality seems like an assumption we can never justify. Calling something fundamental seems to say, “There’s nothing left to explain here. Stop digging.” A lot of progress seems to happen from the people who ignore these prescriptions.
What do I mean by “progress”? None of this is to argue that higher level concepts aren’t useful; thermodynamics, for instance, didn’t cease being a useful concept once it was reduced to particle physics. Or that holistic takes on phenomena can’t be beneficial. Or that in art or daily life, we can’t appreciate things without reducing them.
But reduction aids in acquiring more structurally or causally complete explanations, while assuming something is fundamental often seems to paper over structural or causal gaps. Closing these gaps, when achievable, provides more reliable knowledge, knowledge which gives us new abilities, abilities such as medical scanners, drugs, computers, and many other things. Yes, that does include nuclear weapons and other ills. It doesn’t seem like we can have the good without the bad, although usually the bad can be managed with more reliable knowledge.
At least that’s my view today.
What do you think? Are there benefits to non-reductive approaches I’m overlooking? Or drawbacks to reductionism I’m missing? If you think an alternative approach is better, what are the benefits of that alternative?
SelfAwarePatterns 
I obviously enjoy your posts … and I have comments! :o)
Re “Which gets to why I’m a reductionist. I can’t prove that ontological reductionism is true.”
Wait, we have to be absolutely convinced that reductionism is true before we can use it? It can’t just be a good place to start due to an impressive track record? This smacks of the common (stupid) question “Can science know everything?” Science is the best place to start to understand natural processes, even claimed ‘supernatural’ processes. One doesn’t have to be convinced it is the end all of approaches to understanding to use it. Just as I would start trying to remove a stubborn nail with a hammer, there are other tools designed to remove nails from wood, but a hammer is a good place to start, plus I own one.
People who “believe” without evidence (what I call religious beliefs to distinguish them from ordinary beliefs (If it rains I might need an umbrella) seem to feel that non-religious people must have similar beliefs (Science is the truth, the way, and the life!) when we are more like baseball pitchers who look at a batter they struck out recently on a changeup, but they don’t want to just throw changeups, they need to mix in other pitches that lead to that strikeout pitch. We are just trying what seems to work, not believing in magic because that batter may hit my next changeup over the fence for a home run.
LikeLiked by 2 people
Thanks!
Hopefully you kept reading past that quote, because I take myself to have made points largely agreeing with yours. It really comes down to the classic problem of induction, the idea that while we may have evidence for a particular approach, it’s never iron clad. But that doesn’t mean the evidence is useless. It indicates there is a high probability of the approach continuing to work. And as I noted about all scientific theories being provisional, the best we ever get are models that work for now.
Still, there are black swans. So there’s value in acknowledging that a working model could fail at some point, that in fact it likely will. Although usually the domains where it fails are so outside of anything anyone has imagined, it rarely rescues competing ideas.
LikeLike
@selfawarepatterns.com Your thoughtful toot brought back memories of a *highly contentious* symposium in #NewYork during a 2003 national meeting of the American Chemical Society. Many of the participants and their contributions are included in this review of opposing philosophical views bonding in #chemistry represented by Roald Hoffmann (famously anti-reductionist) and Richard Bader (famous for extending #quantum theory to atoms *in* molecules). https://cen.acs.org/articles/85/i5/Chemical-Bond.html
LikeLiked by 2 people
Remote Reply
Original Comment URL
Your Profile
Thanks. I had no idea chemical bonds admitted to that much interpretation. But I guess it makes sense. It also reminds me of a debate on Twitter a few years ago about whether electrons are real.
LikeLike
what about language? You don’t say anything about that even though the whole post is written in language. How do we reduce the linguistic thoughts in the post to mere particles?
LikeLiked by 2 people
Can’t say I’ve given this much thought, but my initial off the cuff take is that language is communication between minds (although increasingly also with LLMs). So the main lift might be reducing it to thought, although I suspect a lot of sociological and anthropological factors would have to be taken into account.
We’d then have to depend on cognitive neuroscience eventually succeeding in reducing thought to neural processing, which itself will likely require embodied, enactive, and especially extended concepts. After that we’re tapping into the reduction of neuroscience to biology, biology to chemistry, etc.
LikeLike
At some stage perhaps the “reduction” ends and one discovers the fundamental. So “fundamental” can be by no means dismissed. Is it “strings” perhaps”. Or “information” like Wheeler and your Eastern European Oxford Prof. Or is reality actually a bottomless well, never to be reached. In which case nothing is fundamental. The most important thing is never to close one’s mind.
LikeLiked by 2 people
Definitely on not closing our mind. The problem with the fundamental is, how do we know if we’ve hit it? Even if we can’t find anything more basic, we can’t know that future generations won’t, maybe centuries or millenia down the road. The only thing I might find convincing is if we eventually hit pure structure anchored by points that are nothing but that, points, with no qualitative explanation necessary. Even then, there might still be more underlying reality we’ve just never conceived of yet.
LikeLike
Reduction to the point of identifying, maybe not the fundamental, but the usefully composable elements seems the natural stopping point. Stories, reduce to chapters, reduce to scenes, reduce to paragraphs, to sentences, to words. At this level, one can now build upwards, constructing any tale imaginable. Reducing to words was enough. Could we reduce to syllables and then letters, strokes, lines and curves? Pixels? Why? Words get us to an infinitely composable platform.
LikeLiked by 2 people
Though, words as thoughts can be decomposed into their constituent attributes, apple: sweet, round, fruit, seeds, seasonal, etc. Those themselves are words. At this point we have no choice but to use words to reduce other words. Does that make words, and their correlating thoughts fundamental?
