Return to Article Details The realism issue from a Deweyan perspective

1. Introduction

As we should not expect John Dewey, or a contemporary Deweyan pragmatist, to offer any global, overall account of the perplexing controversy over realism, I will approach my topic through a number of specific cases – or, in Deweyan terms, specific “problematic situations” in which this issue is manifested.1 My aim is to show that there is a problem of realism in Deweyan pragmatism (pace David Hildebrand’s important 2003 book, to which I will return in some detail below), although this problem cannot be reduced to any of its standard formulations, either in pre-pragmatist traditional philosophy or in contemporary literature on realism and antirealism heavily influenced by the rise of neopragmatism. I will briefly study the following triad of pragmatically relevant problems of realism, contrasting realist views to a number of different non- or antirealist alternatives, such as idealism, empiricism, and constructivism:

I will begin with the realism issue in the philosophy of science. It will naturally lead us to the other relevant aspects of this problem.

2. Dewey and scientific realism

Dewey’s pragmatic naturalism, as articulated in his major works such as Experience and Nature (1989 [1929]), The Quest for Certainty (1960 [1929]), Logic: The Theory of Inquiry (1938), and Reconstruction in Philosophy (1957 [1948]), a view variously labeled “instrumentalism”, “experimentalism”, or “operational thinking”, accommodates an intriguing tension between various standard realisms and antirealisms. Dewey’s position has clearly instrumentalist and more generally empiricist elements, roughly in the sense in which these terms are used in the philosophy of science, although his talk about instrumentalism should not be simply equated with later philosophers’ of science narrower treatment of the empiricist doctrine carrying the same title. Dewey is not innocent regarding the subsequent association, by Karl Popper and others, of pragmatism and instrumentalism (in the narrower sense). When, for instance, Percy Bridgman (1960 [1927], 5, 31-32) speaks about “operational thinking” and suggests, notoriously, that the meaning of theoretical concepts is to be defined with reference to practical operations in the laboratory, one cannot help noticing a similarity to Dewey’s insistence – roughly at the same time – on practice, operationality, and the instrumental use of theories. However, the similarities between Bridgman’s and Dewey’s views must not be exaggerated, as significant differences remain. For example, when acknowledging C.S. Peirce as a precursor of his operational thinking, Dewey (1936) endorsed, in a qualified manner, a Peircean view of universals (or “generals”), yet avoiding hypostatizing them as real entities.3 For an operationalist like Bridgman, even such a qualified acknowledgment of something like the medieval universals would hardly be acceptable.

Dewey (1960 [1929], 79) celebrates the “overtly executed operations of interaction” that are needed for obtaining scientific knowledge, thereby rejecting the gap traditionally thought to lie between knowledge and action. This seems to lead to a form of antirealism:

There is something both ridiculous and disconcerting in the way in which men have let themselves be imposed upon, so as to infer that scientific ways of thinking of objects give the inner reality of things, and that they put a mark of spuriousness upon all other ways of thinking of them, and of perceiving and enjoying them. It is ludicrous because these scientific conceptions, like other instruments, are hand-made by man in pursuit of realization of a certain interest – that of the maximum convertibility of every object of thought into any and every other. […] [W]hen the physical sciences describe objects and the world as being such and such, it is thought that the description is of reality as it exists in itself. [… However, the] business of thought is not to conform to or reproduce the characters already possessed by objects but to judge them as potentialities of what they become through an indicated operation. […] [T]o think of the world in terms of mathematical formulae of space, time and motion is not to have a picture of the independent and fixed essence of the universe. It is to describe experienceable objects as material upon which certain operations are performed. (135-137)

Science, accordingly, has no more privileged relation to “the real” than (some) other human practices. “There are as many conceptions of knowledge as there are distinctive operations by which problematic situations are resolved”, Dewey says (221). In brief, “scientific conceptions” are not “revelations of antecedent properties of real Being and existence” but “instrumentalities which direct operations of experimental observations” (192). Natural laws, similarly, are “intellectual instrumentalities”, “formulae for the prediction of the probability of an observable occurrence,” instead of being statements about “ultimate and rigid uniformities of being” (205-206; original emphasis). A law that was supposed to “govern phenomena” ought to be understood as “a way of transacting business effectively with concrete existences” (ibid., p. 207). The universality typically claimed to characterize laws and theories “is not that of inherent content fixed by God or Nature, but of range of applicability” (Dewey 1957 [1948], xv).4 Scientific concept(ion)s and theories are tools, “open to development through use” (145).

Similarly, in his naturalist magnum opus, Experience and Nature, Dewey (1989 [1929], 115) equates “the proper objects of science” with “nature in its instrumental characters”. The key move for a pragmatic naturalist is to stop treating the object of science as “complete and self-sufficient”; this will only result in an insoluble problem, viz., the question concerning the relation between the “perceptual order” and the order of “inferred and logically constructed real objects”, with two incompatible kinds of knowledge and of the objects of knowledge (116). It seems, indeed, that mainstream philosophers of science struggled with such a problem throughout the twentieth century. Dewey seems to avoid the problem of scientific realism altogether. In this context, he also reminds his potential misunderstanders that his “instrumentalism” is not a theory about personal satisfaction in knowing, but about “the proper objects of science” (126).5 Precisely as such, it might, however, be regarded as straightforwardly antirealistic.

Dewey’s commitment to instrumentalism is relatively obvious when he argues that the problem of the “two tables” (the so-called “Eddington tables”) is illusory, because the table we perceive and use is “the only table”, “for it alone has both individuality of form […] and also includes within itself a continuum of relations or interactions brought to a focus” (Dewey 1960 [1929], 240). Thus, there is hardly any significant problem of reality (or realism) in Dewey. Reality is simply what is given – or, better, “taken” – in experience. Here Dewey disagreed with the so-called “new realists” of his time, as well documented in his numerous exchanges with his critics (see Morgenbesser 1977). In The Quest for Certainty, he writes: “The world as we experience it is a real world. But it is not in its primary phases a world that is known, a world that is understood, and is intellectually coherent and secure. Knowing consists of operations that give experienced objects a form in which the relations, upon which the onward course of events depends, are securely experienced. It marks a transitional redirection and rearrangement of the real. It is intermediate and instrumental […].” (Dewey 1960 [1929], 295.)6

Is Dewey saying, then, that the scientific table, or the “scientific image” according to which the table is not, in the final analysis, a concrete, perceived, practically used object, after all, but a collection of microphysical particles, is unreal or illusory? If so, he is subscribing to an instrumentalist position, which several philosophers of science, especially following Wilfrid Sellars (1963), have powerfully called into question (cf. Tuomela 1985, Niiniluoto 1999). Or is he, rather, subscribing to idealism, or to what would today more accurately be called constructivism, when he maintains that scientific objects are not independent of inquiry but pragmatic constructions arising out of an intelligent use of the methods of inquiry (cf. also Shook 2000)?

