Marjorie Grene and David Depew
The Philosophy of Biology: An Episodic History
Cambridge University Press, 2004
The Eighteenth Century II
Kant and the Development of German Biology
Although Immanuel Kant produced an important body of occasional writing about biological topics to which we will turn more explicitly at the end of this chapter the bulk of his work on the philosophical aspects of biology can be found in the section of the Critique of Judgment entitled “Critique of Teleological Judgment” (Kant 1790; 2nd ed. 1793). Kant reflects there on the key philosophical issues in the life sciences, especially teleology and reductionism. It must be conceded that these issues were not central to Kant’s lifework, and that they play only a supporting role in the overall line of argument of the Critique of Judgment itself. Nonetheless, Kant’s ideas about the central questions of biological inquiry were influenced by, and in turn influenced, practicing biologists in early nineteenth-century Germany, as well as later philosophers. Indeed, some of those with whom Kant interacted, especially Johann Friedrich Blumenbach, were at that very time laying the foundations of the modern science of biology. For these reasons, Kant deserves a place in a history of the philosophy of biology.
On its theoretical side, Kant’s critical philosophy was an effort to demonstrate that the logical forms inherent in the very act of thinking (like the subject-predicate, if-then, and other such relationships) correspond to categories into which our objective experience - our experience of objects - falls. Our experience can be objective only because this conjunction of sensible matter and categorical form occurs. Kant’s effort to
show how and why this must be so was aimed at securing not only metaphysics, but also scientific knowledge against skeptical attack. Hence, in working out his critical philosophy, Kant was at the same time seeking to demonstrate that categorically constituted material objects moving through space and time are necessarily governed by the “pure principles” that undergird modern physics - the three Newtonian laws of motion, and the dynamical forces, both attractive (gravitational) and repulsive (elastic), that these laws govern. (Forces were multiplying apace in the late eighteenth century; magnetic, electrical, and perhaps chemical forces were becoming conjoined to the gravitational core.) These laws certainly had to be discovered, verified, and applied empirically. But the objectivity, certainty, and necessity of modern physics, which permitted it to join the ranks of fully mature forms of systematic knowledge such as: logic and mathematics, was to be guaranteed by the a priori connections Kant sought to establish between the categories, our pure intuitions of space and time, the meaning of the concept of matter (which for Kant referred to the inherently lifeless stuff that is pushed or pulled around by measurable forces), and, finally, Kant’s slightly reconstructed version of the basic laws of mechanics.
An integral part of this analysis - and something of its motive as well, given Kant’s self-confessed attempt to refute Hume’s skepticism about the objective nature of the causal bond - was the ingenious suggestion that the formal notion of the hypothetical, or if-then, relationship corresponds to the pure concept of cause and effect. When referred to objects in time, this category reappears (in the Second Analogy) as the principle that causes must precede their effects. “If I lay a ball on a cushion, a hollow follows upon the previously flat smooth shape,” Kant wrote in the Critique of Pure Reason. “But if for any reason there previously exists a hollow in the cushion, a leaden ball does not follow upon it” (Kant 1781 [A], p. 203; 1787 FBI, pp. 248-9) (trans. Kant 1929, p. 226).  Even in cases of the simultaneous interaction or “community” that obtains between bodies governed by Newton’s third law, Kant concluded that an exertion of force by an affecting body (not necessarily by crude impact) must precede any change in the motion or rest of an affected body, even if only by an infinitesimally small interval (Kant 1781
1. Citations of Kant’s Critique of Pure Reason refer to the page numbers of the first (A) edition and the second, or B, edition. All translations from the Critique of Pure Reason are from Kant 1929. All other citations of Kant’s work refer to the so-called “Akademie [hence Aki Ausgabe” of the Gesammelte Schriften (Kant 1908-13).
[Al, p. 203). At bottom, there is a single direction of causality. That is a law of nature.
It is not our task to interpret this line of argument, far less to judge its success or failure. Our concern is with grasping the implications of Kant’s theory of causality for the possibility of scientific knowledge of living things - or rather with grasping what Kant took these implications to be, and why. To put the problem simply, Kant thought that the doctrine that causes must precede their effects is in considerable tension with our experience of organisms. For in an organism - a “natural purpose,” as he calls it, for reasons we will explore later - each part is reciprocally means and end to every other. This involves a mutual dependence and simultaneity that is difficult to reconcile with ordinary causality. In fact, so concerned was Kant with the conflict between the efficient causes that uniquely explain natural phenomena and the final causes that seem required in the case of organisms - and with the related question as to whether we might someday find a point at which the laws governing matter also determine and explain the formation of organic stuffs and organs - that he was willing to assert quite confidently:
It is quite certain that in terms of merely mechanical principles of nature we cannot even adequately become familiar with (kennen lernen) much less explain (erklären) organized beings and how they are internally possible. So certain is this that we may boldly state that it is absurd for human beings even to attempt it, or to hope that perhaps some day another Newton might arise who would explain to us, in terms of natural laws unordered by any intention (Absicht), how even a mere blade of grass is produced. Rather we must absolutely deny that human beings have such insight.
Kant 1793, ¶75. Ak. V, p. 400; cf. ¶77. Ak. V, pp. 409-410 
Kant’s concern with potential conflicts between efficient and final causes arises in at least three areas: the mutually supportive web of relationships between various species of plants and animals in what we call an ecological community; the fact, mentioned earlier, that the parts of a single organism are mutually dependent on one another in a radical way; and the process of ontogeny, in which a future end-state seems mysteriously to guide the successive stages of the development of an organism. Although we will deal with all three of these topics, we will be
2. All translations of the Critique of Judgment are from Kant 1987.
especially concerned with the last: how the patterned growth of a plant or animal seems to be influenced by the very end-state toward which the organism is oriented. Kant deals with this question late in the argument of the “Critique of Teleological Judgment” (Kant 1793, § 81). It seems reasonable to consider his substantive, scientific views about ontogeny in connection with the general argument presented there. For on the face of it, there would be less conflict between the Second Analogy, on the one hand, and the reproduction, differentiation, and growth of organisms, on the other, if Kant had subscribed to a straightforward, “encasement” version of preformationism. The fact is, however, that he did not hold to this view, but was influenced instead by a decisive movement toward the epigenetic alternative by biologists in his own time.
On the preformationist family of views, the cause of an organism’s coming-to-be is presumed to lie entirely in a pre-existing germ (whether sperm or egg) that already contains the whole differentiated organism in miniature, which merely unfolds or rolls out (the original meaning of the term evolution [from e-volvere]) under a set of perfectly mechanical, when-then conditions. For his part, Leibniz had no trouble with preformationism in this sense. He was not bothered by the implication that each germ cell must not only contain a whole organism in miniature, but must also encase within itself the germ cells of descendants down to the last generation. On the contrary, Leibniz showed marked enthusiasm for the notion of an infinity of infinitely small systems organized into functionally differentiated parts, his preoccupation with theodicy being supported by the contemporary fascination with microscopy.  For Kant, on the other hand, the notion of an infinity of already articulated and differentiated parts runs up against the requirement that organisms, as natural beings, must at some point be composed of inert, homogeneous, material stuffs that obey purely physical laws. In the Critique of Pure Reason, Kant rejects as “unthinkable” Leibniz’s view that organization can go on until infinity (Kant 1781 [A], ¶526; 1787 FBI, ¶554). But in addition to this philosophical objection, Kant was also affected by
3. Preformationism can easily appear mechanistic, but not particularly naturalistic. Epigenesis, by contrast, can appear naturalistic, but not mechanistic. Clark Zumbach infers from this perception that Kant insists on design in order to qualify the presumptive naturalism of epigenesis (Zumback 1984, p. 92). We would suggest that design is invoked by Kant as an idea not to qualify the naturalism of epigenesis, but to keep naturalism from leading to complete mechanism.
empirical considerations. In his day, embryologists were no longer able (or inclined) to sweep under the rug the problems that fertile hybrids posed for preformationism. If either the egg (ovist preformationism) or the sperm (vermist preformationism) contains the whole organism, how can fertile hybrids possibly arise? And if a combination of factors from both parents is required, what can preformation even mean? Moreover, how could broken-off pieces of the polyp regenerate a whole organism, as they clearly did? In the face of these difficulties, Caspar Friedrich Wolff at the University of Halle and Albrecht von Haller at Gottingen were influential in giving new salience to the notion that each organism is a result of a de novo process of fertilization and development. This is, in a general sense, the epigenetic approach that goes back to Harvey’s reworking of Aristotle’s Generation of Animals. But epigenesis brought with it its own problems. It seemed to require some sort of teleology to guide the formation of an organism. For an Aristotelian like Harvey, of course, this was not a problem; he believed that final causes are truly causes. For Kant, however, the fact that a fully functional, living being is reliably the result of a series of material causal episodes seemed mysterious indeed: potentially a violation of the notion that causes must precede their effects, and inconsistent with any sound definition of laws of nature.
