Creationism: The Theistic Hypothesis as Pseudoscience
One of the more significant social movements of the past few years has been the revival of militant Protestant fundamentalism in the United States. Following the 1925 Scopes trial–a judicial victory but a public relations disaster for Biblical literalists–there came a period of retreat during which fundamentalism continued to thrive as a vigorous sub-culture but made few organized efforts to influence public policy.
However, the 1980’s have witnessed a remarkable resurgence of fundamentalist activism. Rallied by such organizations as the Moral Majority, fundamentalists have lent vociferous support to increased military spending, prayer in public schools, and the demand that “scientific” creationism receive treatment equal to that accorded evolutionary theory in biology courses and textbooks. They have been equally ardent in their opposition to the Equal Rights Amendment (which they played a key role in defeating), gay rights, legalized abortion, sex education, and every other alleged manifestation of what they call “the most dangerous, religion in the world today”–secular humanism. The extent of the social changes entailed by the fundamentalist program and the support that that program has received from some of those in the highest offices of government are sufficient to justify the attention of concerned academics.
Intellectually, fundamentalism has been bankrupt for over a century. Two developments in the intellectual history of the Nineteenth Century served to deprive it of rational credibility. The first was the rise of higher criticism of the Bible in the theological schools and the universities of Germany. The second, and far more famous, challenge to fundamentalism was Darwin’s theory of evolution by natural selection.
Higher criticism showed the Bible to be a thoroughly human document containing many conflicting accounts, divergent traditions, and incompatible interpretations. Though the higher critical arguments are in themselves devastating to the doctrine of Biblical inerrancy, they have never had the public impact of evolutionary theory.
Although evolutionary theory did not achieve complete scientific respectability until the formulation of the evolutionary synthesis in the 1930’s, it had long since served to discredit fundamentalism. By their unyielding adherence to a literal reading of the book of Genesis in the face of mounting evidence for evolution, fundamentalists incurred the ridicule of the likes of H.L. Mencken. As a result of such harshly negative publicity, fundamentalism became a byword for bigotry, rusticity, and superstition.
The recent revival of fundamentalist activism has involved a determined effort to transform the old stereotype into a more sophisticated image. For instance, whereas the creationists of the 1920’s happily repudiated science when it was thought to conflict with scripture, present-day creationists are at pains to emphasize their own scientific credentials. Further, they argue that evolutionary theory is contradicted by the best available scientific evidence and that the only rational alternative is the explanation of origins in terms of a divine creator. In the forum of public debates, these “scientific creationists” (as they refer to themselves) initially enjoyed considerable success in their forensic encounters with evolutionists. However, evolutionists have now counterattacked in a number of books and articles that present cogent defenses of evolutionary theory and detailed rebuttals of creationist arguments.
As noted in the Introduction, this chapter will present “scientific creationism” (hereafter referred to simply, as “creationism”) as a paradigm case of how not to establish theism on a scientific basis. This will be shown by first expounding the main creationist arguments and then showing how, by lapsing into pseudoscience, they forfeit any claim to plausibility. However, before making this claim the notion of pseudoscience must itself be defended since it has been variously argued that it is an illicit concept and so cannot be legitimately employed to outlaw disliked doctrines. Here, the primary task is to argue that none of the “postpositivist” developments in the philosophy of science over the past twenty five years–developments typified by the works of Thomas Kuhn and Paul Feyeraband–rule out either the licitness or desirability of identifying certain doctrines, theories, practices, etc., as pseudoscientific. Further, a rebuttal will be offered to Roger Cooter’s claim that the categorization of certain human endeavors as pseudoscience is merely an instance of ideological oppression designed to serve vested conservative interests. Finally, an adequate set of criteria for the demarcation of science from pseudoscience will be presented.
What then is creationism? Chiefly, it is a comprehensive attack on the theory of evolution, the tacit assumption being that if evolution is eliminated the only alternative is direct creation of the earth and living things by the fiat of a supernatural person. Some of the creationist criticisms of evolution are of a logical, philosophical, or methodological nature. For instance, it is charged that natural selection is a tautologous concept since the “fittest” are those who survive, who in turn are labelled the “fittest.” Further, it is argued–in the tradition of much old-fashioned natural theology–that the first cause of such perfections as intelligence and morality cannot itself have lacked these perfections. It is also contended that evolution is an unfalsifiable dogma since no conceivable observation could serve to refute it (this charge is, of course, flatly inconsistent with the claims that evolution is in fact falsified by the relevant data).
However, the main effort of the creationists is to build a scientific case against evolution. They invoke probability theory to claim that it is infinitely improbable that even the simplest form of life could have arisen by chance from non-living matter. A favourite argument is that the theory of evolution is inconsistent with the second law of thermodynamics since that law holds that systems always increase in entropy whereas evolution sees living systems as increasing in order and complexity.
The creationists get down to details by setting up the two opposing “models”–creation and evolution–and attempting to demonstrate the predictive success of the former and the failure of the latter. They focus especially on the fossil record which they claim reveals precisely the gaps and lack of transitional forms that would be expected if each species were separately created rather than a result of gradual evolution from other forms of life. That is, they charge that since the conjectured “missing links” between groups are in fact missing, the best evidence is that species appeared abruptly with all of their distinctive features fully developed. When faced with any of the standard examples of transitional forms, such as the proto-bird Archaeopteryx, the tactic is to deny its intermediate status and simply assign it to one of the groups in question (this is clearly a fail-safe move since it could be made with respect to any form whatsoever). Other attempts by evolutionists to account for the gaps in the fossil record are dismissed as ad hoc efforts to save the appearances that creationists predict. These and other such arguments are combined with highly selective (and frequently out-of-context) quotes from evolutionary theorists to complete the creationists’ “scientific” case.
Creationists claim that all they want is a fair hearing, i.e., that creationism be presented on an equal footing with evolutionary theory in classrooms and textbooks. By thus making an appeal to fair play, the creationists portray themselves as defenders of intellectual freedom against the purported bias and dogmatism of the scientific and educational establishment. In this way the fundamentalist hopes to shed his obscurantist image and depict himself as the new Galileo struggling to present his case in the face of an oppressive orthodoxy.
As the remainder of this chapter will show, refusal to grant equal time to creationism is entirely justifiable and in no way a manifestation of dogmatism. However, though their charge of bias is entirely unfounded, the creationists do raise an important issue. Those who would proscribe the teaching of creationism in the public schools must bear a considerable onus of justification since those schools are paid for by the taxpayers, a sizable portion of whom sympathize with creationism. Merely showing that creationism is bad science might not do since it could conceivably be argued that citizens have the right to have bad science taught their children if they so desire.
In the United States it would suffice to show that creationism is a religious rather than scientific doctrine since the courts have consistently viewed the inculcation of religious tenets by government institutions as unconstitutional. Thus the question of whether creationism is or is not classifiable as genuine–even if incompetent–science has import extending far beyond the realm of academic debate.