In an earlier work, Dewey (1916, 30) admits that he may sound like an idealist when he holds that thinking, which “is instrumental to a control of environment,”7 in a way constructs its objects: “[P]rocesses of reflective inquiry play a part in shaping the objects – namely, terms and propositions – which constitute the bodies of scientific knowledge. […] Insofar as it is idealistic to hold that objects of knowledge in their capacity of distinctive objects of knowledge are determined by intelligence, [this view] is idealistic. It believes that faith in the constructive, the creative, competency of intelligence was the redeeming element in historic idealisms.” (30-31; see also 60.) Dewey’s Hegelian background is visible here, though he immediately adds that his instrumentalism (or idealism) does not postulate an entity or substance which constitutes the world but “defines thought or intelligence by function, by work done, by consequences effected” (31). Moreover, the ultimate justification of scientific theorization, for Dewey, seems to be humanistic, not narrowly instrumental:

Natural science loses its divorce from humanity; it becomes itself humanistic in quality. It is something to be pursued not in a technical and specialized way for what is called truth for its own sake, but with the sense of its social bearing, its intellectual indispensableness. It is technical only in the sense that it provides the technique of social and moral engineering. (Dewey 1957 [1948], 173)

The charges of instrumentalism and idealism have their obvious justification, but it would be overhasty to judge Dewey to be simply an antirealist in his philosophy of science. For one thing, he – just like James – uses the terminology of “objects”, “concepts”, “conceptions”, “theories”, “hypotheses”, etc., more loosely than is customary in more recent analytic philosophy of science; I have quoted extensively from his writings partly in order to highlight this fact. This habit of usage makes confusions easy. For another thing, Dewey was, perhaps primarily, a naturalist, always arguing that experience and knowledge, including the production of scientific knowledge (and its objects), are natural phenomena in a natural world in which we try to cope, to settle the problematic situations we arrive at. Whatever is natural to our practices of inquiry, our habits of settling indeterminate situations by critically using the intelligence that itself naturally arises as a human capacity of reacting to problematic situations, is to be accepted rather than treated with philosophical suspicion. A philosophical skepticism about, say, the existence of unobservable theoretical entities referred to in our most advanced and most successful scientific theories (and/or practices) would be an utterly unpragmatist and unnatural attitude from Dewey’s point of view. Rather, we should take seriously the natural practices of inquiry we engage in. It was always the actual practice of inquiry itself that Dewey appealed to when, for instance, attacking the ancient ideal of certainty and the “spectator theory of knowledge” (Dewey 1960 [1929]). It was science as it is actually conducted, rather than any abstract philosophical account of science, that he admired when we wrote: “In science the order of fixities has already passed irretrievably into an order of connections in process” (Dewey 1957 [1948], xl).

My preliminary conclusion at this point is that the Deweyan pragmatist is definitely not a scientific realist, if scientific realism is defined as the thesis that scientific theories provide us with the only true (or truthlike) picture of reality, or that the “scientific image” offered by theorization is ontologically prior to the “manifest image” we are more directly acquainted with in our ordinary experience.8 But a Deweyan may be, and Dewey himself may also be regarded as having been, a scientific realist in the sense of rejecting unnatural skeptical doubts about the existence of theoretical entities postulated in the actual course of scientific theorization, as long as such theorization is firmly rooted in its humanly natural practical background, that is, in the natural processes of settling problematic situations that arise in the course of our experience (cf. French 1989). Of course, all pragmatists should admit that the specific postulations of unobservable entities in (current) scientific theories may turn out to be ill-founded. But, as fallibilists, we should adopt such a critical, open attitude to any human claims and ideas whatsoever. The key pragmatist move is to liberate not only science but also scientific realism – and all other philosophical interpretations of science we might practically need – from foundationalist pursuits of certainty, essences, and other remnants of “first philosophy”. Deweyan pragmatism offers, then, a middle path (or, rather, several such paths) between scientific realism and instrumentalist antirealism.

3. Naturalism, anti-reductionism, and pragmatic realism about “emergents”

Dewey’s evolutionary naturalism has sometimes been interpreted as an “emergent theory of mind” (see Tiles 1990 [1988], ch. 3; cf. Savery 1951, Alexander 1992), possibly classifiable as “diachronic emergentism” in Achim Stephan’s (1998, 1999) sense. Though Dewey seldom employed the term, he may be regarded as an emergentist thinker. Indeed, what else could a philosopher constantly emphasizing growth, evolution, and processuality be? What gradually – fully naturally – emerges out of the inanimate world in Dewey’s evolutionary scheme is life, mind, and freedom, hence what we may simply call human culture. For Dewey, these emergent constructions or properties are “real features of […] complex systems which cannot be accounted for in terms that would be adequate if the same constituents were organized in a less complex way” (Tiles 1990 [1988], 148). More generally, the pragmatist tradition as a whole is characterized by the frequent use of notions such as creativity, freedom, evolution, and novelty, which quite naturally find a place in (diachronic) emergentism, too. Insofar as we loosely define emergent properties as “holistic” system properties that a whole or a system possesses but that cannot be deduced from – or perhaps even explained with reference to – whatever could be known about the constituents of the system (or about the behavior of such constituents in systems simpler than the given one), we may say that most pragmatist treatments of mentality, experience, freedom, or inquiry are emergentist in spirit, if not in letter.9

Dewey’s views enable us to put the recently hot controversy over emergence into a pragmatic context. Pragmatic realism, especially in the Deweyan form briefly characterized in the previous section, is an inherently pluralistic and antireductionist position. This antireductionism applies not only to the properties that may be regarded as emergent but also to the concept of emergence itself, which possess no “essence.” For some purposes it might be better – more pragmatically useful – to adopt an ontological scheme in which there are emergent properties; for some other purposes it might be better to adopt another scheme in which emergents do not exist irreducibly. Physicists, for instance, may pragmatically “need” a reductionist scheme treating all entities and processes as ultimately physical. But physicists’ needs and purposes are not all the human needs and purposes there are. Indeed, the concept of emergence itself arises from the need to take seriously various different “levels” of human experience and the world experienced: biological, psychological, and socio-cultural ones, in addition to the physical “bottom level.”

