With all due respect, Ben, I'm getting the feeling you don't actually understand how science - or, rather, scientists - work. "But you wouldn't believe it if it was!" is not actually an argument for the disprovability of a scientific theory.
Lucy, I have a postgraduate diploma in science myself, and have worked as a scientist for at least ten years in my life. I just don't work in your science. Furthermore, I wrote the start of my master's thesis on the philosophy of science, so it's not like I'm ignorant about the matters I'm discussing with you. Yes, I know that's confusing, a philosopher and a scientist. The horror. I'm sorry that the way in which I'm discussing science with you is confusing because it's informed by dozens of modern philosophers and scientists opinions on the epistemology of the discipline. It's meta-science, and it's standing on the shoulders of giants. I left naive inductionism behind decades ago. I left Popper behind then too. When I'm discussing Darwin with you, I'm doing it having read a great deal of the philosophical discussion on the topic, which has left me with a very strong feeling that it's far, far more controversial than you let on. Or perhaps realize. I don't know which is true, which is why I'm discussing this with you, maybe you'll learn something too. Did you understand the idea of the negative heuristic, as expounded by Lakatos? Try it, then you might begin to understand that I'm not trying to deny evolution, nor am I saying that natural selection being nearly a tautology is a bad thing. It's just a curious thing, a theory whose position in science serves a rather unique function.
When I said you have a bunch of recourses that make the disproof of natural selection virtually impossible on logical ground alone, I wasn't just pulling your tit. This exact argument has played out so many times that I'm just bored of doing it again, and cutting to the chase. Have you even considered the point I'm making, or are you so busy giving me a Darwin for Beginners course that you haven't spotted that I already understand the theory? That's how it feels to me.
So, one more time, the hypothetical experiments or observations that you suggested would hardly do any damage to the idea of natural selection at all. They would just join the massive plethora of odd anomalies that have been observed in nature, where some trait would seem advantageous, but hasn't occurred, until the reason that it hasn't occurred is explained by some other contra mechanism. Or, it can just be put down to "hasn't happened yet, oh well". Or it might just involve a slight change to the particular timelines of evolution. My point is that the core belief is pretty much unassailable, and mostly that come down to tautologous nature of "the survival of the fittest" where "fittest" comes down to "what the survivors are like", so it's a theory of the survival of the survivors. Which doesn't make it a bad theory, especially considering what it displaced, it just makes it a rather strange theory, one that could really only be defeated by an entire paradigm shift (read Thomas Kuhn for an explanation of what that is) right across biological science. I make no predictions of the likelihood of this. I think evolution is true, after all. Natural selection is just a little part of it, and the idea of it opened up a whole different way of looking at how things can come to be as they are, which led in turn to the discovery of a great many of the mechanisms along the way. Most of that stuff is the more conventional kind of scientific theory, stuff that can be proved or disproved by experiment and/or observation.
I've personally studied natural selection at work intimately in optimization models, in particular genetic algorithms. That was my work for several years recently. I've not only watched evolution, I've designed it. It works. It's an ironic position to be in, intelligently designing evolution.
I'm still quite keen on Popper / falsificationism, as a minimum standard for a science. I think falisficationism should be rejected as an adequate description of a science, or as a reason for holding scientific knowledge in high esteem. A necessary, but not sufficient description of science, perhaps?
Disclosure / excuse making: I only ever read undergraduate Philosophy of Science, so my understanding of the area is a bit limited.
And when we present you with examples you dismiss them as luck or sexism. Sorry Ben but that isn't productive.
I'm not sure whether you're speaking of one or two argument lines in this thread, or just making a general comment. I didn't "dismiss" anything, I responded to an argument given with a counter argument, which is what my understanding of one part of constructive discussion is. Do you have a counter counter? Can you deny sexism in the sciences? Do you deny that there is a tremendous amount of luck in science, that things are often discovered or not discovered quite by chance, and this can mean years, decades, maybe even centuries before some kinds of progress are made? Have you never gone down a blind alley, and reflected that you could have saved yourself substantial time if you'd taken the other path, but how were you to know in advance? Furthermore, do you deny that a lot of brilliance is often judged by success, so luck may be a big part of it?
Of course it isn't proven, no theory is, merely failed to be disproven.
You're Popperian, then? You do realize that idea came from a philosopher, right?
You are arguing that you should ignore the advice of those with actual experience. That's an experiment that has already been done.
