I'm having a lovely time here at Beyond Belief 2 — you should all be here (and of course, you will be; as they did last year, everything will be available after the meeting on the web.) It's an eclectic mix of all kinds of interesting stuff outside of my usual range: yesterday, we had terrific sessions on the history of the Enlightenment, evolutionary economics, evolution of religion, and some speculation and cosmology. It was vastly entertaining, and lest you think this was a bunch of thugly atheists preaching to the choir, let me reassure you that I disagreed with about a third of what I heard.
Case in point: David Sloan Wilson. The first part of his talk I found to be a wonderful discussion of the importance of multiple levels of selection, and I think he made a persuasive case for group selection (but then, I'm already partial to the idea anyway). Then at the end he made his arguments for the scientific evaluation of religion, and I loved it, but not because I agreed with him — but because he was so vigorously confrontational, strongly chewing out both Dennett and Harris (who were right there in the audience), and taking a few swings at Dawkins (who is not here, so Wilson abstained from going on at length). That's what I like to see at these meetings, an unapologetic airing of grievances that will prompt uncompromising responses. Fabulous! It was like getting a ringside seat at a prize fight!
Still, I think Wilson was wrong. His gripe with Dawkins was explained on eSkeptic, and Dawkins has replied for himself, but Wilson was unsatisfied. He put up a list of interesting questions about religion, with the first being whether god exists, and complained that Dawkins has unapologetically stated that his book was only about the first question…and Wilson then states that Dawkins was practicing bad science. These guys who get so close to religion…it's like he's going to unconsciously overlook the fact that the majority of people on the planet are practicing really bad science, because they don't recognize that their very first premise, that a god exists, is false. To complain that Dawkins only spends a few paragraphs on group selection, which is not a subject that he favors anyway and which was not the subject of his book, while blithely overlooking the fact that his book does address the biggest question of all in his list with an answer that the public to a large degree rejects, is missing the forest for the trees.
Oh, and Dennett sat in a panel with Wilson right after. They mixed it up a little bit (more would have been better), but you'll have to wait for the videos to watch it.
Another flawed highlight of the afternoon was Stuart Kauffman, who gave us a preview of his upcoming book on recovering the sacred. I've been a big fan of Kauffman's work for years — he always has big and radical ideas about the origins of life. In this case, he was making an argument against reductionism: that we have this idea that in principle, that all of biology and all of the mind can be reduced to physics. He argues that while it is true that nothing in biology is going to violate the laws of physics, the range of possible configurations of a single protein is so vast, far greater than the number of particles in the universe, that it is effectively incalcuable. It's an argument that emergence is a real phenomenon that shouldn't be discounted, but also that it is an entirely natural process, an expected consequence of the way the universe works. I liked it, and I'm ordering his book, but then he got a little soft at the end. He wants to "reinvent the sacred as the creativity in the natural universe", and in the interest of finding common ground with diverse communities, he proposes that we call this Spinozoid concept "God".
I really don't think anyone will be fooled. As Dennett mentioned earlier in the day, the religious have a mistrust of the hypocrisy of scientists who claim to be respectful of religious belief yet do not share it themselves, and I think this would be a prime example.
Oh, and before I scurry back to the meetings, I'll mention that the physicist, Sean Carroll, is here. He gave an excellent talk on cosmology and suggested that it's entirely possible that the universe is eternal — there was a Big Bang, of course, but that that might have been an event in an eternal universe, which has some interesting implications for those who are fond of using the Big Bang in their theological speculations.
Sean briefly addressed Kauffman's anti-reductionist proposal, in a way that I find incredibly unsatisfying. These mathematicians and physicists, some days you have to wonder about them. His reply was to propose that if you knew the vector of every single particle in a glass of water, you'd have complete knowledge of what was there; and that if you had an infinitely fast computer and complete knowledge of the initial state of everything programmed into it, you could compute every subsequent state. That doesn't actually address Kauffman's more pragmatic statement that the mathematical possibilities are so immense that a real computer or brain cannot compute them.
It seemed a little ironic that Kauffman was proposing that the complex results of natural processes ought to be called "god", while Sean was rebutting that by postulating a computer with such immense and supernatural powers that it ought to be called "god"…but of course, he wasn't going to call it that.
- Log in to post comments
Not having read or seen any of David Sloan Wilson's comments, one of my biggest complaints about TGD was the scant and unsupported treatment of the evolution of religion. His other books are so excellent at conveying the state of recent research in evolution in a rigorous, but accessible way, that I was surprised to find the section on evolution to be the weakest. I sorta found the book boring because I wasn't learning any new science. It appears that there was a great deal of anger and cynicism (rightly so) present in professor Dawkins when he wrote the book, which biased his normally cool, objective look at both sides of the issue and his role as an educator. I could understand why somebody who seriously took a look at the positive and negative cultural implications of religion and the potential evolutionary advantages/disadvantages it imparts in their research would be peeved at Richard for trampling all over everything. Perhaps he was trying to reach beyond the audience of science enthusiasts and present a non-nuanced display of religion as the poppycock it surely is.
"...let me reassure you that I disagreed with about a third of what I heard."
Sometimes I think this would be the case if you went back and read some of your old blog posts that you'd forgotten about. And that's a good thing! You don't give a free pass to nobody for nothin'.
