Give me creaturely over preacherly any day

You can tell when a dogmatic theist has to review a book by an unapologetic atheist: there's a lot of indignant spluttering, and soon the poor fellow is looking for an excuse to dismiss the whole exercise, so that he doesn't have to actually think about the issues. That's the case with Leon Wieseltier's review of Dennett's Breaking the Spell—it's kind of like watching a beached fish gasp and flounder, yet at the same time he apparently believes he's the one with the gaff hook and club.

It's full of self-important declarations that reduce to incoherence, such as this one:

You cannot disprove a belief unless you disprove its content. If you believe that you can disprove it any other way, by describing its origins or by describing its consequences, then you do not believe in reason. In this profound sense, Dennett does not believe in reason. He will be outraged to hear this, since he regards himself as a giant of rationalism. But the reason he imputes to the human creatures depicted in his book is merely a creaturely reason. Dennett's natural history does not deny reason, it animalizes reason.

One moment he's telling us that just tracing the origins of an idea is insufficient to disprove it (sadly for Mr Wieseltier's argument, there is no sign that Dennett disagrees), the next he's telling us that the origin of Dennett's reason is "creaturely" and "animalized", and therefore of a lesser or invalid kind. I had no idea we could categorize reason by the nature of its source (I'd like to know what varieties of reason he proposes: "creaturely", "human", "divine"? Is there also a "vegetable reason"?), but even if we could, by his initial premise, it wouldn't matter: he needs to address its content, not carp against it because it is the product of natural selection rather than revelation.

Then there's this rather bewildering build-up. Wieseltier carefully builds a case that he has caught Dennett in an internal contradiction, an idea he pounces on with a kind of petty triumphal glee…but all it shows is that he doesn't know what he's talking about.

Like many biological reductionists, Dennett is sure that he is not a biological reductionist. But the charge is proved as early as the fourth page of his book. Watch closely. "Like other animals," the confused passage begins, "we have built-in desires to reproduce and to do pretty much whatever it takes to achieve this goal." No confusion there, and no offense. It is incontrovertible that we are animals. The sentence continues: "But we also have creeds, and the ability to transcend our genetic imperatives." A sterling observation, and the beginning of humanism. And then more, in the same fine antideterministic vein: "This fact does make us different."

Then suddenly there is this: "But it is itself a biological fact, visible to natural science, and something that requires an explanation from natural science." As the ancient rabbis used to say, have your ears heard what your mouth has spoken? Dennett does not see that he has taken his humanism back. Why is our independence from biology a fact of biology? And if it is a fact of biology, then we are not independent of biology. If our creeds are an expression of our animality, if they require an explanation from natural science, then we have not transcended our genetic imperatives. The human difference, in Dennett's telling, is a difference in degree, not a difference in kind—a doctrine that may quite plausibly be called biological reductionism.

To declare that we are not limited by our genetic imperatives does not in any way contradict the statement that we are material, biological beings with behaviors that can be explained scientifically, without recourse to the supernatural or any other kind of immaterial vitalism. Opposing simplistic genetic reductionism—which, by the way, is good to see from Dennett, because he has a bit of a reputation for being far too narrowly reductionist in his views—is not the same as denying a natural, biological basis for behavior. When Wieseltier tries to insist that genetic determinism is the same as biology, he's just flaunting his own ignorance.

The whole review reads this poorly, and I suppose I could take it on paragraph by paragraph…but nah. Brian Leiter has already torpedoed it, so even this much seems like excess. The New York Times really needs to do a better job of finding qualified reviewers—it seems in this case they just found a guy anxious to posture against the ungodly, with no competence to actually judge the book.

More like this

I saw the review. I was tempted to take it on myself, but unfortunately the technical difficulties preventing me from posting to my ScienceBlogs blog made me decide that it wasn't worth the effort.

Reading such material, and finding little within that particularly strongly resembles a coherent argument, I tend to assume the actual line of reasoning followed by the author appears nowhere in the work, and is quite simple:

1. This work contradicts a popular prejudice I happen to hold,

2. Therefore, I must attack it.

... that the arguments used for said attack might be spurious well beyond the point of absurdity is of no great concern. It's a popular prejudice, so the author and the editorial board may safely assume there will be ample applause. Indeed, enough readers will applaud the righteous smiting of the cheeky infidel that squeaks of 'but your criticisms make exactly no sense', however well-reasoned those squeaks might be, may just as safely be ignored.

To declare that we are not limited by our genetic imperatives does not in any way contradict the statement that we are material, biological beings with behaviors that can be explained scientifically, without recourse to the supernatural or any other kind of immaterial vitalism.

That's exactly what I thought, too. All that bluster and build-up, then a conclusion that just doesn't compute.

As the ancient rabbis used to say, have your ears heard what your mouth has spoken?

Er, right, Mr Wieseltier, and my mouth is agape that you so neatly conclude that Dennett's recanting on his humanism. Is it that hard to understand that human consciousness has developed to a point that we're able to observe and analyze our own behavior, then modify our own behavior, formulate moral guidelines (for better or worse) etc?

Interesting, too, that Wieseltier casts his ire as a defense of philosophy and not just religion. Philosophy is great. Sometimes it may even be helpful in real world terms. But some philosophy is junk, too, and however enjoyable the mental calisthenics involved in philosophizing may be, it certainly helps if they're rooted in reality. And science is the discipline that seeks to determine reality.

Seems Wieseltier's review has been met quite favorably across the blogosphere, too. Guess I shouldn't be surprised. But his simply adopting a withering and erudite tone does not make him correct. I'm sure the clergy railed against Galileo with similar erudition. Which is not to say that Dennett's arguing scientific theory in his book or that his position is above criticism. Just to say that those on the side of fiction and superstition are quite capable of some impressive speechifying when doing so.

To declare that we are not limited by our genetic imperatives does not in any way contradict the statement that we are material, biological beings with behaviors that can be explained scientifically, without recourse to the supernatural or any other kind of immaterial vitalism

Well, what do you mean by a "scientific explanation of a behavior"? I know what Pinker and Dennett think it is, but what do you mean by it? You don't seem to buy their EP adaptationist stories. You don't have to assume a supernatural force or being to doubt whether most complicated human behaviors like religion have a "scientific meaning".

I wish Wieseltier had done a better job but a Dennett's a purblind idiot.
With that in mind:
Why do you insist on arguing logic with priests? Priests begin from their definition of all meaning being social. They argue not only from tradition but from the moral imperatives of tradition: you arguing that blood can not not become wine is irrelevant. And yet you and Dennett wonder why you cannot convince the faithful? That's simply illogical.

You're more interested in the world than in perception, but then perception is how we experience the world. Computers don't perceive, animals do, and you're an animal. Those of us who are observant animals- who do not therefore pretend to be machines- pay close attention to and question ourselves, our motives and intentions. We try to remind ourselves to doubt, lest we begin to respond only out of reflex or habit. Scientists are not science any more than policemen are the law. Dennett doesn't understand that simple fact and Wieseltier doesn't respond to that arrogant stupidity clearly enough. But he's right.

Science does not explain Brian Leiter's fixation on academic social status or his unwillingness to respond to the blatant contradictions between his semi-leftist politics and his dedication to the academic priesthood. He's a Nietzschean snob. His tastes and manner define him more than his 'philosophical ideas.'
Why is Brad DeLong so god damn phobic of anything that has to do with Noam Chomsky, when the have so much in common? Delong's responses are frankly irrational. He goes of like a loon. Science can't explain that. Or perhaps it can, but it's not a science that we have access to.

Try to explain out loud to yourself why we defend the rule of law and not of men. Laws are nothing but traditions written on parchment. Why not just have people like you or Dennett make decisions for us? Why go through all this absurd ritual? Since lawyers are nothing but amoral craftsmen (and con-men) why not have scientists debate among themselves to resolve court cases?
Think about what the rule "of law" means and why we have it.
Then go back and take a high school literature class.
Philosophy is for adults son. Grow the fuck up,
and we can talk.

Wow Seth. I have to say that was one of the least coherent arguments that I have ever read. I will have to read it again but there might not have been a single well formed idea in the entire screed. Maybe you could try that again if you have time.

Slight nitpick that might be a "No true Scotsman" fallacy, Robert S.: Philosophy, at least the type my brother's gotten a degree in, is actually about critical thinking, sometimes using it to untangle paradoxes. The problem with the perception of philosophy is that there's a lot of pseudophilosophy out there, not interested in discussing the steps of logic, just like pseudoscience isn't willing to discuss evidence or methods of gathering that evidence.

It was once painful to my brother's psyche when he asked if there was a philosophy section in a bookstore. The ditzy employee he asked responded, "Well, we've got a New Age section."

I'm struck by Seth's insistence that "science doesn't explain..." certain things. Does he have some other canon of explanation he'd like to propose? All I see in his remarks are dubious interpretations like: "He's a Nietzschean snob. His tastes and manner define him more than his 'philosophical ideas.'"

It's easy to hurl brickbats-- but Leiter is entirely right when he points out that the etiology of a belief matters. And the point of Dennett's book is to argue that there are naturalistic etiologies for religious belief, etiologies which already explain far more and fit far better with our scientific understanding (the only understanding of the world that actually passes any empirical tests) than any religious metaphysics.

As for the rule of law, I think we have a very clear case of 'red herring' there. It has very little to do with science or religion-- it's more a matter of common sense: Do you trust individuals to get things right reliably and fairly? Which individuals? In any large and complex society, a system of rules we can rely on to structure these processes (and use to ensure our own point of view is heard) is way more reliable and resistant to abuse. Sadly, nothing can stop individual humans from undermining any such system (look at Bush and Cheney's over-reaches and the supine response of your Congress...). But we do the best we can despite them.

I had read this review on Saturday, and was waiting to hear what you would say. I think I took away something different from Wieseltier's review than you did. Perhaps it was because I was reading Roger Penrose the previous evening. In The Road to Reality, Penrose says that

...mathematical existence is different not only from physical existence but also from an existence that is assigned by our mental perceptions.

The patterns of the Mandelbrot set ( and I suppose he would extend this to all of mathematics)

were already 'in existence' since the beginning of time in the potential timeless sense that they would necessarily be revealed precisely in the form we perceive them today, no matter at what time or in what location some perceiving being might have chosen to examine them.

I take that to mean that that we humans have the ability to do mathematics is to be explained by biology, but that mathematics transcends our particular biological facts and mathematics is objective in that any biology (or electronics) that can do math. will arrive at the same mathematical facts. Mathematical truth cannot be explained by biology.

Since science is concerned with physical existence, perhaps biology can explain our science. But I take it that Penrose's argument is that it cannot explain mathematics.

Thus, when it came time to ascribe meaning to "to have the ability to transcend our genetic imperatives?" I took it to mean a similar thing - namely, to have access to something objective that transcends our biology, that would be the same for another biological or electronic entity with similar capabilities though of vastly different design.

---

I think this also makes it clear as to what the reviewer is objecting to:

Everything we value - from sugar and sex and money to music and love and religion - we value for reasons. Lying behind, and distinct from, our reasons are evolutionary reasons, free-floating rationales that have been endorsed by natural selection.

If everything we value has been endorsed by natural selection, then we haven't transcended our genetic imperatives, have we?

And the above quote doesn't square with the theory of evolution as I understand it. I thought that in general biological pathways, organs, organisms have a versatility far greater than the qualities acted upon by natural selection, which is how complexity can increase. Thus our brains and ultimately our behavior, while selected upon to optimize survival in an African savannah setting or wherever, might be capable of far more than that. All we can say is that the extra stuff didn't give us a significant selection disadvantage. If science and math and music did not exist during the bulk of our evolutionary formative period - if e.g., humans 60,000 years ago were essentially biologically indistinguishable from ourselves - then how can we say that science and math and music abilities were determinative in our evolution? Rather, evolution has fashioned a brain that turns out to have a much richer repertoire than what was selected for, and again, we have transcended our genes.

The problem with the perception of philosophy is that there's a lot of pseudophilosophy out there, not interested in discussing the steps of logic, just like pseudoscience isn't willing to discuss evidence or methods of gathering that evidence.

Good point, BronzeDog. And I certainly would've groaned along with your brother had I heard that New Age remark!

I haven't read Dennett's book and I certainly haven't agreed with everything he's said in the past, so I have no opinion on that.

But I notice that Wieseltier hauls out an accusation of "scientism" in his third sentence. I've found that, while scientism as usually defined--the treatment of science as a kind of religion--is a genuine fallacy that it might in principle be useful to accuse people of, in practice when somebody throws around accusations of scientism it is a remarkably reliable marker of pious crankery. The word's been terribly debased.

Just to let you know, P.Z., but there most certainly is "vegetable reason." I use it all the time to find the right light and soil conditions! *LOL*

I'm 3/4ths of the way through Breaking the Spell, and what strikes me about it is the way Dennett bends over backwards to try not to put off people like Wieseltier. The book is punctuated with asides that beseech the thoughtful religious who might be reading not to put it down. I wonder how soon Wieseltier put (or threw) the book down?

His review is the standard theistic response to an intellectual criticism of theism. Much ado is made over the person making the criticism ("In his own opinion, Dennett is a hero."; "Dennett flatters himself that he is Hume's heir."), over perceived flaws in the form of the criticism ("It is not based, in any strict sense, on empirical research."), or the unpleasant thoughts that they associate with atheism ("They cannot fortify you, say, after the funeral of a familiar intentional system.") -- but not once does he actually provide a positive defense of theism, because there is none. There never was, and there never will be. The same goes for the recent religion threads here.

You don't have to assume a supernatural force or being to doubt whether most complicated human behaviors like religion have a "scientific meaning".

There's a difference between having a "meaning" and being explainable.

Religion doesn't have to have a meaning to be perfectly explainable by science - all it needs is to not conflict with the observed laws of nature in any way ... and for people to have percieved things they didn't understand, and/or tell people such is hardly contrary to science.

Any likely scientific "explanation" of religion will be fundamentally unsatisfying, since it would be equivalent, in some ways, to a scientific "explanation" of why I chose the clothes I did today, for instance.

It basically boils down to: "There's no reason it couldn't happen, and thus, well, did."

Perhaps it's better to compare it to "explaining" why, when rolling this die, I got a "3". Fundamentally, it boils down to: "Something had to happen. Why not that?"

By Michael "Sotek… (not verified) on 19 Feb 2006 #permalink

"You cannot disprove a belief unless you disprove its content. If you believe that you can disprove it any other way, by describing its origins or by describing its consequences, then you do not believe in reason."

I think this has gone on long enough. We need to take a nice, heavy textbook on reasoning, one with a nice solid explanation between deductively disproving a proposition and inductively arguing against a proposition's likelihood of being true, and beat this man senseless with it.

This absolute inability to grasp the difference between the two simply cannot continue.

"One cannot disprove an idea (an idea like, perhaps, a religion) by describing its origins. However, one can make a decent argument against that religion's likelihood of being true if one can prove that the religion originated with the statements of someone who, upon examination, doesn't seem very credible."

That wasn't that hard. Honestly.

I certainly don't have any objection to evolutionary theory or thinking, but I do take objection to Dennett's version of it. He claims that evolution by natural selection is an "algorithmic process". Hello? I should use that as a pick up line. And it should be perfectly obvious that religion is not a natural phenomenon; chimpanzees don't have religion. Religion is a response to a problem set that only emerges with the linguistic reduplication of the world and the insistency of symbolic thinking, together with the integration and transmission of social orders. It's a socio-cultural phenomenon and it's not obvious that it requires or is amenible to natural-scientific explanation. It's in our alienation and not in our genes. And though I wouldn't want to revert to a Kantian style dualism, there's much to be said in favor of the distinction between facts and norms, which Dennett tends to subvert, which, in however a muddled and wishful way, I think the reviewer was trying to get at. The delimitation of different questions and different applicable criteria is something the rationality of science itself can not do without.

By john c. halasz (not verified) on 19 Feb 2006 #permalink

"You cannot disprove a belief unless you disprove its content."

Oh yeah?

How about my first testable principle then, namely:
"No god is detectable (even if that god exists)"

Unless, and until someone can detect a god, or a valid, repeatable message or communication from that god, then the valid assumption must be that the god does not exist.

Put up, or (preferably) shut up, believers!

By G. Tingey (not verified) on 19 Feb 2006 #permalink

Oh, and "Vegatable reason" ...

How about Andrew Marvell in "To his coy mistress ....

" Vaster than empires, and more slow, our vegetable love should change and grow ...."

By G. Tingey (not verified) on 19 Feb 2006 #permalink

And it should be perfectly obvious that religion is not a natural phenomenon; chimpanzees don't have religion.

It should be perfectly obvious that hairlessness is not a natural phenomenon; chimpanzees are not hairless.

It should be perfectly obvious that walking upright is not a natural phenomenon; chimpanzees don't walk upright.

It should be perfectly obvious that prominent breasts are not a natural phenomenon; chimpanzees don't have big tits.

etc...

John Cornwell has a similar attempt at smothering Dennett in the Sunday Times. Apparently it is an insidious book because it proceeds from "emotive, ill-informed prejudice" and is therefore combatted by wheeling out Nazism and Stalinism - even if this were apt, it is a tu quoque.

Cornwell himself seems driven by pique at the playful suggestion that the Communion wine be DNA tested and the genome of Christ obtained. It is clearly a splendid idea. Despite the accession of the Catholic church to scientific advances made in recent times, the refusal to recognise the metaphor of transubstantiation strikes me as a perfect example of how the faithful miss, or more accurately are directed to miss, the whole point.

The Eucharist is part of an ancient, 'multilayered' doctrine - multilayered being Cornwell's nod to doctrinal schism spanning centuries wherein the only actual bodies broken and blood spilt were all too human.

Let's put it this way - the reviewer says that Dennett is arguing that "religion is an artifact of our biology". Then equally, reason, science and mathematics are also artifacts of our biology.

Then it is time to read Coturnix's articles on reductionism.

I don't usually like Dennett, but this argument of his:

"This fact does make us different. But it is itself a biological fact, visible to natural science, and something that requires an explanation from natural science"

touches on a point I don't see made often enough. Evolution has given humans capacities and freedoms that it has not given to other creatures. You do not need a transcendent source to get these capacities.

For example, genes give horses hooves, and genes give humans hands. We have no choice in the matter. Both are defined by evolution, but the horse is constrained to about six simple hoof-behaviors, whereas humans are gifted with an essentially infinite array of hand-behaviors. We didn't evolve piano-playing per se, but we did evolve hands capable of piano-playing.

Vegetable reason, PZ?

Of course! Haven't you read anything produced by the Discovery Institute? Hail Great Pumpkin, designer of the Patch.

Throughour nature we see traits seected for one thing being adapted and co-adapted for things they weren't originally selected for. Our evolution produced an animal capable of high level social interaction (culture). Everything else that flows from that ultimately *depends* on our genes, not on "transcending" our genes.

John Halasz: cultural processes are a subset of natural processes and are not shared by all species. Biological reductionists miss this, but there's nothing paradoxical or difficult about saying that the capacities for language and culture are hereditary and limited to one species.

Is there also a "vegetable reason"?

I think you just saw it in action in the review.

Actually, "vegetable reason" sounds like a perfect description of how logic and critical thinking is taught to school children. To clarify, let me propose an experiment:

Step 1) Find out who in your area is in charge of setting curriculum standards.
Step 2) Write them a letter asking if it is a requirement for graduation that high school kids know what a sophistry, a fallacy and an argument are.
Step 3) Evaluate the response.

My hypothesis, based on personal experience,is that you will get the following response or some version of it.

"Thank you for your interest in the education system of _____________(insert region here). While we do not require that branch of philosophy or even offer it, we are always interested in encouraging critical thinking by students. I am sure they get that kind of thing in language arts class, or math or something."

Vegetable reason - Plant the seeds and hope for the best. It is sort of like teaching reading by hurling heavy books at students from a great height.

"Despite the accession of the Catholic church to scientific advances made in recent times, the refusal to recognise the metaphor of transubstantiation strikes me as a perfect example of how the faithful miss, or more accurately are directed to miss, the whole point."

Indeed. I've discussed this with a Catholic philosopher who fancies himself as a bit of a modernising liberal and who claims not to believe in transubstantiation. Nevertheless he defends it as not prima facie absurd, despite the fact the doctrine claims a) the wafer and wine do literally turn into the body and blood of Christ, b) transubstantiation is not metaphor or symbolism, and c) the wafer and wine do not physically (or in any observable way) change. Now I'm not a philosopher, but I really can't see how all three can be true. His attempt to explain it consisted of comparing the wafer to a table that you crouch under in the rain, turning it into an umbrella. Never mind that the use of 'umbrella' in this context is a metaphor.

By Ginger Yellow (not verified) on 20 Feb 2006 #permalink

Everyone (neurologically normal) has philosophical opinions, but not everyone has spent the time to be informed about some of them, much less make an attempt at a synthesis of many of them. The funny thing is, though, a little knowledge can be a dangerous thing, here, for the same reason it is in pseudoscience. Pseudophilosophy is just as dangerous and I suspect, as people get stuck between "old and new" increasingly, it will only increase.

John C. Halasz "[Religion is] a socio-cultural phenomenon and it's not obvious that it requires or is amenible to natural-scientific explanation."

All functions of society are socio-cultural phenomena and one cannot simply use one such function to analyze another. Or rather one can argue for a scientific explanation of religious belief but that will never eliminate its roots in our animal tendency to pattern thought based on previous experience. My point about Leiter is that the tastes and manners that precondition his contradictory arguments are the equivalent of faith.