In fact, the situation was somewhat more complicated than that. Kant agreed with Buffon, von HaIler, and Charles Bonnet that there are pre-existent germs for primordial parts. But he did not believe (any more than Haller did) that these germs are floating free in the world, as Buffon had, or that they are formed into a living whole by something as mechanical as an interior mold (moule intérieur) (Kant 1763. Ak II, ¶68; see Chapter 3 on Buffon). Kant’s contribution to the scientific question (proposed as early as 1763) was to postulate that, in addition to germs, certain predispositions (Anlagen) are also heritable. These predispositions make themselves felt only under the specific conditions to which the organism-to-be is to be adapted, thereby guiding its formation toward its fully functional end. Germs are species-specific and constant; dispositions are conditional and variable. Kant called his approach “generic preformationism” rather than something like “germ-grounded epigenesis.” In this phrase, the adjective takes back some of the noun; Kant’s preformationism is “generic” because he recognizes that only “the form of the species is preformed virtualiter” or potentially (Kant 1793, ¶81. Ak V, 423, our italics). When it comes to individuals, Kant’s position is fundamentally a version of epigenesis. The individual
has no prior identity in ovo or in utero; it is a product of a complex interaction between germs, predispositions, and environmental Contingencies. “Nature,” Kant writes, “produces organisms rather than merely developing them” (Kant 1793, ¶81. Ak V, p. 424). He even admits explicitly that this position is substantially epigenetic. “The system that considers individuals as products is called the system of epigenesis,” he writes. “It is what we call generic preformationism” (Kant 1793, ¶81. Ak V, p. 423).
Whether it was philosophical or empirical considerations that predominated in his thinking, however, what is important in the present context is that Kant’s subscription to epigenesis, qualified though it may have been by his acceptance of pre-existent germs and rendered more plausible by his postulate of inherited predispositions, was still in considerable tension with the views of causal explanation countenanced in the Critique of Pure Reason, as well as with the sort of science whose foundations are grounded in the Metaphysical Principles of Natural Science (Kant 1786b). One part can certainly trigger off the development of another in a causal sequence under particular conditions within the womb of the mother as well as under external environmental conditions. But in a living thing, the existence and balanced functioning of developed organism?
Kant might have relieved the tension arising from the goal-orientation of ontogeny and the strict when-then causality to which he was committed by the doctrine of the first Critique if he had been willing to believe that organisms are artefacts: productions, presumably, of a divine creator. The idea of a house, for example, is the “ground of the possibility” of the production of a particular house, as is the idea of any other technical product in which functionally different parts are subordinated to an Overall intention. Such a product, Kant says, is a purpose, goal, or end (Zweck): “A purpose is the object of a concept insofar as we regard this concept as the object’s cause... We think of a purpose if we think... of the object itself... as an effect that is possible only through a concept of that effect” (Kant 1793, ¶10. Ak. V, p. 220). There is no violation
of the Second Analogy in regarding organisms as objects brought into being in this way. For a guiding idea occurs in the mind of the architect before the house comes into being, and initiates a sequence of actions that results in and guides its production. If organisms were artefacts, they would, in this sense, be purposes. (Note that, for Kant, the concept of a purpose [Zweck] is applied first to the things or states of affairs brought into existence by means of an intentional idea that individuates these objects (Gegenstande), and not to the guiding idea itself. A purpose, he writes, is the “object of a concept,” not the concept of an object. This does not coincide with English usage.) For several reasons, however, Kant was unwilling to say that organisms are, or indeed are crucially like, artefacts.
For one thing, the inference to a superhuman artificer depends on the validity of the “physico-theological proof” of God’s existence, commonly known as the argument from design. Unquestionably, Kant had a great deal of respect for this argument. In the Critique of Pure Reason, he calls it “the oldest, clearest and most accordant [of all such arguments] with the common reason of mankind” (Kant 1781 [Al, ¶623; 1787 FBI, ¶565, trans. Kant 1929, p. 520). Indeed, Kant was always eager to assign the argument from design a certain heuristic function in our thinking about living things; even in the first Critique, where arguments for the existence of God are pilloried, he remarked that the argument from design “enlivens the study of nature, just as it derives its existence and gains ever new vigor from that source” (Kant 1781[Al, ¶623; 1787 [BI, ¶65 1, trans. Kant 1929, p. 520). How precisely to construe the heuristic role of the design argument is, in fact, one of the central issues of the “Critique of Teleological Judgment.” Still, given the limitations of our cognitive apparatus, which grants genuine knowledge only when we succeed in organizing sensory data under determinate concepts, Kant thought it presumptuous (vermessen) to “posit a different, intelligent being above nature as its architect” (Kant 1793, ¶68. Ak. V, p. 383). Moreover, as a fascinated reader of Hume’s Dialogues Concerning Natural Religion, he was keenly aware that a vicious circle results “if we introduce the concept of God into the context of natural science in order to make the purposiveness in nature explicable and then turn around and use this purposiveness to prove that there is a God” (Kant 1793, ¶68. Ak. V, p. 38).
Perhaps more significant in the present context is the fact that Kant was unwilling to let the artefact line of reasoning even get off the ground, because he was also aware of certain disanalogies between organisms
and artefacts. These render the inference to a divine maker questionable from the outset. There cannot be an argument from design to a designer unless there is an argument to design in the first place. But Kant was alert to the consideration that the parts of an organism are so mutually dependent and so tightly connected with the whole that it is difficult to say what, if anything, should come first and what should come later, as we must do when we design, build, and analyze (“reverse engineer”) artefacts. In this respect, Kant says that organisms are - or at least must be grasped by us as - self-formative, bootstrapping operations, in which each part appears to be the joint product of all the other parts. This is what Kant means when he says that an organism is “a product of nature in which everything is both an end and also a means” and in which the parts are “reciprocally cause and effect of [one another’sl form” (Kant 1793, ¶65-66. Ak. V, pp. 373, 376).
By contrast, the parts of a machine are not reciprocally related as cause and effect, or means and end, of one another. They are connected to one another in a looser, more “decomposable” way. True, the parts of a machine may all be necessary if the other parts are to perform their proper roles in the working of the whole object. The legs of a table, for example, must not be wobbly if the surface is to be stable. But none of these parts is a means that brings the others into existence. “In a watch,” Kant writes, “one part is the instrument that makes the others move, but one gear is not the efficient cause that produces another gear” (Kant 1793, ¶65. Ak. V, p. 374). In this connection, Kant denies that organisms are even like the self-moving machines envisioned by Descartes. “A machine has only motive force,” he writes, “whereas an organism has self-formative force” (bildende Kraft) (Kant 1793, ¶65. Ak. V, p. 374) - a statement from which we might well infer that epigeneticism is imprinted into Kant’s very conception of an organism. Precisely because they are not self-formative, moreover, the parts of an artefact must be “the product of a rational cause distinct from the matter of the thing, that is, distinct from the thing’s parts” (Kant 1793, ¶65. Ak V, p. 373). They compel an inference to a designer. The complex reciprocal causality of living things, on the other hand, requires us to disavow the very notion of an external agent. “We say far too little,” Kant writes, “if we call [organismsl an analogue of art, for in that case we think of an artist apart from nature” (Kant 1793, ¶65. Ak. V, p. 375). Nor does it help if we transfer the idea of an external artificer into the nascent organism itself, as if there were some crypto-intentional agent lurking within it that steers it toward what it will require down the road.