That's why I keep coming back.
Xopher
I haven't read anything on Kauffman's anti-reductionism, but I'm going to comment on it anyway since they all go the same way. All thought and all language is reduction. The novelty of a theoretical model is a product of reduction. Right out the gate it should be obvious that a position opposed to reductionism isn't even tenable, we cannot even advance the opinion without committing to reduction. Besides that, a good model will predict and explain the emergent phenomena, self-organization and complexity that the anti-reductionist crowd likes to use as a case against reductionism.
I don't think that the idea of a computer which will determine, precisely, every subsequent state of that glass of water (aside from taking the view that the glass of water is a computer for computing states of a glass of water) is even one that can be entertained as a thought experiment. Supposing a computer were built, it would take on a physical configuration based on what it was computing. That configuration will have an effect on its surroundings. That effect may not be signifigant in the evolution of a glass of water to any reasonable approximation, but even extremely reasonable approximations aren't precise computations.
What's more we cannot, even in principle, know everything about a glass of water to begin with. And I could be misunderstanding something here, but it seems that Carroll is flirting with an ensemble interpretation.
Reinvent the sacred as variation? Hmmm...more appeal to multiculturalism.
Glad to hear that there is an afterlife to Beyond Belief.
Yes, well, Sean Carroll is for my money a clear-headed thinker and atheist to boot. About the only thing he discusses which I personally have a hard time to see as entirely agreeable is his and Jennie Chen's time-symmetric version of cosmology.
Sure, that the universe looks low entropic when looking back can be due to a simple symmetry breaking, we happen to be here instead of in a part of the multiverse where the arrow of time goes backward. But then again, we can see symmetry breaking without having the other possibilities realized.
It is therefore with some joy I note that he has started to discuss more regular eternal inflation scenarios, a cosmology that makes my head hurt much less. :-P
There are two cultures that clash here, on several levels.
One level is that Kauffman and others, often solid state physicists, are promoting emergency as a unique phenomena and sometimes as a possible path towards a deeper quantum theory.
AFAIU theoretical physicists in general thinks this is misguided. I don't think they say that for example biology can be understood effectively on a physical level, but that emergence is observed on systems that still interacts according to simple physics. As Carroll says, the system states still have a physical model.
The other level is that physicists and for example computer scientists may use idealized machines or gedanken experiments to broaden knowledge. Carroll's Deep Thought doesn't look as much as a religious agent to me as an oracle that can decide non-computable functions.
Calling the universe "God" is about as effective in the long run as calling reason "a faith" or calling humanism "a religion" or calling feelings of awe and wonder "an appreciation of the sacred." Religious people gladly accept all the new terms and proceed to do what they always do: blur together superficially similar concepts and let the less reasonable one ride on the back of the reasonable one so that they both seem reasonable. If they're really just different versions of the same thing, it seldom weighs in favor of what was already credible.
Is prayer a petitionary request to God, or is it an inner meditation which seeks acceptance of what happens? It's either; it's or; it's both at once.
Does the Power of Positive Thinking lie in optimism, focus, and hard work, or do positive thoughts have a force which can magically change the cosmic structure of the universe? It's either; it's or; it's both at once.
Is God a disembodied, conscious, personal intelligence who created and helps us like a father, or is it the Ground of Being, Mystery, and the Actuality of the Potential? It's either; it's or; it's both at once.
Are the stories in the Bible myths and metaphors which express truths about human nature, or did they really mostly happen? It's either; it's or; it's both at once.
Religious thinking usually involves holding two opposing ideas and believing them both. Stuart Kauffman's grand plan of redefining God into being something reasonable is not going to make religion more rational and nonbelievers-in-traditional-God-only more acceptable. It will simply be blithely retranslated into woo and the stigma against "atheism" will continue being encouraged.
Actually, it seems Kauffman is putting forward an opinion that's based on pragmatic considerations, and in that case there's little to argue with.
I jumped the gun there.
PZ is right, you should all be here! It's a stimulating meeting, partly because there are smart people who occasionally disagree.
I don't understand why anyone thinks reductionism vs. non-reductionism (whatever you want to call it) is an interesting question. In principle, you could imagine accounting for everything that happened in the universe in terms of the microscopic laws of physics. In practice, you can't. And almost everyone agrees on both of those statements, yet they continue to invent an "argument" by choosing to emphasize one thing or the other.
You are correct--this is precisely what Dawkins was doing with TGD. It was not meant to present a rigorous scientific treatment of anything; anyone who read it with such an expectation completely misjudged its author's intent.
Beyond Belief was wonderful last year (except for the anethesiologist kook). The best bit for me was the exchange between Richard Dawkins and Neil deGrasse Tyson.
http://onegoodmove.org/1gm/1gmarchive/2006/11/beyond_belief.html
I've read a lot of arguments "against reductionism" from reputable scientists, and they always leave me shaking my head, asking "Um, ok, I agree with everything that was said... but what does it have to do with reductionism?"
Kauffman's argument seems to be that way too. Sure, ermergence is real, and he's probably right about the unimaginably huge number of possible protein configurations... but so what? How is this an argument against reductionism?
Either my definition of reductionism is wrong, or his is, or (more likely), reductionism is one of those words, like naturalism, that there is no consensus about what it means.
