Loose Ends of Science

John Horgan's recent position paper is certainly powerful — a rhetorical masterpiece. But on closer inspection I'm not sure the argument hangs together. Science seems to be ending for so many different reasons that it's hard to keep track of them all.

Cosmology and biology are coming to an end, we're told, because their reigning theories — the Big Bang and Natural Selection — are all but complete. Particle physics, on the other hand, is said to be ending largely because Congress cancelled the Superconducting Supercollider, making it impossible to proceed. The study of complex systems, John argues, never even got off to a start because it depends on computer simulations, which he sees as little more than very complicated video games. (There is something inherently fishy about computer models, he suggests, that somehow doesn't apply to modeling with differential equations. I really don't get the distinction.)

Scientific questions that don't fit into the above categories are declared to be unanswerable for metaphysical reasons. Neuroscience, for example, is supposed to be coming to an end because the nature of consciousness is forever unknowable. The deepest biological question of all — whether life is an aberration or something universal — will probably never be answered, John says, because we're stuck inside the solar system and can only wait, in vain, for someone out there to contact us.

Is it true that the proposed title of the book was originally "The Ends of Science"? A different endtime scenario has been tailored to fit each scientific frontier.

Maybe I'll write a sequel called "The Loose Ends of Science." Our theory of particle physics, the ingeniously jury-rigged Standard Model, stops far short of unifying the electroweak and strong nuclear forces and depends mightily on the existence of a particle, the Higgs boson, that has never been seen. The Big Bang theory cannot explain something so basic as how structure arose in the universe without declaring that most of it is made from a kind of "nonbaryonic" dark matter not included in the Standard Model. If science is over it is not because it is complete. The present theories are just running out of steam.

Because of the insatiable human hunger to find pattern, the search for better theories will continue. And because no map can ever encompass all of creation, science can never really end.

Re: Mr. Horgan's Claims:Useless and Mildly Pernicious

I am not sure why Mr. Horgan's remarks forwarded to me by Norman Levitt via Alan Sokal seem worthy of consideration by an active research scientist. Just as the response to the man who pronounced "an end to history", is that history continues nevertheless, so any claims that science is limited or ending or has an end, are refuted by "it" continuing nevertheless.

Certainly Mr. Horgan is correct that each fact known so to speak, is one less to discover, so the supply of facts, theories, or explanatory paradigms etc. is presumably drying up. But it is tantamount to saying that each day the universe is winding down and we humans are one day closer to extinction. (Or that each poem written is one less that can be written, although the analogy here is less exact.) How useful is such an observation? The question is, where are we now on this huge time scale?? We are operating now on a temporal microscale. Mr. Horgan is talking about events on a temporal macroscale.

I don't regard Mr. Horgan's claim as a very fruitful area of discussion unless Mr. Horgan could demonstrate how his thesis directly and practically had some implications for current research strategies and programs. As to the argument that scientists are just "filling in details now", that same argument could have been made after the periodic table was discovered as well. The point is that yesterday's dismissed "details" often have had the seeds of tomorrows great discoveries. So let us not worry if we are only working on what to Mr. Horgan appear only to be details.

Moreover, Mr. Horgan's claim may well be pernicious in its consequences, if he can convince aspiring students that since "science" has limits or is coming to an end, that therefore there is not much point in individuals pursuing a scientific career. Certainly, some individuals have retrospectively cited Mr. Horgan's or similar arguments to justify dropping out of science, and doing something else usually easier and less challenging — once they had tenure at least — but these are often decisions by individuals who didn't enjoy doing science so much in the first place and whose primary goals may have been the search for solutions to particular problems that have been answered since they entered the field. Once answered. they lost or lacked curiosity to go on in other areas.

The question really is, it seems to me, is the issue really worth discussing. If for the sake of argument one granted his claim, insofar as it is correct, what bit of difference would it make to the practice of science? Or to science policy?

My own view is Mr. Horgan's claims are dangerous to the extent that anyone might actually be deflected by them to change any practice or approach to science or science policy.

-Ernest B. Hook

Bravo John for a robust and entertaining defense of your thesis. I have a couple of points to make:

1. Life on Mars. As I am writing a book on this myself, I have thought a lot about the significance of the recent NASA "evidence". You are right that, if the features in the meteorite do turn out to be evidence for life on Mars, the chances are it came from Earth or vice versa. Clearly the planets are not isolated. However, it is possible to discriminate between contamination and an independent origin. Suppose Mars life was based on left-handed DNA, rather than right-handed as is Earth life. That would be strong circumstantial evidence that life had happened twice in the solar system. Then it is a dead cert that it has happened wherever conditions allow, and that the universe is teeming with life. This would surely be a major advance in science and a transformation of our world view, and would also demonstrate that the laws of nature are "rigged" to make the emergence of life inevitable. I agree that the latter position is regarded as ludicrous by most biologists (though not by Christian de Duve), but that is why the discovery of an independently-arising life form elsewhere would be so iconoclastic.

