Geum.ru

Iv российский философский конгресс

Вид материалаДоклад

Содержание


On scientific realism
Science, scientific realism and epistemology
Scientific realism and Kant’s critical philosophy
Constructivism construed
Kantian constructivism
Conclusion: Scientific realism, idealist constructivism and history
Подобный материал:
1   ...   7   8   9   10   11   12   13   14   ...   25

Третье заседание


Конструктивизм и дискурс


Third session


Constructivism and Discourse


Том Рокмор (США). Научный реализм, идеалистический конструктвизм и история

Tom Rockmore (USA). Scientific Realism, Idealist Constructivism, and History


Though it is widely believed that science is the main source of knowledge in our time, there is still no agreement on how science functions as a source of knowledge. One view, very popular at the moment, is formulated in terms of scientific realism, which is often understood as the correct alternative to constructivism. Proponents of scientific realism, many working scientists, philosophers of science and philosophy make versions of the familiar claim that science uncovers, discovers or reveals the mind-independent world as it is. Some observers (Boyd 2004) go so far as to identify scientific realism with the self-understanding of the sciences.

This paper will examine some forms of scientific realism. Though those interested in scientific reason often turn away from the history of philosophy, I will point out that the argument against scientific realism, though not under that name, is already clear in Kant (Kant 1998). Since Kant is a German idealist, one aim is to call attention to German idealism for current problems in epistemology, including philosophy of science. Another is to promote epistemological (or epistemic) constructivism (and empirical realism) in place of scientific realism.


On scientific realism

“Realism” is routinely understood in widely different ways. Ordinary realism is the view of the ordinary individual, which can be paraphrased as the belief that in our daily lives we in fact know the way the world really is. Metaphysical realism, sometimes also called Platonic realism, is is a more sophisticated philosophical reformulation of ordinary realism. It is the philosophical view that under appropriate circumstances at least some of the time we succeed in knowing the mind-independent world as it is, or again the way the world is. Empirical realism is the comparatively weaker philosophical view that knowledge concerns no more than what is empirically-given in experience, as opposed to unobserved or unobservable entities. Scientific realism is sometimes allied with scientism, or the view that science and only science provides knowledge of what is real (Sellars 1991).

Even a minimal acquaintance with scientific realism is sufficient to show that the conviction that science is the sole or at least the main source of knowledge in our time is not accompanied by widespread agreement of what this consists in with respect to realism. Views about realism and its importance for science vary widely. According to Fine, realism of any kind is simply dead (Fine 1986, 112). On the contrary, according to Putnam realism is the only philosophy of science that does not make the success of science a miracle (Putnam 1975, 73).

Scientific realism is understood in many different ways in an ongoing debate. In order to provide some of the flavor of the current discussion, I have arbitrarily selected recent accounts by Psillos (Psillos 2000), Devitt (Devitt forthcoming) and Boyd (Boyd 2004). Psillos’ article, which describes the main points in a recent book (Psillos 1999), proposes metaphysical, semantic, and epistemic theses. Metaphysically we can say that the world has a definite and mind-independent structure. Semantically we can report that scientific theories should be taken at face value. Epistemically we can claim that mature and successfully confirmed predictive scientific theories are at least approximately true of the world. Psillos, who is conscious of the challenge based on the past, further suggests the need to reconcile the historical record with some form of realism.

For Devitt, like folk theory, which centers on observable entities, the constructivist approach featured in Feyerabend and Kuhn concerns only observables. According to Devitt, scientific realism is best seen as a metaphysical doctrine based on unobservable entities or facts. Scientific realism features a commitment to current scientific theories in the belief that science is mainly right. The arguments for scientific realism are the success argument and related explanationist arguments, and the arguments against it are the underdetermination argument and (following Putnam) the pessimistic meta-induction argument.