Working in the complex-systems field (I'm in the hotel lobby at ICCS 2007 right now, uploading pictures) one hears the word "reductionist" thrown around a fair amount, almost always as an insult. This bothers the heck out of me, actually, because everybody here is doing reductionist science, in the only sense of the word I can find reasonable. It's no less reductionist to say "I can predict the behavior of this system because it's a scale-free network" (or because it's on the period-doubling route to chaos, or whatever) than it is to say, "I can predict what's going to happen because it's made of protons, neutrons and electrons."
I doubt any physicist who has paid attention to statistical physics since at least the 1970s would be in the least shocked by this. Ken Wilson's formulation of effective field theories was a long time ago, people. Even the Feynman Lectures on Physics talk about laws of physics which do not "reproduce themselves at higher scales". If you think this is a revolutionary concept, then you must have slept through freshman physics in 1963.
A proper answer to Kauffman, or at least a better one than Carroll gave (on the spur of the moment, I expect), is that we implement Kauffman's programme every time we teach how to derive thermodynamics from statistical physics. We do not consider the position and momentum of every atom in a gas to predict its behavior; instead, we look at the emergent phenomena of temperature, pressure and volume. Douglas Hofstadter points out that trying to build a simulation which tracks every particle is actually a step backward, since pressure isn't a property of any individual atom, but the collective effect of many atoms, bashing themselves into the walls (and it doesn't even matter which particular atoms those are).
John Armstrong has suggested that one could interpret this in the framework of category theory. Emergent behavior, like the laws of thermodynamics, would be the decategorification of the micro-scale physics. I don't know if anybody has tried to formalize this or wring any useful ideas out of it, but it's interesting to think about.
I have a theist friend who insists on using "reductionism" in his arguments against atheism--"If there is no god, then we're all just molecules interacting with other molecules, and someone torturing a puppy is just one set of molecules acting on another, and you have no basis for deciding that's bad."
I never really know what to say to that because it's such an absurd "reduction" of the issue, because no one, no one, looks at life like that. No one looks at anything like that. I don't know what's up with the torturing a puppy thing, either--those theists can have the most sadistic imaginations.
How could you do that even in principle? Do you know for sure what all the laws are?
Here are the remarks by John Armstrong I mentioned above. Curse this one-link-per-comment limit.
My first guess at implementing his idea would be to consider the phase space of a classical particle (technically, this would be some manifold with a symplectic geometry, I suppose, but I don't think that matters right away). Construct a category by taking the points in this space as objects, and use for morphisms the time evolution of one point to another. A morphism exists from A to B if the system can progress from A to B under Hamilton's equations of motion. This morphism would be an isomorphism if the system can progress from B back to A. In a simple system like a pendulum — with nice, closed phase-space trajectories — all the points along a trajectory would then fall into an isomorphism class. Decategorify, and bam! You've got yourself the macrostates of the microcanonical ensemble.
Or, well, maybe. I'm basically making this up as I go along.
...if you knew the vector of every single particle in a glass of water, you'd have complete knowledge of what was there; and that if you had an infinitely fast computer and complete knowledge of the initial state of everything programmed into it, you could compute every subsequent state.
That sounds a lot like the machine Frank Tipler is building in his basement.
We've known since the 1970's that the conditions under which group selection can counter individual selection ("deme sizes of less than 25... and migration, could not be too much greater than five percent per generation" Levin BR, Kilmer WL. 1974. Interdemic selection and the evolution of altruism: A computer simulation study. Evolution 28: 527-545) almost never occur. Kin selection theory generates lots of experimentally verified predictions every year; group selection none. But people want to believe in it because it makes them feel good. Sound familiar? Group selectionism is a religion.
Sean Carroll (#8):
You'd think so, but you do find people denying the first claim -- in academic philosophical circles at least. Dualism (i.e., property dualism, of the sort advocated by David Chalmers) seems to be getting an increasingly large following. They're wrong, of course, but it isn't easy to explain where their arguments go astray.
You're certainly right to say that there's a good deal of confusion, especially among scientists, about how to understand "reduction." And I think you're also right that much of this confusion can be resolved by distinguishing what can be "in principle" and what limitations we face "in practice."
Sean, how do you justify the statement that we could, in principle, account for everything that happens in the universe via basic Physics laws? Doesn't Heisenberg uncertainty tell us that we cannot do this even in principle?
The God Delusion was not a book intended for other scientists, but for the general public. Dawkins had a point to make: that there is no evidence for, and considerable evidence against, the existence of the kind of god to whom most people pray. He chose, wisely in my opinion, to make this case using the most well-established scientific knowledge available: physics, astronomy, biology. The fact of evolution and the fact of the Big Bang are far better supported than the propositions we have concerning the origins of religion. In that area, we have a paucity of hard data and a profusion of overlapping hypotheses.
Making one's case with the most solid science possible is not a crime. Wilson wishes Dawkins had written a different book, but the one which Dawkins chose to write is the one the world needed.
Exactly.
I wanna hear more! :-)
WTF? I'm certainly no physicist, but if my layman's understanding of the uncertainty principle is correct then it's physically impossible to precisely know both the position and acceleration of a particle. So it's impossible to precisely know the vector of any particle in that glass let alone all of them.