LikeLiked by 1 person
I think it depends on our goals. For a story, we might start with a plot, setting, characters, the protagonist’s goals, their obstacles, such as an antagonist, etc. Then focus on the presentation, which for literary fiction would get us into the chapters and all the rest. But if our goal is to understand story, we might end up looking at psychology.
There’s an argument that all words ultimately reference either sensory perceptions or actions. Even words referring to abstract phenomena tend to start with sense and action terms as metaphors. For example, “democracy” comes from the Greek terms for “people rule.” Of course, there are people who insist perceptions are irreducible.
But that’s a frequent thing in reducing. Often there are limits in a particular domain to how far things can be reduced. I always use the example of a bit in software. Within the operations of software, it’s often the most fundamental thing. We know if we want to reduce further, we have to switch to the hardware view. But if we’re only trying to understand a piece of software, bits are usually as farthest down we need to go.
LikeLiked by 1 person
Bits are as far down as you can go … unless … the software under consideration is analog and you are just simulating with digital computing, in which case the farther you want to go down, the more bits you need.
LikeLiked by 1 person
Right. But I think you’re talking about reducing inside the simulation being run on the software. But, on a digital system, it terms of the software’s own operations, even if hand coded in machine language, the bit is the lowest thing it can operationally deal with. (At least it was in my assembly language days. I’m not up on all the latest instruction sets and architectures.)
LikeLiked by 1 person
When it comes to consciousness, there are two ways to resist a reductionist account. One is to refuse to accept the reduction of consciousness to something else—typically matter. The other is to refuse to accept the reduction of everything else to consciousness. But those who advocate an ultimate reduction to consciousness are not usually thought of as reductionist, because “reductionist” is understood also to imply reduction to “the physical,” and “the physical” is assumed to be matter, when all is said and done (though now matter is often qualified as encompassing “energy”).
LikeLiked by 1 person
Good point. Strictly speaking, what we’re reducing to could, in principle, be something other than physics. Ontological reductionism is merely the idea that everything reduces to a minimal number of parts. There’s nothing in that view itself requiring that those parts be physical. For an idealist, it’s all reducing to mental ideas.
But I think that’s where the other types of reductionism come in. In science, there are typically bridging laws between reduced and reducing theories, mapped relationships. Chalmers talks in terms of an a priori relationship between the reduced and reducing theory.
His argument, as a property dualist, is that no such relationship has been or can be established. But this then gets into what we think consciousness itself is. If we think it’s functionality, something structural and relational, then there’s no barrier, at least in principle, to a logical mapping between the two. But if we think it’s something beyond functionality, something non-relational, then that mapping starts to look impossible, even in principle.
The question is how we ever establish which is reality.
LikeLike
If the choice is between something possible in principle and something impossible in principle, I know which one I’d choose. And In fact, consciousness as the basis of a structural reality seems like a promising approach. It leaves us explaining the physical in terms of the functional relationships of consciousness. The other road is to explain consciousness in terms of the functional relationships of the physical. But as soon as we talk about “functional” relationships, we enter the realm of purpose and meaning, and are left grappling with the problem of how to introduce this to the merely physical. If we posit consciousness as fundamental, the notion of something having a function seems less puzzling.
LikeLike
Funnily enough, this story has just shown up on slashdot.org:
https://science.slashdot.org/story/25/12/27/042221/should-physicists-study-the-question-what-is-life
“There are, however, ways other than reductionism to think about what’s fundamental in the universe.“
LikeLike
If something is actually possible in practice, who’s more likely to discover it, the person who believes it’s possible, or the one who’s concluded it isn’t?
Functional relationships, that is, cause-effect relationships, aren’t necessarily goal related, although it seems reasonable to assume most in the mind are. But it also seems like evolution through random variation and natural selection gives us everything we need for purpose and meaning in biological systems.
For me, the question about positing consciousness as fundamental is what lines of research does it encourage? Is it useful in medical research? Psychological therapy? AI research? What new abilities could it lead us to?
On Adam Frank’s article, my question for him is, what specific changes does he want to see in scientific research? He was asked that in a podcast interview recently, and he didn’t really have anything, just a strong pitch that it needs to happen. My impression is that the people who have historically changed how science was conducted typically did it by first achieving success with their new methods, then publishing about it.
LikeLiked by 1 person
“If something is actually possible in practice, who’s more likely to discover it, the person who believes it’s possible, or the one who’s concluded it isn’t?”
It wasn’t me who suggested that one of the alternatives might be impossible. Taking such a position may well be short-sighted, and perhaps there is something to be said for a non-relational consciousness after all. But it’s not a position I need to take, so I’ll leave that up to you. I’m happy with the alternative idea that consciousness is relational, which as you say leaves no barrier to the mapping problem you raised.
Now you have raised another problem of what good it might to do posit consciousness as fundamental. The major problem it helps solve concerns consciousness itself: what purpose it serves in a world that supposedly could run just as well without it, on purely mechanical terms. We posit it not because our mechanical account really needs it, but because there it is, and so we had better fit it in somehow. We may even find ourselves calling it an illusion for exactly this reason.
The physical world seems obvious and necessary, and we are naturally drawn to a reductionism in that direction. Having started on such a course, we find ourselves having to account for consciousness, as if it must therefore be secondary. But we forget too quickly that consciousness is also obvious and necessary, and perhaps is equally deserving of consideration as something that might be primary.
LikeLiked by 1 person
“in a world that supposedly could run just as well without it”
If by “world” we mean to include agents with the same abilities we have, then to me this seems like a major assumption. As you describe, a lot of problems follow from it. It seems like a postulate that should receive more justification by those who hold it. Without it, consciousness seems like a part of the world that can be studied like any other.