Transferring the notion of an artificer from outside to inside in this way merely adds mystification to an already inappropriate analogy. “The organization [of living thingsl,” Kant concludes, “infinitely surpasses our ability to exhibit anything similar through art” (Kant 1793, ¶68. Ak. V, p. 384).
The concept of a “self-formative force” refers explicitly to Blumenbach’s concept of “formative force” (“Bildungstrieb”), as we shall see. In the “Critique of Teleological Reason,” Kant attempts to clarify this idea by employing the related concept of “self-organization.” An organism, he says, “must be both an organized and a self-organizing being” (Kant 1793, ¶65. Ak V, p. 374, our italics). By this, Kant means three things. First, he means that organisms make from inorganic matter, rather than merely find, the organic materials out of which they are formed as organisms, such as tissue, blood, and bone (Kant 1793, ¶64. Ak. V, pp. 371-2). Second, the process of ontogeny requires the pattern of reciprocal causation we have already noted. The self-organizing aspect of this process can be seen empirically, Kant argues, by considering the compensatory growth of one feature when another has been stunted in the developmental process. “If birth defects occur, or deformities come about during growth, certain parts... form in an entirely new way, so as to preserve what is there and so produce an anomalous creature” (Kant 1793, ¶64. Ak. V, p. 372). Finally, organisms are said by Kant to be self-organizing in the sense in which a tree, for example, by reproducing itself, helps sustain the species-lineage that makes it possible (Kant 1793, ¶64. Ak. V, p. 371). Organisms are dissimilar to artefacts in this sense as well. They cannot be assembled from independent and exogenous parts, as artefacts can, but must come into existence by the reproductive activity of progenitors of the same lineage of which they are members, or even parts.
Nevertheless, Kant’s denial of the stronger claim that organisms are artefacts, as well as of the weaker claim that organisms are in any important way like artefacts, did not prevent him from asserting that any hope we may have of alleviating the tension between finality and efficient causes in judging living beings does depend on our bringing to these puzzling beings a model of means-end causality that is drawn from our own experience as agents, particularly in contexts of making or production. Kant calls this model of causality “purposiveness.” The goal-orientation and part-whole “fittingness” that allows us to identify and begin to investigate organisms means that they are “to be covered by a concept or idea that must determine a priori everything that the
thing is to contain” (Kant 1793, ¶65. Ak. V, p. 373).  Unless this were so, we could have scarcely any awareness at all of these curious entities, which seem different from inanimate objects, let alone any clue to explaining how they work. But where, other than our own experience as designers, makers, and choosers, could we ever acquire the idea of a being in which “the concept of the whole determines the form of the parts” as fittingly, usefully, efficiently, even beautifully as we clearly observe in the case in organisms? (Kant 1936/1993. Ak. 22, p. 283). The very idea of an organism, accordingly, demands that, as in an artefact, an overall intention (Absicht) must be presumed to determine each constructive step, even though in organisms the parts come to be and come to be related to one another in ways that defy our experience with artefacts. “We cannot even think of them as organized things without also thinking that they were produced intentionally” (Kant 1793, ¶76. Ak. V, p. 398).
Kant’s way of putting this crucial point is to say that even if we do not and cannot know that organisms are “purposes” - objects brought into existence in accord with an antecedent concept - we must admit that they cannot appear to us except as “purposive” (zweckmassig). “We call objects, states of mind, or acts purposive,” he writes, “even if their possibility does not necessarily presuppose the presentation of a purpose... All that is required is that we be unable to understand their possibility except on the assumption that [they] were produced according to design” (Kant 1793, ¶10. Ak V, p. 220).  Thus Kant argues that organisms are “explicable” (erklärbar) and “cognizable” (erkennbar) only “by using the idea of purposes as a principle” (Kant 1793, ¶68. Ak. V, p. 383). We do this by using the teleological maxim, “Nature
4. The term used here is “Zweck,” which means “end” or “finality” as well as “purpose.” Generally, we prefer to accept the conventional translation of the term as “purpose” to stress the semantic tie that Kant (unwisely in view of cybernetics) makes between the notion of an end-directed system and the structure of an intentional, means-end action.
5. The “Critique of Teleological Judgment” is preceded in the Critique of Judgment by a “Critique of Aesthetic Judgment.” In addition to being a contribution to the growing discipline of aesthetics in eighteenth-century thought, Kant’s analysis of judgments of the beautiful seems meant minimally to establish, in the context of the overall aims of the Critique of Judgment, that a distinction can be drawn between purposiveness and purposes. In aesthetic judgments, and especially in judgments of the beautiful, purposiveness is ascribed without reference to purposes, and indeed in their complete absence. This prepares the way for Kant’s ascription of purposiveness to living things, where purposes and purposiveness do not appear quite as separable.
does nothing in vain,” to discover the suitability or fitness of each of the parts to bring about and sustain the whole, and vice versa. To the extent that organisms are identifiable and explicable only by means of this model - an epistemological necessity, even if not an ontological one - Kant is willing to say that “the [living] thing is a purpose [or end, Zweckl, even if it is not an artefact” (Kant 1793, ¶65. Ak. V, p. 373).
We come now to a key point. All along, Kant has been asking in his “Critique of Teleological Judgment,” not whether there are any living things, but whether there are any natural purposes (Naturzwecke). It is now clear that the answer is yes. Organisms are natural purposes. Kant says they are purposes in order to register our utter dependence in “grasping” them and “cognizing” them on an analogy between organisms and our own means-end reasoning activity, as we have just noted. The stress here is epistemic. Still, organisms are not merely purposes for Kant. They are, ontologically considered, natural purposes - the only case, in fact, in which Kant can finds natural purposes among objects in the world.
This characterization supports two important inferences. First, it allows Kant to invoke the nature-artefact contrast in order to stress that, as self-organizing and self-formative wholes, whose parts appear to be reciprocally means and ends of one another, organisms are not, and are not very much like, machines, and so cannot straightforwardly be assimilated to the design model or the physico-theological inference. Second, Kant’s identification of organisms as natural purposes suggests that since organisms are more fully embedded in the natural order than, say, rational agents, there is every reason to try as hard as possible to bring natural laws to bear on explaining them. Characterizing organisms as “natural purposes,” and not as artefacts, thus sounds a directive as well as a jarring, even potentially self-contradictory, note. “Nature,” writes Kant in the Prolegomena to Any Future Metaphysics, “is the existence of things so far as it is determined according to universal laws” (Kant 1783, ¶14. Ak. IV, p. 294; see also Kant 1786b, “Preface.” Ak IV, p. 467-8; and Kant 1788. Ak VIII, p. 159). The laws of nature by means of which we have scientific knowledge are mechanistic in the sense that “we regard a real whole of nature only as the joint effect of the motive forces of the parts” (Kant 1793, ¶77. Ak V, pp. 407-8). As natural beings, accordingly, organisms must clearly be under the sway of natural laws - presumably mechanistic laws - of various sorts. Yet Kant’s point in characterizing organisms as natural purposes was to deny
that as wholes they are under the control of parts in the way mechanisms are.
Oddly enough, it was probably this very tension that attracted Kant to his conception of organisms as “natural purposes.” The reference to nature in this formula has the effect of tethering our inherent tendency to move illegitimately from seeing organisms as purposive beings to dogmatically and prematurely bringing God into the picture, a procedure that Kant believes will hinder rather than advance our pursuit of knowledge. On the one hand, the design inference puts too positive a spin on what the analogy with artefacts can achieve. On the other, it obscures the ways in which organisms are in fact deeply embedded within a system of natural laws. “The expression ‘purpose of nature,’ Kant says,
keeps us from mingling natural science, and the occasion it gives us to judge its objects teleologically, with... a theological derivation [of these objects]... We must carefully and modestly restrict ourselves to an expression that says no more than what we know, namely “natural purpose.” For even before we inquire into the cause of nature itself, we find that nature contains such products and engages in their production. They are produced there in accord with known empirical laws; it is in terms of these laws that natural science must judge its objects... Hence the causality in terms of the rule of purposes... must be sought within natural science. Natural science must not leap over its boundary in order to absorb... something to whose concept no experience whatever can be adequate.
