Scientism: ‘Yippee’ or ‘Boo-sucks’? — Part II

[Editor’s Note: This essay is part of Scientia Salon’s special “scientism week” and could profitably be read alongside other entries on the same topic on this site, such as this one by John Shook and this one by yours truly. My take on the issue is very different from that of the authors who contributed to this special series, and indeed close to that of Putnam and Popper — as it should be clear from a recent presentation I did at a workshop on scientism I organized. Also, contra the author of the third essay in this series (but, interestingly, not the author of the first two!) I think the notion that mathematics is a part of science is fundamentally indefensible. Then again, part of the point of the SciSal project is to offer a forum for a variety of thoughtful perspectives, not just to serve as an echo chamber for my own opinions…]

by Robert Nola

Hayek, Popper and Methodological Scientism

Another Dictionary definition we have seen (B3 in the previous essay) suggests that a kind of scientism arises when the methods of the natural sciences are inappropriately extended to other sciences (especially the social sciences). A broader view of B3 can be taken if it is understood to concern methods developed in one area which are then extended to other areas (or domains). This is the already mentioned ‘methodological scientism.’ Its content is specified by the methods actually extended from one area to another; such extensions may be legitimate and so applauded by their advocates or deemed illegitimate and so condemned by their detractors as scientistic. Such a broad view of B3 is taken by Putnam when he considers the extension of many kinds of formal studies found in logic and semantics to other areas of philosophy. For Putnam this is another kind of scientism: “… as long as we are too much in the grip of formalization we can expect this kind of swinging back and forth between the two kinds of scientism I described [viz., positivism and relativism]” [14].

Now some do deplore the extent to which formal methods have invaded philosophy while others welcome what clarity and understanding they can provide. Some of the heat generated by such extensions can be dissipated if it is viewed as part of the development of a logical and semantic research program (in the sense of Lakatos [15]) which can be applied to many areas of philosophy. Viewed in this way we can discover the program’s successes in its progressive phase and its failures in it degenerative phase. But in making assessments of such a research program we need to have agreed criteria for assessing progress and degeneration in a rational way. Calling the application of formal methods ‘scientistic’ is, at best, merely an expression of an attitude and not a contribution to the research program itself (though in some ways the critic Putnam himself has been an able contributor to aspects of such research programs in philosophy!). But for Putnam this is important because what he thinks is at stake concerning scientism in this area “are attempts to evade the issue of giving a sane and human description of the scope of human reason” [16].

A quite different form of methodological scientism is alleged by Hayek and Popper to infect the human and social sciences. Hayek agrees that there are such things “as the methods of Science in their proper sphere.” But he does not say much about what these methods are though extensions beyond their “proper sphere” are illustrated shortly. This leads to his contrast between science and scientism: “we shall, whenever we are concerned, not with the general spirit of disinterested inquiry but with slavish imitation of the method and language of Science, speak of scientism or the scientistic prejudice” [17].

Hayek locates at least three slavish imitations of the methods of natural science in the social and human sciences. One such imitation can be found in the sort of objectivism underlying behavioristic and physicalist approaches in psychology which eschew all mental phenomena. Another lies in the methodological collectivism which supposes that social wholes exist and obey sui generis laws that social scientists can discover. And a third manifests itself in a historicism which supposes that these laws govern the stages of development of the “wholes,” and thus a scientific history is feasible in which the prediction of future “wholes” becomes possible [18]. The issues raised by these kinds of scientism have been debated extensively in the literature; the task here is not to continue the debate but to flesh out the implicit science-scientism contrast.

