Rationalism

views updated

RATIONALISM

The term rationalism (from the Latin ratio, "reason") has been used to refer to several different outlooks and movements of ideas. By far the most important of these is the philosophical outlook or program that stresses the power of a priori reason to grasp substantial truths about the world and correspondingly tends to regard natural science as a basically a priori enterprise. Although philosophies that fall under this general description have appeared at various times, the spirit of rationalism in this sense is particularly associated with certain philosophers of the seventeenth and early eighteenth centuries, the most important being René Descartes, Benedict de Spinoza, and Gottfried Wilhelm Leibniz. It is rationalism of this type that will be the subject of this entry.

Two other applications of the term should, however, be distinguished.

Rationalism in the Enlightenment

The term rationalism is often loosely used to describe an outlook allegedly characteristic of some eighteenth-century thinkers of the Enlightenment, particularly in France, who held an optimistic view of the power of scientific inquiry and of education to increase the happiness of humankind and to provide the foundations of a free but harmonious social order. In this connection "rationalistic" is often used as a term of criticism, to suggest a naive or superficial view of human nature that overestimates the influence of benevolence and of utilitarian calculation and underestimates both the force of destructive impulses in motivation and the importance of such nonrational factors as tradition and faith in the human economy. Jean d'Alembert, Voltaire, and the Marquis de Condorcet, among others, are often cited in this connection. Although there is some truth in these criticisms, the naïveté of these and other Enlightenment writers has often been grossly exaggerated. Also, insofar as "reason" is contrasted with "feeling" or "sentiment," it is somewhat misleading to describe the Enlightenment writers as rationalistic, for many of them (Denis Diderot, for example) characteristically emphasized the role of sentiment. Reason was praised in contrast with faith, traditional authority, fanaticism, and superstition. It chiefly represented, therefore, an opposition to traditional Christianity.

Here there are two contrasts with the seventeenth-century rationalism of Descartes and others. First, this rationalism is not characteristically antireligious or nonreligious; on the contrary, God in some sense, often in a traditional sense, plays a large role in rationalist systems (although Spinoza's notion of God was extremely unorthodox, and it is notable that the opposition of reason and faith is important in his Tractatus Theologico-Politicus ). Second, the view of science held by such Enlightenment thinkers as Voltaire was different from that of rationalism, being much more empiricist. The central contrast embodied in the term rationalism as applied to the earlier systems is that of reason versus experience, a contrast that is certainly not present in the Enlightenment praise of the "rational." Parallel to this difference, there is a difference between the characteristic seventeenth- and eighteenth-century views on the nature and importance of system; the eighteenth century declared itself against the esprit de système of the seventeenth century, with its elaborate metaphysical systems, and in favor of an esprit systématique, which could be orderly without being speculatively ambitious. (See d'Alembert, Preliminary Discourse to the Encyclopédie and Condillac, Traité des systèmes.)

Rationalism in Theology

The Enlightenment spirit of rational criticism directed against the supposed revealed truth of the Scriptures also had effects within Christianity itself. In this connection the term rationalism is used in a specific theological sense to refer to the doctrines of a school of German theologians that was prominent roughly between 1740 and 1840, and which had great influence on the development of biblical criticism. With their spirit of antisupernaturalism can be associated Immanuel Kant's Die Religion innerhalb der Grenzen der blossen Vernunft (1793), in which rational morality is the basis of religious belief.

However, the best-known use of "rationalism" in a religious connection is an entirely negative one, in which it stands for an antireligious and anticlerical movement of generally utilitarian outlook, laying great weight on historical and scientific arguments against theism. This use of the term, a popular rather than a technical one, seems now to be obsolescent, its place being taken by humanism.

Rationalism versus Empiricism

Rationalism as it will be discussed here is standardly contrasted with empiricism. This contrast (which rests on that contrast between reason and experience which has already been mentioned) is now so basic to the use of the terms that no account can afford to ignore it, and a number of comparisons between views associated with these two outlooks will be made in the course of this entry. It is of course impossible to give a detailed comparison of the two outlooks, and in general comparisons will be introduced incidentally to the account of rationalist ideas. There is, however, one issue, that of innate ideas, which embodies a central disagreement between the two, and regarding which an account merely from the rationalist side would be particularly unilluminating. On this issue the disagreements will be considered in rather greater detail than elsewhere. At the same time, it is hoped that the treatment of this issue will give slightly more insight into the rationalist outlook than can be achieved by what is at other points inevitably a very selective summary of rationalist opinions.