Kant 1793, ¶68. Ak V, pp. 381-382
By following Kant’s line of reasoning in the “Critique of Teleological Reason,” we have come to the following realization: No matter how much of an enthusiast he was for assimilating systematic and secure knowledge to the methods and concepts of the natural sciences, it is hard to avoid the impression that Kant was delighted to find within the sphere of nature itself some objects - organisms, considered as “natural purposes” - that he was confident would forever resist full subsumption under mechanical laws. There are, it would seem, two main reasons for this apparent delight in limiting the claims of physical science on living beings.
First, the general project of the Critique of Judgment is to find a way of mediating the tension between theoretical reason, which places us fully within a deterministic causal order and the freedom and responsibility that practical reason confers on us (Kant 1793, ¶9. Ak V, p. 196). The clash between these competing perspectives is at its most intense when we reflect that in a deterministic world, our virtuous actions cannot reliably be counted on to grant us happiness - a thought that is only partially alleviated by the faith that a congruence of virtue and happiness awaits us in a later life.  The mere possibility that nature itself contains, and indeed produces, beings that are as thoroughly purposive as organisms must appear to us as a sign, then, that the world as a whole might conceivably be arranged with a view to making it possible for us to realize our ends as rational beings. Kant writes:
Organized beings are the only beings in nature that, even when considered by themselves and apart from any relation to other things, must still be thought of as possible only as purposes of nature. It is these beings, therefore, which first give objective reality to the concept of a purpose that is a purpose of nature rather than a practical purpose, and which therefore gives natural science the basis for a teleology.
Kant 1793, ¶65. Ak V, pp. 375-6
Continually encountering, as we do, natural entities that cannot be grasped without construing them as purposes, Kant thinks we are entitled, even compelled, to reconsider even the idea of universal teleology, a notion that the advance of modern science had been discounting for some time:
Once we have discovered that nature is able to make products that can be thought of only in terms of the concept of final causes, we... may thereupon judge its products as belonging to a system of purposes - even
6. Since early in the nineteenth century, Kant has been interpreted as putting a premium on reconciling virtue and happiness in the afterlife. It was Kant’s idealist successors, especially Schelling and Hegel, who revolted against this inference, which was insisted upon by their theological teachers, demanding instead that social and political changes that can either be anticipated or precipitated tend to create a world in which humans can be at home - that is, in which moral activity is not systematically at odds with happiness. In recent years, scholarship has suggested that Kant himself might not have as been much of a “Kantian” in this respect as he has been made out to be. This interpretation gives new salience to the Critique of Judgment’s project of reconciling theoretical and practical reason by finding a way to see mankind as the ultimate end of nature. See Guyer (2001).
if they... do not require us [in accounting for] their possibility to look for a different purpose beyond the mechanism of blind efficient causes
Kant 1793, ¶7. Ak V, pp. 380-1, our italics
The claim is no longer that in order to support the moral destiny of humans, the world must be purposive all the way up and all the way down, as it was in the days of naive universal teleology. It is merely that even as more and more aspects of the world system come to be explained in terms of natural laws, the overall arrangement of that system can still be thought of without contradiction, and indeed productively, as having been hierarchically arranged to make the ends of beings such as ourselves realizable. Viewed in this light, “the whole of nature appears as a system in accordance with the rule of purposes, to which all mechanism of nature. . . is now subordinated (at least for examining the appearances of nature by means of it)” (Kant 1793, ¶67. Ak V, pp. 378-9).
The purposive systematicity of the world, Kant assures us, is not a known, or a knowable, fact. Far from it. For the idea of universal teleology, or even the more basic notion of organisms as natural purposes, cannot yield what Kant, in his technical vocabulary, calls “determinate” judgments.  In other words, we cannot employ teleology to subsume particular cases or facts deductively under general laws, thereby generating scientific knowledge, and we certainly cannot make a valid inference, on the basis of end-oriented cases, facts, and patterns, to a divine maker. We cannot do these things because teleology of a determinate sort simply does not fall under the sway of the a priori concepts that, in unifying our sensory intuitions, also give us knowable objects and natural laws. Instead, the teleological idea is merely a heuristic, or what Kant calls a “regulative,” idea: a construct that can help us think productively about particular objects and relations that we encounter, but that we cannot subsume under law-covered explanation. Against this background, Kant seems to have thought that the Aristotelian maxim, “nature does nothing in vain,” as well as maxims such as “nature takes the shortest path (lex parsimoniae),” “makes no leaps (lex continni in natura),” and is “parsimonious” in the sense of Ockham’s razor (principia praeter necessitatem non sunt multiplicanda) ,  might pay their way
7. For Kant, the term “judgment” refers to the cognitive ability to relate the particular to the universal. In determinate judgments, the direction is from securely established generalizations to particulars subsumed under them. In reflective judgments, particulars are given and a covering generalization is sought or entertained.
8. On these maxims, see Kant 1793, “Introduction,” ¶V. Ak V, p. 182.
by helping us discern empirical regularities that these guiding ideas happen to bring into view, as well as providing us with resources for drawing the various dimensions of our experience together into a whole, which is the overall aim of the Critique of Judgment (Kant 1793, Introduction ¶V. Ak V, p. 182).
In making this claim about the regulative use of teleology, Kant is presupposing the important distinction he makes in “The Critique of Teleological Judgment” between relative, or extrinsic, and intrinsic forms of teleological judging (Kant 1793, ¶63. A V, p 367; see ¶82, p. 425). Living as he did on the flat littoral of the Baltic Sea, Kant was aware of benign ecological arrangements that allow humans to make a living from the fertile soil and nutrient-laden waters that are washed down - by rain-swollen rivers from mineral-rich mountains. Did these arrangements come into existence because they were, or would in the future be, useful to human beings dwelling, farming, and fishing on those alluvial floodplains and coastal waters? (1793, ¶63. Ak V, p. 367). Even more fundamentally, “Did Nature have a purpose in depositing ancient layers of sand, namely, to make spruce forests possible there?” (Kant 1793, ¶63. Ak V, p. 367).
There are dangers in giving a dogmatic answer either way. If we insist on an unqualified affirmation of the teleological perspective, we can easily end up canceling the scientific knowledge we actually have, or can plausibly acquire, about how the sand, the sea, the mountains, and everything else in such a sequence actually have come into existence, and how these elements are causally related to one another in a straightforward, when-then way (Kant 1793, ¶63. Ak V, pp. 367-8). For if we persist in pursuing a universal teleology in the full sense, we must be prepared to conclude that, as in an organism or natural purpose “each intermediate link” - the sand, the sea, the mountains, the trees, and anything else - “must be regarded as itself a purpose and its proximate cause as a means to it” (Kant 1793, ¶63. Ak V, p. 367). But in this case, we must drift either toward the absurd “Panglossian,” if not Leibnizean, results that Voltaire satirized in Candide, or else fall into the error that Kant associates with Herder, who, in thinking of the whole world as a single purposive organic whole, ends up denying that matter itself is inherently inert, and so threatens the fundamental principles of physics that have been so carefully established.