The laws of science are not socio-cultural phenomena, but the uses to which science is put are. What atheist humanists and religionists both are offended by in Dennett's form of secular anti-humanism is the putting forth of the tools od science as themselves answers to questions of morality. But his claims to observational clrity are belied by the fact that he can't realize that religion only concerns itself marginally with the study of the world: it's primary interest is in law and social order. To hear him lecturing the faithful is akin to listening to a mathematician lecture fans of literature for reading lies.
"What's with you people. Don't you know, stories aren't real!!?"

The defense of religion is the defense of social order. Dennett argues that science can displace that order- displacing an order of language with one of number- and be both moral and just. But the activity of scientists- as opposed to the "laws" of science- is a social activity and as much an order of language as is religion or a court of law. Human beings experience life as a function of a social reality. We do no experience the world as number and computers do not 'experience' the world at all.

Religion is not the problem, it is those who claim unmediated access to the 'real' world whether the world of numbers or of gods. Dennett makes Nino Scalia's arguments for clarity and order but bases them on science rather than Catholic doctrine.
Big fucking shit. It still don't fucking work.
That I should have to explain this shit to adults...

PZ Myers wrote:

One moment he's telling us that just tracing the origins of an idea is insufficient to disprove it (sadly for Mr Wieseltier's argument, there is no sign that Dennett disagrees), the next he's telling us that the origin of Dennett's reason is "creaturely" and "animalized", and therefore of a lesser or invalid kind. I had no idea we could categorize reason by the nature of its source (I'd like to know what varieties of reason he proposes: "creaturely", "human", "divine"? Is there also a "vegetable reason"?), but even if we could, by his initial premise, it wouldn't matter: he needs to address its content, not carp against it because it is the product of natural selection rather than revelation.

Again, you and I take away a different meaning from the reviewer. Note that I haven't read Dennett's book, and cannot say whether Dennett does what the reviewer charges him with. Nor can I speak for the reviewer; I speak only for the meaning I received from reading the review.

Consider me, reasoning (don't laugh!). One day science will be able to explain in terms of brain structures and neuro-transmitters the entire process going on inside my head, and the origin of all these things in terms of evolution. However, the validity of reasoning doesn't depend on this description. Indeed, we wouldn't be able to arrive at this description without a priori taking reason to be valid, independent of the details of my brain apparatus.

When the reviewer says Dennett is animalizing reason, he charges that Dennett is making the validity of human reasoning be contingent on whether evolution brought about the right structures in our brain or not.

YMMV, and I look forward to hearing your take on it.

The laws of science are not socio-cultural phenomena, but the uses to which science is put are. What atheist humanists and religionists both are offended by in Dennett's form of secular anti-humanism is the putting forth of the tools od science as themselves answers to questions of morality.

That's not what Dennett does at all. In previous books (I have yet to read this one), Dennett goes out of his way to stress that morality (should) is separate from biology (is), even though the former is generated by the latter. In a similar manner on a meme-based view of culture the popularity of an idea is not based on the reproductive benefits it provides to its holder, but rather on the meme's own reproductive ability. What he's saying is that the tools of science can help us understand how human morality came to be, and thus help us shed dangerous illusions (in this case, morality comes from this book because the invisible sky monster says so) about the nature of morality. Understanding the biological origin of morality/religion does not dictate what our morality/metaphysics should be, but it lets us better use our reason to make that judgement.

By Ginger Yellow (not verified) on 20 Feb 2006 #permalink

Hey Ginger Yellow. Transubstantiation is explained by simple aristotelian logic. All substance has an unchanging essence and changeable accidents. You know, water changes phase, but it is still water. So in the miracle of the eucharist, the wafer changes its essence without changing its accidents. Only a heretic would fail to see that obvious truth. (I don't know smileys, but just imagine that I have inserted a sign for sarcasm.)

Why is our independence from biology a fact of biology? And if it is a fact of biology, then we are not independent of biology.

I'm guessing this guy has never worked with computers. Computers, like humans, are set up so that not everything they can do and achieve has to be hard-wired into them (although a few basic processes are). I know from experience that computers, again like humans, don't work too well if you stick a pickaxe through them*. A computer's independence from electronic engineering is a fact of electronic engineering. This is not self-contradictory.

If Mr Wieseltier feels that in fact this is self-contradictory, may I suggest that he test his hypothesis by putting a pickaxe through his computer?

* My experience here is of course with computers not humans. It's very stress-relieving.

....there most certainly is "vegetable reason."

Actually, it's the name of one of Santayana's earlier and more obscure works, written while he was still working out the basics of his ideas ;-).

"If Mr Wieseltier feels that in fact this is self-contradictory, may I suggest that he test his hypothesis by putting a pickaxe through his computer?"

Why not his head, if he's so confident?

By Ginger Yellow (not verified) on 20 Feb 2006 #permalink

John C. Halasz writes:

I certainly don't have any objection to evolutionary theory or thinking, but I do take objection to Dennett's version of it. He claims that evolution by natural selection is an "algorithmic process". Hello? I should use that as a pick up line.

I think I know what Dennett means by "an algorithmic process" and I don't understand your objection to it.

"Lather, rinse, repeat" is an algorithmic process, too. (Albeit less clearly so than "mutate, select, repeat" because it's usually assumed that the instructions are decoded by an intelligent interpreter.)

I don't see what's wrong with pointing out that natural evolution is reasonably viewed as a naturally-occurring instance of a genetic algorithm. It can be illuminating, e.g., in a discussion of evolution as a heuristic search system, where the heuristics emerge from a brutally algorithmic "implementation."

(This has been useful to me, in understanding the effectiveness of natural evolution compared to most genetic algorithms; evolution through randomly-varied and randomly-connected environments reduces the greediness of the search by reducing premature committment. It is also useful, for many audiences, in de-mystifying evolution by showing how utterly mechanical and trivial processes can indeed implement sophisticated searches of "design space.")

If you have some good argument against this view, it'd be better to come out and state it than to simply ridicule it, as though it was just patently wrong and stupid. It seems to just be your "this guy is so dumb..." prelude to making similarly unclear objections to other things. That reminds me of the tone of the NYT review, and makes me wonder if you understand what Dennett is actually getting at.

I don't see what's wrong with pointing out that natural evolution is reasonably viewed as a naturally-occurring instance of a genetic algorithm. It can be illuminating, e.g., in a discussion of evolution as a heuristic search system, where the heuristics emerge from a brutally algorithmic "implementation."

No, that's backwards. Genetic algorithms were specifically created to *simulate* our understanding of natural selection. But so what? Anything can be simulated on a computer. Ever play SimCity? Does that mean cities are algorithms too?

There's a difference between having a "meaning" and being explainable

Well, I'm not sure there is, but I should have used consistent terminology. "Scientific explanation" it is.

Any likely scientific "explanation" of religion will be fundamentally unsatisfying, since it would be equivalent, in some ways, to a scientific "explanation" of why I chose the clothes I did today, for instance. It basically boils down to: "There's no reason it couldn't happen, and thus, well, did."

Exactly my viewpoint.

"[Religion is] a socio-cultural phenomenon and it's not obvious that it requires or is amenible to natural-scientific explanation."

Well, it's nice of Dualism to show up.

I don't see what's wrong with pointing out that natural evolution is reasonably viewed as a naturally-occurring instance of a genetic algorithm. It can be illuminating, e.g., in a discussion of evolution as a heuristic search system, where the heuristics emerge from a brutally algorithmic "implementation."

No, that's backwards. Genetic algorithms were specifically created to *simulate* our understanding of natural selection.

I don't think that's right. I don't think that either historical priority or the simulated/real distinction is crucial. (Or even as tenable as many people assume.)

Genetic algorithms are not just simulations of physical/biological evolution; they are actual examples of evolution in a more general sense. The principles of evolution are not intrinsically tied to the substrate of the original examples.

Consider addition. Addition in a computer is not just a simulation of addition in my head---and addition in your head is not just a simulation of counting on your fingers.

Likewise, when physicists talk about light waves, that's not just a metaphor; light waves are not merely analogous to waves on the surface of water---there's a general class of phenomena that was first recognized in one context, but literally occurs in other contexts as well. The naming of these concepts is done by linguistic metaphor, but the resulting named concepts are not just metaphors for (or simulations of) the initial kinds of examples. When physicists talk about waves of light, these days, they're not just making an analogy about water motion.

It just doesn't much matter in which context a general phenomenon was discovered.

If I use a genetic algorithm to evolve a digital electronic circuit design, I'm not just simulating evolution or simulating designs. And I'm certainly not simulating biological evolution. I'm actually using evolution to search for actual designs.

More subtly, when Robert Axelrod uses evolutionary algorithms to explore basic issues of the evolution of cooperation, he's not just simulating biology. The principles he's exploring apply to things besides biology, e.g., anthropology and economics. He's studying how certain abstract patterns of interaction can emerge from other abstract patterns of interaction, given certain formal properties, irrespective of the underlying medium. (E.g., microbes, humans, multinational corporations.) His "simulations" aren't just simulations---they're examples of those very general kinds of things, which can be interpreted as models of other things.

Whether something is a literal example of something, or a model of something else, is generally a matter of interpretation.

This is even true when using biological models. For example, if I'm studying fish fighting and cooperating in aquaria in a lab, presumably I'm not just interested in how fish fight and cooperate in aquaria in a lab. I am hoping that is a reasonably accurate model of phenomena that occur in the wild as well---and likely with different kinds of fish, or even entirely different kinds of organisms.

(When Darwin wrote so much about barnacles, his main points were not "barnacle science." The main utility of barnacles to Darwin was that they were an easily-studied "natural experiment," i.e., a naturally-occuring simulation of what happens with many other kinds of organisms.)

To the extent that you explore general, abstract principles with specific, concrete examples, you are using one thing to simulate another. This is what makes the simulated/real distinction a whole lot weaker than it seems at first.

That's one of Dennett's basic points, and I think is a good one. Too many people miss that basic point, and think he's just wantonly conflating things. He's generally not; he's got very good reasons for discounting certain distinctions that most people overrate.

The myth of a unipolar consciousness "I say what I mean and I mean what I say" is as American as Henry Ford and 'can-do' anti-intellectualism. And of course, the parallel to Dennett is not Scalia but Posner's Law and Economics. And Posner and Leiter are buddies of course.
But the rule of law is not the rule of reason; we choose the former because the latter is impossible. Dennett attacks not only religion but language itself; language which can not escape ambiguity.
Under the rule of reason, if it were possible, no one would ever get off on a technicality.

I looked up the Dr. Seuss reference just to make sure I was right, and look what I found: "A plea for the Humanities." by another Seth too!
I guess god must be smiling on me today.

LEON WIESELTIER wrote:

Scientism, the view that science can explain all human conditions and expressions, mental as well as physical, is a superstition, one of the dominant superstitions of our day; and it is not an insult to science to say so.

Dictionary.com defines superstition:

1. An irrational belief that an object, action, or circumstance not logically related to a course of events influences its outcome.
2.
a. A belief, practice, or rite irrationally maintained by ignorance of the laws of nature or by faith in magic or chance.
b. A fearful or abject state of mind resulting from such ignorance or irrationality.
c. Idolatry.

It isn't clear which definition he might be using. The only one that would not be absurd in this context is 2c, which would still be quite a stretch.

Genetic algorithms are not just simulations of physical/biological evolution; they are actual examples of evolution in a more general sense. The principles of evolution are not intrinsically tied to the substrate of the original examples. Consider addition. Addition in a computer is not just a simulation of addition in my head---and addition in your head is not just a simulation of counting on your fingers.

Addition is different because it is a process that has nothing to do with the physical world. Mathematicians have a set of axioms and operations that allows one to perform the operation "addition". Because it can be defined exactly, the results can be perfectly performed by a computer.

Natural evolution can't be, because it is something that we have only a very imperfect knowledge of. Any algorithm that we can make is going to different from the real thing. It's the same problem that economists and weathermen have -- they can't predict recessions and rain with any real accuracy because their models are flawed simulations of the real world.

"Under the rule of reason, if it were possible, no one would ever get off on a technicality."

Again, continuing what is apparently my own little personal crusade, I'd like to remind everyone of the existance of INDUCTIVE REASONING.

Those "technicalities" exist for two reasons, generally speaking.

First, because they guide our inductive assumptions away from tempting but generally incorrect inductive conclusions (example, no bringing up the rape victim's sexual history because people frequently conclude that a girl who has sex with men at all has sex with any man at any time and can't be raped),

and second, because sometimes there are values more important than a conviction at a trial (convicting a man of a crime is generally considered less important than stopping the police from torturing confessions out of people, so we risk losing the former by excluding confessions produced by the latter).

Once you accept that not all reasoning is deductive, these make perfect sense.

The myth of a unipolar consciousness "I say what I mean and I mean what I say" is as American as Henry Ford and 'can-do' anti-intellectualism. And of course, the parallel to Dennett is not Scalia but Posner's Law and Economics. And Posner and Leiter are buddies of course.
But the rule of law is not the rule of reason; we choose the former because the latter is impossible. Dennett attacks not only religion but language itself; language which can not escape ambiguity.

I have no idea what your point is here. Dennett has written several books attacking the idea of a unipolar consciousness. And I really fail to see how Dennett is attacking either language or the rule of law.

By Ginger Yellow (not verified) on 20 Feb 2006 #permalink

there's nothing paradoxical or difficult about saying that the capacities for language and culture are hereditary and limited to one species.

The other point being missed is that recent research in animal behaviours shows that a) all of them develop some language specific to their needs and b) all of them have some level of culture. What differentiates humans from other animals is not the capacity to form these, but the level of complexity that can be developed, and the physical capacities to do so. If a chimpanzee had the vocal cords for human speech they would easilly manage language equal to that of a severally mentally retarded person, and they can manage a lot of the socializations we can. On the other hand, some parrot can't develop the complex social system, but can do complex problem solving, are better at math than apparently most fast food cash register operators and only lack the memory for an understanding of language equal to that of a chimp. Other animals are all across the board, with some huge gaps in capacity, but at the same time having significant jumps in ability in other areas. It just that for most of the time, since the rules of science where seriesly codified and applied, people like Leon Wieseltier where defining what we should *expect* to find in other animals. If you don't look for something, you obviously are not going to find it and for a long time no one wanted to find social systems, complex problem solving or language in "mere" animals.

Addition is different because it is a process that has nothing to do with the physical world. Mathematicians have a set of axioms and operations that allows one to perform the operation "addition". Because it can be defined exactly, the results can be perfectly performed by a computer.

I think that's a red herring. Exactitude isn't, or shouldn't be, the issue, if the question is whether something is merely a simulation, rather than an actual instance of something.

A simulation is simply a process interpreted as a model of another process, and most kinds of models are not exact.

What makes something a simulation is the interpretation. Being a simulation is not an intrinsic property of one thing, or an exact thing, but a statement about its interpretation, relative to something else.

Most practical genetic algorithms do not simulate anything else, even if the process is sometimes misleadingly called "simulated evolution." They are not simulations at all, and what's going on in the computer is as real as anything else.

If that's not real evolution, what is it?

I think many people tend to think of computers as simulation tools, and think that what happens in a computer isn't real. They don't realize that for something to be a simulation, it also has to be a real thing of some sort; what goes on in a computer isn't imaginary, or intrinsically about something else. (No more than barnacles are about other kinds of parasites, even if Darwin chose to use them as that sort of model.)

You might say that what happens in genetic algoritms not real biological evolution by natural selection, but I don't think that should matter. For example, if somebody evolves microbes in a lab by rigging the selection factor, we don't say it isn't evolution. The theory of evolution clearly transcends whether the "natural selection" is literally and purely natural.

Likewise, I think it transcends whether the things being evolved are biological. For example, if we discovered complex replicating patterns of electron flux evolving by natural selection on the surface of a neutron star, we'd say that was evolution, even if the entities weren't "biological" in our earthly sense---they might have no cells, nothing corresponding directly to metabolism, etc. But if we saw the general pattern of mutation and selection of replicators leading to complexity, function, and diversity, we couldn't withold the term "evolution"---and wouldn't want to. (We'd say: see, it even happens to wildly different kinds of things in wildly different environments.)

Evolutionary theory is about what happens to replicators, given mutations and selection pressures.

Those things can happen to patterns inside computers, too, without their being a simulation of something else, so I don't understand why anybody would think that's not real evolution.

Time for an experiment. A test.

Get some horsies. then mechanically alter their germ line to give them voice boxes and improved versions of the neural structures connected to homo sapien sapien's Higher Powers. No mere copying of human genes! First principal DNA mechanics only!

Tinker away over the generations until the first principles are clear and the horsies are all acing the SAT's (in the special hanicappied administered test sessions, since the horsies haven't been given digits and opposable thumbs.)

Then sit back and have all the clowns explain how there is no mechanical basis for the horsies' souls and waxin on about Reason, citing authority that goes waaaaay back and is wrong as it ever was. Proving once again by geometric logic that what exists does not.

What will have been tested is the human capacity to twist and distort truth to whatever the inferior human is comfortable with. Mileage will vary. I predict.

I also predict Jesuits will prove to have the greatest capacity as a social (and therefore non-mechanical in the fools' view) gang of refuse to believe the truthers. But this is not for certain, what with fatwas here and there and Buddhists who evade the whole issue.

The bottom line: when a believers religion is threatened by evidence, the believers lash out to remove the threat by any means necessary to maintian their self acknowledged supremacy based on their belief. Leave God out of it.

I know all y'all are into defending science and stuff, and perhaps LW gave DD an uncharitable read, but it seems to me that LW's review can be given a more charitable read than what I'm seeing here. Set DD and the great promise of scientific research aside for a minute and consider LW's main points:

(1) The hypothesis that everything can be explained by natural science is just that, a hypothesis, and
(2) to assume that that this hypothesis is true is not to make a scientific assumption but a philosophical one, and
(3) to assume that everything can be explained by natural science is ultimately self-refuting.

I've yet to see any of these points really addressed here.

In other words, as I read LW's review it is not about DD or the details of his particular philosophical project (we're talking about the NYT, here, not the APA), but the general worldview DD's book evokes. That worldview (not inappropriately called "scientism" or "scientific reductionism" to a broadly educated audience) is self-refuting.

We've seen persistent but self-refuting theories before. Logical positivism was one -- if it were true it would have been an exception to the logical positivist rule that all true statements are empirically verifiable. Positivism died, and LW is pointing out that scientific reductionism will die too--only, for whatever reasons (prejudice? knee-jerk protection of one's intellectual idols?) it sometimes takes a long time for incoherent theories to go away.

By Philosophy Prof (not verified) on 20 Feb 2006 #permalink

...The hypothesis that everything can be explained by natural science is just that, a hypothesis...

Natural science refers to everything that has effects. Unless you know of something that can explain stuff without dealing with effects, I don't see a problem.

John Cornwell is utterly bewildering here:

"Religion persists not because it is a meme, but because it is evidently human to believe in something beyond what one can perceive, just as it is human to dance, to make music, paint pictures, and tell stories."

Whatever you may think of the term "meme", this sentence gets things completely backwards! Under Dennett's usage this sentence could be rewritten like this:

"Religion persists because it is a meme: it is evidently human to believe in something beyond what one can perceive, just as it is human to dance, to make music, paint pictures, and tell stories."

..because being infected with memes (or "passing on ideas" or whatever other meme-neutral term you wish to use) is precisely what we humans have going for us. Cornwell thinks his usage of "it is human to X" somehow means something different than Dennett's usage of "X is a meme" -- when in fact they mean the same thing!

r says: "when a believers religion is threatened by evidence, the believers lash out to remove the threat by any means necessary to maintian their self acknowledged supremacy based on their belief"

No doubt LW would agree. He's pointing out that scientism is a religion, and the reaction of certain of its believers on this site might confirm r's point.

By Philososphy Prof (not verified) on 20 Feb 2006 #permalink

(1) The hypothesis that everything can be explained by natural science is just that, a hypothesis, and
(2) to assume that that this hypothesis is true is not to make a scientific assumption but a philosophical one, and
(3) to assume that everything can be explained by natural science is ultimately self-refuting.
I've yet to see any of these points really addressed here.

1) The word Wieseltier used was superstition. Maybe you could justify his use of that word. Besides, since this idea isn't testable in the scientific sense, isn't it more of a philosophy than a hypothesis?

3) I don't get it. I'd rather you made your point before I tried to respond to it.

"Natural science refers to everything that has effects. Unless you know of something that can explain stuff without dealing with effects, I don't see a problem."

I suppose, our language being conventional, you could define natural science so that it is the same thing as philosophy generally. But most people think natural science has as its object everything natural (physical) and their natural effects. Natural science doesn't include logic (concepts in and conceptual effects) and it doesn't include mathematics (mathematical objects and their effects) and it doesn't include metaphysics (beings whether physical or not, and their effects). Of course natural science can and should draw on logic, mathematics, and metaphysics in appropriate ways, the way optics draws on geometry, biology draws on chemistry, etc. But you can't define natural science as if it were the science of everything -- that's just the mistake that LW points out. And I don't see what any scientist would have to gain by it.

By Philosophy Prof (not verified) on 20 Feb 2006 #permalink

wamba: That everything is testable by natural science is a hypothesis. If a hypothesis is maintained irrationally, it is a superstition. Many people who believe that everything is testable by natural science believe so superstitiously -- they wouldn't give it up regardless of evidence (like rational arguments). Presumably it is possible to believe the hypothesis non-superstitiously -- some of my college students might qualify.

You are right that it is a hypothesis that cannot be empirically tested--which is what makes it a philosophical position, not a scientific one. This is the distinction with which LW begins his review.

Hypotheses that cannot be empirically tested still may be susceptible to proof or disproof. Take the Pythagorean Theorem or Fermat's Last Theorem.