Moreover, different notions of causation – especially “downward causation” needed for emergence (see again Stephan 1999) – may be acceptable and differently pragmatically functional in different schemes. One may, the Deweyan pragmatist might suggest, give up the futile search for the correct concept of emergence (which is typical of current discussion of emergence) and define a variety of different notions of emergence; the exciting issues that remain to be settled are, in any case, pragmatic, having to do with how well those notions “work” for the scientific or philosophical purposes they are designed to meet. When confronting a definition of emergence (or anything else), we should pragmatically ask what kind of philosophical work can be done with the concept, and how the proposed definition helps us in understanding, reformulating and (possibly) solving (or dissolving) our age-old philosophical dilemmas, such as the mind-body problem or downward (mental) causation. It is partly a terminological issue what kind of properties or structures are called “emergent”. But terminological or conceptual issues are certainly not unimportant in philosophy, as philosophical problems and views are largely constituted by the traditions within which they are conceptualized, or spoken and argued about. Pragmatists encourage us to turn our attention to such problems, in order to find out what we are actually claiming when we claim, say, that certain properties are emergent.

Deweyan pragmatism might significantly help us in developing a relaxed, pluralist, and context-sensitive attitude to the notion of emergence. Perhaps a notion of emergence based on nonreductive physicalism, as discussed in recent emergence literature (e.g., Stephan 1999), does some interesting, pragmatically valuable work in certain specific fields, e.g., in the philosophy of biology – possibly in accounting for the relation between biological and physico-chemical properties. But when we move on to other ontological regions, particularly the mental and cultural realms, we do not seem to have a sufficiently clear idea of how the program of nonreductive physicalism could be carried through (with or without emergence).10 A stronger notion of emergence than the weak ones grounded in physicalism is needed for an adequate account of our self-image as intentional, consciously and purposively acting, free and responsible agents – for an account we may want to give of ourselves as human beings, irreducible to physics or biology. We might, then, have use for at least two different concepts of emergence: a relatively weak one to be employed within the factual realm in which humans are undeniably parts of law-governed physical and biological nature, and a stronger one to account for the qualitative difference between this factual level of investigation and the normative one that is our “second nature.”11 The applicability of such concepts of emergence would of course have to be assessed in more detail.

Let me add a few more remarks on pragmatism’s relation to emergentism. As manifested in the interpretations of Dewey’s naturalism as an emergent theory of mind, the evolutionary character of pragmatist thought is obvious, and this feature of pragmatism has been noted in connection with emergentism (see, e.g., Patrick 1922, 701; Goudge 1973; cf. Tully 1981). Yet, little explicit scholarly work has been done on Dewey’s or other pragmatists’ actual conceptions of emergence or on their relations to major emergentist thinkers, although, as Thomas Goudge (1973, 133) notes, the pragmatists were “the first group of philosophers to work out in detail a philosophy of mind based on evolutionary principles.”12 A number of insights into the issue of emergence, and the further issue of realism as applied to emergent properties and structures, can be drawn from both Dewey’s and G.H. Mead’s works. Dewey (1960 [1929], 214-215) writes: “The intellectual activity of man is not something brought to bear upon nature from without; it is nature realizing its own potentialities in behalf of a fuller and richer issue of events.” What we have here is an ordinary and commonsensical rather than a technical notion. In an early study on Dewey, William Savery (1951, 498) described his view by saying that “our perspectives [on nature and experience] are emergent natural events” which “have a continuous flow.”

Although Dewey was a naturalist, he rejected metaphysical realism, as we already saw in the previous section discussing his attitude to the issue of scientific realism. For him, the ontological structure of reality – whether postulated by science or by ordinary experience – was a humanly established structure, itself emerging in the course of human experience and inquiry, rather than being already “there” in an absolute sense, as the metaphysical realist would have it. In this sense his realism of emergents, too, was thoroughly pragmatic. Dewey was not entirely happy with the term “emergence”, however. Late in his life, jointly with Arthur Bentley, he argued that the “natural man” who talks, thinks and knows should not, “even in his latest and most complex activities”, be surveyed “as magically ‘emergent’ into something new and strange” (Dewey & Bentley 1989 [1949], 45). He contrasted his “transactional” view of emergence to previous views thus:

At a stage at which an inquirer wants to keep ‘life,’ let us say, within ‘nature,’ at the same time not ‘degrading’ it to what he fears some other workers may think of ‘nature’ – or perhaps similarly, if he wants to treat ‘mind’ within organic life – he may say that life or mind ‘emerges,’ calling it thereby ‘natural’ in origin, yet still holding that it is all that it was held to be in its earlier ‘non-natural’ envisionment. The transactional view of emergence, in contrast, will not expect merely to report the advent out of the womb of nature of something that still retains an old non-natural independence and isolation. It will be positively interested in fresh direct study in the new form. It will seek enriched descriptions of primary life processes in their environments and of the more complex behavioral processes in theirs. (121)

Thus, while being careful with the word “emergence”, Dewey did not reject the idea altogether but only what he saw as its magical overtones. He simply called for advancing scientific research on the emergence of life and the mind. A pragmatic scientific realism about emergent properties in the sense of the characterization given in section 2 above is possible from a Deweyan perspective. Mead’s pragmatism, influenced by Dewey’s, includes a notion of emergence as something “social”, as well as an analysis of sociality as a character of emergent evolution. In his prefatory remarks to the posthumous publication of Mead’s The Philosophy of the Present, Dewey insisted that Mead took the doctrine of emergence “much more fundamentally” than “most of those who have played with the idea” (Dewey 1932, xxxviii). According to Dewey, Mead “felt within himself both the emergence of the new and the inevitable continuity of the new with the old” (xxxix).13

According to Goudge’s (1973, 142) in my view plausible interpretation, both Dewey and Mead employed the notion of emergence in order to resist “the classical thesis that (1) since mental phenomena now exist, they must have been implicitly or potentially present in evolution from the very start; and (2) their potential presence played an active part in their later realization, and was not merely an abstract possibility”. But the evolutionary emergence of mentality is not a sudden leap; it is “prolonged, successive” (142) – to the extent that one finds in these pragmatists’ philosophy of mind “a conceptual tension between the category of emergence and that of continuity” (144). Goudge is probably right when he says that Peirce gave more priority to continuity in his metaphysics (just think about Peirce’s famous metaphysics of synechism),14 while Mead emphasized emergence and Dewey remained somewhere in between.

