Friday, October 8, 2010

Jerry Ravetz - a reply to various comments

Response from Jerry Ravetz to several comments in this blog

There is so much to discuss here, that I am nearly overwhelmed. That did happen to me after my first post on Wattsupwiththat. Hundreds of comments, of every possible variety and attitude, came pouring in. Just going through them was a full day’s work; I couldn’t begin to sort them, to say nothing of making a systematic reply. One result of this was that I was drawn too much into the debate as it developed around me, and quite forgot some key points of my own scheme of things. Just now I have a big stylistic problem: if I pick off points from the discussion it will be disjointed and incoherent, and if I give my own synthesis it will be just another essay! I have read and re-read the discussion, and hope to have something worthwhile here.



First, a few words on Kuhn. My understanding of his philosophy of science has developed over the years. For me the key passage now is the part of the preface to the book where he defines ‘normal science’; that is on page 5 of the original edition. There he defines normal science as: “the strenuous and devoted attempt to force nature into the conceptual boxes provided by professional education.” The irony here is quite devastating – strenuous and devoted, force nature… This could not be more different from Popper’s heroic scientist risking all in his dedication to truth. For Kuhn, the ‘puzzle-solving’ normal scientist is merely a process-worker in an industrialised Intellectual Property business, but worse, one who really loves his boss. I have found affinities between this character and the subjects in the book The Child Buyer by the popular author John Hersey. This man bought highly intelligent children from their parents, so that they could solve mathematical problems for National Defense. To make them work more efficiently, they would be manacled to their desks and blinded. When these details emerged, some parents protested, but eventually the combination of patriotism and profit won the day. The book appeared in 1960, when Kuhn was still working out his ideas.

My own ideas about science were derived from my experience as a research student in Cambridge, where I set and then solved (or failed to solve) my own problems, with minimal intervention by my supervisor. What Kuhn described (with such exquisite ambiguity, perhaps irony) as ‘normal’ science was for me merely technician’s work, and I described it as such in my book. That’s one of the reasons that I failed to take Kuhn’s work as seriously as I should have. Although Kuhn dealt only with ‘pure’ science, he was actually the first philosopher of industrialised science, when research changed from being seen as a calling, to a career or only a job. When I wrote my book I failed to draw out the consequences of this disenchanted vision. For I had worked myself into the conclusion that the successful pursuit of objective knowledge about the world depends on a subjective, moral commitment to doing work of good quality. Otherwise, there would be a sort of ‘Gresham’s Law’ of quality in science, with the bad work driving out the good. The roots of this commitment had to lie in some ideology, and so perhaps a ‘false-consciousness’ of science as the noble endeavour of seeking Truth was important for quality-assurance of the whole enterprise. I lived with that paradox, and eventually decided that ‘industrialised science’ lacked the means for its own justification, and hence a ‘critical science’ would be necessary for the rejuvenation of science. All that is subject to criticism, of course. What is relevant here is that in Kuhn’s ‘normal’ science, especially as described on p.5, there is no place for that moral commitment, and hence no support for quality-assurance, and hence no protection against corruption. Although he wrote some sixty years ago, it is highly relevant today. Kuhn described the defence of the paradigm by scientists in decidedly non-classical terms: “Normal science, for example, often suppresses fundamental novelties because they are necessarily subversive of its basic commitments. … (p.6) … when, that is, the profession can no longer evade anomalies that subvert the existing tradition of scientific practice …”. In the light of that description, I can now put the very subversive question: suppose that the embattled, anti-critical science done by the Climategate scientists is really just ‘page-5 normal’ - devoid of idealism or indeed of ethics, and moreover the ‘norm’ of practice. Then what future is there for science? I will return to this question later.

Bringing Kuhn up to date, quite recently I have coined the term ‘paleo-normal science’. This is puzzle-solving, but with the extra defining assumption that the natural world is like a Tinkertoy – interchangeable solid bits that can be taken apart and put together again. All thoughts of context, be it philosophical, ethical, social, natural-systemic or environmental, are excluded from its paradigm. Paleo-normal scientists are the assembly-line workers for the military-industrial-scientific complex, just as mainstream economists have been its ideologues. There may be somewhere a textbook or examination in a mainstream natural science that provides students with some protection against recruitment to paleo-normal science, by emphasising those contextual aspects. If so I would be delighted to hear of it. I use the term ‘paleo-‘ to emphasise that now, in this post-normal age, to persist with that sort of science is to be involved in an enterprise that is obsolete, however dominant and dangerous. It is analogous to persisting in neo-classical economics after the Credit Crunch. It is an empirical question, whether paleo-normal science has been dominant in the areas that are crucial for technological development.

