Mistake of the Week: Belief is King

A couple days ago the Times ran a much-debated story about Marcus S. Ross, a young-earth creationist who completed a PhD in geosciences at the University of Rhode Island. Apparently his thesis was a perfectly-legitimate study of marine reptiles that (as he writes in the thesis) went extinct 65 million years ago. Ross merely disavows the entire materialistic paradigm of which his thesis is a part.

If you want some long, acrimonious flamewars about whether the guy’s PhD should be revoked, whether oral exams should now include declarations of (non)faith, whether Ross is a walking illustration of Searle’s Chinese Room experiment, etc., try here and here. Alas, most of the commentary strikes me as missing a key point: that to give a degree to a bozo like this, provided he indeed did the work, can only reflect credit on the scientific enterprise. Will Ross now hit the creationist lecture circuit, trumpeting his infidel credentials to the skies? You better believe it. Will he use the legitimacy conferred by his degree to fight against everything the degree stands for? It can’t be doubted.

But here’s the wonderful thing about science: unlike the other side, we don’t need loyalty oaths in order to function. We don’t need to peer into people’s souls to see if they truly believe (A or not(A)), or just assume it for practical purposes. We have enough trouble getting people to understand our ideas — if they also assent to them, that’s just an added bonus.

In his Dialogue Concerning the Two Chief World Systems, Galileo had his Salviati character carefully demolish the arguments for Ptolemaic astronomy — only to concede, in the final pages, that Ptolemaic astronomy must obviously be true anyway, since the church said it was true. Mr. G, of course, was just trying to cover his ass. The point, though, is that his ploy didn’t work: the church understood as well as he did that the evidence mattered more than the conclusions, and therefore wisely arrested him. (I say “wisely” because the church was, of course, entirely correct to worry that a scientific revolution would erode its temporal power.)

To say that science is about backing up your claims with evidence doesn’t go far enough — it would be better to say that the evidence is the claim. So for example, if you happen to prove the Riemann Hypothesis, you’re more than welcome to “believe” the Hypothesis is nevertheless false, just as you’re welcome to write up your proof in encrusted boogers or lecture about it wearing a live gerbil as a hat. Indeed, you could do all these things and still not be the weirdest person to have solved a Clay Millennium Problem. Believing your proof works can certainly encourage other people to read it, but strictly speaking is no more necessary than the little QED box at the end.

The reason I’m harping on this is that, in my experience, laypeople consistently overestimate the role of belief in science. Thus the questions I constantly get asked: do I believe the many-worlds interpretation? Do I believe the anthropic principle? Do I believe string theory? Do I believe useful quantum computers will be built? Never what are the arguments for and against: always what do I believe?

To explain why “belief” questions often leave me cold, I can’t do better than to quote the great Rabbi Sagan.

I’m frequently asked, “Do you believe there’s extraterrestrial intelligence?” I give the standard arguments — there are a lot of places out there, the molecules of life are everywhere, I use the word billions, and so on. Then I say it would be astonishing to me if there weren’t extraterrestrial intelligence, but of course there is as yet no compelling evidence for it.

Often, I’m asked next, “What do you really think?”

I say, “I just told you what I really think.”

“Yes, but what’s your gut feeling?”

But I try not to think with my gut. If I’m serious about understanding the world, thinking with anything besides my brain, as tempting as that might be, is likely to get me into trouble.

In my view, science is fundamentally not about beliefs: it’s about results. Beliefs are relevant mostly as the heuristics that lead to results. So for example, it matters that David Deutsch believes the many-worlds interpretation because that’s what led him to quantum computing. It matters that Ed Witten believes string theory because that’s what led him to … well, all the mindblowing stuff it led him to. My beef with quantum computing skeptics has never been that their beliefs are false; rather, it’s that their beliefs almost never seem to lead them to new results.

I hope nobody reading this will mistake me for a woo-woo, wishy-washy, Kuhn-wielding epistemic terrorist. (Some kind of intellectual terrorist, sure, but not that kind.) Regular readers of this blog will aver that I do have beliefs, and plenty of them. In particular, I don’t merely believe evolution is good science; I also believe it’s true. But as Richard Dawkins has pointed out, the reason evolution is good science is not that it’s true, but rather that it does nontrivial explanatory work. Even supposing creationism were true, it would still be too boring to qualify as science — as even certain creationists hunting for a thesis topic seem to agree.

Or anyway, that’s what I believe.

60 Responses to “Mistake of the Week: Belief is King”

  1. Matt Hellige Says:

    These points are fair and worth making, but there’s another useful way of interpreting “What do you believe?” questions asked of scientists. Given a particular question, an expert in the field presumably has a greater grasp of the evidence for and against than he or she could explain to a layperson in simple, concise terms. So asking such an expert what he or she “believes” is a way for the layperson to use the expert as an evidence filter, and can be a much more useful question than “What is the evidence for and against…?”

    This underscores a second role of the scientific establishment: to serve not only as evidence producers but also evidence interpreters, when the nature or quantity of the evidence places it out of the casual understanding of the layperson. Of course this “priestly” role of the scientific establishment is one which produces a certain amount of discomfort at times, but it is a necessary role. If I ask you “Do you

  2. Matt Hellige Says:

    [misfire on the last comment, sorry…]

    Anyway, in summary, it often matters more to me as a layperson how a knowledgeable expert understands and interprets the evidence than how I might (ignorantly) interpret the evidence, or even in many cases what the evidence is.

  3. Scott Says:

    Thanks, Matt! You make an excellent point. Indeed, if you read this paper by Cowen and Hanson, you’ll see that there’s actually some theory behind the idea that knowing someone’s opinion can be almost as useful as knowing the evidence on which their opinion is based. (Crucially, though, this theory works under the assumption of Bayesian agents with common priors — an assumption that emphatically doesn’t seem to be satisfied in cases like evolution vs. creationism.)

    If Ross (for example), when asked to evaluate Darwin’s arguments about the geographic distribution of species, were to say that they’ve been conclusively rebutted by ID theorists, I would indeed call that academic fraud. But if he said that, while Darwinism certainly seems compelling from an evidence-based standpoint, you should nevertheless believe creationism because otherwise you’ll burn in hell, I would not call that fraud, just perversity.