So even if we did have a magical, all-powerful computer to run the calculations wouldn't we need some kind of magical, all-seeing sensor to get the data for the calculations in the first place?
Nevermind. I didn't see Sean's #8 comment when I posted that. I'm more clear on what he was getting at now.
I think the reductionism issue is really over supernaturalism. Natural viewpoint are distinguished from religious ones by the basic direction in which explanations go: are they cranes, or are they skyhooks?
Did we start out with morals and meanings and minds? Are they forces which act down on matter and energy? Or did those things evolve, the result of what Michael Shermer called "a stepwise, bottom-up natural self-organized complexity out of simplicity."
Sure, there are various ways of explaining or defining "reductionism," and there are quarrels in science on explanations for the relationship between what goes on at the small level and what goes on at the big level. They argue at conferences and in papers.
But the only argument against reductionism which really matters is the one that says that there is more to reality than matter in patterns of motion, because Pure Mental Being and the Life-and-Love Force are not physical.
@Donnie B. & SeanH
What Sean Carroll presumably has in mind here is that if we knew the complete quantum state of the system -- even of the whole universe -- (where this state would be represented by a vector in an infinite dimensional complex Hilbert space), and we knew the Hamiltonian that governed the system (i.e., the law of time evolution), then this would tell us all there is to know about the universe.
The uncertainty principle (as standardly understood) places limits on our ability to ascribe classical properties like position and momentum, but the evolution of the quantum state (which gives the probabilities of measurement outcomes) is completely deterministic (though there is a question of whether it can be empirically discovered if you don't have an ensemble of identical systems).
We do, however, still face the challenge of the so-called quantum measurement problem when trying to interpret the significance of the quantum state -- especially when we're trying to talk about the entire universe as our system of interest. From what I've seen, cosmologists typically like to adopt a many-worlds interpretation of quantum theory, in which all possibilities encoded in the quantum state are actualized in different "worlds."
Sean H - no, you were right to object. To say that in principle, we can account for everything in the universe is hubris. Science is an approximation of reality, not reality itself. The universe doesn't obey our laws. Our "laws" are consistent patterns of behaviour that we observe - they are what we can say about the universe, not what it actually "is" (Bohr). Sure, we can simulate aspects of the universe using our laws to arbitrary levels of accuracy, but we can't re-create it 100%.
@ jeff (#24)
a) I'd like the reference to that Bohr quotation if you have it handy. Bohr certainly did think that quantum mechanics required a revision in epistemology, and an abandoning of the classical view of what can be known. But I'd say he also insisted that quantum theory is complete and accurate (for the domain of phenomena at the atomic, molecular, chemical, etc. scales).
b) The relevant point for reduction vs. anti-reduction is whether we have a good account of what the limits of our physical theories are. Here the relevant point is that (e.g.) quantum electrodynamics (together with classical gravity) is going to govern all biological, psychological, sociological, etc. processes. Thus in principle we should think that we can account for all these processes with known physical laws. However, this is obviously far from possible in practice.
This doesn't require us to say that science is reality, or even that we know all there is to know about the world (obviously we don't). But we do know enough to say that everything we see around us is physical (i.e., that it obeys physical laws), and in that sense everything is reducible to physics. But obviously we could never understand everything that goes on in the world using only Q.E.D.
Reductionists claim the special sciences are reducible in principle; anti-reductionists deny this. The current vogue in philosophy, for example, is to claim that the special sciences, psychology in particular, needn't reduce to the fundamental sciences even in principle. I think the problem is that scientists like Kauffman see this and try to reinterpret it into something scientifically meaningful. Reductionism must be making a stronger, practical claim; anti-reductionism must be a response to this; and thus announce themselves as anti-reductionists and "holists" when they're actually just straightforward reductionists.
The same is true of emergence. Emergentism claims that novel non-reducible laws govern "higher" processes; there's basically "non-commuting" descriptions of the same physical situation that aren't causally connected to one another in anyway. The language of "emergence" was adopted by scientists but really they're talking about things typical in physics. The talk of "anti-reductionism" and "emergence" is a rhetorical flourish designed to gain attention.
Yeah, I made the same mistake.
If anyone is seriously claiming this I will oppose them out sheer laziness... I refuse to go back to undergrad science just so my advanced degrees in social sciences are worth something. I didn't even hold onto my old labcoat.
"Reduction" in itself is an essential element of most experimental design, the 'divide-and-conquer' strategy that typically pays the most dividends. I have no interest in any character who rules out 'reduction' a priori on this sense, since they are inevitably a purveyor of pseudoscience, often of the quantum variety.
At the same time, "reductionism" and "holism" as rigorous philosophies of biology are, as I understand it, really kind of useless in that each makes claims that are either not testable or which are incoherent.
However, if "reductionism" and "holism" are thought of rather broadly as different points on a spectrum of ways of thinking about problems, then both can be useful---but the former is of greater utility in experiment, and the latter in hypothesis formation.
Darwin's personal history provides a good illustration of how both strategies can operate on different levels simultaneously in the same person. Darwin was firmly committed to the notion of living things obeying physical law like the rest of the natural world, and this idea (as Eldridge points out) is well-developed in his thought early in his career. At the same time, however, Darwin thought nothing ill of looking for the connections, for the large-scale, emergent properties of systems that he believed to be the product of evolution.