Devitt points to but does not discuss social constructivism, which Boyd considers in some detail. According to Boyd, scientific realists hold that successful scientific research provides knowledge of largely theory-independent phenomena, whether or not they are observable. We are justified in accepting scientific claims at face value, but also as fallible. Foe Boyd, scientific realism is the sciences’ own philosophy of science, considers various challenges. He rejects criticism based on the underdetermination of theories, which reduces explanation to prediction; for an acceptable theory must also fit into pre-established theories. Any defense of scientific realism must eschew a priori grounds. In answer to the so-called pessimistic meta-induction he points to the idea of approximate truth.

These and other forms of scientific realism seem to have two main characteristics. First, they associate a specific form of realism with the undeniable accomplishments of modern science. Second, on the basis of scientific realism as the criterion of knowledge, modern science is depicted as successfully uncovering, discovering or revealing the mind-independent world as it is. A scientific realist is someone who formulates scientific theories with the intention of knowing the way the world is, to which we refer and for which we have evidence and knowledge on the basis of scientific predictions, which are themselves true or false. Scientific realists of all stripes share the belief that scientific theory correctly approximates to the way the world is.


Science, scientific realism and epistemology

There seems to be no relevant difference between epistemology as it arises in science and in general. The conviction that to know means to know it as it is go all the way back in the Western philosophical tradition to Plato and perhaps to Parmenides. In the Republic, Plato already features the influential view that some among us—call them men of gold—on grounds of nature and nurture can at least some of the time directly grasp the real. In modern times Platonic realism is widely known as metaphysical realism. As a species of metaphysical realism, the main distinction of scientific realism lies in the claimed association with science as successful in knowing the way the world is.. Thus Steven Weinberg, the quantum physicist, insists against Kuhn that science would be irrational if it is did not in fact really discover the structure of the mind-independent real. “If I agreed with Kuhn’s judgment about the progress of science, that there is no sense in which science offers a cumulative approach to some sort of truth, then the whole enterprise would seem rather irrational to me, even if not to Kuhn” (Weinberg 1999).

The view that knowledge consists in grasping the mind-independent world as it is is not unique to modern science. It is a staple of the entire Western philosophical tradition. In reacting against the Platonic view that we directly grasp what is as it is, the so-called new way of ideas features variations on the theme that we grasp reality as it is indirectly through the medium of ideas. The old way of ideas identifies ideas and reality through the thesis that reality and ideas are one and the same. The new way of ideas distinguishes ideas and reality; ideas stand in for and under appropriate conditions correctly represent reality. The latter thesis is common to such disparate thinkers as Descartes and Locke. The rationalist Descartes argues for knowledge running from the mind to the world through ideas. The empiricist Locke argues for knowledge running from the world to the mind through ideas. The claimed connection between ideas in the mind and the way the world is is still strong at present. Brandom, for instance, claims: “For the properties governing the application of those concepts [e. g. physical things such as electrons and aromatic compounds] depends on what inferences involving them are correct, that is, on what really follows from what. And that depends on how things really are with electrons and aromatic compounds, not just on what judgments and inferences we endorse” (Brandom 2000, 27).

The claim featured in scientific realism—there is a way the world is and we know it as it is—is a staple running throughout the entire Western tradition, including modern philosophy. The difficulty lies in making out this claim. The issue is already directly joined in Kant’s consideration of the relative merits of representationalism and constructivism.

Scientific realism and Kant’s critical philosophy

In order to grasp Kant’s relevance to the viability of scientific realism, it is important to see the link in his position between philosophy and science. The link between Kant’s grasp of science and his philosophical position is often evoked but just as often misunderstood. According to Reichenbach, “Kant’s solution of the epistemological problem was at the same time the last one in which science played a role” (Reichenbach 1957, xii). This judgment is echoed by Habermas. “Hence, I should like to put forth the thesis that since Kant science has no longer been seriously comprehended by philosophy” (Habermas 1971, 4). The belief that philosophy was later unaware of or at least oblivious to the importance of science conflicts with the contributions of Cassirer and Whitehead, Kuhn and Feyerabend, as well as numerous contemporary analytic philosophers of science.