  4. Robin Blume-Kohout Says:

    Scott, I’m really glad you posted this. As I’ve read the media coverage of this story for the past couple of days, I’ve been looking in vain for somebody to clearly say the obvious thing: “What the hell does it matter what he believes? The whole point of science is that it’s not relevant what he believes!” I was starting to doubt my own sanity.

    (Okay, maybe I should, but that’s a separate topic.)

    Point is, I agree wholeheartedly. What’s particularly terrifying to me about this whole episode is the fraction of people (greater than 10%, at least) who seem to believe that there should be some kind of litmus test, where you have to affirm that you believe some kind of party line in order to get a PhD, for crying out loud. Presumably, these people haven’t devoted, oh, 5 minutes of thought to what that would do to academic creativity, objectivity, and credibility.

    Matt: I’m sorry about this, but I think you’re painfully wrong. Academics should not have a priestly role, and non-experts should not ask an expert “What do you believe?” in order to get an expert answer. That way lies great danger.

    A priest is evaluated precisely on what [s]he believes, and how [s]he will guide others. This requires dogma — to become a certified priest, you must believe certain things, and you are examined to ensure it. A scientist is evaluated on what [s]he knows and how [s]he thinks. We try to select the absolute best people at those tasks. We do not select them for their beliefs.

    Treating scientists as priests (i.e., asking them for guidance on beliefs) means one of two things must happen. Either you’re asking someone for something they’re not necessarily qualified to give (belief advice), or the scientific establishment is going to have to start selecting for priestly qualities, which will compromise our ability to choose the most knowledgeable and thoughtful people.

    So, please, don’t just ask Scott what he believes. Ask him “What’s your evidence for believing that?” and make up your own mind. He’s got a lot more evidence for “P is in BQP” than he does for Flying Spaghetti Monsterism.

    [apologies to the proprietor of this establishment…]

    P.S. Scott, didn’t this page used to have a “preview” button?

  5. Scott Says:

    Robin: Thanks for the comment! The reason I ceded some ground to Matt is that I know that, against their best instincts, scientists do occassionally have to grit their teeth and summarize their evidence in the form of a belief: “Will the bridge fall down?” “Does tobacco cause lung cancer?” “Is the earth getting warmer?” The mistake arises in seeing beliefs as central to how science operates.

    I do indeed have very strong evidence that P is in BQP. 🙂

    No, I never had a “preview” button — can anyone familiar with WordPress tell me how to add one?

  6. wolfgang Says:

    > Even supposing creationism were true, it would still be too boring to qualify as science

    Yes indeed. A very good point which does not get enough attention in this debate.
    If ID and creationism would win and schools have to teach it: What exactly would they teach about it ?

    How many times can you read the first two chapters of Genesis?

  7. milkshake Says:

    What they could teach is a long list of examples taken from anatomy and botany that demonstrate and re-confirm the perfect wisdom of the Creator. (This would be more economical than doing experiments and excavations, also monk teachers don’t eat much so the tenure would be cheaper).

  8. Robin Blume-Kohout Says:

    Scott writes, “I know that, against their best instincts, scientists do occasionally have to grit their teeth and summarize their evidence in the form of a belief.” Okay, okay, fine. You’re right, and I know it. And, furthermore, I tend to think that if you are going to go and ask somebody what to believe, then a scientist is probably, on average, a better bet than a randomly selected person. And, if I were pulled by wild horses, I would admit that we all have to take experts’ words for most things, most of the time.

    But, darn it, I can be idealistic, and pretend that we all gather and evaluate the evidence (with some help from experts) on our own. Maybe a more fair suggestion on my part would be, “Don’t ask a scientist what he believes; ask him what he believes the evidence indicates.” And scientists should answer the question “What do you believe?” as if it were “What do you believe the evidence indicates?”

    I’m just very leery about anything that implies a responsibility for scientists to be anything extracurricular… and trusting us to speak ex cathedra would do just that. However, your point about the Cowan & Hanson paper is well taken (we cross-posted; I hadn’t seen your comment) — perhaps our responsibility can be boiled down to “When speaking professionally, attempt to simulate a Bayesian agent whose prior has bounded relative entropy with the mean prior over your professional society.” Now that we could test at orals — probably efficiently!

    Wolfgang: I think your argument misses an important point — certain folks believe that everything you need to know can be learned by 8th grade. ID is great from this point of view. By age 13, you can get on with being fruitful and multiplying (you’re don’t need to know that you’re actually exponentiating…)

  9. Bemused... Says:

    I guess Marcus Ross also deserves an honorary degree in compartmentalization. 🙂

  10. Anonymous Says:

    Clearly science should not have any litmus test for belief. If you’re a young-earth creationist but do good particle physics, your physics publications should not be judged on the basis of your crazy beliefs.

    On the other hand, there’s a real issue of intellectual honesty here. If you publish a paper or a thesis, there’s an implication that you at least believed the results as of the time you were writing it. Here’s a thought experiment:

    Imagine that your branch of science relies on Test X, a procedure widely believed to test for the presence of Chemical Y. Researchers believe this has been well established and papers in the field make frequent use of Test X (always drawing the conclusion that Chemical Y is present when Test X gives a positive result).

    After running some unorthodox experiments, you come to the conclusion that Test X is unreliable, and that it may fail for any of several subtle reasons. When you mention this to Dr. Smith, an expert in the field, he says that he has known that for years. Long ago, he discovered that Test X didn’t do what everyone else thought it did, but he didn’t publicize this discovery because he was afraid of controversy. Instead, he continued to write papers using it without comment, in the standard way. In the meantime, if anybody happened to ask him in private whether Test X really worked, he would say no, but there was no hint of this in his publications.

    Dr. Smith was wrong to do this: it’s a grotesque example of intellectual dishonesty to write papers that agree with the mainstream consensus while ignoring evidence against it.

    How is the case of Marcus Ross different? The only difference I see is that we all think Ross’s private creationist beliefs aren’t in fact based on evidence at all. That may be true, but there’s a good chance he at least thinks they are evidence-based. If asked to justify his beliefs, he would likely reply with a long list of inconsistencies in evolution, objections to carbon dating, etc. If he thinks these beliefs are justified, then isn’t it wrong of him to write scientific papers that deliberately exclude what he considers to be compelling evidence?

  11. Chris W. Says:

    WGBH’s Radio Open Source devoted its show tonight to a discussion of Baruch Spinoza (Spinoza: Mind of the Modern). Christopher Lydon’s guests were Rebecca Goldstein and Antonio Damasio.