Innate Ideas

Descartes distinguished three classes of "ideas" (by which he meant merely whatever it is in a man's mind in virtue of which he can be said to be thinking of a given thing): adventitious, factitious, and innate. The first type came to the mind from experience, the second were constructed by the mind's own activity, and the third were created by God together with the mind or soul itself. The last included what were for Descartes the three fundamental ideas of the basic types of substance: God, mind, and matter (or extension). For the most part Descartes argued negatively for the view that these ideas are innate, trying to show that they could not be derived from experience (where this means, fundamentally, sensation).

His argument had two main points. First, the ideas are pure, containing no sensory material; these ideas are not images, reproductions, or copies of sensory experience. Descartes regarded this as fairly obvious in the cases of God and of mind; and he made a particular effort (as in the argument of the wax in Meditations II) to establish the same claim for matter. Second, the fundamental ideas implicitly contain, in different ways, some idea of infinity, and in grasping the idea one thereby grasps the possibility of infinitely many and various modifications to which mind and matter can be subject. In the case of God this argument goes further, for here we grasp an actual infinity of perfections implicit in the idea. The same point, however, holds for all these ideas: The grasp of infinitely many possibilities must transcend what has been given to us in experience, since experience could have given us at best only a limited set of such conceptions, corresponding to what had actually been experienced.

Even if both points of the argument are granted as showing that these ideas are not totally derived from experience, it might be doubted whether they are enough to show that the ideas are innate. For might they not be grasped in some nonempirical manner at a later stage of life—for example, when (or if) someone comes to think in these very general terms? In Descartes's philosophy there is at least an implicit answer to this objection. Descartes thought that the pure ideas of mind and matter are used in the comprehension of experience even before they become conscious in reflection. It is by reference to these ideas that one forms the ordinary unreflective conceptions of oneself as having a series of thoughts or of a material object as enduring, occupying space, and having various characteristics, even though, before reflection, one's conceptions of these things will be confused. Thus, the operation of the pure idea is implicit in ordinary prereflective experience, and such experience begins to be acquired from the moment of birth; therefore, there is ground for calling the pure ideas innate.

In the case of God the argument is slightly different, since it is less clear that this idea is "put to use" in any prereflective way. Here Descartes may have meant to claim merely that it would be natural to the power and economy of God's operations that he should implant the idea of himself in the soul at its creation, "the mark," as Descartes put it, "of the workman on his work." There is indeed a difficulty in seeing how, for Descartes, there could be an idea in the mind of which the mind is not fully conscious (as this account implies), since for Descartes "mind" and "consciousness" were virtually equivalent. And this difficulty also arises for the ideas of mind and matter, since Descartes explicitly denied (presumably there was no alternative) that the infant or young person is fully conscious of his innate ideas; they are latent and emerge only later—in the process of learning language, for instance. Nevertheless, Descartes's claims for the operation of fundamental ideas in prereflective consciousness, although not quite consistent with his metaphysics of the mind, became an important element in later theories of innate ideas, especially in the debate with empiricism.

innate principles

Descartes appealed only to innate ideas, or concepts, the materials of judgments and beliefs. He did not invoke innate principles, or propositions, his view apparently being that granted innate ideas, we have only to grant in addition a certain power of the mind to elicit features implicit in these ideas in order to explain how necessary knowledge could be derived from innate ideas (as he supposed it could).

Leibniz, however, who continued the Cartesian insistence on innate ideas, added a requirement for innate principles. His argument was of the same general type as that ascribed to Descartes with respect to the ideas of mind and matter: If there were no innate and unlearned propositions, we could learn no propositions at all—at least not by way of logical deduction. For, he argued, confronted with any valid inference of the form "P, so Q," we could not see that Q followed from P except by having already grasped the necessary truth of the proposition "if P then Q." Thus, in order to follow any inference and to learn anything by deduction, first premises are required that must themselves be unlearned.