In the same passage, Reichenbach asserts: “Neither in Kant nor in the prevailing schools of philosophy does he [i. e. the working scientist—T.R.] find an epistemology that enables him to understand his own scientific activity. Philosophy still acts like a stranger toward the gigantic complex of natural science, even to the point of rejecting it” (Reichenbach 1957, p. xii). Reichenbach’s intention is to defend Einstein’s view that physical space is non-Euclidean by undermining Kant’s view of space as necessarily Euclidean. According to Reichenbach, space cannot adequately be studied through an a priori philosophical method. Physics determines the geometry of real physical space. Yet we can accept the need to reject Kant’s view of space for reasons linked to the later development of modern physics while defending a certain Kantianism, suitably revised, with respect to the problem of knowledge.

Kant’s epistemology addresses the theme of what we can know, including the precise problem, central to scientific realism and all other forms of metaphysical realism, of whether we can claim to know the mind-independent world as it is. The inability to make out this claim is shown, according to Kant, by the failure of all previous efforts to arrive at knowledge in basing claims to know on objects. “Bisher nahm mann an, alle unsere Erkenntnisse müsse sich nach den Gegenständen richten; aber alle Versuche über sie a priori etwas durch Begriffe auszumachen, wodurch unsere Erkenntniss erweitert würde, gingen unter dieser Voraussetzung zu nichte” (Kant 1998, B xvi).

Though Kant’s claims for knowledge are a priori, his argument is strangely based on experience. He objects that if the cognitive object is in fact mind-independent, there is no epistemological link to it. I see no way to answer Kant’s objection to metaphysical realism, hence, no way to defend scientific realism.

If this argument is granted, then two consequences follow immediately. First, all form of metaphysical realism, including scientific realism, must be rejected, since there is no way to know that we in fact ever succeed in grasping mind-independent objects. Second, Kant’s own way of stating the problem of knowledge must also be rejected. In the famous letter to his friend Marcus Herz, written toward the beginning of the critical period, Kant asks: “What is the ground of the relation of that in us which we call “representation” [“Vorstellung”] to the object [Gegenstand]?” (Kant 1967, 71). Kant’s question suggests that representations relate mind-independent objects. Yet his critique of efforts to know mind-independent objects as they are in themselves, in his language things in themselves, leads to the rejection of his own way of posing the epistemological problem, hence of representationalism of any kind.


Constructivism construed

Representationalism, the new way of ideas, scientific realism, and metaphysical realism are intimately related. In his critique of art, Plato long ago seems to reject what later became known as epistemological representationalism. This approach to knowledge presupposes two main theses: there is a way the mind-independent world is, and under the proper circumstances we in fact know the way the world is by correctly representing it. The new way of ideas rehabilitates representationalism, hence metaphysical realism, including scientific realism. Yet the crucial argument pointing to the utter failure to make out metaphysical realism, hence scientific realism, was already formulated by Kant. Kant’s other, more promising constructivist solution to the problem of knowledge is contained in the often-mentioned, little-understood, so-called—Kant never uses the term—Copernican revolution in philosophy.

It is not by accident that those committed to scientific realism rarely say much about constructivism. Though Devitt, who treats it elsewhere (Devitt 1997) at least indicates he takes a dim view of it, neither Psillos nor Devitt discusses constructivism in their recent essay,. Boyd, who devotes half his discussion to this theme, differentiates three forms of what he calls social constructivism. These include (1) adoption of a scientific paradigm, which imposes a quasi-metaphysical causal structure on the phenomena scientists study; (2) the view that the production of scientific phenomena is subject to a variety of social influences; and (3) the idea that the findings of science depend on power relations within the scientific and broader community, but not on the facts. Boyd, who labels the first view as neo-Kantian, associates it with a large number of thinkers (e. g. Hanson, Kuhn, Scheffler, Shapere), who share the view that “any change in the fundamental laws involving a scientific term must involve a chence in referent” (Boyd 2002, 9). He links the second view, which he also labels as neo-Kantian with thinkers (e. g. Kripke and Putnam) committed to a causal theory of reference (Boyd 2002, 10). And he connects the third view, which he describes as the post-modern challenge, with recent work in the relatively new discipline of science studies ( e. g. Galison, Latour, Pickering, Shapin, Schaffer) and feminist philosophy of science (e. g. Alcoff, Harding, Longino, Hintikka) (Boyd 2002, 12).