LikeLike
When I talk of a world that supposedly could run without consciousness, I am of course talking about one that could run without agents. This is in fact the standard reductionist account. Agents do not appear until the world is well along, and everything that happens in that world can be explained without agents—even after those agents have appeared. The behaviour of the agents themselves can be reduced to things that are non-agentic.
This, I think, is the true spirit of reductionism as we usually conceive it. Once we turn to agents as a requirement to explain some part of the world, we have admitted the defeat of the more ontologically compact world where “all of reality consists of a small number of building blocks.” Either this totality requires agents for its explanation, or it doesn’t. So I have to ask: does your reduced world “mean to include agents with the same abilities we have,” or can it get along without them, as initially proposed? Or again, perhaps we need to include agents after all among the small number of building blocks needed to explain all of reality.
LikeLiked by 1 person
“does your reduced world “mean to include agents with the same abilities we have,” or can it get along without them, as initially proposed?”
It currently includes them but, as you noted for the reductionist view, the universe existed for a long time before they emerged. And it will exist after there’s no longer energy for agents. How and why do we get agents like us in a universe once without them? Before Darwin, this did seem like a pressing question. Today the answer is evolution through variation and natural selection.
But that answer only works if consciousness is part of the causal chain.
LikeLike
The point I’m trying to make, the one I think you need to hear, is that you have some ontological choices to make. Either you commit to a simple ontology that does not need agents, or one way or another you have to get more sophisticated about the idea of an ontology. I think some others here have proposed alternative ways of modelling ontology that might help. But you can’t just add agents to your simple ontology as needed and pretend to be fully consistent.
LikeLike
I’m always interested in hearing something I need to. But the statement by itself isn’t really landing for me. I can’t see that reality requires agents like us, but can see how we arise in a reality without them through an evolutionary history. My question would be, in what specific way do you see that as inconsistent?
LikeLike
Well, you say that you can’t see that reality requires agents like us. This is consistent with the idea I proposed earlier of “a world that supposedly could run just as well without [consciousness].” Yet when I made this point, you apealed to the fact that the world as we know it contains conscious agents:. “If by ‘world’ we mean to include agents with the same abilities we have, then to me this seems like a major assumption.”
The assumption you seem to be questioning is my statement that the world could run just as well without conscious agents. And yet just now you’ve said that you can’t see how reality requires agents like us. All it needs, according to this post, is a small number of building blocks, and agents need not be among them. So the question is, do we need agents or not?
I don’t know how I can put this more clearly.
LikeLike
Ah, ok. Let me clarify what I meant by: “If by “world” we mean to include agents with the same abilities we have, then to me this seems like a major assumption.”
I was calling into question the idea we can have agents like us and not have them be conscious. In other words, I took you to be saying we have to explain why we don’t live in a zombie world. My point was that the possibility of zombies seems like a major assumption. But if you were just saying we have to explain why the world has conscious agents in it at all, then my response is back to the evolution one.
Hope that clears things up.
LikeLike
I’m afraid it doesn’t, because those two questions are the same question.
If the world runs mechanically, by pure cause and effect, then what seem to be agents are more parsimoniously explained as zombies, machines made out of simple building blocks (in elaborate configurations). Then we have to explain why the world has conscious agents in it at all.
I hope you see how they are the same question. The appeal to evolution must apply in either case. But if agents can evolve from the simple building blocks by the mechanical, cause-and effect operations of evolution, why can’t zombies evolve the same way?
The possibility of zombies is not a major assumption that I’m making, but a serious consequence of the logic you are presenting. The real major assumption here is that anything but zombies is to be expected of the reductionist account. The only reason anyone has to make it is that there are non-zombies, or agents, in the world, and this unexpected fact needs explaining.
Thus we still have the same contradictory claims:
– We can explain everything in the world as simple building blocks, without reference to agents.
– To explain some things in the world, we need to appeal to agents. (Oh, but we can explain them by evolution!)
The trouble is that the agents reduce to non-agents and non-agentic causes and effects, and logically we are back to the first statement. “Evolution,” for all its power to explain the progress of agents once they appear, is in this case a magical invocation for getting around the problem..
LikeLiked by 1 person
As always, it depends on what we mean by “consciousness.” If we mean functionality, then the idea of a zombie seems like a contradiction, an attempt to posit a functionally conscious being without the functionality of consciousness.
For zombies to be a meaningful concept, we need to be talking about something other than functionality. But now we need to consider the implications for this view of consciousness. It appears to entail something causally impotent, that can’t be having any effect on this conversation we’re currently having about consciousness because it’s one zombies could just as easily have. All the arguments I’ve seen against this seem to involve sneaking functionality back into the picture.
And consider that, in a zombie world, the zombies would hold a model of themselves as conscious beings. They would argue that they are conscious. They’d resist admitting that they aren’t as much as we would. Which raises the question; if zombies are possible, how can we ourselves ever be sure we aren’t ones? What argument can we make that the zombies themselves couldn’t?
So for me, non-functional consciousness is a redundant concept. I’m not even inclined to call it an illusion, because I can’t see any valid reason to think our internal impressions are about anything other than functional consciousness.
So to be clear, when I say evolution can explain conscious agents, I mean functionally conscious agents, the only ones I think are there, at least for now.
LikeLike
I think it depends rather on what we mean by “functionality.” Earlier in our conversation, you said that “Functional relationships, that is, cause-effect relationships, aren’t necessarily goal related.” With this I can agree. A machine like a car can be understood as having functional relationships with the world, without invoking goals on the part of the car. Any goals associated with it are introduced by our human uses for its functionality.
When you say ‘For zombies to be a meaningful concept, we need to be talking about something other than functionality,’ I have to wonder if you mean the same thing by ‘functionality.’ A zombie is exactly like a car, at least for the present purpose: it has a cause-and-effect functionality, without having goals. That’s all we need to talk about, and exactly what we need to talk about. So here again we have a confused position. As expressions of pure cause-and-effect relationships, zombies are functional, and yet to talk about zombies meaningfully we have to be talking about something else than functionality.