By Philosophy Prof (not verified) on 20 Feb 2006 #permalink

Philosophy Prof, your point (2) is certainly correct; it is a philosophical assumption. But Dennett's whole project is to work out the philosophical implications of scientific realism; he's very explicit about that. Also, you are relying on the phrase "everything can be explained" in (1) for too much: can the rules of baseball be explained by natural science? In principle, sure; but in practice, no. Dennett has called this kind of demand "greedy reductionism", and it is mostly used as a caricature against science. Scientific reductionism does not mean that "everything" can be explained in terms of simple physics; it does mean that you can explain phenomena at one level in terms of interactions of phenomena at a lower level, and at no level do you need to introduce anything miraculous.

I do not get your charge of self-refutation, either. Science is not a set of statements that includes one like "Science is correct". It is just a method that makes one assumption: that the universe is regular; i.e., that future will be like the past. This is just the assumption that we call "being rational". If you have a novel criticism of this assumption (i.e., one that Hume didn't make), I would like to hear it.

Dennett's methodological naturalism may be nuanced, and LW may have missed the nuances, but they seem irrelevant to LW's main point. If science is just a method, "natural science" is that method applied to the natural world. To assume that all things are a part of the natural word, and can be explained according to the (natural) scientific method, is a philosophical assumption, and one warranted by neither empirical evidence or rational argument.

Hume is on point: the method of science presumes metaphysical commitments. Hume argued that these metaphysical commitments are neither grounded in reason alone nor in empirical observations. So, when science claims to know things that in fact it assumes (e.g. the regularity of the universe), it is actually going beyond reason. (What is strictly irrational may still be true, it just can't be proven by theoretical argument or empirical observation.) When science claims to know certain other things that it actually only assumes (e.g. that there is no real non-physical thing) it is departing even farther from reason.

Critics of scientism can still be "rational" as you define it. The universe is regular, the future will be like the past -- but none of this means that we can give a physical explaination of everything.

By Philosophy Prof (not verified) on 20 Feb 2006 #permalink

Philosophy Professor equals cheater?

Where is the "evidence" that has refuted science?

I offer the horsie experiment that is designed to have the strongest possible cause and effect significance. The experiment stipulates it will only use mechanical devices, and mortals mechanically derived and empirically verified knowledge - to create beings with reason. No spirit will be breathed into the creature.

Every informed person today know that this experiment is doable, and, probably, unfortunately, inevitable.

Who is threatened by the results? Not science. Science is what follows where results lead.

Those with vested interests are threatened. The day's Coperinicans, the earth is five thousand years old crowd, the genesis fairy tale is real crowd. Now the losers hide behind Reason and cherry pick their metaphysics.

Cheaters consistently and necessarily confuse the questions of whether we know (and we do know all mental activity in homo sapiens has a mechanical and emotional basis) and how knowing changes us (no one can say with authority though many claim to, so the answer we want is inaccessible to us) and whether humans with the day's knowledge can be trusted as decision makers (they clearly cannot as has been well established many ways including the persistent cheating by the over educated).

But on dog that didn't bark is the god that didn't show itself. In over four hundred years of science, much has been determined, but the omniscient omnipotent has not barked. It has neglected to allow itself to be be revealed by mortal means. Who would have guessed four hundred years ago?

Science is science. It goes where it goes. Cheaters equate the pursuit of science with religion. Science would be happy to confirm a divine being. Most scienctests have been believers. Many still are. Only an evil person would create a false catagory and round up and cram decent people into it just to suit their own twisted purposes.

Science has not ended up where it now is because of some agenda of preceonceived beliefs. Science has followed. Following leads to

Philosophy Professor equals cheater?

Where is the "evidence" that has refuted science?

I offer the horsie experiment that is designed to have the strongest possible cause and effect significance. The experiment stipulates it will only use mechanical devices, and mortals mechanically derived and empirically verified knowledge - to create beings with reason. No spirit will be breathed into the creature.

Every informed person today know that this experiment is doable, and, probably, unfortunately, inevitable.

Who is threatened by the results? Not science. Science is what follows where results lead.

Those with vested interests are threatened. The day's Coperinicans, the earth is five thousand years old crowd, the genesis fairy tale is real crowd. Now the losers hide behind Reason and cherry pick their metaphysics.

Cheaters consistently and necessarily confuse the questions of whether we know (and we do know all mental activity in homo sapiens has a mechanical and emotional basis) and how knowing changes us (no one can say with authority though many claim to, so the answer we want is inaccessible to us) and whether humans with the day's knowledge can be trusted as decision makers (they clearly cannot as has been well established many ways including the persistent cheating by the over educated).

But on dog that didn't bark is the god that didn't show itself. In over four hundred years of science, much has been determined, but the omniscient omnipotent has not barked. It has neglected to allow itself to be be revealed by mortal means. Who would have guessed four hundred years ago?

Science is science. It goes where it goes. Cheaters equate the pursuit of science with religion. Science would be happy to confirm a divine being. Most scienctests have been believers. Many still are. Only an evil person would create a false catagory and round up and cram decent people into it just to suit their own twisted purposes.

Science has not ended up where it now is because of some agenda of preceonceived beliefs. Science has followed. Following leads to

By the way, another strength of LW's review is that it criticizes scientism not from the point of view of religion or faith (though you might not be able to tell from the comments here) but from the point of view of humanism. His main charge is that DD's worldview calls into question not belief in God, but meaningful belief period.

That in mind, can someone please tell me what non-question-begging empirical evidence can be aduced for the thesis that a given person's belief is a physical event in that persons brain?

As far as Im aware, natural science could at best establish a correlation between a persons entertaining a belief and a kind of physical event in that persons brain. But since when is correlation identity? Correlation isnt even causation. How could one empirically show that the physical event causes the belief and not the other way around?

By Philosophy Prof (not verified) on 20 Feb 2006 #permalink

r is upset, which may or may not confirm my point about lashing out.

I never said science was refuted. I said scientism was refuted. It is inconsistent to believe that every truth can be empirically tested, that everything can be explained by natural science. Read a good obituary of positivism. But don't worry, science survived the death of positivism, and it will survive the death of scientism or reductionist materialism or whatever you want to call it.

By Philosophy Prof (not verified) on 20 Feb 2006 #permalink

Not sure if it answers any questions, but I do seem to recall something about an experiment that could trigger "religious experiences" electrically. Technically, it doesn't prove that those experiences are caused by the trigger, but so far, it has failed to disprove, which is how science works: We can never have absolute proof, but each failure at disproof builds confidence.

Not sure if it answers any questions, but I do seem to recall something about an experiment that could trigger "religious experiences" electrically. Technically, it doesn't prove that those experiences are caused by the trigger, but so far, it has failed to disprove, which is how science works: We can never have absolute proof, but each failure at disproof builds confidence.

Philosophy Prof,

It seems a bit disignenuous on your part, or maybe just the NYT reviewer's, to cast this in terms of whether "natural science" can explain everything. Dennett's not that "scientistic," even if he doesn't thing supernaturalism correctly explains much of anything.

So far as I know, that's not what the controversy is about.

It's really about whether naturalistic philosophy can do better than religious philosophy, and explain some important things that are widely believed to be immune to scientific study or naturalistic explanations---such as consciousness and morality.

Dennett doesn't think that giving a naturalistic etiology of such things is sufficient to refute them; he doesn't think that having science means you don't have to do philosophy.

For example, if he's roughly right about people having what amounts to a hyperactive deity detector, that undermines certain influential arguments for the existence of god(s). It doesn't prove that there's no god, but it does explain why people would think so, even if there wasn't---and that's got to be important in weeding out the good arguments from the bad ones.

Likewise he can show that many aspects of consciousness are quite explainable in information-processing and evolutionary terms. Even if that doesn't prove we don't have souls, it surely undercuts many traditional arguments for non-physical souls.
(And even if Dennett's particular theory isn't quite right, which I think it's not in some "details," it still serves as a "don't care theory," which shows that naturalistic explanations can go a lot further than is widely supposed.)

Many such arguments, taken together, substantially shift the burden of proof in the conflict between naturalism and supernaturalism. Naturalism is better at explaining many interesting things than is widely supposed, and supernaturalism doesn't clearly to explain much that's real, or really explain it correctly.

BTW, I don't know why you bring up positivism, or exactly what you mean by "positivism." Near as I can tell, Dennett isn't a positivist, at least not in a strong sense that supports any simple rebuttals of his philosophy.

PZ, have you read the book? That might be helpful.

Dennett is clearly on the right track but does a really, really bad job of presenting the argument.

Essentially the entire book can be summed up in one sentence: "Religion is really stupid, and people just believe in it for historical reasons, and everyone should stop believing in it right now, because I'm really smart and I say so."

I'm no defender of religion, quite the contrary. But that shouldn't prevent me or you from criticizing people who are sloppy in their support of "our" side or the other. Their contribution is not useful in the overall context of things.

Dennett spends the first chapters arguing for a scientific approach to religion, but then presents *no* scientific evidence in the rest of the book. Instead, we are treated to hundreds of pages of him rambling on and on, basically just making up a story, apparently imagining himself to be so gifted that he doesn't need to actually validate any of his arguments. But then again, that shouldn't surprise us, since that's what philosophers do for a living.

--
Bob Myers

I don't read LW as defending "supernaturalism" (whatever that is). And you all may be right that Dennett himself is much more moderate in his own naturalism than LW perceives. Fine. But LW's main point is that a worldview that holds that every question is ultimately a question for natural science to answer is both irrational and inhumane. (With all due respect to the egos of authors, it is irrelevant whether Dennett actually holds this view. LW has judged, I think rightly, what the popular reception of his book's thesis will be. And, for that matter, whatever the subtleties of methodological naturalism that philosophers of science may appreciate, Dennett seems to relish the reductionist label -- look at the interview of Dennett in the NYT Magazine several weeks ago, where he embraced the charge of "Darwinian fundamentalist.")

I bring up logical positivism -- which stated that statements are either empirically verifiable or meaningless -- as another example of a theory of everything that can't account for itself.

Again, what is at stake for LW is not God or religion or the supernatural, but whether our theories make any sense in real life -- he is defending the ground of "humanism" and "rationalism", not theology.

By Anonymous (not verified) on 20 Feb 2006 #permalink

Many people who believe that everything is testable by natural science believe so superstitiously -- they wouldn't give it up regardless of evidence (like rational arguments).

I dare say you have never run that experiment; or are you claiming to know of evidence for the existence of something supernatural? According to you, that makes your statement superstitious.

Bob wrote "But then again, that shouldn't surprise us, since that's what philosophers do for a living."

Overgeneralization, anyone?

Still, your points about Dennett are fine. But why overgeneralize? Sounds like you could use a course in logic. Guess what? Logic is taught in philosophy departments.

By A fan of logic (not verified) on 20 Feb 2006 #permalink

wamba: On superstitious scientism, I do speak from experience, but if its public or reproducible data you want, I think it's been generated over the past day on this thread. LW has given rational arguments about why not everything is testable by natural science. For some reason, his arguments have occasioned more attacks on his character than critical self-examination among the principled naturalists on this site.

By Philosophy Prof (not verified) on 20 Feb 2006 #permalink

I think we can say this much: Everything natural is testable by natural science. Of course, the definition of natural doesn't leave much ground for the supernatural: Anything that has an effect is, by definition, natural.

"Anything that has an effect is, by definition, natural."

False.

Even if some specialized materialistic philosophers work with this definition, it is not the definition that normal people have of "natural" or of "cause" and "effect". A supernatural cause is not a contradiction in terms. Whether or not there are such things as supernatural causes (or effects) is another matter, but at least one can't determine that there aren't by building it into your definition of "natural."

Just speaking of the meanings of terms:

"being" = "material being"
"cause" = "physical cause"
"natural" = "what is real"

Don't try testing this in a lab. You may want to try using a dictionary though.

By Philosophy Prof (not verified) on 20 Feb 2006 #permalink

Even if some specialized materialistic philosophers work with this definition...

These people are known as "scientists." The work they do with that definition is called "science."

...it is not the definition that normal people have of "natural" or of "cause" and "effect".

Do you mean that, like in the sense that normal people refer to "theory" as meaning a wild guess?

Whether or not there are such things as supernatural causes (or effects) is another matter, but at least one can't determine that there aren't by building it into your definition of "natural."

My definition renders the question meaningless. There's no line in the sand to determine what effects are "supernatural" or "natural" under the "normal" definitions. If it has an effect, it can be studied by science.

It's really about whether naturalistic philosophy can do better than religious philosophy, and explain some important things that are widely believed to be immune to scientific study or naturalistic explanations---such as consciousness and morality.

I certainly hope that its not a choice between "naturalistic" philosophy and "religious" philosophy; and this is a large part of why I was more than a little bit concerned at both PZ's and Leiter's reviews. On further reflection last night, I was less convinced that LW was occupying a "humanist" point of view (as opposed to a theist one), but, generally speaking, I worry that, in some people's minds, the only alternative to a natural-scientific account of something is a supernatural/religious account. And I just think that clouds over a very rich philosophical tradition that has aimed to situate itself between those two extremes, by giving non-dualist, non-theist, non-naturalist accounts of morality, reason, &c. What the insights of that tradition come down to, I think, is roughly that while we perfectly well can give scientific accounts of whatsoever we want to, there are some phenomena (beliefs, actions, attitudes) of which we demand certain kinds of explanation that in turn outstrip the conceptual resources of natural science. Or something like that; I'm pretty terrible at expressing the core insight of anything, but the point being, denying expansive philosophical naturalism doesn't always amount to substance dualism, theism, etc. At least... not in the minds of self-identified non-naturalists like myself.

Hope I'm not misreading the dimensions of debate. I'm running on very little sleep at this point, so making sense of anything is a significant struggle (not that it isn't for me normally!).

By T.W. McKinney (not verified) on 20 Feb 2006 #permalink

As far as I'm concerned, "naturalist", "religious" and "humanist" arguments are irrelevant here, and I'm not taking sides on that (although you can guess where I'd fall if I were): I just found Wieseltier's argument abominably incoherent and contradictory, and would say so whatever side he was arguing for.

I'll also mention that I am not a fan of Dennett. I didn't care for Darwin's Dangerous Idea at all, and would like to see Dennett and Ruse locked together in a room somewhere far, far away where they blather away at each other without confusing the rest of the world.

I don't read LW as defending "supernaturalism" (whatever that is). And you all may be right that Dennett himself is much more moderate in his own naturalism than LW perceives. Fine.

Not fine. If LW is mischaracterizing Dennett, it's not fine; that's not a book review; it's a hatchet job.

And Dennett is very strongly naturalistic; if that's what LW and you really want to criticize, go for it.

If you want to say that naturalism per se is self-refuting, why beat around the bush? Or if it's reductionism, say that.

Don't just assert that reductive materialism is self-refuting and will die, as though you'd know and we should take your word for it.

But LW's main point is that a worldview that holds that every question is ultimately a question for natural science to answer is both irrational and inhumane.

If that means what I think it means, it's false. Some explaining is in order, and you're not doing much better on that account than LW.

(With all due respect to the egos of authors, it is irrelevant whether Dennett actually holds this view. LW has judged, I think rightly, what the popular reception of his book's thesis will be.

Wow. I find it hard to believe this stuff is actually coming from a Philosophy Prof.

OK, from now on we can misrepresent the views of authors and assassinate their character, if we think their audiences will draw the wrong lessons from their books.

None of my philosophy profs would have let me get away with this kind of argument.

And, for that matter, whatever the subtleties of methodological naturalism that philosophers of science may appreciate, Dennett seems to relish the reductionist label -- look at the interview of Dennett in the NYT Magazine several weeks ago, where he embraced the charge of "Darwinian fundamentalist.")

I think there's a reason for that. "Reductionist" has become a code word for bashing materialism, without bothering to distinguish between greedy reduction, functionalism, etc. In other words, pretty much any actual explanation of anything can be slandered as "reductionist." If that's the way "the other side" is going to play it, I think it's reasonable for people like Dennett to say "fine, then, I'm a reductionist."

I'll go that way, too. If what you mean by "reductionism" is simply naturalism, I'm a reductionist, too. Two cheers for reductionism!

One of the ironies in this is that most of the people who take "reductionism" to be a bad thing are supernaturalists who are hyper-reductionists about the very same phenomena. E.g., we have consciousness because we have the part that just does consciousness, or we have morals because god gave a bit that just does morals, with no actual explanation.

And many others are simply mysterians, who assert, with no real evidence, that phenomon X just can't be explained in anything like the ways we explain other things, or can't be understood at all.

I bring up logical positivism -- which stated that statements are either empirically verifiable or meaningless -- as another example of a theory of everything that can't account for itself.

But you haven't justified the comparison. You've just asserted that reductive (whatever that means) materialism is self-refuting, without saying how. And you've made what seems like a bad analogy; I can understand verificationism being self-refuting. But Dennett isn't a verificationist, by any means, so I don't know what your point is.

Again, what is at stake for LW is not God or religion or the supernatural, but whether our theories make any sense in real life -- he is defending the ground of "humanism" and "rationalism", not theology.

I have no idea what you're talking about. You seem to be implying that scientific explanations don't make sense in real life.

All I get out of this is that you more or less agree with LW because you think he more or less agrees with your perception of the popular reception of Dennett's views, and you think reductive materialism is somehow self-refuting and harmful.

That's just not amounting to much.

several distinct concepts need to be distinguished: reality, our perception of reality, our models of reality, and the probabilistic correspondence of those models with reality.

I would bet that since roger penrose is an intellectual giant, he does that. presumably he is arguing that the behavior of the mandelbrot set is a latent "reality" that is only lacking instantiatation by an intelligent being (an abstract version of a tree falling unheard in the forest?). of course, while it seems clear what it means for a concrete reality like a sound to "exist" whether heard or not, one may question what it means for a latent reality to "exist" in the absence of instantiation.

[Note: having just revisited the comments and read Paul W's discourse on simulations, I realize I've made an assumption based on ignorance that may be untrue, viz, that mandelbrot sets don't manifest themselves in nature but only as computer-generated fractals. but if that assumption is incorrect, then I don't understand penrose's point since the natural manifestation would be physical existence.]

"mathematics is objective"

I think this is incorrect. I view mathematics as just human modeling. the well-known unequivocal "fact" that 2+2=4 is a human convention, no more "objective" (which I take to mean "true") than that "mary dog mary cat frank" which could be made true in an (extremely awkward) alternative model for integer counting. here too, such models are unnecessary in the absence of an intelligence needing to count, so it isn't clear to me that from the fact that such models arise when such an intelligence emerges, it logically follows that there is an underlying reality that transcends that intelligence.

"The laws of science are not socio-cultural phenomena"

I think they are, strictly speaking. we attempt to construct models (laws of science) that correspond well with observations thereby giving us high confidence in their predictive or explanatory value. but note that there's a lot of fuzz here: observations (equipment limitations, interpretive bias), correspond well (margins of error), confidence (probabilities). there may be an underlying reality (unsupported objects released above the ground routinely end up on the ground) but the "laws of science" describing how they get there are human constructs with only approximate and probabilistic correspondence to that reality.

biologically determined vs sociologically constructed

is this "free will vs determinism" in disguise? if so, the either-or modeling may be similarly wrong. quantum physics has suggested (to me, anyway) that a more realistic model for free will vs determinism is probabilistic and dependent on the information available. eg, model a person raising an arm when asked to raise either. in a model with no a priori information about the person, the probability of choosing the left arm should be 0.5, ie, unfettered free will. in a model with more a priori information (left or right handed, complete personal history, psychological state, etc.), the probability might be 0.99, pretty close to determinism, but not quite. and even if science is one day able to analyze "the entire process" in arun's neurons, due to quantum effects the probability will presumably never be 1. in this sense, free will or determinism is not a meaningful question. my guess (admittedly based on total ignorance of the relevant disciplines) is that biology vs sociology is analogous.

[Note: also from reading new comments, I see that Pete - unlike me - understood what "greedy reductionism" meant. if I now have it right, trying get the model fidelity necessary for 0.99 might be an example.]

"To assume that all things are a part of the natural wor[l]d ... is a philosophical assumption ... [not] warranted by empirical evidence"

but when does the "empirical evidence" supporting that assumption (or more to the point, the absence of evidence supporting the contrary) suggest that the probability that the assumption is true become high enough that for all practical purposes it can be assumed "true"?

in any event, the strawman implicit in the phrase "when science claims" is the scientist (or more generally, "rationalist") who positively denies the existence of supernatural phenomena, ie, has an unprovable faith-based (ie, religious) posture. however, my impression is that many, if not most, people who might label themselves "rationalist" (or something roughly equivalent) really have one of two postures vis-a-vis supernatural phenomena: they see no credible evidence to suggest that the probability that they exist is not vanishly small, or they have no interest in them at all. neither posture is faith-based, ie, religious. they are operating assumptions that have proven workable but would willingly (though maybe not happily) be abandoned given negating evidence.

"none of this means that we can give a physical explaination of everything"

you do, I assume, recognize that this is a version of the "argument of the gaps". ie, neither does the current failure to explain everything mean that we can't explain so much that what's left has little if any practical consequence. eg, why would anyone care whether or not there is a deistic god other than as a matter of curiosity?

...my impression is that many, if not most, people who might label themselves "rationalist" (or something roughly equivalent) really have one of two postures vis-a-vis supernatural phenomena: they see no credible evidence to suggest that the probability that they exist is not vanishly small, or they have no interest in them at all.

Just to make it clear for everyone: My stance is that "supernatural" is a nonsense word. I've never heard a good definition of "supernatural" outside of D&D (doesn't work in an antimagic field, usually doesn't provoke attacks of opportunity). Now, if someone wants to claim the existence of something (A designer, Uri Geller's ability to bend spoons with his mind, etc.), my type of naturalism doesn't care if something is "supernatural" by common definitions. If it has an effect, it can be studied by science.

P.P.:

But LW's main point is that a worldview that holds that every question is ultimately a question for natural science to answer is both irrational and inhumane.