This is not the place to study constructivism in detail (Rockmore 2005). Boyd evokes social constructivism. Constructivism comes in many different types, including social, asocial, contextual and acontextual, epistemological and ontological subforms. Mathematical constructivism goes back at least to the origins of Euclidean geometry in ancient Egypt and ancient Greece. Epistemological constructivism arises in modern philosophy in Hobbes, who influences Vico, and again independently in Kant. Since Hobbes’ constructivism was not well known, and Vico was not well known outside of the Italian language debate, later forms of constructivism, for instance Husserl’s conception of constitution and Carnap’s conception of rational reconstruction are mainly influenced by Kant’s asocial form of constructivism.

Husserl’s conception of constitution, emerges in a long series of writings including at least the Philosophy of Arithmetic, Ideas I, Cartesian Meditations, and in two posthumously published books The Crisis of European Sciences and Transcendental Phenomenology and Experience and Judgment. A typical statement occurs in Ideas I where he contends, in refuting subjective idealism, that all reality exists through the dispensing of meaning (Husserl 1962, § 55, 152-154). Husserl, who considers the subject, understood as consciousness, as self-contained and absolute, hence dependent on nothing, seems to be saying that the spatio-temporal world only is for a subject as what is intended (Husserl 1962, § 49, 136-139). He seems to be claiming there is a mind-independent external world with which we come into contact and which we know insofar as it is constituted in our consciousness through the intention, or way in which consciousness is directed toward its object.

In the Aufbau, Carnap works out what he later described as “the rational reconstruction of the concepts that refer to the immediately given” (Carnap 1967, v). The Aufbau presents a so-called constructional system of objects or concepts, in which the term “object” is taken in the widest possible sense. In following the logicist example of Principia Mathematica, Carnap proposes to derive all conceps from, as he says, no more than a “few fundamental concepts” (Carnap 1967, § 1, p. 5). This approach rests on the idea of reduction, later important in physicalism. An object or concept is said to be reducible if and only if statements about it can be replaced by statements about the other object (Carnap 1967, § 2, 6). Carnap here applies a theory of relations to problems of pure theory in respect “to the task of analyzing reality” (Carnap 1967, § 3, 7). As in his protocol theory, he substitutes logical constructions for sense data. Carnap distinguishes between concepts as objects and objects falling under concepts, and points to the difference between idealism and realism, for instance the Marburg Neo-Kantian view that thinking “creates” objects, and the realist view that thinking merely “apprehends,” or grasps, them. According to Carnap, the conception of construction is neutral with respect to this difference in suggesting that objects are neither created nor apprehended but rather constructed. In working out a theoretical way to reduce reality to the given, he suggests a model in which, on the basis of what is directly given to mind, the observer can produce a logical construction which is logically equivalent to, hence can stand in for, or replace, inferred but unobserved (and in principle unobservable) entities (Coffa 1993, 214-218).


Kantian constructivism

Kantian constructivism, which motivates such later constructivists as Husserl and Carnap, develops through his extension of mathematical techniques to wider epistemological themes. Mathematics concerns objects insofar as they can be exhibited, hence constructed, in pure, or a priori, intuition (Kant 1998, A 4). Mathematics constructs concepts, which “means to exhibit a priori the intuition which corresponds to the concepts” (Kant 1998, B 741), on the basis of a non-empirical intuition, but philosophy reasons on the basis of concepts.