“Pseudoscience” may be easily and somewhat trivially defined as ‘any inquiry, practice, or set of doctrines that lays claim to scientific status while in fact failing to achieve that status.’ Pseudoscience is intellectual fool’s gold; it has the superficial glitter of scientific respectability but its substance is only a worthless amalgram of baser elements. However, to define “pseudoscience” in this way is no more helpful than to define “fake art” as ‘not genuine art.' What is needed are demarcation criteria–standards, values, rules, methods, etc. taken as defining characteristics of genuine science but routinely ignored or repudiated by pseudoscientists.
Only a few decades ago the task of delineating such demarcation criteria would have been thought quite easy. It was believed that there was a unique, universal, and algorithmic “scientific method” that prescribed the necessary and sufficient conditions for all genuine science. This belief naturally led to the aim of formulating a permanent, ahistorical criterion for the demarcation of science from pseudoscience. Now it is generally held, even by those who advocate the maintenance of a clear distinction between science and pseudoscience, that no single criterion will do the job. Mario Bunge puts it this way:
Most philosophers have attempted to characterize science, and correspondingly pseudoscience, by a single feature. Some have chosen consensus as the mark of science, others empirical content, or success, or refutability, or the use of the scientific method, or what have you. Every one of these simplistic attempts has failed. Science is far too complex an object to be characterizable by a single trait–and the same holds for pseudoscience.
Further, the sobering lesson of history is that scientific practice is protean rather than static. That is, what counts as legitimate science can change quite radically from an era to the next. Hence, no one should be surprised if any or all members of any set of demarcation criteria turn out not to have permanent or universal application.
On the other hand, there is no a priori reason to think that the task of spelling out demarcation criteria is either impossible or useless. True, science is now seldom conceived as the incremental accumulation of knowledge through the application of a single, unvarying methodology. However, that a currently more plausible conception of science need not rule out demarcation criteria can be seen if we follow Alasdair Maclntyre, in viewing science primarily as a cultural practice. MacIntyre defines “practice” as:
…any coherent and complex form of socially established cooperative human activity through which goods internal to that form of activity are realised in the course of trying to achieve those standards of excellence which are appropriate to, and partially definitive of, that form of activity, with the result that human powers to achieve excellence, and human conceptions of the ends and goods involved, are systematically extended. Tic-tac-toe is not an example of a practice in this sense, nor is throwing a football with skill; but the game of football is, and so is chess. Bricklaying is not a practice; architecture is. Planting turnips is not a practice; farming is. So are the enquiries of physics, chemistry and biology, and so is the work of the historian, and so are painting and music.
By thus conceiving of science as a socially established practice we can readily account for the historically bounded, protean, and variegated character we now construe it as having. However, the fact that MacIntyre’s practices aim to realize distinctive internal goods and are guided by partially definitive standards of excellence indicates the possibility of basing demarcation criteria on those goods and standards.
It is clearly possible to develop criteria to distinguish some of the practices mentioned by MacIntyre from similar and perhaps rival activities. American and Canadian football are readily distinguished as are chess and checkers and farming and gardening. Of course, there is no reason to think that the distinction between science and pseudoscience will be nearly so easy to draw as in these cases. However, it should be clear that the mere fact that our demarcation criteria cannot be permanent and ahistorical does not preclude their soundness or usefulness.
Nevertheless, the revolutionary developments in the philosophy of science during the past three decades have raised serious questions about how, or whether, science can articulate and apply successful demarcation criteria. The writings of N.R. Hanson, Michael Polanyi, Thomas Kuhn, Stephen Toulmin, Paul Feyerabend and others, usually characterized as “postpositivist” or “postempiricist” philosophy of science, have had the effect of radically challenging the old, established (and, with the public at large, still popular) image of science. That image was the depiction of science as the quintessentially rational enterprise, not only committed to objectivity but possessing the methodology to consistently achieve it.
The effect of postpositivist thought on the image of science has been remarkably like the influence of higher criticism on the image of scripture. Science, like scripture, has been brought down from the heavens and viewed as a thoroughly human product reflecting all the foibles, ideological biases, and ulterior motives of its creators. Scientists themselves have been decanonized and their rules and methods are no longer revered as if they had been inscribed by the finger of God. Even the rationality of science has been challenged in a variety of ways and it is this fact that makes the question of demarcation such a touchy issue.
Attempts to distinguish science from pseudoscience have typically been concerned to show the superior rationality of the former over the latter. Yet postpositivism challenges precisely this traditional conception of scientific rationality.
As detailed by Richard J. Bernstein in his book Beyond Objectivism and Relativism, the postpositivists have argued for the repudiation of logical empiricism and the restructuring of our conceptions of rationality along hermeneutical and pragmatist lines. Since efforts to formulate demarcation criteria have standardly presupposed logical empiricism (or something closely akin), anyone now wishing to propose such criteria must first decide how to deal with the postpositivist view of rationality.
One move would be to accept the postpositivist conception of rationality and then to argue for demarcation criteria based on that new conception. Of course, such criteria will not be claimed to possess the same permanence and universality as when they were believed to derive from ahistorical canons of rationality. It will be seen below that Kuhn (explicitly) and even Feyerabend (implicitly) offer criteria that would single out for criticism many of those pursuits deemed pseudoscientific by old-fashioned rationalists.
Another response would be to reject the postpositivist conception of rationality and to uphold some version of the older ideal as the correct one. From this perspective, the charge that science does not conform to the older standards of rationality will constitute a de facto denial that science is rational. The task of the more traditional rationalist would then be to show that science does in fact conform to the standards of rationality that the postpositivists wish to repudiate. This done, it will only be necessary to reaffirm certain of those standards as providing the legitimate demarcation between science and pseudoscience. Hence, whether the avant garde rationalism of the postpositivists is accepted or whether a more old-fashioned notion of rationality prevails, the possibility of demarcation criteria of some sort need not be ruled out.
Some of the more radical postpositivist charges against orthodox notions of the rationality of science are succinctly, though very unsympathetically, summarized by Israel Scheffler:
The notion of a Fixed observational given of a constant descriptive language, of a shared methodology of investigation, of a rational community advancing its knowledge of the real world–all have been subjected to severe and mounting criticism from a variety of directions… The extreme alternative that threatens is the view that theory is not controlled by data, but that data are manufactured by theory; that rival hypotheses cannot be rationally evaluated, there being no neutral court of appeal nor any shared stock of meanings; that scientific change is a product not of evidential appraisal and logical judgment, but of intuition, persuasion, and conversion; that reality does not constrain the thought of the scientist but is rather itself a projection of that thought.
A thorough examination of the strengths and weaknesses of each of these claims is not required here and would far exceed the scope of this thesis. Here it will be sufficient to examine only the two arguments most likely to be advanced in the name of postpositivism against the effort to produce demarcation criteria.