What puzzles me more is your assertion that “zombies would hold a model of themselves as conscious beings.” I suppose that depends what we mean by zombies! By definition (so I thought) they hold no models at all, being pure automatons, more complicated than cars but the same in kind. Reading this, I have a feeling that if we explored your thoughts about zombies, we might run into more contradictions.
I would feel better about it if these contradictions were addressed frankly, but instead we seem to be moving from subject to subject without ever seriously grappling with the difficulties raised,
As for non-functional consciousness being a redundant concept, I’m not even sure what that means, but no one is contesting the functionality of consciousness. What’s being contested, to bring us back to the point, is the need to invoke consciousness at all if the world is purely a machine, like a car or a zombie but even more complex.
Evolution, as the science of biological development, does not even pretend to address this problem; it begins and ends with biology. The attempt to apply it to the transition from non-life to life is a side project, a hobby of reductionists who are looking for a way to explain why their machine-world has acquired consciousness—in this case by pleading, illogically, that it did so because it needed consciousness to be what it now is. The machine doesn’t need consciousness, but it does. Borrowing the genuine authority of evolution for this considerably more speculative purpose is dubious at best, and seems to be a nest of confusions when examined.
(I apologize, by the way, for the slow pace of my replies. I’m travelling in Scotland for Christmas, and being plied with whiskey and tall tales by many a generous host. It’s hard to find moments to write.)
LikeLike
I’m happy to address any contradiction you see me making. Hopefully something in what follows does that. (If not, just specify the specific contradiction you see and I’ll zero in.)
My notion of zombies is the one from the SEP article on the subject. (emphasis added)
For a zombie to talk about consciousness, it must have information about the subject matter and a working model to produce behavior associated with it. If we’re talking about a version of a zombie that can’t have that, then I don’t see how they can physically meet the requirement of acting just like a conscious being.
In particular, if it can’t behave with (at least apparent) purpose or talk like it has consciousness, then it no longer seems to match the traditional definition of a philosophical zombie. Once there are behavioral differences between a zombie and a conscious being, then the question of why we don’t live in a zombie world seems moot. Unless I’m missing something?
“in this case by pleading, illogically, that it did so because it needed consciousness to be what it now is.”
I’ve never made this argument, at least not intentionally, or even seen it made by anyone in the origin of life research, at least not anyone credible. I certainly don’t believe life evolved because “it needed consciousness.” Sorry, not even sure what you’re trying to say here.
(Totally understand. Take your time. Sounds like a cool trip!)
LikeLike
The SEP article is far from definitive. Indeed, it concludes, “In spite of the fact that the arguments on both sides have become increasingly sophisticated — or perhaps because of it — they have not become more persuasive. The pull in each direction remains strong.”
The statement you’ve highlighted, that “zombies behave just like us, and some even spend a lot of time discussing consciousness,” doesn’t support your idea that zombies “hold a model of themselves as conscious beings.” The pertinent paragraph (in Section 4.2) reads:
“Zombies’ utterances. Suppose I smell roasting coffee beans and say, ‘Mm! I love that smell!’. Everyone would rightly assume I was talking about my experience. But now suppose my zombie twin produces the same utterance. He too seems to be talking about an experience, but in fact he isn’t because he’s just a zombie. Is he mistaken? Is he lying? Could his utterance somehow be interpreted as true, or is it totally without truth value? Nigel Thomas (1996) argues that ‘any line that zombiphiles take on these questions will get them into serious trouble’.
The objections raised involve a hidden, or obscured, assumption that the zombie has some kind of consciousness, either access or phenomenal. Most of the article rests on this assumption, which is derived from the argument that in this world, when you assemble certain physical components in a certain way, consciousness appears; therefore, zombies assembled in this physical world must have consciousness. The article mostly begs the question of whether the physical components, as we understand them, are sufficient to give rise to consciousness when assembled. That is taken as an observed fact, rather than the unexplained assertion it turns out to be when examined closely. Here and there, the author (Robert Kirk) responsibly considers alternatives, but in the end they are given short shrift: “Abandoning causal closure appears to conflict with empirical evidence; while the idea that phenomenal or quasi-phenomenal properties are fundamental remains obscure.”
Reading the article reminds me why I took a dislike to analytic philosophy. To someone who has neen exposed to scholastic philosophy, it sounds very much of a piece. I attribute this to a similarity of situations. The scholastics were struggling to make logical sense of a received wisdom that inherently defied logic. The analytics are struggling with a different received wisdom, but they are in the same predicament regarding its problematic nature. Thus, for all their erudition, many of them are like dunces counting angels on the head of pins.
—
‘ “in this case by pleading, illogically, that it did so because it needed consciousness to be what it now is.”
‘I’ve never made this argument, at least not intentionally…’
Perhaps I should have said “agents” instead of “consciousness.” This goes back to when I said, “in a world that supposedly could run just as well without it”, and you responded, “If by ‘world’ we mean to include agents with the same abilities we have, then to me this seems like a major assumption.” That is, if we are to explain a world with agents reductively, we need to remember that it contains agents, and this goes to prove the explanation that agents can be explained reductively. It’s the same question-begging I mentioned above, and really the essential contradiction that needs to be addressed. The others are variations of it. To repeat, the contradiction is:
– We can explain everything in the world as simple building blocks, without reference to agents.
– To explain some things in the world, we need to appeal to agents.
It’s a clear question, deserving of a clear answer. To which of these statements do you subscribe?