No, I'm not arguing that, nothing like it at all. You present a completely false binary:
either you let scientists who have had the experience of working with brilliant people make the choice on who should get funding or you let an accountant in wtgn make that choice on who should get funding.
which is a hobby horse of yours, but by no means covers the options. Another option is that many people can have input into the decision, including the shop floor scientists, the scientists who are somewhere up the chain towards management, the specialist managers, and the people who have to pay for it all decide. Also, the books have to balance, or it's unsustainable, so accountants really do need to be involved. That's a fact of life you have to deal with, I'm sorry.
You have repeatedly made the contention that we suffer from a lack of brilliant people in the sciences, but you're coming entirely from only one of those perspectives, the guy on the science shop floor. Which makes you like some footy player in a club saying that they really want Dan Carter in their team. Sure, great, that would be nice if you could afford him, and if it would make a difference, really. The rest of the team might be so crap that it would be a complete waste, and you'd still lose every game. Of course it would make some difference, but the question of the quantity of difference is important and you've really come back with nothing on that, other than to say it's difficult to evaluate. If it's so difficult to evaluate, why are you so certain of it, then? THAT is really my main point. For every anecdote, there's a counter anecdote, so we moved rapidly to an impasse which can only be solved by harder evidence, rather than a longer list of more anecdotes.
I’ve also studied both science and philosophy.
Popper’s falsificationalist model is certainly flawed and incomplete.
But we don’t have to agree with Feyerabend and conclude that there is no reliable “scientific method”.
One way falsificationalism can be made more applicable to real science is to acknowledge that we usually can’t test hypotheses in isolation. Instead, in practice, scientific experiments and observations are used to test hypotheses in bundles (as argued by Kitcher). If we get an unexpected result, then that means that something is wrong with at least one of our starting assumptions – but we don’t know which one(s), and the flawed assumption(s) might be some that we didn’t even know we were making. We then need to test different combinations of hypotheses in order to work out which one(s) are flawed.
So, progress in science requires that we can recognise hidden assumptions and devise ways of testing them.
It’s not a perfectly automatable process. It involves creativity. It involves some trial and error. But that doesn’t mean it’s not a reliable method of finding out about the universe.
I'm still quite keen on Popper / falsificationism, as a minimum standard for a science.
I still rate the guy extremely highly. For starters, his writing is highly accessible, and that's a sign of brilliance to me, when someone can explain a very sophisticated idea without very much jargon. Falsificationism is, to my eyes as a philosophy student, one of the most original and profound ideas to have entered the field in the twentieth century. It's the best answer I've ever heard to Hume's problem of induction, the problem that effectively ended the entire school of empiricism. Prior to Popper, it was all about induction and legions of philosophers trying vainly to reconcile that the principle of induction can't stand on deductive grounds with horrible theories like "we are programmed to believe this irrational thing". Popper opened the possibility of putting science back onto a deductive footing to explain it's extraordinary access to truth.
Lakatos builds on Popper, puts theories into the context of entire research programs, rather than individual ones, which goes a long way to redressing the main problem with Popper, that practically every scientific theory would not have got off the ground because it would have been refuted early on, before it could be developed. Feyerabend argues very strongly that Lakatos was effectively an epistemological anarchist like himself, though, masquerading behind Popper's demarcation criterion as a strong believer in method. When I read that, I had my first real experience of aporia, in the Socratic sense, to this day it has me totally confused and feeling like I don't really understand anything at all about science, despite having forged through all that theory. I'm essentially a skeptic, I can see that anarchism can't be refuted, but it does seem rather useless. So now science is incomprehensible. I don't mean that I don't understand scientific theories or believe them. I just mean that I'm buggered if I can say clearly where they get their power from. I haven't kept up with the theory since then, I understand postmodernists engaged in an all out war to have their views accepted/understood, and that many of the softer social sciences have benefited from that. To me, that kind of war is like a paradigm shift, indeed such ideas as paradigm shifts and incommensurability are part and parcel to that angle. I'd love it if someone who did get that stuff could comment.
ETA: Linger! Snappy snap.
I didn’t “dismiss” anything
Yes you did. Lucy and I have presented examples which you have simply dismissed as either luck or sexism. We presented the examples as instances where a person of exceptional ability made a change in the field that few were able to understand. And you ignore those examples.
It's hard to have a discussion with someone who is willing to ignore and dismiss data.
That luck and sexism exist in science does not mean that all brilliant women are merely average and suffering from sexism and that all brilliant male scientists are merely lucky.
Another option is that many people can have input into the decision
Ben have a look at the way science funding is managed in New Zealand. Apart from The Marsden Fund the overriding deciding factor in all other funding in New Zealand science is economic benefit to New Zealand as determined by accountants. It is not that I demand that accountants have no say in science funding but that I want a system where accountants are not the ONLY ones who decide science funding.