2. Consciousness. Maybe it is transitory, but we still don't know how it arises, or what it takes for a system to be conscious, or why qualia (assuming you believe in them) exist, as they serve no evolutionary purpose. Even if consciousness is not a fundamental aspect of our universe, it is still a mystery yet to be solved. You can't just shrug it aside as of no consequence because it may be limited to a tiny region of spacetime.

Congratulations on a stimulating essay!

Sincerely,
Paul Davies

(John Horgan said:) But I get frustrated (and I think George Johnson does too) by the excessively fawning stance of much science writing these days

If you can't fawn over Stephen Hawking (who is attacked by Horgan as being an overly conceptual post-scientist), who can you possibly fawn over? Or is fawning too undignified for the stultifying self-consciousness Horgan wishes to import from art theory into the sciences? Are we supposed to succumb to the same arms race of cynicism that has made generations of artists afraid to be comprehensible? The sad thing about Horgan is that we can make his ideas seem to be true on a temporary basis by not funding the superconducting supercollider, for instance, or ruining the motivation of young people who might bring their passions to the study of science mysteries.

All the best,
Jaron

J. D. Bernal, in The World, the Flesh, and the Devil: An Enquiry into the Future of the Three Enemies of the Rational Soul (New York: E. P. Dutton, 1929, p. 28), observed that "we are still too close to the birth of the universe to be certain about its death." Seventy years later, we may still be too close to the beginning of science to be certain about its end.

George B. Dyson

John's synopsis of his ideas is great — both challenging and revealing. The comments that follow refer to this synopsis, not the book as a whole.

In some ways it has always struck me that John's real issue is "the end of science stories": as in "I've been a science journalist a fair while now, and I seem to have written every big story there is, twice - there ain't much left to do." It's a feeling that I have every sympathy with. But I never had the nous to take it and expand it as john has done. It is from that sense of personal omission I offer these gentle observations.

Lets start with Kuhn. We all know that in The Structure of Scientific Revolutions Kuhn was a slippery bugger who can say more than one thing at once. He uses the terms in his book very loosely, which means that people (like John) who follow him base themselves on shifting ground. For example, my copy of SSR has a necker cube on the cover, and in various places within it Kuhn uses the gestalt shift model of such now-you-see-it-one-way, now-you-see-it-the-other optical illusions as an analogy to scientific revolutions. Its a good way of showing the incomensurability of the before and after world views — but at the same time it offers a view of scientific revolutions with no structure at all.

In short, Kuhn's book doesn't actually offer what it describes in the title. It is full of insights into scientific revolutions, but loose on how they are actually built. The one undeniable corollary that I drew from it was that if nature itself cannot be used as a way of distinguishing between before and after views (a point on which K is rightly insistent) then you have to look at some sort of explanation external to science for scientific revolutions.

To continue the niggling about titles, John's should in fact be "The end of scientific revolutions". John downplays Kuhn to some extent (especially when vague ideas of new paradigms are raised as objections — see infra) and says that revolutions are overplayed. but at the same time he has internalised Kuhn thoroughly enough to think that the revolution is the only worthwhile mode of major scientific advance. It is because he thinks that the intensive, massively funded and highly professionalised cadres of modern science have effectively revolution-proofed their sciences that he sees them as basically coming to an end. He allows that there's lots of good science on the structure of relativistic jets in Seyfert galaxies, the detection of gravitational waves from coallescing neutron stars, the unravelling of secondary messenger pathways, the identification of suspect terranes and so on, but it's a matter of filling in within the established frame rather than changing the frame.

Pre-Kuhn John's analysis would have seemed distinctly odd. People did not accept that world views before and after scientific revolutions were incommensurable. They saw only one sort of science, and it was pretty much what John calls "filling in the blanks". For most of the sciences they saw the frame which held the blanks as being pretty fixed. They knew there were atoms, there were forces, there was natural selection and so on. They saw all science as, in Kuhn's terms, normal science, though they realised its tempo changes.