Continuity, along with emergence, is thus a major theme in pragmatism; one important task in Deweyan naturalism is to find a middle ground between the doctrine of continuity (Peircean synechism) and a realist view of emergent layers of reality. This issue is related to more general ones about whether metaphysics – about continuity, about emergence, or about anything else – is so much as possible in pragmatism. I cannot settle such issues here. Neither Dewey nor Mead neither carefully analyzes the notion of emergence, nor argues for the reality of emergents, in any straightforwardly pragmatic manner. We cannot here determine the correct interpretation of either Dewey’s or Mead’s emergentism, but their antireductionist naturalisms are so closely related to typical emergentist ideas that they should be taken into account as some of the most creative, yet also critical, representatives of this peculiar philosophical orientation.15

Dewey’s evolutionary, emergent, pragmatic naturalism is not restricted to the philosophy of mind or even to the theory of ethical and aesthetic values.16 In Dewey’s (1991 [1934]) view, religious values can be “inherent in natural experience” (28) – and thus also “emergent” in a sense. As he writes: “Any activity pursued in behalf of an ideal end against obstacles and in spite of threats of personal loss because of conviction of its general and enduring value is religious in quality” (27). Hence, we should distinguish between religions and “the religious” (or, specifically, the religious aspect(s) of experience). The latter expression “denotes attitudes that may be taken toward every object and every proposed end or ideal” (10). Such attitudes may emerge in the context of natural human life.

The concept of the religious, then, ought to be liberated from the supernatural commitments of actual historical religions, from dogmas and doctrines that are pragmatically unnecessary (44). Indeed, supernaturalism is an obstacle in pursuing the kind of natural changes that are in our power to bring about (80). The values and ideals that belong to the religious attitude and emerge from natural human practices are not imaginary but real; they are “made out of the hard stuff of the world of physical and social experience” (49). The religious is, through this rearticulative maneuver, rendered part of nature itself – and nature, for Dewey, is of course all-encompassing, everything there is, yielding innumerable emergent features and constructs, including even “the religious” which traditional religions misconstrue by placing it outside nature altogether, into the “supernatural” realm.17

We may speculate that Dewey, had he been happier with the term, could even have embraced an emergentist philosophy of religion, with religious value qualities emerging through the natural, worldly experiences of human beings within their purposive practices, facing their problematic situations. Neither religious or other “higher” aspects of human cultural practices can be reduced to mere psychological, let alone physical, regularities. A pragmatic realist view of such aspects ought to be developed, and the notion of emergence, pragmatically and pluralistically articulated, may be helpful in this task.

Finally, one might wonder what the discussion of Dewey as a nonreductive naturalist and emergentist in the end has to do with the realism issue, the basic problem of this paper. The connection between emergence and realism has remained relatively implicit in this section. This connection is nevertheless tight. The reality of emergent structures or properties is what is ultimately at issue here. This problem can be seen as the problem of realism as applied to whatever emergent features might be postulated. I have been suggesting a pragmatist re-examination of this fundamental problem. It is part of my suggestion that we should not let unpragmatic metaphysical realism dominate the emergence debate in the philosophy of mind, in the philosophy of science, or in general metaphysics (see Pihlström 2002). Rather, we should be prepared to argue that any entities, structures, or processes we may regard as real emerge out of purposive human practices (of inquiry) aiming at the resolution of problematic situations. Thus, the concept of emergence, far from being just a minor issue in the philosophy of mind, may in fact help us in offering a plausible Deweyan compromise between realism and constructivism, as outlined in section 2. If we already had a pragmatic theory of emergence at our disposal, and if we had already developed a viable pragmatic realism about “the emergents”, then we might be able to construe the dependence of scientific objects upon practices of inquiry as the “emergence” of those objects out of inquiry. The construction work for such a pragmatic realism about both emergence and the objects of inquiry has, however, been barely begun here.

4. Naturalism and realism in the neopragmatist reception of Dewey

As we know, Dewey’s pragmatism has undergone a revival in the philosophy of the past few decades. Richard Rorty and Hilary Putnam are undoubtedly the two most influential neopragmatists writing today. Their criticisms of traditional (metaphysical) realism and especially their mutual debate on the significance of pragmatism as a philosophical movement of our times, and on Dewey in particular, have received continuous attention.18 Only seldom have Rorty’s and Putnam’s interpretations of the classical pragmatists themselves been subjected to detailed critical scrutiny, however. True, in Rorty’s case, there is a growing amount of literature available addressing his (highly problematic) ways of employing the classical pragmatists’ ideas, but in Putnam’s case, such historically informed discussions have been relatively rare. Thus, David Hildebrand did, in his book on Dewey and the neopragmatists (Hildebrand 2003), a great service to the philosophical community by taking up a key issue in these two neopragmatists’ thought: their understanding and appropriation of Dewey’s pragmatism, especially in relation to the issue of realism, with which we have already been concerned above.19

Hildebrand’s volume is a careful study on Dewey (in his own historical context) and an equally careful assessment of Rorty’s and Putnam’s versions of pragmatism in relation to Dewey’s. Hildebrand seeks to show that Dewey’s pragmatism was and remains superior to these two philosophers’ neopragmatisms and that, in particular, it can lead us out of the realism/antirealism controversy that both Rorty and Putnam have unsuccessfully tried to avoid and in which they have repeatedly entangled themselves (155). These two well-known thinkers’ endless quarrels about reality, truth, and inquiry can be set aside, as soon as we adopt a Deweyan form of pragmatism in which, Hildebrand argues, the problem of realism does not even arise. Chapters 2 and 3 in his book analyze Dewey’s debates with the realists and idealists of his own days, laying the ground for the claim (substantiated in the later chapters) that neopragmatists tend to misread Dewey in a manner similar to his early misinterpreters.