    Rebecca Goldstein referred to the fact that as an young Orthodox Jew she was advised that if she must (as a woman) go to college, then she should at least avoid philosophy, because it sews doubt.

    As for scientists being pressed to state their beliefs, as guidance to others on what to believe, one should mention Einstein’s famous remark:

    To punish me for my contempt for authority, fate made me an authority myself.

    One more thing: I just found an interesting interview with Goldstein:

    Luke: “You’ve seen a bit of Jewish life around the world and around the United States.”

    Rebecca: “I’ve even been a scholar in residence at various synagogues.”

    We laugh.

    Rebecca: “I always feel like a terrible fraud.”

    Apparently Marcus Ross doesn’t see any reason to feel like a fraud.

  12. Mihai Says:

    Since the flame war is already starting strong, I won’t feel morally guilty to add my own dissenting opinion 🙂

    Scott, science kind of existed in the middle ages, and likely it will always survive somehow, due to human nature. But the great (one could almost say quantum) leaps that have happened in the past century, or centuries, are due in no small part to a changed perspective among non-scientific decision factors. We came from burning scientists in plazas to giving them government grants to do whatever they want.

    This was the consequence of a long and assiduous effort to publicize science to laymen, to promote it among governing bodies, etc. We need scientists who are willing to talk ex cathedra, act as experts, and share with people the occasional interesting “scientific beliefs”. This is not for the spirit of science, it is for the well-being of science in the world. And as a scientist, I think you will agree that this is ultimately for the betterment of mankind.

  13. Chris W. Says:

    That may be true, but there’s a good chance he at least thinks they [Ross’s beliefs] are evidence-based.

    I don’t know about that. The whole ID thing seems to come out of a long tradition of sophisticated Christian apologetics that originated in the Catholic Church. Stephen Meyer, a Cambridge Ph.D. at the Discovery Institute, also teaches a course in Christian apologetics at a Christian university in Florida. The general strategy, it seems to me, is to exploit the psychological impact of evidence or putative evidence, while undermining belief in its ultimate relevance (“either you believe or you don’t”). If one starts to view any alleged evidence as merely a tool someone is using to manipulate belief, then it becomes progressively more difficult to take evidence, and appeals to evidence, seriously.

  14. Greg Kuperberg Says:

    My beef with quantum computing skeptics has never been that their beliefs are false; rather, it’s that their beliefs almost never seem to lead them to new results.

    Exactly! This cuts both ways. It is exactly what I like to tell mathematicians who (a) are skeptical of quantum computation, and (b) defend skepticism partly by the argument that belief doesn’t even matter. So far, the believers have made vastly more progress than the skeptics. Now, part of the mathematical tradition is that believing and disbelieving a conjecture come to the same thing as long as you work on the conjecture. I accept this too, but it is not the whole truth, because some conjectures are a lot more interesting if they are true. (Or in some cases, false!)

    The same can be said of string theory, by the way. Although in both cases, “your mileage may vary”. There is no guarantee that belief in either string theory or quantum computing will remain useful.

    There is belief for the sake of loyalty and belief for the sake of utility. Belief for the sake of loyalty is not part of science. Belief for the sake of utility is a valid part of science, but at the end of the day, it’s just advice, it’s not something that a strict scientist can demand from others.

    It’s clear enough that the creationists want belief for the sake of loyalty. Scientists should not make of responding in kind. Certainly Ross deserves his PhD.

    The one thing that I will grant to the opposing view is that scientists are only human. How would you respond to an able student with a an odious agenda? Suppose that a student told you, “I plan to write a first-rate thesis in quantum computation because I am a lobbyist. I plan to use my credentials to denounce the field and cut off funding.” What Ross is trying to do is almost that bad. I might say, “As an ideally ethical research, I support you. But as a mere mortal, I hope that I don’t see you again.”

  15. Greg Kuperberg Says:

    If ID and creationism would win and schools have to teach it: What exactly would they teach about it?

    They would teach that we should be grateful to God for creating the universe, that Jesus died for our sins, and that evolution is not only unproven speculation, but also antithetical to Christianity. As the Dover decision noted, the real goal all along is not to teach any kind of science, not even bad science, it’s to supplant science with religion.

    I like to point out that it is not impossible to entertain intelligent design as science; on the contrary, it is commonplace. Archeologists very often do encounter objects that were intelligently designed. But they also have a model of the intelligent designers, who are invariably hominids. The difference between intelligent design, the “duh” lowercase form, and Intelligent Design is that you’re not supposed to analyze the latter. The Intelligent Designer is, by design, an unfathomable article of faith.

  16. Chris W. Says:

    Greg, that’s an interesting question. Given his credentials in the “paradigm” of conventional paleontology, would Ross start working to undercut research in the field? The NY Times mentioned a young earth Creationist who is also a geophysicist at LANL and an acknowledged expert on the earth’s mantle. Would young earth Creationists ever consider suppressing a paradigm relied upon by, say, the oil industry? I don’t think so…

    (Then again, I guess finding and burning as much oil and coal as possible, and accelerating climate change, could be taken as a fulfillment of Biblical prophecy; bring on the Tribulation!)

  17. Greg Kuperberg Says:

    In particular, I don’t merely believe evolution is good science; I also believe it’s true.

    Actually there is a philosophical question here that applies to string theory and quantum computation, among other topics. How great a distance do you ever see between exciting research and the truth? I believe that if a scientific theory looks interesting, it is at least a little more likely to be true; scientists would be unlucky blockheads if their interest is negatively correlated with the truth. But how strong has the relationship been, historically?

    Maybe the most discouraging case that comes to mind is controlled fusion. It would be so cool if tokamaks produced electricity. Just like skeptics of QC, skeptics of controlled fusion have never had a strong impossibility argument. And yet controlled fusion hasn’t worked. Is there a lesson in that?

  18. Greg Kuperberg Says:

    Given his credentials in the “paradigm” of conventional paleontology, would Ross start working to undercut research in the field?

    According to the New York Times, Ross is already flaunting his credentials on the creationist circuit. In my view of creationism, that is an attempt to undercut research in the field. Admittedly it could be an unwitting attempt, if Ross is the weird duck who can truly believe contradictory “paradigms”. (I suppose that it is at most 1/3 of the White Queen, who can believe six impossible things before breakfast.)