An objection to this argument can be seen from the famous difficulty raised by Charles L. Dodgson (Lewis Carroll) that if there is necessarily a difficulty in seeing the validity of the original inference as it stands, the same difficulty will recur with the inference obtained by the addition of the "innate" major premise; to grasp the validity of this inference, another major premise would seem to be required, and so on, thus starting a vicious regress. Dodgson's point makes it clear that no multiplication of premises can be adequate to extricate the validity of an inference; what is needed is something of a different category, a rule. At this point a characteristic empiricist rejoinder to Leibniz's puzzle is to claim that the rules of inference are not unlearned but are learned in the course of learning a language (they are the rules implicit in the correct use of "if," "then," "not," and so on). This illustrates the natural and perhaps inevitable tendency of empiricism, in contrast with rationalism, to turn to a linguistic account of logical necessity. (Such an account, however, even if adequate in itself, may not dispose of the issues as thoroughly as empiricism has tended to believe; the question remains of what is involved in learning a language.)

Leibniz, in introducing the argument just considered, explicitly stated that he was of the "Platonic" opinion that a priori knowledge (at least) is innate and "recollected" (New Essays, Book I). There is a difficulty, however, in knowing how far the doctrine is supposed to range: Leibniz's doctrine that the soul is a monad and that every monad only develops its own inner potentialities, being unaffected by anything outside, implies that in one sense all thoughts, of whatever kind, are innate. This problem involves major questions in the interpretation of Leibniz—in particular, of his views on sense perception. However, it seems reasonable to say that at least in his remarks on innateness in the New Essays Leibniz was distinguishing between kinds of knowledge and ideas, such that some (the pure and a priori) can be said to be innate and others cannot.

Leibniz's remarks were in criticism of the First Book of John Locke's Essay concerning Human Understanding, and they constitute a subtle consideration of the issues lying between rationalism and empiricism at this point. Locke's First Book, although called "Of Innate Ideas," is in fact chiefly concerned with innate principles (and in some part with the alleged innate principles of morality that had been advanced by his adversary, Lord Herbert of Cherbury). Locke considered various characteristics supposed to show that a given proposition is innate (that it is universally believed, that it is assented to as soon as understood, and so forth), and had little difficulty in showing that these are inadequate. He then turned to the consideration that tiny children do not display elaborate conceptions of logic and mathematics such as are alleged by rationalists to be innate. His principle in this instance was "There is nothing in the mind of which the mind is not conscious"; if these conceptions were innate, they would be in the infant's mind, hence it would be conscious of them and (presumably) could display this consciousness. Leibniz, in reply, claimed that this so patently follows that Locke, in insisting on the principle about consciousness, was in effect begging the question: This principle is what the issue turns on. But, as has been seen, this was not how Descartes put the matter. Leibniz here made the cardinal point of the discussion his own non-Cartesian doctrine of subconscious perceptions (connected with his general doctrine of continuity).

debate with empiricism

Once the obvious fact is granted that the allegedly innate ideas do not manifest themselves temporally before other experience, it may be wondered whether any point remains to calling them innate. It has sometimes been suggested that the doctrine of innate ideas merely depends on a confusion between a logical and a temporal sense of "prior." However, this is to underestimate the force of the rationalist claims that the allegedly innate material is such that its operation is a precondition of our learning anything else. It is not easy to decide how to evaluate these claims, as against the central empiricist claim that no such preexisting material need be postulated (the so-called tabula rasa theory of the mind). For one thing, empiricism in its first developments tended to make up for the lack of original raw material by crediting the mind with a very elaborate set of operations. This was evidently the case with Locke, who used such notions as "abstraction," "reflection," and "intuition"; who spoke of "ideas" that are not evidently mere copies from sense perception; and who admitted a nonempirical notion of "substance" and its powers. His position retained a number of rationalist elements. The much more economical apparatus of David Hume, which in effect admits nothing but sensations, their copies, and the operations of association, defines a quite distinctive empiricist theory.