Looking forward to these videos. Enjoyed it thoroughly last year.
"There is no quantum world. There is only an abstract physical description. It is wrong to think that the task of physics is to find out how nature is. Physics concerns what we can say about nature..."
-- As quoted in "The philosophy of Niels Bohr" by Aage Petersen, in the Bulletin of the Atomic Scientists Vol. 19, No. 7 (September 1963); The Genius of Science: A Portrait Gallery (2000) by Abraham Pais, p. 24, and Niels Bohr: Reflections on Subject and Object (2001) by Paul. McEvoy, p. 291
-------
All this talk of reductionism and reductionism can be interesting. It seems to me that whenever you are explaining something in terms of something else, you are reducing it. Whether the reduction is valid or not depends on whether or not something is lost in the explanation.
And of course, you have the problem of when to stop reducing (like a child asking "why", until you're exasperated). There are (3) options: 1) reduce until you reach some atomic fact or primitive that can't be reduced any further, in which case you have something that can't be explained, or 2) reduce forever and ever amen - I'd like to think the universe isn't structured that way, or 3) reduce in a circular or graph-like pattern - doesn't seem applicable to physics, but who knows?
LOL! I wouldn't go quite that far, but I've been trying to get Blake and the other group selection enthusiasts to comment on this article. Let's see if I get a bite this time.
More contentious but also amusing is this exchange between DSW, Sober and Trivers.
Steven Weinberg has a good essay on reductionism ("Reductionism Redux"), in which he distinguishes between "grand" reductionism and "petty" reductionism. That split is nothing new (it's what Sean, and most everyone else, does), but his discussion about it is nice.
Nobody advocates writing down the Hamiltonian of a group of deer to study their behavior. But anyone sensible believes that emergent phenomena really do emerge, from the usual physical laws. (And I don't think quantum uncertainty has much to do with life in our macroscopic world. That's pretty deterministic.)
Yeah, reductionism, I think Dennett has is right. Reductionism is Ok as long as you don't get too greedy, and if you go the other way that is OK as long as you only see cranes and never sky-hooks.
Re: Number 6, that was eloquently stated. Theists who bother to argue about their beliefs seem to use the Fallacy of Equivocation almost exclusively. This dates even back to Anselm, who basically equated the word or concept with the physical thing and called it a proof.
I cant wait to see these, but I havent even finished watching last year's conference!
@ Jeff (#32 -- though comment numbers are apparently subject to change):
Thanks for the reference; I should have realized that it was the Petersen quotation. I'll just point out that this is a paraphrase of a statement Bohr is claimed to have made, and it goes well beyond anything he says in his many (well-considered) published writings on the subject -- which is presumably why it is so widely quoted. I'd say that his considered opinion does lie in this general direction, but his views aren't as extreme as this second-hand paraphrase implies.
On reduction:
It's helpful to ask what it is that supposed to be reduced. I argue for a form of reductive physicalism (hence the handle) -- which for me means the claim that everything obeys the laws of physics. But I don't think that this implies reduction of all explanations, because I see explanations as supplying information that enables understanding -- and an account in terms of quantum physics obviously does little if anything to allow us to understand (e.g.) the evolution of species. Likewise, I wouldn't claim that we could in general derive higher level theories from lower level ones -- so I don't insist on theoretical reduction.
For this reason (@ John B, #29), there's no reason to think that the social sciences require physics for their validation (i.e., to be "worth something"). Nonetheless, we do live in a physical world, and if social facts are real facts (i.e., if they do anything) then they are ultimately facts about extremely complex physical systems -- obeying physical laws.
Interesting. I would at a hasty guess think that you capture macrostates in ergodic systems that way. But I don't see how dissipative, fundamentally time-asymmetric processes and possibly chaos would not deviate from such isomorphism behavior.
Re comment #24:
IIRC Bell test experiments have verified that there are no local hidden variables to ~ 15 sigma. And I believe I read a press release on a new experiment that lie more in the line of the original EPR suggestion, presumably with fewer potential loopholes, that pushed it to ridiculous 23 sigma or so. It must be one of the best verified "no-go" result or result of any kind we have.
For my money, there is definitely no more "more" there. But again, identifiable variables is the principle, and knowing the actual states is the practice. Let's see if we can decide whether particles are particles or strings first.
I don't know, it's not hard to find wild quotes from these quantum guys - some are well sourced. For example, from Bohr:
"For a parallel to the lesson of atomic theory regarding the limited applicability of such customary idealisations, we must in fact turn to quite other branches of science, such as psychology, or even to that kind of epistemological problems with which already thinkers like Buddha and Lao Tzu have been confronted, when trying to harmonize our position as spectators and actors in the great drama of existence."
- Niels Bohr, Speech on quantum theory at Celebrazione del Secondo Centenario della Nascita di Luigi Galvani, Bologna, Italy (October 1937)
And from Heisenberg:
"Any concepts or words which have been formed in the past through the interplay between the world and ourselves are not really sharply defined with respect to their meaning: that is to say, we do not know exactly how far they will help us in finding our way in the world. We often know that they can be applied to a wide range of inner or outer experience, but we practically never know precisely the limits of their applicability. This is true even of the simplest and most general concepts like "existence" and "space and time". Therefore, it will never be possible by pure reason to arrive at some absolute truth."