Dewey, Hildebrand points out, could not accept, for example, the American “New Realists” (e.g., W.P. Montague’s) views on the absolute thought-independence of things, because according to his pragmatic naturalism, the thinking organism is not isolated from its environment (nor vice versa). Dewey’s major assumption, viz., that “knowledge and experience are not coextensive”, was opposed to both realism and idealism, a point which neither his critics nor his admirers, labeling his view either realistic or idealistic, sufficiently appreciated (19, 22-23).20 The very division between subject and object that various realisms and antirealisms trade on is not a primary philosophical datum but “the result of an inquiry” – of a philosophical inquiry, to be sure, but an inquiry nonetheless (26). As already noted in section 2 above, inquiries are always carried on in concrete, practically problematic situations, never in total abstraction from what Dewey used to call the “problems of men”; indeed, “what is primary [for Dewey] is a whole situation – ‘subject’ and ‘object’ have no a priori, atomistic existence but are themselves derived from situations to serve certain purposes, usually philosophical” (27; see also 83, 119). When we remember that philosophical and scientific inquiry are not two poles of a dichotomy, we can apply Dewey’s insights into the constitution of scientific objects with full force to the issue of the constitution of “philosophical objects of inquiry” (or the “objects of philosophical inquiry”), such as the subject/object distinction.

For Hildebrand, the entire realism/antirealism controversy “begins from subjective premises”; thus, since Dewey rejects subjectivist construals of experience and knowledge, emphasizing that these are “social at root”, his pragmatism cannot be placed in a context defined by the realism/antirealism issue as commonly conceived (5). Here, however, questions begin to emerge: why shouldn’t the realism issue arise even on the basis of a nonsubjectivist, socially oriented theory of knowledge and/or experience? For instance, Wittgenstein’s (1958 [1953]) emphatically social conception of public language-use as the basis of linguistic meaning has often (correctly or incorrectly) been taken to be relativistic or antirealistic. At least, Wittgensteinian voices – Putnam’s (1994, 1999) among them – in the realism discussion have been voices that must be taken seriously.

Moreover, even Dewey’s views can and perhaps must be described in terms of their realist and idealist dimensions: nature is “not generated ab extra by knowing” but constrains knowledge (realism); yet, nature is changed as it becomes known, and “objects (existences with meaning) come to be through the act of knowing” (Hildebrand 2003, 60). It is extremely hard to get entirely rid of the realism/antirealism vocabulary, even if one is willing to grant (correctly) that a complex philosophical position such as Dewey’s (or Peirce’s, James’s, or Wittgenstein’s, for that matter) cannot be reduced to either of these opposing poles.21 One might even claim that a Deweyan will not get rid of metaphysical realism, the view that reality possesses its own ultimate ontological structure, as long as s/he simply replaces “substance” by “process” as “the basic ontological category” (82).22 For a more radical pragmatist, there is no such thing as “the basic ontological category”; all categories are human, purpose-laden constructions that serve some of our interests in coping with the world we live in. Metaphysical realism can surely be connected with a processual ontology, too.

Still, Hildebrand is careful to admit that he cannot and is not even attempting to go “beyond” Rorty’s and Putnam’s realism/antirealism debate in the sense of declaring that the debate is about nothing; for him, going beyond means “superseding” or “supplanting” the debate (178). In such a “going beyond”, Dewey offers him a helping hand. His skill at making use of Dewey is so great that I have no doubt at all about the success of his argument against Rorty’s project. He perceives, among other things, that Rorty may in the end have to embrace a form of linguistic idealism (Rorty’s own claims to the contrary notwithstanding), given that Rorty’s “causal talk” about the merely causal (viz., nonrepresentational) relations between language and (supposedly) extra-linguistic reality is, “in the end, just talk”, hence something that falls within language itself (109; see also 162-163, 182-183).23 Language alone, according to Hildebrand, cannot do the explanatory and descriptive work that Dewey’s notion of experience does (108). Rorty even commits the “hypostatization fallacy” (the typical philosophical fallacy according to Dewey), “forgetting that sentences and vocabularies are eventual products of inquiry, not its primordial medium” (126).

I almost completely share this critical attitude to what Rorty is attempting to do with Dewey, and with the classical pragmatists more generally.24 My only complaint is that Hildebrand’s uncompromising attack on Rorty’s linguistic turn, which he claims to be incompatible with the proper “practical starting point” of pragmatism, makes it impossible for us to employ Wittgensteinian resources in developing a properly linguistified form of neopragmatism. There is, I would argue, a sense in which pragmatism, even Deweyan pragmatism, can accommodate the Rortyan turn toward language (instead of, or perhaps rather in addition to, experience) in a late-Wittgensteinian setting, in which the focus is on the actual or possible use of language in various humanly relevant and historically changing practical circumstances, as distinguished from the supposedly stable syntactic or semantic structures of language as such.

In short, Hildebrand does not seem to be entirely fair to the possibility of such a linguistically oriented form of pragmatism. This is because he – apart from occasional general references (e.g., 174-175) – neglects the remarkable (both historical and systematic) connection between Wittgenstein and pragmatism, a connection which is of enormous importance, though in different ways, for both Putnam and Rorty. I believe that pragmatism can, with the help of Wittgenstein’s later philosophy in particular, be developed into a philosophy that (instead of, say, simply subordinating metaphysical problems to semantic analysis) takes seriously our actual human linguistic practices as ineliminable contexts of whatever experience, non-linguistic experience included, can ever be meaningful for us. If this is correct, then Hildebrand’s rejection of Putnam’s version of pragmatism, along with Rorty’s, as a species of the unpragmatist “theoretical starting point” is overhasty. Perhaps even Rorty, despite all of his serious shortcomings, is right to the extent that pragmatism is a plausible option in our post-linguistic-turn philosophical culture, even though his “ethnocentrist” and uncritically relativist conception of what this option specifically involves is (I agree with Hildebrand) deeply flawed.