The first argument arises from the standard interpretation (an interpretation seriously challenged below) of Thomas Kuhn’s view that scientific revolutions (such as the establishment of the Darwinian theory of evolution) consist of “paradigm shifts.” Such shifts constitute changes in perspective so radical that they change even the accepted standards of rationality and hence cannot themselves be judged rational by pre-existing standards.
Given this construal of scientific change, it is possible to argue that those condemned as “pseudoscientists” are in fact only investigators who base their work on a paradigm different from the one currently recognized by establishment scientists. If this alternative paradigm works, i.e., if it can be taken up by a community of researchers, there may be no reason to think that alleged pseudoscientists are any less “rationaI” than their more respected counterparts. Defenders of creationism could claim that there is indeed a thriving and growing community of researchers who base their inquiries on the creationist paradigm.
The second argument is Paul Feyerabend’s charge that the rationalist effort to formulate rules for the justification of theories and hypotheses is hopeless since any non-vacuous formulation would have ruled out most of those discoveries now viewed as cornerstones of scientific progress. Hence science has been and ought to be pursued in the absence of any rules that attempt to demarcate between the rational and the irrational. But if anything goes then so does (to engage in a bit of Feyerabandian invective) what the jaundiced and vulgar rationalist scurrilously dubs “pseudoscience.”
The task of interpreting Kuhn and Feyerabend is not one for the faint of heart. Even sympathetic readers complain that they are inconsistent and unclear in the development of some of their key arguments. In Feyerabend’s case the difficulty is exacerbated by the fact that he is so frequently intentionally outrageous and flippant.
The harshly polemical nature of many of the first responses to Kuhn and Feyerabend indicates that most of their early readers viewed them as blatant irrationalists whose works led downwards into the abyss of relativism and nihilism. Some contemporary writers continue to hold to that assessment, but such an interpretation has recently been vigorously challenged by Bernstein.
Critics of Kuhn have frequently focused on his claims that conflicts between rival paradigms cannot be settled by appeal to neutral observations, the canons of logic, or methodological principles. They view his characterization of scientific change as parallel to religious conversions and “gestalt switches” as reducing the process of theory choice to “mob psychology.”
Bernstein contends that Kuhn does in fact distinguish rational forms of argument from irrational forms of persuasion and that he never denies that the former always has a place in the debates that surround the emergence of new paradigms. Rather, according to Bernstein, Kuhn only wishes to attack the view that there are permanent, ahistorical, and algorithmic criteria that can be applied to provide automatic decisions in situations of theory choice. That is, Kuhn does not wish to deny that there are rational rules and guidelines that ought to be employed; he is only maintaining that individuals may reasonably differ as to the application and relative merits of those rules and guidelines.
Bernstein sees Kuhn as arguing that theory choice is a matter of practical reasoning, Aristotelian phronesis, in which certain values guide deliberation but do not mechanically determine its outcome. When consensus finally emerges it is the product of a dialogical process in which rational discussants sift through their disagreements by applying practically-learned judgmental skills. Like any practical skill, the ability to make such judgments must be learned through experience and not by the rote following of precisely formulated rules. But this fact in no way vitiates the rationality of a procedure that incorporates such judgments.
If Bernstein is right, Kuhn emerges as a defender of the rationality and progressiveness of science. As such, we should expect that he would strongly abjure any interpretation of his work that excuses pseudoscience as simply a research program based on a nonstandard paradigm, as a matter of fact, we do find that Kuhn in one of his later essays endorses five of the standard demarcation criteria: accuracy, consistency, scope, simplicity, and fruitfulness. Though he still denies that these criteria can be applied algorithmically, he maintains that they should be employed in choosing between an established theory and any upstart rival.
However, he maintains that the sharpest contrast between science and pseudoscience is not found within the context of revolutionary science–the tumultuous period when proponents of competing paradigms are struggling for ascendancy. Rather, it is the puzzle-solving capacity of normal science–science conducted during the period of stability when a single paradigm is accepted by all practitioners–that most clearly distinguishes science from pseudoscience. During periods of normal science the practitioners within a field attempt to extend the scope of the ruling paradigm by employing it to solve new problems that arise in the course of inquiry. Genuine science is marked by the success of ruling paradigms in providing solutions to such problems; we know that a field is pseudoscientific if it consistently fails at puzzle-solving.
On the basis of the puzzle-solving criterion, Darwinian evolution is one or the greatest successes in the history of science. The most dramatic illustration of the problem-solving capacity of Darwinian theory was the emergence of the evolutionary synthesis in 1930. It had been thought that Mendelian genetics and Darwinian selection were incompatible until R.A. Fisher showed that, on the contrary, they are mutually indispensable. Indeed, Darwinian theory has proven so successful in explaining the data of paleontology, molecular biology, anatomy, embryology, and other fields that it can be fairly said that “nothing in biology makes sense except in the light of evolution.”
Creationism, on the other hand, is completely unsuccessful in providing the barest outline of a problem-solving strategy. Instead of seeking to explain the mysteries of creation, creationists quickly take refuge in the inscrutable will and unknowable actions of the deity. Further, none of the central creationist claims can be confirmed by repeatable experiment. When creationists do attempt to solve puzzles they only underscore the weakness of their position, as when they invoke the Noachian flood in a strained effort to explain the features of the fossil record. Further, the notion of a benevolent creator brings with it its own set of thorny problems, such as why the purported creation is so fraught with pain, waste, and ugliness. The insolubility of these last difficulties will be the theme of the final chapter of this thesis.
It seems, therefore, that Kuhn’s arguments need not be interpreted as inimical to the effort to distinguish science from pseudoscience or indeed rationality from irrationality. On the contrary, Kuhn himself provides demarcation criteria that unambiguously vindicate evolutionary theory as a genuine and highly successful scientific enterprise while condemning creationism as utterly inadmissible.
However, it can be argued that Kuhn’s rationalism does not go deep enough. Although Kuhn does sanction the application of standards of rationality, he gives no deeper justification for those standards than that they are in fact the ones employed by the scientific community. This exposes him to a version of the Euthyphro dilemma: Are those standards rational because they are the ones sanctioned by the scientific community, or is the scientific community rational because it follows those standards?
In the former case it seems that “rational” refers to whatever scientists take it to apply to. However, if the standards of rationality are thus, in the final analysis, matters of arbitrary decision, why should anyone feel abashed about opting for different standards (unless, like the anxious parvenu, they are terrified of being snubbed by their social betters)? If the second horn of the dilemma is taken, some account needs to be given of why the scientist’s standards of rationality should be considered the most productive ones. But the effort to do this would seem to take us back to the attempt, now under attack from many directions, to formulate a permanent, ahistorical matrix of rationality. The rationalist, if he/she no longer seeks such a matrix, therefore has the formidable task of showing how scientific standards of rationality are themselves rationally justifiable while maintaining that they need no transcendent, ahistorical grounding.