  19. Chris W. Says:

    Well, okay, there you are. I guess that’s the religious fundamentalist’s version of Lenin’s “the Capitalists will sell us the rope with which we will hang them”.

  20. Robin Blume-Kohout Says:

    Ah, Chris W., thank you. That was the allusion that tickled at the back of my mind for the past few days, but wouldn’t stay still long enough for me to figure it out.

    See, the lovely thing is that we did sell them the rope, and yet somehow they got own their heads in the noose. China appears to have headed in a similar direction, except in this case they decided they liked the rope so much they’d go into the business themselves (note: still a repressive, authoritarian government –but not so communist). The Dutch did much the same thing (though even more literally — they sold ’em war supplies) to the English for about a century or so in the 1650-1750 range, and to some degree the English did it to the French.

    I rather like the image of science as a sort of seductive ideology akin to trade, of the “Why, sure, feel free to try and use our weapons against us… but caveat emptor” sort.

    However, there is the cautionary tale of deconstructionism and postmodernism, which started off as an invention of the Left, and (IMHO) has been very successfully co-opted by the right these days (Fox News, anyone? Swift Boating?).

  21. Joshua Ball Says:

    Scientists should minimize neither their own influence on public opinion nor the shortage of informed minds in their respective fields. Doctors, lawyers, and engineers are called “professionals” because they belong to a self-regulating body that maintains a code of ethics and a standard of excellence. Such bodies exist because the public needs to be protected from fraudulent or otherwise incompetent professionals. A similar mechanism exists in the scientific community; it just goes by the name “peer evaluation”. However, the damage that bad scientists can do is not in the form of bad prescriptions, obstruction of justice, or falling bridges. While pure scientific advancements often take a generation to trickle down to technology, the most immediate effects of scientists are on public opinion. While Ross’s contribution to science is appreciated, his hypocrisy is unprofessional, and can be damaging to public opinion.

    That being said, I would hesitate to say that ID can never lead to good science. An extended analysis of the probabilities of certain organelles arising from natural selection is most certainly a worthwhile study, and would bring cohesion between biology and the dating methods of other branches of science, like geology. But the validation of ID does not necessarily lead to the existence of God. It can also suggest the intervention of extraterrestrial life, a la Arthur C. Clarke.

    Also, I wonder if the QED box will always be sufficient to convince others of the validity of one’s own work. We already live in a time when the foundation of set theory is not universally agreed upon (Axiom of Choice). Also, if it is ever proven that P does not equal NP, the proof will have to undergo a philosophical, not just mathematical, scrutiny, because as Razborov and Rudich showed, no natural proofs of P!=NP are likely to exist. The same thing happened to Cantor when he revealed his diagonalization proof.

    Scientists trust the work of other scientists because they have shared assumptions: mathematics is universally true and the scientific method is trustworthy. However, if I were to publish a paper which inferred a conclusion from epirical evidence and from the Axiom of Choice, it would not be enough to present the procedure and the evidence. I would also have to convince you that the Axiom of Choice is true in the natural world (whatever that means).

    When Newton published his discoveries on gravity, many scientists (Newton included) were dissatisfied because he failed to explain the mechanism by which gravity worked. Of course we can conceitedly brag that we are more enlightened now as to what the true nature of science is, but that is just strong evidence of the shared beliefs among all physicists today – namely that mathematical descriptions capture the essence of forces.

    So science IS about backing up your claims using evidence and possibly other means. We are just fortunate enough to live in a time when “evidence” is well defined and “satisfactory explanation” is seldom debated.

  22. Leonid Kontorovich Says:

    Thanks for introducing the usual clarity into a hopelessly muddled debate, Scott. I agree entirely!

    I’d only add that when working on a proving a hard math conjecture, it is immensely helpful to actually believe it’s true.

    Oh, and you may be amused to hear that I am a hard-core Platonist who believes that mathematical statements have no logical connection to the physical world, yet my work has already found some pretty concrete applications in neurobiology. And I do this without an internal contradiction!

  23. John Sidles Says:

    Modern cognitive science and modern information theory are IMHO converging on a similar picture of belief, which in its strongest forms are called “faith”.

    One of the paradoxes that emerges is the principle that the stronger one’s faith, the less one is aware of it!

    Also paradoxically, the more logical a person is, the less aware they are of their own faith structure!

    For example, isn’t it true that a fundamental faith of the science and mathematics community is simply this: P=NP?

    Like I said earlier this week in a response to Scott’s thread “A quiver springs his voice and breast”: “Aren’t the Bill Perry poem, the George Dantzig story, and the Fields Medal interviewees all speaking about the same topic, namely, the touching faith of the young that P does equal NP? That faith is at the heart of the Dantzig story. The professor assures the youth that a certificate exists, and buoyed by this faith, the student finds the certificate in polynomial time. To embrace this story, therefore, is to embrace the community-binding ideal that students should have faith that P=NP, and moreover, that students are themselves the embodiment of that P=NP algorithm! I would never condemn that faith — perhaps I still embrace it myself!”

    In medical school we teach our students (very seriously) “Never take away a patient’s faith, until you can offer them a better faith!”

    Recognizing that the technical community’s near-universal faith that P=NP does have some notably severe adverse consequences—also that it is both mathematically incorrect and silly too!—but not having presently having a better faith to offer, I will end this post! 🙂

  24. Torbjörn Larsson Says:

    Scott, thank you for articulating a view that where rarely espoused in other commentary. And I certainly think the use of conventional science in the PhD work is a rope that can be tied around the creationist neck as well.

    But your post leads to another concern I hadn’t thought of, voiced in a few comments here. What does this say about our justifications for using facts? If a creationist may put aside his own PhD work he devalues science in the view of the public. Then it isn’t only a case of that he will “use the legitimacy conferred by his degree to fight against everything the degree stands for”. He will also succeed in some measure. Bummer!

    Btw, when asked for my “belief” in some question, I invariably answer that “my view is…”. It gets rid of the whole fact-trust-faith issue for me. It is practically impossible to not have a current view, the trick is to be able to change. Anyway, that is my view.

  25. island Says:

    you’ll see that there’s actually some theory behind the idea that knowing someone’s opinion can be almost as useful as knowing the evidence on which their opinion is based.

    I find this to be nearly 100% true when it comes to the anthropic physics, and it is predictably ideology dependent, depending entirely where you happen to be talking about it. But what I find to be most alarming is the either/or attitude, because this is not right, and I easily prove it all the time. As soon as you back them into a corner, they go to silent denial mode in lieu of a better argument, rather than to willingly recognize that you have a point.