Nor were they entirely wrong. Kuhn's ideas are very valuable, especially when applied to some aspects of the history of the exact sciences, and when used as a pointer to the importance of external factors in the history of scientific thought. But the bimodal revolution/normalcy model simply doesn't do justice to all of science, and so can't be used to justify John's claims about its end. Take one of the oddest of those claims:

Darwin's achievement-especially when fused with Mendelian genetics into the new synthesis-has rendered all subsequent biology oddly anticlimactic, at least from a philosophical perspective.

In short, if you never thought that Kuhnian revolutions were the be-all and end-all of serious science, its hard to get too worked up about the fact that the exact sciences don't have them anymore.

That said, the claim that revolutions don't happen any more is not that convincing. The last set of events that might really be called a revolution in particle physics culminated in october 1974. It was before John or I started writing but it's not that long ago. John may think that the establishment of the standard model was a minor revolution or not even a revolution at all — in which case physics ended with QED. I don't think that anyone could deny that plate tectonics was a revolution, and that happened within both our lifetimes. John would point out that I can't induce anything further from these recent revolutions; but I would reply that he can't induce anything from an open-ended lack of revolutions that happens to cover our relatively brief professional careers.

When we look back it may well be that the lull never really existed. It's far from clear to me that cosmology, for one thing, is as stable as John assumes. With 90% of the mass within our horizon not currently observable there seems to be still some possibility for a pretty big rethink of how it all fits together. Its not clear to me that our current cosmological history is that much more convincing than geology before plate tectonics (when the bottoms of the oceans were as uncharted as the missing mass is now).

Meaningless analogy? Then how about a thought experiment or two. Imagine a world with neutrino observatories and gravitational wave observatories; imagine an event "seen" by these and by traditional observatories that emits a gravitational wave signature bluntly incompatible with general relativity. I suggest that this is not inconceivable and would trigger a revolution. If general relativity is correct, then it won't happen. But general relativity, like all theories, is only correct so far. The fact that it is used by GPS satellites, though undoubtedly a Very Neat Thing About The World, does not mean that it is necessarily correct, any more than the fact that Newtonian physics after 1750 produced the most spectacularly accurate lunar tables for navigation means that absolute space and time was correct.

John might argue that the change from general relativity to some other theory of space time wouldn't matter to the "basic narrative". General relativity mattered only in the most tiny way to the "basic narrative" of the solar system. But it mattered a lot to science. Scientists aren't that concerned with the basic narrative; they have to have theories that are right. (Journalists communicating with the lay public, on the other hand, have to care about the basic narrative.) Science is not just about the cast of characters — its about explanation and prediction. You can have science that believes in galaxies, quarks and natural selection for centuries with ever shifting explanations and prediction underneath.

Or imagine an experiment that through some neat insight no one has yet had differentiates between the many-worlds view of quantum mechanics and a copenhagen or allied interpretation. I'm thinking of something like the Aspect experiments here. Now suppose it showed that we do in fact live in one of an infinite number of very loosely interacting parallel worlds. That, I submit, would be a revolution.

They may not happen; but to be convincing John has to argue that they can't happen, and i don't think he does.

A couple of other points: John throws around the idea that things are unverifiable a fair bit. Without going the whole Popperian hog, falsifiability counts too, as the two thought experiments above show. And some of what John sees as irony is falsifiable. Unless I have misinterpreted Jim Lovelock a lot, the discovery of a stable crypto-ecosystem in some Martian nook or cranny that has been there for billions of years but not done much else would be a falsification of the whole Gaia project, which is based on the insight that life is necessarily a planetary phenomenon. Similarly, if the LHC reveals a range of hypercolour particles and no trace of the proposed supersymmetric particles then I would suggest that superstrings, of which supersymmetry is taken to be a low energy manifestation, would be in a lot of trouble. Both Gaia and superstring theory could be rejigged as a result of such discoveries, but all theories can be rejigged to meet the facts, as Duhem and Quine pointed out. And as Lakatos added, such ad hoc rejigging is the sort of thing that loses your research programme its credibility.

And he's also not above egregious appeals to authority. When Stephen Gould ("a real expert on life") says that most reruns of history won't lead to consciousness, he strikes me as indulging in profoundly "unverifiable" "ironic" science. At least the SETI teams are trying to find some evidence. And the fact that Louis Wolpert says embryology is basically over is fine, but my deep fondness for Louis, and my respect for his absolutely vital foundational work in the field, doesn't mean that I don't think he can be wrong. In fact, I think he's very frequently wrong. The idea that Louis Wolpert is more likely to be right than Natalie Angier, regardless of the arguments the two employ, just because he has been a leading light in development research is a bit like saying that John Casti is necessarily right and John Horgan necessarily wrong because John Casti is a leading light of Santa Fe.