A possible way of mediating between the two extremes of arbitrariness and ahistoricism has been adumbrated in an important article by Robert Feleppa. Feleppa argues that scientific criteria of theory-choice (which, by extension, also serve as demarcation criteria) can and should be justified, but not by attempting some sort of a priori grounding. Justification begins by recognizing that we have in fact succeeded in establishing some rationally acceptable theories. We then attempt to codify those practices that led us to accept those paradigmatically rational theories and to reject others that we still consider unsound. The rules generated by this process of codification can then be compared both with scientific practice itself and with our presystematic (i.e. prior to codifying) intuitions about rationality. This allows for a dialectic that aims to achieve a balance between principles and practice much like the state of “reflective equilibrium” envisioned in Rawls’ A Theory Of Justice. In other words, codification gives us a clearer understanding of those principles that we have in fact already successfully employed. We then can employ those principles to new cases as guides to distinguish good theories from bad, and indeed to rule out certain claims as not scientific at all.
However, these rules are not viewed as Platonic absolutes. It is recognized that they can and sometimes will conflict with scientific practice and our presystematic judgments about rationality. In such a case, if the countervailing reasons are strong enough, the rules will have to be modified. Thus, scientific standards of rationality are grounded and not arbitrary. However, their grounding is not in an ahistorical matrix, but in the historically contingent facts of scientific practice and the prevailing presystematic intuitions about rationality.
Felappa’s proposal presents an attractive via media between groundlessness and ahistoricism. However, as Feleppa admits, the truly hard-bitten foundationalist might regard this proposal as “having all the advantages of theft over honest toil.” Such an unmollified foundationalist might argue that even if the move from practice to principles and back to practice is a virtuous rather then vicious circle, no reason has been given for moving in one circle rather than another. For instance, why not engage in voodoo or witchcraft and derive our regulative principles from those practices and the notions of rationality found within them? Indeed, are not our hopes for justification merely illusory unless scientific practice and its imbedded conceptions of rationality are themselves grounded?
In reply, it should be reiterated that there are many good reasons to believe that this sort of ultimate justification is a will o’ the wisp. However, though foundationalism dies hard, it is not necessary to lay it to rest in this thesis. Feleppa has provided all the tools needed to dispose of pseudoscience, so further inquiry into foundations is irrelevant to our present purposes.
In other words, even if Feleppa has only succeeded in pushing groundlessness one stage further back, he has pushed it far enough to allow the essential failure of pseudoscience to stand out. That failure is simple inconsistency. Pseudoscientists profess admiration for scientific practice and invoke the presystematic intuitions of rationality imbedded in those practices, yet they blithely ignore or unreasonably reject the principles derived from those practices and intuitions. The failure of pseudoscience is therefore not that it pays insufficient homage to some transcendent, ahistorical ideal of rationality, but simply that it is inconsistent. It affirms certain practices and intuitions and then refuses to abide by the rules that derive from them.
Demarcation criteria may therefore be invoked against pseudoscientists since those criteria are the very ones that the pseudoscientists would themselves affirm if only they were consistent. (Of course, nitpickers could argue that the demand for consistency is itself a transcendent, ahistorical ideal. However, it is one such ideal the present author is prepared to stand by.)
Whether the presystematic practices and intuitions from which demarcation criteria derive are themselves more deeply grounded is a different question and, again, one which doesn’t need to be answered here. Hence, even if there are no rational grounds for preferring science over utterly alien practices and intuitions, those who want to dance with scientists without paying the methodological piper can justly be condemned.
Suppose, though, that codification fails to produce any nonvacuous set of theory-choice criteria that have or can be actually employed by scientists. That is, suppose that scientific practice has been and must be a spontaneous, unguided endeavor that can only flourish when completely unfettered by (even historically grounded) rules and methods. In such a case, no examination of scientific practice could yield any codifiable rules and any effort to invent and impose such rules would only succeed in stultifying science.
This, so far as it can be clearly ascertained, is one of Feyerabend’s characteristic claims–a claim perhaps best expressed by Bernstein in his examination of Feyerabend:
Name any rule, algorithm, decision procedure, method, or value that is supposed to guide scientific inquiry, whether it be accuracy, verification, falsification, or some other standard, and I will show you that it has been violated not only in advancing and discovering new hypotheses and theories but in supporting and justifying them. Name anything that you think is or ought to be irrelevant to scientific inquiry, whether it be social context, metaphysical beliefs, or personal idiosyncrasies, and I will show you cases in which it is relevant to a scientist’s investigations.
If Feyerabend is right that the history of scientific practice reveals no codifiable rules or methods, there would appear to be no legitimate distinction between science and pseudoscience (or at least no distinction that cannot be cashed out in terms of the historicist/pragmatist understanding of productivity). If there are no rules, no one can be charged with failing to conform to them. However, buried deep within Feyerabend’s strictures against rules and methods, it is possible to detect a metacriterion that provides the rationale for his attack. What Feyerabend most abhors is intellectual stagnation; what he most admires is creativity, spontaneity, and the proliferation of ideas. It can therefore be expected that Feyerabendian vituperation would be directed against any doctrine which, if widely adopted, would lead to intellectual rigidity and the stifling of innovative impulses. Further, it is hard to imagine any set of doctrines more inflexible, restrictive, and intransigent than those termed pseudoscientific by traditional rationalists. When was the last time an astrologer revised his/her theories in the light of new evidence? Do Velikovskyans welcome scientific criticism? Did parapsychologists applaud illusionist James Randi when he showed how Uri Geller’s purportedly paranormal feats could be duplicated using simple stage magic? Does creationism stimulate puzzle-solving and open new vistas of research?
Although Feyerabend condemns the cavalier dismissal of “pseudoscience” by orthodox scientists, he is quite aware of the unproductiveness of, for example, astrology:
Modern astrology is in many respects similar to early mediaeval astronomy; it inherited interesting and profound ideas, but it distorted them and replaced them by caricatures more adapted to the limited understanding of its practitioners. The caricatures are not used for research; there is no attempt to proceed into new domains and to enlarge our knowledge of extra-terrestrial influences; they simply serve as a reservoir of naive rules and phrases suited to impress the ignorant.
Carl Sagan could not have put it better. Further, though he offers “three cheers” for fundamentalists who opposed a purportedly dogmatic statement of evolution in school textbooks, Feyerabend suspects that if the fundamentalists were in power they would be just as repressive as the scientific establishment.
The upshot is that it is quite plausible to construct a purely Feyerabendian critique of astrology, creationism, et al. These doctrines evince an appalling lack of productivity. That is, they do not lead to useful knowledge nor do they promote fresh ideas or exhibit any tendency to push research into new domains. They encourage unquestioning acceptance on the part of adherents and regard all criticisms with unmitigated hostility. The sterility of such doctrines may be contrasted with the best scientific theories which, in addition to their many practical applications, have been marvelously productive of new insights, creative approaches, and liberating expansions of vision.
