Comments on: Mumbo jumbo

By: sennoma

sennoma — Mon, 16 Feb 2004 01:35:10 +0000

Keith: I think I’m just saying that those things are a lot more of the reality of day-to-day science than is commonly supposed. Well, I can live with that. Interestingly, my wife, who is not a scientist, was suprised to hear your point of view and even more suprised by the extent to which I agreed with it. In particular, she imagined replication of individual observations to be rather stricter and more routine than my experience indicates it is. It seems that the lay view of science may be rather different from what I imagined it to be (although I should add that my wife has a deep and abiding interest and autodidactic background in science and is probably not a good example of an “average” view). Allen: Isn’t it at the moment when it would get dropped (when the paradigm is in danger from contradictory evidence, and it can’t be, or isn’t, ignored) that social factors—the politics—are most important? I would think so, yes; but my contention is that all the politics can ever do is delay acceptance of the new model, not prevent it: the evidence always wins. Also, I was explicit about the length of time I am talking about: months to years, not decades or lifetimes. Years of delay because of politics might seem—well, it is—miserably inefficient; but even so, the enterprise of science can claim remarkable success, and no other system that I know of has ever come close to matching it. (I should confess here that I have Kuhn’s ideas only at second hand; a copy of The Structure of Scientific Revolutions is on its way to me as I write, since this conversation has revived my interest.)

By: allen claxton

allen claxton — Sat, 14 Feb 2004 00:17:57 +0000

Just thinking out loud here. When Kuhn claims that sociological forces play the determining role, I think he is taking too short term a view. In the longer term, he is flat wrong: politics must give way to empiricism I don’t know if there is an “in the long run” for Kuhn. I assume that you mean that in the long run, faulty “normal science” gets dropped. But dropped in favor of what? Isn’t it at the moment when it would get dropped (when the paradigm is in danger from contradictory evidence, and it can’t be, or isn’t, ignored) that social factors—the politics—are most important?

By: Keith M Ellis

Keith M Ellis — Fri, 13 Feb 2004 00:48:26 +0000

“That’s very true, but it’s not a common insight in my experience. You have spent a lot of time around scientists. :-)“—Sennoma

Yep. I have. :) Seriously, no one could work even remotely as hard as a scientist does for so little pay without being motivated by a deep idealism. Well, okay. One woman I was involved with, an astrophysicist, was one of the most deeply pragmatic people I’ve ever known. She once said that she does what she does because she “gets to play with the big toys”. Which, since I’m pretty idealistic about science, sort of took me aback. (Had I continued in physics I would have gone into cosmology, so I think had a little case of hero worship going on and I was disapointed in her.) All the scientists I’ve known are amazingly dedicated and hard working people.

By: Keith M Ellis

Keith M Ellis — Fri, 13 Feb 2004 00:39:35 +0000

Whoops. Extra “http:” in that link.

By: Keith M Ellis

Keith M Ellis — Fri, 13 Feb 2004 00:38:15 +0000

Sorry. I talk about where I went to school a lot since it’s so unusual. I’m almost certainly obnoxious on the subject, actually. That’s why I mentioned it here only in an off-hand way. Kinda backfired, huh?

“But when you say ‘Authority and credentialism win the day’, and you’re talking about the ‘community of trust’ (your phrase!) method of doing day-to-day science, you’re using terms that I just don’t think are appropriate and that don’t match my experience.—Sennoma

Well, perhaps it’s a connotation of what I’m saying that you have trouble with. I think I’m just saying that those things are a lot more of the reality of day-to-day science than is commonly supposed. You really should spend some time on a USENET newsgroup like sci.physics.relativity. You’ll be adopting an credentialist/authoritarian position pretty quick, I bet. I say that jokingly but also half-seriously. This is my point about how the over-romanticized view of science plays into the hands of crackpots.

By: sennoma

sennoma — Thu, 12 Feb 2004 22:16:37 +0000

I was a physics major before I went to SJC What’s SJC? I hate to give in to credentialism but now I’m curious as to what you do and how you got qualified to do it. (I work here via a PhD in molecular parasitology and four years in HIV replication.) there’s a hell of a lot that scientists don’t know that one would expect them to know; and they are especially naive about the philosophy of science I confess that I’ve long thought it odd to earn a PhD without reading any philosophy. I mean none, nada, not a jot or tittle; there was no philosophy requirement, not even in ethics, in my graduate degree. (I understand that the US system is a bit more rigorous than the Aus one in this regard.) I don’t think many people can become scientists without being in some sense deeply idealistic about the enterprise of science That’s very true, but it’s not a common insight in my experience. You have spent a lot of time around scientists. :-) I think it’s sort of odd that people like yourself who are normally quite skeptical could imagine that practicing science equates to a deep understanding of the enterprise of science. This is rarely the case with anything else—what would be so special about science? I’m not really following this line of argument; perhaps it’s “experience” vs “deep understanding”. I’m really only arguing for the former, and not trying to dismiss the value of a shifted or outside perspective. But when you say “Authority and credentialism win the day”, and you’re talking about the “community of trust” (your phrase!) method of doing day-to-day science, you’re using terms that I just don’t think are appropriate and that don’t match my experience.