Hayek’s attack on historicism resonated with Popper who also made a similar attack at about the same time. In a footnote to later editions of The Open Society and its Enemies Popper says: “If by ‘scientism’ we mean the tendency to ape, in the field of social science, what are supposed to be the methods of the natural sciences, then historicism can be described as a form of scientism” [19]. Popper says ‘supposed’ because he thinks that the methods employed in the natural sciences have been systematically misunderstood by the practitioners and advocates of these sciences themselves and, subsequently, by those who would adopt them in the social sciences. It is part of his substantive account of scientific method to correct this misunderstanding. Popper continues after the above quotation: “But if by ‘scientism’ we should mean the view that the methods of the social sciences are, to a very considerable extent, the same as those of the natural sciences, then I should be obliged to plead ‘guilty’ to being an adherent to ‘scientism.’” So Popper endorses methodological scientism in a descriptive sense; a good many of the methods of the natural sciences can be extended to the social sciences while some cannot. This leads to a research program of working out what are the legitimate and illegitimate extensions of the methods developed in the natural sciences to the social sciences. And back the other way, viz., whether there are methods of the social sciences which can be extended to the natural sciences. (Popper’s encounter with some philosophers of social science, e.g., Adorno, shows that he thinks there are none since there is a methodological unity to all sciences [20].).

Popper’s point that there is a widespread misunderstanding of what are the methods of natural science is taken up in the ‘Preface’ Hayek wrote to a 1967 collection of papers which he dedicated to Popper:

Readers of some of my earlier writings may notice a slight change in the tone of my discussion of the attitude which I then called ‘scientism.’ The reason for this is that Sir Karl Popper has taught me that natural scientists did not really do what most of them not only told us that they did but also urged the representatives of other disciplines to imitate. The difference between the two groups of disciplines has thereby been greatly narrowed and I keep up the argument only because so many social scientists are still trying to imitate what they wrongly believe to be the methods of the natural sciences. The intellectual debt which I owe to this old friend from having taught me this is but one of many [21].

We can conclude from the above that, according to Hayek, his and Popper’s views converge in two ways over the science/scientism distinction, viz., what are the correct methods of science (when they are not misunderstood) and where some methods might be illegitimately applied in some areas. This is illustrative of how some have understood methodological scientism, along with a story about how different dividing lines might be drawn between science and an alleged scientism (in the methodological sense). However, more recent accounts might give a different story about what the methods of science are and draw a different dividing line concerning their application in different sciences (if one is to be drawn at all) [22]. The position adopted by Hayek and Popper provides a novel re-run of aspects of the old Naturwissenschaften–Geisteswissenschaften debate in which advocates of the latter proposed special methods of their own and resisted attempts by advocates of the former to introduce models and concepts of the natural sciences into psychology and social sciences [23].

Science, Scientism, Ways of Knowing and Epistemic Scientism

The dictionary definition (B1a from the previous essay) expresses the epistemic thesis that scientific knowledge is the only kind of knowledge. Some might take a pro-attitude (e.g., Rosenberg) to this while others adopt an anti-attitude (Putnam) calling it ‘scientistic.’  Quine is a little more careful when he says that the methods of science are an extension of those of common sense. Our more sophisticated scientific methods go well beyond those of common sense; but in adding to them they do not undermine them, though they might refine them. Let us adopt the more liberal Quinean view about the methods of science in which much of our everyday knowledge and the methods whereby it is acquired pass muster along with our more sophisticated science and its methods [24].

Opponents of the imperialism of science say that science is not the only way of obtaining knowledge — there are others just as good or better. Let us abbreviate ‘ways of obtaining knowledge’ as ‘WoK’; and allow that the scientific way of getting knowledge [25] is included in WoK along with other ways. Here are three claims that can be made in this area of which the first two could be dubbed ‘scientistic’:

1. In every domain, science is the only way of obtaining knowledge; there are no serious competitors as they fall short in various respects.
2. In each domain there is an associated WoK (which includes science); but there is nothing superior about science — all members of WoK are on a par (relativism).
3. There is at least one domain with its associated WoK such that at least one member of WoK is superior to science.

Which of these claims is correct? And what are the methods that rival science?