This is the background to Post-Normal Science as I devised it with Silvio Funtowicz. In much belated retrospect, I now recognise a crucial ambiguity, indeed confusion, in the doctrine as we presented and developed it. We ask, just what is PNS? Do the ‘four horsemen of the scientific apocalypse’ define Post-Normal Science or a Post-Normal situation? The quadrant-rainbow diagram compounds the confusion, since the outer band is labelled ‘PNS’ while the whole diagram is labelled ‘PNS’. I always found that somehow unsatisfactory, but thought that the point was sufficiently obvious. I recall now that when I was doing mathematical research, the phrase “it is obvious that…” was the great elephant-trap. When I had reassured myself for the fifth time that a point was indeed ‘obvious’, I would then take a deep breath and check it. Revising the ‘final’ text meant tedious re-typing plus entering symbols in India-ink! And how many times did I do that!

Whatever I might have written down, it is clear from the present discussion that there is a lot of confusion around. The mantra ‘facts uncertain…’ describes a situation, that, with Roger Pielke Jr., we might well call the PN situation. I do not recommend the ‘Post-Normal’ for everyone; I just say that sometimes this is where we are. In those cases, we need a different style of doing science, the Post-Normal, which I describe only in the most general terms, as arguing ‘not by conclusive demonstration but by inclusive dialogue’. So far so good; and at this point I might take the liberty of criticising my critics. Because an essential aspect of the practice of PNS is the participation of the Extended Peer Community (EPS). Some have reduced this to ‘Extended Peer Review’, but that’s OK. This is where PNS goes political, and Silvio and I have related this extension of scientific practice to the advance of democracy in other fields. Indeed, I immediately saw (and repeatedly said) that one of the most significant aspects of Climategate was the presence of the EPC on the blogosphere; otherwise the emails would have not be interpreted so deeply, and could have been dismissed as the injudicious remarks of scientists exasperated by the unprincipled harassment by oil-company stooges. So I would suggest strongly that anyone who discusses PNS should please take account of the EPC. This has not happened in this discussion so far.

Having said all that, and thanks to everyone who has got this far, let me discuss the postings. First, there is Dennis’s criticism of the definition. I would stick by the four criteria as demarcating this from ‘applied science’ where, I insist, facts are (relatively, pragmatically) certain, values implicit, stakes not high and decisions not urgent. He makes an important point about our present situation, something that surprised and confused me at first. This is, that for many critics, the CAGW science is mistaken at best, or fraudulent at worst, since there is no genuine uncertainty, etc. about climate change. For them, the whole thing is a scare story backed by bogus science. Some of those critics, looking for an explanation of how scientists could have been so misled and corrupted, found the explanation in the writings of Schneider and Ravetz, as brought up to date by Hulme! This theory even got as far as a full-page article in the right-wing British weekly The Spectator. I might have hoped that this would restore my creds with the left-leaning warmistas, but that doesn’t seem to have happened.

The denialists did have a principled basis for their criticism of me, in the way that in my explanations of PNS, I rather airily said that we must replace Truth by Quality as an ideal for science. Some who explored this found the writings of Mike Hulme, which seemed to be defining quality in mushy politically-correct terms, such that Lysenko could have been a valid ‘extended peer’. My management of that issue was one of the worst blunders of my career. I failed to point to my work on quality assurance with Silvio and Jeroen, but I just apologised briefly and obscurely. My only explanation for this was that I was under extreme pressure at the time, and was so engaged with my critics that my other work slipped out of focus.