On the other hand, an unqualified negative is equally suspect. In eschewing all cognition-oriented employment whatsoever of the teleological maxim, we fail to recognize the extent to which the world must
appear to us to be an orderly, law-governed place, in which various sorts of beings are in fact systematically useful to one another even to the point of being necessary for each others’ existence. Having missed the evidence for these relationships, we might also forfeit the opportunity to track down the when-then patterns of cause and effect that undergird, say, the ecology of the Baltic littoral. Such explanations are devoutly to be wished for. Whenever they are possible, in fact, mechanistic explanations should be allowed to trump appeals to purposive causation. For they alone conform to the conditions necessary for scientific knowledge of objects, toward which our reason pushes us. (While Kant seems willing to envision at least two kinds of causality, purposive and mechanistic, he certainly appears to have limited the term “explanation” exclusively to the latter. )
We must, then, make use of the teleological idea in our pursuit of scientific explanations. To use the idea productively, however, it does not seem to be enough for Kant that we merely entertain it as a convenient fiction that might help us find mechanistic explanations. Kant’s view is not that “pragmatic.”  The appearance in our midst of beings that we must construe as thoroughly purposive - organisms - gives us just enough justification to at least think of the world as conceivably having been put together in such a way that mechanical laws serve as the means for constructing a cosmic hierarchy in which the well-being of its inhabitants, and especially of human beings, can best be realized. With the aid of the teleological idea, we might eventually unravel ecological systems that we discover into determinate chains of cause and effect. But if organisms - natural purposes - were ever broken down into
10. Here we are skirting discussions in recent secondary literature about what precisely Kant meant by mechanism, whether he meant different things at different times, and whether at some stage he denied that mechanism is the only kind of causality that is consistent with the Second Analogy. Peter McLaughlin asserts that Kant, as he drifts under the influence of chemistry toward the Opus postumum, moves away from restricting empirical causality to mechanism, but that he never denies that, given our conceptual scheme, explanation is restricted to mechanism (McLaughlin 1990). Zumbach makes a case that Kant revised his view of explanation as well to accommodate what we now call teleological explanation (Zumbach 1984). Zumbach’s intriguing suggestion seems to us to read into Kant a late nineteenth-century neo-Kantian distinction between explanation and understanding.
11. Kant’s regulative ideas, when applied to human affairs, are called pragmatic ideas. Kant’s concept of a pragmatic idea may have inspired pragmatism. But he was no pragmatist. Ideas cannot be so arbitrary that they can be thrown away without threatening the knowledge acquired and stored by their means.
linear, deterministic chains of cause and effect among merely physical or chemical components (a contrary-to-fact conditional), Kant’s justification for using the teleological conception for considering the bare possibility that the world is, in the final analysis, oriented toward purposes would, in his own view, crumble. Ironically, then, if we are to keep learning about nature, it is important that we rule out from the beginning a purely mechanistic view of organisms like that of Descartes, or even Buffon, and that we refrain from hoping or expecting that “perhaps some day another Newton might arise who would explain to us, in terms of natural laws unordered by any intention, how even a mere blade of grass is produced” (Kant 1793, ¶75. Ak V, p. 400).
Kant was aware that the history of thought is full of mistakes in both an overly teleological and an anti-teleological direction. The anti-teleological mistakes are of two sorts. In order to preserve the status of organisms as natural beings, the atomists, in one way, and Spinoza, in another, mismanage the concept of nature by holding that it contains no purposive items at all. Democritus and Epicurus take organisms to have come into existence by way of an improbable confluence of chance collisions. Kant regards this view as “manifestly absurd” (Kant 1793, ¶72 Ak V, p. 391). Far from offering us a path toward scientific explanation, these thinkers minimize, overlook, or even negate the high degree of lawful order that we find not only in organic development, but in experience as a whole. Spinoza, for his part, attempts to preserve the lawfulness of nature. But in his fatalism, he regards nature as a deterministically self-unfolding, self-positing system that moves inexorably in accordance with underlying principles that extend even to human actions. Where atomism exaggerates the role of chance in experience, fatalism underestimates it, making our experience as agents impossible. These are the two ways in which the teleological aspect of experience and its objects are unreasonably denied.
Other systems, meanwhile, attempt to preserve the purposive element in “natural purposes,” but only by subverting the natural element. One of these we have already mentioned: Herder’s attempt to resolve the conflict between mechanism and teleology by extending intrinsic teleology to nature as a whole, thereby obliterating the distinction Kant had drawn between them. Kant’s opposition to this “hylozooic” interpretation (which he saw springing up in the proto-Romantic youth culture of his later decades) knew few bounds (Kant 1793, ¶¶72-3, pp. 392-4.) 
11. On Herder and Kant, see Zammito 1992; 2002.
His horror at the idea was driven, in the first instance, by his conviction that such a view was no less atheistic than Democritean chance or Spinozistic fatalism. In the eighteenth century, in fact, atheistic materialism was more closely associated with the notion of living matter than with atomism. Hence, Kant was incensed at Herder’s claim that “perhaps even in so-called dead things one and the same disposition (Anlage) for organization, only infinitely cruder and more muddled, might preside” (Kant 1785b. Ak VIII, p. 62). Kant’s opposition was supported by his settled conviction that matter is by definition inert, capable only of inertial motion, whereas “hylozooists endow matter as mere matter with a property of life” (Kant 1793, ¶¶73, 65. Ak V, pp. 394, 374). To follow young, proto-Romantic intellectuals down this path would be to surrender much of the ground that the Critique of Pure Reason and the Metaphysical Principles of Natural Science had won.
The fourth system that Kant criticizes is “theism.” Unlike the other three errors we have mentioned, Kant does not regard theism as incoherent or ideologically suspect. On the contrary, as we have mentioned, he respected this position for its utility in giving aid and comfort to the moral point of view. However, where Herder’s hylozooism overextends the concept of internal teleology, theism, by way of its uncritical reliance on an analogy with artefacts, tends to introject external teleology into the very concept, and metaphysical status, of an organism. In failing to see the ways in which organisms are self-formative, self-organizing systems, in which every part is reciprocally cause and effect of every other, theism “does not make [theirl manner of production a whit more comprehensible to us” (1793, ¶73. Ak V, p. 395). Indeed, as Hume pointed out in his posthumously published Dialogues Concerning Natural Religion (Hume 1779), theism, even if it were able to shed detailed light on how living things actually came into existence and on how they work, would still have to prove decisively that the apparent end-directedness of organisms “could not possibly result from a mere mechanism of nature” (1793, ¶73. Ak V, p. 395). As Hume argued, this is a tall order. It is difficult to see how theism could rule out the possibility of mechanism without begging the question not only against mechanism, but against a host of other possible, if unknown, accounts of the origin of living things - all of them probably beyond the range of knowledge.
It is nonetheless fairly easy to understand why, in spite of their various deficiencies, mistakes of all four sorts have been made, and doubtless will continue to be made. In being compelled by our cognitive weaknesses
to construe organisms as natural purposes, we are inevitably pulled in two contrary directions, a mechanical and a teleological. How far in each direction can we go? If teleological ideas enable us to identity ecological patterns that we can trace to mechanical causes, who is to say that the same heuristic process cannot productively be projected into the interior of the living being itself? If so, might we not hope to find mechanisms that support embryological, physiological, and other organic functions in the same way that alluvial flows support plant, animal, and human life? At the limit of such inquiry, might we not stumble upon the point at which inert matter is worked up into living tissue in a mechanistic way? How, then, can we rule out the very possibility that a living thing might show itself to be “an effect of the motive forces of the parts”?
As we have already noted, the possibility that we might be able to penetrate the organism to this degree - the possibility, in other words, that there might be a “Newton of a blade of grass” - would be fatal to Kant’s delicately poised regulative use of the teleological idea. Our justification for using that idea was said to be our experience of organisms as natural purposes. Finding that organisms are merely the result of the forces that govern their separate parts would undermine that support, plunging us headlong into materialism of one sort or another, and hence into various outrageous claims. It might seem preferable to tilt in the other, teleological, direction, as hylozooists and theists are said to do. But this, too, has its dangers. In this case, our anomalous experience of organisms as natural purposes will be interpreted as affording us a glimpse of how in organisms “the presentation of a whole also contains the basis that makes the connection of the parts possible” (Kant 1793, ¶77. Ak V, p. 407). Enjoying this kind of insight into the underlying nature of organisms depends, however, on affirming something that the Critique of Pure Reason categorically denies - that we have intuitive knowledge, however unclear, of a purposiveness that constitutively pervades organisms and the world order generally (Kant 1793, ¶77. Ak V, p. 407). For Kant, our intuitions are restricted to unorganized and disunified flows of sensory data, and to our pure, contentless intuitions of space and time. Our scientific knowledge is wholly constructed, with the help of the categories, from these forms of intuition. In moving from categories to laws, and from laws to explanations, we must presume that wholes are determined by forces operating at the level of parts. We have no intuitions, no matter how fleeting or feeble, of systems in which the moment we grasp the whole, we “intuit” how all the parts
are brought into being and disposed. We have no intuitions, that is to say, of purposive systems.