    “It is an undeniable fact that the universe appears designed”
    so…
    “as things stand now, we will be hard pressed to answer the IDists if the landscape fails”
    -Lenny Susskind

    Lenny was clearly trying to pressure/blackmail the string community into accepting the multiverse and anthropic selection. That’s my *supported* opinion, because it is a hard fact that Lenny isn’t about to find god if the landscape fails.

    So Lenny wasn’t doing science, as much as I’d like to claim that he’s supporting a strong interpretation if the landscape fails. The difference being that I’m being honest about what I believe. Richard Dawkins has recently jumped straight on Lenny’s multiverse bandwagon, so is he being honest?… or is he ***reacting*** through pre-prejudiced conditioning?

    Creationism plagues the anthropic physics on one side of the coin, but, equally fanatical and highly reactionary anticentrism rules mind of neodarwinians… and most others who interpret the evidence through their typically liberally slanted worldview.

    Brandon Carter said that his peers were infected with “anticentrist dogma”.

    … and he was right, so the creationsists side of the debate is necessary to the balance of reality… unfortunately.

    “But when you look at CMB map, you also see that the structure that is observed, is in fact, in a weird way, correlated with the plane of the earth around the sun. Is this Copernicus coming back to haunt us? That’s crazy. We’re looking out at the whole universe. There’s no way there should be a correlation of structure with our motion of the earth around the sun — the plane of the earth around the sun — the ecliptic. That would say we are truly the center of the universe.”
    Lawrence Krauss

    Crazy, Lawrence?… Or is it supporting evidence for a strong interpretation that will never be given equal time?

    Please don’t tell me that anybody is being honest with their interpretations, because I don’t buy it for a minute.

  26. roland Says:

    i am for sure not a creationist. but is darwinism a sufficient explanation ? Mathematician Marcel-Paul Schützenberger didn’t think so. read the interview.

    http://www.arn.org/docs/odesign/od172/schutz172.htm

  27. Scott Says:

    Leonid:

    Oh, and you may be amused to hear that I am a hard-core Platonist who believes that mathematical statements have no logical connection to the physical world, yet my work has already found some pretty concrete applications in neurobiology. And I do this without an internal contradiction!

    And why shouldn’t you? Not even G.H. Hardy (or Plato himself) thought that math had no applications to this imperfect world of shadows — it just isn’t logically necessary that it should.

  28. Scott Says:

    Joshua:

    Also, if it is ever proven that P does not equal NP, the proof will have to undergo a philosophical, not just mathematical, scrutiny, because as Razborov and Rudich showed, no natural proofs of P!=NP are likely to exist. The same thing happened to Cantor when he revealed his diagonalization proof.

    No, checking that a lower bound proof doesn’t naturalize is just ordinary mathematical scrutiny. In Cantor’s time set theory hadn’t yet been axiomatized, but now it has, and (within ZFC for example) we know exactly what we mean by a proof of P!=NP.

  29. Scott Says:

    I guess Marcus Ross also deserves an honorary degree in compartmentalization. 🙂

    I’ll say! There must be millions of politicians, salespeople, and guys in singles bars who wish they could compartmentalize that well.

  30. Southern Fried Skeptic Says:

    Ha! Rabbi Sagan. Damn Kabbalistic Konspirators.

  31. Kurt Says:

    Roland:
    i am for sure not a creationist. but is darwinism a sufficient explanation ? Mathematician Marcel-Paul Schützenberger didn’t think so.

    Hmmm, I wonder if Bill Dembski cites Schützenberger in his work? Information-theoretic attacks on evolution don’t carry much weight with biologists. The Pharyngula site that Scott links to near the top of this entry provides many examples.

  32. Scott Says:

    Chris W.: Thanks so much for the Rebecca Goldstein interview! I’m sure there are ‘greater’ novelists than her, but with the possible exception of Mark Twain, I can’t think of any who’s had a bigger personal effect on me. (I just finished Mazel, having previously read The Mind Body Problem [her best by far], The Late Summer Passion, and Strange Attractors.)

    My favorite snippets from the interview:

    Luke: “Do you find more to love [in the Jewish tradition] than to hate?”
    Rebecca: “Yes, especially when I’m not living in a Jewish community.”

    In [third] grade, Yael said to me about some story or explanation her teacher had given, ‘This doesn’t make any sense. What do you think?’
    “I looked at her and said, ‘Do you really want to know what I think about all this?’ There was this long pause. We looked into each other’s eyes and she said, ‘Not yet.’

  33. Scott Says:

    Mihai, Greg, Robin: you know what? Y’all are onto something. Your comments forced me to recognize that there’s an important empirical component to my position. Namely, I don’t think a few lunatics like Ross will in fact succeed in destroying science — I think they’ll merely make fools of themselves. And I also think that, for all but an inevitable minority, science really is corrosive of dogmatic beliefs.

    There’s a strong analogy here to free speech laws. If, for example, the neo-Nazis were to take power in America (no, c’mon, guys — I mean the actual ones, not Bush), then it would have turned out that protecting their First Amendment rights was an unmitigated disaster. But I don’t think that’s going to happen.

  34. Scott Says:

    So far, the believers have made vastly more progress than the skeptics … The same can be said of string theory, by the way.

    Greg: Would you agree that there’s a difference between the two cases, i.e. that string theory skeptics are trying to make progress in a way that quantum computing skeptics are not? (I know we disagree about the magnitude of the difference… 🙂 )

  35. Boxing Jewels Says:

    […] Scott Aaronson […]

  36. Douglas Knight Says:

    to see if they truly believe (A or not(A))

    What are you, a physicist or something? I thought CS dogma was constructive.

  37. Scott Says:

    What are you, a physicist or something?

    If you continue posting vicious personal attacks, I will have no choice but to start deleting them.

  38. Greg Kuperberg Says:

    Would you agree that there’s a difference between the two cases, i.e. that string theory skeptics are trying to make progress in a way that quantum computing skeptics are not?

    Sure, that’s true of some string theory skeptics, and that’s fine. Some of them do good mathematics and I am perfectly happy to interact with them, or even hire them if the conditions are right. I don’t know if they do good physics, but I do not rule out the possibility entirely. They may also not be entirely set in their skepticism.

