This is a section of a text I’m going to be presenting at a conference in a couple of weeks. I’d appreciate comments from anyone who may happen to read it.
Schelling’s initial conception of naturephilosophy can be expressed by the phrase: critique of grounds. More precisely, this means distinguishing between ideal and real grounds, the confusion of which is according to Schelling characteristic of mechanistic physics. This first stage of naturephilosophical inquiry fits perfectly into the Kantian conception of critical philosophy:
The task of a philosophical science of Nature largely consists in just this, to determine the admissibility as well as the limitations of such fictions in physics, which are absolutely necessary for the continued advance of investigation and observation, and only obstruct our scientific progress when we seek to use them outside their proper limits. (Schelling 1988 : 78)
Critical naturephilosophy is thus an extension of the Critique of Judgment’s examination of the limits of validity of teleological concepts to, for example, the concepts of quality (ibid. : 21) or of force (ibid. : 175). To what extent can the subjective necessity of using these concepts for the “continued advance of investigation and observation” be regarded as an objective determination of Nature?
An example from Kant’s Metaphysical Foundations of Natural Science helps to make clear the distinction between real and ideal grounds. Against the atomists, Kant argues that matter “fills a space, not through its mere existence, but through a particular moving force” (Kant 2002 : 210) that prevents other “movables” from entering that space; to which he makes the atomist object that we need not posit any such force because it is simply contradictory for two matters to occupy the same space. “But,” Kant replies,“the principle of noncontradiction does not repel a matter advancing to penetrate into a space where another is found” (ibid.). That two matters occupying the same space would be contradictory is a reason for us to hold that this is impossible (ideal ground), but this fact as such has no effect in reality; only a “particular moving force” (real ground) can prevent this state of affairs from coming about.
The first edition (1797) of Schelling’s Ideas for a Philosophy of Nature can be regarded as largely devoted to this critical naturephilosophy, but to the second edition in 1803 Schelling adds the subtitle “as Introduction to the Study of this Science” (Schelling 1988 : 3 n. 1), suggesting that critical naturephilosophy should be regarded simply as an introduction to the science of naturephilosophy itself. It is now necessary to separate ideal from real grounds not only because the former “obstruct our scientific progress when we seek to use them outside their proper limits”, but also in order to make possible a science of real ground as such, naturephilosophy proper. Since what distinguishes real from ideal grounds is that though the latter provide an explanation of what they ground, they do not themselves produce it, the essence of real ground is seen to be productivity. Naturephilosophy as science of real ground therefore means the science of productivity.
This science of productivity is developed systematically in the 1799 text “On the concept of speculative physics and the internal organization of a system of this science” (Schelling 2004; references by page number alone are to this text.). Speculative physics means a system of the concepts of natural science — not only physics in our modern sense but also chemistry and biology — with productivity as its principle. The concept of productivity, as principle, must itself be made to produce all concepts of what is productive, i.e. all real grounds.
Speculative physics is distinguished from its empirical counterpart in that the latter can treat only of mechanical motion, that is, motion which itself “results only from motion” (195), whereas the former “occupies itself solely and entirely with the original causes of motion in Nature”; or more generally, empirical physics can derive the phenomena of Nature only from other already-existing things, whereas speculative physics “aims generally at the inner clockwork and what is nonobjective in Nature” (196), the unconditioned or “unthinged” (unbedingte) that is the condition of every thing (Ding): “the first inquiry of speculative physics is that which relates to the unconditioned in natural science” (202).
The true contrary to speculative physics, however, is not empirical physics as historically given but “pure empiricism” or “history”, which is merely a “collection of facts, of accounts of what has been observed, what has happened under natural and artificial circumstances” (201) and is therefore not science; whereas in empirical physics “empiricism and science run riot together, and for that reason they are neither one thing nor the other” (ibid.). Schelling here has in mind the employment of such fictions as the ‘gravific fluid’ that would be the cause of Newtonian attraction, the examination of which is the task of critical naturephilosophy. The speculative naturephilosophy therefore provides the foundation for the critical in separating “science and empiricism as soul and body, and by admitting nothing into science which is not susceptible of an a priori construction” (ibid.). It is only insofar as a concept of physics can be constructed in productivity that it can be regarded as having valid objective use and thus as constituting the real ground of a phenomenon of Nature.