By: Keith M Ellis

Keith M Ellis — Thu, 12 Feb 2004 21:46:22 +0000

Very, very nice and thoughtful comments, Sennoma, and I appreciate them. I think overall you and I are pretty much in agreement. I got stuck arguing the position I did because, yes, Seth made an overstatement but then so did Tim and Abiola in response.

“That second sentence smells suspicously like an appeal to authority to me, and the first is just plain wrong.—Sennoma

Well, I respectfully disagree. My education is deep in the history and philosophy of science, and not in the sense of reading someone’s survey. We read the original works and did a lot of the experiments up to mid-twentieth century science. There’s six semesters of laboratory science and eight of math. And I was a physics major before I went to SJC. Anyway, I’m not arguing primarily from a “science studies”/Kuhnian point of view. I am familiar with a fair bit of his and related work; but my opinion on this arises from a combination of my understanding of western science in its progression, and in my experience training in science and being (then and later) among and observing scientists. It’s my opinion that there’s a hell of a lot that scientists don’t know that one would expect them to know; and they are especially naive about the philosophy of science. It’s especially weird to me given that many of them are quite cynical about the daily practice of science; but I don’t think many people can become scientists without being in some sense deeply idealistic about the enterprise of science. It’s my observation that this creates a sort of intellectual blind spot. Anyway, I think it’s sort of odd that people like yourself who are normally quite skeptical could imagine that practicing science equates to a deep understanding of the enterprise of science. This is rarely the case with anything else—what would be so special about science? Yes, I understand you’re on the defensive from the attacks of the pomo relativists and that most of them don’t know jack shit about science. But, you know, there are people that do. And they’re not all scientists.

By: sennoma

sennoma — Thu, 12 Feb 2004 21:02:29 +0000

Seth: Scientists imagine themselves as ‘science.’ Tim: Nonsense. And you know it. That seemed to set the whole pile ablaze. Seth's painting with too broad a brush, IMO, and in fact there are relatively few scientists who conflate themselves with the larger endeavour. Not to say that we (better get that out of the way: I'm a biologist by trade) don't have ego problems in the scientific community, but I think all but the most egotistical scientists feel that they are only a part of something larger than themselves. Keith (to Tim): there’s quite a bit of discussion and inquiry that make a strong argument that science is closer to what Seth says it is than to what you (I’m guessing) think it is I'd like to argue for a position much closer to Tim than Seth, but perhaps not all that far from Keith if I read him right. But I'll start by disagreeing pretty strongly with Keith on this: Science is not a body of knowledge, nor is it a methodology. Those things help define it; but it is a community far more than it is anything else. I think that one of the basic mistakes in sociology of science is to confuse the community with its purpose, and one of the basic mistakes in knee-jerk reaction to sociology of science is to try to insist that the two are entirely separate. From the dictionary (why yes, I do remember usenet; why d'you ask?):

1 : the state of knowing : knowledge as distinguished from ignorance or misunderstanding
2 a : a department of systematized knowledge as an object of study science of theology> b : something (as a sport or technique) that may be studied or learned like systematized knowledge science>
3 a :knowledge or a system of knowledge covering general truths or the operation of general laws especially as obtained and tested through scientific method b : such knowledge or such a system of knowledge concerned with the physical world and its phenomena : NATURAL SCIENCE
4 : a system or method reconciling practical ends with scientific laws science>

The word "science" covers quite a bit of ground in common use, so it's as well to say up front that I want to talk about definition #3, a body of systematic knowledge. That's not a community, or a methodology, it's a thing that can be (even if it usually isn't) examined separately from the way it was produced and the people who produced it. It's a cultural artefact, and to fully understand it you have to understand the culture that made it; but it stands on its own in relation to any observer from outside that culture. That's because it's not just any "knowledge", it's essentially a systematized set of observations and models built from those observations. The observations can be, and are, repeated; and the models can be, and are, tested; anything found to be false is discarded. I don't much care whether you choose a realist version of "false" or equate falsehood with "lacking predictive power about future observations", and I don't think academic definitions of what it means to test a model count for squat in the long term. Anything that doesn't work -- that is contradicted by empirical evidence -- is removed from the system. But to bend Keith's way a little, that's not to say that how and why and how fast things are discarded doesn't matter. Those questions are important, and sociology of science can have important things to say about them. On falsification, we find Abiola saying Popperian falsificationism is actually a rather decent description of how good science is done, and Keith replying:

No, it is not. [...] Go ten years back and pick an issue of Nature. Use the papers that appear in that issue as a small sample. Now see:
1) How many times those papers are subsequently cited; and, 2) How many of those experimental results have been replicated?

I decided I'd have a go at that, and if you care for the grisly details they can be had here. I only got through two papers, but I found that the essential results in both had been both directly replicated and confirmed in other ways. In one case, a paper with 124 citations, it took about four years -- although I didn't try to find the earliest replication, so it may have been done earlier. In the other case, I got verification in the same year (1993); that paper had been cited 444 times. I don't think Keith expected that when he wrote Replication of results is not routine, but to be fair, two is a tiny sample and among all those citations there are almost certainly a goodly number of studies which accepted the original finding and built on top of it without replication. I think this is what Keith means when he says Science relies upon authority; not, as a practical day-to-day matter, falsification. It's what Kuhn called "normal science": working within a paradigm, because challenging everything in strict Popperian fashion just takes too much damn time. I don't think it's "authority", I think it's better called "trust", and I don't mean trust in whoever made whatever observation one is relying on. I mean trust in the long term working of the empirical method, regardless of (even despite) its cultural context. Anything that comes into conflict with new empirical evidence will be re-evaluated and, if false, discarded. Anything, no matter who proposed it. Keith agrees with me even as he is disagreeing when he says: most experiments are not replicated because there is a community of trust that allows that if an experiment has all the trappings of a good solid experiment, then replication is assumed possible but unnecessary. Authority and credentialism win the day. This is how the majority of science is done. Yes, it's how most science is done; no, it's not authority and credentialism. Because of the way new work builds on old, in Kuhnian "normal science" you very rarely get tripped up by assuming replication to be possible but unnecessary: everyone knows that, sooner or later, a dodgy result will be found out and having published it will come back to bite them in the ass, hard. As between Popper and Kuhn, it seems that the difference between Tim/Abiola and Keith is in which set of factors, empirical or social, is thought to be the most important in determining the fate of any given theory. If you take Popper to be saying "this is how it should be" and Kuhn to be saying "this is how it is in the real world, Karl", some of the tension disappears. I don't think science progresses by strict Popperian falsification from theory to better theory, but I do think that, in the long term -- and I mean months to years, not decades or, pace Planck, lifetimes -- the requirements of Popperian falsification are satisfied by the somewhat more ad hoc way that real scientists go about things. When Kuhn claims that sociological forces play the determining role, I think he is taking too short term a view. In the longer term, he is flat wrong: politics must give way to empiricism, simply because we want stuff to work -- improved crops, better sources of energy, cures for disease, and so on. If Popper is unworkably restrictive, Kuhn (and Feyerabend!) go too far. A looser view of Popper's model seems to me able to account for both the vagaries of real scientists working in a real, social, political context, and for the technological success of the scientific endeavour. Part of the problem seems to be pomo relativists who want to wave Kuhn as a flag of triumph over truth -- sorry, "truth" -- and the understandable backlash against them from, well, anyone with half a clue. I note that Keith keeps pointing out that he's not a pomo relativist, but talking about "authority and credentialism" doesn't help him to make that case. In particular, I disagree with this:

That you practice science no more means you have an aquaintance and facility with the study of science than does living in the physical world make one a physicist. The points I am making are well known and uncontroversial to people that do study science

That second sentence smells suspicously like an appeal to authority to me, and the first is just plain wrong. As a working scientist, I might not have read Popper, Kuhn, Feyerabend et al., or their contemporary inheritors; but I do have some idea as to how real science gets done, and I'm not being proprietary by insisting that my experience is valid and useful in discussions about what science is and how it works.

By: Keith M Ellis

Keith M Ellis — Wed, 11 Feb 2004 08:15:27 +0000

Well, Robin, I suggest you read Davis and Hersh’s book. It’s very interesting. It’s been many years since I’ve read it, but I do believe that they argue that there would be practical consequences for a wholly consistent platonist or formalist mathematician. And this is why I brought it up in this context. A completely culturally relativist science would be something quite different from what it is, as Tim was pointing out. But also so would a pure empirical realist science. The reality is more complex; and, I think, more interesting.