The problem here is that no one, to my knowledge, holds that "every question" can be answered by natural science. Ethics, for example: while ethical questions can certainly be illuminated by a scientific understanding of human nature, any ethical answers could not be dictated by science, even in principle.

It seems to me that LW is made uneasy by materialism, because he feels there must be "something else" in order for life to have meaning. Despite thousands of years of people believing in "something else", no one has ever convincingly demonstrated it. All we see is our natural world.

T. W. M.:

while we perfectly well can give scientific accounts of whatsoever we want to, there are some phenomena (beliefs, actions, attitudes) of which we demand certain kinds of explanation that in turn outstrip the conceptual resources of natural science.

I agree, but I do not think this makes this kind of inquiry "non-natural". All these phenomena are still manifestly carried out by biological beings, equipped only with evolved biological machinery. The only reason science would be extremely strained to come up with answers to why we do some things, is that we are able to do arbitrary things.

Ultimately everything we do has a physical explanation, but to make sense of what we do is another story. Dennett, while taking atheism as a given, argues that our best chance of making sense of religion is to study our biology/psychology. I don't see the coherence in LW's response against this.

Does the current state of science justify Dennett's investigation of the issue he explores or is he beyond science's warrant and violating Humanities' and Religion's turf?

The reviewer says in part:
"He [Dennett] remarks that the question cui bono? who benefits? "is even more central in evolutionary biology than in the law," and so we must seek the biological utilities of what might otherwise seem like "a gratuitous outlay." An anxiety about the reality of nonbiological meanings troubles Dennett's every page. But it is very hard to envisage the biological utilities of such gratuitous outlays as "The Embarkation for Cythera" and Fermat's theorem and the "Missa Solemnis.""

However.
Science has excellent evidence that life on earth does not tolerate "gratuitous outlays." See that other atheist irk, Dawkins. So science is either wrong or wherever there is a gratuious outlay, there is some serious splaining to do.

Dennett has directly adressed one area where this scientific problem of fundamental biology is manifest, religion. Obviously all the religions aren't right. They are costly. So why are they all there? Could be God made them so. But how are humans made so that they spontaneously generate religions (mutually exclusive religions at that)?

A similarly fundamental scientific problem based on the current state of knowledge, that gets people upset so they refuse to see the issue since it invovles their actual lives and beliefs: homosexuality. Why is it there?

Dennett tries to explain what science demands an explanation of. Within the natural science tradition supernatural explanations are not acceptable.

No critic here has justified criticism of Dennett as pursuing what science cannot legitmately pursue. Nor does it seem any defenders are saying that Dennett has found The Truth.

Scientism, like secular humanism, is a scam invented and maintained so people with threatened vested interests can use the nouns as Just So explanations of complex divergent effects, while inconsistently announcing, e.g., "Correlation isnt even causation." And while correlation indeed need not be (but typically does signal)causation labeling isn't causation either and accusing Dennett of scientism or communism just evades the issues.

By Anonymous (not verified) on 20 Feb 2006 #permalink

Dennett has directly adressed one area where this scientific problem of fundamental biology is manifest, religion. Obviously all the religions aren't right. They are costly. So why are they all there?

It's useful for social purposes, which helps on biological purposes. The benefits are good enough to outweigh the costs.

homosexuality. Why is it there?

I don't know for sure, but one hypothesis I've heard (a decade ago, so my memory's probably distorted) is that there's some necessary genes for all us of one gender or another to be able to get along without getting too far into alpha male zone or whatever. Homosexuals are people who get a few too many.

Of course, to draw conclusions from a lack of knowledge of these things would be an argument ignorance and/or lack of imagination.

Within the natural science tradition supernatural explanations are not acceptable.

Because they aren't explanations. They're just giving up. Plus, as I said, "supernatural" is a nonsense word.

Scientism, like secular humanism, is a scam invented and maintained so people with threatened vested interests...

Propaganda tactic/logical fallacy: Appeal to motivation. Try staying on the topic of the arguments, not changing it to the arguers.

windy: You responded with an obvious fallacy. The point about "chimpanzees" was that no animals possess a religion, because they lack the thematic self-consciousness and discursive reasoning capacities to raise such questions, which are an effect of language. Further my basic point concerned appropriate levels of explanation, which is a rational-normative question and not simply an empirical matter.

John Emerson:

I have no problem with the thesis that human language emerged from natural evolutionary processes, presumably on the basis of a prior analog system of animal communication and that specific phenotypic "adaptions", (enlarged brain, descended larynx, etc.) were requisite for the emergence of symbolic, syntactically organized and semantically self-stabilized language. However, whatever the synergistic convergence of factors in its emergence, it does not follow that language is simply innate, (as opposed to species-specific), since it derives just as much from the exteriority of communications across the world as from the brain processes of its users. Further, to causally explain a capacity is obviously not to constitute a causal determinism of the exercise of that capacity, else that capacity is merely epiphenomenal rather than a genuine phenomenon, to be explained away. Once language emerges, it evolves together with the socio-cultural form of life in which it is embedded and which it mediates, and that evolution is not a case of natural selection rooted in genes. (Strange as it may seem to some,"meaning" is a non-causal relation, irreducible to material processes, although there are other ways in which it can be gotten at and "explained" non-reductively and criticized at its "roots": e.g. cf. Wittgenstein.) And while I have no problem per se with the notion of a biological substrate to human social life, the socio-cultural form of human life does at the very least add on new "problem sets" (or conflicts and questions) that modify the expression of that biological substrate. (Actually, I would hold to Arnold Gehlen's thesis that human beings are born organically deficient, that is, lacking in adequate functional specification of instincts due to the natural evolution of "excessive" instinctual "plasticity", such that they require cultural structuration of their biological/instinctual potentials in order to acquire the behavioral adequacy and adaptive functionality that is pre-given, in large degree, for animals by their instinctual endowments.) At any rate, the main point here is that there are limits to the explanatory adequacy of naturalistic explanation in terms of efficient material causes that render efforts to totalize such explanations arguably, (i.e. normative-rationally), fallacious.

jbark: There is no "dualism" showing up in my account, unless it be opposition to a crude, dogmatic and reductive monism. The only "dualism" appealed to is the (empirically well-warranted, I take it) one between language and reality and between the counterfacctual status of norms, which are anchored in the recognitions and agreements between language users and the empiricity of facts, which are agreed upon through the differentiation of normative frameworks.

Paul W.:

The wisecrack about "pickup line" was that "algorythmic process" is an awfully odd way to refer to biological reproduction. Not just odd though, but specifically wrong. An algorithm is a fixed reiterable rule, which breaks down a more complex intractable calculation into more managable subroutines. It's a term from mathematics that obviously applies as well to computer science. Now I think I get what Dennett wants to say by invoking the term. He wants to follow up on Dawkins' gene selection account of Darwinian evolution, whereby genes are the main, if not sole, objects of natural selection. Now there are significant problems with Dawkins' claims,- (see Prof. Myers post on 2/18/06 at 8:25 pm below to get some idea of that),- but the idea seems to be that the algorithms operate on sets of genes and their rules of recombination. However, in the long-run of evolution by natural selection, there are no pre-existent rules, nor is there any fixed combinatorial of genes: any "rules" or extension/addition/alteration of genes emerge only from natural selection, and there is no ready generalization of such as can be empirically found and substantiated. (That's an empirical question, not a matter to be dictated by high theory.) But still more fundamentally, Dennett claims to be giving an account of Darwinian theory, that is, evolution by natural selection, when in fact he is not giving a strictly selectionist account at all, but rather an instructionalist account, based ultimately on the model of digital computational devices, (or, similarly, computationalist cognitive-functionalist theories in "cognitive psychology", that derive from formal analysis and then are somehow claimed to be embedded in material processes). An instructionalist account involves fixed predefined functions, which then operate upon information that is predefined as transparent to those functions and their operations. But in natural reality per se, there are no predefined functions, nor any unambiguous environmental information or signals. A properly selectionist explanation involves contingent selection events among statistically variant populations that redistribute or re-enforce the distribution of probabilities. (The conceptual and formal means did not fully exist in Darwin's time, but I think his account is basically a stochastic account, in terms of accounting for the probability of the improbable.) The basic point here is that such contigent selections themselves are what generate the information that brings about the covariation between organisms and their metabolic processes and their environments, which information would not otherwise "exist". And that is the source of the tremendous explanatory resources of Darwinian, selectionist explanations. Computers are exceedingly useful tools, but make for bad metaphors, and, though formal modelling is indispensible, their applications to empirical data is a procedural question that requires independent justification.

In general, the anthropological reduction and ad hominem criticism of religion has been around in full dress form since Feuerbach, and probably was partially and variously anticipated before that. Now ad hominem argument was classically classified among the fallacies, but I think it has a legitimate rational use as an obviatory procedure in the service of some larger, independently established truth. And I think Feuerbachian style criticism of religion can go about 99% of the way toward achieving its aims. But it's that last 1% that prevents its absolutization/totalization, which in a plural world is probably a good thing. But that last 1% will not submit to being bludgeoned by irrational passions claiming to derive from "logic" or "science". It's not Dennett's materialist "reductionism" that bothers me and which, at any rate, is entirely unsurprising, not the least bit shocking. It's his instrumentalist reductionism that I would object to, as not remotely adequate to the questions involved in religion, and, for that matter, to an account of the rationality of science. One shouldn't let an anti-religious animus blind one to the crudities and dogmatisms of his account, when others have already done a much better job of it.

By john c. halasz (not verified) on 20 Feb 2006 #permalink

john c. halasz:
You responded with an obvious fallacy

No, I was pointing out your fallacy. If you didn't intend the chimp part in "It should be perfectly obvious that x is not a natural phenomenon; chimpanzees don't have x" as support for your argument, then I'm sorry, but in that case you should phrase it some other way. And ease up on the "obvious", because it's not.

And why do you think the following are 'obviously' not 'natural phenomena':

Religion is a response to a problem set that only emerges with the linguistic reduplication of the world and the insistency of symbolic thinking, together with the integration and transmission of social orders.

Seems to me those things most likely have at least a very strong biological basis, even if they may or may not have been naturally selected in themselves. It's obvious to me, at least! ;-)

I hate the word "obvious." If something were truly obvious, it wouldn't need stating... And then, it'd probably need to be challenged, anyway.

Re:
"Propaganda tactic/logical fallacy: Appeal to motivation. Try staying on the topic of the arguments, not changing it to the arguers."

Bronze Dog. Thanks for the instruction but I'll stick to my position that bias and motive of the source are the most valuable information one can have in evaluating the value of the source. Corrupt data, double binds and all that.

If perchance you ever have a jury or judge decide your welfare, be sure to invite me so I can enjoy the delicious pleasure of hearing your lawyer openly reject all inquiry into motive and character of the judge, jury and your accusers, as logical fallacies to which she will not stoop on your behalf, even as the same are used against you. Realizing the man who will decide your fate is out to get you is something the uneducated grasp easily. The stuff of comedy in fact.

A war is being fought against the enlightenment. One ground is "scientism." The Dennett review was an attack in this war. But as you pefer, wake up or sleep on.

As to grappling with the issues, Mr. Halasz is as close as I expect an academic to come. While the terms of art he uses are simply not up to the task at hand and lack the certain power presumed therein, the faith they are is, from an uninvolved distance, touching. If we have but world enough and time enough it will all work out.

By Anonymous (not verified) on 20 Feb 2006 #permalink

I think the root of this goes to a denial of the impact of subjective awareness, a denial that goes deeper than what some would call simple 'anti-freudianism' to become a form of anti-humanism. If we treat people as machines then we are under no obligation to be more than -less predictable than- machines. As in Law and Economics, we are judged by our lowest common denominators. And at the same time, much as modernists would claim to be outside of history because they understood its processes, scientism proposes a similar privileged position for its own faithful. But what behaviorist would argue that being a behaviorist renders one immune from conditioned response?

Consciousness is the spontaneous product of the conflict between a conditioned response and a clearly exceptional event. When what had previously been, in the experience of an organism, a reflex, an automatic mechanical activity, is challenged by an observation, made possible by a capacity for objective computation and reason, the resulting INDECISION is the first act of consciousness. Consciousness is not reason, it's the conflict between reason and conditioned response, or the memory of past experience. All Consciousness is flawed consciousness. It is neurosis.

Build a neurotic computer and you'll have artificial intelligence. Exclude subjective experience and you're unable to explain the processess of the animal mind.
Dennett the menace doesn't get it.
Fuck'em

windy:

I wasn't writing a treatise in my first comment. I later emended that omission. But yours was an uncharitable/incompetent misinterpretation, bad hermeneutic form. Sorry. To put it in syllogistic form: only language-bearing beings are capable of the thematic consciousness of the world required to raise religious or mythic questions or problems: chimpanzees- (choosen because they are the cognate species, but any other species would do),- don't have a language; therefore chimpanzees don't have religion. Your counterproposals were fallacious since they fail to establish a major premise; e.g. blonds have fun; i'm not blond; therefore I don't have fun. But it's not merely a matter of logic; what's obvious is that there is no empirical evidence for religious behaviors- (say, burial of the dead, which is not a human universal)- among other animal species.

As for your unargued claim (see above paragraph) that language and socio-cultural forms have a "strong biological basis", if that means that they have a physical, biological substrate, as a necessary condition, I've got no problem with that; if that means that that is a sufficient condition for their explanation, I do. In general, I'm suspicious of large monolithic explanations, whether it concerns the self-development of Absolute Spirit or the explanation of human cultures and societies as the expression of a pre-determined, evolutionarily pre-adapted human biological nature. "Natural phenomenon", by the way, in this context means "explainable through efficient material causal mechanisms alone." The emergence of human natural language from prior natural history is a difficult and tickish question, the subject of current debate and research. I myself would emphasize the social-relational roots of language,- (look up "analog communication", to see why)- rather than instrumentally adaptive accounts, though the two are not mutually exclusive and could well have fed back into each other.But claims that language is simply a pre-defined function based on biological pre-adaptions strike me as partaking of the "instructionalist" fallacy I mentioned above.

anonymous:

There is no "war on the Enlightenment". There are various reasonable criticisms and arguments exchanged publicly from various different positions or points-of-view. That was supposed to be what the Enlightenment was all about.

By john c. halasz (not verified) on 20 Feb 2006 #permalink

"There is no empirical evidence of non-physical reality, so therefore there is no non-physical reality." This is a fallacy, used to demonstrate circular arguments to intro philosophy students. The suppressed premise is, "if there is no empirical evidence of something, it doesn't exist" -- but this would only be true if we accept the conclusion of the argument, that there is no non-physical reality. By definition, non-physical causes and effects cannot be detected empirically. A blind person can't argue that there is no such thing as colors just because he can't see them; likewise, a materialist can't argue that there is no such thing as non-physical reality just because he makes appearance in the physical world a precondition of his admitting that it exists.

LW isn't fighting for God or scary religion, he's fighting for the fact that some things -- ideas, minds, love, rational argument -- don't have mass or show up under microscopes.

By Philosophy Prof (not verified) on 20 Feb 2006 #permalink

I read Wieseltier's argument a second time and tried to read it more sympathetically, to figure out if there could be some more coherent interpretation of what he's saying.

It occurs to me that the contradiction that PZ immediately noticed, in which Wieseltier denounces the attacking of ideas based on origins and then immediately puts down "creaturely reason", might actually be not a contradiction on Wieseltier's part but a very badly phrased attempt to demonstrate what Wieseltier sees as another internal contradiction of *Dennett's*.

It would go something like this: Dennett tries to banish religion by showing that it has a natural, biological origin; but Dennett also thinks reason has a natural, biological origin; therefore, in order to be *consistent* with his disbelief in religion, Dennett must also disbelieve in reason. The devaluation of "creaturely" reason would then be not a devaluation on Wieseltier's part but an implied logical consequence of Dennett's attitudes.

If that's what he's driving at, though, I don't think it works. I'm pretty sure that Dennett believes in the possibility of strong AI; he thinks that machines could in principle reason. Therefore, manifestly, Dennett does not see any contradiction between reason and mechanism. But Dennett may well see a contradiction between divine inspiration of religious thought and mechanism; if the brain is a reasoning mechanism then direct communion with God would be a kind of neural miracle, a violation of the laws of nature of the sort Dennett doesn't believe in. The validity of either of these views would be a whole separate discussion, but I don't see any logical incompatibility between these two views. So if that *is* the argument, I think it still doesn't stand.

The point is that the natural biological origin of religion or of reason does not logically say anything about the validity of either.

Dennett's natural history does not deny reason, it animalizes reason. It portrays reason in service to natural selection, and as a product of natural selection. But if reason is a product of natural selection, then how much confidence can we have in a rational argument for natural selection? The power of reason is owed to the independence of reason and to nothing else. (In this respect, reason is closer to mysticism than it is to materialism.) Evolutionary biology cannot invoke the power of reason even as it destroys it.

But if reason is a product of natural selection, then how much confidence can we have in a rational argument for natural selection?

Problem: Natural selection favors a (generally) reasonable mind because of reason's validity. If it weren't valid, it'd be useless or even dangerous and, therefore, selected against.

See, though, in that first passage Arun quoted it sounds as if Wieseltier is actually stating something he believes instead of inferring it as a logical consequence of Dennett's beliefs; he seems to think that attributing reason to natural selection really does lessen one's confidence in it.

Which reminds me very much of C. S. Lewis's argument for the existence of the supernatural, that (I am probably grossly oversimplifying here) if our reason is just the working of natural forces then we have no reason to trust that our reason is saying anything true or important, so only if reason has a supernatural origin can we take it seriously.

I tend to side more with BronzeDog's idea that the validity of reason is what made it adaptive, so that natural selection is actually what saves mechanism from Lewis's attack. But of course one could equally argue (though I've never heard of anyone doing so) that Darwinian natural selection favored religion because God is real and smites people who don't worship him (though his taste in worship seems curiously eclectic).

BronzeDog:

Huh? Natural selection favors the covariance of statistical regularities in its environment and the processes of a path-dependent organism. "Reason" has nothing to do with it: History 101. The belief in the pre-existence of reason is a metaphysical belief par excellence.

By john c. halasz (not verified) on 20 Feb 2006 #permalink

Matt McIrvin:

Huh? Norms are not derivable from causal explanations, but does that mean that they are thereby "supernatural"? And the question at stake was whether scientific explanations themselves can be derived recursively from scientific explanations, or rather, whether the possibility of scientific explanations does not derive from historically established norms of rationality. In fact, I think, that natural scientists themselves, in the course of their investigations, establish rational norms in their respective disciplines, qua cannons of explanation, that are neither empirical, nor immune from empirical considerations and possibilities.

By john c. halasz (not verified) on 20 Feb 2006 #permalink

only language-bearing beings are capable of the thematic consciousness of the world required to raise religious or mythic questions or problems: chimpanzees- (choosen because they are the cognate species, but any other species would do),- don't have a language; therefore chimpanzees don't have religion.

OK. But the same applies to grammar. Chimps don't have language, and therefore no grammar. Language is a prerequisite for grammar, and grammar is largely culturally determined, but that doesn't mean grammar is exclusively socio-cultural.

It seems that ability for grammar has been selected for after language arose. Personally I don't think religious behavior has been specifically selected for, but that suggestion can't be 'obviously' dismissed.

But it's not merely a matter of logic; what's obvious is that there is no empirical evidence for religious behaviors-(say, burial of the dead, which is not a human universal)- among other animal species.

Now you bring animals into the mix again - that only humans do something doesn't prove that something is not biological. (Granted, burial of the dead isn't)

I myself would emphasize the social-relational roots of language,- (look up "analog communication", to see why)- rather than instrumentally adaptive accounts, though the two are not mutually exclusive and could well have fed back into each other.But claims that language is simply a pre-defined function based on biological pre-adaptions strike me as partaking of the "instructionalist" fallacy I mentioned above.

Ah -if you believe that people are saying that, then I see why you have a problem. But nobody claims that language and religion are simply pre-defined biological functions. Feedback between cultural and biological processes, like you propose, seems most likely. There is simply disagreement on the amount of feedback. Happy?

windy:

I have heretical views on language formation. Roughly,I think that syntactical organization devolves from embedded structures of action and interaction. The upshot would be that syntax, as the marker for language formation, would not be, as in the "classical" Chomskyan account, innate, but rather "enate", that is, transfered from the organism/environment relation onto the "plane" of communicative interaction. Semantic abilities, which would involve the neural re-categorization of perceptual categorizations, would already have been developed, but would have been further re-categorized syntactically though the further effects of communicative interaction. That's a very crude, rough-sketch account of the emergence of language, but the upshot is that it can ilustrate the notion that biological capacities can give rise to extra-biological "realities", which they subtend without reducing to their determination.

By john c. halasz (not verified) on 20 Feb 2006 #permalink

"mathematics is objective"

I think this is incorrect. I view mathematics as just human modeling. the well-known unequivocal "fact" that 2+2=4 is a human convention, no more "objective" (which I take to mean "true") than that "mary dog mary cat frank" which could be made true in an (extremely awkward) alternative model for integer counting.

Mathematics is about the truth about integers that lies behind "2+2=4" and "mary dog mary cat frank", and not about specific conventions.

Bronzedog wrote:

Problem: Natural selection favors a (generally) reasonable mind because of reason's validity. If it weren't valid, it'd be useless or even dangerous and, therefore, selected against.

So now you get the reviewer's point. If Dennett has a story about the biological origin of religion, should we say "natural selection favors a (generally) religious mind because of religion's validity. If it weren't valid, it'd be useless or even dangerous and, therefore selected against"?

In general, natural selection and evolution cannot tell us anything about the validity of our ideas (or why do physics, just see if natural selection selects for brains that believe in string theory or not)? The validity or invalidity of religion is to be determined by reasoning about it, not in spinning stories about its origin. Since Dennett does precisely the latter, the reviewer charges Dennett with not believing in reason.

