This is the Nature Network and therefore Terribly Extremely Very Serious foothold for Nature Senior Editor Henry Gee. If you want fun and games, visit http://cromercrox.blogspot.com/
I’ve just come across David Sloan Wilson’s inaugural blog on SciBlogs which is entitled, in part,
Science as a Religion that Worships Truth as its God
I don’t think I’ve read or heard anything more misleading all day, and in this post I hope to explain why I am so concerned.
The short answer (don’t worry, I have a longer one handy) is that it is not the business of science to discover ‘truth’, because ‘truth’ cannot be judged to be such, in any absolute way. To put it another way, were we to stumble upon the ‘truth’ we could never know that we had done so.
What science is all about, in contrast, is the quantification of doubt.
It is doubt, friends, that fuels science: the testing of hypotheses; the subjection of scientific ideas, grant applications, papers and presentations, to exacting scepticism.
Let’s try this out with a familiar example, albeit set in a fictional setting. Epping Forest Virus (EFV) is established as the cause of the debilitating disease Big Rock Candy Mountain Spotted Fever (BRCMSF), a dreadful pestilence in which the teeth of bulldogs turn to rubber, policemen affect wooden prostheses, and (most distressingly of all) the hens all start to lay hard-boiled eggs.
Scientists at MegaPharma, by dint of looking up anteaters’ behinds working all hours pipetting very small quantities of clear liquid from one vial to another and back again, have come up with WonderDrug™, a small molecule that interferes with EFV’s replication cycle. After lots more work, the expenditure of truly eye-watering quantities of money, and Clinical Trials I, II and II (after which they’re thinking of filming a sequel), WonderDrug™ makes it to market, a terrific potion that has no side-effects whatsoever and kills all known germs dead. Tautologically.
But is it true? By this I mean – can one really claim 100% success for WonderDrug™? Will those bulldogs get back their bite? The answer, of course, is ‘no’. Clinical trials only work on a finite sample of people. And even were WonderDrug shown to work in a million cases, there is always the possibility that the million-and-first patient will find their skin turning blue and bits of them falling off as a result of taking WonderDrug™.
What’s the lesson here? It is this – that the effects of WonderDrug™ cannot be known – all we can do is frame various degrees of doubt based on the tests we have carried out.
I am amazed, actually, that anyone can doubt that science is all about doubt. After all, any statistical test you carry out – any test at all – ends in the computation of a probability value, which is rated against an arbitrary significance estimate. If the effect you’re after comes out with a P of less than 0.05, it does not mean that the phenomenon you’re after is true. All it means is that there is a chance of less than one part in twenty that the result you see will have been a fluke. This is not a measure of truth – it is a measure of doubt.
Here is a more concrete example that’s closer to my own backyard, and which I describe at length in my book In Search of Deep Time.
In that book I talk about cladistics, which for the purposes of this post is a method for estimating the degree of doubt inherent in schemes of evolutionary relationships. Cladistics was especially appealing to palaeontologists, who were constrained to produce such schemes from very fragmentary evidence – and who woke up to the fact that you could not simply draw lines between fossils and make claims of ancestor-descendant relationships, replete with evolutionary trends, and gaps obligingly filled with missing links. Such schemes were as unfalsifiable as bedtime stories and did not count in any way as serious science.
Energised by such lessons, palaeontologists at the Natural History Museum in London put on (in 1980) Man’s Place in Evolution , an exhibition of human evolution in which the various extinct species of hominin were treated comparatively, rather than presented in the canonical cliche of a parade of creatures walking across the stage, getting steadily more evolved as they did so, and culminating in Modern Man. Instead, visitors were invited to draw their own evolutionary conclusions. This enraged one commentator, the late and much-missed Lambert Beverly ‘Bev’ Halstead, who wrote
‘The well-attested sequence of human fossils representing samples of succeeding populations has … been taken as a classic example of the gradual evolution of a single gene pool. Certainly there is not any serious doubt about Homo erectus being directly ancestral to Homo sapiens’ (Nature 288, 208; 1980)
The equally late and much-missed Colin Patterson responded by asking Halstead to be more specific about his sources, or, rather, challenging Halstead to examine the basis of what he regarded as undeniable scientific truth:
‘Confronted with these statements, one must either bow to Halstead’s scholarship, or ask ‘attested’ by whom? ‘been taken’ by whom?’ (Nature 288, 430; 1980).
But it was Donn Rosen, Patterson’s colleague (again, late and much-missed), who put the boot in, bringing us to the punchline of this post.
‘Halstead states his convictions that human evolution was gradual, that its gene pool of the past had certain knowable characteristics, and that the ancestry of living species can be determined with deadly accuracy. Halstead’s convictions arise from his discovery of a new form of doubt, such that he can contend that ‘there is not any serious doubt about Homo erectus being directly ancestral to Homo sapiens ’. This is certainly the news that Biology has waited for, the moment when the Truth can at last be known so that all this difficult and extremely tiresome theory can be dispensed with. Until now my colleagues and I had always imagined that to doubt something was ’to be uncertain as to a truth or fact’ and that the notion that distinguishes science from, say, politics is that in science uncertainty about the truth must remain or progress ends. Halstead’s discovery can only mean that he has in hand a new form of truth – a kind of truth that can be known. Many of us puzzled over what kind it might be until it finally dawned on us that it emerges from false doubt and, to honour its discoverer, I call it Halsteadian Truth ’ (Nature 289, 8 and 105; 1981. Original emphasis).