And to get back to the analogy. I don't mind if we can't afford Dan Carter in the Foulds Park 15, unless of course the reason we can't afford him is because we have to pay the club accountant $300k.
Quantity of difference??? Lucy and I have presented you with examples of people who either changed science utterly or advanced it by decades if not hundreds of years. You've casually dismissed those examples as irrelevant. We've tried to explain the hands on effect of these people in the lab and you dismissed our experience as so trivial as to be immeasurable.
But that doesn’t mean it’s not a reliable method of finding out about the universe.
Well, what else do we have? And if someone found something better, that would just become the new thing we called science. It must be able to evolve, or it could go into a long stagnant period. But if I got anything from all of those thousands of hours of tweaking evolutionary algorithms, it was that there's no ideal set of forces and pressures for solving all objective functions. Sometimes massive diversity in the evolving population led to rapid convergence. Sometimes the opposite, a sudden rush of convergence would be coupled with a very low level of difference, as the algorithm discovered the exit to a local optima, and all the population crowded into it into the rush to the next plateau. At those times, a population of two converged fastest. And at the end, there's no knowing if it's done, it doesn't prove that it's found the optimal solution - maybe it has, or maybe there's another rush just around the corner. When that is suspected, that we're in vicinity of the optimal, the evolutionary algorithm became very inefficient, and brute force methods made a lot more sense. I'd flip those on, and see the objective start climbing again. Always, unless you got to a solution that was perfect (like a classification algorithm perfectly partitioning the sample data), you couldn't really be sure that you hadn't climbed the wrong mountain entirely, and were fighting for every inch of uphill when just across the valley some towering colossus's mere foothills were higher than where you had got to. There's a very deep analogy to the human search for truth in all of that.
I'm totally undecided on the value of philosophy anywhere in this. Ideas like Popper's stand a good chance of becoming counterproductive dogmas (Popper foresaw this, and consider his own theory to also be a scientific theory, which could be refuted one day). Science that's right on the cutting edge is often struggling to be recognized as science, because it involves rejecting a lot of current principles, and possibly the tolerance of being less accurate on a number of point for a while as the theory develops.
Yes you did.
No, I didn't and I've answered this criticism, whereas you're just reiterating your last post and now inventing positions for me. You're not engaging. Good day, sir.
This seems apt.
LET US TURN NOW to the celebration of scientific rigor. Individual sciences develop rules and standards for appraising evidence—as they learn about aspects of nature, they learn more about how to learn. At any particular stage of inquiry, communities of scientists agree on the canons of good inference, so that the work of certification of new results goes relatively smoothly. To the extent that the agreed-on rules are reliable, knowledge accumulates. It is important to understand, however, that at times of major change the standards of good science themselves are subject to question and discussion. And this observation, amply demonstrated in the history of the sciences, has important consequences.
From " The Trouble with Scientism", in The New Republic, by Philip Kitcher, who is Professor of Philosophy at Columbia University.
Philosophistry seems to be a brand new (mine)field...
Surely a brand new (mind)field.
We'd have to do a double-blind belief test on that - blinkered and ya miss it!
I wonder if this (very interesting) impasse could be partially resolved if we just agreed that particularly brilliant people are useful because they save time. Ben's contention (if I understand it) that one brilliant person could be replaced by several diligent people with a bit more time implies this anyway, and Bart's experience confirms it.
The debate seems to turn on the point of quantification. For what it's worth, I'm not sure that most decisions around hiring / funding etc really require a precise quantification of ability, as Ben is demanding. In most situations, a comparative judgement seems to be enough. A gets job or funding over B because s/he is better, not because s/he is 1.8 times better.
I know that 'better' is problematic and unsatisfactory, but the reality is that in any given field (science, rugby..) there are so many qualities contributing to the overall package that quantification is impossible anyway. In both fields, for example, the value of forming good 'combinations' is every bit as important as brainpower or speed, but far harder to quantify. Or at least, harder to quantify to those who don't understand how the combinations work. Which I think is the point Bart is making about accountants holding the purse strings..
now inventing positions for me
Read your response to Lucy on page 4 Ben.
quantification of ability
This is the bit that is causing difficulty. I, and I believe Lucy as well, am saying that there is more than just a quantitative difference at the top end, it's a qualitative difference in ability. In my experience those people can never be replaced by any number of good people.
My problem with funding students the way we do now is that those qualitatively different people will choose to avoid science. We don't have person A to give funding to because they have chosen a profession that will pay their loan back faster.