Either the reviewer is right, or is easily refuted by showing that that is not what Dennett does. Instead we degenerate into a "vegetable" and "mineral" reasoning rant, which I attribute to too much interaction with IDists. Associate with people (even those you oppose) too much and you take on their characteristics.

So now you get the reviewer's point. If Dennett has a story about the biological origin of religion, should we say "natural selection favors a (generally) religious mind because of religion's validity. If it weren't valid, it'd be useless or even dangerous and, therefore selected against"?

The problem with this is that invalid religion isn't dangerous enough to outweigh its social benefits.

Wieseltier is merely the kazillionth person in history to fall victim to the good old genetic fallacy. (We have a few more in this comment thread, of course.) Hey Leon, explaining something does not in any way mean explaining it away, so chill, OK? Yawn.

By Steve LaBonne (not verified) on 21 Feb 2006 #permalink

"Philosophy Prof" writes:

"There is no empirical evidence of non-physical reality, so therefore there is no non-physical reality." This is a fallacy, used to demonstrate circular arguments to intro philosophy students.

That's not a fallacy in the relevant sense, because it's a straw man.

You've made it clear that you find straw man arguments against Dennett congenial and acceptable, because of how you think his ideas are likely to be received.

That's pathetic.

LW isn't fighting for God or scary religion, he's fighting for the fact that some things -- ideas, minds, love, rational argument -- don't have mass or show up under microscopes.

Another straw man. Dennett frequently talks about things that don't have mass or show up under microscopes---in particular ideas, minds, love, rational argument---and their reality and importance. Unlike you, he says intelligent things about them.

I don't believe that you're a philosophy professor. And you don't seem to have ever read Dennett. You're no better than LW at this, and he sucks.

"There is no empirical evidence of non-physical reality, so therefore there is no non-physical reality." This is a fallacy, used to demonstrate circular arguments to intro philosophy students.

This is a fallacy, used to demonstrate straw men to intro philosophy students.

John,

I'm going to have to punt on arguing about Dennett's particular claim that evolution is algorithmic, and his spin on it; I don't have my copy Darwin's Dangerous Idea handy, and don't trust my recollection of my reading of Dennett.

I do share concerns about his overly Dawkinsish focus on gene selection to the exclusion of things like group selection, etc. That did not keep me from recognizing the validity of viewing natural evolution as an algorithm---but I can't say that my reading of how Dennett meant that is correct.

So all I'm going to say at this point is that I don't think the basic statement that "evolution is algorithmic" is patently absurd, irrespective of how Dennett put it and where he was going with it.

I think it depends very much on what you think "algorithm" means, and how you apply that term to evolution. Many people assume that an algorithm has to be sequential, or invoke subroutines to solve problems top-down using a divide-and-conquer strategy. That's not right.

There are lots of bottom-up algorithms, where the logic of the strategy is not reflected directly in the control structure, and complexity emerges by glomming things together into larger and larger structures in a data-driven way. (Bottom-up parsers are the most common example.) There are some that are also very simple, massively parallel, and implemented in hardware, such as routing algorithms for massively parallel hypercube networks.

So when somebody says "evolution is algorithmic," that sounds about right to me. And it's not because I don't understand evolution; it's because I do understand computers and algorithms.

If my use of the circular argument is part of a straw man argument, then my point is still granted: treating everything as made up of or caused by material substance is neither strictly "scientific" nor generally rational. We are commonly aware of non-material causes -- like the will -- that are not a part of "nature" in the sense that they are available for the empirical observation of natural scientists.

If Dennett doesn't want to reduce thought, belief, and choice to epiphenomena of matter, I'm happy for him, but again, he invites this criticism, by glibly trying to explain religion and mind as evolving from matter. When the criticism comes, it's not very convincing of his partisans to insist on the technical difference between explaining and explaining away.

In a philosophy class, the details of Dennett's position would matter more. But Dennett clearly conceives of his project not just as an academic philosophical enterprise but as an entry in the culture wars. Why else would he submit to the NYT Magazine treatment, or hope for his book to make waves outside of graduate philosophy departments? For his fans to complain about his arguments for being recognized for their cultural significance, and for being given prominent, extensive critical attention in the NYTBR, seems a bit disingenuous or insecure -- especially because, by any reasonable standard, LW does give genuine philosophical arguments against Dennett. (Those of you who think this review is just emoting haven't read freshmen papers in a while, and if you think LW's perspective is fringe religious you haven't read the New Republic for a while.) It is an interesting sign of how far the culture wars have come that to find a critic of atheistic materialism, the NYT doesn't have to go find some marginal conservative Christian but a respectable liberal humanist Jew.

I have read Dennet and have heard him speak. (I haven't read this latest book, but Bob Myer's comments above confirm my expectation.) Frankly, I know that he's not just man enough to dish it out but man enough to take it. I doubt he's half as vexed over the LW review as are his eager defenders on this blog.

By Philosophy Prof (not verified) on 21 Feb 2006 #permalink

If my use of the circular argument is part of a straw man argument, then my point is still granted: treating everything as made up of or caused by material substance is neither strictly "scientific" nor generally rational. We are commonly aware of non-material causes -- like the will -- that are not a part of "nature" in the sense that they are available for the empirical observation of natural scientists.

If the will has an effect, it is material. If the will has an effect, it is natural. If something has effects, it is available for empirical observation.

If my use of the circular argument is part of a straw man argument, then my point is still granted:

No, it isn't.

treating everything as made up of or caused by material substance is neither strictly "scientific" nor generally rational. We are commonly aware of non-material causes -- like the will -- that are not a part of "nature" in the sense that they are available for the empirical observation of natural scientists.

Question-begging.

If Dennett doesn't want to reduce thought, belief, and choice to epiphenomena of matter, I'm happy for him, but again, he invites this criticism, by glibly trying to explain religion and mind as evolving from matter.

You glibly dismiss things that he's written entire books to explain and justify. He's written two whole books, Elbow Room and Freedoms Evolves explaining his views on the will, and justifying them.

And you say he just glibly asserts those things. Whether he's right or wrong, he doesn't do that---you just glibly assert that he just glibly asserts them. I call foul.

Your descriptions of Dennett and his work are exactly what you falsely claim his are: glib and dismissive.

When the criticism comes, it's not very convincing of his partisans to insist on the technical difference between explaining and explaining away.

It's not just a "technical difference," and you should know that. Everybody else here knows that explaining something is not the same thing as explaining it away. There's a little thing called functionalism that is not ephiphenomenalism. As a "Philosophy Prof," you should know those things, and care about the distinctions. If you have a problem with, say, functionalism, you should explain it rather than falsely conflating it with epiphenomenalism and using it as an excuse to glibly dismiss anything you don't like.

In a philosophy class, the details of Dennett's position would matter more. But Dennett clearly conceives of his project not just as an academic philosophical enterprise but as an entry in the culture wars.

You seem to think that Dennett knowingly engaging in the culture war means that LW and you don't have to address his actual views or actual arguments; you can glibly dismiss whatever you like.

We, here, don't think so. We actually care about the reasons for believing or disbelieving things; we're not just culture warriors.

If you're going to argue that way, please find a more appropriate forum. We're not interested in that kind of false and dismissive propaganda.

In a philosophy class, the details of Dennett's position would matter more. But Dennett clearly conceives of his project not just as an academic philosophical enterprise but as an entry in the culture wars.

Sorry, buddy, but philosophy/critical thinking/science have applications outside the classroom: There's no reason to bring up Dennett's motivations (appeal to motivation), or even Dennett himself (that's just begging for genetic fallacies), except possibly, as Paul pointed out, for propaganda purposes.

Dennett, in the New York Times Magazine, January 22, 2006:

"Ugh. I certainly don't believe in the soul as an enduring entity. Our brains are made of neurons, and nothing else." Non sequitur. And glib.

"Belief can be explained in much the way that cancer can. I think the time has come to shed our taboo that says, 'Oh, let's just tiptoe by this, we don't have to study this.' People think they know a lot about religion. But they don't know." Straw man. And glib.

"Churches make a great show about the creed, but they don't really care. A lot of evangelicals don't really care what you believe as long as you say the right thing and put a lot of money in the collection box." Unsubstantiated claim. (Wouldn't you love to see the empirical study to back this up? Come on scientists, show me the evidence!). And glib.

Come on guys, if you want to stand up for him, at least know what you're dealing with. "He wrote books about his big ideas!" Great. Look where it's got him.

To BronzeDog: insisting that only material things can have material effects doesn't make it so. Note: I'm not trying to show you non-material things (for that, you need to get beyond the limits of your imagination, and allow "show" to apply to something other than the senses). I'm only pointing out that, in principle, no amount of empirical observation can establish that there are no non-material beings. I repeat: The proposition, "There are no such things as non-material beings" can't be discovered to be true empirically, and it can't be made true by defining reality as material. These are clear philosophical points. Please be assured that they don't threaten the practice or theory of natural science in the least.

For those who care about "reasons for believing or disbelieving things" -- I have yet to see a materialist produce the non-question-begging reason to believe that all reality is composed of or caused by matter.

By Philosophy Prof (not verified) on 21 Feb 2006 #permalink

The LW review, "ismism" short form:

Scientism
scientism
materialism
biologism
theism
protectionism
obscurantism
Theism
theism
theism
animism
theism
criticism
originalism
rationalism
rationalism
mysticism
materialism
humanism
humanism
biological reductionism
organism's
materialism
naturalism
supernaturalism
anthropomorphism
criticisms
criticisms
fideism

Clearly LW has a hyperactive -ism detection device. Must be genetic. Probably a Jewish thing -- Jews are always getting all theoretical-like.

Or is wamba just admiting that this philosophy stuff gets hard when ideas are treated as, well, *ideas*, and not as, say, synapses firing?

By Philosophy Prof (not verified) on 21 Feb 2006 #permalink

Or is wamba just admiting that this philosophy stuff gets hard when ideas are treated as, well, *ideas*, and not as, say, synapses firing?

Absolutely not, it's much easier that way. Just as it is much easier to consider a computer application as an abstract algorithm, or even as a piece of higher-level-language code that it is to consider it as machine code, or as NAND and NOR gates flipping. That doesn't mean the latter is impossible or incoherent, just more difficult.

P. P.,

Golly, it sounds like you're saying Dennett shouldn't give interviews.

I've read the NYT Magazine piece, and it's a very brief interview, or edited-down version of an interview. He's being asked his basic opinions, and he's giving them. He doesn't have the space to give many of his reasons, or explanations to make those reasons clear. (And likely, a lot of stuff he said got left on the cutting room floor by someone else.)

http://www.nytimes.com/2006/01/22/magazine/22wwln_q4.html?ex=1295586000…

You seem to think he shouldn't state his opinions that you find objectionable, if he doesn't have the time and space to cash them out to your satisfaction---or if he does, he's just an unprincipled, propagandistic culture warrior, and that gives you an excuse to act likewise here.

Don't go blaming your rhetorical shortcomings here on Dennett's very-brief-interview style. It's absurd and pathetic.

As the NY Times' Deborah Solomon is rather infamous for her interviewing style, I wouldn't be at all hard on Dennett regarding his answers to her as reported.

By David Wilford (not verified) on 21 Feb 2006 #permalink

I'm happy to admit the limitations of the genre of journalistic interview, but I don't think Dennett's claims here can be written off as just products of those limitations. The views he is expressing are an integral part of his project. Sheesh, I seem to be the only one here giving Dennett credit for what he actually believes.

I'm willing to be corrected. Perhaps you could please refer me to the books or articles where Dennett cites the relevant empirical evidence and offers extensive rational argument in defense of his claim that most "evangelical Christians" don't really believe what they say they believe, and for the claim that the only reason we have not historically seen fit to give biological explainations of religion is because of a "taboo". And please, please, please show me the source where he gives the non-question-begging argument that because the brain is made up of neurons, it follows that there is no such thing as an immortal soul. That I'd love to see.

By Philosophy Prof (not verified) on 21 Feb 2006 #permalink

To BronzeDog: insisting that only material things can have material effects doesn't make it so.

Missing my point entirely. If something has effects, it is material. By definition.

If psychic powers exist and have effects, they are, by definition, material.

If souls exist and have effects, they are, by definition, material.

If deities exist and have effects, they are, by definition, material.

Non-material causes and effects are impossible for the same reason that round squares and four-sided triangles are impossible: The only way something could be non-material is if it doesn't have effects.

I'm only pointing out that, in principle, no amount of empirical observation can establish that there are no non-material beings. The proposition, "There are no such things as non-material beings" can't be discovered to be true empirically...

Duh. You can't prove a negative. I never claimed a negative, except, maybe, as a null hypothesis to be falsified. Besides, non-material things, unlike material things, have no effects, and therefore can't produce evidence.

... and it can't be made true by defining reality as material.

I'm not defining reality as material. You seem to be trying to arbitrarily define some hypothetical material things as non-material.

Paul W.:

I have no objection to formal modelling, which is essential, nor to computer simulations, which can be extremely useful and often indispensible, if quantitatively large and complex systems of to be investigated empirically. Nor is there any problem with functional explanations of organism, which, pardon me, is obvious, and I think biological structures and processes would tend to be highly distributed and could in many cases be describable in terms of complexes of simple repeatable rule sequences. The problem arises when an functionalist/instructionalist account is projected on to biological evolution as a whole. An instructional system is latently directional, and that violates the atelic canon of Darwinian explanation. Simply put, animals have conation, but evolution does not. Nothing does the selecting in natural selection, unless it be, to absurdly generalize, something like local thermodynamic conflicts. You spoke of "search algorithms", but there is nothing doing the search, nor anything to be searched: stuff just happens. The other related problem is that if there were, say, an algorithm seaching the combinatorial possibilities of the genome, then only pre-existent possibilities could be derived. But biological evolution involves the emergence of novelty, diversity and complexity and the tremendous explanatory power of selectionism is that it can account for such without, as it were, seeking it. So the absurdity of Dennett's claim that evolution by natural selection is an "algorithmic process" partly from its artificiality, but also partly from its bumptious eagerness to explain, which mis-explains the whole point. Dennett elsewhere criticizes various putatively opposed thinkers for "sky-hooks", that is, constructing a preferred account of "mind" and then projecting it on the world, as if the world must be such as to accommodate their theory of "mind". But Dennett claims, (again, wrongly, I think) that consciousness is an "algorithmic process" and, lo and behold, it turns out that natural selection is too! So there's some reason for thinking him a fairly absurd fellow. Just because he's an atheist, doesn't mean we should all rush to defend him because he's on "our" side.

Philosophy Prof's basic claim, I think, has been that there are "things" that are not supsceptible to material explanation that are nevertheless not illusory or supernatural: e.g. a belief. One can explain a belief by recourse to the history of its holder and how he acquired it, but that does not necessarily go to the validity of the belief, which depends on the reasons and evidences that would justify the belief, which, yes, could include procedures of belief acquisition. But all three are intersubjective normative-rational matters,- ("deontic score-keeping" Brandom would call it),- and neither they, nor the beliefs they would justify can be reduced to correlations with underlying material causes. That is simply a bald category mistake.

By john c. halasz (not verified) on 21 Feb 2006 #permalink

Evidence that consciousness (or "mind") is strictly material in nature:

When we fall asleep, consciousness fades yet the brain remains active. Why is this so?
Massimini et al

Science, Vol 309, Issue 5744, 2228-2232 , 30 September 2005

To investigate whether changes in cortical information transmission play a role, we used transcranial magnetic stimulation together with high-density electroencephalography and asked how the activation of one cortical area (the premotor area) is transmitted to the rest of the brain. During quiet wakefulness, an initial response (~15 milliseconds) at the stimulation site was followed by a sequence of waves that moved to connected cortical areas several centimeters away. During non-rapid eye movement sleep, the initial response was stronger but was rapidly extinguished and did not propagate beyond the stimulation site. Thus, the fading of consciousness during certain stages of sleep may be related to a breakdown in cortical effective connectivity.

By David Wilford (not verified) on 21 Feb 2006 #permalink

David Wilford- or any other vulgar materialist here- explain how the sentences in your paragraph above were generated by cause and effect, as an apple falls from a tree.
Then explain- and here is the hard part- how to act upon such an assumption, in daily life, work politics etc.

Then answer this: Why do Dennett and our host argue with religious fundamentalists in the way they do? Why not examine religious arguments as Ronald Dworkin does in terms of the abortion debate. Most people opposed to abortion, many of whom will say 'abortion is murder' are not opposed to it in cases of rape or incest. How can this be, Dworkin asks, if they believe as the they say that "abortion is murder"? The answer is that they don't mean what they say- or what they think they mean- but something else, that they refuse to articulate even to themselves This has to do not with a simple crime but with the sense these people have that others do not or will not take the responsibilities of abortion seriously enough. Only a small minority of those who are opposed to abortion are opposed to it under any circumstances.

So in this case, what is it that keeps PZ Meyers, Dennett the Menace, or Richard Fucking Dawkins, from looking behind others' statements at what they might mean? Why this opposition to the possibility of subtext? Is subtext for mystics?

Have your little chats about free will if you want. Stuck as I am in the world, I am interested in communicating with others and in finding out about them, about how they live and respond to things, to each other. This is a political interest of course, but I don't want to simplify it. I'm human: I like company when I eat and I like to fuck.

As an atheist and a humanist I try to be agnostic towards myself. I consider it an intellectual and moral responsibility to make the attempt. Curiosity is a moral value.
By these measures Dennett has shown himself a failure.
The above words may be the result of pure cause and effect but we'll never know so I won't sweat it.

And all my other criticisms stand.
john c. halasz and P.P. keep up the good fight.

The problem arises when an functionalist/instructionalist account is projected on to biological evolution as a whole. An instructional system is latently directional, and that violates the atelic canon of Darwinian explanation.

I have no idea what this really means, and I'm afraid we're trying to deal with a bunch of different issues.

An algorithmic system is not necessarily "instructional", if I understand what you mean by that. (And I may not.) The relatively simple algorithms commonly implemented in hardware are algorithmic, but are typically not implemented by explicit instruction-following. For example, the steps in a sequential algorithm may be sequenced by a simple output signal from one piece of hardware, which triggers an operation by the hardware unit receiving that signal. No "instructions" are necessary, beyond simple signaling or clocking, i.e., something that causes the right thing to happen at the right time.

So to say that something is algorithmic is not to imply that it's "instructional" in anything like the strong sense of instruction reading/dispatching in a von Neumann architecture. The kind of genetic regulatory networks described in the P.Z. Myers article you referenced are quite sufficient to directly implement algorithms---and I would guess that most of them do. They are also sufficient to implement various higher-level instruction-following schemes, and I'd be rather surprised if some of them didn't do that, too.

And I don't see how any of this violates an atelic convention. I'm not assuming any of it is literally purposive, even if it may look that way if the right things systematically happen at the right times, and especially once adaptive things have evolved.

Nothing does the selecting in natural selection, unless it be, to absurdly generalize, something like local thermodynamic conflicts. You spoke of "search algorithms", but there is nothing doing the search, nor anything to be searched: stuff just happens.

I don't get this. If you look at any complex algorithm implemented in hardware, up close, there's "nothing implementing the algorithm" except little bits of hardware doing their local things and signaling each other.

Implementing an algorithm isn't a local property of particular bits of hardware. For example, in a hypercube routing network, there's no piece that's "doing clever routing across the network." There's just a bunch of bits of hardware at the corners of the hypercube, switching packets left, right, up or down according to entirely local criteria. They do not follow instructions, but they each implement a trivial local algorithm; taken together, they implement a much more interesting distributed algorithm. Routing across the network is an emergent property of those simple parts doing their local things in a mindless (and instructionless) way.

Now, of course, in hypercube routing, somebody designed a distributed algorithm and the local pieces of hardware so that that sort of distributed property would emerge. They probably wrote it down on piece of paper in some formal notation. But they didn't write it as instructions for the computer to follow. They just build the hardware, and the algorithm just happens.

Similarly, you could build an evolutionary algorithm directly in massively parallel hardware, with no instruction-following whatsoever. Just build a hardware cellular automaton, capable of supporting replicators, load in some random (potential) replicators, and turn it on. You would be implementing an evolution algorithm directly in hardware, with no instruction-following at all. And it would implement a search---the same exact search you'd get from a von Neumann computer actually executing the instructions of an equivalent program.

In the all-hardware case, there'd be "nothing doing the search"---at least, not in the obvious sense that the von Neumann computer is decoding and executing a search program. The search would "just happen," by the interactions of the various bits of hardware---which are admittedly designed, but not programmed.

My point about evolution being algorithmic is that nature seems to have done something basically similar, by accident---it set up a situation where the local interactions of matter result in an evolutionary algorithm that "just happens," with no instruction-following.

The fact that this was accidental doesn't mean that we can't recognize it as algorithmic.

Given the world as it is, I wonder at the subtexts in the arguments of those who turn blue in the face arguing logic with priests.
Given their inability to see what's obvious to others, that priests are not bad logicians but something else entirely, the subtext is all there is to think about.

I am disappointed in Leon Wieseltier's review of Dennett's �Breaking the Spell�, as much for its poor analysis, as for its closing, ad hominem insult. As a scientist, I know of no others who meet Mr. Wieseltier's definition of Scientism. They and Dennett are more accurately characterized as believing that science is the only arbiter for describing the properties of things in the natural world � things like liquid water, and theoretical constructs like the particle theory of subatomic phenomenon, and the evolution of religious behavior.

There is no problem in Dennett's assent to Hume's two questions regarding religion (its foundation in reason, and its origin in human nature), while not accepting Hume's response to the first. How many of us agree on a question while differing on our enlightened responses and discourses? Yet, Mr. Wieseltier uses the distinctions in Dennett's thought process to accuse him, inappropriately and unfairly, of misquoting and misrepresenting Hume.