Here is a list of rivals to science within WoK (the list can be extended): intuiting; looking to one’s cultural tradition; guessing; hoping; wishing; throwing a die or tossing a coin; praying; consulting a guru; having a séance; looking up an ancient religious text which is regarded as authoritative; having out-of-body or near-death experiences; and the like. In addition, sociologists of knowledge would like to put emphasis on the social conditions which cause belief playing down the role of evidence supposed in science. The definition B1(b) mentions beliefs (or knowledge) arising from religion; the scientistic would reject these since they are amongst the cases ruled out by B1(a), viz., the epistemic scientism they endorse.

While there may be no such thing as the scientific method, there are a number of distinctive methods characteristic of the sciences when it comes to hypothesis testing, such as: clinical trials; the hypothetico-deductive method; various inductive methods; the rival methods of Popper and/or Lakatos; varieties of Bayesianism; Kuhn’s theory of weighted values; and the like [26]. What makes these methods superior to those rivals also listed in WoK? A quick answer (not argued for here) is their reliability in achieving their epistemic goals. Wishing, intuiting, hoping, etc are notoriously unreliable for whatever epistemic goal they might attempt to realize. The rivals to science in WoK need to improve their reliability if they are to count as even remotely acceptable as methods.

So where does this leave us concerning the epistemic scientism expressed in claims like (B1a)? It would appear that the only acceptable items in WoK are the methods of science themselves (understood broadly in Quine’s sense). There may be social, psychological or religious causes of belief; but these are causes only and provide no rational ground for belief of the sort provided by our best principles of scientific method.

Discussions of WoK are often bedeviled by the use of the word ‘knowledge’ in the abstract. But this is not helpful as there are a host of different contexts in which the word ‘know’ is used that require a more refined approach to issues concerning “ways of knowing.” Here are several: (1) ‘know how to …’ where the blank is to be filled by an expression denoting an ability or a capacity such as ‘play the violin,’ ‘speak Chinese,’ ‘integrate an equation,’ and the like. (2) Know … (by acquaintance) where the blank can be filled by expressions such as ‘the way home,’ ‘the Pope,’ and the like. (3) Cognitive knowledge of the form ‘knows that p’ (where p can be any proposition whatever). (4) Explanatory ‘know why …’ where the blank is to be filled by an expression referring to something that is to be explained (e.g., the heart pumps blood). (5) There is also the different explanatory context ‘know how …’ (e.g., the heart pumps blood). Knowing why and knowing how clearly call for different explanations. (6) ‘Know what … is’ where the blank can be filled by expressions such as ‘electrons,’ ‘differentiation’ (in mathematics), etc. And so on for other uses of ‘know.’

There is an important lesson here for how we are to understand WoK. One can well question whether epistemic thesis B1(a) applies in the case of knowing how to … Consider the case of, say, knowing how to play the violin. There may well be a trial and error method when one attempts to develop a certain bowing technique or find the right finger placements on the strings. People might be divided over whether such trial and error methods are scientific. But keep in mind Quine’s view that the methods of science, however sophisticated they become, are an extension of common sense methods including trial and error. Such trial and error is also at work when scientists try to get an experiment to come out right. Consider a different “object” of “knowing” in the case of knowing how to differentiate some equations of motion; this is something to which scientific training does apply.

Again, think of the question when one attempts to know why a Stradivarius violin has the particular tone it does and other violins do not. In order to find the right explanation, and thus acquire some knowledge why, much scientific knowledge of the knowing that p variety has to go into answering questions about the kind of wood used, the varnish, the construction, and the like. So science can have a role here. Finally consider knowing which of two performances of a work one prefers, or knowing which of two violin concertos one prefers (say, the Brahms versus the Tchaikovsky). Here it is not immediately evident that science has anything to say about the proximal causes of such particular preferences. But science could well have a distal role in spelling out how human cognitive evolution got us to respond to and appreciate music in the first place [27].