If the debate about innate ideas is cast in terms of a Humean empiricism, there remain principally two issues, one logical and one psychological. The logical issue concerns the question whether highly general concepts, such as those used in mathematics and the sciences, are reducible to or analyzable into those sorts of empirical concepts that can plausibly be said to be derived from sense experience. It would be widely agreed that the answer to this question would be "no." The second, psychological issue is whether the acquisition of concepts, such as occurs in language learning—and this would include even the supposedly straightforward empirical concepts—can be adequately explained by a psychological model postulating only the minimum empiricist requirements of sense perception, retention, association, and so forth. There is influential opinion (held by Noam Chomsky and others) that the answer to this, too, must be "no"; any adequate model may well require stringent innate constraints on the direction and nature of generalization from learning situations. How far these restraints might be supposed to approximate to the rationalists' conceptions of innate ideas—or, in other words, whether the model demands an innate analogue to the possession of concepts—remains to be seen.

If this is indeed an open question, then there is an explicitly psychological version of the rationalist view that is still worth serious consideration. This is not, of course, to say that the innate elements in an adequate model would be likely to correspond to the particular sorts of "ideas" that the rationalists selected for this status—such as the metaphysical notions of God, matter, and mind. Also, there was certainly an endemic confusion, in both the rationalist and the empiricist position on this issue, between psychological and logical issues. Nevertheless, there is still some life in the question, in both its logical and its psychological aspects, the occurrence of the psychological term innate in the original debate not being merely the result of confusion.

Knowledge

It was remarked above that there would now be wide agreement that many general theoretical concepts of mathematics and the sciences do not admit of total reduction to empirical concepts. In contrast with positivist or operationalist views it would be agreed by many that such concepts as "mass," for instance, are not a mere shorthand for sets of possible observation data. Such agreement, however, although it would constitute a rejection of strict empiricism, would not in itself constitute an acceptance of rationalist views about such concepts. It is possible to think that these concepts "transcend," or "go beyond," the empirical merely in virtue of conventional elements—that they are parts of humanly constructed models of reality which relate the observable by imposing a structure on it.

Essential to rationalism, however, is a realistic view (incompatible with even a modified empiricism) about the relation of these concepts to reality and about the necessary relations obtaining between these concepts themselves. The intellectual grasp of these concepts and the truths involved in them is seen as an insight into an existing and unique structure of the world. It is not easy to express this picture (which in varying degrees dominated the rationalists) in less figurative language, but the picture has at least two consequences: that there is a unique set of concepts and a unique set of propositions employing these concepts that adequately express the nature of the world, and that these propositions form a system and could ideally be recognized as a set of necessary truths. There are, admittedly, difficulties about the last point, particularly with reference to Leibniz (these will be considered in the next section). However, something like this general picture is central to rationalism and leads immediately to the question of how anyone can come to know this uniquely correct representation of the world. This invites two more specific questions: What, in general, is the guarantee that knowledge of the world is possible? how can any individual tell in a particular case whether he has hit on some genuine piece of knowledge?

descartes's epistemology

Most rationalists tended to answer the first of the above questions by referring to God; some, but not all, did the same for the second; and they varied in the priority that they assigned to the two questions. Descartes started famously with the second question and found the answer in the "clear and distinct perceptions" of the intellect. Proving, as he supposed, the existence of God via clear and distinct perception, he then employed God's perfection of "being no deceiver" to establish in general terms the reliability of beliefs that went beyond clear and distinct perception. He was, however, so impressed by the thought that it was only in virtue of humanity being created and sustained by God that he could know anything at all, that he was constantly tempted to double back and use the divine perfection to guarantee even the basic clear and distinct perceptions, thus laying himself open to the charge of arguing in a circle.

However this may be, it is notable that in Descartes "clear and distinct perception" is a thoroughly epistemological category. The truths that can be clearly and distinctly perceived do not constitute one homogeneous logical or metaphysical class of truths; the class includes at least statements of contingent existence (his own, in the cogito ) and of necessary existence (that of God), contingent statements about immediate psychological experience, and necessary truths about the relations of ideas. The status of these last, which Descartes called eternal truths, is somewhat obscure. Descartes held, in the Augustinian-Scotist tradition, that they were the products of God's will; but it is left unclear what it is that God has brought about in creating eternal truths, and hence what it is that one knows in knowing them.

the cartesian tradition

The development of the Cartesian tradition within rationalism tended to emphasize to an even greater extent the theological elements in Descartes's theory of knowledge. Thus Nicolas Malebranche retained for the individual case the test of "clear and distinct perception" in a style that seems to assimilate it to moral perception and the promptings of conscience: "One should never give one's complete assent except to propositions which seem so evidently true that one could not reject them without feeling an interior pain, and secret reproaches of the reason" (De la recherche de la vérité, I, Ch. 2; for the moral analogue, see Bossuet, Traité de la connaissance de Dieu et de soi-même, Ch. 1, Sec. 7).