    The catch is that a few of the active skeptics are also bitter, and write lame attempts at debate in the guise of surveys, or issue misleading broadsides in the venue of public opinion. It’s not my professional concern since I am not a string theorist, but I gather that if these few try hard enough to irritate string theorists, they may moderately succeed.

    It’s fair to say that there is much less of both in the case of quantum computing, but not quite none. For example, Ed Fredkin is a QC skeptic and you could say that he actively pursues alternatives. Meanwhile Robert Laughlin has a short broadside against QC in one chapter of his popular physics book. Laughlin isn’t bitter, merely arrogant, but it sort-of counts.

  39. Greg Kuperberg Says:

    Information-theoretic attacks on evolution don’t carry much weight with biologists.

    That’s because there aren’t any. There are attacks by people who have studied information theory, but there are no attacks which are materially based on any serious information theory. Dembski in particular is just bluffing.

    My friend Dror Bar-Natan likes to distinguish between the theory of evolution and the fact of evolution. The theory of evolution has a lot of things wrong with it, and in any case has moved far past Darwin’s simple model of natural selection. Indeed, it is an exciting research area precisely because it is unsettled. In fact, I’m sure that it can be improved with information theory. On the other hand, the fact of evolution is irrefutable if you want to do good science, just like the fact of quantum mechanics.

  40. Larry Moran Says:

    Ross was working on a Ph.D. in the field of paleontology. Ross rejects the very core principles of geology, namey that the Earth is billions of years old.

    Imagine that he had tried to defend that in his thesis by attempting to refute all of the evidence for an old Earth and presenting bogus scientific evidence for a 10,000 year old Earth.

    Should be get a Ph.D. under those circumstances? I hope everyone will agree that the only possible answer is no. What this means is that Ross would be flunked because he’s ignorant of basic concepts in geology. You confuse the issue by referring to his stupidity as a “belief” because as soon as you start using that word everyone gets up on tippy-toes in order to aviod looking like they’re anti-religious bigots.

    We flunk people all the time when they demonstrate that they are incapable of reaching rational conclusions based on the best available evidence. This has nothing to do with “beliefs” or “loyalty oaths” (another loaded word). It has to do with how science is done and with intellectual integrity.

  41. Ian Durham Says:

    All very good points. Much of this boils down to the way we argue about things in general as a society. Things too often get politicized. If someone asks me if I “believe” in the Big Bang or evolution or something like that I always correct them on the use of that term (while attempting to not sound belittling) and explain that it is about not just evidence but simple logic – if A implies B and B implies C, then A implies C etc. etc. or whatever.

    Unfortunately, as scientists, we are not immune (though physical scientists, mathematicians, and computer scientists are usually better than others at it). I made a point about this a few months back on my blog (which no one reads but I still post to, perhaps simply to amuse myself), noting that the global warming debate is often misrepresented by the scientists themselves. In essence (and this harkens back to something I posted to the Quantum Pontiff and included in an editorial in the latest issue of The Quantum Times), highly probable statistical correlations are presented to the public as absolute truths and the public’s response is usually “But it’s only a statistical correlation! Even I, Joe Idiot, know that means there’s a chance it’s wrong!” Granted the chance may be small, but by presenting it as fact the scientist immediately gives a doubter an opportunity for argument.

    On the other hand, I will say it irritates me to no end (and I see this in many of my students who are not physical or mathematical sciences majors) that people can’t make simple logical connections. Evolution isn’t just some grand idea for arguing about old bones, it forms the basis for genetics, right from the geinning in fact (cf. the fruit fly experiments at Columbia in the early 20th century) and is used in drug development. So, on the one hand, people will deny evolution occurs and yet will gleefully take that vaccine that was developed using evolutionary principles. And therein lies the danger: if evolution and the Big Bang and other such scientific claims are taught to be doubted in schools, what’s going to happen to the future of medicine and technology? It’s an extreme, but could we end up as backward as those religious sects who disavow education and technology altogether? The only hope is that the innate greed of the west knocks sense into some people (“I still need my cell phone and my vaccine,” says Joe Idiot.)

    Well, that’s my incoherent rant for the day…..

  42. roland Says:

    i started reading “The Mind Body Problem” but stopped.

    A woman marries a mathamatician but after a while wants to get divorced for some reason. Just then her husband reveals his
    new intellectual impotence. Suddenly she (unwillingly) sees him as a person.
    ohmygod, she got married without seeing him that way – that is really shallow,

    Because of that and general boredom i stopped reading.

  43. island Says:

    Greg, I think that those are excellent points, and your assessment of Bill Dembski rings true to my experiences with him. He is thoroughly dishonest.

    I still think that there is an ideological wall standing in the way of improvents to evolutionary theory that can be defined by whatever it is that causes Lynn Margulis to call them, “neodarwinian bullies”, which is not a normal disagreement among peers, it’s a shot at antifanaticism.

    Both sides need their diploma’s screened.

  44. Greg Kuperberg Says:

    Ross rejects the very core principles of geology, namey that the Earth is billions of years old.

    Or rather, the right half of his brain rejects it. The left half of his brain accepts it as the truth. Moreover that half of Ross helped the field with valid research. So the right thing to do is to give him the PhD, because he deserves it, and then hope that he doesn’t cause too much trouble. PhDs are for achievement, not rectitude.

    It irritates me to no end that people can’t make simple logical connections.

    What I like to say is that experiment without theory is just as bad as theory without experiment. Just about anyone knows to ridicule the latter, but there is not enough cultural resistance to the former. For instance, consider the recurrent fear that radio waves from cell phones or power lines could cause cancer. Of course, it is our practiced reaction to say, look, there have been plenty of studies that show no relationship. But my real gut reaction, before I learn much about negative experiments, is that these fears are theoretically impoverished. There is no reason to conjecture that cell phones cause cancer, only illusions of reasons, some of which spring from nebulous mistrust of technology. This is not to say that no reason to worry will ever arise, only that existing epidemiological surveys just grasp at straws.

  45. Scott Says:

    You confuse the issue by referring to his stupidity as a “belief” because as soon as you start using that word everyone gets up on tippy-toes in order to aviod looking like they’re anti-religious bigots.

    Larry: I know there are people who think “beliefs” have to be respected regardless of their content. Those people are idiots. (Now there’s a belief for you. 🙂 ) If you’re looking for wussified postmodernism, you’ve come to the wrong place.