Which leads me to a final point. John takes the current lack of novelty in some areas of science as one sort of evidence for his claim; he takes the ferment in others as a different sort. He's entertainingly rude about consciousness and complexity as new areas to look to for the future, and I share some of his doubts. But I also suspect that these areas are exactly what paradigm shifts look like early on. One of Kuhn's most important points was that science — a body of social practise as well as one of reliable knowledge — changes its mind in revolutions not just about the way the world is, but about what tools, practises, concepts and moral stances are necessary to perceive it correctly. Its not inconceivable to me that in 30 years all scientists will see computer simulations in a new epistemological light. Nor is it that strange to me to think that the distinction between objectivity and subjectivity will be radically renegotiated in the course of consciousness research. Either of those things would be revolutionary.

Again, they don't have to happen, but they could happen; the nature of what counts as science could change. And that would not be an end of science, just and end to one view of science, the outmoded view in which the new conceptual world made no sense.

In short: John's view of what science is requires revolutions as signs of progress. He argues that these grand conceptual shifts are no longer happening. However, he does not show that they can't happen. He does not show that sciences that have not been subjected to revolutions are moribund in any practical sense — to treat biology as "philosophically" over is just evidence of a poor choice of philosophies. And when confronted with areas of potentially profound conceptual change he does not accept that they may be revolutions in the making, because he does not see how their product can count as useful according to the standards of the pre-revolutionary world view.

He does show that there is something deadened at the moment about the exact sciences allied to physics. And he does show that calling for new paradigms is pointless if the researchers in the field can't be made to take up the call. He shows that science is faddish, and that critical journalism is very entertaining and probably quite useful. And he shows that having been an English major can be turned to profitable use later in life.

As I've said several times before, John's argument is not silly, and I don't think he is making it in bad faith. He is also an interesting person, who I enjoyed meeting some time ago. But I believe he is wrong, and it is not hard to explain why. The basic reason is that the "map of reality" and "narrative of creation" that he described, while enormous achievements, are full of holes, unanswered fundamental questions and, in some cases, basic inconsistencies. This is because the scientific revolution that produced these achievements is not yet finished, but has some way to go. An indication of how much of this revolution remains unfinished can be gotten by writing down a list of questions that we cannot yet answer:

How do cells differentiate into different cell types?

How does a single cell develop into a coherent organism?

Why are there procaryotes and eucaryotes, but apparently nothing in between?

What is the exact story of how life began?

How does the brain work?

How did the galaxies form?

What is responsible for the large scale structure of the galaxies?

What keeps the star formation rate of many spiral galaxies constant in time?

Why were the initial conditions in the early universe so symmetric?

What are the reasons for the values of the twenty-odd parameters of the standard models of particle physics and cosmology?

Why do those values have the property that they make it possible for stars, galaxies and complex chemistry to form?

How is gravitation consistent with quantum phenomena?

What happens inside of black holes?

What happens at the end point of black hole evaporation?

Each of these questions is the focus of intensive work by thousand of very bright young and not so young scientists. Among people engaged in this work there is a strong sense of optimism that the next years will see dramatic breakthroughs. Each of these will add to the "map of reality" knowledge as fundamental as anything discovered in the twentieth century and all will, sooner or later, lead to theories that are verified by observation and experiment. Even in quantum gravity and string theory (where there have been dramatic breakthroughs in the last years) there is a growing list of experimental predictions. (I wrote a paper some years ago cataloging the experimental predictions of quantum gravity and string theory, and the list has grown since.) It is true that these cannot yet be carried out, but I would not want to be in the position John is of betting against the possibility that the thousands of bright people working in these areas around the world will not find a way to carry out these tests.

For more arguments against the "end" of science, please see my last exchange with John in Time In Quantum Cosmology. (By the way, I notice that John seems to have stopped defining ironic science as science that could not even in principle be tested experimentally, given the ease with which even string theory evades that.)

But in closing, I would like to remark that I find John's stance disturbingly characteristic of the present moment. We are in a time of extraordinary change, with positive developments all around us. In the last years democracy has expanded dramatically and the danger of war and reach of totalitarianism has receded, the economy is stable and growing, our nation is being reinvigorated by a new wave of immigration, amazing things are happening in the arts, theater, dance, while in science and medicine there have been a slew of breakthroughs, from effective treatments for AIDS and certain forms of mental illness to all the new observational data in astronomy and cosmology.