Let us briefly turn to Hildebrand’s Dewey-inspired criticisms of Putnam, having been convinced that his critical analyses on Rorty are on the whole well defended. While Hildebrand admits that Putnam is “far more adroit than Rorty at capturing the spirit of Deweyan pragmatism,” that he has read Dewey more carefully than Rorty has (127), and that Putnam thus “fares better” in developing neopragmatism (154), he finds a number of things to criticize in Putnam’s position, too. Above all, Putnam, he says, finds the emphasis on “the primacy of practice” in pragmatism a “theoretical strategy” (150), thus in a sense – in a manner resembling Rorty’s – betraying the true primacy of practice that pragmatism requires. Putnam, Hildebrand points out, has written surprisingly little on the notion of experience, despite its absolutely central role in Dewey’s pragmatism and despite the fact that, in the Deweyan scheme, it is only in the course of experience that language can so much as arise (151-152; see also 217, n76). Putnam, then, adopts a theoretical (instead of a practical) starting point in his neopragmatist philosophizing by discussing our conceptual schemes, pictures of the world, theories, etc. – and their pragmatic orientations – instead of focusing on the primary experiences that are (in the famous Deweyan slogan) “had, suffered, and undergone” in our lives. Like Rorty’s, Putnam’s approach to pragmatism is linguistic or “conceptualistic;” in both cases, this can be regarded as “the residuum of their analytic pasts” (181). Hildebrand elaborates:

Though [Putnam’s] intent […] is to stress that sensation and conception are interdependent […], his argument starts with a theoretical discontinuity (inputs vs. concepts), which is only possible because of a previous abstraction from particular and concrete situations. This method is typical of Putnam, who relies upon categorizations such as “inputs,” “outputs,” and “conceptual schemes” to explicate the meanings of terms rather than giving a functional explication of how the terms are embedded within the inclusive and not necessarily cognitive contexts of experience and experiment. (184; emphasis in the original.)

It is not my primary task here to defend Putnam, because I do admit that Hildebrand perceptively questions his strong reliance on language and the resulting neglect of “experience”, but at this point one should remember, again, that Putnam, after all, is not only a Deweyan but also a Wittgensteinian thinker. The Deweyan description of the irreducible functionality of meaning is actually quite accurate in a Wittgensteinian context, too.25 If one compares Putnam’s “conceptual schemes” to Wittgenstein’s “language games”, then one is not forced to view them as an unnecessary “explanatory interface” but may instead see them as something that provides the vital “connections to practice” through which meaning, for pragmatists, is to be explicated (cf. 184). Thus, Hildebrand’s way of assimilating “language games” to (again theoretically – instead of practically – conceived) conceptual schemes (190) is misleading. For a Wittgensteinian philosopher of language like Putnam, language games are not (or at least not primarily) theoretical interfaces; they constitute the practices through which we experientially transact in and with our world. Hildebrand’s way of ignoring this possibility of saving Putnam from the Deweyan criticism results from his neglect of Wittgenstein as a crucial source for Putnam (and for Rorty). One may – just as in the case of Dewey – argue that Putnam reads Wittgenstein more carefully than Rorty, though again this is not to say that his reading is unflawed. What I am claiming is merely that there is much more to be found in the idea of a language game than Hildebrand is willing to find there, perhaps even something that Deweyan philosophers of experience ought to take seriously. Pragmatists celebrating the primacy of experience need not, then, entirely reject the linguistic approach, given that a practical starting point can be built into that very approach itself, as Wittgenstein and many of his followers have attempted to do.26

As Hildebrand cites Dewey attacking the all too common “assumption of the identity of objects of knowledge and ultimately real objects”, an assumption usually shared by realists and idealists alike (64-65), one begins to wonder whether Dewey is not in fact attacking basically the same assumption that Kant himself criticized (and which Kant saw as the “transcendentally realistic” flaw common to both his empiricist and rationalist predecessors). In other words, can’t we say that the distinction between “antecedent existences” and objects of knowledge is a new version of, or at least analogous to, the Kantian division between things in themselves and empirical appearances, provided that the latter is interpreted along the lines of the “one world – two aspects” reading rather than the more robustly metaphysical “two worlds” interpretation?27 Is Dewey so very un-Kantian, then, after all? Why, in brief, are many pragmatists and scholars of pragmatism so afraid of Kantian notions, particularly of the notion of a thing in itself, that no pragmatist reconceptualizations of such notions are ever even attempted? Couldn’t we even reread Dewey, like some scholars like to read Wittgenstein, as a Kantian-like transcendental idealist and empirical realist (though, admittedly, this would be a clear departure from his own favorite philosophical terminology)?

The following passage is, again, illustrative of Hildebrand’s anti-Kantianism:

Commentators like Rorty and Richard Bernstein read Dewey as a kind of Darwinian Kant, someone attempting to give a naturalized metaphysics of experience. They are right to see that kind of project as doomed to failure. But this is simply not Dewey’s project. Dewey’s pragmatism was not offering an updated version of Kant’s Copernican revolution (thinly disguised with Darwinian accouterment) but an authentic empirical alternative to Kantian and Cartesian metaphysics. It is, R. W. Sleeper reminds us,28 not a “First Philosophy” but a “Last Philosophy” […]. (120; emphases in the original.)

The problem with this is the assimilation of the Kantian “project” to a Cartesian form of first philosophy. True, Kant himself may be vulnerable to the charge, but present-day Kantians-cum-pragmatists (such as, arguably, Putnam) need not be. By questioning the rather unpragmatic and unhelpful dualism between experience and existence – and the corresponding dualism between the two allegedly different kinds of metaphysics focusing on one or another pole of this oppositon, a dualism often assumed in Deweyan scholarship – we may retain a (quasi-)Kantian reading of Dewey (and of pragmatism in general), without sacrificing the empirical spirit of his naturalistic, “Darwinian” metaphysics.29 It is, indeed, by being a metaphysics of (natural) existence that Dewey’s pragmatism succeeds in being a metaphysics of experience (and vice versa), insofar as Kant was right in insisting that the conditions necessary for the possibility of cognitive experience are equally necessary for the possibility of the objects of such experience.

Just as in Kant’s own philosophy, the problem of realism is absolutely central in this reinterpreted Deweyan pragmatism, as it also is in the neopragmatist reception of Dewey – even in Rorty, his own claims to transcend this issue notwithstanding. The problem of realism is not crucial for a pragmatist as a traditional philosophical question calling for a traditional answer in standard terms requiring us to choose between realism and some of its alternatives, but as a constant challenge to our attempts to develop pragmatism further in a Deweyan spirit. Not only new answers to the problem are needed, but new ways of pragmatically formulating the realism issue are required as well.

5. Conclusion

Having briefly gone through these three cases in which the realism issue becomes manifest in Deweyan pragmatism, we may conclude that there is a realism issue in Dewey – indeed several. They are not reducible to any essential, global formulation but must, rather, be investigated in a piecemeal fashion, case by case, respecting the pragmatic plurality of the cases themselves and the quite different needs that our practices in the relevant contexts serve. For example, we need to be sensitive to the different aspects of the realism dispute as it arises in relation to science, to religion, or to morality, to name just a few humanly central contexts in which it may arise. There is, if my brief comparison to Kantian and Wittgensteinian perspectives in the previous section is even remotely plausible, even a sense in which Deweyan pragmatic naturalism might be reconstructed as a form of transcendental idealism, though this will require a reconsideration of the standard meanings of “pragmatism”, “naturalism”, and “transcendental idealism”. We might even find resources for a novel perspective of “transcendental pragmatism” in Dewey (and the other pragmatists).