    Ross’s beliefs can be judged, should be judged, and are utterly without merit. The question is whether one should take his (doofus) beliefs into account in evaluating a thesis where he never once mentions them.

  46. Ian Durham Says:

    And my answer to the latter point, i.e. whether or not he can and should have his PhD revoked for this, I say no since it potentially opens a can of worms from which springs the ability of an academic institution to revoke credentials for comments or actions utterly unrelated to those comments or actions. So, should some university suddenly have a board of governors (or directors or whatever) and administration that is pushing a particular agenda, they could theoretically pull credentials from anyone they didn’t like. It sets a dangerous precedent. If the work wasn’t plagiarized and met the existing standards then it should stand. Why the bonehead is now spouting dogma antithetical to his original ideas is a mystery, but it shouldn’t disqualify those original ideas. And even if he was just “playing along” with us rationalists the whole time, there’s no way to prove it.

  47. Gil Kalai Says:

    After the successful “Reason to Believe” and the equally successful sequel “Reason to Believe II”, I find the current post by Scott that can be called “Reasons! not Beliefs” surprising.

    I did have some problems with the notion of “belief” in Scott’s earlier posts. It was a mixture of three things:

    1) I believe X because of the evidence that X is true.

    2) I believe X because I think the universe is nicer if X is true.

    3) I believe X because I think the universe is nicer if people believe X.

    Take the question if NP is P.

    Perhaps, a primary evidence that NP is not P comes from our long experience in impossibility (pessimistic) results. We cannot write the square root of 2 as a fraction; and we cannot trisect an angle with a ruler and a compass, and we cannot prove every true statement, even in principle. The long list of pessimistic results and the insight that often what we cannot do relatively easy we cannot do at all, and often the insights *why* we cannot do it comes much later in the game, is a major reason to believe that P is not equal NP.

    This pessimistic instincts were the motivation for conjectures preceding the NP/P formulation – that certain computational tasks requires complete (exponential) enumeration. (In fact these conjectures are stronger than NP=!P.) Complexity theory is largely developed based on the assumption that NP is different than P. The theory does give some evidence that NP=!P and also some evidence that proving this conjecture will be very difficult but overall it is safer to regard NP=!P more as an assumption than a prediction of complexity theory. (The theory does have strong predictions. It gives rather strong evidence that RP=P.)

    Scott did make some weak case for why the universe is nicer with NP being different than P. Those are not very convincing. One argument is that creativity is manifested by the gap between P and NP. Creativity is (philosophically) relevant to complexity theoretic distinctions, whether between decidable and undecidable statements, or between statements in P vs NP. But *not* in the sense that our mind can make NP-complete computations (or even decide the undecidable.) The gap between deterministic and randomized algorithms may well be more relevant. And, in any case, a world where our computers can solve any optimization problem would have enough advantages to compensate for some decline in the status of creative people.

    Greg added some new ingredient related to 2) and 3). A reason to believe in X is that believing X leads to more interesting results. This is interesting, but I don’t believe it.

    Scott: “Science is fundamentally not about beliefs: its about results.”

    While I agree with Scott’s overall sentiment, here Scott is pushing it. “results” is too narrow term to describe science.

    Greg: “If a scientific theory looks interesting it is at least a little more likely to be true.”

    Hmmm, interesting belief. Maybe it is not monotone: It should look interesting but not TOO interesting.

  48. mtraven Says:

    The flap about whether or not he gets a Phud is not that interesting to me. I want to know is how this guy does it, and why. Maintaining two incompatible belief systems is something I guess we all do to some extent, but usually not to the level of getting a doctorate and landing the Times. Dawkins has labeled this sort of thing “Virtuoso Believing”, and at some level you have to applaud.

    There’s a long and boring go round between me and some of his defenders here. For some reason, hypocrisy is considered a virtue among the religious. I gave up after awhile, if anyone else wants to go in for a round, feel free…

  49. Tyler DiPietro Says:

    Scott, I think you’re absolutely right. In fact, I don’t even know why this is even a news story. A guy completed his graduate work and earned his PhD., thus accomplishing what thousands of others do every year. I guess it just has that “dog bites man” effect so coveted by today’s tabloid journalism establishment.

    This guy will just join the handful of credentialed individuals who abuse their positions for pseudoscience. There is an astrophysics PhD. named Gerardus Bouw who advocates not only YEC but geocentrism, for FSM’s sake. Like you say, science is about results, religion is about belief. That the rubes in the pews will swallow whatever this guy says and be amazed is a given, almost as much of a given as the possibility that he’ll be useless as an actual scientist.

  50. Larry Moran Says:

    Scott says,

    Ross’s beliefs can be judged, should be judged, and are utterly without merit. The question is whether one should take his (doofus) beliefs into account in evaluating a thesis where he never once mentions them.

    Let’s try and focus. First off, I’m not advocating that his degree be revoked. I don’t think anyone is trying to take away his Ph.D.

    What I’m concerned about is whether he should have got a degree in the first place. I’m glad we agree on one thing. Ross would never have gotten a Ph.D. if he had been truthful in his thesis. That’s probably why he wasn’t.

    Now the question is the following. Knowing that Ross was lying in his thesis&mdash:because he didn’t make a secret of it—would it have been fair to question him about his true understanding of the age of the Earth during the Ph.D. oral? What if he admitted that what he wrote in his thesis does not reflect his true understanding of the age of the Earth? Should he pass?

    And what if he lied during the Ph.D. oral and said that he accepts the scientific evidence for the age of the Earth? Does intellectual integrity count for anything in your mind or should we just give him the degree knowing that he’s a good liar?

  51. Greg Kuperberg Says:

    Ross would never have gotten a Ph.D. if he had been truthful in his thesis.

    But he was truthful in his thesis. His thesis has a series of true statements about paleontology. They aren’t statements about his own beliefs, they are statements about the history of the planet. The real issue is that he’s untruthful at other times, when he promotes creationism.

    Maybe this discussion has moved from Ross himself to what a PhD means or should mean. I think that it should mean that you have done valid research in the field. You’re describing it as a description of what people understand. Even aside from political controversies such as creationism, your suggested interpretation bothers me. I would prefer that people see my own PhD as a demonstration of achievement, or at the most that I master(ed) certain skills, rather than that I know, believe, or understand any particular thing.