So why are people so pessimistic? Why is there so much talk of the end of this and that? This for me is the great unanswered question of the present moment.

The following are a few relevant comments from my article in the May, 1997 issue of Prospect magazine entitled "Exaggerating the Death of Science:"

The first question to ask the purveyors of the idea of the end of science is how they know that there are no deeper layers of the particle world yet to be probed. Of course, they do not. To put things in perspective, there is some reason to think that the smallest scale on which any particle could have a meaningful existence would be the so- called Planck scale, where quantum effects make even the concepts of space and time fuzzy. I won't bore you with all the zeroes that have to be written out after a decimal point before you get down to the Planck scale, but in terms of ratios the distance we still have to go from the scale of quarks to the Planck scale is the same as the ratio of the diameter of an atom to the orbit of the Moon. The orbit of the Moon was first satisfactorily described by Isaac Newton a little over 300 years ago; quarks were explained a little over 20 years ago. Even allowing for the greater pace of science today (something which is far from assured in these days of financial constraints), it seems to smack of overconfidence to expect as much progress in the next 10 years as in the past 300 years.

....

Even if all the particles and forces are already known, and even if a TOE is found in the next 10 or 20 years, would that leave physicists out of work? Far from it. The great physicist Richard Feynman used to make an analogy with a game of chess. A child of five can learn the rules — how a knight moves, the role of the pawn, and so on. Indeed, a child of 13 has just become an International Master. But the greatest chess player who ever lived can spend a lifetime applying those rules, and still find new ways for them to interact, producing new games of chess. The equation that could be written on a T-shirt would not be the last word in physics, but the basic rule book, from which you would still have to explain the complexity of the Universe around us.

That, indeed, is the key to the next development in physics — complexity. It's all very well speculating about probing deeper into the structure within the atom, but what about the structure in the Universe at large? If the Universe began in the hot fireball of a Big Bang, some 15 billion years ago (as an overwhelming weight of evidence suggests), how did it evolve to produce galaxies and stars, planets and people?

The key word here is "evolve". Forget the books that tell you the end of science is in sight. The most important science book published in 1997 is undoubtedly one which carries the opposite message, and which also carries the rather startling title The Life of the Cosmos. It comes from Lee Smolin, a physicist based in New York, and it elaborates a theme that has been developed over the past few years in serious scientific circles by Smolin himself, Andrei Linde, in California, and a handful of other researchers. Their thesis is that the way the Universe works can best be understood not simply by applying the rules of physics worked out by Newton and Einstein, but by taking account as well of the rules of evolution worked out by Darwin — the theory of natural selection. The Universe itself, and its major components (notably galaxies like our own Milky Way) may literally be alive, on this picture, and, more to the point, may have evolved by natural selection from a simpler state to produce the complexity we see around us.

-John Gribbin

As a human chauvinist, I find life and consciousness to be the most exciting things that are happening in this Universe. Nothing in the chemistry of the mixture of small molecules and tars that is found in common meteorites or in space suggests that such substances could give rise to the marvelous intricacy of a bacterial cell; yet that has happened. We know nothing of the process. We are led to wonder: what other combinations of matter and energy, perhaps very different than ours, could also support such a process of complexification? What unexpected qualities would the results of such a process possess?

Further, nothing about a bacterium suggests that it could evolve into conscious beings capable of supporting a technological civilization. Yet that has occurred, over four billion years. Further questions come up. What marvels might be produced if evolution followed another path, rather than the one toward consciousness? What possibilities exist for further evolution of our species in the direction of enhanced consciousness?

We will not find the answer to these questions quickly, nor is there any need to do so. No urgent deadlines await us (though we should remember to get off the Earth before it becomes uninhabitable, 1-5 billion years from now).The Universe is 12-15 billion years old. We have been carrying out the business of science for a few centuries at most. Perhaps after a millenium or two, it will be time to look back and ask "How're we doing". But there is no need now to talk about the end of science (unless we mean an abrupt end, such as the explosion of a nearby supernova or the permanent collapse of civilization). We are only at the beginning.

The following is my response to the responses of Kevin Kelly, George Johnson, Ernest B. Hook, Paul Davies, Lee Smolin, Jaron Lanier, George Dyson and Oliver Morton to "The End of Horgan."

1. Kevin Kelly, you say you cannot believe certain theories will never be verified unless I can propose a method for conclusively demonstrating unverifiability, a priori. See, I think I have such a method. It's called common sense. Common sense tells me — and the vast majority of other scientists, by the way — that we will never have any empirical evidence for the existence of parallel universes. I therefore conclude that such theories will forever remain mere speculation.