In his Copernican revolution in philosophy, Kant generalizes his view of mathematics as necessarily yielding certain knowledge to all cognition in obscurely contending that the subject constructs the cognitive object as a necessary condition of knowing it. As a priori, Kantian constructivism is independent of context, hence time and place. Knowledge requires that one or more subjects in fact know the cognitive object, and a theory of knowledge must explain conditions under which this is possible. Kant’s solution consists in making the object dependent on the mind. According to Kant, if the subject constructs its cognitive object, then, since that object corresponds to the structure of the human mind, we can know it. He famously suggests in a passage on modern working scientists: “Sie begriffen, dass die Vernunft nur das einsieht, , was sie selbt nach ihrem Entwurfe hervorbringt ….” (Kant 1998, B xiii).

In place of the canonical view that knowledge consists in grasping a mind-independent object as it is, Kant suggests that knowledge consists in grasping a mind-dependent object we construct as a condition of knowing it. If, as Kant contends, the subject constructs the cognitive object, knowing and being must coincide. In this way, we reach the main German idealist thesis in its Kantian version.

Kant’s statement of the constructivist solution is helpful, but problematic. A main difficulty is that he provides no account of how construction actually takes place, and how it relates to what human beings are capable of doing. Kant avoids any conflation between the logical and the psychological conditions of knowledge at the cost of understanding the epistemological subject as a mere epistemological function (Kant 1998, §§ 16-17). Yet no one has ever shown how to bring Kant’s transcendental logical account of the conditions of knowledge together with what human beings are in fact capable of doing. Indeed, Kant suggests such an account is impossible. Though he specifies the conditions of knowledge, he unclearly claims that an account of how the subject constructs the object lies beyond our reach (Kant 1998, B 181).

Kant shows that scientific realism is indefensible but fails to clinch the argument in favor of his constructivist alternative. Post-Kantian German idealism generally rejects metaphysical realism while wrestling with the problem of how to carry forward and complete Kant’s proposed constructivist solution for the problem of knowledge.

The views Boyd considers under the heading of constructivism are relatively distant from Kant’s. So-called post-modernist constructivism is in fact a form of epistemological skepticism following from the idea that cognitive claims arise within relations of power. Neo-Kantian efforts based on a causal theory of reference restate but do not resolve Kant’s own suggestion that phenomena are appearances of things in themselves. This approach fails because of the notorious inability to show that representations in fact represent. The most promising form of constructivism considered by Boyd is the neo-Kantian idea that the scientific paradigm imposes a structure on what we study. This view, which is not found in Kant, is formulated by Hegel, who suggests that a change in scientific theory results in a change in its object. According to Hegel, knowledge is the result of a process in which theories based on experience are tested and modified through further experience with a view toward formulating a theory which, at the limit, which is not necessarily ever reached, would fully correspond with its cognitive object (Hegel 1977, 46-57).

In comparison to Kant, Hegel goes further in working out the constructivist approach to knowledge central to German idealism. Hegel follows Fichte in understanding the subject of knowledge as one or more finite human beings in a social and historical context. Epistemological construction is social, and, since a distinction between the social context and history cannot be drawn, also historical. In this way, Hegel continues the German idealist shift away from all forms of metaphysical realism, including scientific realism, to empirical realism, or again from Platonism to idealism, now understood as fully historical.


Conclusion: Scientific realism, idealist constructivism and history

I come now to my conclusion. I have been arguing that scientific realism is a species of metaphysical realism that, after Kant, should not be defended. I have further been suggesting that Kant’s constructivism, though not as he formulates it, is very promising. There are still numerous defenders of scientific realism, Yet in Kant’s wake, and despite his own interest in metaphysical realism, we should turn away from any form of the traditional assertion that to know means to grasp the mind-independent world as it is and toward the idea that we can know only what we in some sense construct.