Ben is arguing (I think) that there is no qualitative difference and that the quantitative difference is insignificant. We don't need the best and brightest we can make do with more of the good ones.
Oh and BTW Darwin’s contribution wasn’t a Theory of evolution it was the Theory of natural selection.
Thank you Bart.
This is the bit that is causing difficulty. I, and I believe Lucy as well, am saying that there is more than just a quantitative difference at the top end, it’s a qualitative difference in ability. In my experience those people can never be replaced by any number of good people.
That's been my experience in various settings.
You think wrong. I'm arguing against something far more specific - your point about attracting the best and brightest by paying for the through education funding, justified by their higher scientific output. You're trying to turn it into a question of whether having more brilliant people is good, with your fantasy version of me saying it's not. I'm asking the question "How good is it? Is it really worth the cost?". You can't answer, except through pointless anecdotes about how brilliant people did brilliant stuff, which don't even need to be addressed because they are not in dispute. I raise the obvious counter anecdote that non-brilliant people sometimes do brilliant stuff, and don't get me started on the number of brilliant people who came to absolutely nothing. I also raise the point that brilliance is a nebulous quality, quite possibly largely accounted for by simply working hard, or measured only after the fact by the quality of the output. You tried to turn this into me saying there is no such thing as brilliance, which is patently absurd, and I've denied it several times, and yet you persist. This is failing to engage.
Yes, I know it's hard to quantify brilliance. Which is exactly what's wrong with making claims that you must have more of it. You can't even measure when you've got it, nor what it was worth. Some of your anecdotes are even explicitly to that point - brilliance that was unacknowledged. How are you going to convince me that you could find a Margulis, or that you would recognize her when she interviewed? More likely, you'd be one of the people writing something back to her from the journal saying her ideas are too whack to publish, because you don't get them, being a self-confessed B grade.
Furthermore, the idea of encouraging more brilliant scientists into the NZ workforce via education is so indirect that it might do nothing at all, except produce brilliant scientists for other countries at great cost. Indeed, that is what you identified as the problem right from the start, that you're not seeing the good CVs. Which suggests that the problem is not the education system, but your actual industry, here in NZ, which may simply be unappealing to the brightest graduates. Maybe it's the pay. Maybe it's the funding generally which means crappy labs and not enough equipment. Maybe it's just that this is an isolated, impoverished, small backwater, and talent always goes to the center of where it's happening in their thing, in every field, and you're always going to have to make do with the B students.
This isn't tall poppy syndrome. It's a simple acknowledgment that you have an unrealistic idea about what's involved in convincing people to part with their money. Everyone wants the best, until they have to pay for it. Your say so that you need more brilliance in the lab, therefore science degrees should be more heavily subsidized needs more work. I'm actually trying to help by pointing out the shortcomings, believe it or not.
Generally, I think we need more funding for science, sure. I have never agreed with student loans, period. And I think you may have something of a point that in this country we have a problem attracting people to degrees for which incomes are not high, and that is to the detriment of the entire society. But I couldn't really care so much about brilliance - my take is that it's fairly random where it shows up. The problem of making students think about the financial outcomes of their education affects ALL student numbers in the impecunious subjects. I actually care about ALL people, rather than just brilliant people, who I think have considerably less need than most people to be given extra advantages. We already give them far, far more help than other people. This is not tall poppy syndrome, it's my sense of fairness and social responsibility.
Yes, I know it's hard to quantify brilliance. Which is exactly what's wrong with making claims that you must have more of it. You can't even measure when you've got it, nor what it was worth.
Trying to define value only by what you can measure is the usual flaw of accountants and other counters of beans, magic or otherwise.
The accountancy and law degrees are cheaper and easier to get and offer access to an employment market with many customers – a no-brainer on a personal level. The science and engineering degrees are harder to get and offer access to a smaller employment market and lower incomes. So, clearly, all sensible people should become lawyers and accountants. But this isn’t so good on the level of national economic development – and, crucially, it’s every bit as much of an artefact of the system you’ve chosen to distribute the public resources as, say, people choosing science degrees because there’s a 100% subsidy for the cost of science degrees.
Factually, this is not really true.
1 EFTS of Law is $5,243*; a year of bench science $5,743. But a BSc takes three years, and a LLB four. So in fact a BSc is about $5,000 cheaper than an LLB. (Except you have to add in (a) the cost of living for a year, which is probably around the 10K mark, and (b) the opportunity cost of that year, which is probably floating around the 30-40K mark. So I would imagine an LLB costs roughly 15k more ignoring opportunity cost, and maybe 50k more including it.) Roughly the same maths applies for anyone wishing to become a chartered accountant. While the EFTS are slightly cheaper, it is a four year degree. A BEng EFTS is $6,217, and the BEng is also four years, so that’s $4,000 more than an LLB or Accountancy.