The realism issue, in my view, is ineliminable from serious philosophy, pragmatist or not, and the pragmatist tradition more generally can even be said to flourish through its fruitful tensions related to the realism vs. idealism controversy (in its many different forms). Nonetheless, we should avoid the illusion that pragmatist views, Dewey’s or others’, could simply be defined in terms of traditional (non-pragmatic) construals of these positions and tensions.30

Notes

1 I have borrowed some material from the following works of mine dealing with these issues from a Deweyan and more generally pragmatist perspective: Pihlström (2002), (2004b), and (2007), as well as El-Hani & Pihlström (2002). [N.B. My references to Dewey in this paper are to easily available paperback editions. If the conference papers are published, the references will be converted into references to the critical editions.] ↩

2 I am certainly not implying that it would be uncontroversial to classify Wittgenstein as a Kantian transcendental thinker. I do find this approach fruitful (cf. Pihlström 2003, ch. 2), but I am not going to explicitly deal with problems of Wittgenstein scholarship in this paper. In any case, Dewey and Wittgenstein might be interestingly compared in this regard. ↩

3 According to Peirce’s “extreme scholastic realism”, generals do not exist but are real. According to Dewey (1936, 532), such Peircean generals perform a vital function as “formulae of operations”, guiding us in our inferences; thus, their status is logical, not metaphysical (in Dewey’s broad sense of “logical”: cf. Dewey 1938). ↩

4 For Dewey’s instrumentalism about laws, see also, e.g., Dewey (1989 [1929]), 121-123. Shook (2003b, 328) connects the Deweyan view of laws as revisable “working hypotheses” with the more recent position among philosophers of science, such as Ronald Giere and Nancy Cartwright, that scientific realism need not be committed to the truth of exact laws of nature. It may be noted that Shook’s (2003a) volume contains a number of relevant papers on (scientific) realism and naturalism, understood in a Deweayn fashion. ↩

5 The entire ch. 4, “Nature, Means and Knowledge”, of Experience and Nature (Dewey 1989 [1929]) is essential to Dewey’s views on the status of scientific objects. ↩

6 On the importance of distinguishing between knowledge and experience in Dewey’s naturalism, see, e.g., Hildebrand (2003). ↩

7 See Dewey (1916, 331), for his characterization of instrumentalism as a “logical version of pragmatism”, according to which “knowing is literally something which we do”. On this intimate relation between knowledge and action, see especially Dewey (1960 [1929]), passim. Given Dewey’s lifelong effort to avoid unpragmatic dichotomies between these two, it is odd indeed to note that a relatively early French critic and further developer of pragmatism, Gaston Bachelard, has been interpreted as criticizing James’s and Dewey’s pragmatism for a model of instrumental or practical rationality in which reasoning (intellectual work) is separated from practical work (Tiles 2005, 172). Bachelard’s antireductionist emphasis on the richness and diversity of reality, to be encountered in a variety of historically developing ways, is obviously close to pragmatism, but his view that pragmatism failed to seek a “detailed understanding of particular conditions” and discuss the use of technology in experiments (160) is narrow, if not downright false. Dewey, in particular, devoted a life’s work to such issues. Perhaps Bachelard’s major novelty in relation to pragmatism was his insistence on the processual dynamics of science, coming close to views later known as Thomas Kuhn’s (1970). ↩

8 For such a strong conception of scientific realism, see Sellars (1963) and Tuomela (1985); for some criticism, see Pihlström (1996). According to Rorty (1997, pp. 92-93), scientific realism and religious fundamentalism are “products of the same urge”, the pursuit of an absolute conception of reality. It is this pursuit that the pragmatist who follows Dewey should abandon; and one certainly need not be a Rortyan in order to be able to appreciate these insights. ↩

9 More detailed characterizations of the notion of emergence are obviously beyond the scope of this paper. For the most comprehensive discussion of different concepts of emergentism, see Stephan (1999); cf. also Stephan (1992) and (1998). On the notion of “emergent evolution”, also relevant to the pragmatist tradition, especially Dewey, see Blitz (1992). For more explicit attempts to bring pragmatism and emergentism together, see El-Hani & Pihlström (2002) and Pihlström (2002). (A lot of relevant literature is cited in these papers, upon which the present section is partly based.) ↩

10 See, for instance, the debate between Kim (1998) and Stephan (1998, 1999). ↩

11 I am obviously indebted to McDowell’s (1996 [1994]) account of “second nature” and the “space of reasons” here. On McDowell’s relations to pragmatism, see Pihlström (2003), ch. 4. See also several relevant contributions to de Caro & Macarthur (2004). ↩

12 Goudge’s essay is a valuable introduction to Peirce’s, Dewey’s, and Mead’s evolutionary philosophies of mind. Goudge (1973, 134-135) points out that, in the pragmatists’ Darwinian picture, the mind “must have a causal capacity of its own” in order to be able to respond to goals or ends and to initiate purposive actions. He thus sees downward causation as an essential ingredient in these pragmatists’ emergentism. ↩

13 For Mead, the problem of emergence is intimately connected with the notions of the past and the present. He defines emergence as “the presence of things in two or more different systems, in such a fashion that its presence in a later system changes its character in the earlier system or systems to which it belongs” (Mead 1932, 69; see also 66). In Stephan’s (1998, 1999) terms, Mead’s, like most pragmatists’, emergentism is diachronic rather than synchronic, because temporality plays a key role in his position. Because of their dynamic attitude to metaphysics in general, pragmatists were not much interested in synchronic dependence relations, which contemporary emergence and supervenience theoreticians usually focus on. An emergent property (or “the emergent”) cannot, by definition, follow from the past “before it appears”, but when it appears it always does this (Mead 1932, 2, 11). Emergent properties, pragmatically viewed, do not have any definite ontological status prior to our conceptualizing them in terms of past and present: the emergent, Mead says, “has no sooner appeared than we set about rationalizing it”, showing that it can be found in the past (14). The emergent, then, is both a conditioning and a conditioned factor (15). Mead joins those emergentists who think that even exactly determined events can scientifically be thought of as emergent (17). Emergence does not require indeterminism. As for Dewey, the emergence of life is an important theme for Mead, as it “confers upon the world characters quite as genuine as those it confers upon living beings” (35). In any case, the emergent arises out of conditions that make its arising inevitable: “What we seek in the environment [of experience] is a statement of the world out of which the emergent has arisen, and consequently the conditions under which the emergent must exist, even though this emergence has made a different world through its appearances.” (42). Here Mead (43-44) refers to the famous British emergentists’ Samuel Alexander’s and C. Lloyd Morgan’s views. Moving to the social aspects of emergence, Mead somewhat puzzlingly notes that the “social nature of the present” arises out of its emergence through a process of readjustment: nature “takes on new characters”, e.g., life, and this process of readjustment is what “social” refers to (47). Such a social character can, he argues, “belong only to the moment at which emergence takes place, that is to a present” (48). This even leads to an acknowledgment of the “social character of the universe.” ↩