Although Feyerabend’s rhetoric makes it tempting to dismiss him as an irrationalist, Bernstein plausibly construes the ultimate aim of all the scorn and invective to be the liberation of scientific creativity from the bonds of rigidity, closure and fixed method. Yet even if he is thus a new sort of rationalist, scientists may justly feel that with friends like Feyerabend they stand in no need of enemies. In fact, the belief that spontaneity and creativity within any field is well served by abolishing all rules and restrictions is a puerile notion at best. Some of the indisputably greatest works of art and literature were produced within rigid stylistic idioms. The creative energies of Petrarch, Shakespeare, and Spenser were not hampered by the formidable restrictions on the sonnet: fourteen lines or lambic pentameter with fixed rhyme scheme treating only a single sentiment or emotion. To see whether Mozart was stifled by the demands of the sonata allegro form one need only listen to the “Jupiter” Symphony.
Of course, stylistic demands can become so excessively rigid that artistic creativity is stifled. When this happens a dessicated traditionalism emerges and art declines until a revolutionary decides to kick over the traces. Nevertheless, it is clear that channelling creative energies, far from making them stagnant, often makes them flow more rapidly. The interrelationships of rule, freedom, and creativity are thus subtle and dialectical; such nuances are not elucidated by Feyerabend’s ham-handed attacks.
Of course Feyerabend might reply that he is fully aware of this dialectic; he is not a serious anarchist, only a playful Dadaist trying to restore the balance obscured by the rationalist obsession with rules and methods. However, that their doctrines actually distort the true picture of that balance is itself a charge that most traditional rationalists would be eager to challenge. They could agree to put the matter to precisely the sort of test that Feyerabend demands. That is, to see whether, as Feyerabend charges, the adoption of theory-choice criteria would retard science, history may be examined to see whether the application of the recommended criteria would have precluded past discoveries. However, to prove that it is unwise to adopt a particular criterion, it is not enough, as Feyerabend assumes it to be, to show that it would on one or more past occasions have led us astray:
Of course, our principles of comparison may on occasion point us in the wrong direction. No rationalist need dispute this truism. If historical evidence is to count against a particular principle of comparison we need more than anecdotes about failures, we need a proof that it has led us wrong more often than not. That would require a massive historical investigation which Feyerabend does not even begin to provide.
Hence, Feyerabend does not appear to have given any grounds for thinking that the creativity of individual scientists or the progress of science as a whole will be in the least vitiated by the adoption of theory-choice (and hence demarcation) criteria.
It may now be confidently asserted that postpositivism poses no insuperable barrier to the endeavour to establish legitimate demarcation criteria. First, the postpositivists themselves offer a number of arguments that are useful in distinguishing science from pseudoscience. Second, those of their views that would endanger or weaken this distinction have been shown faulty. However, postpositivism has certainly shown that no set of demarcation criteria will be as precise, immutable, absolute, or universal as had previously been thought. Such criteria will serve as guides to rationality rather than as algorithms, and hence their application in any given case can only be determined by a judicious and detailed inquiry.
In summary, nothing will be automatically dismissed as pseudoscientific as soon as it runs afoul of a demarcation criterion. Demarcation criteria distinguish between science and nonscience and to say that something is nonscientific is not per se pejorative. However, the term “pseudoscience” is always pejorative since it implies imposture–nonscience masquerading as science. Hence, if the proponents of a suspected theory, doctrine, etc. do not claim to be practitioners of science, i.e. if they do not, in MacIntyre’s terms, affirm the internal goods and standards of excellence characteristic of scientific practice, they cannot be pseudoscientists. However, if they inconsistently claim to be practitioners of science–as the creationists most definitely do–while running afoul the demarcation criteria that derive from scientific practice, then they can only be termed pseudoscientists.
Though debunkers of pseudoscience must proceed more cautiously and make more modest claims than in the heyday of positivism, they need not fear that they will now lack for targets. True, certain disciplines previously stigmatized as pseudoscientific, e.g. the two best known objects of Popper’s censure–Marxism and Psychoanalysis, might now slip through the widened interstices in the net of demarcation criteria. Nevertheless, even thoroughly “postpositivized” criteria are narrow enough to haul in quite a catch of unquestionably pseudoscientific pursuits.
However, the way is not yet clear for the would-be debunker of pseudoscience. Another challenge must be met, this time from the direction of conceptual relativism and Marxist historiography. Roger Cooter charges that the practice of affixing the pejorative label “pseudoscience” to certain fields is both epistemologically unwarranted and socially deleterious. Cooter argues that the characterization of science as rationally superior to pseudoscience cashes out into the claim that the former is value free and objective while the latter is distorted by ideology. However, this view of ideology as something extraneous to science that can infiltrate its objectivity and turn it into pseudoscience is mistaken:
… by “ideology” we mean the partial view of nature and human nature expressed by a group or class which informs perception and conceptualization. It is not meant that ideology must be false nor that the reality of nature is necessarily distorted through ideology. This usage is Marxist-informed but it is non-pejorative and non-epiphenomenal, denoting simply the social, political, metaphysical, and theological or philosophical superstructure that must accompany every economic system. According to this view, science as an intellectual formation and ideology as worldview can never be separate realities or autonomous “things” merely interacting, but must always be mutually constitutive of each other or interpenetrating to form a seamless web.
Hence the scientific worldview, far from being a detached realm of non-ideological truth, is in fact just one of many ideologies. Specifically, it is the ideology that enshrines the view of nature and human nature that arose conjointly with capitalism in the seventeenth century. Hence, the use of the pejorative term “pseudoscience,” even when deployed by left-wing scientists against such targets as I.Q. testing and sociobiology serves to obscure the ideological nature of science. Not only does such labelling wrongly depict science–merely one among many possible worldviews–as the ideology-transcending touchstone of rationality, it reinforces capitalist political and social relations.
Cooter supports these arguments with a thorough study of the rise and fall of phrenology in the nineteenth century. He shows that, by and large, the advocates of phrenology were among the more progressive and even radical elements of the population. By contrast, those quickest to condemn phrenology as “pseudoscience” were establishment scientists whose wealth and status constituted a major investment in the prevailing capitalist order. Interestingly, when some phrenologists did succeed in attaining a degree of social respectability, they lost no time in deploying the label “pseudoscience” against their radical counterparts. All of this suggests that the characterization of phrenology as pseudoscientific, by mystifying the ideological nature of establishment science, tacitly supported the capitalist political and social relations enshrined by that ideology.