    I might have attributed this difference in interpretation to a difference between physical sciences (or mathematics) and biology, but I’m not sure that that’s fair. Certainly in mathematics, I’m perfectly happy to credit people who, for whatever reason, reject their own valid contributions. If you have a good idea, it doesn’t really matter if you also disavow it. I would suppose that biology should be the same.

  52. Greg Kuperberg Says:

    I want to know is how this guy does it, and why.

    I can’t really put myself in the shoes of deeply religious people, but here is what I think anyway. I think that the guy believes creationism not for any empirical reason but out of loyalty. But he is also reasonably smart, and even interested in evolutionary biology despite his loyalties. He therefore also understands empirical scientific truth, just as a different mode of “truth” than literalist Christianity.

    Now I think that believing things out of loyalty amounts to either stupidity, or dishonesty, or denial (which is just self-dishonesty). I also think that it’s too fashionable in the United States, especially in national politics. For example, the polls tell me that most Republicans still believe that the US is “winning” the war in Iraq. I interpret most of this sentiment as sheer nationalist loyalty. It bugs me. But the problem is not just conservative ideology. I have heard people say, “in this country, you ARE innocent until you ARE proven guilty”. I.e., it’s not just a legal presumption, it’s an injustice if we don’t personally believe that all acquitted defendants are innocent. This bugs me too, because it’s stupid.

    So okay, belief for the sake of loyalty is a brain malfunction. But it is also only human. Undoubtedly I also believe some things just because I’m supposed to, and maybe social pressure also prevents me from identifying these false views.

  53. Science vs. Truth « Blunt Object Says:

    […] Mistake of the Week: Belief is King […]

  54. Scott Says:

    Larry: As Greg pointed out, the real question here is not about Ross; it’s about what a PhD is supposed to signify. I tend toward Greg’s view, whereas you tend toward a different view.

    Actually, the more I think about it, the more I realize this Marcus Ross story is optimally designed (intelligently designed?) for being blogged about.

    I mean, look. We’re talking about a story that’s got

    (1) evolution vs. creationism (always a winner),
    (2) a slightly new twist, but not so new that everyone won’t immediately have an opinion about it,
    (3) a compelling human-interest angle, and
    (4) an academic ethics question, which even those who accept the fact of evolution can argue about till the sun goes cold (or for Dr. Ross with his mortarboard off, the Rapture).

    So, on behalf of all of us bloggers: thank you, Dr. Ross!

  55. Gil Kalai Says:

    Another case study for “beliefs, results and ideas” is the nice idea that Scott and Greg promote about applications on non-commutative (quantum) probability in various areas of mathematics and computer science. (Which are *not* a priori related to quantum physics.)

    The amazing success of the “probabilistic method” in so many areas of mathematics (even in proving theorems which seem completely unrelated to probability) make this idea of Scott and Greg very appealing. However, results in this direction are still very sparse. Some reason for skepticism is the insight from statistical mechanics that models based on ordinary probability and on non-commutative probability behave overall in a very similar way. I suppose we will have to wait and see.

  56. Greg Kuperberg Says:

    The amazing success of the “probabilistic method” in so many areas of mathematics (even in proving theorems which seem completely unrelated to probability) make this idea of Scott and Greg very appealing.

    Thanks for the positive statements here! But I would say “stance” rather than “idea”. If anyone can be credited with the idea, it would be von Neumann.

    Some reason for skepticism is the insight from statistical mechanics that models based on ordinary probability and on non-commutative probability behave overall in a very similar way.

    There is no such insight from statistical mechanics. The experience in both statistics mechanics and field theory is that classical and non-commutative probability are similar in some ways. But in other ways, they aren’t. For example, there is no explanation of superconductivity and lasers in classical probability. Although there is no strict impossibility theorem in the classical setting, the mere existence of lasers is significant evidence in favor of quantum probability.

    Or to pick a more initial example, one of the paradoxes of physics a century ago was how electromagnetism, which is an attracting force, can possibly lead to the existence of rigid structures in chemistry. For comparison, there are no rigid structures in astrophysics. The answer is Pauli exclusion, which is purely a statement of quantum probability and not a new force law.

  57. Greg Kuperberg Says:

    Gil: Concerning the limited or not-so-limited appearance of the probabilistic method and its quantum counterpart:

    In a way, the probabilistic method does not really exist in mathematics. It is technically valid to say, “oh, that is really a counting argument” when you see the probabilistic method in use. In any of these results, probability theory is just one way to explain the real proof. Some people do not like this scaffolding. I disagree with them, but the equivalence to non-probabilistic explanations is valid.

    The presence of non-commutative probability in mathematics is similar, but even more hidden because most mathematicians do not study non-commutative probability. There are a lot of non-quantum results in mathematics whose proofs can be phrased quantumly, but traditionally aren’t. People generally won’t be convinced unless you use quantum probability to prove something new, and even then someone else might rephrase the proof in non-quantum terms. That is a high hurdle. If everyone were equally conservative about classical probability, the probabilistic method would not be a standard concept either.

  58. Drew Arrowood Says:

    There are philosophers (see http://www.nyu.edu/gsas/dept/philo/courses/consciousness97/papers/stich.html) who have tried to argue that the whole concept of belief is a bad one, even in explaining human psychology.

    A naive person would therefore assert that they believe that there aren’t such things as belief.

  59. Ze Says:

    I would prefer that people see my own PhD as a demonstration of achievement, or at the most that I master(ed) certain skills, rather than that I know, believe, or understand any particular thing.
    Greg I agree with you especially since the other view has a dangerous logical consequence.

    If a PhD was about what you understood,believed and knew what happens if that research your PhD was based was found to be incorrect? Would we have to revoke everybody’s PhD’s that were based on science later found to be incorrect? Since science should be impartial and apply the same standards to everybody.

  60. Gil Says:

    “GK:…make this idea of Scott and Greg very appealing.”

    “GK: Thanks for the positive statements here! But I would say “stance” rather than “idea”.”

    Once again you send me to the dictionary, Greg, let’s see
    [Merriam-Webster online]

    STANCE

    “Etymology: Middle English stance, staunce, from Middle French estance position, posture, stay, from Old French, from Vulgar Latin *stantia…”

    (yes, it sounds like vulgar latin)

    “..the position of the feet of a golfer or batter preparatory to making a swing”

    What can I add? Swing, GREG, swing !!!

    (Anyway your nice idea/stance of “non-commutative probabilistic method”, Greg and Scott, deserves a separate discussion sometime.)