Remember the scene in Animal House when Donald Sutherland got a couple of students high on pot and then started telling them that our entire universe is just a speck of dust in the pocket of a giant, and that each atom in our universe is also a universe in its own right with its own inhabitants (or words to that effect)? Superstring theory and baby-universe models are pretty much the same thing, in my mind. But if you want to believe in their potential actuality, that's cool. By the way, do you also believe that there is a benign God watching over this vale of tears?

2. George Johnson, you don't give me much to disagree with. You note that I use different arguments for dispatching different fields of science. Quite true. You also mention that my book was originally called "The Ends of Science." Also true. Originally I was worried that calling it "End of Science" would make it too easy for reviewers to accuse me of jumping on the end-of-something-big bandwagon. "Ends" was more subtle. It evoked different kinds of endings as well as the different reasons for doing science. But since the central message of my book is that pure science has in a certain sense already ended, and since my years as a journalist have taught me that subtlety is often lost on the masses, I decided to go with the blunt instrument: "End of Science." I have no regrets.

By the way, I also agree with you that science is not complete, and that there are many mysteries left unexplained. You say you may write about these mysteries in "sequel" to my book called "Loose Ends of Science." Great idea, but you should know that the afterword for my book's new paperback edition is called "Loose Ends."

3. I don't know who you are, Ernest P. Hook, but you certainly have a talent for blurbs. I hope to get "Useless and mildly pernicious!" on the next printing of my book, along with Lee Smolin's "not silly!" You say first that my end-of-science thesis is not worthy of consideration by research scientists, because it has no practical consequences. You then express concern that students might be discouraged from entering certain fields because of my arguments. These are of course contradictory statements.

If you had been nicer in your posting, I might have told you about my views on science funding and on science as a career for young people, both of which you probably had in mind when you raised this issue of practical consequences. But because you were so snippy, I'm not going to tell you what I think here. You will have to buy the paperback edition of my book and read the afterword. Or remain forever ignorant. Your choice.

4. Paul Davies, you raise a couple of important points in your usual lucid fashion. If we find life on Mars, that would be very cool, especially if it employs a genetic mechanism significantly different than ours. That discovery, which as you say would suggest that life arose independently on the two planets, would certainly provide more substance to speculations about the likelihood of life elsewhere in the cosmos. But I bet reasonable people would still disagree, as they do now, on whether the universe is "teeming with life," and settling that issue definitively would still be extremely difficult. I hope we find a way to do it, but we can't assume we will.

As for consciousness, I did not mean to imply that I don't think it's an important or interesting problem. I'd say it's the MOST important and interesting problem, excepting perhaps the more general question of why there is something rather than nothing. I just don't think science can "solve" consciousness in the way that people want it solved, any more than molecular biology "solved" the mystery of life. Daniel Dennett recently proposed that the problem of consciousness is really just a sub-problem of the larger problem of life. That makes sense to me. There is nothing in the laws of physics or chemistry that says either life in general or conscious life in particular must exist (although romantics such as Christian DeDuve and Stu Kauffman strain to demonstrate otherwise). Thus, even when we have "explained" them in material terms — with DNA and neurotransmitters and the like — they will remain profoundly mysterious.

5. Lee Smolin, you've compiled a nice list of questions left for science to ponder. But some of these questions, such as how cell differentiation occurs, I would put in the category of filling in details of pre-existing paradigms. Others, such as why the early universe was so "symmetric" and how "exactly" life began, I classify as probably unanswerable in a definitive sense given the dearth of reliable data.

Lee, you say you are also disturbed by my pessimism, which you see as part of a larger intellectual trend. Come on. Pessimists are a vanishingly small group in this gee-whiz, can-do culture, and especially in the realm of science. Do you really think we'd all be better off if these few voices of dissent were silenced?

6. Jaron Lanier, like, lighten up, dude. Just because science is over doesn't mean we can't have fun any more. As the science editor of the Economist told me recently when I met him in London, "We still have sex and beer!" And virtual reality, of course.

7. George Dyson, nice quote from Bernal, whose essay on the long-term evolution of intelligence (humanity is just a stepping stone, we will eventually turn into really smart clouds of conscious gas) inspired your father's wonderful musings on the distant future. I love this stuff, which in my book I call scientific theology, but you've got to admit it's kind of escapist. Our fate, I suspect, will be forever tethered to this tiny planet, and these frail bodies.