By: robin green

robin green — Wed, 11 Feb 2004 02:22:53 +0000

I’m not sure that there is anything to the distinction between Platonism and formalism, beyond the worshipping of concepts. Platonism says that mathematical facts “really exist” in some “platonic sphere” which is independent of any real-world contingencies. Whereas, formalism says that mathematical facts follow as logical – or rule-following – consequences of the formal systems in which they are “embedded”, and that they have no existence independent of those systems. When stated that way, there is no necessary incompatibility. The Platonic sphere simply is formal mathematics!

By: Keith M Ellis

Keith M Ellis — Mon, 09 Feb 2004 19:45:12 +0000

From another thread:

"The Mathematical Experience, Phillip Davis and Reuben Hersh. A nice mix of history, philosophy and the odd bit of actual mathematical insight. Useful for dispelling the notion that maths is the dry stuff taught to engineers and economists."—Abiola

A book I, too, highly recommend (as well as Innumeracy). Perhaps you've forgotten a central theme of the book: that what mathematicians feel they do is not exactly (or not at all) what they believe they do which may, in turn, be quite different than what they claim they do. Particularly, Davis and Hersh argue that most mathematicians are secretely Platonists while publicly, and rigorously, presenting themselves as formalists. And this is a very important matter to Davis and Hersh because, to them, understanding the tension between these two dual-minded views is essential to understanding what mathematics is. That is the theme of their book. Science is this same situation turned on its head. Scientists present themselves as naive realists the way mathematicians present themselves as formalists. But as a practical matter, scientists are consensualist pragmatists, where mathematicians are idealists. In both cases, to really get to the heart of what this means for mathematics and science, you have to understand the whole of this apparent dichotomy and not just a part.

By: seth edenbaum

seth edenbaum — Mon, 09 Feb 2004 19:23:23 +0000

I think the problem many people have is that they can only imagine one kind of machine. Machines are logical so we must be logical as well. But consciousness is flawed awareness. There’s no reason to argue any metaphysical mumbo jumbo, we make our own. We make mistakes. We act out of misconceptions and fear. Ambiguity makes some of you nervous. And there’s no comparison between Godel’s theorum and the questions in a courtroom. Am I going to argue that the uncertainty principle means that I can’t rely on data from a hidden camera? Did the camera cause my wife to be in the hotel room with that man? I saw the tape!! Lets leave Godel out of this. Respond to something else on my bill of particulars. Remember I’m not talking about science, but about its use as general principle. I’ll leave Kuhn to Ellis. I’ve never read him.

By: Keith M Ellis

Keith M Ellis — Mon, 09 Feb 2004 19:21:53 +0000

Abiola, I repeat: in common parlance, "implies" does not mean "requires". You well may know Godel's proof better than I, but you seem to have difficulty parsing a simple sentence. And we both know Godel's proof well enough to know that the Godel statement certainly does not imply that in general logical statements are undecidable. It demonstrates only that a certain kind of statement is undecidable. Godel is inappropriate as refuation of Seth's statement. But it was nicely authoritarian. Isn't that ironic?

"Evidence? Don’t ask me to do your homework for you; provide evidence for this claim or hold your tongue."—Abiola

This is not my homework, it's yours. You're dismissing Kuhn and countless people who have done work you have not. You claimed that:

"Popperian falsificationism is actually a rather decent description of how good science is done."—Abiola

Unless you want to aver that most actual scientific work isn't "good science", then for this statement to be true most scientific results would need to replicated so that their falsifiability is relevant. But most experiments are not replicated because there is a community of trust that allows that if an experiment has all the trappings of a good solid experiment, then replication is assumed possible but unnecessary. Authority and credentialism win the day. This is how the majority of science is done. By necessity, it's not how revolutionary science is done. But revolutionary science is no more the model for the entire activity of science than is a consitutional convention a model of the entire activity of representational government. Tim keeps challenging me to demonstrate something I have not asserted. I never once claimed that empirical truth does not play a decisive role in science; I have no need, then, to demonstrate how opinion necessarily must win the day. I never claimed such a thing. You both are arrogant and propietary about science in the way that a basketball player might be arrogant and propietary about physics. That is to say, improperly. That you practice science no more means you have an aquaintance and facility with the study of science than does living in the physical world make one a physicist. The points I am making are well known and uncontroversial to people that do study science, however, and when you waltz onto this field of study and make loud, obnoxious and contrarian claims:

Thomas Kuhn was full of bunk. Fashionable theorizing to the contrary...I do know the odd thing about how science is done...I suggest you tak a look at one or two actual scientific publications. If you bother to read some of the papers in these two journals

...the onus is on you to back up your assertions with proof. Furthermore, your demeanor is arrogant and rude and fully warrants a "you're full of shit" from someone who's lost patience with it. There is nothing "wrong" with Popperian falsificationism; but as a model for the enterprise of science it is inappropriate. I have bent over backwards for you and Tim and repeatedly reminded you that I agree with you in the essentials of empiricism and that I am not the least some pomo trotting out a vulgar relativism in an attempt to knock down the "false god of Science". Here, there, and everwhere else I play your present role as Science's defender. Yet you insist on responding to me as if I were arguing things I am not, as if they were your strawman version of the pomo argument you clearly have a chip on your shoulder about. And, Abiola, our exchange about Godel is revealing: you quite inappropriately invoked Godel as an authority to contradict Seth Edenbaum. Whether it was a willful or unwitting misreading of Seth doesn't matter. When challenged on it you double down and yet again invoke authority. In doing so you are demonstrating mine—and even his!—point.

By: Abiola Lapite

Abiola Lapite — Mon, 09 Feb 2004 18:05:43 +0000

“Like you, the crackpots see the revolutionaries—the Einsteins, the Newtons, the Darwins—as exemplary of exactly what science is all about. Authority, consensus, the community of science is immaterial, the truth will out. Except they believe that science has been hijacked by the conservatives and the authoritarians and is in the service of lies, not truth (but only temporarily, for surely their genius will inevitably be recognized). You, in contrast, believe that no such hijacking has taken place; indeed, that such authoritarianism as exists embraces truth wherever and whenever it appears.” A fancy bit of ad hominem wrapped up in mind reading. All there is at the heart of Popperian falsificationism is modus tollens; if there’s something wrong with it, why not say so in plain english? Why all the blather about “authoritarianism” and so on? It seems clear to me that you don’t really know what you’re talking about. You lecture others about Gödel’s Incompleteness Theorem even though you probably couldn’t give an accurate outline of Gödel’s proof if your life depended on it, much less tell us what his Compactness Theorem or the Löwenheim-Skolem Theorems imply. You dismiss the evidence of people with real experience of what science is about in favor of a nice long rant about “authoritarianism” and so on. You engage in cheap name-calling and peremptory commands rather than fill in the gigantic gaps pointed out in your “reasoning.” Engaging in further discussion with you looks to be a losing proposition. Ars longa, vita brevis, and I have better things to do.

By: tim

tim — Mon, 09 Feb 2004 17:54:40 +0000

Now it is getting ridiculous. ‘An unintended consequence of your romanticized view of science is that it gives credibility to the crackpots.’ The simple truth is that whether or not the crackpots’ perpetual motion machines work has nothing – absolutely, positively nothing – to do with the authoritarian structures of the scientific community. As for other ideas the crackpots may have, I’m willing to let the evidence do the talking there, too. Speaking of which, you were going to provide some evidence for my challenges, weren’t you? QED, de Broglie, etc.? It was going to be something more than ‘You’re both full of shit.’ I hope. ‘You, in contrast, believe that no such hijacking has taken place; indeed, that such authoritarianism as exists embraces truth wherever and whenever it appears.’ No, I, in contrast, believe that such authoritarianism may be part of the sociological structures in which scientists work, but that it is not science; that a musicians union is not music; that the federal agencies sponsoring grants aren’t mathematics; that a tail is not a leg. Naive of me, perhaps. ‘And here we come back to being fully on-topic. By engaging in the public sphere this naive and false view of the nature of science, you and those like you are inadvertently encouraging the milleu in which pseudoscience thrives.’ On the contrary. Pseudo-science thrives when people believe that ‘science relies upon authority’ and not (for example) empirical evidence or falsification; when people believe that science and truth come from having a prestigious endorsement, not from having experimental results. Pseudo-science – like, for example, Lysenkoism, Keith – arises when people accept the authorities instead of evidence. Those people selling pseudo-science are selling it based on authority, on testimonial, on charisma, on claims that it comes from members of the scientific community (preferably in white lab jackets). If in fact there is a problem with the understanding of science that leads to this undesireable state of affairs, it comes from the people telling us that ‘science relies on authority,’ that science is whatever someone who is a part of the scientific ‘community’ says it is. And I think we know who has asserted these things in this debate. But I’ve had my say, time to get back to doing science. You get the last word.