LikeLiked by 1 person
I take the purpose of the SEP article to be to educate the reader on the philosophical concept of zombies, not to argue for or against the idea. I know there are articles that do go into full advocacy, but I actually prefer the ones that just discuss all the pluses and minuses.
I find this contention about a zombie having a model puzzling. The device you’re using to read this has models, structures of data allowing it to respond productively to certain inputs and outputs, and most people don’t see it as conscious. But we can change what I said to “has a physical configuration that leads to it insisting it is conscious.” We still end up with the issues I noted.
“To repeat, the contradiction is:
– We can explain everything in the world as simple building blocks, without reference to agents.
– To explain some things in the world, we need to appeal to agents.”
Not sure what “appeal to agents” means. But if we’re explaining the world with agents in it, then obviously agents will be part of the account, the same way coffee makers would. But there’s no requirement for them to be included in the simple building blocks. They’re among the patterns those building blocks form.
Hope that helps.
LikeLike
Happy new year!
The SEP article makes a responsible effort to cover all the positions, but the sentiment or attitude I sensed throughout seemed to take a physicalist ontology for granted. Perhaps the skeptical eye is oversensitive to such things.
My point about models concerns not so much the modelling function of things. As you say, a computer incorporates models; so does a toaster or a car. What troubles me is the idea that a zombie would hold a model of itself. One wouldn’t normally say of a computer that it possesses a self, much less models one. At some point you want to say that a physical object transcends this limitation and acquires a model of itself. The criterion for crossing this threshold seems to be whether it “has a physical configuration that leads to it insisting it is conscious,” but to be generous, I think this must fall short of what you mean. One can easily configure a tape recorder to do that, but the tape recorder is certainly not modelling itself for the duration of the configuration.
By “appeal to agents” I mean basically the blind spot in your argument. If we’re explaining the world with unicorns in it, then obviously unicorns will be part of the account. You assume agents in order to account for them; to me this seems illegitimate.
LikeLike
Happy new year Jim!
Ok, so you see me with a blind spot, which if so, I’m obviously blind to. But the contradiction I see you holding is that you don’t seem to buy the traditional concept of zombies, that they are physically and behaviorally indistinguishable from a conscious being, but you still see it as a mystery why we don’t live in a zombie world.
Not sure if there’s anyway to break out of this logjam. But I guess that’s the conclusion to most philosophical disagreements. 🙂
LikeLike
I can accept that zombies would be physically and behaviourally indistinguishable from conscious beings. This implies either that there or no zombies, or there are no conscious beings. There are conscious beings; therefore, there are no zombies, and we do not live in a zombie world. There is no contradiction in this account, and it does not worry me.
But taken alone, it’s not sufficient to explain why there are conscious beings instead of zombies. The mere fact of conscious beings is not an explanation for them. This is where I see the reductionist account as confused. It assumes that the existence of conscious beings somehow suffices for the reductionist explanation of them, despite the fact that its building blocks do not predict them, and everything that develops from those building blocks can be described completely in terms that do not involve them.
Increasingly, I feel that breaking out of this philosophical logjam is important to the future of science and philosophy, and so I’ll drag this out a little longer, although like you I sense it’s time to give this particular exchange a rest. Let’s try looking at it another way.
For a reductionist, coffee makers can be explained without reference to coffee makers, in terms of the basic building blocks coffee makers are made from. One can also explain coffee makers in terms of their behaviour or functionality, but these are not considered to be part of the reductionist account; it does not pretend to explain where coffee makers came from or why they are needed in the world.
When it comes to agents, the situation is different. The ambitious reductionist tries to explain why they came to be, why they have appeared in the world. In the case of coffee makers, it’s easy enough to discuss their appearance in terms of the agents who use them. In the case of the agents, though, there can be no appeal to anything prior; they have to be bootstrapped from nothing. Pointing out that there happen to be agents, who bring to the world purposes for themselves and other things, is not a satisfactory answer, but s form of begging the question. It seems to me that something else must be added to the building blocks, either at the very beginning, or bootstrapped from nothing at some later point, if we are ever going to explain coffee makers fully. You can call them “agents,” and I don’t mind. It’s where they show up and why that interests me.
I acknowledge that there are various stories about how they “evolved,” and while I don’t want to rain too much on those admirable and interesting efforts, I think they’re barking up the wrong tree. I reiterate that evolution assumes some sort of agency; it cannot be invoked to account for itself. Either something about it is built in to the world, or it, too, shows up inexplicably at an undefined later point. Invoking it to explain agents is just a way of prolonging the circularity.
LikeLike
Ok, let’s look at a reductive scenario for how agency evolves.
The building blocks available to a reductionist include the laws of thermodynamics, particularly the second one, emergent from particle physics which mandates that entropy always increases. Another name for entropy is information. (See the similarities between Boltzmann and Shannon’s equations. https://en.wikipedia.org/wiki/Entropy_(information_theory) ) And another name for information is complexity.
So we have an inevitable increase in complexity over time. In the right substrate, that will lead to processes that replicate. (Think crystals or clay for simple versions.) Although the replications are never perfect. There are always slight variations due to stochastic jostling from the environment. Inevitably there will be shortages in the substrate that limit that replication, leading to some patterns being more successful than others. Over time, variations that increase replicative success proliferate.
That’s really all evolution needs to get started. Eventually the random variations plus selective pressure from the environment make the patterns increasingly sensitive to conditions in the environment that might affect their continued existence and replication. These patterns store information (more complexity) which effects their activity. Minimal agency has arrived.
LikeLike
As I said, I acknowledge these interesting efforts, and I wish them luck. Thanks for the lesson, and thanks for not starting in on the Friston free energy principle. As the discussion moves from one ground to another, the same problem keeps coming up: it’s all just chemistry, isn’t it? At some point the accrual of information with maximum efficiency seems to become a project something cares about, instead of a blind process devolving from entropic principles and carrying on in its own relentless, automatic, unconscious way. It may be an amazing process, but as the saying goes, where’s the beef?