Dennett is very clear, if not forthright to a fault, by saying he is offering his own speculation on what science may find in a study of religion as a natural phenomenon. Is he not explicit about doing so from the perspective of evolutionary (instrumental and functional) biology. Wieseltier seems to delight in uncovering Dennett's words on this, as if he has uncovered a secret, revealing passage, and hitting Dennett with a Gotcha!

Wieseltier dismisses Dennett's reasoning because Dennett's view presupposes human reason to be a natural phenomenon, based in biology. Then when Dennett uses the word 'transcend' to describe high levels of human reasoning, Wielseltier gives him another Gotcha!, and attaches the opprobrious label of 'animal' to Dennett's human reason. Wieseltier assumes an 'obvious truth' that human reason is a faculty that exists apart from its biology, a la Descartes. Well, here is where the discussion should begin. Instead, Wieseltier chose to end it, not prematurely, but before it even started.

By Norman Costa (not verified) on 21 Feb 2006 #permalink

There is a war on against the enlightenment that the "cultural war," in which some struggle to impose their beliefs on others, is but a part.

Bizarrely, the war predictably erupts amongst the eggheads whenever evolutionary psychology/mind/material issues come up. This should give the haters pause, but, of course, does not. I have seen the issue blow up so often now I now find it reassuring, the way a boil about to crest promises the cure is near, or perhaps like the pain of a tootache relished, or hearing 2 plus 2 is 5 and knowing it is not.

Agent A publishes some argument about the material basis of human nature, which is typically but speculations that will come naturally to anyone familiar with the work being done in the field, and, who is inclined to speculation. So Dawkins. So Dennett. So Pinker. (The Evil Tooby and Cosmides!) No big deal. The fun of the chase. True, when they speculate they irk other people and go beyond the evidence. So what. How ..... familiar. But, they at least work off the evidence of the day, where most all future winners will be found at work.

But then in response to these good faith speculations, Agent B (almost invariably from the humanities, which has been missing in action for generations) angrily quotes or references some received wisdom about mental processes, which is invariably unique to the western tradition. Seems about half the time the angry one works a dead end marxist tradition. (Say, Chomsky. Say Gould.) The received wisdom is sometimes profound, but more often dogmatic, and irrelevant on its face, because it is transparently contingent, with a proven history of zero predictive value, and proven useless in the field.

I haven't isolated the particular immune responses from the tenured eggheads humanities received wisdom, but the Mecca being defended by those cornered has something to do with the idea that one can be free of history and biology and create oneself by reason or some liberating choice that emerges from the study of whatever they have studied and prefer to study. They defend some secret liberating grasp on the human condition but for which evil will reign. They rudely defend some well mannered tradtion of Reason and civility that I have never seen in action, though I do see smug recitations. Somehow the way some have attached to the western tradition is threatened in a widespread manner. I don't really know why since I know my attachment to the western tradtion isn't threatened by Dawkins and Pinker and Dennett. To the contrary, it is a fun time to be around. The liberal arts and sciences are converging. It is a great time.

Once upon a time I endeared myself to an evangelical in an ev psych grad class by saying on her behalf what she was afraid to say but believed, that an alternative to an evolved psychology of the species explanation to religious belief (and the one offered in the class was far shallower than Dennett's) was that God did what he said he did. The trouble is, in that class, while the sneering at the Divine explanation was rude and unjustified, neither was there a place for the divine explanation.

Mr. Costa is right, those who want to should argue Descartes since that seems the core issue. Which I think is a waste of time. Descartes was wrong. Numerous traditions east and west that have never made that mistake. Those who want to go with Descartes should at least acknowledge that.

Paul W:

I lack your knowledge in computer science, but I think I understand what an algorithm is: it's a fixed repeatable computational rule or procedure that produces a fixed and repeatable outcome. And I grant that many biological structures and processes can be described in those terms. Cellular automata, for example, have been used successfully to explain the placement of breathing holes on the bottom of plant leaves, and Boolean algebra can be used to describe the hierarchies of gene regulation. But all your examples are designed, admittedly; even a computer simulation of evolutionary processes using, say, strings of code, is in some sense specifically designed, (and, to my knowledge, at least, such experiments have come across distinct limits in terms of the evolution of complexity). That's the danger of the metaphor of "reverse engineering", since one is tempted to forget that in evolution there is no design. Of course, it makes perfect sense to look at individual organisms in terms of their "design", since organisms are all more or less teleonomic or goal-seeking systems. But the question was whether it was legitimate to view evolution as a whole as an "algorithmic process", which I take it Dennett meant , not metahorically or heuristically, but in some sense literally, (reductively?), as if evolution itself were some "massively parallel" computational process. I'll illustrate the difference between an "instructional" or directed process and a selectional one with the immune system. When scientist started to investigate it in the '50's, they assumed in was in some way a directed system. An antibody would somehow identify an antigen and mold itself onto it to take on the "proper" shape and then be selectively reproduced. But that's not how it turns out. Instead, the body produces a very large number of antibodies of different shapes and types and whatever antibody happens to bind with an antigen gets selectively reproduced, the increase in the size of that antibody population among the total antibody population being the acquired immunity. The immune system neither "knows", nor "seeks" antigens, but it gets the job done anyhow. That style of explanation was dubbed "internal Darwinism". Similarly, long-run biological evolution does not "seek" anything except the immediate sufficient adaption of an organism to its environment, including the other organisms in that environment. The problem with a model of a variant population of algorithmicly defined replicators, (which for Dennett would be genes rather than organisms or metabolisms), is that over the long haul not just the configuration of replicators, but the rules of their replication would have to change or evolve, (I'm, of course, thinking in terms of organisms or matebolisms, rather than unit-genes). In other words, there would have to be algorithms for the transformation of algorithms, which can't be. The model is at once too conservative and too parimonious, too mechanistic and insufficiently biological. And it also fails to take into account the centrality of the organism/environment relation,- (which is why the above-mentioned computer simulations fall short, since they model populations but not environments),- since, as I mentioned above, there are no unambiguous signals in nature, (which even your hardware algorithms presuppose), whereas the strength of selectionist explanations is that they can account for the "fit" between organism and environment through the generation of new information, which is central to the features of the explanandum: novelty, diversity and complexity. Dennett is a philosopher by trade and he evinces philosphical vices, excessive cognitivism, for one. But philosophy traditionally concerned itself with the identity of "eternal" substances, which recur to themselves, which is encoded into the logic that they deploy. Hence novelty, new being, has always been a difficulty for philosophical thinking, as if it were a violation of philosophical reason. But I'm just being philosophically suspicious of the overgeneralizations of a philosopher.

By john c. halasz (not verified) on 21 Feb 2006 #permalink

I find this interesting with BronzeDog. Maybe we can clear up a confusion. In my experience, we philosophers tend to use "material" to signify extension in space and time. It is thus not redundant to say that an effect is a material effect, and materialism, the theory that everything that is real is material, is not be trivially true. I think this usage goes back to Aristotle (and this is also why "nature" and "physical" can be taken to signify what has extension in space and time). On these meanings of the terms, it is an open question whether there is anything real, any causes and effects, that are not material, and a proposition like "An immaterial God causes immaterial effects in immaterial creatures," such as we might find uttered by Aquinas, say, is not ill-formed (although it may be false).

I think Hobbes is the only major philosopher I know who tried to define "material" and "being" so that they were necessarily co-extensional.

I'm curious, are there whole language communities in which BronzeDog's alternative meanings of these terms are the norm? Surely this isn't an alternative discourse of science generally? If we could get straight this confusion over language, maybe we could clear up a lot of the "disagreements" over "materialism."

And BronzeDog, if you want to keep using "material" in your way, can you tell me what word you use to designate extension in space and time, and what word, if any, you use to designate what does not have extension in space and time? If materialism is trivially true for you, then I want to know how to state the interesting thesis that there is no cause or effect that is not extended in space and time.

By Philosophy Prof (not verified) on 21 Feb 2006 #permalink

...I think I understand what an algorithm is: it's a fixed repeatable computational rule or procedure that produces a fixed and repeatable outcome.

There are plenty of algorithms that incorporate randomness. From Wikipedia:
"Deterministic algorithms solve the problem with exact decision at every step of the algorithm. Random algorithms as their name suggests explore the search space randomly until an acceptable solution is found. Various heuristic algorithms (see below) generally fall into the random category."

Like your example of the immune system: it can be better understood/modelled (take your pick) as a random algorithm than a deterministic one.

The problem with a model of a variant population of algorithmicly defined replicators, (which for Dennett would be genes rather than organisms or metabolisms), is that over the long haul not just the configuration of replicators, but the rules of their replication would have to change or evolve (...) In other words, there would have to be algorithms for the transformation of algorithms, which can't be.

Why not? Look up genetic algorithms/genetic programming.

And BronzeDog, if you want to keep using "material" in your way, can you tell me what word you use to designate extension in space and time, and what word, if any, you use to designate what does not have extension in space and time? If materialism is trivially true for you, then I want to know how to state the interesting thesis that there is no cause or effect that is not extended in space and time.

I'm not sure what you mean by "extension" in this context.

My primary point about materialism is that I'm not going to allow "supernatural" or "non-material" as special pleading words. You can't have something that has an effect (an is therefore observable by science) and is simulatenously unobservable by science (having no effects).

I lack your knowledge in computer science, but I think I understand what an algorithm is: it's a fixed repeatable computational rule or procedure that produces a fixed and repeatable outcome.

I'm afraid you don't really know what an algorithm is. It doesn't have to produce or a fixed or repeatable outcome, and it doesn't have to be a "fixed" algorithm.

There are many algorithms that are nondeterministic or even intentionally randomized. (There's an entire field of study of randomized algorithms, where randomness in the algorithms makes them statistically immune to certain unfortunate interactions with systematicities in the data. These are often analyzed in game-theoretic terms, by showing that even a maliciously constructed data set could not foil the algorithm.)

In other words, there would have to be algorithms for the transformation of algorithms, which can't be.

Algorithms that transform algorithms are extremely common; most programmers use them every day---e.g., every time they use an "optimizing" compiler to translate source code algorithms into equivalent-but-not-identical algorithms implemented in machine code. Program optimization is all about substituting algorithms for other algorithms to improve performance without changing the results.

There are also self-modifying algorithms, which actually transform themselves, to adapt or "learn." These are not used as often, because they're powerful but hard to understand. (You can get much the same effect, more comprehensibly, with programs that generate new programs and execute those, rather than literally modifying themselves.) Such things are done under the covers in a variety of other kinds of software, ranging from graphics rendering to database query processing to checkers-playing.

One of the reasons I find Dennett's talk of search algorithms especially appealing is that a heuristic search algorithm may be "designed to" solve a particular kind of problem, but it's often quite unclear whether it really does it well, or why, and what heuristics would actually solve the problem better.

Having done research on various heuristics, I'm all too familiar with the fact that an algorithm is only a specified procedure, which may not implement the heuristic (guessing principle) that it was intended to. Even if it does, that guessing principle may not be a good one, depending on the actual problems you encounter.

The study of heuristics is largely the study of what algorithms actually do, given different "environments" they may encounter, i.e., input data or problem spaces. That is typically not quite what they were meant to do, and fairly often it's not at all what they were meant to do.)

The fact that heuristics may be intelligently designed counts for a whole lot less than you might think. Typical heuristics turn out to have unintended biases and failure modes. Even the intended biases may have important emergent effects that were overlooked in designing the algorithm. At the bottom line, intentions count for nothing.

In actually understanding a heuristic algorithm, you always have to be aware that it's just an algorithm; it is not magical, or intelligent. It's just a "machine" doing some stuff, and your job is to figure out what that machine will really do in practice, under a variety of circumstances.

Just like understanding evolution. Once you recognize that evolution implements a search algorithm, that shouldn't blind you to questions of what it's really doing, as you seem to assume. It should do pretty much the opposite.

It should make you ask questions like: what's the algorithm, really? What heuristic(s) does that really implement? E.g., what kind of local "guess" does a particular algorithmic action really amount to? And under what circumstances will that heuristic "guess well," or systematically "guess wrong"? What are the long-run and global implications of such patterns of good and bad guesses, in the face of different possible patterns of inputs?

It's largely about trying to characterize unintended and emergent phenomena, as opposed to blindly accepting an intelligently designed off-the-shelf solution and its teleological presuppositions.

This talk of "guesses" and so on may seem like it violates an atelic canon, but I don't think it really does---no more than typical talk of "function" in biology. For example, when I analyze a heuristic, I generally heavily discount the actual designer's intentions, and may reframe what the actual algorithm is really doing in terms of an obviously imaginary designer who designed it to do that, or even a malicious one who intended to get systematically poor performance or bad answers.

Because that "designer" is obviously imaginary, I'm constantly alert to the possibility that such a view overlooks important realities, and ready to reframe my characterization.

The more interesting point is that I treat real designers the same way, and that's what analyzing algorithms is largely about: pretending that there isn't a real designer, or at least not one who actually knew what He was doing.

This isn't so very different from how biologists think about evolution; I'd say it's essentially the same.

For me, the appeal of describing evolution as algorithmic is to stress that it only does what it mechanically does. Irrespective of any intentions, goals, or heuristics of any real or imaginary designers, the algorithm will do what the algorithm will do, which may not be what was intended or imagined.

Calling it algorithmic, as opposed to heuristic or adaptive, is a way of keeping in mind that it's not really smart, can't see into the future, and doesn't really care about anything. It's a way of staying grounded while entertaining useful "intentional stances" that may be enlightening, too.

If something is such that for it to be is for it be characterized in spatial dimensions, and so have size, shape, location, etc, we philosopher-types say it is extended in space or has extension. We also say that it is "material", i.e. made up of or determinable in terms of matter (which is the most general kind of thing that is extended in space).

Roughly, if something is available to our sense organs (which are themselves "material" in this sense), that thing is material or has extension.

So, on this understanding, if there is something such that for it to be is not for it to be characterized spatially, it is not material. Most people who believe in God take him to be immaterial in this sense, by which of course they do not mean that he has no effects. Such thing would not be available to the sense organs (if it were, it would be material), although we may still have some faculties, besides our five senses, which could help us to apprehend it.

This is why over the door to Plato's Academy was the inscription, "Let no-one ignorant of geometry enter here." The point is, one can't do philosophy until one is able to use one's intellect, rather than depending on the senses alone to grasp reality. Geometry, which asks us to apprehend universal truths (e.g. about triangularity, such as the the Pythagorean theorem), truths which do not depend on any particular sensible object and the necessity of which is not determined by testing samples, takes us beyond our senses and imagination by exercising the intellect.

By Philosophy Prof (not verified) on 22 Feb 2006 #permalink

Well, then. Extension into space doesn't seem to factor into my definition at all.

Most people who believe in God take him to be immaterial in this sense, by which of course they do not mean that he has no effects.

Well, then, if he has effects, that means he's material under my definition, and thus, subject to science.

Such thing would not be available to the sense organs (if it were, it would be material), although we may still have some faculties, besides our five senses, which could help us to apprehend it.

The "five" senses don't matter in my argument. If you can come up with some faculty to detect God or "immaterial" things, then, by all means, present it.

Philosophy Prof: Psychoneural dualism has the (apparently) insurrmountable difficulty of running afoul of conservation laws. (Remember that this was brought up against Descartes, who had the out that the conservation laws known at the time were not as all-encompassing as the ones known now.) As for the positive direction, I'm with Paul Churchland, Dan Dennett, Mario Bunge and many other contemporaries - contemporary neuroscience confirms the hypothesis every time a schizophrenic is partially treated, someone gets incarcerated for drunk driving, etc. (This is all elliptical, but you should be able to fill in the rest.) I also note (later on) that you claim that you cannot find ideas in the material world. Have you been looking? What is wrong with the central hypothesis of cognitive neuroscience? Do you think it false? Why? As for will (mentioned yet later), there is a tremendous amount of work being done on the neuroscience of will, some of which is appearing in the work of philosophers, including a modest student paper of my own. (See my web site.)

Seth Edenbaum: Dennett does not deny subjective experience. He, rightly in my view, attempts to explain where it arises from. Whether that project (in Consciousness Explained, etc.) is successful to the extent that he believes is debatable, but that's independent of the project's merit. (Bunge calls this the "objective study of subjectivity" and extends this to the social sciences as well.) Would you have us leave consciousness forever a mystery? I don't think that's compatible with scientific world views.

Paul W.: There are often two uses of "algorithm." In CS, algorithms do not have to produe fixed, repeatable result; in mathematics generally algorithms do.

The problem with a model of a variant population of algorithmicly defined replicators, (which for Dennett would be genes rather than organisms or metabolisms), is that over the long haul not just the configuration of replicators, but the rules of their replication would have to change or evolve, (I'm, of course, thinking in terms of organisms or matebolisms, rather than unit-genes). In other words, there would have to be algorithms for the transformation of algorithms, which can't be.

As has been previously noted, you're just wrong about the last bit, most strikingly with regard to genetic programming. Genetic programming systems are exactly algorithms that evolve algorithms. That's the whole point of genetic programming.

You also seem to assume that if something is algorithmic and involves replicators, those replicators and their manner of replication must be predefined programmatically and must be fixed.

That is just not true; replicators can emerge, as they do in natural evolution, from the underlying physics. Once a certain sort of replicators emerges, they can follow the rules of that sort of replication, including evolving new variations in their manner of replication.

The model is at once too conservative and too parimonious, too mechanistic and insufficiently biological. And it also fails to take into account the centrality of the organism/environment relation,- (which is why the above-mentioned computer simulations fall short, since they model populations but not environments),-

That's just not true. Many evolutionary algorithms model environments; although you're likely to see them under the heading of "artificial life" rather than "genetic algorithms." Even a simple genetic algorithm models some trivial sort of "environment," simply by using a fitness function. More interesting environments are, naturally more interesting. (Especially when you view evolution as a search algorithm; rich environments broaden the search, and that has many implications for what evolution can and can't do.)

since, as I mentioned above, there are no unambiguous signals in nature, (which even your hardware algorithms presuppose),

No, they don't. The whole point of heuristics is to deal with ambiguous signals.
E.g., given the bits of data or parts of the search space you've seen so far, what can you guess about other data or other parts of the search space?

whereas the strength of selectionist explanations is that they can account for the "fit" between organism and environment through the generation of new information, which is central to the features of the explanandum: novelty, diversity and complexity.

That makes no sense to me. Algorithms can generate novelty, diversity, and complexity. They can do it bottom-up, inducing new, higher-level algorithms by low-level mechanical actions. (Just like natural evolution.) This has been demonstrated many times with genetic algorithms, genetic programming, and artificial life systems; it can all happen inside a computer, just like it can outside a computer.

Algorithms are not nearly as limited or limiting as you think; they can be generative and creative.

Recognizing that doesn't sound like a "philosophical vice" on Dennett's part.

It seems to me that you have a lot of stereotypical ideas about algorithms and computation, which limit your ability to see the validity of what Dennett is saying.

Before dismissing Dennett as simplistic and blindered by his love for algorithms, you should realize some things that Dennett realizes:

Algorithms don't have to be deterministic.
Algorithms don't have to have a determinate output for a given input.
Algorithms can be implemented directly in hardware.
Algorithms don't have to involve instruction-following.
Algorithms can be massively parallel and distributed through a medium with no complex local operations.
Algorithms can induce novel and complex phenomena, bottom-up.
Algorithms can induce other algorithms, or even modify themselves.
Algorithms can process ambiguous signals.
Algorithms can be all about interactions of things with their environments.

If you don't understand at least these things about algorithms, you're in no position to condescend to Dennett as though he was naive about what can be an algorithm, and what algorithms can or can't do. You literally do not know what he's talking about when you dismiss his talk of algorithms as a bad metaphor.

This is helpful, BronzeDog, and while I'm talking with you I'm happy to use "material" in your way. But then, when I want to refer to extension in space, as a property belonging to rocks, brains, molecules, etc., but not, say, to justice, or right triangularity*, or (as traditionally conceived) God and angels (whether they exist or not!), then what words can I use? And generally, how would you name or formulate what *I* call materialism, the position that all that is real is material (in *my* sense of "material"), a position which if true would still not be *trivially* true? I just want to get our language straight before I take up your invitation to a substantive discussion about the existence of (what I call) immaterial objects.

And for everybody else: Is BronzeDog's usage of "material" common in your circles? In my circles, it's pretty idiosyncratic.

*And just to be clear, I am happy to regard "justice" and right triangularity as "abstract objects" or abstractions, which are not "real" except as they are said to characterize individual (material) objects -- but still, as "abstract objects" they are things that we can talk about, form true judgements about, and though they are not material (in my sense) they do have effects (e.g. justice causes whatever has it to be virtuous, the rightness of a triangle causes the square of the triangle's hypotenous to equal the sum of the square of its two sides, etc.).

By Philosophy Prof (not verified) on 22 Feb 2006 #permalink

Keith writes:

Paul W.: There are often two uses of "algorithm." In CS, algorithms do not have to produce fixed, repeatable result; in mathematics generally algorithms do.

Yes. That isn't the sense of algorithm that Dennett is using, though. He's clearly talking about the more general sense common in CS.

In CS, we do teach that sense when talking about making certain distinctions. Then we generally ignore that and use it in the more general sense---except when making certain distinctions about certain things.

My impression is that mathematicians end up doing the same thing for the same reasons, if they actually work with algorithms a lot. The narrow nit-picky mathematical definition is awkward even for mathematicians, because it's only useful for making a particular distinction that usually isn't interesting. (I have heard mathematicians start out by saying "I don't actually have an algorithm for this, just a heuristic procedure that usually works" and after a few minutes, consistently referring to their procedure as an "algorithm.")

