The source of the Article:
Science 25 August 2006:
Vol. 313. no. 5790, pp. 1047 - 1048
DOI: 10.1126/science.1131579
Information on the book:
PHILOSOPHY OF SCIENCE:
Science Undermined by Our Limited Imagination?
Tim Lewens*
Exceeding Our Grasp
Science, History, and the Problem of Unconceived Alternatives
by P. Kyle Stanford
Oxford University Press, Oxford, 2006. 248 pp. $45. ISBN 0-19-517408-9.
Science’s Review of the book:
Kyle Stanford’s admirably clear and engaging Exceeding Our Grasp addresses the most basic question in the philosophy of science: Should we believe what scientific theories tell us about the world? Stanford is not asking the trivial question of whether our theories are correct in every detail. Everyone will agree that many of the fine-grained claims in molecular genetics, quantum physics, and biological anthropology, for example, are likely to need substantial revision in the future. The question is instead whether we should think our best theories–in chemistry, physics, biology, and elsewhere–are even close to the truth. So-called scientific realists say yes. Stanford says no.
In defense of this striking claim, Stanford’s book develops what he calls “the problem of unconceived alternatives.” His argument is a close relative of an older philosophical argument known as the “pessimistic induction,” which begins by claiming that the history of science is predominantly a history of failure. Time and again, theories that enjoyed impressive predictive and practical successes, and that were regarded as beyond doubt, have later been rejected as fundamentally mistaken. The argument concludes that the theories we now hold to be true will eventually go the same way.
Stanford (a philosopher of science at the University of California, Irvine) diverges from a simple defense of the pessimistic induction by shifting the argumentative focus from scientific theories to scientists. He tries to show that past scientists have typically failed to consider (let alone evaluate) important alternatives to the theories they have ended up espousing. The central chapters of the book consist of a series of case studies in neglect, all in the domain of 19th-century theories of development and inheritance.
Consider Darwin’s theory of pangenesis, which he defended in his 1868 work The Variation of Animals and Plants Under Domestication (1). Darwin argued that the accepted phenomena of inheritance can be explained on the assumption that each adult organ produces tiny particles, which Darwin called “gemmules,” of a character specific to that organ. Particles produced by different organs collect together in the sex cells and are passed on to offspring, where they direct the formation of new organs resembling those which produced them. Stanford argues that Darwin failed to consider a whole class of alternative explanations for parent-offspring resemblance that attribute similarities between parents and offspring to the production of each organism by a common cause. Something like the common cause view is endorsed by many theorists today; on this view, the offspring’s traits resemble those of its parents not because the character of the parent’s traits causally influences the character of the offspring’s traits, but because the genetic material that originally produced the parent’s traits also produces the offspring’s traits.
In cases like this we can use today’s theories to expose the theoretical blind spots of earlier thinkers. But Stanford’s aim is not to congratulate modern scientists on how much more perceptive they are than their predecessors. He argues that there is no reason to think that we are any less susceptible than Darwin to the pervasive phenomenon of cognitive oversight. According to Stanford, history suggests that modern scientists, too, are currently overlooking alternative theoretical options of a wholly alien sort, which will only be apparent to scientists of the future. This persistent failure of the scientific imagination means that we should expect the truth to lie in the vast space of theories to which we are blind, rather than in the small areas that we are able to survey.
Stanford’s book deserves to be widely read. Its central argument is clearly stated, its conclusion is radical, it engages in a productive fashion with detailed case studies, and it lays down several substantial challenges to scientific realism. Lastly, it is consistently thought-provoking.
One of the thoughts it provokes is that Stanford’s argument may prove far more than he wants it to. At various points, he implies that his antirealism is built on the belief that cognition in Homo sapiens is simply not up to discovering the workings of the world. This belief, in turn, is justified on the basis of the documented history of human scientists’ failures to conceive of important alternative theories. But which specific human weaknesses are supposed to make our species’ scientific claims vulnerable to Stanford’s sceptical argument? Is it a question of our having a maximum IQ that is too low? or imaginative faculties that are somewhat too restrictive?
Another way of asking this question is to consider how, on Stanford’s view, an imaginary species of alien scientist would need to think in order to generate reliable scientific knowledge. Suppose we assume, with Stanford, that the pervasive failure to conceive of important alternative theories truly licenses scepticism about theoretical claims. It seems to follow that the theoretical claims of any imaginable species that acquires concepts by interacting with the world around it are likely to come into the sceptical firing line. For any such species, the formulation of advanced theories is not possible when science begins. Rather, interactions with the world enable the gradual expansion of the repertoire of concepts its members can use to describe the world. For these species, earlier scientists are consequently unable to conceive of theoretical possibilities entertained by later scientists. A species of this kind might be able to shrug off Stanford’s argument if its history of science is one of pathological hedging. Its early theories are logically as weak as can be; hence they need never be rejected, for they are never shown false. But it is hard to see why such a way of doing science is more likely to get to the truth than our own.
So far as I can see, the only other kind of species that might be able to elude Stanford’s argument is a cognitive super-species, which finds out about the workings of the world by exhaustively considering, at the very beginning of its own history of science, every relevant theoretical possibility there is, no matter how complex. It might then settle on just one theory, which would not change much for the rest of that species’ existence. A species like this, which has the miraculous ability to formulate advanced theories of the world prior to interacting with it, would not be vulnerable to Stanford’s induction, for it would have no history of unconceived alternatives.
My worry, then, is that in spite of the detailed case studies of human science Stanford presents in Exceeding Our Grasp, the apparent conclusion to draw from his argument is not that our species has an idiosyncratic set of weaknesses that handicap our abilities to find out about the world. Instead his argument points to a far stronger claim, namely that if a scientific community’s repertoire of concepts expands over time as it interacts with the world, then that community cannot generate theoretical knowledge effectively. I do not know whether Stanford would embrace such a sweeping conclusion. In any case, perhaps my own failure to conceive of how, on Stanford’s view, theoretical scientific knowledge is possible for any non-miraculous species, is merely another confirming instance of the induction he so tenaciously defends.
Reference
- C. Darwin, The Variation of Plants and Animals Under Domestication (John Murray, London, 1868).
10.1126/science.1131579
The reviewer is in the Department of History and Philosophy of Science, University of Cambridge, Free School Lane, Cambridge CB2 3RH, UK. E-mail: tml1000@cam.ac.uk
Your thoughts?