LikeLiked by 1 person
Frank is just arguing for more complexity theory less reductionism.
LikeLiked by 1 person
He wrote a book called “The Blind Spot”, which I’ll admit I haven’t read, but in many interviews and write ups, he makes clear the missing ingredient, the “blind spot,” is subjective experience. What’s less clear is exactly what scientists are supposed to do about it. It seems like he’s arguing for more than just complexity theory.
LikeLike
I am referring to the specific article in the link, not arguments he has made elsewhere, although they are clearly related.
The article is about explaining life in terms of physics and he is suggesting complexity theory as possible avenue for physics of life.
LikeLiked by 1 person
The wording “reduction of consciousness to something else, typically matter” can make the process unnecessarily complex.
Reduction of consciousness to life looks, I feel, as more pertinent as a subject of study. The positioning of the reduction of life to matter is another subject. The two can be addressed separately. (I feel we can agree that evolution on earth went through the steps: matter, life, human consciousness).
Would you agree on such a simplification for the concept of reduction applied to human consciousness?
LikeLiked by 1 person
I do agree. Every problem is easier if we break it into manageable chunks. And in the case of consciousness, we have many layers we can lean on: psychology, cognitive neuroscience, neurobiology, biology, chemistry, and then finally physics. So the real question is, how do we reduce consciousness to psychology and cognitive neuroscience?
Of course, if your notion of “consciousness” is something fundamentally separate from psychology, then it doesn’t help much. But if we just want to reduce self monitoring, episodic memory, deliberation, working memory, and attention, then I think we’re in business.
LikeLike
If this was addressed to me, I’m not sure we can all agree on the sequence “matter, life, consciousness.” Some now argue that before matter there is structure and relationship. But if these things are called upon to account for matter, even more so are they called upon to account for life. This has become something of a theme in modern biology.
That consciousness should come last, after life, no longer follows from the rethinking of the world as fundamentally structural and relational. The entities that make up relations could easily involve a primitive mutual sensitivity, which could explain where consciousness comes from better than the old matter-based acccount.
LikeLike
Reduction of consciousness to life is the best approach. However, the evolution of life has to be framed within a reducible ontology as well, one that does not draw an arbitrary line of demarcation that cannot be justified.
In an ontology that posits life as universal, we are essentially dealing with host evolution. In this context, one should also consider that not only is the evolution of new life forms unpredictable, but we don’t even know what to look for.
Is it possible that nature has thrown us a curve ball, a completely novel, unique and unexpected form a life that we did not expect, let alone even consider? Is it possible that this uniquely novel life form is the only form of life that is conscious?
LikeLike
The evolution of life up to human consciousness can be addressed starting with the last common ancestor we share with chimpanzee (about 8my ago). See ISPSM 2023 https://philpapers.org/rec/CHRSCR-3
LikeLike
As a tool maker myself and a beneficiary of the craft, I am a reductionist as well. However, neither the art of making tools nor the modality of reductionism tells me anything about the “true nature of reality”. The only thing reductionism tells me is that the art of making tools is mechanical, deterministic and causally closed.
My question is this: since all of our scientific knowledge is derived from inventing and making tools, should we therefore infer that reality itself is mechanical, deterministic and causally closed?
LikeLiked by 1 person
Some of Galileo’s critics charged that since his discoveries happened through the telescope, and there was no way to get close enough to validate what the telescope was showing him, they shouldn’t count. Of course, telescopes could be pointed at things on Earth, and the resulting observations validated by getting closer to what was being observed. But there was no way to be sure what worked on Earth would work for the heavens.
Today there remains the criticism that science tells us what things do, of their structures and relations, but not what they are. Of course, this assumes there is something about what they are outside of their structure and relations. The problem is how we find find out about things beyond structure and relations?
In the end, it seems like we have to make assumptions. The question is which assumptions are more fruitful?
LikeLike
Again, it depends on what one means by “fruitful”. Does fruitful mean the ability to make more accurate predictions or does it mean that our vision of reality aligns more closely with the actual landscape?
LikeLike
To the extent accurate prediction diverges from alignment with the landscape, how can we know? Granted, the scope of accuracy for those predictions could turn out to be less than we currently think, but until we have data alerting us to those limits, how could we know?
LikeLike
You appear to be leveraging our ability to make accurate predictions as the arbiter of whether our vision of reality matches the actual landscape. Is this correct or am I missing something?
LikeLike
I am. But open to learning about any alternate ways to do that arbitration.
LikeLike
Reductio ad absurdum is a modality of reductionism is it not? Reduction to absurdity is the most efficient and the most reliable way that I know of to do that arbitration for one simple reason. It is much easier to determine what something “is not” than it is to determine what it actually is.
Once reductio ad absurdum empties the bank account of dubious ideas, then it’s time to start making and testing reasonable assumptions to see if they stand up under the scrutiny of reductio ad absurdum.
LikeLiked by 1 person
When it’s used to identify a logical contradiction, I think reductio ad absurdum can be useful. Although too often it’s used just to appeal to a subjective absurdity. And it’s not always easy to tell the difference. Einstein and Schrodinger, when they identified quantum entanglement, meant it to be a RAA, but it turned out that reality really does work that way, and three physicists won the Nobel a few years ago for experiments demonstrating it.
But in general it’s a good point that logical deduction is a valid way to adjudicate. Although if you think about it, it can be seen as an alternate way to test predictions / fit the data, when physical experiments aren’t possible or practical. The solution to the black hole paradox, if it’s ever found, will have to be a mathematical deduction, a logical test rather than an empirical one. (Of course, the paradox itself was only discovered mathematically.)