-Physics and Philosophy: The Revolution in Modern Science (1958) Lectures delivered at University of St. Andrews, Scotland, Winter 1955-56
The following quotes are widely attributed to Heisenberg in many science articles, but their actual sources elude me (at least in a quick web search):
"The hope that new experiments will lead us back to objective events in time and space is about as well-founded as the hope of discovering the end of the world in the unexplored regions of the Antarctic."
"Some physicists would prefer to come back to the idea of an objective real world whose smallest parts exist objectively in the same sense as stones or trees exist independently of whether we observe them. This however is impossible."
"Materialism rested on the illusion that the direct "actuality" of the world around us can be extrapolated into the atomic range. This extrapolation, however, is not possible - atoms are not things. [emphasis added] "
-Werner Heisenberg
And from John Wheeler:
"I do take 100 percent seriously the idea that the world is a figment of the imagination," -remarked to physicist writer Jeremy Bernstein ("It from Bit")
Sounds to me like Kauffman has a good point, but didn't quite make it as clearly as one could wish (with no disrespect, typical of him).
One could take the typical 'reductionist' POV and correctly note that the dynamics of a complex system are merely the product of individual parts interacting, and that with sufficient knowledge of the state and computational power the overall behaviour can be modeled....
However, since knowing that enough of the state is generally infeasable, and the computation appears to be intractable generally, treating the 'emergent' phenomena as an operational entity itself appears to be in order. It is at least plausible that predictive models at that level may be tractable.
Makes sense to me. Though the reductionist approach analyzing the individual components and their interactions seems also seems like a reasonable way to discover the parameters/variables which the emergent behaviour is dependent upon.
"Anti-reductionist" comments crop up from time to time in physics talks, but I never know what position they think they are objecting to, or how emergent behaviour is anything more profound than the fact that unexpected phenomena can arise in complex systems because we can't compute them exactly.
It is not the position of anyone in science that, e.g., computing the exact trajectories and collisions of a mole of particles is a smart way to analyze a gas. Reductionism to me is more of a requirement of logical consistency. If a gas is a collection of constituent particles, then the behaviour of the gas can't be inconsistent with the behaviour of the individual particles. Hence, in principle , to exactly compute the motion of all the constituents, based on the rules governing those constituents, is to completlely determine the behaviour of the gas. There can be no "new" rules for the gas which aren't also expressed as rules for the particles.
It's bleedingly obvious that this isn't a realistic program for working with complex ensembles. One, we have finite computing power with finite precision; two, we have finite accuracy in knowing the total configuration of particles, i.e. the initial conditions to plug into our calculation, (quantum mechanics introduces some interesting twists to this, a la Heisenberg, but it's not like you could have done it classically); three, we can never know that our equations and laws are exactly complete, they certainly aren't now and you could only ever prove them to arbitrary experimental accuracy. All of these are, however, matters of practicality, which no one disputes. They don't pose a problem to the philosophy of reductionism. As someone observed above, any explanation, that is, any science that goes beyond basic observation is a form of reductionism.
Now the great thing is that you don't need a complete reduction to fundamental physics to do everything. There are many interesting collective properties which depend only on math and on some, possibly approximate, properties of the constituents, but not all. This is statistical mechanics in physics, and every "higher order" science outside of physics. That's really cool, but it's not in conflict with reductionism, except when it comes to budgets.
I don't know if many people in the sciences would actually disagree with my position above. In religious and psychological discussion, anti-reductionist rhetoric is sometimes employed as a get-out-of-jail free card to magically produce free will without having to explain anything, or to elevate someone's pet beliefs above rational analysis. This is part of why I really dislike the language of emergence and the disparagement of reductionism.
Anyone?
Back in the American Enlightenment, Spinoza's whole "God == Reason" thing got the movement a lot of buy-in from various moderate and extreme godbelieving types like Deists, Puritans, Calvinists and Anglicans. Maybe the lesson is that some sorts of people are more willing to be fooled, so why not invite them along for the ride if it makes them feel good about it?
Such converences are great. The previous Beyond Belief tried to answer some really important questions, like 'Should science do away with god', 'can we be good without a god' and 'can science replace religion', all questions that can be answered with a big great enormous YES!
windy (#33):
As far as I can tell, the take-home message of that article is that the terminology is hopelessly confused, and as an outsider trying to learn about the subject, I'm tempted to agree. However, it doesn't address at all the subjects which I personally find most interesting, in particular the effect of spatially extended populations and the breakdown of mean-field assumptions. For the record, I don't give a damn whether environmental inheritance, group selection or kin selection is philosophically "nice"; I don't understand why people care which explanation is "warm and fuzzy". We begin with the observation that human beings and other animals can perform acts of cooperation and self-sacrifice. Explaining that in terms of maximizing the fitness of microscopic components doesn't change the validity of the original assumption, any more than the discovery of atoms reduces the pain I feel when I slam my foot into a rock.
Again, I wanna hear what PZ thinks.