The trouble with WoK is that its various claims about knowledge are underspecified; once it is made clear what kinds of knowing are involved (which is not done in its various formulations above), then some of the tension about scientism can evaporate. The same can be said about the kind Descriptive Scientism (DS) I described in the previous essay.  Insofar as scientific theories and explanations are to be developed for each domain, then the kind of knowledge that science is meant to produce is mainly cognitive knowing that p and cognitive knowing why p. Though the shaping up of some skills in the case of knowing how to is important in science, knowing how to need not be an essential part of what (DS) claims. Some restrictions were proposed on how (DS) was to be considered; here is another restriction on the appropriate kind of “knowing” produced by science that is to be considered in understanding what (DS) says and how it is to be applied.

Literature, Criticism and Scientism

It might be thought that in evaluating (DS) literature and our response to it along with literary studies and criticism would be outside its scope and thus a counterexample to it. But if one makes clear what aspects of literature are being considered the case is much less clear cut. Here are just four of the many points that could be made about the common ground one can find between literature and science thereby removing some of the heat from the scientism debate and supporting the application of (DS) to some aspects of literature.

1. Determining a text and authorship. There is a time wasting enterprise which claims that the plays of Shakespeare were really authored by someone else (perhaps they were written by another person but with the same name!). Much more interesting is the claim that a number of well-known plays by Shakespeare, when closely investigated, show the hand of a collaborator; he did not write all of the play himself! Then again, plays attributed to some of his contemporaries show the hand of Shakespeare himself. Much of the research in the area of Shakespeare collaboration involves the painstaking noting of patterns of versification, the kind of vocabulary used, various verbal indicators, and the like and then applying the methods of statistics to determine whether there is one author of the play, or two or more collaborators at work. As a result of these investigations the amount of collaboration between Shakespeare and his contemporaries is quite surprising. So are the methods for determining Shakespearean authorship recognizable in science or are they quite different? Reading the works of some of these textual investigators shows that they are not doing anything very different from what can be found in science (in Quine’s sense of methods of science). So here is one domain, that of authorship, that falls happily within the scope of (DS) [28].

2. Interpretive hypotheses. In a fine but perhaps not well known article called “Hermeneutics and the Hypothetico-Deductive Method” [29], Dagfinn Føllesdal argues that the hermeneutic method, commonly supposed to underlie some aspects of literary interpretation, is nothing but the hypothetico-deductive method (HD) as used in science (it is also found in courtrooms and everyday life). Føllesdal illustrates his case by setting out five of the many interpretations literary critics have been given of the role of the Stranger who appears in the fifth act of Ibsen’s play Peer Gynt. The Stranger is variously said to represent anxiety, death, Ibsen himself, the Devil and the ghost of Lord Byron. The task is then to consider just how well these various interpretations of the Stranger, and their consequences, fit the text of the play. Føllesdal does an excellent job of setting this out by investigating the text and other matters as any literary critic would. But the difference is that Føllesdal is aware of the methods of science whereas many other commentators are not. My own suggestion is that the interpretative method he describes is not so much the HD method (which at best deals with one hypothesis at a time) but rather a form of inference to the best explanation, which can deal with several rival hypotheses at once.

Føllesdal also does a good job of showing how various advocates of hermeneutics, such as Habermas, misunderstand what the HD method is. Here we have an example of the point made by Popper and endorsed by Hayek, that there is a systematic misunderstanding of what are the methods of natural science and how they might be carried over to other sciences and even literary studies. So there is much obscuring bush to be cleared away to get a proper understanding of how modes of inference can apply across a wide range of studies, including the humanities, without invoking alleged unique kinds of inference making applicable only to a special domain such as the humanities. It also removes much of the obscurity surrounding talk of “hermeneutics,”

3. Zola’s experimental novel. The 19^th century French novelist Émile Zola was a strong advocate of a form of naturalism in both his novels and more theoretical writings [30]. Zola’s naturalism was strongly influenced by Claude Bernard’s Introduction à la médecine expérimentale (1865) the aim of which was to establish the scientific method in medicine. Similarly, Zola wished to establish something like the experimental method in his novels.