8. Oliver Morton, I only wish all critics could be as interesting and witty as you are. On the other hand, you make so many good points that I really don't know where to begin. I'll just make a few general observations, in the hopes that someday we'll be able to jaw at each other further over beers. You offer a list of what-ifs that, fulfilled, could jumpstart science in the future. All of these things — which range from discovering evidence for the many-worlds interpretation to contacting intelligent aliens — are indeed possible. My argument is just that they are not inevitable or even probable, as most popular writing about science suggests.

Also, while it's conceivable that general relativity and even quantum mechanics could yield to even more effective conceptual and mathematical formalisms — superstring theory, say — I suspect such a development will ultimately matter only to specialists. We will still live in the same old material world, made of matter and energy, particles and forces, time and space. That paradigm, if you will, is rock solid.

And I very much doubt — and I know you do too, having discussed the issue with you — that such new-and-improved formalisms would lead to blazing revelations about consciousness, as Roger Penrose and others of his ilk have suggested.

You say that superstring theory, while perhaps not confirmable, is at least falsifiable. No supersymmetric particles, no superstrings. But remember that Andrei Linde and Alan Guth and others once insisted that inflation would definitely, positively be falsified if the cosmic mass-density did not equal one. Well, now that it looks like the mass-density falls well short of that, that prediction has fallen by the wayside. This kind of thing is rampant in physics and cosmology these days, and it's one reason why these fields are obviously in such trouble.

You were right to jump on my remark about post-Darwinian biology not offering much of philosophical significance. Dolly the sheep certainly had some interesting implications (although not as interesting as some overheated commentators suggested). I guess my point is that the hard-core atheism and materialism of a Richard Dawkins or Steve Gould (I have a theory that these guys only pretend to disagree about stuff to make their field seem more exciting) could have been expressed immediately after Darwin, and was, pretty much, by Neitzsche.

I love talking about Kuhn, as you obviously do, too. I'm rather fond of the part of my book that deals with Kuhn and his slippery ideas. If you read that, you'll see that you overestimate the degree to which I demand that really big scientific advances be Kuhnian in nature. My view of science is much more straightforward than Kuhn's. For example, Kuhn would consider it terribly naive to compare scientific progress to geographical discovery. I don't think it's really such a bad analogy. We have done a good job mapping out the universe, and we are now speculating about things that are off the map, as it were: parallel universes, the Planck realm, extra dimensions of space and so on. There be dragons.

I think more in terms of big questions than of Kuhnian revolutions and incommensurability (a concept whose usefulness has been greatly exaggerated). The biggest questions now are what I identified in my essay as "inevitability questions." How inevitable was the universe, the laws of nature, life and so on? Most of us want to believe that we were inevitable; religion is a manifestation of this longing, and so are all our "scientific" speculations about other universes and other biosystems. My own feeling is that, even if we transform the entire universe into an omniscient cosmic computer, it will still be stumped by the brute fact of its own existence — proving once again that AI can't live up to its hype. But remember, we still have sex and beer.

Life after Death - Response to Horgan:

I can't quite warm up to Horgan's death sentence for science. My chills are not from a deep need to protect the concept of science but from a really specific instance. I just can't see what he's getting at when he talks about my field - neuroscience.

The main argument about the end of science, if I've got it right, is that the progress has been so great that we have nothing big left to figure out. But when it comes to neuroscience, where he says we face one of the biggest problems - the mind, our field is ending because we'll never figure the big one out.

Can't we get a youth discount? Neuroscience is infantile. We can't have a paradigm shift since we don't have a paradigm. Maybe we'll never have one. But maybe it's too soon to tell. Either way it seems terribly closed minded to say we'll never figure out how the mind works. If we give up at this early stage we'll certain never get there.

Still, it's important to point out that the study of consciousness is just a minute part of neuroscience (though the only part of the field he discusses in his book). There is, after all, the question of how the brain works, in addition to the one about the mind. In fact, many neuroscientists think they are working on the brain rather than the mind. The brain does many non-mental things that are important, like keeping our lungs inhaling and exhaling at the right speed, making sure the heart pumps away, controlling posture and locomotion, regulating digestion, and on and on. Though certainly less sexy than consciousness, these are more important for survival. We can live a long time without a belief, but not very without a breath.