I believe I've seen it here, at the JREF forums, and from my brother's philosophy courses. (B.A., seeking a Masters in Library Sciences)

Justice, right trianglarity, etc. are concepts/abstractions/whatever. They exist in the minds of humans, but, as far as I know, nowhere else. Of course, from my point of view, concepts are material, or at least have a big material component that creates effects: They exist as brain chemicals flying around in our heads or whatever.

God, angels, the Flying Spaghetti Monster, etc., if they exist, are still material if they have effects. I'd call their subcategory "unknown," "hypothetical," or "weird."

Nitpick: I don't think the rightness of a triangle "causes" the Pythagorean Theorm. It's a derivation inherent in its definition.

Kieth Douglas: thanks for your answers. I don't think I'm proposing "psychoneural dualism". I'm a fan of Aristotelian thought, and Aristotle's theory of the soul is perfectly compatible with, indeed would have predicted, e.g. our ability to treat schizophrenics, the effect of alcohol on behavior and perception, etc -- indeed, as far as I can tell Aristotle's conception of the soul is compatible with much neuroscience research (although please formulate for me what you take to be its central hypothesis).

Aristotle's conception of the human soul and its powers is just not susceptible to the same famous (and I agree devastating) criticisms of Cartesian substance dualism, and yet Aristotle still insisted that the intellectual operation of the soul did not directly depend on matter and so was, strictly speaking, immaterial. I have yet to hear an experiment described that could even in principle disprove his conception of intellectual activity as immaterial (and so, in principle, as something that could take place without any body at all). What we have evidence of is that, in human beings, intellectual activity is correlated with brain activity and brain stimula can effect consciousness -- but all this would be predicted by Aristotle's description of the human intellect's relationship to imagination and sensation.

By Philosophy Prof (not verified) on 22 Feb 2006 #permalink

BronzeDog, I'm trying to be patient with you, so please be patient with me. Surely you can think of some other word for the concept of (what I call) non-material objects other than "unknown" or "hypothetical" or "weird." (I'll agree that "Flying Spaghetti Monster" is weird, but probably if it flies and is made of spaghetti it would be material, again in my sense, unlike, say, God or angels, if they exist.) I just want to know how to say in BronzeDog speak, for instance, "A mind without a body, if such a thing exists or could exist, would be a _______ being." I would fill in the blank with "immaterial" (or "non-physical"). Even if you think such a being would be weird or unknowable or impossible, I want to know the general category of the kind of thing I could call it in conversation with you.

Surely you must allow for such a term, even if you have to make it up on the spot. Even if you think a bodiless mind is physically impossible, it isn't *logically* impossible -- we can *conceive* of it (Descartes at least had that much right). Indeed, if you think that notion is absolutely incoherent, a logical impossibility like the notion of a round square, then it would turn out that you have no non-circular arguments for your belief that all beings are material beings.

By Philosophy Prof (not verified) on 22 Feb 2006 #permalink

For me, it's simply drawing a fictional line. I guess maybe "extra-spatial" or something like that, sort of like how things that happen on Earth are terrestrial, and how things happening elsewhere are extra-terrestrial.

Good, "extra-spatial" works. So, for instance, would you agree that traditional conceptions of God have it that God is "extra-spatial", and Descartes thought that the mind was "extra-spatial"? I know you don't see the point, since you don't think there are such things as extra-spatial things, but, at least this way we can characterize the views of other people (like Cartesians or traditional theists) accurately and without ridiculing them or accusing them of commiting some fundamental logical fallacy. Right?

By Philosophy Prof (not verified) on 22 Feb 2006 #permalink

"Good, "extra-spatial" works. So, for instance, would you agree that traditional conceptions of God have it that God is "extra-spatial", and Descartes thought that the mind was "extra-spatial"?

Yes, they'd be "extra-spatial".

I know you don't see the point, since you don't think there are such things as extra-spatial things...

Straw man. There's the stuff that M-theorists deal with, that seem promising and scientific, at least to this cosmological layman.

...but, at least this way we can characterize the views of other people (like Cartesians or traditional theists) accurately and without ridiculing them or accusing them of commiting some fundamental logical fallacy. Right?

Sorry, but I'm still seeing the special pleading. Changing the word for the seemingly imaginary line isn't changing it.

Where's the special pleading? I'm not arguing any position with you yet, just trying to clarify our terms. It seems to me that insofar as we're all interested in rational argument we should want to be able to characterize the views of people we disagree with without also insulting them attaching unnecessary pejorative connation to their views. I don't call you an anti-Semite just because you disagree with LW; supply-siders shouldn't call social democrats "Commie pinkos" just because they have different economic theories and different views of the role of the state; etc.... I'm just trying the lay out a framework for civilized conversation.

By Philosophy Prof (not verified) on 22 Feb 2006 #permalink

From my point of view: It's pointless to come up with a term for "supernatural" or "immaterial" or what-have-you. It's just a big distraction. If you have a hypothesis for what causes certain events, you should test it scientifically. If you don't know how to test it scientifically, you should be asking for help coming up with a test.

The special pleading that I perceive is: "I have something with all the characteristics that define an apple (generates effects, and is therefore subject to scientific testing). But it's not an apple (it's supernatural), and shouldn't be treated as such (it's not subject to scientific testing)."

Paul W. and windy:

I erred slightly when I said that algorithms that generate other algorithms "can't be." What I meant is that they can't ultimately be, that they can not be expected to "regenerate" themselves in the face of unbounded contingencies. I'm working from the understanding that any system of rules is bounded, finite. There can, of course, be meta-rules, that is, rules that in addition to their own basic function serve to generate other rules. When a new domain in mathematics is opened up, say, imaginary numbers, certain basic rules are laid down, which then generate a system of implications. (That goes to the question of whether mathematics is invented or discovered: it's both. But there also has to be the human agency of some mathematician involved, who "sees" or intuits the mathematical "sense" of doing mathematics that way.) My basic point could be put in terms of general systems theory, that is, the abstract mathematical theory of self-delimiting, self-maintaining information processing systems, which some biologist apply in modelling organisms. An eco-system is not a "true" system, but rather an interlocking series of system/environment relations, and there can be no overall system of systems, for the intrinsic reason that such a total system would lack any delimitation from an environment, which establishes a system in the first place. I take Dennett's claim about evolution as an "algorithmic process" to be such a totalizing claim, rather than just a heuristic, in which case it could be analyzed where and how it fails. And for all the wonderful job that Paul has done of explaining himself, (though perhaps not so much Dennett rather than the reasons he finds the latter persuasive), he hasn't addressed my main point about the power of selectionist explanation: there are no unambiguous signals or information in nature, and selectional processes, by redistributing and re-enforcing the distribution of probabilities, in effect, generate new improbable probabilities that constitute the "information" required for sufficient adaptive "fit". The process is not just atelic, but generative, as well, (and, of course, path-dependent and constrained by prior organic structure). And that's a crucial feature of the explanandum, since any rules presuppose the information on which they operate. In fact, insofar as biological structures and metabolic processes do exhibit algoritmic or rule-governed features, it could be said that it is the complete absence of rules in natural selection as a whole that generates rules. So my basic objection to Dennett is that he misrepresents some of the real power of Darwinian explanation, while declaring it, in self-congatulatory fashion, to be "dangerous". I also take some objection to "machine" explanations of organic phenomena, explanations of a whole consisting of interacting parts which produce a result. It's true that it's essential to understand the discrete causal mechanisms that comprise the pieces of the puzzle and "irreducible complexity" is a hoax, if its meant to imply that complexity can't be explained. But that much abused phrase does have merit insofar as it points to the cohesion and self-organization of organic/metabolic processes: what is to be explained is life and not "dead" matter and ultimately the sheer improbability of the emergence of life from matter, which contravenes the conservation principle of sheer matter, specifically, the 2nd law of thermodynamics. A certain holism belongs to such explanation, even if one must be careful not to project the teleonomy of organisms onto the evolutionary process itself. (This would come up in Dennett's theory of consciousness, which, I think, fails to account for why such neural systems would emerge/evolve in interaction with other organic processes: the philosophical assumption that consciousness is a cognitive state is perhaps slightly askew.) I can understand why people in CS, or more generally, through the popular diffusion of IT, might be tempted to apply the metaphor of digital processing systems to other areas, but, aside from that analog processing is likely to be more apt in biology, it's just a metaphor, no different from the mechanical watch, whose maker was said to rule the universe.

By john c. halasz (not verified) on 22 Feb 2006 #permalink

"Mathematics is about the truth about integers ... and not about specific conventions."

well, I don't know about "truth", but I think the reality behind integer addition is this: I've got a pig and a dog; my neighbor has a pig. we swap my dog for his pig. how many pigs do I have? if we're on my farm, I can just point. but it would be convenient to be able to convey to someone not on the farm how many pigs I have. now I can say I've got my old pig and a new pig, but that gets cumbersome. etc, etc. until you have a set of "conventions". then along comes someone who says "this is too cumbersome, let's see if we can generalize these conventions into a manageable system". postulate some axioms and walla! - integer addition.

the role of intelligence in creating such mathematical systems seems critical, so if "objective" means roughly "exists independently of instantiation by intelligence", I don't see mathematics as objective. this wikepedia entry suggests (to me, anyway) that it's a lot harder to model integer arithmetic than would be the case if it were just there (ie, instantiated) waiting to be discovered as is the speed of light:

http://en.wikipedia.org/wiki/Addend

Norman Costa:
"As a scientist, I know of no others who meet Mr. Wieseltier's definition of Scientism. They and Dennett are more accurately characterized as believing that science is the only arbiter for describing the properties of things in the natural world: things like liquid water, and theoretical constructs like the particle theory of subatomic phenomenon, and the evolution of religious behavior."

Keith Douglas: 1-"Dennett does not deny subjective experience. He, rightly in my view, attempts to explain where it arises from. Whether that project (in Consciousness Explained, etc.) is successful to the extent that he believes is debatable, but that's independent of the project's merit. (Bunge calls this the "objective study of subjectivity" and extends this to the social sciences as well.) Would you have us leave consciousness forever a mystery? I don't think that's compatible with scientific world views."

Again and again and again! I ask these questions: Why are Dennett et al. so ideologically opposed to ethnography? Why is it that they find it necessary to believe[!] that people will not in good faith defend ideas that they can be shown not in actuality to hold? If Dennett does not deny subjective experience why does he ignore the examples of it that appear before him? I'll say the same of PZ Myers [and the whole school of rationalist analysis in general] The answer seems to me at least to be that to study the subjective experience of others -as such- is to have it reflect back on the observer. So no, I do not think Dennett accepts subjective experience. I can't believe Dennett thinks he will convince any believers to his side after treating them with such contempt (and having so much fun doing it) Dennett strikes me as someone who who deny the existence of his own arrogance. But isn't his frustration a symptom of the same 'disease' of metaphysical desire? And shouldn't he be able to diagnose these symptoms in himself?

Ideas may be a combination of neurotransmitters, proteins, temperature and timing, but neither Dennett nor anyone else has proven themselves able to construct a hard scientific model. And even then consciousness can not map itself entirely. Philosophers would call this an aporia.

"Would you have us leave consciousness forever a mystery? I don't think that's compatible with scientific world views."
Build yourself a computer with a primary algorithm- if that's the word- for self preservation, and one for collecting data via conditioned response. Give it a nervous system. Add these to any complex computational structure and your machine will achieve sentience, as insecurity indecision and fear.

You don't acknowledge psychological states by pretending they don't apply to yourself.

Seth. You crazy?

Where is it Dennett picks a fight, or treats believers with contempt?

As to taking shit, those that have delighted in the wonder the science pursues, will have a lifelong history of having their true beliefs being mocked, misrepresented, and misunderstood, by fools who take full advantage of all the benefits of science yet while scorning the discipline. Yet, for the most part, science fans remain cheerful and uncomplaining, if just left to go on doing good, advancing understanding even while others struggle to hold back the advance. Especially those biologists, who seem to be the labrador retrievers of science. Always up for some fun exploration with whoever else is up for it. Ethnology?!

Seth - you refer to this discussion in your blog as "a little fun with a high school science geek." I'm curious - whom do you mean?

Again and again and again! I ask these questions: Why are Dennett et al. so ideologically opposed to ethnography? Why is it that they find it necessary to believe[!] that people will not in good faith defend ideas that they can be shown not in actuality to hold?

Why do you think Dennett opposes ethnography? What is that last sentence even supposed to mean? Is there an extra 'not' somewhere, or why do you think people would defend ideas they don't hold?

Dennett thinks that we naturally adopt an "intentional stance" with respect to other humans, which means that we think they act based on the subjective ideas and beliefs they hold, just like ourselves. It's ironic that you propose that Dennett himself isn't capable of adopting this stance...

Paul W:

Since I cross-posted with your last post, I'll just stubbornly add a bit more, then leave off. Everything you say about algorithms, taken from advanced levels of CS, is fine: they can exploit randomness, whether "true" randomness or not, they can disambiguate signals, they can evolve and generate novel, unpredictible results, etc. (In the case of neural-net programs, their unpredictable evolution of problem-solving is, however, constrained by what is programmed into its modules or "cells", which thus are not analogous to a neurons, or better, neronal groups, which though they may well have "defined" re-enforced circuits, do not have anything like "instructions", only synaptic connections which alter over time.) But they are all cases of something purposely implemented in an architecture, of what ever sort, to deal with difficulties and complexities in that architecture, which is why they're difficult to successfully achieve, (and presumably why you get paid so much). The case of disambiguating signals is a crucial one, because any digital processing device can freeze and/or crash when confronted with undecidable signals. Now, I suppose one could take DNA, proteins, cells, etc. as the bits of architecture in and by which algorithms would be implemented. But what then generates those bits. If you say that they too are algorithms, you're entering into the pain of infinite regress. What I've been trying to point out is the real power of Darwinian selectionist explanation is that it obviates or finds its way around what would otherwise be seemingly intractible problems. ("Darwinism" is a radically anti-metaphysical, but robustly successful theoretical paradigm). Though I readily grant that algorithms can be useful in specific cases of specific features of orgamisms, Dennett's appropriation of Darwin , which he loudly trumpets, (though he is using Darwin to extend a philosophical/cognitive program of his own), in terms of an overall "algorithmic process", somewhat distorts the matter and is also obviated by a better account of what's "really" going on. I'll offer a simple example of a simpler explanation. In a digital processing system, of whatever kind, everything must be coded in yes/no or on/off bits, such that bounded categories must formed and borderline cases are a distinct problem. But in an analogy system, there is no yes/no, only more or less similar or different, and it doesn't make any difference which way a borderline case is "categorized" since even a whole series of such cases will simply evolve the "categories" or patterns. (Biological systems would also tend to exhibit a fair degree of redundancy, such that multiple pathways can achieve the same or similar "result".) The upshot is that "simpler" systems can acheive more complex results. I also think his rendition of Darwin somewhat skews the explanatory aim of biology. He wants to show how complex machines that we call organisms can be "built" from natural selection, including ones with "minds". But I would say the aim is to explain how those peculiarly cohesive things that we call organisms and their idiosyncratic behaviors emerge and evolve through contigent interactions with their environments and other organisms in it and the focus is not just on the "mechanics" of individual organisms, but also on the eco-systems that they co-evolve and on the biological "values" that various organisms develop. The latter are key to offering a biological evolutionary account of the emergence of any mental properties or capacities organisms might evince, up to the development of consciousness, such as perception, cognition and emotion. and the specific constraints that they may involve. In short, organisms are not just machines, and we have to distinctly recognize that in our orders of explanation.

I have absolutely no problem with "materialism" nor "physicalism" nor with the notion that life, qua autocatalytic, self-regulating, self-reproducing chemical systems, emerged out of the physical would and is constrained by and compatible with the laws of that world nor with the notion that "mind" or emergent mental properties and capacities evolved out of underlying biological processes. Those are all things that I believe, but within the broad circle of "physicalism" there are many possible positions, and I don't happen to buy into Dennett's version. If I'm guilty of "condescention" toward Dennett, then that's because I don't like his triumphalism, which is an excess that could be analyszed. (Still less, do I enjoy the enthusiasm of those of his fans, who think that "materialism" of " ev. psych." can explain everything, and if one points out that that is not and can not be so, one is immediately a dualist-Marxist-post-modernist-feminist-crypto-religious-effete-no-nothing-apologist for everything that wrong with the world and human culture.) But I also have a distaste for his reductionism, which I don't think one can deny is there. On the one hand, to explain, in the relevant sense, a phenomenon is to reduce it to its causes, and in that sense, all explanations are "reductive". But, on the other hand, any valid explanation must conserve the "identity" of the phenomenon it explains, else its not clear what the explanation is an explanation of. There is a difference between "x is A, because y and z" and "x is just A, because really y and z". In the latter case, A drops out of the equation and you really only have x, y, and z. Now I believe that reality, natural or not, consists of a number of emergent levels and is relevantly real all the way through. The problem with reductionist explanations, (which should be distinguished from reductionist research strategies), is that they attempt to explain a "higher" phenomenon at a level too "low" and that they tend to substitute implicitly an explanation of another "thing" for the "thing" that's claimed to be explained. I think something of that is going on in Dennett's "algorithmic process all the way through and all the way down."

By john c. halasz (not verified) on 22 Feb 2006 #permalink

I'm not going to ask my landlady why she's crying when I know her mother's just died.
I could of course say that since she's Catholic and goes to mass 4 or 5 days a week I'd have thought she would be happy, since her mother has certainly gone to heaven. If I were an android of the sort so many Star Trek watching technogeeks pretend to be, I would just stand if front of her impssively and tell her I was confused. Of course if I were human I'd probably be laughing to myself: passive aggresives enjoy that sort of cruelty.
I don't argue with the faithful because I know that at any given moment I might become one of them, by making an assumption based not on logic but my wishes. None of us can escape the tendency to believe, though I suppose autistics might come close.
Dennett's defense of the 'Brights' is predicated on the assumption that this is possible, and even, I gather, on the assumption that someone's already done it.
Who, Joe Stalin?
(see: 'rule of law' above)
---
R- "Yet, for the most part, science fans remain cheerful and uncomplaining..."

Scientists do tend to be optimists. But they also tend to use words like 'truth.' as in 'ultimate truth' but truth is a term of metaphysics, and science is not concerned with truth but FACTS; facts which are mundane until someone has the desire to discover them and then revert to it after the post coital glow of discovery has faded.

I'm tired. I'm done.

John,

Suppose I said that I thought the universe was "quasi-algorithmic," i.e., that basic physics at a very low level could be described precisely and accurately by an algorithm.

In that case, the generativity, creativity, or holistic properties of anything real could be precisely and accurately modeled by an algorithm, in principle, provided that we do it at a low enough level, with a big enough computer, and let the higher-level phenomena emerge bottom-up from that.

I do think that the universe is likely something like that. So do many physicists; so far as we know, the workings of fundamental physics are quasi-algorithmic. (Dennett and I and some cosmologists would just go ahead and call it "algorithmic," but I won't beg the question of whether that's "just a metaphor," or a literal truth with a lexical ambiguity messing things up.)

For all we actually know, the laws of physics could be literally algorithmic, in the very strong sense that our universe could in fact be a simulation running on an algorithmic digital computer in another universe; we don't know that it's not.

So far as we know, there are no observed phenomena whatsoever---however relational, holistic, emergent, environmentally reactive, interactionist, "irreducible," etc.---that are not ultimately perfectly modelable by an algorithm running on a digital computer. Not one.

What I want to know is whether believing that the entire universe is quasi-algorithmic is something you find objectionable, in itself. Is that at the root of your objections to Dennett's philosophy? Do you find it objectionably "totalizing" because it's a claim, or a defensible guess, about the totality of everything we know anything about? (Or is it irrelevant?)

I personally do believe something sort of like that. I think that the universe is likely fundamentally digital and quasi-algorithmic, or that if it's analog or not-quite-quasi-algorithmic, that low level difference likely doesn't really matter much to most things we're interested in---e.g., evolutionary theory, or theories of belief or consciousness.

I strongly suspect that either the universe is digital and quasi-algorithmic from the bottom-up, or that if it's not, it "might as well be"---the differences are not terribly philosophically important.

Does this make me a bad guy like Dennett? Is it a problem for you if I think that the universe is "just like a computer program" in that sense---or maybe similar enough that the differences aren't especially interesting?.

Am I a horrible "reductionist" for thinking that the whole universe is very much like a computer program, or close enough for rock 'n' roll?

Seth wrote:

If I were an android of the sort so many Star Trek watching technogeeks pretend to be, I would just stand if front of her impssively and tell her I was confused. Of course if I were human I'd probably be laughing to myself: passive aggresives enjoy that sort of cruelty. I don't argue with the faithful because I know that at any given moment I might become one of them, by making an assumption based not on logic but my wishes. None of us can escape the tendency to believe, though I suppose autistics might come close. Dennett's defense of the 'Brights' is predicated on the assumption that this is possible, and even, I gather, on the assumption that someone's already done it.
Who, Joe Stalin?
(see: 'rule of law' above)
I'm tired. I'm done.

"A little fun with a high school science geek"?

I dunno, I think "Interesting discussion interspersed with babble from a rude, incoherent troll" would be a better description for this thread.

PS. Stalin had no intention to "escape his tendency to believe". He was a complete paranoid nutcase who thought that believing in the right stuff makes science work the way you want.