In other words, there is ‘scientific’ truth, which is really a form of doubt – and ‘Halsteadian’ truth, expressed as false doubt, and which really means a truth that can be known, absolutely and forever.
Having discussed all this, I am now in a position to ask a particular question – no doubt rhetorically, because nobody will give a damn what I say, I might as well be screaming into an uncaring void, that’s the trouble with this blog, nobody loves me, it’s all that Richard Grant’s fault, the great ‘airy fool – about the nature of the truth in David Sloan Wilson’s headline. Is the Truth which he claims Science Worships based on genuine doubt, or false doubt? On provisional solutions, or on Halsteadian Truth?
I shall close by repeating a point raised by Donn Rosen, which I find particularly telling.
’ … the notion that distinguishes science from, say, politics is that in science uncertainty about the truth must remain or progress ends.’
To understand this point is vital, for in it lies the distinction between gullibility and scepticism; between science and religion; between the vibrancy of a society in which free enquiry is practiced and valued, and one in which the truth is prescribed from above, whether by a religious leader, a militant atheist, or a totalitarian dictator, and the result is stagnant, stultifying and oppressive.
Last updated: Friday, 23 Oct 2009 - 00:14 UTC
© 2009 Nature Publishing Group
Doesn’t that sound incredibly arrogant? Or to put it differently, would Galileo’s Pope have said exactly the same thing?
scientifically (and humbly) yours,
BJB
It reminds me of the Heisenberg uncertainty principle, should exist.
Excellent post Henry. I was writing something on this topic at the exact time you were: Science and the Worship of Truth
However, at the same time, I don’t think the specific example of WonderDrug applies to the way Wilson is using “truth” in his post. He’s talking about objective reality in the same way that physicists are seeking a Grand Unified Theory that accurately explains natural phenomena. I suspect he would agree with you on the level that you’re talking about.
Thanks, EMJ. In general I found Sloan Wilson’s post unobjectionable. I agreed with quite a lot of it. It was the headline, though, that grated.
David Berreby points out one pertinent consequence of our failure to understand that science is about doubt, rather than truth. The Anti-Vax community profit from this by promoting the view that if science hasn’t discovered the ‘truth’ then all is confusion and discord. Creationists do much the same.
Nice post Henry. As Indiana Jones said to to his class in the first film: “Archaeology [and by extension science] is the search for FACT. If you want truth, try the philospohy class down the hall…”
I remember that scene, Arran, – I also remember whooping for joy at the time.
Well, is science a search for ‘fact’? It’s more the process of searching, innit?
“The Anti-Vax community profit from this by promoting the view that if science hasn’t discovered the ‘truth’ then all is confusion and discord.”
I’m amazed by the ideological breadth that this postmodern claptrap has attained: it seems like every nut-job from the far-left to the far-right invokes it.
Well said, Mr. Gee.
Thanks.
One endeavours to give satisfaction.
The assumption that evidence equals truth also has implications in situations opposite to to your WonderDrug example. It means when there isn’t evidence, people are confident it’s not true, like with chronic fatigue sydrome. I just wrote about it and mentioned your rant, which saved me elaborating on the doubt/truth idea. Thanks!
The Heisenberg uncertainty principle suggests that God does not play dice. There is always a space in which various scenarios can be developed according to the observed object, there is always a correction factor for other discoveries waiting behind them. At this point nature is beautiful. Always there is a chance. There are always new possibilities endless, I say.
Stirring stuff, Henry – for head and heart.
@ Cobi: thanks for pointing me to your blog. And what you say is entirely right, and boils down to that old chestnut ‘Absence of Evidence isn’t Evidence of Absence’.
Now, that doesn’t mean that any
loonymonomaniacindependent thinker has a license to bang on about their own complex ideas, built without reference to the existing body of knowledge.All it means is that one must keep an open mind to new possibilities for which some evidence might conceivably might exist. For example, your blog notes a new and tentative finding that links chronic fatigue syndrome with a particular virus. This shouldn’t be controversial – after all, somatic disorders have a history of being linked with psychiatric complaints for time immemorial – until causative agents were discovered that provided a more plausible cause. All the way from medieval diagnoses that attributed neonatal malformation to the effects of being cursed by a bggar – to the association of stomach ulcers with stress rather than Helicobacter pylori.
@ Alejandro: Quite right, too. If the findings published in every paper in Nature or anywhere else published were ‘true’ in an absolute sense, we could soon pack up and go home.
@ Stephen – thanks. Your comment crossed with mine. And very appropriate your comment is too. The cladists (all those late and much-missed people) felt very much from the heart that science was a matter of doubt. They were genuinely passionate about it. I had the great good fortune to have worked for a few weeks with Colin Patterson and his colleagues, when I was at an impressionable age, an I have inherited the same passion. I also knew Bev Halstead personally – he was every bit the colourful rogue that his reputation painted.
Henry – sorry, I meant at the natural phenomena no to the journal Nature. We do not control any natural phenomenon, there will always be doubts and other new phenomena, ad infinitum.
On the other hand one must be crazy to think that all is said, even the Theory of Evolution of Darwin is no absolute. With all respect I say.
It just struck me, this should be
Just as with God, we worship them at the appropriate time, and forget about them for the rest of the week.
When the scientists want that they are published your works, Bob.
It’s just as well that we editors no longer require human sacrifice. Ah, those were the days…