In this way, or something like it, there arises what Kant calls “The Antinomy of Teleological Judgment.” For Kant, an antinomy is an apparent contradiction: not a stupid mistake, to be sure, but instead a conflict that is subtle enough to arise over and over again by way of Reason’s self-generated demand for answers to questions that go beyond “the bounds of sense” (Kant, 1787 FBI, p. 490, in trans. Kant 1929, p. 422). The four cosmological antinomies that Kant discusses in the first Critique illustrate the general idea of an antinomy. Kant summarizes them as follows:
Whether the world exists from eternity or has a beginning; whether cosmical space is filled with beings to infinitude, or is enclosed within certain limits; whether anything in the world is simple, or everything [is] infinitely divisible; whether there is generation and production through freedom, or whether everything depends on the chain of events in the natural order; and finally whether there exists any being completely unconditioned and necessary in itself, or whether everything is conditioned... and contingent.
1787 [B], p. 509, in trans. Kant 1929, pp. 433-4
Convincing arguments from seemingly secure first principles can be given for both alternatives in every case. Nonetheless, antinomies can be resolved either by bringing to light a hidden assumption under which both alternatives labor, or by shifting the ground from which the apparent contradiction arises. When this is done, the apparent contradictions turn out either to be contraries, in the logician’s sense, which might both be false, or subcontraries, which might both be true. The apparent contradiction in the second cosmological antinomy is resolved, for example, in the first way; it could turn out that the world is neither finite nor infinite. In the case of the third antinomy, on the other hand, both alternatives might be true. The same events might be conditioned and free from different points of view: what, from a theoretical, scientific point of view, is determined by laws, including psychological laws, is revealed as free, from a practical point of view.
The systematic errors about organisms that Kant criticizes in the “Critique of Teleological Judgment” might be viewed as analogous to the transcendental illusions that arise when our speculations about cosmology push us beyond the sphere where our logic-chopping, concept-dependent minds can use secure principles to sort out sensory intuitions
into true and false statements.  So viewed, errors about natural purposes can be represented as arising from the apparent contradiction between the following statements:
(1) All production of material things and their forms is possible in terms of merely mechanical laws.
(2) Some production of material things is impossible in terms of merely mechanical laws.
Kant 1793, ¶70. Ak V, p. 387
In view of the seeming intractability of this contradiction, it is easy to see why it might invite such desperate, and self-undermining, measures as it receives from the four contending parties that Kant has criticized. Democritus and Spinoza might be construed as attempting to secure (1) by denying (2), Herder and the theists as preserving (2) by denying (1). None of the advocates of the systems Kant criticizes seems to have realized, however, that organisms are not presented to us by way of seemingly secure lawful generalizations that conflict with one another when they are extended beyond the bounds of sense. Seen through the somewhat problematic lens of the notion of a “natural purpose,” they are instead presented to us as puzzling particulars, whose governing laws we do not (yet) know, but which we hope to learn more about. That is to say, organisms are presented to our reflective, not to our determinate, judgment. Hence, although there may well be an antinomy of some sort between (1) and (2), it is not an antinomy of teleological judgment.  Given the way organisms are “grasped” by us and “cognizable” by us, the antinomy of teleological judgment holds instead between the
12. Support for this way of explicating the antinomy of judgment as like a transcendental illusion is provided by Kant’s claim that “going to the extreme of explaining everything merely mechanically must make reason fantasize and wander among chimeras of natural powers that are quite inconceivable, just as much as a merely teleological kind of explanation that takes no account whatever of the mechanism of nature makes reason rave” (1793, ¶78. Ak V, p.411).
13. Kant says an antinomy results from (1) and (2), yet “this antinomy would not be one of judgment, but a conflict in the legislation of reason” (1793, ¶70. Ak V, p. 387). It is not quite clear what he might have meant, but it is clear that (1) and (2) do not constitute the antinomy of teleological judgment.
(3) All products of material nature and their forms must be judged to be possible in terms of merely mechanical laws
(4) Some products of material nature cannot be judged to be possible in terms of merely mechanical laws. (Judging them requires a quite different causal law, namely that of final causes.)
Kant 1793, ¶70. Ak V, p. 387
From the reflective point of view, (3) and (4) are consistent after all. They are merely maxims that pull us in opposed directions when we think about organisms or ecological systems, and can even be made to do so productively if we manage them right. The first tells us that “we ought always to reflect in terms of the principle of mere mechanism of nature, and hence ought to investigate this principle as far as we can, because unless we presuppose it in our investigation of nature, we can have no cognition of nature at all in the proper sense of the term” (Kant 1793, ¶70. Ak V, p. 387). (We can have no cognition in the proper sense because explanation, it would seem, depends on mechanical causality.) “None of this,” however, Kant goes on,
goes against the second maxim - that... in dealing with certain natural forms (and, on their prompting, even with all of nature) we should probe these in terms of... the principle of final causes. For this leaves it undecided whether in the inner basis of nature itself, which we do not know, the physical-mechanical connection and the connection in terms of purposes may not, in the same things, be linked in one principle.
Kant 1793, ¶70. Ak V, p. 388
Many commentators have claimed that the antinomy is resolved the moment we recognize the difference between reflective and determinate judgment.  However, this is unlikely. First, Kant himself says that
14. This interpretation can be found in Adickes 1924-5, and, partly following Adickes, in Cassirer 1938, p. 341. McFarland 1970 finds the same inadequate interpretation in several other Kant scholars, and, in a qualified way, in Reinhard Low 1980. Low thinks that the chapters in which Kant treats the distinction between regulative and determinate judgment as a “preliminary” resolution of the antinomy, especially ¶¶70-71, do have the consequence of reducing both mechanism and teleology to regulative status. For Low, however, these earlier chapters were actually written after the substantive resolution presented in ¶¶77-78, which should be regarded as definitive. For Low, the considered doctrine of ¶¶77-78 has the consequence of making teleological judgment intimate the actual constitution of organisms as things in themselves, anticipating a [turn toward a constitutively teleological view of the cosmos in the Opus Postumum. We are inclined to deny that the resolution of the antinomy has the consequence of making mechanistic judgments merely regulative, and hence to think that the doctrine of ¶¶77-78 is not as different from that of sections ¶¶70-71 as Low takes it to be.]
HHC: [bracketed] displayed on page 114 or original.
the move from (1)-(2) to (3)-(4) is only a “preliminary solution to the antinomy” - and he takes seven more chapters to reveal, in ¶78, “How the principle of universal mechanism of matter and the teleological principle can be reconciled.”  Second, it has long been noted that merely shifting from a determinate to a reflective point of view for both sides of the antinomy has the effect of demoting (3) to a regulative principle - in effect retracting, or at the very least weakening, what Kant had said about the constitutive force of mechanistic causality and the determinate status of mechanistic explanation both in the Critique of Pure Reason and at other places in the Critique of Judgment itself. It seems clear then, that the shift from a determinate to a reflective perspective is a necessary, but not yet a sufficient, condition for resolving the antinomy of teleological judgment. Something else needs to be done.
What needs to be done in addition is to demonstrate how the necessity under which we labor of “grasping” and “cognizing” organisms through reflective rather than determinate judgment implies, in and of itself, that “the same product and its possibility” appearing to us in two non-coincident causal frames - “the physical-mechanical connection and the connection in terms of purposes” - might conceivably be fused into a unity when organisms are considered as things in themselves. Kant’s claim is not that from a “supersensible” perspective - the perspective of “things in themselves” - mechanical causality and explanation are merely regulative, or, conversely, that the teleological way of judging is uniquely true. It is that from this higher perspective, they are two approaches to what is substantively the same thing.