14 See Peirce’s essays in metaphysical cosmology, especially the 1892 essay “The Law of Mind”, reprinted in vol. 1 of Peirce (1992-98). ↩

15 Other authors who recognize, but do not elaborate on, the connections between pragmatism and emergentism include Blitz (1992, pp. 133-135, 200), McLaughlin (1992, 57), and Emmeche et al. (1997, 89). ↩

16 For pragmatic forms of moral realism insisting on the reality of values as entangled with facts, partly inspired by Dewey, see Putnam (2002) and (2004), as well as Pihlström (2005a). ↩

17 For a critical discussion of “religious naturalism” in some of its recent forms, including Dewey’s, see Pihlström (2005b). ↩

18 For these neopragmatists’ most important writings on Dewey, or employing and further developing Deweyan ideas and arguments, see the essays available in Putnam (1994), (1995), and (1999), as well as Rorty (1982), (1991), (1998), and (1999). It was already in Rorty’s famous 1979 book, Philosophy and the Mirror of Nature, that he, to the surprise of most analytic philosophers, declared Dewey to be among the three most important philosophers of the twentieth century, along with Wittgenstein and Heidegger. ↩

19 This section is partly based on my review of Hildebrand’s book (Pihlström 2004b). ↩

20 For a figure beautifully illustrating Dewey’s division of experience (which itself falls entirely within nature) into existences “had” and objects “known”, see Hildebrand (2003, 62). One might ask, however, whether the had/known distinction is not itself just a new dualism to replace the earlier ones, which Dewey and his followers ought to give up. ↩

21 See also Pihlström (2004a). On Dewey’s idealistic commitments, see especially Shook (2000). ↩

22 Incidentally, we may also note that James, another classical pragmatist, problematically thought that reality ultimately consists of “pure experience” – a “radical empiricism” not too far from Dewey’s celebration of “primary experience”. These metaphysical theories may be in tension with the pragmatist view that reality does not ultimately, all by itself, consist of anything; i.e., that the world remains whatever ontological structure(s) it may have through its being a world for us, practically and conceptually structured by us, a reality possessing (in Dewey’s words) a “practical character”. James’s (1975) pragmatist writings provide a better glimpse on the practical character of reality than his radically empiricist metaphysics. ↩

23 On the other hand, Putnam accuses Rorty of falling into metaphysical realism, as Hildebrand notes (2003, 168-169). ↩

24 See Pihlström (1998), ch. 7. Another author who has continuously examined Putnam’s and Rorty’s pragmatisms, arguing that Dewey’s version is superior to them, is Joseph Margolis (see his 2002, 2003). ↩

25 It is not necessary to speculate about Dewey’s possible influence on Wittgenstein’s thought here. As is well known, Wittgenstein knew William James’s work to some extent, and pragmatist ideas may have reached him through James. On Wittgenstein’s relation to pragmatism, see Pihlström (2003), ch. 2, and the research cited therein. ↩

26 Another, related point about which I am willing to take issue with Hildebrand is the use of the word “transcendental”. Hildebrand (like so many others) frequently uses this word synonymously with the word “transcendent”, although these terms surely do not mean the same thing in post-Kantian literature. Roughly, “transcendent” refers to what Hildebrand takes “transcendental” to refer to as well, namely, to something that lies beyond or external to experience (or nature); “transcendental”, however, should be taken to denote the necessary conditions for the possibility of (say) experience (or, from a Wittgensteinian point of view, meaningful language). Thus, if a notion or principle or argument is “transcendental”, this by no means makes it “transcendent” or commits its author to any assumption about the “transcendent”, at least if we follow the Kantian usage of these notions. I would be the first to admit that pragmatism – either Dewey’s or someone else’s – is not, and should not be, committed to any transcendent powers of the mind, or refer to transcendent objects in the explication of the meaning of signs, but when Hildebrand starts to talk about “transcendental (or magical) powers” and “transcendental objects” (42-43), the “absolute or transcendental” subject-object dichotomy (27), or “transcendental gaps” between mind and matter or between “us” and nature (60; cf. also 149), it seems that he simply has not paid due attention to the undeniable Kantian background of pragmatism (including Dewey’s). ↩

27 For these two ways of reading Kant, see the literature cited (in relevant contexts, in relation to pragmatism) in Pihlström (2003). ↩

28 Hildebrand is, of course, referring to the major Dewey scholar’s Sleeper’s 1986 book. ↩

29 This is not the proper place to examine this topic further, as I cannot here survey Dewey’s reading of Kant, let alone Hildebrand’s interpretation of that reading. I attempt to develop a version of “transcendental pragmatism” (which does not reject but rather embraces Deweyan pragmatic naturalism) in Pihlström (2003). For an attempt to reconsider the traditions of pragmatism and Kantian transcendental philosophy with the aim of synthesizing them, despite Dewey’s as well as James’s relatively strong anti-Kantian rhetoric, see also Pihlström (2006). ↩

30 I should like to thank Alexander Kremer for his kind invitation to contribute this paper to the Szeged Dewey Conference (May 30-31, 2007). ↩

Works cited

[N.B. The works by John Dewey listed in this bibliography are mostly to easily available paperback editions. For critical editions, see The Collected Works of John Dewey, 1882-1953, published by the Southern Illinois University Press, Carbondale, and divided into three series (Early, Middle, and Later Works), 37 Volumes, 1967-1990. Also an electronic edition on CD-ROM is available.]


Copyright (c) 2007 Sami Pihlström

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.