This apologia has really gone on too long. I would only respond to a remark by Dennis: “In old fashioned pre-post-normal situations, it was was old fashioned objective science that told us what would harm us and what would not, not some post modern discussion.” This takes me rather further afield than I discussed explicitly in connection with PNS, but it is a very important point. I think it relates to the dominant conception of risk. Once upon a time it was assumed by scientists that all probabilities and harms could be assessed and quantified, and either prevented or remediated appropriately by science. The counter-examples to this belief were not significant. But for those on a critical fringe, there were plenty of examples in public health, disease, and nutrition, where the causes, cures and even existence of harms were controversial. One great example is pellagra, which had been assumed to be the result of a yet undiscovered germ, until a very brave physician fought to demonstrate that it is actually a disease caused by a bad diet. Lead in petrol is another case, and even lead generally (poor children eating the flaking lead paint in their houses because it tastes sweet). But during the period of rising confidence in science as the solution to all our ills, it was believed that the outstanding problems of health & safety would naturally in the course of things be solved by science. Part of the significance of Beck’s ‘risk society’ was the acceptance that modern technology systematically creates new risks along with reducing old ones. Our (partial) loss of faith in old fashioned science is partly a result of a recognition of the complexity of life, disease and death in the modern world.

In conclusion, could I just hammer the point home, that when there are insufficient data for a genuine significance test, and there are no theories but only dubious models, and when values and presuppositions influence the setting of priorities (determining knowledge and ignorance), the framing of problems, the choice of methods, etc. etc., where is ‘old fashioned objective science’? Somewhere I give the example of road safety research, focused on the protection of people inside a car when it hits something, but neglecting those on foot or on bikes who are getting hit. Yes, there will be high-quality objective scientific research on in-car restraints etc., for the favoured class on the inside of the vehicle, but the other road users may well complain that the whole research programme itself is rigged, perpetuating ignorance about their plight while apparently producing objective knowledge. In making this observation, am I betraying the ideals of Truth in Science?
I can now answer Hans’s question about the young researcher in climate science. Should s/he aim at finding the truth, or only try for quality? Well, a philosopher friend of mine has given me a distinction: the scientist should of course aim at being truthful, that is saying honestly what they believe the facts to be. But to believe that one can find Truth in climate science, on a par with 2 + 2 = 4, is to be back with Galileo and his incoherent and incorrect proof of the motions of the earth which he had to believe was True. On the other hand, Quality as term reminds us of the complexity of knowledge and the moral engagement that is necessary for its production, diffusion and wise application. I have not written extensively about Quality, but there is Jeroen’s work and also the whole Quality Assurance movement which has influenced every sphere of human productive activity with the paradoxical but significant exception of scientific knowledge.

Thanks for bearing with me… I look forward to further discussions…

And my apologies for this prolonged delay, for which I can only plead overwork on too many interesting things, so that the draft of this essay got overlooked in my computer.

3 comments:

PolyisTCOandbanned said...

tl: dr.

Anonymous said...

Science is interested in true theories, but is only capable of proving that theories are false.

Which theory has more Quality, a true theory or a false theory?

Assume a false theory has more quality. Now, the number of false theories about some event is trivially shown to be infinite. On what grounds are we now going to prefer one false theory above all others?

Assume a true theory has more Quality? We do not want false theories, so we would want to eliminate all false theories from the collection of not-yet-falsified theories. This is exactly what Popper has been saying all along.

Werner Krauss said...

I really enjoyed reading this post. I especially enjoyed poetic expressions such as 'military-industrial-scientific-complex' - a term that is almost forgotten after the end of the cold war. Each time deserves its own science discussion.
With bringing 'climate' into play, I think Ludwik Fleck is also very interesting. Climate research is very much dominated by the question of anthropogenic influence. There are very good 'quality' results already. But, of course, this is only one small aspect of climate. There are so many other things to know about the functioning of climate - that is what we can learn from Fleck. When we know one thing, we don't know everything about the subject. Climate, as far as I understand, is still a bag full of surprises.
What I want to say (with Fleck) is that our knowledge of climate change is, of course, partial. Partial knowledge, partial truths.

The political relevance of climate brought back the good old 'truth', 'objectivity' and even 'genius' attributes to science. I think these values are kind of outdated. In a world of partial knowledge it is about 'good' or 'bad' science. That's pretty sober and simple.
For example, obviously Mann's hockey stick was not really well crafted. Others corrected it, now the hockey stick doesn't look like one anymore. Maybe more realistic than before - that's fine.
This is exactly what we can expect from science. Good work, improvement of previous work. It's not that dramatic. In my understanding, post normal science also means saying good-bye to that super-romantic image of the scientist as a hero and impersonator of pure values. Instead, science is producing partial truths that politicians and citizens have to deal with and to draw conclusions from. It is as imperfect as these bureaucrats of knowledge are - and from peer review to research funding, there are permanent improvements (and the opposite, too) under way.