To make this argument, Kant - at least the Kant of the Critique of Judgment - begins by agreeing with the maxim, individuum est ineffabile. Even in determinate judgments, which subsume particulars under laws sufficiently well to produce scientific knowledge, the sensuously given particular can contain something “contingent,” as Kant puts it, with respect to those laws, something that eludes their determination. “For the universal supplied by our (human) understanding does not determine the particular; even if different things agree in a common characteristic, the variety of ways in which they may come before our
15. These claims are embedded into the titles Kant provided for ¶¶71 and 78, respectively.
perception is contingent” (Kant 1793, ¶77. Ak V, p. 406).  This is more obviously the case when it comes to organisms. Their “contingency,” or perhaps indeterminacy, with respect to general laws is apparent in our constitutional inability to bring the mechanistic and the purposive frames together. This might not be so if we had the intuitive knowledge that God presumably enjoys - a “power of spontaneity,” Kant calls it, in which the cognizing of an object differs not a whit from its creation (Kant 1793, ¶77. Ak V, p. 406). If our minds were intuitive in this sense, our awareness of an organism would not only bring it into being, but would, for that very reason, contain and specify all its parts as expressions of a whole. As things are, however, the only kind of intuition we possess (other than that of the pure forms of space and time) is our capacity to receive sensory data, which the discursive, logic-chopping, non-intuitive apparatus of our mind works up into categorized objects. Indeed, our only clue to the way in which the parts of an organism are a priori integrated into, and expressive of, the whole is the way in which, as practical and productive agents, we subordinate means to ends. This may not get us very far. But the very fact that we must identify and learn about organisms by reflecting on them as “natural purposes,” and must regard mechanistic laws as subserving those purposes in the way in which means serve the ends of an agent, presupposes that “in nature’s supersensible substrate,” mechanism and teleology might be fused. For in using one maxim to compensate for the deficiencies of the other, we must be presupposing the possibility of a deeper unity in which both frames are brought together to an intuitive intelligence.
It is easy to make either too little or too much of this argument. The appeal to the supersensible ground of the possibility of organisms is more than a mere logical possibility. It is a presupposition of our way of grasping and learning about living things. It falls short, however, of assertorically affirming such a transcendent unity. For it makes no claim that stands independently of our learning procedures. Nonetheless, the appeal to the possible unity of the teleological and the mechanistic in the supersensible constitution of organisms is strong enough to prevent Kant from defensively setting a limit beyond which the mechanistic notion cannot go, for fear that it would go too far. “It is quite undetermined,” he writes, “and for our reason forever undeterminable, how much the mechanism of nature does as a means toward each final intention in
16. We leave aside the issue of whether the claim that all human cognition contains traces of the “contingent” is consistent with the doctrine of the first Critique.
nature” (Kant 1793, ¶78. Ak V, p. 415). We can, indeed should, use mechanism, therefore, to penetrate as far as we can into the interior structure of organisms. We may do so, moreover, with little or no fear that we will end up as materialists. For the objects that we investigate in this way are given to us by means of an idea, “natural purpose,” which by itself excludes the possibility of successful reduction on pain of losing our grasp on the entities we are talking about. From this perspective, proper method in biological inquiry is to manage our seemingly opposed maxims in such a way that teleological reasoning is used to identify biological phenomena and mechanism is used to explain the means by which they are realized. The explanatory work will never eliminate the referential and descriptive.
The influence of contemporary biological theorizing on Kant’s “Critique of Teleological Reason” was not restricted to his views about the epigenesis-preformation issue. In fact, the interpretive framework set forth in this text was in part suggested by Kant’s own interventions, over a period of several decades, into issues about the unity of the human species and the reality of different races that were being debated in the learned journals of the day, such as Berliner Monatschrift, Teutsche Merkur, and Allgemeine Literarische Zeitung. In fact, it is difficult to avoid the impression that the doctrines set forth in the “Critique of Teleological Judgment” were in part calculated to support Kant’s positions in these controversies. In these occasional writings, Kant proposes to solve, or at least mediate, what look like empirical quarrels by offering conceptual distinctions and methodological preachments. This work of conceptual clarification had an impact on the formulation of methodological aspects of several research programs within German biology, especially at the University of Göttingen. In almost no case, however - including that of Gottingen - did the reception of Kant’s interventions in biological and anthropological issues accord with his own intentions.
The origins of Kant’s interactions with the emerging community of German biologists lie in the fact that, beginning in 1757, he regularly taught a course in “physical geography.” Generally intended as a sort of “animal-vegetable-mineral” survey, Kant’s first prospectus for this course proposed as its primary topic a survey of man, with special reference to the biogeographical distribution of human types in accordance with climatic considerations. Both Kant’s prospectus and
his subsequently published writings on these topics show direct familiarity with the successively issued volumes of Buffon’s Histoire naturelle (1749-1767) - texts that had already stimulated Kant to develop his General Natural History and Theory of the Heavens (Aligemeine Naturgeschichte and Theorie des Himmels), in which he proposed in 1775 the “nebular” hypothesis about the formation of the solar system.
The influence of Buffon on Kant’s lectures on physical geography shows itself in his efforts in his course, as well as in four articles published over a period ranging from 1775 to 1788, to refute the notion that humans belong to different subspecies, or even species (Kant 1775 [rev.1777], 1785a, 1786a, 1788). This notion, “polygenism,” was from the start not devoid of racist implications, since Africans were invariably listed as departing furthest from the Caucasian norm, and as most likely to have had a different origin. It had begun to gain favor through use of the Linnaean classification scheme to pigeonhole different varieties of human beings as, in effect, subspecies. (Linnaeus himself recognized, at one point, an extinct or only residually extant subspecies of “troglodydes” in the genus Homo; see Sloan 1995, p. 12; and Chapter 11.) Kant, for his part, was deeply opposed to polygenism. In order to shore up the monogenetic alternative, he brought forward precisely what Linnaeus and others had denied - namely, Buffon’s genealogical species concept and its implied criterion of species membership, according to which interbreeding organisms whose own offspring are fertile are, by definition, members of the same species (see Chapter 3). By this standard, all human beings are members of the same kind (Art); even potentially there are no distinct kinds (Arten) of our species.
This is not to say that Kant agreed with Buffon on every point connected with his species concept (and he certainly disagreed with his conception of ontogeny). Kant was less opposed than Buffon, for example, to efforts at systematic classification like those of Linnaeus (on Buffon’s view, see Chapter 3). That is because, in one of his most important conceptual interventions, Kant distinguished, in a way that breaks with a tradition going back to Pliny, between natural history (Naturgeschichte) and natural description (Naturbeschreibung) (first articulated in Kant 1775. Ak II, p. 434). Natural description, Kant says, includes not only natural history in the traditional sense, but what Linneaus was up to - using the hierarchical categories that human language seems so spontaneously to generate to help us identify, compare, and otherwise orient ourselves in relation to the diversity of the living world. There is no harm in such an enterprise as long as it retains its fundamentally pragmatic
status as a resource for what Kant calls “Naturkunde,” or skill in dealing with nature, sometimes of a very humble, barnyard sort. The harm comes in conflating natural description with what Kant (not Pliny or Buffon) calls natural history. Natural history, practiced as he prescribes, extends to the origins and dispersion of biological kinds, and indeed to the history of the changing earth itself (Kant 1775. Ak II, p. 434). But Kant thinks very little of this curiosity can be satisfied (Kant 1775. Ak II, p. 434). True enough, illusory satisfactions abound. But they are often the result of confusing the notion of a lineage, on which natural history depends, and classificatory concepts that are appropriate solely to natural description.