Paul W.:

I don't think "reductionism" is horrible; I think its arguable, and perhaps ultimately such arguments come down to a matter of taste and tact. (Idealism, of course, is just as much a reductionism as any materialism.) As for your claim to an algorithmic universe, I would regard that as somewhat fantstically speculative and thereby highly reductive. I think explanations need to be tailored and differentiated and that there are no total explanations. (For one thing, explaining is something we do and is intricated with other activities, so we would never get to the end of it.) And I do think of the universe as plural, not in the sense that there are multiple universes, but in the sense that there is no singular overarching perspective on it. The so-called "laws of nature" are in the first instance grounded on the statistical regularities to be found there. But the laws of physics are simply not the same os the laws of biology. Biological negentropic organization does not "violate" the laws of physics, if one realizes that increases of biological negentropy must be offset by increases in environmental positive entropy, and that a "biosphere", a collection of co-evolving organisms, (since that is the condition of evolution and a single form of life would scarcely amount to much, nor likely endure and develop), can only emerge under quasi-permanent conditions of thermodynamic disequilibrium. But other than that, the laws of physics do not determine, but only condition and constrain the laws of biology. The usual account is that, if one ran the "script" of biological evolution n times, one would get n different outcomes. Even if physicists were to actually achieve a unified Theory of Everything, that would be just a theory of physics and would not serve other explanatory aims. (A good negative example would be those who try to explain consciousness through exotic appeals to quantum mechanics. As a matter of principle, nothing but "ordinary" material causal mechanisms should be required for such an explanation, and it is simply a matter of putting them together in the "right" way, which is done selectively.)

By john c. halasz (not verified) on 22 Feb 2006 #permalink

Paul W.:

I do think that you are taking a single model, one that you are professionally involved with and highly knowledgeable about, and reifying and overextending it. That's a not uncommon tendency among educated professionals: doctors think everything is about functional health, lawyers about rules and regulations, mechanical engineers about weights and measures, philospohers about universal reason and logical arguments, etc. You might call that "conceptual realism". But one thing that natural science, as well as, much else, should teach us is that the world is very different from ourselves.

By john c. halasz (not verified) on 22 Feb 2006 #permalink

"Problem: Natural selection favors a (generally) reasonable mind because of reason's validity. If it weren't valid, it'd be useless or even dangerous and, therefore, selected against."

This is a commission of the fallacy of presupposition in that you take as a priori that because something is conducive of survivability (that is, because it is "useful" within the context of natural selection), it is also conducive of truthfulness. Perhaps there is some environmental regulation of which we are unaware because of our possibly corrupted paradigm of intellectualization that actually greatly favors ignorance and grants it, as opposed to knowledgeableness, the greatest chance of survival. Even within an empirical framework, anti-empirical beliefs favored by natural selection are not absent in the historical record: The Commandment "Thou shalt not kill" could certainly improve one's chances of survival and believing in its value would be "useful," but we wouldn't take its religious basis as being factual, or, to use your term, "valid," as a consequence of those observations.

By fritzerbuster (not verified) on 22 Feb 2006 #permalink

don't think "reductionism" is horrible; I think its arguable, and perhaps ultimately such arguments come down to a matter of taste and tact. (Idealism, of course, is just as much a reductionism as any materialism.) As for your claim to an algorithmic universe, I would regard that as somewhat fantstically speculative and thereby highly reductive.

OK, so quantum cosmology is tasteless and tactless, I guess. (?)

You seem to be avoiding addressing my point head on.

There is every reason to think that the entire universe is what I'd call algorithmic, or at least quasi-algorithmic if you think it's tacky to go ahead and call it algorithmic.
By that I mean that the laws of physics are likely isomorphic to the workings of an algorithm.

I am not asking about the wildly speculative idea that it's actually a program running in a computer; I don't believe that myself, or particularly care to defend the idea---and of course that's "wildly speculative," but it's not what I am interested in.

I'm also not claiming that much interesting follows from that, for typical scientific purposes---or even for most philosophical purposes, if you're already a materialist and are not allergic to mechanism.

I am not talking about radical "algorithmism," where you greedily reduce every kind of phenomenon to the workings of an algorithm at the next level down in functional organization. (E.g., reducing natural evolution to the workings of some easily-recognizable genetic algorithm that reifies classical Darwinian selection, or reducing consciousness to some clear aspect of an "intelligence algorithm" that looks like a classical AI program, or whatever.)

I'm not talking about any of that when I ask you whether the idea that the entire universe is quasi-algorithmic, and entirely mechanistic, is something that bothers you. I'm talking about that idea itself, and not any particular strategy for reducing particular high-level phenomena to low-level physics, or even any assumption that it's interestingly possible. I'm not talking about whether any greedily reductionist theories are adequate descriptions of any of the very complicated kinds of phenomena you can find in such a universe.

So please do not go on and on about what has or doesn't have teleonomic this or negentropic that; I'm not ready to process that information, and it's a waste to keep spewing it when I'm asking a more basic question.

I know full well that the laws of biology are not the laws of physics. I think that should be apparent by now. And I agree that Dan Dennett (and Steve Pinker) are too greedily reductionistic about certain particular things sometimes. I don't want to talk about them, or which things they're too greedy about, or any of that, at the moment.

I am only asking, now, whether you consider the idea that the universe as a whole is quasi-algorithmic objectionably "totalizing," and/or "reductionistic" in a bad sense, in itself.

Such a universe is entirely mechanistic in a straightfoward, fundamental way, but that does not necessarily imply that any interesting high-level phenomena in that universe can be described by simple, isolable algorithms. (I know that, but I'm not sure you really do; you seem to run things together across levels a lot, making it impossible for me to follow you.)

One reason that I ask is that I sincerely don't know what you mean by "reductionist," or whether you're really comfortable with materialism or mechanism per se. Maybe you're entirely comfortable with those things, as a general framework, but skeptical of particular greedily reductive explanations within that framework. Or maybe the problem goes deeper.

But maybe, just maybe, the problem is that you are phobic of giving me even an inch. Maybe you don't want to acknowledge that you think the universe is likely entirely mechanistic and quasi-algorithmic, because if you give me an inch I'll take a mile.

If so, please stop it, or let's end this conversation right now. It's really tedious to argue with somebody who obscures good basic points because he's afraid of where somebody will go with them. It's just not worth it.

In particular, I don't want to hear any more condescending etiologies of my looking-for-a-nail thought processes like this:

I do think that you are taking a single model, one that you are professionally involved with and highly knowledgeable about, and reifying and overextending it. That's a not uncommon tendency among educated professionals: doctors think everything is about functional health, lawyers about rules and regulations, mechanical engineers about weights and measures, philospohers about universal reason and logical arguments, etc. You might call that "conceptual realism". But one thing that natural science, as well as, much else, should teach us is that the world is very different from ourselves.

Do you think you know me? I think you don't.

You seem to think you have me pegged as a computer geek who wandered into philosophy and biology looking for a nail. BZZZT! WRONG!

Please cut the condescending crap, okay? Until you can tell me what I studied, and what undergraduate and graduate degrees I got, and in what order, please spare me this insufferable talking-down-to-the-engineer foolishness.

Just because I obviously know a hell of lot more about engineering than you do, that doesn't mean I'm a worse philosopher or psychologist or anthropologist of science geeks than you.

So if you keep talking that way, I'll have you pegged as a pretentious, insecure poseur.

I think you have a certain theory about why people might take algorithms very seriously, which you're overly fond of, and you're trying to apply it to me---like somebody with a hammer, looking for a nail.

I think you picked the wrong geek.

not to presume to referee a debate that is way beyond my knowledge base, but it might be useful to the participants to know how a moderately educated and intelligent reader reacts.

I have just reread almost all of this exchange. with only a basic understanding of algorithms (a comm EE with some experience with primitive H/W and S/W algorithms), I got the gist - if not in every instance the details - of paul W's discourse, and my reaction is, I suppose, the ultimate compliment - "duh". seems obvious.

on the other hand, my reaction to mr halasz is concisely captured in "you seem to run things together across levels a lot, making it impossible for me to follow you". admittedly, that in iself doesn't mean much; there is the possibility - ok, likelihood - that it's because I'm too ignorant to follow - but if paul W has the same problem, maybe some self-reflection is in order.

in my experience, over-jargoning is a pretty good indicator of a problem: sometimes benign - not being a good communicator or a degree of confusion; but sometimes not so benign, eg, being a "pretentious, insecure poseur". but at the least, since some of us read (and, BTW, comment on, expecting to be shot down) this kind of blog for self-educational purposes, it's a disservice not to be understandable. from reading john w, I have a new idea or two. not so from mr halasz, even though for all I know his comments are actually the more idea-ladened.

Paul W.:

In the first place, "tact" does not mean "tacky", unless you're making a pun. I was trying to be polite with respect to your rhetoric about "horribleness", though I probably missed its strategic deployment. What I meant was that everyone has a perspective to which prejudices or prejudgments are attached, and whereas one should try to reflect on one's own prejudices and become thematically aware of them, as well as, attempt to understand the perspectives of others and the prejudices that may be involved in them, I don't think such prejudices are ultimately dissoluable into some pure, mutually transparent, presuppositionless understanding. Call that just a fact of life, but I also think it goes to something of the meaning of "Enlightenment". Now some people take satisfaction in "reductive" explanations that show that some A is really just the effects of some set of "lower" order causes. Others might insist the A is an irreducible whole, and, if they are not being sheerly naive or dogmatic, they would offer a set of reasons to justify the claim. But what counts is not what is "rationally" satisfying but what is true, even if we may never have a complete and absolute knowledge of the truth. I would be of neither disposition. As to the latter type of claim, I think it likely involves a failure to "break down" and analyse a real phenomenon and try and see to what extent it can be further explained or explicated. As for "reductionist" explanations, I think what they show is that A is not a real phenomenon, but rather an epiphenomenon, a mere "appearance", and that can be valid, but is not a paradigm of "adequate" explanation. As I tried to explain, I don't see how the "universe" or "nature" is conceivable except in terms of a series of emergent levels, which a real all the way through. So for me adequate explanation involves the methodical delimitation and "identification" of a real phenomenon and the framing of an explanation in terms of a set of causes at the relevant level and kind or pattern, (which might involve several levels and patterns as needed). But "reductionism" as a scientific research strategy is essential, since we need to know the causal bits and pieces that would comprise the laws, structures or organizations that form an underlying explanations of kinds of phenomena. And often enough a "reductionist" explanation is the best understanding a science can achieve at its current state of research, which is fine, so long as any sense of incompleteness or puzzlement is not forgotten. We might not know, say, how gene regulations lead to brain development that fixes anatomically instinctual programs; we may know many bits at several levels without understanding the whole process and pathway. But, as I indicated, I view explanation as a human activity intricated with other activities, which is a quasi-pragmatic or praxiological view, such that, while causal explanations must be rigorously atelic in their "logical form", it makes sense to speak of and recognize "explanatory aims", which does speak to the question of adequacy. In explaing "things", we are trying to introduce rational order and intelligibility into the world, starting from our experience and moving out from there. So I think its legitimate to ask what an explainer is trying to do with his explanation, which does come up in disputes over "holism" and "reductionism". But perhaps more to the point, there arises the question of the limits of our knowledge beyond the world of our experience, which might not be timeless or ahistorical, but do come across the impossibility of any "metaphysically" complete knowledge beyond the contigencies of our experience: we come to a point where we have to answer, we just don't know" and "that explanation has no possible fit with our uses or needs". So I guess I should be called a "weak" reductionist- (sorry, if that's not hard-headed or "manly" enough),- and an inconsistent naturalist, since I don't believe that language, meaning and socio-cultural forms of life and their development can be wholely, adequately and relevantly explained solely in terms of material causal processes and since I don't see any rational or philosophical imperative or systematic necessity toward the development of a "consistent" naturalism, else we're logically damned. (I think such a singular systematic drive is prone to lead to errors, a Wittgensteinian point). So much for my epistemic views, which I don't think are unreasonable or obscure, but might not be to others' tastes.

Now to address directly the "challenge". I was admittedly taken aback by your post, because I thought the topic was evolution as algorithm and not the universe "entirely", and having raise an "infinite regress" point, it struck me as raising the stakes, even as I failed to perceive your intent to perform a "litmus test". And one of my first reactions was to be reminded of those old idealist systems that claimed to "produce" the world from the self-developing intelligibility of "mind": hence "fantastical". That, of course, is not your meaning, nor intent. But I do have some difficulting in construing the claim and making sense of it, (which might reflect my limitations and not yours). Did you mean that the universe in its entirety could be construed as if it were a vaste machine implementing algorithms, or that it is somehow itself one single total, if complex, algorithm, or that it is a self-generating series of self-generating algorithms? But then it turns out that you think the claim is "not very interesting" and it's not a claim to explain "interesting high-level phenomena". If the claim then is that at least some physical cosmologists think that algorithms are the best formal means of integrating the theoretical explanation of the most basic structures of the physical universe or that they are "isomorphic to the laws of physics", which presumably would mean that they are otherwise cast in a different formalization, then I will grant you that agnostically, since I would have only the dimmest and most rudimentary understanding of advanced physics. But I'll drop the question about "algorithms", in that I take it the point you want to make is about "mechanism". If that means that the "universe" or natural reality thoroughly is non-teleological, indifferent to human purposes and through and through comprised of the effects of immanent material causal mechanisms or processes, then I would hold to that and would simply call it "realism". It's less something I believe than a presupposition of anything I would see fit to invest any belief in. If the further question is whether the physical universe, "expressed" in the basic laws of physics are the primary reality, then obviously without the physical universe there is nothing at all, no "higher-level phenomena". But perhaps I have a quibble there. If the laws of physics, through local/regional disturbances or disequilibria induce the formation of causal processes, they don't necessarily suffice to completely determine or adequately explain further developments on the road to "higher-level phenomena". The causal events that somehow interact or link up to form causal processes, through those linkages, form structures that constrain the causal "flow". And understanding how those "structures" are causally formed and "causally" constrain is, I think, key to understanding how causal processes generate "high-level phenomena", whether or not that too is "algorithmic". (A later additional note from preview: just to try to be clearer, I'm emphasizing "structure" as a "key" because I think it goes at once to "identity", that is, what a real phenomenon is, "efficacy", how a phenomenon impacts and is impacted or controls and is controlled by us, and "intelligibility", what about a phenonmenon grants the terms by which we would grasp and understand it.) The laws of physics are not "designed" to do that; that's not their "explanatory aim". So maybe I would say that the laws of physics are essential, but not fundational, if that would answer your putative question about their primacy. I'm not really interested in any question about an absolute singular origin to the universe and I think of the universe "as a whole" as in some sense contigent; it's not turtles all the way down for me, but rather at bottom stochastic. If that sort of explanation does not rationally satisfy some, I can see no other basis. Of course, you apparently want to claim that the universe is "entirely" mechanical, and that's where I "cope out", since, as I said, I don't think that language, meaning, and human socio-cultural life can be relevantly and effectively explained in such terms. (Philosophically, if one adds in social interaction, a lot of puzzles and problems generated by the model of monologic consciousness/external physical world are obviated as unnecessary. And Dennett is pursuing a philosophical program, into which he enlists Darwin, of systematic naturalism, with a heavy dose of psychologism, which I disagree with. But that's a whole 'nother can of worms.)

This exchange began with my being called out for dismissing Dennett's construal of evolution as an algorithmic process. And I've tried to argue that such a construal is at once too much and too little to do justice to the explanatory power of Darwinian selectionist explanation, arguing that Dennett's claim gives short shrift to the stochastic, locally episodic and ecological dimensions of such explanation, which don't require any overall structure of rules, and I've also tried to indicate how "simpler" mechanisms or processes might give rise to more complex and more robust results. I've also indicated something of why Dennett's organisms as machines is reductionistic, in a pejorative sense, and distorts what I see as the explanatory aims of biology, and also hinted that it damages his effort to give a credible account of the emergence of "mind" or consciousness from biological processes, through not adequately taking account of the organic interactions between physiology, behavior and emergent mental functions and rushing to an account of consciousness as an adaptive function of cognition. And I don't think I'm wrong in construing that Dennett takes his understanding of biology from the model of machine intelligence or formal cognitive-psychological analogs. For all that I've been instructed here on the types, ranges, and power of algorithms, I don't think it's been effectively counter- argued that the transfer of such models onto the domain of biology is warranted. I'm sorry if I'm too prolix, but I'm trying too be clear, if unsuccessfully so, and if I'm "condescending", "posing, or "phobic" or whatever ad hominems are applied, I've tried to be rationally accountable to views or aims that were not always themselves clear. My views might not rationally satisfy others, but then there's no law that says they have to. Are we through with the de-ontic score-keeping" now?

By john c. halasz (not verified) on 23 Feb 2006 #permalink

Philosophy Prof: The central hypothesis of neuroscience is that psychological functioning (from involuntary control of organs and gland secretion) to the most abtruse theological or mathematical musings is performed by systems of neurons (not necessarily all in the brain, I might add) acting in concert. Aristotle himself is of course ambivalent about materialism, as I am sure you know (the above hypothesis is, of course, one specific materialist hypothesis - one could think that another organ or systems were involved (as the endocrine system is coupled strongly, for example) but I'll let you have your strong antimaterialist Aristotle. But note, though, he says that many psychological characteristics and processes require the body to perform and gives quite a list. We know quite assuredly now that the list can be extended to every known class of process. (Probably not every individual process; but to appeal there is an idealism of the gaps.) One example: your discussion included "the will". As I mentioned, there is a lot of work on willed action. Similarly, if we take a look at what happens to patients with emotional impairment, not only do they fail to feel, they also fail to reason correctly. For example, they have great difficulty in changing their behaviour in the light of new information. Where is the dividing line between the cognitive and the affective?

Of course, if you expect, like an "ordinary language philosopher" of a ridiculous sort to find entities that correspond to the nouns in our psychological vocabulary, you will be disappointed. "The nervous system minds" is parallel to "The circulatory system moves blood". Digestion is similar; it is what the stomach and intestines do. (Note, if you're curious, that my position is compatible with functionalism, elimative materialism, and Bunge-style emergent materialism, and certain versions of Kim's view, whatever you want to call it. See my paper on neuroscience and philosophy.) The language of corrolation implies either another stuff (idealism) or another bed of properties. I have no objection to saying that there are Bunge-style emergent properties of nervous systems, but still are what the nervous system does.

Seth Edenbaum: None of the three people under consideration (myself, Bunge or Dennett) is opposed to enthography. I don't understand why you bring it up, either, nor do I understand most of your message. This philosopher does not find aporia in the notion that any model of anything but perhaps the most trivial factual system is necessarily incomplete.

Incidentally, to the mechanismic/algorithmic debate. There's a decent, though not terribly good, case that the two are disjoint. (They are certainly disjoint from materialism.) More later if time permits. (Curious may view my web page here. I don't mean to toot my own horn constantly, but these are complex issues which I have addressed at great length elsewhere, so I might as well make use of the work. :))

update from the "de-ontic scorekeeper": that was totally understandable (well, I did have to enlist webster for help with "praxiological") and the score is now one "duh" each. plus a bonus point for the turtles analogy, which occurred to me as well.

update from the "de-ontic scorekeeper": that was totally understandable (well, I did have to enlist webster for help with "praxiological") and the score is now one "duh" each

On that note, I think I'm basically going to bail.

I think that to do this conversation right, we'd need to start over and separate out some basic issues, and proceed very, very carefully.

First, we'd need to clearly talk about functionalism, and in what sense that is or isn't reductionism. (As opposed to a kind of holism, even. It certainly isn't epiphenomenalism or eliminativism.) There are several ways a functionalist explanation can fail, or fail to convince, and all of them tend to get turned into accusations of "reductionism." I think there's a less loaded and ambiguous way to describe most of them.

We'd also need to thoroughly discuss several clear examples of "Dennett," as opposed to just generalizing and calling him "a reductionist," or a bad reductionist as opposed to a good one, or whatever.

And from that, we'd have to address the issue of whether Dennett is really a totalizing (bad) reductionist, or just justifiably more optimistic about functional explanations of certain things (e.g., the mind, religion, morality) than a lot of other people... and whether he's justified in being as optimistic as he is about various specific theories of various specific phenomena.

And from there, we could explore the issue of whether he's right that there's a taboo around these subjects, which must be broken, or just a justified skepticism.

Whew. I really don't have the time or energy for all that right now, and it's not clear this is the right forum for it.

Keith Douglas says: "We know quite assuredly now that the list [of psychological functions performed by bodily organs] can be extended to every known class of process." I don't see how this can be asserted, scientifically or philosophically (without begging the question). Basically, Aristotle's psychology (as interpreted by Aquinas anyway) has it that all human cognitive processes depend on the body, either because they are physical processes (like sensation and imagination) or because they are naturally accompanied by physical events (like intellectual understanding, which draws on imagination). However, one activity cannot be reduced to another activity that it depends on, and the activity of intellectual understanding is strictly speaking immaterial (the act of my intellect is not a physical act), even if it is (in natural life) always necessarily accompanied by some material acts which it draws on (acts imagination). (This is why Aquinas could find Aristotle's relatively "materialistic" analysis of the soul not only congenial but already supplied with the necessary premises for an argument for the continued existence of the soul apart from the body, at least in principle.)

I can't imagine what the scientific experiment would look like whose results would either confirm Aristotle's hypothesis (that intellectual activity itself is immaterial) or the more strict materialist's hypothesis (that all cognitive function, including intellectual activity, requires bodily activity). Again, putative evidence for the latter -- correlating thoughts with brain activity -- is not only consistent with but predicted by Aristotle's treatment of the relationship between imagination and intellect.

A positivist might claim that Aristotle's isn't a scientific hypothesis, since it can't be empirically verified. But then, since it is a hypothesis which competes with the more strict materialist hypothesis, neither is the materialist hypothesis scientific -- it can't be proven, since the phenomena that it claims to explain can be explained by the alternative Aristotelian hypothesis.

By Philosophy Prof (not verified) on 28 Feb 2006 #permalink

from one of the letters:

"As a composer, I am weary of being commandeered as evidence of supernatural forces."

zing!