For our construction of a phenomenon to be valid, it must be a repetition or re-construction of the “inner construction” by which it is produced in Nature; but the acts of Nature by which phenomena are produced “are never isolated, but performed under the concurrence of a host of causes which must first be excluded if we are to obtain a pure result. Nature must therefore be compelled to act under certain definite conditions, which either do not exist in it at all, or else exist only as modified by others” (196-197). Because of this interaction of natural forces, we can only know the grounds of a phenomenon through an “invasion of Nature” (196) that isolates the contribution each makes to the result, that is, an experiment. Experimentation is a “question put to Nature, to which it is compelled to reply”, a “production of phenomena” (197) in which we “put [Nature] into conflict with herself and set her own forces in motion against her” (Schelling 1988: 57).
“But,” as Schelling remarks, “every question contains an implicit a priori judgment; every experiment that is an experiment is a prophecy”, or is made on the basis of a hypothesis. The hypotheses implicit in an experimental design can be confirmed by other experiments, but these will necessarily imply further hypotheses, which must be confirmed by further experiments, and so on ad infinitum; experimentation by itself can therefore provide us only with knowledge conditional upon other experiments, rather than with knowledge of the unconditioned (197). “Since the final causes of natural phenomena are themselves not phenomenal, we must either give up all attempt ever to arrive at a knowledge of them, or else we must altogether put them into Nature, endow Nature with them” (ibid.).
To put final causes (or rather a final cause) into Nature means to subordinate the study of Nature to an “absolute hypothesis” (197) which can be expressed most concisely as: the existence of Nature itself. This hypothesis maintains that over and above the various natural things that make up the world, there is also Nature itself: we must assume “that the sum of phenomena is not merely a world, but of necessity a Nature (that is, that this whole is not merely a product, but at the same time productive) […]” (ibid.). That Nature itself is productive implies, first, that it cannot be separated into inert matter in which forces are implanted only secondarily (Schelling 1988 : 154-157): “[matter] and bodies, therefore, are themselves nothing but products of opposing forces, or rather, are themselves nothing else but these forces” (ibid. : 156); and second, that we can know Nature only by making our knowledge thereof productive: “our knowing is changed into a construction of Nature itself, that is, into a science of Nature a priori” (198).
If, however, the system of speculative physics is to be more than merely hypothetical knowledge, this hypothesis by which we “endow” Nature with its final causes must be “as necessary as Nature itself” (197). The necessity of this hypothesis must be demonstrated by putting it to an “empirical test”: “as long as there is in the whole system of Nature a single phenomenon which is not necessary according to that principle, or which contradicts it, the hypothesis is at once shown to be false” (197-198). The hypothesis is shown to be necessary by deriving all phenomena of Nature from it; or rather, since this is clearly an infinite task, the necessity of a hypothesis can only be provisionally accepted so long as it has not been falsified.
This extremely classical conception of the nature of science is not satisfactory by the standards of Schelling’s own philosophy. On Schelling’s account, natural science, though it is the science of productivity, itself exists only as product: only the complete deductive system in which all phenomena of Nature have their place is in fact a science. The process of producing this system however is left out of the account. What should the speculative physicist do upon discovering a phenomenon that contradicts her hypothesis? How should she correct her system? Schelling’s remark that the hypothesis must “bear its necessity in itself” (197) suggests that he regards the empirical test as something of a formality: it is in the conceptual demonstration of necessity that the real work of speculative physics lies, and once this has been carried out, there is little danger of falsification. This neglect of the process of correction is particularly strange given that Schelling’s own work is (in)famous for its mutability. Furthermore, it is hardly compatible with the “first maxim of all true natural science, to explain everything by the forces of Nature” even if this has as consequence that “what we call ‘reason’ is a mere play of higher and necessarily unknown natural forces” (195). A more satisfactory account of science will therefore have to study this play of forces, or productive science, rather than science as product, having productivity only in its object.