Kant in part misunderstands the thrust of his own position. He indicates he is a deep Platonist in claiming to know Plato better than he knew himself (Kant 1998, B 390). Yet in rejecting the canonical claim that to know is to grasp what is as it is in favor of the replacement view that we know only what we construct, Kant is an anti-Platonist who points to, but does not work out, a very different, better constructivist approach to evidence, truth and knowledge. I would like to suggest that early in the new century the development of a historical conception of constructivism is a main task on the epistemological agenda.


Bibliography


Boyd, Richard, 2002. “Scientific Realism,” in Stanford Encyclopedia of Philosophy, 17 pp.

Brandom, Robert, 2000. Articulating Reasons, Cambridge: Harvard University Press.

Carnap, Rudolf, 1967. The Logical Structure of the World and Pseudoproblems in Philosophy, translated by Rolf A. George, Berkeley: University of California Press.

Coffa, Alberto, 1993. The Semantic Tradition from Kant to Carnap: To the Vienna Station, Cambridge: Cambridge University Press.

Devitt, Michael, 1997. Realism and Truth, Princeton: Princeton University Press.

Devitt, Michael, “Scientific Realism,” in The Oxford Handbook of Contemporary Analytic Philosophy, Frank Jackson and Michael Smith, eds., Oxford: Oxford University Press – forthcoming.

Dummett, Michael, 1978. “Realism,” in Truth and Other Enigmas, Cambridge: Harvard University Press, 145-165.

Fine, A., 1986. The Shaky Game, Chicago: University of Chicago Press

Habermas, Jürgen, 1971. Knowledge and Human Interests, translated by Jeremy J. Shapiro, Boston: Beacon Press.

Hanson, N. R. 1958. Patterns of Discovery, Cambridge: Cambridge University Press.

Hegel, G. W. F., Hegel’s Phenomenology of Spirit, translated by A. V. Miller, New York: Oxford University Press.

Husserl, Edmund, 1962. Ideas: General Introduction to Pure Phenomenology, translated by W. R. Boyce Gibson, New York: Collier Books,

Kant, Immanuel, 1998. Critique of Pure Reason, translated by Paul Guyer and Allen Wood, New York: Cambridge University Press.

Kant, Immanuel, 2001. Prolegomena To Any Future Metaphysics, translated by James W. Ellington, Indianapolis: Hackett.

Kant, Immanuel, Philosophical Correspondence, 1759-99, 1967. Translated and edited by Arnulf Zweig, Chicago: University of Chicago Press.

Kuhn, Thomas, 1970. The Structure of Scientific Revolutions, Chicago: University of Chicago Press.

Laudan, Larry, 1981. “A Confutation of Convergent Realism,” in Philosophy of Science 48, 218-249.

Peirce, C. S., 1998. “What Pragmatism Is” (1905), in The Essential Peirce. Selected Philosophical Writings, volume 2 (1893-1913), edited by Nathan Houser et al., Bloomington: Indiana University Press, 331-345.

Psillos, Stathis, 1999. Scientific realism : how science tracks truth, London: Routledge, 1999.

Psillos, Stathis, 2000. “The Present State of the Scientific Realism Debate,” in British Journal of the Philosophy of Science (2000), 705-728

Putnam, Hilary, 1978. Meaning and the Moral Sciences, London: Routledge,1978

Putnam, Hilary, 1981. Reason, Truth and History, Cambridge: Cambridge University Press.

Reichenbach, Hans, 1957. The Philosophy of Space and Time, translated by Maria Reichenbach and John Freund, with an introduction by Rudolf Carnap, New York: Dover Publications.

Rockmore, Tom, 2005. On Constructivism, Lanham: Rowman and Littflefield.

Sellars, Wilfrid, 2000. Empiricism and the Philosophy of Mind, introduction by Richard Rorty, study guide by Robert Brandom (Cambridge: Harvard University Press.

Sellars, Wilfrid, 1991. “Philosophy and the Scientific Image of Man,” in Sellars, Science,

Perception and Reality, Atascadero: Ridgeview, 1-41.

Weinberg, Steven, 1999. Letter in the New York Review of Books, vol. XLVI, no. 3, February 18, 1999, 49.