Suppose that science is indeed ideological in precisely the way that Cooter charges. Recall that Cooter employs the term “ideology” in a “non-pejorative” and “non-epiphenominal” way that does not imply that ” … ideology must be false nor that the reality of nature is necessarily distorted through ideology.” Now what exactly Cooter means by the “reality of nature” is not clear. In a footnote Cooter indicates belief in a world of non-ideological facts that exist objectively outside of human consciousness. If that world is the “reality of nature,” then surely it makes no sense to say that such a world can “either be or not be distorted by ideology–it simply is as it is. What Cooter must mean is that our understanding of the “reality of nature” (presumably that world of non-ideological facts) is not necessarily distorted by ideology. However, if this is so it would seem that some ideologies could present an at least approximately true (in whatever the appropriate sense of “true” would be) account of that reality. If so, defenders of science need not claim that their worldview is non-ideological, only that its conceptions of nature achieve a higher degree of verisimilitude (however construed) than any opposing ideology.
Perhaps, though, Cooter would want to claim that conflicting ideologies comprise such radically divergent conceptual schemes that there is no possible neutral ground upon which their competing claims can be adjudicated. Without some means of rational comparison no ideology can claim superiority for any of its conceptions. Yet Cooter’s entire critique presupposes that certain of the conflicting claims between his view and establishment science can be settled in his favour. In fact, his whole case collapses unless his understanding of the ideological nature of science is objectively right and science’s self-understanding is concomitantly objectively wrong. Hence, unless his view is the only objective, non-ideological position, Cooter cannot deflect the conclusion of the last paragraph by an appeal to conceptual relativism.
That even deeper difficulties attend the sort of conceptual relativism that Cooter seems to espouse is shown by Donald Davidson in “On the Very Idea of a Conceptual Scheme.” According to Davidson, the notion that there can be different conceptual schemes (Cooter’s ideologies) that differentially perceive and conceptualize a common reality (Cooter’s “reality of nature”) is incoherent. Such a view presupposes that conceptual schemes other than one’s own can be identified (otherwise it makes no sense to talk about different conceptual schemes). However, for two conceptual schemes to differ each must comprise a language that is either wholly or partially untranslatable into the other. But how can anything that is wholly untranslatable be identified as a language at all? On the other hand, if we fail to understand part of another’s language, how can we ever know that our communicative breakdown is not due to presently inadequate but correctable translations rather than in-principle untranslatability?
Concerning the social role played by the deployment of the label “pseudoscience,” it may, for the sake of argument, be admitted that those who deploy that label generally do so in defense of vested interests. However, even if this is so it has nothing to do with the soundness of the arguments advanced in favor of that deployment. For instance, information about the vested interests of debunkers of phrenology is just irrelevant to the question of, say, whether we can determine that someone is a born criminal by feeling the bumps on his head.
Further, if cranial bumps provide no indication of latent criminality, social policies based on the belief that they do are sure to fail. Hence, debunkers of pseudoscience can provide a considerable social service by preventing the squandering of energy, time, and money on projects that are bound to prove fruitless.
Finally, it is simply untrue that the deployment of the label “pseudoscience” always serves to reinforce a conservative ideology: “.. if pseudosciences such as astrology serve as opiates distributing people from a real understanding of themselves and their society, where such an understanding would contribute to desirable social change, then attacking pseudoscience can only be seen as politically progressive.”
The way is now clear to return to the topic of creationism and to develop the demarcation criteria that show it to be a pseudoscience.
When the search is begun for plausible demarcation criteria to invoke against creationism, we discover that we suffer from an embarrassment of riches. Many reasonable criteria are offered in the literature and creationism eases the task by obligingly failing almost every test. We have already seen that creationism fails utterly to meet the Kuhnian criterion of providing a workable puzzle-solving strategy. Further, the Feyerabendian celebration of freshness, creativity, and spontaneity is diametrically opposed by the fundamentalist Bible-as-biology-text approach. However, rather than simply cataloguing creationism’s vices, it will be more illuminating to systematically develop a plausible set of demarcation criteria.
Clearly, any set of demarcation criteria will have to be drawn in terms of certain methodological rules that, by definition, genuine science satisfies and nonscience fails to satisfy. Daniel Rothbart makes a distinction between two types of methodological standards: “(1) those testing standards used during and after the experimental test, such as repeatability of experimental results, versus (2) those testing standards used before any testing to determine which theory should be selected for testing.” The latter pre-testing standards are indispensable since there are, in principle, an infinite number of theories available for any given test and hence there must be some a priori means of eliminating all but the most promising. Within the category of pre-testing requirements, some standards, such as simplicity and concordance with accepted background theories, are not strictly necessary for a priori selection for testing. A given theory may be lacking by such standards yet still possess countervailing virtues sufficient to make it eligible for testing. Other pre-testing standards must be met:
These can be called necessary requirements for a priori selection, as illustrated by the condition that, trivially, each hypothesis must produce test statements purportedly describing some future event. Any hypothesis that fails to yield such statements is immune to experimentation and clearly must not–and in a sense cannot–be selected for testing.
Rothbart claims that these “eligibility requirements,” as he refers to all such necessary pre-testing requirements, serve to provide a basis for a demarcation criterion:
I thus propose that the demarcation between science and non-science is the satisfaction of all eligibility requirements (all necessary requirements) from our methodology for “rational science.” On this view the defining attribute of genuine science is its candidacy as a viable hypothesis for testing or its plausibility to be selected for experimentation–in a word its test-worthiness. Scientific theories must exhibit enough merit and promise to render them serious contenders for a priori selection. Although a scientific theory need not be the best testing candidate, it must be a viable candidate by satisfying all eligibility requirements. Conversely, the defining feature of non-science is its implausibility for testing.
As it stands, Rothbart’s proposal is not yet a demarcation criterion; it is a metacriterion stipulating that the distinction between science and nonscience is that the former meets all eligibility requirements, whatever those turn out to be. We will not have a demarcation criterion until those eligibility requirements are spelled out.
To ascertain these necessary conditions of science, it must be asked what purpose an experimental test is designed to serve. Rothbart contends that the function of testing is to determine the standing of a hypothesis in relation to rival hypotheses rather than to achieve absolute verification or falsification:
Rather than producing some final decision on a hypothesis, by either absolute verification or absolute falsification, the primary function of a test is to contribute to the preferential ranking for the hypothesis in relation to some rival theory. Testing provides empirical grounds for an epistemic ordering of rival hypotheses. Scientists usually perform tests in a context of a competitive struggle between some new hypothesis and some previously accepted background theory.
At this point it might be objected against Rothbart that sometimes new theories and hypotheses are accepted without being tested against older ones. For instance, it is unquestionably true that Newton’s laws of motion and gravitation were accepted long before anyone thought of a way to test them. The great appeal of Newton’s theories lay in their comprehensiveness and simplicity rather than in any experimental triumph over opponents. Thus, there seem to be cases in which such considerations as comprehensiveness and simplicity override the demand for testability.
However, the above example is misleading. Rothbart’s metacriterion is testworthiness, not that the new hypothesis or theory must be immediately capable of testing. In the third chapter it will be argued that an utterly untestworthy hypothesis does not merit acceptance no matter how comprehensive it is. Newton’s laws, though not testable in practice when they were first introduced, are completely testworthy and have in fact been tested many times.