Torbjörn Larsson, OM (#39):
Time asymmetry would be a problem, yes. I think you'd have to incorporate some kind of coarse-graining, yielding a weaker definition of "isomorphism". Maybe we should build a 2-category, in which functors exist to take time-evolution morphisms to other morphisms whose targets are in the same coarse-grained region of phase space?
It would be neat if category theory provided a formal framework for discussing issues of time asymmetry, ergodicity, emergence and such.
But there might be a "there" somewhere else. Do the Bell experiments, profound as they are, rule that out yet?
It seems to me that locality or size is not a constraint of "reductionism"- it just happens to go in a downward direction in our observations - so far.
I don't know about simulating a glass of water, but it seems like simulating protein conformations is a good task for a quantum computer. In fact, these sorts of simulations were the original motivation for quantum computing.
Now, the total information flow out of a quantum computer can't be any larger than out of a classical one of the same speed and number of (qu)bits. So you couldn't compute ALL the conformations, any more than you could with a classical computer. But you can't compute the highest integer either, and I bet you don't lose sleep worrying over that.
I'm certainly not arguing against emergence here. It just sounded like Kauffman's protein folding example was an argument from "inconceivably large numbers". I tend not to believe those arguments because advancing technology invalidates them with astonishing regularity.
The reductionism debate reminds me of a point that came up recently in a discussion about consciousness and the brain. Unless you subscribe to dualism/magical thinking, then your consciousness must be the outcome of all the physical activity in your brain. But actually modelling consciousness at that level would be an activity akin to modelling turbulent flow in terms of the exact motion of all the individual water molecules. Impossible in practice; probably impossible even in principle (Heisenberg, plus from relativity you can't tell what's happening in the Elsewhere that will affect you in future).
By coincidence, I happen to have gotten into a discussion of reductionism with a Christian friend on my blog. In the course of this, she referred me to this essay by a professor of religion. Me, I think the guy is confused. (My position in the discussion, modulo a few corrections, has been essentially what several people have said on this thread).
@49, he's a professor of religion... of course he's confused.
"because they don't recognize that their very first premise, that a god exists, is false."
Um,so god has been scientifically proven to be "false?" That's interesting. Or is it that you don't want to believe in god?
Agnostics say that they don't know - that's honest.OK, let's say there is no god. Please tell me what DOES exist. Everything dies and breaks down into simple compounds, except plastic. Thus, either plastic is the force that holds the universe together, or god. Whew! I'm glad that's settled.
Already Laplace said that long time ago:
"We may regard the present state of the universe as the effect of its past and the cause of its future. An intellect which at a certain moment would know all forces that set nature in motion, and all positions of all items of which nature is composed, if this intellect were also vast enough to submit these data to analysis, it would embrace in a single formula the movements of the greatest bodies of the universe and those of the tiniest atom; for such an intellect nothing would be uncertain and the future just like the past would be present before its eyes."
And it is not true, not only because of quantum effects because the sensibility of certain systems to initial conditions. In the long term the predictions diverge from reality. Chaos rules the world, yes, even Ctulhu is ruled by it.
Sorry for google-stalking, but I was trying to find some of your earlier comments and found this at another blog:
Your assumption about kin selection is wrong. Kin selection requires the breakdown of uniformity (but at the level of interactions, not necessarily mating) since individuals need to be more likely to help kin! Structuring is simply one way (perhaps the most important way) to accomplish this in practice.
And Bar-Yam's invocation of 'non-mean-field theory' has a whiff of re-inventing the wheel. Metapopulation biology? Other spatial models in ecology? Island biogeography? Assortative mating? Perhaps that's a parochial criticism, but another thing that bothers me about that article is that there is no discussion on how "evolution of reproductive restraint by group selection" relates to K-selection? Every adaptation that results in less offspring being produced must superficially look like "reproductive restraint" when it starts getting more common. So are most of the features of long-lived organisms now to be explained by group selection?
Well, I have a serious complaint about this year's Beyond Belief conference.
I had been looking forward to it since last year, and there was a real possibility that I could even get someone to pay for me to get there. When I saw a few weeks ago that there were no details on the site other than the dates announced a year ago, I emailed the people who seemed to be the organisers to see if it was on again. For obvious reasons I wasn't about to buy any return tickets between Sydney and LA if there was nothing happening at the other end.
Nobody got back to me, so I assumed that it wasn't on.
Well, maybe next year ...
#3: It is sometimes useful to distinguish between monism and reductionism. Materialism is one species of monism - i.e. that there is one basic stuff. Reductionism, by contrast, would be the family of positions opposed to various notions of emergentism.
Janus: See above.
jeff: I might add that Bohr's position as articulated in his words has been cricitized for decades as not corresponding to what he actually did as a physicist and on other grounds to boot.
Physicalist: Except that there might be laws at higher levels. Consider a chemical case: the substituents you can place on a benzene ring differ depending on the existing case. This seems (to my mind at least) as a case of "downward causation". Or, put another way, a pattern at the lower level isn't to be found. Of course it is all material, whence my first comment.
Dr. Myers, even though I deeply respect you and your work, I completely disagree with you on 'he (David Sloan Wilson) made a persuasive case for group selection'.