Towards the beginning of his The Experiemental Novel, Zola asks in a manner inspired by his reading of Bernal: “The first question which presents itself is this: Is experiment possible in literature, in which up to the present time observation alone has been employed” [31]. And at the end of chapter 2 he sums up his position, which is best expressed in his own words:

I have reached this point: the experimental novel is a consequence of the scientific evolution of the century; it continues and completes physiology, which itself leans for support on chemistry and medicine; it substitutes for the study of the abstract and the metaphysical man the study of the natural man, governed by physical and chemical laws and modified by the influence of his surroundings; it is in one word the literature of our scientific age, as the classical and romantic literature corresponded to a scholastic and theological age. Now I will pass to the great question of the application of all this, and its justification. [32]

The justification he seeks is the same as that of Bernal’s medical scientist who wishes to understand the human medical condition: “… this dream of the physiologist and the experimental doctor is also that of the novelist, who also employs the experimental method in his study of man as a simple individual and as a social animal. … We are, in a word, experimental moralists …” [33]

Novelists, critics and scholars have proposed many theories about the nature of novels. Zola’s theory of the novel, sketched above, is striking in that it comports closely to a way in which (DS) might be applied to the theory of novels as well as their writing. Though not all novelists and scholars will agree with Zola, he says many things about the kinds of observation novelists make of humans; and following Bernal, he emphasizes their propensity to experiment with “forms of life” in their novels. Much of this many novelists could endorse. Though more needs to be said than these brief comments, the application of (DS) to literature is a project that Zola would endorse. The charge of scientism would be descriptively correct but any pejorative connotations need not be accepted, especially by Zola.

4. Darwinian evolution and story-telling. Darwin’s theory of evolution has been applied not only to the evolution of our bodies (e.g., our possession of opposable thumbs or our ability to walk upright) but also to the evolution of our minds in which various cognitive mechanisms are claimed to have populated our brains in order to enhance our survival. We need not resolve the question as to whether these cognitive mechanisms are direct adaptions in the course of evolution or the by-product of other evolved psychological mechanisms; but they are certainly adaptive. Nearly all humans engage in some form of artistic endeavor (passively or actively), whether it be music, dancing, the visual arts or stories. What is proposed by some theoreticians of evolutionary psychology is that there are specific evolved cognitive mechanisms without which we would simply not engage in the arts. A particular thesis, advanced by Brian Boyd, is that art is a cognitive adaptation derived from play; in fact he defines art as “cognitive play with pattern” [34].

His particular target is story-telling (including the stories of religion) and a consideration of both the evolutionary processes that give rise to story-telling and its adaptive function in our daily lives. This tells us something of the distal causes of our engaging in the general activity of story-telling. But it also allows for the more proximal causes of story-telling through an account of how particular cases of story-telling fit into the overall theory. The cases Boyd considers are Homer’s Odyssey and writings by Dr. Seuss such as Horton hears a Who! How this occurs is a complex story that cannot be summarized here. But it is fairly clear that the application of Darwinian evolutionary psychology leads to the progressive scientization of the domain of story-telling; as such it serves to account for aspects of story-telling, thereby providing yet another positive instance on behalf of the naturalism implicit in (DS). In fact the four aspects of literature briefly set out here serve to show that there is a lot to be said for a partial extension of science into the domain of literature before condemning the extension as ‘scientistic.’

_____

Robert Nola is a professor of philosophy at the University of Auckland in New Zealand. His interests include philosophy of science, metaphysics, epistemology, selected areas in social and historical studies of science, atheism and the relationship between science and religion. With David Braddon-Mitchell he co-edited the volume Conceptual Analysis and Philosophical Naturalism (2008).

[14] Hilary Putnam (1983) “Why there isn’t a ready-made world,” Chapter 12 of Realism and Reason: Philosophical Papers Volume 3 (Cambridge, Cambridge University Press): 199.