Malebranche gives a strongly Augustinian and indeed Neoplatonist turn to the general account of God's guarantee of the possibility of knowledge. His doctrine was that all our knowledge of the external world is mediated by God; the mind of God contains paradigm ideas in whose form he created the world, and it is these same ideas of which we are conscious when thinking about the world. This is the meaning of Malebranche's saying that we see all things in God. This doctrine, apart from serving religious purposes, was also an attempt to get around the difficulties inherent in Descartes's own causal account of relations between matter and mind (which will be considered more generally later in this entry).

The role of God in the foundations of knowledge takes different and less extreme forms in other areas of the rationalist tradition. The greatest contrast to the Malebranche development of Cartesianism might plausibly be said to be Spinoza's system. It is true that Spinoza did assert that it is the nature of God that guarantees the correspondence of our thoughts to the world, but he so transmuted the notion of God that the doctrine is only verbally similar to Cartesianism. "God" is one name ("Nature" is another) for the one substance, that is, everything that there is. This substance has infinitely many attributes, of which we can comprehend only two, mind and matter. These two attributes are necessarily parallel to one another, and corresponding to any mode of the one attribute there must be a mode of the other. Hence, thought and the material world are inherently adjusted to one another, and the development of knowledge consists in the project of rendering the thought component of this relation as clear (in Spinoza's term, as "active") as possible. It admittedly remains obscure how, within the constraints of Spinoza's determinism, this can be regarded as a "project" at all. Despite this and the other notorious difficulties, Spinoza's system is particularly interesting in the present connection as a thoroughgoing attempt to answer the crucial question that was left very much in the air in Descartes's thought, namely, how any knowledge of a necessary truth, regarded as knowledge of the relations of ideas, could also constitute knowledge of the world.

Leibniz's system, for all its radical differences from Spinoza's, resembles it in one respect having to do with the foundation of knowledge: The general possibility of the correspondence of thought to the world is guaranteed metaphysically by the existence of a correlation between the two. The monads are not affected by anything outside and each develops its own activity from within, but a correspondence between the activities of the monads is given by the "Preestablished Harmony"; and knowledge, the correspondence between "conscious" states of certain monads and other monads, is a special case of this. The Preestablished Harmony, however, depends on God's optimal choice, that is, on God's benevolence. Thus, in a less explicitly epistemological form, Leibniz (in contrast with Spinoza) reverted to the original Cartesian standpoint, in that there is a transcendent and personal God who has a will, and it is a result of his will that there is an ultimate guarantee of the possibility of knowledge.

In general, however, Leibniz was not much concerned with epistemological problems; in particular, he was uninterested in the question that was the starting point for Descartes: How can the individual be certain of the truth of anything? Spinoza was concerned with this question, and tried to develop a theory of knowledge that would avoid the regress latent in Descartes's method, arising from the question of how one knows that one knows. In Spinoza's "degrees of knowledge" it is an essential property of the highest, or intuitive, degree that it is self-guaranteeing. Even so, there is an evident shift in the Spinozistic outlook away from the Cartesian question "What do I know, and how do I know it?" Spinoza, like Leibniz and many other rationalists, gave the metaphysical description of the world from "outside," from a "God's-eye" standpoint rather than from the subjective epistemological standpoint from which Descartes (although unsuccessfully) tried to work. It is, perhaps, a mild irony of the history of philosophy that Descartes's attempt to start with subjective questions of epistemology and to "work out" from there had more influence on the development of empiricism than on later rationalism.