Rothbart suggests that if any newly proffered theory is to be considered scientific it must (at least in principle) be testable vis-a-vis its rival background theory. This in turn suggests that for a new theory to meet this condition it must at least match the explanatory power of the older theory. Unless a new theory accounts for–at the very least–all of the phenomena accounted for by the old theory, it will not constitute an improvement with respect to the evidence. In such a case the new theory will not merit testing against the background theory and hence cannot be considered scientific (in Rothbart’s sense of “scientific”). Thus, Rothbart’s first eligibility requirement is that a new theory may be considered scientific only if it emulates the explanatory power of its rival background theory.
The demand for in-principle testability leads to the further requirement that any candidate theory must be capable of clashing empirically with its rival background theory. That is, the new theory must be of a sort that is able to yield test implications that are inconsistent with the old theory. Clearly, one theory is not testable vis-a-vis another unless some empirical results are relevant to the choice between them. Rothbart’s second eligibility requirement is thus that any candidate for a priori selection for testability must imply (or, at least, be capable of implying) test statements which the relevant background theory would regard as very probably false.
Rothbart’s eligibility requirements, which pari passu serve as demarcation criteria, present a number of attractive features. First, they accord very well with our intuition that science must ultimately, even if at a considerable remove, be testable against experience. This is perhaps the intuition that Popper attempted to capture in his falsifiability criterion. However, unlike Popper’s criterion (or, at least, the way that criterion is frequently interpreted), Rothbart’s eligibility requirements do not attempt to make once-for-all, ahistorical categorizations.
For Rothbart “science” and “nonscience” are relational terms rather than absolute categories. A theory is scientific in relation to another if it at least equals the other’s explanatory success and if it generates test implications incompatible with that other theory. Hence, whether or not a given theory is scientific depends upon which rival background theory is in place when we consider the former’s eligibility for testing. This accounts for the fact that what is scientific at one time can be unscientific at another: Galilean mechanics is scientific with respect to Aristotelian physics but unscientific with respect to the theory of relativity. Hence, Rothbart’s criteria capture the intuitions that made Popper’s criterion so appealing while according much more readily with the protean, revolutionary nature of science.
Further, Rothbart’s development of demarcation criteria closely follows the via media between ahistoricism and groundlessness outlined by Felappe. Rothbert’s eligibility requirements are a priori only in the sense that they are invoked prior to testing, not in the sense that they are deduced from a transcendent matrix. On the contrary, he makes no appeal beyond our presystematic intuitions and the rules latent in established scientific practice. Consequently, his demarcation criteria are neither algorithmic nor absolute: No formula is given for mechanically determining when one theory emulates the explanatory power of another or whether the test implications of one theory are in fact incompatible with another theory. These issues can only be settled by reasoned debate between qualified parties. Further, there seems to be no reason to expect that scientific practice will never change so radically as to undermine the testability metacriterion or perhaps to render the eligibility requirements no longer strictly necessary. In the meantime, however, Rothbart’s criteria do seem to fill the methodological bill.
If we take creationism as the candidate for eligibility assessment and make evolution its rival background theory, it is clear that the former is unscientific vis-a-vis the latter. With respect to explanatory success, the natural world is replete with phenomena that find wholly adequate explanations in terms of evolutionary theory but which creationism cannot even begin to account for. For instance, very many organisms possess useless structures that a creator would have no reason to endow them with but which evolutionary theory explains quite easily as vestiges of past adaptations. Some of the present adaptations of animals (such as Stephen Jay Gould’s favorite example, the Panda’s thumb), though functional, serve their purposes very poorly. Evolution explains these clumsy adaptations as modifications of the only organic structures that happened to be available; creationism leaves us wondering why God did not do a better job of engineering.
Comparative anatomy shows that homologous bone structures are employed by bats for flying and by moles for digging. Evolution explains such homologies in terms of descent from a common ancestor; creationism can only appeal to the inscrutable decision of the creator to employ the same design for radically different purposes. Whale embryos develop teeth and then absorb them before birth. This makes sense if their genes bear the imprint of their evolutionary history, but creationism is again reduced to appealing to the unknowable purposes of the deity.
Perhaps creationists would argue that the above begs the question against them by assuming that the decisions of a divine creator cannot count as legitimate explanations. Yet the scientific rejection of explanation in terms of the volitions of a putative deity is not, as might be charged, an arbitrary rule reflecting a materialistic bias.
First, from a historical point of view, science could never have gotten off the ground if it had rested content with the prevailing supernatural explanations of phenomena. Books such as Bertrand Russell’s Religion and Science may be consulted for a plethora of examples showing that nearly all of the early advances in science involved a struggle against entrenched supernaturalist explanations.
Further, as books such as Russell’s show, we now possess numerous historical examples of naturalistic explanations supplanting supernatural ones and none going the other way (creationism is excluded as a counterexample since it is precisely the case at issue). Hence, we have the basis for an extremely strong inductive argument that runs as follows: In cases C, C+1, C+2, C+3, C+n, where C…C+n are past instances in which supernatural explanations were thought adequate to account for phenomena, each of those supernaturalistic explanations has since been supplanted by a naturalistic one. Further, in no case has a naturalistic explanation been supplanted by a supernaturalistic one. Hence, when we discover a new case, C+(n+1), in which a supernaturalistic explanation is proffered, we have very solid inductive grounds for rejecting that proffered explanation and confidently awaiting a naturalistic account.
However, even deeper problems attend the effort to explain phenomena in ways compatible with science by invoking the will of a supposed deity. Scientific explanation typically proceeds by subsuming a particular explanandum under a general nomological proposition. That is, an individual occurrence is explained scientifically by showing it to be the product of specific initial conditions and the operation of a law-like regularity. Further, it should be possible to employ that regularity to make predictions about future circumstances in which occurrences of that sort will again come about.
By contrast, an appeal to the particular volitions of a deity only succeeds in introducing a scenario, not in subsuming an event under the sort of law-like regularity that can serve as the basis for predictions. In other words, it might be that an event happened the way that it did because God so willed it. But since there are no “laws of supernature” under which to subsume the event, and hence no way of predicting in which future circumstances God is going to will such an event, such an account does not conform to the scientific model. Hence, though the will of God might serve as a legitimate explanation in some circumstances–this need not be denied–it is hard to see how it could count as a scientific explanation.
Finally, the deepest difficulty for the effort to explain in terms of divine volition is that it is notoriously difficult to derive test implications from explanations in terms of God’s will. The gravamen of Antony Flew’s famous essay “Theology and Falsification” is that theists employ such utterances as “it was God’s will” in ways that appear compatible with any conceivable set of circumstances. Clearly, insofar as a theory is compatible with any possible experimental or observational result it cannot generate any test implications. How can there be a crucial experiment between creationism and evolution if no matter how things turn out creationists will claim the results consistent with their view?