LikeLike
We can almost always find component parts of any object. The question always is whether the decomposed parts of a larger explain by themselves the larger object. And, in fact, they almost never can because objects in the natural world are products of chaotic processes. That is, they are determined by their component parts but they can’t be predicted by those parts. This contrasts with the mechanical.
LikeLiked by 1 person
It seems like chaos can be tracked and quantified. Isn’t that the main goal of chaos theory? And while the chaos prevents us from making precise predictions, it doesn’t really stop us from understanding the systems in question. For example, we can’t precisely predict the weather, but that doesn’t mean we don’t understand a great deal about it, an understanding that still allows for useful probability estimates.
LikeLiked by 1 person
The only way we understand weather is by applying our understanding of the higher-level order that arises in atmospheric chaos. We don’t do it by reducing the atmosphere to its components, even if in theory, that might be possible.
Causally I can agree the bigger arises from the smaller, and if that is sufficient to constitute a complete explanation, then I would count myself as a reductionist. Whether it is or isn’t sufficient, it leaves out the interactions and temporal aspect of the natural world.
LikeLiked by 1 person
Right. There are different levels of certitude for reduction. One is just the assumption that everything does in fact reduce to a small number of building blocks. A stronger one for any particular theory is empirical data that implies one theory reduces to another. Honestly, this is all we have in many cases.
The strongest is a priori or logical reduction. This one requires bridging laws, so that we can logically map the constituents of the reduced theory to the reducing one. I don’t know enough about meteorology to know if that’s happened with weather. Even if it has, I’m sure it would only be for some aspects and not others.
LikeLiked by 1 person
This paper claims to bridge from particle to fluid.
Hilbert’s sixth problem: derivation of fluid equations via Boltzmann’s kinetic theory
Explained in Scientific American:
https://www.scientificamerican.com/article/lofty-math-problem-called-hilberts-sixth-closer-to-being-solved/
Critique here:
https://arxiv.org/abs/2504.06297
Make sure you check the critique.
LikeLiked by 1 person
if qualities are just neural categorizing habits, do you think there’s any level where the explanatory ladder actually ends, or is it more like an infinite ikea bookshelf where we keep discovering new screws and half-finished instruction pages?
LikeLiked by 1 person
Remote Reply
Original Comment URL
Your Profile
There may well be a fundamental foundation to reality. The problem is I don’t know if we can ever be confident we’ve reached it. A candidate might be something like propertyless points anchoring relations. Even then, we could never be sure it wasn’t a limitation of our current view or paradigm.
LikeLike
I think the point Mr Owens is trying to make, and he can correct me if I’m wrong is that reductionism glosses over the “hard problem of life”. And this approach is easy enough to explain because the physical sciences are not in the business of realism, they are only interested in utility.
Nevertheless, if individuals like ourselves want to get serious about an ontology, the “hard problem of life” has to be addressed. Now, whether you are personally interested is addressing the “hard problem of life” is another question.
LikeLiked by 3 people
For me, the hard problem of life is similar to the hard problem of consciousness. It doesn’t exist as a discrete problem, but just as all the “easy” problems combined. And the “easy” problems are just the “hard” problem broken into manageable chunks. And definitely, this is the reductive approach.
LikeLike
Which “easy” or “hard” problems would they be?
LikeLiked by 1 person
In the case of life, the ones that immediately come to mind would be: replication, homeostasis, allostasis, metabolism, the functioning of proteins, DNA, RNA, ribosomes, molecular chemistry, etc. And of course the initial origins of life are a puzzle, but I wouldn’t put it in the same category as the one the hard problem of consciousness is supposed to be in.
LikeLiked by 1 person
Is it reasonably possible to draw a definitive line of demarcation between life and non-life that is not arbitrary?
Since organic life emerged from naturally occurring self replicating chemicals, at what stage of this process can we assert with any authority that these chemical compounds are now alive? This dilemma is the “hard problem” of life for the physical sciences, a problem to which the sciences cannot reach a consensus.
The hard fact is this: all of the boundaries that the physical sciences attempt to draw between life and non-life are arbitrary and cannot be justified for two simple reasons.
First, the current justifications are “constituted” by the observation of high-level properties, which themselves are contingent on a certain level of perception, a perception which leaves some things out and leaves others in.
Second, there are indefinite levels of perception. So, from whichever one science uses to describe life, science will necessarily neglect other aspects. The reason science’s selection of one or another level of perception is used for a description is fundamentally pragmatic, and not ontological; confirming once again that the physical sciences are in the business of utility, not realism.
LikeLiked by 1 person
It seems like most biologists are happy to leave the demarcation of life to the philosophy of biology. Although astrobiologists have an interest, for obvious reasons. Myself, I don’t think there’s a fact of the matter whether viruses are alive, but they remain securely in the purview of biology. Viroids and prions have far less of a claim, but still are recognized as biological.
The demarcation of species is another issue I think philosophers of biology focus on more than actual biologists.
Eventually if it becomes important for everyone to be consistent in their terminology, we might see a moment like the IAU issuing a fairly arbitrary definition of “planet.” If so, I’d be surprised if isn’t just as contentious. Although the controversy does have the benefit of calling people’s attention to just how hazy these concepts are.
LikeLiked by 1 person
But how about you Mike; are you a biologist or an armchair philosopher like the rest of us? Dismissing “the hard problem of life” as inconsequential to a reducible ontology that will stand up under the scrutiny of analysis isn’t philosophy.
LikeLike
Lee, I have a hard time caring too much about definitions. The main thing is to understand how the systems in question work. Whether we choose to call viruses “life”, “replicators” or something else is, to me, just a convention. I actually think the main thing is to understand just how fuzzy the boundary is. Again, it’s sort of like the debate on whether Pluto is a planet. As long as there’s some rational to the labeling, I’m okay with it.