Science and Scientific Method

No attempt will be made here to give an account of the detailed developments of the philosophy of science within rationalist thought, or of the actual scientific conceptions held by or associated with rationalists, although these are of course of great importance, most notably in Leibniz's critique of Cartesian physics and in the development of his concept of force. We shall consider only one or two general points about the rationalists' conception of a completed science and associated notions of scientific method.

Rationalist developments in these matters can usefully be seen in the light of an unresolved conflict within Descartes's system between the method of approaching scientific inquiry and the expected shape of the final product. Descartes favored in principle an approach to inquiry that might be called systematically exploratory. This he called the analytic method; and the straightforward exposition of the results of such an inquiry would be heuristic in style, explaining the resolution of difficulties as they were encountered in the systematic progress. He seems, however, also to have had a picture of a completed science as a complete deductive system, ideally expressed in a unique system of theorems with necessary truths (of a metaphysical character) as its axioms; this he termed the synthetic method of exposition. There is, perhaps, no essential clash between these two ideas of method and result; but Descartes seems not to have been clear about the relation between the two or how this specific method, fully pursued, would yield this specific result. Ambiguities about this question emerge in Descartes's accounts of the role of experiment, in which he sometimes gives the incoherent impression that he is both engaged in logical deduction of scientific laws from self-evident metaphysical premises and doing experiments to assist him in this deduction. On the whole, it is probably better to regard the idea of a complete formally deductive metaphysico-scientific system as less important in Descartes's thought than is sometimes supposed, and to see him as using certain limiting principles of scientific explanation, within which he constructs models to explain particular phenomena.

The idea of the total deductive system, however, had a powerful effect on rationalism and reached its most extreme expression in the work of Spinoza, where the "synthetic" method of Euclidean demonstration is explicitly regarded as necessary to the highest form of understanding. This was not just an expository preference; it was an expression of the basic Spinozistic outlook, which regarded the relation of cause to effect as that of logical ground to consequence—for Spinoza all explanatory relations were logical and timeless. The parallel orders of thought and matter, remarked on earlier, supposedly guarantee that the logical relations of ideas will constitute a totally adequate expression of the nature of the world. (A singular application of this notion of total parallelism is to be found in Ehrenfried Walter von Tschirnhaus, who in Medicina Mentis [1687] argued that an adequate definition of laughter should be able to produce laughter.)

Leibniz, partly under the influence of Erhard Weigel, was also attracted to the "geometrical method." He devoted a good deal of effort to the project of a universal calculus, which would enable arguments on any subject matter to be cast into a rigorous demonstrative form. However, the idea of such a calculus in no way presupposes an ideal of being able to demonstrate scientific truths from metaphysical or other supposedly self-evident axioms, which was the Spinozistic and, on occasion, the Cartesian ideal. Even if Leibniz started with the notion that it should be possible to settle any argument by appeal to the self-evident, he abandoned it in his mature philosophy, in which he made fundamental the distinction between "truths of reason," which can be established by logical insight on the basis of the law of noncontradiction, and "truths of fact," which depend on the principle of sufficient reason and cannot be established on logical grounds alone. There are some notorious difficulties about this distinction, especially concerning the question of the nature of the contingency of "truths of fact," since Leibniz also held the further general principle that in all true propositions the predicate is contained in the subject. It does seem clear, however, that there is an ineliminable contingency about "truths of fact," and hence that the aspiration of reducing all knowledge to a system of deductions from self-evident premises must be impossible in the Leibnizian system.

Francis Bacon said in his Cogitata et Visa (1607), "Empiricists are like ants, they collect and put to use; but rationalists, like spiders, spin threads out of themselves." Bacon, of course, preferred the ants. Although there is some element of truth in the image of the spider, as applied to some rationalist thinkers, it does less than justice to the substantial empirical work done under rationalist inspiration. This is all the more so if one counts Galileo Galilei's view of science as fundamentally rationalist. He certainly rejected any kind of Baconian empiricism and shared the rationalist vision of a mathematical structure of reality that intellectual insight could grasp; but he perhaps had a more sophisticated feeling than any of the philosophers for the balance of imagination and experiment in physics. The rationalist tradition certainly embodied fundamental insights (lacking in empiricism) about the nature of science; above all, it saw the importance of mathematical structures in physical explanation and the vital possibility of a theory's making a conceptual jump beyond the observations and not merely (as in empiricism) an advance in generality. Its sense of the activity of the scientific mind, of its restructuring of observations through concepts and models, was very significant. At the same time, empiricism rightly fought for a clearer distinction between pure mathematics and natural science, undermined the aspirations to final certainty that dogged the rationalists, and emphasized the role of laborious observation and experiment in contrast with the rather dreamlike quality of rationalist visions of the universe. No clearer case exists in the history of philosophy of the need for, and eventual occurrence of, a synthesis; one aspect of that synthesis is neatly summed up in a remark of Giorgio de Santillana that "the true scientist has an empiricist conscience and a rationalist imagination."