This is not to say that in actuality there are no states of affairs that are incompatible with theism. Indeed, the final chapter of this thesis claims that certain evils are in fact incompatible with the theistic hypothesis. Hence, there does not seem to be any reason why God’s will must in principle be regarded as compatible with all possible states of affairs. Nevertheless, theists do exhibit a strong tendency to speak in this way, and insofar as they do they cannot hope to compete scientifically with rival naturalistic hypotheses.
From the above, it is clear that creationism fails to pass either of Rothbart’s eligibility requirements and hence is nonscientific. Further, since creationists lay claim to scientific status, their position is correctly characterized as pseudoscientific rather than merely nonscientific simpliciter. Hence, all of those attempting to erect theism on a scientific basis ought to take heed of creationism as an object lesson showing what traps they must avoid if they wish to keep from lapsing into pseudoscience. If, therefore, anyone else claims scientific status for theism but ignores these lessons, e.g. by not making theism testable against a rival background theory, their efforts will also be pseudoscientific. Whether certain much more sophisticated and subtle proponents of the “theistic hypothesis” in fact succeed any better than their creationist counterparts remains to be seen in the next two chapters.
 For an excellent account of the significance of the evolutionary synthesis see Ernst Mayer, “The Triumph of the Evolutionary Synthesis” in The Times Literary Supplement, (2 November 1984), pp. 1261-1262.
 The two best known pieces of “scientific” creationist literature are H.M. Morris, ed., Scientific Creationism (San Diego: Creation-Life Publishers, 1974) and D.T. Gish, Evolution? The Fossils Say No! (San Diego, Creation-Life Publishers, 1979). The former work comes in both a “general” edition and a “public school” edition. The difference between the two is that the “general” edition contains scriptural references and overtly Christian doctrine while these are omitted from the “public school” edition. All references are to the same page number in each edition.
 These include Douglas J. Futuyma, Science on Trial (New York: Pantheon Books, 1981); Philip Kitcher, Abusing Science (Cambridge, Mass.: MIT Press, 1982); Niles Eldredge, The Monkey Business (New York: Washington Square, 1982); Chris McGowan, In The Beginning… (Toronto: Macmillan of Canada, 1983); Laurie Godfrey, ed., Scientists Confront Creationism (New York: W.W. Norton, 1983); Michael Ruse, Darwinism Defended (Reading, Mass.: Addison-Wesley, 1983), chapts. 13 and 14. These books are all reviewed by Michael Fox in The Queen’s Quarterly, Vol. 91, #3 (Fall, 1984), pp. 755-758.
 A longer version of Cooter’s article appears under the title “Deploying Pseudoscience: Then and Now” in Marsha P. Hanen, Margaret J. Osler, and Robert G. Weyant, eds., Science, Pseudo-Science, and Society (Waterloo, Ontario: Wilfrid Laurier University Press, 1980) pp. 237-272 and a shorter version appears as “The Conservatism of ‘Pseudoscience'” in Patrick Grim, ed., The Philosophy of Science and the Occult (Albany: State University of New York Press, 1982), pp. 130-143.
 Morris, pp. 19-20. The most pungent comment on this line of reasoning comes from Ruse, p. 305. He notes that such a construal of causality entails than since putting manure on rhubarb makes rhubarb tasty and nutritious, manure itself must be tasty and nutritious.
 Morris, pp. 59-69. Futuyma replies, p. 223: “The formation of any particular nucleic acid sequence by chance is very improbable, but the chance of forming one or another visible form is very high. Under conditions resembling those on the prebiotic earth, simple organic molecules actually form from elementary constituents (ammonia, methane, etc.), and assemble themselves into self-replicating nucleic acids which mutate and are altered in frequency by natural selection, all in the laboratory.”
 Morris, pp. 123-125. Futuyma, p. 223: “The second law [of thermodynamics] applies only to closed systems. Organisms which exist in open systems can capture energy and use it to build greater chemical order, and they do so all the time.”
 For an account of this naive view of scientific method and the understanding of science that has come to replace it, see Stephen Toulmin, “The New PhiIosophy of Science and the ‘Paranormal'” in The Skeptical Inquirer, IX, #1 (Fall, 1984), pp. 48-55.
 See Sheffler; also Imre Lakatos, “Falsification and the Methodology of Scientific Research Programmes” in Criticism and the Growth of Knowledge, ed., I. Lakatos and A. Musgrave (Cambridge, England: Cambridge University Press, 1970), pp. 91-195.
 For a recent interpretation that continues to view Kuhn as an irrationalist see W.H. Newton-Smith, The Rationality of Science (Boston: Routledge and Kegan Paul, 1981) pp. 102-124. Bernstein challenges this interpretation in pp. 52-61.
 For a succinct account of the historical development of Darwinian evolution see Antony Flew, Darwinian Evolution (London: Granada Publishing LTD, 1984); for the reference to Fisher and the evolutionary synthesis see p. 23 of this work.
 The most famous attack on this sort of foundationalism is Richard Rorty’s Philosophy and the Mirror of Nature (Princeton: Princeton University Press, 1979). All of the most powerful objections to the foundationalist effort to ground knowledge in a transcendent, ahistorical matrix are presented in Bernstein.
 This is in no way a denial that science can rationally claim superiority in certain respects to, e.g., witchcraft. In fact, I would argue that insofar as science and witchcraft make clashing truth-claims, such as those relating to the etiology of diseases, the former can definitely be shown superior to the latter. For instance, the witch-doctor’s explanation of a case of sleeping-sickness as the result of a hex can be justifiably deemed inferior to an explanation in terms of an excessive number of trypanosomes in the bloodstream.
 See “How to Defend Society Against Science” in Introductory Readings in the Philosophy of Science, E.O. Klemke, Robert Hollinger, and A. David Kline, eds. (Buffalo, N.Y.: Prometheus Books, 1980), p. 62.
 Then again, they might not. For a cogent critique of the claimed scientific credentials of psychoanalysis see Adolf Grünbaum, “The Role of Psychological Explanations of the Rejection or Acceptance of Scientific Theories” in Hanen, Osler, and Wayant, pp. 29-53.
 A comprehensive set of demarcation criteria are developed by Daisie and Michael Radner in their book Science and Unreason (Belmont, California: Wadsworth Publishing Company, 1982). The Radners’ demarcation criteria are in no way incompatible with the ones developed below. However, the two criteria offered in this chapter are adequate for the demarcation of science from nonscience and have the advantage of being developed more systematically than the Radner’s criteria.
 It is sometimes charged that evolutionary theory does not fit this model since it makes no predictions about future evolutions. True, evolutionary theory cannot predict which sorts of species will emerge from present ones; there are simply too many imponderables at work in the evolutionary process. However, advances in a number of fields have confirmed predictions that would be expected on the basis of evolutionary theory. For instance, molecular biology has recently shown that species believed related on the basis of evolutionary theory evince previously unknown biochemical similarities (see Futuyma, p. 55).