First, I fail to see how he could do that, when he completely lacks any experimental evidence for group selection. His main strategy is to obfuscate existing data with a lot of verbal gymnastics on top. Truly bright minds have already shown, demonstrated and convinced everyone else (well, almost) that group selection, thought theoretically plausible, is NOT a significant force in evolution. Period. Let the philosophers rumble on.
Second, being vigorously confrontational is ok if you have a name and a career behind you already, but building up a career just on confronting serious scientists and researchers is what I would call charlatanism. And this DSW guy is just such a person. He thinks he is in the same league as Trivers, Williams and Dawkins. Come on.
Third, we can doubt the scientific integrity of someone who accepts funding from the Templeton Foundation (suffice to say that it also funded the Discovery Institute?) and next thing we know, surprise surprise, he claims that religion can have an evolutionary explanation. Through group selection, of course.
Someone should once and for all tell him to shut the f up and do some serious research for a couple of decades before putting his boxing gloves on.
Cheers
V
Thank you for your summaries of this event, which I found very interesting - I'm sorry I'm a little late in finding them.
I'm going to start with a real clanger - I am a theist. But I am not going to try to argue with your atheism per se, I think that would just waste both our time. But I would like to comment on just one comment of yours: "the majority of people on the planet are practicing really bad science, because they don't recognize that their very first premise, that a god exists, is false."
If you wish to provide a reasonable alternative to the beliefs you find so objectionable, I suggest this is not the way to do it, for several reasons:
1. You assume that belief in God is the "very first premise" for many believers, but I wouldn't have thought it was the first premise of any believer I know. It is certainly not my first premise. A quick read of a few of the books of ex particle physicist John Polkinghorne would I think demonstrate that it is not true for him either. And while it may be true of some or even many believers, you have not offered any evidence for it. You are subtly putting down believers by alluding to their less logical representatives, when fairness would demand that you engage with the more thoughtful ones. If I made similar statements based on my assumptions about unthoughtful atheists, I think you would object, so is it unreasonable for me to request the same courtesy in return?
2. There is in fact a long history of philosophical, historical and personal argument, based on accepted facts (including some scientific ones) and claimed facts which support belief in a god in many people's minds. You obviously think the facts point to other conclusions, and you may find the arguments "unscientific", but there are many thoughtful people who would contest your conclusions, and some of them are eminent scientists. None of which proves anything about the actual (non)existence of god, but it does mean that many people start with something other than God (i.e. the world as they experience it) and use reason to arrive at their conclusions. Please don't pretend none of that is there, or just ignore it, or misrepresent belief as being totally non-logical.
3. You say with great authority "that a god exists, is false", when in reality that is a metaphysical statement that you really cannot justify. Repeating it in an authoritarian manner doesn't change that. Even Richard Dawkins recognises this and says that in his view God almost certainly does not exist.
Why do I raise these points? Because I believe some scientists are speaking with all the authority of their scientific study into areas where they can no more speak with certainty than I can. Such a view provokes misunderstanding, and could in the end lead to just as adverse consequences as the excesses of religion atheists often decry. By all means criticise our view, but please base it on a correct understanding of the beliefs of thoughtful theists, and please don't claim more than you can actually know. It may make good polemics, but it is poor science and poor logic and poor humanity.
Thank you for allowing me to make these comments. I think discussion of these issues is important, and I appreciate the opportunity to read about "Beyond Belief" and respond in this way. Best wishes.
Glad to hear that there is an afterlife to Beyond Belief.
Yes, well, Sean Carroll is for my money a clear-headed thinker and atheist to boot. About the only thing he discusses which I personally have a hard time to see as entirely agreeable is his and Jennie Chen's time-symmetric version of cosmology.
Sure, that the universe looks low entropic when looking back can be due to a simple symmetry breaking, we happen to be here instead of in a part of the multiverse where the arrow of time goes backward. But then again, we can see symmetry breaking without having the other possibilities realized.
It is therefore with some joy I note that he has started to discuss more regular eternal inflation scenarios, a cosmology that makes my head hurt much less. :-P
There are two cultures that clash here, on several levels.
One level is that Kauffman and others, often solid state physicists, are promoting emergency as a unique phenomena and sometimes as a possible path towards a deeper quantum theory.
AFAIU theoretical physicists in general thinks this is misguided. I don't think they say that for example biology can be understood effectively on a physical level, but that emergence is observed on systems that still interacts according to simple physics. As Carroll says, the system states still have a physical model.
The other level is that physicists and for example computer scientists may use idealized machines or gedanken experiments to broaden knowledge. Carroll's Deep Thought doesn't look as much as a religious agent to me as an oracle that can decide non-computable functions.
Interesting. I would at a hasty guess think that you capture macrostates in ergodic systems that way. But I don't see how dissipative, fundamentally time-asymmetric processes and possibly chaos would not deviate from such isomorphism behavior.
Re comment #24:
IIRC Bell test experiments have verified that there are no local hidden variables to ~ 15 sigma. And I believe I read a press release on a new experiment that lie more in the line of the original EPR suggestion, presumably with fewer potential loopholes, that pushed it to ridiculous 23 sigma or so. It must be one of the best verified "no-go" result or result of any kind we have.
For my money, there is definitely no more "more" there. But again, identifiable variables is the principle, and knowing the actual states is the practice. Let's see if we can decide whether particles are particles or strings first.