Substance and Causality

In the history of classical empiricism the concepts of substance and of active causal power together became progressively weaker and were finally abandoned. Thus Locke employed the full Cartesian array of both material and mental substances, both possessing causal power; George Berkeley banished material substance, partly on the ground that it could not be conceived of as possessing causal activity, which belongs only to mental substance; Hume maintained that the notions of substance and of causal activity are unintelligible. By contrast, in the rationalist tradition the notion of substance has not declined; developments in the idea of causal activity, although partly parallel to the idea of substance, are very different; in general the fortunes of "substance" and of "causal activity" have not been directly linked, as they have proved to be in empiricism—both have undergone considerable and partly independent variations.

In the case of substance (which will be very briefly considered here) the concept has not so much been criticized as used in differing ways to express differing metaphysical views of the world. On one measure, at least, the extremes in this respect are represented by the philosophies of Spinoza and Leibniz. Spinoza gave what he claimed was an a priori demonstration that there could be only one substance (Deus sive Natura, God or Nature); this was intrinsically neither material nor mental, these distinctions arising (as noted above) only at the level of the different attributes of this same substance. Essential to Leibniz's outlook, on the other hand, was an infinite set of substances, the monads, each of them different from all the others. In their character, although there are difficulties of interpretation on this point, they are more of a mental than of a material kind.

On the question of causality an important stream in the history of rationalism stems from the problem left by Descartes, concerning the causal interaction of mind and matter. Descartes's own view, which postulated simple efficient causation as holding between the two types of substance, failed to appeal to even the most ardent Cartesians, and their attention was particularly directed to this question, although difficulties about the meaning of causation even between material bodies also were considered. The natural tendency in the Cartesian tradition was to move toward attributing all causal power to God, and this movement of thought culminated in the doctrine of occasionalism—that both physical and mental events in the world are occasions for the application of God's power, which itself directly produces what would normally be called the effects of those events. This doctrine is most thoroughly expressed in the writings of Malebranche. Similar views, however, are to be found in Louis de la Forge (Le traité de l'esprit et de l'homme, 1666) and Géraud de Cordemoy (Le discernement du corps et de l'âme, 1666), whose work was known to Malebranche.

The theory of occasionalism can be usefully contrasted with Berkeley's empiricist account of causation. For both the only genuine activity was spiritual. For Berkeley the effects of such activity were also spiritual (mind can affect only mind), and indeed there was no other type of substance. The occasionalists retained material substance and did not find it unintelligible that mind can act upon matter; however, they held that the only mind for which such action is intelligible is the infinite mind of God. Here, as elsewhere, the questions of the gulf between mind and matter and of causation as activity emerge as of common concern to both rationalist and empiricist metaphysics, the influence of Descartes being clearly discernible in both.

Another writer who inclined to occasionalism was Arnold Geulincx (Ethics, 1665; 2nd ed., 1675); however, he also suggested a different model for causality, in which God did not, as in occasionalism, make a constant series of miraculous interventions into the natural order but had established ab initio a series of coordinated developments, the relations between which are what is taken for causal interaction. In this connection Geulincx introduced the example of the two clocks, perfectly adjusted to keep the same time, one of which strikes when the other shows the hour; the appearance of causal connection between them is only a result of precise prearrangement.