[15] On Lakatos.

[16] Putnam, Loc. cit.

[17] Hayek (1942): 269. Or Hayek (1952): 15.

[18] For these varieties of scientism see Hayek (1943) “Scientism and the Study of Society. Part II,” Economica, New Series, Vol. 10, No. 37 (Feb., 1943): 34-63. Or Hayek (1952): 44-79.

[19] Karl Popper (1962), The Open Society and its Enemies: Volume 1 Plato (London Routledge and Kegan Paul, fourth edition) note 4* to chapter 9: 286.

[20] See the Popper-Adorno confrontation in David Frisby (1976) The Positivist Dispute in German Sociology (London, Heineman); Popper’s paper “The Logic of the Social Sciences” is at pp. 87-104 with an addendum at pp. 288-300.

[21] Hayek, F. A. (1967) Studies in Philosophy, Politics and Economics (London, Routledge and Kegan Paul): viii.

[22] Popper gives a different characterization of scientism when he says: “Despite my admiration for scientific knowledge, I am not an adherent of scientism. For scientism dogmatically asserts the authority of scientific knowledge; whereas I do not believe in any authority and have always resisted dogmatism.” Karl Popper (1994) In Search of a Better World (London, Routledge): 6. Popper’s anti-authoritarianism and anti-dogmatism provides a different version of epistemic scientism.

[23] For another account of scientism in historical and social sciences see the five points made by Kitcher (p. 21) and his critique of them which ends with the conclusion that “the five points of scientism rest on stereotypes” (p. 22). Philip Kitcher (2012) “The Trouble with Scientism: Why history and the humanities are also a form of knowledge,” The New Republic, 24th May, 2012.

[24] “… science is itself a continuation of common sense. The scientist is indistinguishable from the common man in his sense of evidence, except that the scientist is more careful.” W. V. Quine, W. O (1976) “The scope and language of science,” in The Ways of Paradox and other Essays (Cambridge, Harvard University Press; revised edition): 233. Note that Quine allows for cognitive incompetence in our attempts to discover knowledge.

[25] Here I do not consider those who would drop out of the debate about ways of getting knowledge and say that scientific methods at best give us a (high-ish) rational degree of belief in our scientific claims.

[26] It should be noted that while there is considerable agreement over what are some of the methods of science, there are still areas of disagreement. The most widely (but not universally) accepted account of method emerges from Bayesianism (though there are disputes amongst Bayesians as to what form this should take). But none of this allows for the ascendency of rivals to science within WoK.

[27] There is a vast amount of research into how cognitive evolution helped produce human responses to art and religion. For a sampler see Steven Mithen, The Prehistory of the Mind (London, Phoenix, 1988) or The Singing Neanderthals: The Origins of Music, Language, Mind and Body (London, Phoenix, 2006).

[28] On Shakespearean collaborations see: James Shapiro (2010) Contested Will: Who Wrote Shakespeare? (London, Faber and Faber): 286-95. For related work on trying to determine a text of the New Testament, see: Bart Ehrman (2005), Misquoting Jesus: The Story behind Who Changed the Bible and Why (New York, Harper Collins).

[29] Dagfinn Føllesdal, “Hermeneutics and the Hypothetico-Deductive Method,” reprinted as chapter 15 of Michael Martin and Lee McIntyre (eds.) Readings in the Philosophy of Social Science (Cambridge, The MIT press, 1994); 233-45.

[30] See Le Roman expérimental (1880; The Experimental Novel) and Les Romanciers naturalistes (1881; The Naturalist Novelists).

[31] Émile Zola (1964) The Experimental Novel and Other Essays, translated by Belle Sherman (New York, Haskell House): chapter 1 p. 6.

[32] Op. cit., 23.

[33] Op. cit., 25.

[34] Brian Boyd (2009) On the Origin of Stories: Evolution, Cognition and Fiction (Cambridge MA, Belnap, Harvard University Press): 15.