I really only get worked up about it when people refuse to clarify which definition they’re using, or use one that’s so far from conventional understanding that it amounts to misinformation.
I do realize this is something that sets me apart from most philosophers (professional or amateur).
LikeLike
Your vision of reality is mechanistic and I get that; not saying I understand the rationale, but I get it. There are many individuals who share your perspective. I’ve had prolonged conversations with a radicalized materialist on Suzi Travis’ substack. Suzi herself is more pragmatic and she even took it upon herself to moderate our discussions which helped cool the temperament of my protagonist.
Based on my limited experience with hard core materialists, it is my belief that the fallout from post modernity is that individuals are becoming radicalized by scientific extremism as the institution struggles to reestablish its status as the proxy savior, a position that it held during modernism.
With that, I’ll quit badgering you……
LikeLike
On rationale, my personal experience and impression from reading history is that it’s the mechanist explanations, in the sense of a causal or structural account with minimal gaps, that work, and that best stand the test of time. So it’s my bias for what a successful theory will have.
But I try not to be rigid about it. I’ll accept a pragmatic explanation with structural gaps if it works, but as a placeholder until a better explanation becomes available. Of course, in the end all theories are placeholders.
Maybe someday I’ll have to revise that view.
LikeLike
I think the biggest problem for any radical materialist is that there is only one thing they can contrast a mechanistic paradigm against, and that is idealism with it nonsensical “consciousness is everything” ideology. In all seriousness, that is a real problem.
We’ve lived with the mind/matter dichotomy for so long that “we can’t even imagine” that there might be an alternate paradigm; a literal third option. Whenever I attempt to bring this subject up, without even attempting to explain what that paradigm might include or how it might work, all I get is an incredulous stare; go figure.
I hope a challenged you in some small way without being too offensive.
LikeLiked by 1 person
I wonder if I count as a radical materialist. I am what Chalmers calls a Type-A Materialist, seeing no epistemic gap between the mental and physical, as opposed to Type-Bs who do see a gap but not an ontic one. It does seem like these views can be compared to idealism, but also dualism and panpsychism in all their different flavors.
On a third way, there is neutral monism, which seems to map to a type of panprotopsychism in contemporary terminology. Don’t know if that’s what you had in mind.
I love being challenged in a thoughtful way. Thanks! Hope I wasn’t too offensive either.
LikeLike
A radical materialist would be anyone who is convinced by scientific extremism that reality is mechanical, deterministic and causally closed.
Nope….. it ain’t neutral monism. And all of the other varieties you listed are sub-sets derived from the causally closed mind/matter dichotomy. My ontology breaks the causally closed cycle imposed by those restraints and charts a different path.
LikeLiked by 1 person
More than one scientist has been quite interested in the topic.
https://en.wikipedia.org/wiki/What_Is_Life%3F
LikeLiked by 1 person
Sure, although I think Schrodinger was more than interested in just the definition of the word “life.” He wanted to know how it worked. And reportedly he did inspire the search for DNA.
LikeLiked by 1 person
The definition of life and how it works aren’t different subjects. What life does and how it does it is what distinguishes life from non-life definitionally.
Schrodinger suggested that genetic information might be stored in a aperiodic crystal, which is more or less what DNA is.
LikeLiked by 1 person
“Since organic life emerged from naturally occurring self replicating chemicals, at what stage of this process can we assert with any authority that these chemical compounds are now alive?”
I don’t really see this as a deal breaker. Water at 0 C transitions from liquid to solid, at what point do we call it ice and not water? At exactly 0 degrees, it is both melting and freezing. It can manifest properties of ice and water during the transition. Very pure water can remain liquid in a supercooled state below 0 C but then freeze instantly with a slight disturbance.
Boundaries in the natural world are fuzzy. I think understanding the transitions, using complexity theory, is what Adam Frank is talking about in the article referenced.
LikeLike
Good to hear from you Jim, it’s been a while.
From my limited understand of complexity theory, according to AI’s summary, it appears to rely heavily on the concept of agents as fundamental building blocks. If that’s the case, I can get on board. I believe the core idea of agents is what Mr. Owens was emphasizing as well.
LikeLiked by 1 person
Somewhere I think I said that life is a bigger mystery and one we are less close to solving than consciousness.
LikeLiked by 1 person
“I think the point Mr Owens is trying to make, and he can correct me if I’m wrong. . .”
You’re not wrong, although “glossing over” is a gentler way of putting it than I might choose. At this point “cognitive dissonance” comes to mind .
LikeLiked by 1 person
Cognitive dissonance has to be a foundational premise for scientific extremism because of all the contradictions in their ontology. I think what separates a radicalized materialist from a pragmatic materialist is the ability to think for oneself; same goes for religion.
Be it religious and scientific extremism, both promote radicalization, and I think this is a symptom of post modernity.
LikeLiked by 1 person
I’m a reductionist myself, but an egalitarian reductionist, when it comes to ontological reduction. That is, I think ontological reduction simply relates categories to one another, for example as part and whole. It doesn’t introduce fundamentality and derivative-ness – those apply to our epistemic grasp of the objects, properties, or processes in question. The map has a starting point and a direction – but this doesn’t automatically transfer to the territory. I would also like to note that on some reductive explanations, such as Mad Dog Everettianism ( https://arxiv.org/pdf/1801.08132 ), the starting point for understanding the world is the whole universe and its laws.
LikeLiked by 1 person
I do think there’s a lot of overlap in theory reconciliation and reduction. And successful theories that are superseded could be said to “reduce” to the replacement theory, although I don’t usually see it put that way. But I can see where you’re coming from on the egalitarian point.
LikeLiked by 1 person