This same analogy was frequently employed by Leibniz in explaining his own very thoroughgoing version of this thesis, in which all appearance of causal interaction is an instance of the preestablished harmony between the several developments of the monads. Here again there is a notable contrast with and a similarity to empiricism: Both Leibniz and Hume, each representing the culmination of one of the two traditions in its classical form, deny the existence of "transeunt action" between different things and see what is called causation as a correlation between phenomena. Leibniz, however, emphasized some kind of spontaneous activity within the monad, while for Hume neither such activity, nor any notion of a substance, such as a monad, was acceptable. The views of these two philosophers are also worthy of comparison on other subjects, such as space and time; and the points of contact between them are the more significant in the light of the radical and very obvious differences in the spirit, method, and presuppositions of their two philosophies. These differences in the two culminating figures constitute a paradigm, almost a caricature, of the divergent styles of thought associated with rationalism and empiricism, while at the same time similar pressures in the history of thought produced partly parallel developments in each.

Bibliography

For an understanding of rationalism, writings by and about particular rationalist philosophers should be read; references to these will be found in the bibliographies of the appropriate entries.

general works

Bouillier, Francisque. Histoire de la philosophie cartésienne. 3rd ed. Paris, 1868.

Cassirer, Ernst. The Philosophy of the Enlightenment. Translated by Fritz Koelln and James Pettegrove. Princeton, NJ: Princeton University Press, 1951. As the title indicates, this is not directly on seventeenth-century rationalism but relates to it.

Cottingham, John. Rationalism. London: Paladin, 1984.

Doney, Willis. "Rationalism." Southern Journal of Philosophy, Supp. 21 (1983): 1–14.

Lecky, W. E. H. History of the Rise and Influence of the Spirit of Rationalism in Europe. London: Longman, Green, Longman, Roberts, and Green, 1865; rev. ed., London, 1910.

Santillana, Giorgio de, and Edgar Zilsel. "The Development of Rationalism and Empiricism." In International Encyclopedia of Unified Science. Vol. II. Chicago: University of Chicago Press, 1941.

Windelband, Wilhelm. A History of Philosophy. Translated by J. H. Tufts. New York, 1901. Part IV, Ch. 2; Part V, Ch. 1. Cites the works and—summarily—the views of many philosophers, including minor ones.

philosophy of science

Burtt, E. A. The Metaphysical Foundations of Modern Physical Science, rev. ed. New York, 1952; paperback reprint, Garden City, NY: Doubleday, 1954.

Dijksterhuis, E. J. The Mechanization of the World Picture. Translated by C. Dikshoorn. Oxford: Clarendon Press, 1961.

Gillispie, C. C. The Edge of Objectivity. Princeton, NJ: Princeton University Press, 1960.

innate elements in language acquisition

Fodor, J. A., and J. J. Katz, eds. The Structure of Language. Englewood Cliffs, NJ: Prentice-Hall, 1964. Chs. 21 and 22. Papers by Noam Chomsky and E. H. Lenneberg, respectively.

other recommended titles

Bealer, George. "Modal Epistemology and the Rationalist Renaissance." In Conceivability and Possibility, edited by Tamar Szabo Gendler. Oxford: Oxford University Press, 2002.

BonJour, Laurence. In Defense of Pure Reason. Cambridge, U.K.: Cambridge University Press, 1998.

BonJour, Laurence. "A Rationalist Manifesto." Canadian Journal of Philosophy, Supp. 18 (1992): 53–88.

Brink, David. "Kantian Rationalism: Inescapability, Authority, and Supremacy." In Ethics and Practical Reason, edited by Garrett Cullity. New York: Clarendon Press, 1997.

Carruthers, P. Human Knowledge and Human Nature. Oxford: Oxford University Press, 1992.

Gendler, Tamar Szabo. "Empiricism, Rationalism, and the Limits of Justification." Philosophy and Phenomenological Research 63 (2001): 641–648.

Markie, Peter. "Rationalism vs. Empiricism." Stanford Encyclopedia of Philosophy. Available from http://www.plato.stanford.edu.

Peacocke, Christopher. The Realm of Reason. Oxford: Clarendon Press, 2004.

Peacocke, Christopher. "Three Principles of Rationalism." European Journal of Philosophy 10 (2002): 375–397.

Bernard Williams (1967)

Bibliography updated by Benjamin Fiedor (2005)

Encyclopedia of Philosophy