But even if we go back to the mind, there's much more to figure out than consciousness. Most of the mind works unconsciously. That's not to say its operations are repressed or otherwise hidden from consciousness. Instead, it means that consciousness (at least what we humans refer to when we talk about consciousness) is something that was added to the brain recently (in evolutionary terms). It was layered on top of all the other stuff that was already there. Consciousness has access to some of that stuff, but not all of it. In fact, most of it is inaccessible. Much of our brain's functions operate unconsciously simply because those processes are not available (neurally) to consciousness. And many of these processes fall into the domain we call "mental." Speaking grammatically, for example, is done without willful participation of consciousness, as is our initial response to danger or beauty. And the breakdown of mental life that occurs in mental disorders is due at least as much to changes in these implicitly operative systems as to alterations in consciously controlled processes.

The foregoing implies that we know a lot about the brain, otherwise how could I say so much about it. But sadly we know very little. We have no idea how our brains make us who we are. There is as yet no neuroscience of personality. We have little understanding of how art and history are experienced by the brain. The meltdown of mental life in psychosis is still a mystery. In short, we have yet to come up with a theory that can put all this together. We haven't yet had a Darwin, Newton, or Einstein.

Don't get me wrong, I'm not proud that our field has yet to achieve a grand theory. On the other hand, I'm not even sure that we need one. Maybe what we need most are lots of little theories. It would be great to know how anxiety or depression works, even if we don't have a theory of mental illness. And wouldn't it be wonderful to know how we experience a wonderful piece of music (be it Bach or rock), even in the absence of a theory of perception. And to understand fear or love in the absence of a theory of emotion in general wouldn't be so bad either. The field of neuroscience is in a position to make progress on these problems, even if it doesn't come up with a theory of mind and brain. In Horgan's terms, this may mean the field is dead. If so, then we can look forward to a long and wonderful life after death.

Horgan will surely have something clever to say in response. I'm prepared to be ripped to shreds. But I'm not prepared to concede infant mortality to neuroscience.

In his December posting, Joseph LeDoux argued that when it comes to his own field, neuroscience, my "death sentence" for science is grossly premature. His main argument is that "neuroscience is infantile." That is simply not true. Galvani showed two centuries ago that nerves emit and respond to electric current, and around the same time Gall produced a precursor of the modular theory of mind that Steve Pinker and others are now touting. William James wrote Principles of Psychology in 1890, and Freud began setting forth his psychoanalytic theory shortly thereafter (after producing a solid book on aphasia). Meanwhile, Cajal and others were already unraveling the structure and function of neurons.

Here we are 100 years later, in the era of MRI and PET scans and DNA probes and computer models and microelectrode recordings from individual neurons. We have compiled a huge amount of information about the brain and its role in perception and behavior. Yet as LeDoux acknowledges, "we know very little. We have no idea how our brains make us who we are. There is as yet no neuroscience of personality... The meltdown of mental life in psychosis is still a mystery."

So LeDoux and I agree on the current plight of his field. The question is, just how far will neuroscience go in the future, given how little progress there has been to date? I grant that my treatment of neuroscience in THE END OF SCIENCE was a bit superficial; as LeDoux points out, explaining consciousness should not be the be-all and end-all of neuroscience.

I hope to atone for my sins in a new book on neuroscience and other mind-related fields, including psychiatry, behavioral genetics, evolutionary psychology and even artificial intelligence. I fear that, given their poor record to date, some of the most critical problems addressed by these fields may be intractable. I'm not just talking about the old thumbsuckers like consciousness and free will and nature/nurture but practical problems like mental illness.

A couple of months ago, I spent a disturbing morning watching patients at the Psychiatric Institute in New York City receiving shock therapy. If I were suicidally depressed, I might submit to shock treatment myself; incredibly, it is the most effective treatment for severe depression that we have. But the relapse rate is extremely high, as much as 90 percent. And no one can pretend that this is a "scientific" therapy; it's the equivalent of kicking a TV set on the blink.

I am certainly not the only observer who has come to a pessimistic conclusion about the future of mind-related science. The Harvard psychologist Howard Gardner, a contributor to this site, states in a 1992 essay, "Scientific Psychology: Should We Bury It Or Praise It?": "Psychology has not added up to an integrated science, and it is unlikely ever to achieve that goal." Gardner speculates that in the future psychology, rather than becoming a "hard" science like physics, will become more literary in nature. Gardner tries his best to make this prophecy sound hopeful; to me, it sounds like an acknowledgement of defeat.

LeDoux is absolutely right that we will never know if such pessimism is warranted if we give up now. Let me assure him that I believe, and hope, his field has a vital future. If I were advising my own kids on what field of science to study, neuroscience would be my first choice (and particle physics the last). The problems of neuroscience are profoundly important - both intellectually and pragmatically. The fact that these problems may also be unsolvable means that the field will never die.