While claims about relative difficulty of degrees are a minefield, I would note that both law and engineering have very high failure rates in first year. Science and accounting are both technical, math heavy subjects, but to become an accountant (obviously not the same as an accounting degree) you must do four years not three, and complete a professional qualification on top of that. (As in fact you must for law.)
So in fact no, law and accounting are not cheaper and easier than science or engineering. Possibly, law and accounting are cheaper and easier than engineering.
But it isn’t like engineers are badly paid! I would say that engineers tend to be paid as well, if not better, than lawyers or accountants.
And it isn’t an arbitrary fact that training engineers costs more. To make someone an engineer (or a scientist, for that matter) is intrinsically expensive — you must provides labs, materials, practical experience, etc. To make someone a lawyer or accountant, you must provide teachers and books. So it isn’t an artefact of the funding system; it is an artefact of the world. If you give a accountancy department and engineering department the same money, the accountancy department ought be able to produce more accountants than the eng department engineers.
Further, if in fact we have an undersupply of scientists, we ought see the wages paid to them rise, until it becomes an attractive career, and then the supply will increase until equilibrium is reached. Is there a reason this process shouldn’t occur? (And the reverse, of course, for lawyers or accountants.)
* All figures from the University of Canterbury for 2012, found here. I have assumed degrees are purely composed of the relevant efts, which isn’t in fact true. I would guess the effect would be to make all the degrees slightly cheaper, but not noticeably.
My reading of Feyerabend left me convinced that he was describing a social process, not engaged in the project of trying to work out why we ought to give scientific thought special status. Hmm... actually, I take that back. In effect, he was saying, we ought not to give it special status, because it's just one social activity among many. However, that simply doesn't gel with my confidence in aspirin, or epidurals, or any of the magnificences of contemporary medicine (c/f the woo of homeopathy etc. And yes, if homeopathy works for you, that's jolly nice, but all the double blind studies to date show that it's really the placebo effect. Which is great! Just not something you can prescribe without playing on people's credulity.)
Back to Phil. of Science... I get off the boat somewhere between Kuhn and Lakatos, 'though I am probably somewhere closer to Lakatos. One of the things that works about science is the extent to which we build a web of knowledge, so that things that we learn in one area make sense, or fit in with what we learn in another area. Evolution and botany and geology and physics all help to explain and/or confirm the movement of continental plates. If we were to find that the flora and fauna on either side of the Atlantic were totally discontinuous, then we would have powerful evidence suggesting that plate tectonics was not correct as a theory. As it turns out, at present, all the knowledge we have to date tends to confirm plate tectonics.
Of course, what would really happen is that if the flora and fauna were discontinuous, we would start looking for fossils, or start researching speciation, or whatever, to save the core hypothesis (i.e. that continental plates move). That is because science is also a social process. But over time, scientific truth (believe me, I'm wincing as I write that word) will out.
There's a heap of other examples I could use rather than plate tectonics, but it's one I happen to love, because it is so new. 1960s, I think. Suddenly, when plate tectonics became widely known, within scientific circles, a whole lot of other things made sense.
So I don't go as far as Feyerabend, and certainly not into a state of aporia. But that could just be because I have not read enough Phil of Science to get into the state in the first place.
Keir, unless I'm mistaken, the 'costs' per eft are in fact what the University charges NZ students- not what a years study in that field costs the University. That's closer to the overseas costs- (domestic fees, plus Govt money- the part Joyce is intent on tweaking- of approx 2x the fees- so in the region of $15k per EFT). There's a pretty big margin added for foreign students- which has turned NZ public education into a business, at least partially.
A big chunk (about a third) goes to support central services- AV and HR, student services, facilities- gardeners, and cleaners; vice-chancellors and marketers, etc etc. It's arguable you could run a university without some of this- but not without all of it. The rest goes direct to Colleges/Depts.
<disclaimer>I think- subject to observer error and bias- and I'm not the registrar :)
Well, the question of what a year’s study in a field actually costs the university per student is somewhat tricky. (Because of accounting for centrally provided services, opportunity costs, non-linear costs etc.)
But yes, the fees paid are roughly half of what the government contributes per student. But that’s a bit too generalised, because different degrees get different government contributions. You can look at those here, and observe that per EFTS, you are looking at a 6K contribution from the gov’t for the arts/law/business/accountancy, and a 10K contribution for science/engineering.
Yep- more for science and engineering, and even more for medicine, I gather.