The confusion arises, in the first instance, because species are both kinds (Arten) and lineages that are maintained by faithful chains of successful breeding (Gattungen).  The notion of lineages that breed true corresponds to the third sort of organic self-formation distinguished by Kant in the “Critique of Teleological Judgment,” according to which species “form themselves” by renewing their membership through generation. Difficulties arise, however, when both concepts - the classificatory and the generative - are conflated and then projected together up the scala naturae to classificatory genera and higher taxa. In this case, we imagine a “consanguinity which originates from one single generative mother-womb,” as Kant puts it in the “Critique of Teleological Judgment,” alluding to Herder’s Ideen (Ideas for a Critical History of Mankind) as well as to his own critical reviews of that work (Kant 1793, ¶80. Ak V, p. 419; see Kant 1785a. Ak VIII, p. 54). But in doing so, we imagine that our classificatory system (which is only a descriptive apparatus) is a mirror of the actual origin of taxonomic categories. That is to say, we imagine with Herder that kinds have origins in the same way that individuals do. In his reviews of Herder, Kant goes out of his way to say that he regards this as a “monstrous” idea (Kant 1785a. Ak VIII, p. 54). Given his distinction between natural history and description of nature, however, there really isn’t much to worry about. The idea is a category mistake, pure and simple.
Kant argues that the confusion between natural description and history is just as mischievous when we project it downward to differences within a species - only this time the problem lies, not in treating classificatory distinctions as historical, but in treating historical
17. The linguistic usage is a bit more complicated than this, but, on the whole, Gattung for Kant does not describe a classificatory genus, as in normal German, but has a historical ring that is intended to capture Buffon’s species concept.
distinctions as if they were distinctions in kind. This does not mean that Kant, in his Buffon-inspired monogenism, did not recognize differences within the single human species. Indeed, Kant distinguishes rather sharply between races (Rassen), varieties or local variations (besondere Schlage), and evanescent sports (Spielarten) (Kant 1775).  He defines races as “deviations that are constantly preserved over many generations and that come about as a consequence of migration” (Kant 1775. Ak II, p. 430). They are departures (Abartungen) from the same historical stem (Stamm), caused by the expression long ago of certain inherited predispositions in a given environment, and the suppression of others (Kant 1775. Ak 2, p. 434). Following Linnaeus, Kant says that there are four such stable races, recognized by the skin colors white, yellow, red, and black (Kant 1775. Ak 2, p. 432; see also Kant 1785a, 1786a, 1788). There are, of course, plenty of mixes between the races, Kant acknowledges, referring to mulattos and other intermediates, and plenty of locally different populations with this or that quirk. But Kant argues that the very fact that these differences are less constant than races means that they are not races. “A specific diet can surely produce a stock of humans,” he writes, “but the distinctions that identify such a stock as distinct quickly disappear when this stock is transplated to another place” (Kant 1775. Ak 2, p. 431; on this topic, also see Chapter 11). Because he gave the concept of race a clearer definition than others had done, and because he treated races as more stable than local variations, Kant can be linked with the articulation of the ill-fated concept of race (Bernasconi 2001). However, to be fair, it should be acknowledged that Kant worked out his concept of race in order to support a strong, potentially “anti-racist” version of monogenism. His key idea is that races are not classificatory subspecies, as they were, at least potentially, for Linnaeus. For Kant, the concept of race becomes useful for reflection on the genuine questions posed by natural history, such as questions about the geographical distribution of populations of the same species, only when one abandons the tyranny of the classificatory impulse.
Herder ranks as high among Kant’s opponents in issues about human diversity as he does in questions about the “monstrous” idea that new species come from a single “mother.” In one of his critical reviews
18. Kant makes the same distinctions in Kant 1785a, l786a, and 1788. We cite the 1775 essay (revised in 1777; the 1777 text is available in an English translation by J. M. Mikkleson in Bernasconi and Lott 1999, pp. 8-22. Where the 1777 text is identical to the 1775 version, we use Mikkleson’s translation.)
of Herder’s Ideen, Kant writes, “The division of the human species into races does not find favor with our author” (Kant 1785a. Ak VIII, p. 62). For Herder the distribution of human traits across different environments is simply a blur. It is more open to accidents than Kant would admit, and so, potentially, to changes in species boundaries. Not surprisingly, Herder entertains a very porous version of epigenesis; there are no species-specific germs (Keime). (Elsewhere, Kant remarks that Herder “rejects the system of evolution” (Kant 1785a. Ak VIII, p. 62); he means, of course, evolution in the original sense: the mechanical rolling out of the preformed organism.) For Kant, on the other hand, not only do germs keep ontogeny within species boundaries, but heritable predispositions (Anlagen) keep races adapted to specific environments. As he makes clear in his 1788 essay, the constancy of races also provides Kant with an opportunity to affirm his conviction of the indispensability of teleological reflection in biology and anthropology. In virtue of their rationality, bipedality, and related distinctions, human beings are far more physically mobile than other species. They need to become adapted to different climates. At the same time, human beings need to be protected from the potentially degenerative effects of novel climates (and, relatedly, of the fruits of intermarriage). The concept of race, and the notion of predispositions on which it depends, answers both needs. It keeps races inclined to remain in, or to find, congenial climates. At the same time, it limits the long-term, degenerative effects of whatever uncongenial conditions members of a specific race happen to find themselves in (Kant 1788. Ak VII, pp. 157-184).
Clearly, then, Kant’s concept of inherited predispositions was developed in conjunction with his views about race. As early as his 1775 essay “Of the Different Races of Mankind,” Kant was arguing that “numerous germs and predispositions must lie ready in human beings either to be developed or held back in such a way that we might be fitted to a particular place in the world” (Kant 1775. Ak II, p. 435). It was probably to sustain this analysis, too, that Kant entered into the thorny thickets of the preformation-epigenesis debate in the first place. Because his own notion of ontogeny was somewhat close to epigeneticism, Kant worried about the tendency of many epigeneticists, including Herder, to flirt with, or actually to embrace, materialism. To block this worry, he insisted that we must always begin our reflections (and end them as well) with the species as given, and with the individual organism as coming to be within a fixed context of species renewal, albeit with a wide variety of inborn variability that we must understand as purposively adapted to specific regions.
Against this background, it is not hard to understand why Kant thought he had found a kindred spirit in Johann Friedrich Blumenbach, a former student of Hailer’s and professor in the medical faculty at Gottingen. In his dissertation, De Generis Humani Nativa Varietate - which appeared later in the same year, 1775, as Kant’s “Of the Different Human Races” - Blumenbach had come down roughly on the same side as Kant in the matter of races (although in later editions he added a fifth race, Malays; see Chapter 11). Moreover, Blumenbach, like Kant, wanted to find an interpretation of epigenesis that preserved something of the stress on heredity that had been one-sidedly embodied in the preformationist tradition. It was in this connection that Kant felt the influence of Blumenbach most fully. The second edition of Blumenbach’s essay Über den Bildungstrieb (“On the Formative Impulse”) appeared in 1789. (The first edition was published in 1784.) In the first edition of the Critique of Judgment, Kant wrote, “No one has done more by way of proving this theory of epigenesis than Privy Councilor Blumenbach” (Kant 1793, ¶81. Ak V, p. 424). Indeed, Blumenbach’s proof of epigenesis makes use of the idea of a “formative impulse”; when Kant, in the “Critique of Teleological Judgment,” sees organisms as self-forming and self-organizing, he is alluding to Blumenbach’s Bildungstrieb. For both thinkers, too, this is a purely theoretical posit designed to account for facts such as these: Although the lopped-off parts of the polyp do indeed produce a whole organism, the organisms in question becomes smaller and smaller with each generation; and when, in the process of embryo-genesis, one part is deformed, another tends to be built up in compensation. This is not the old concept of “soul,” for which Kant had no use; soul would presumably not wax and wane as it does in generations of polyps. But neither is the Bildungstrieb anything as potentially capable of giving birth to kinds as well as individuals as Herder’s “genetic power” (genetische Kraft). Instead, for both Blumenbach and for Kant, the formative force is very much like Newton’s concept of gravity. It is known entirely by the effects it organizes and, to that extent, explains. Thus, with Herder probably in mind, Kant praises Blumenbach for
avoiding too rash a use of [his Bildungstrieb] ... He starts from organized matter. For he rightly declares it contrary to reason that crude matter on its own should have structured itself originally in terms of mechanical laws, that life could have sprung from the nature of what is lifeless, and that matter could have molded itself on its own into the form of a self-preserving purposiveness. Yet by appealing to this principle of original organization - a principle that is inscrutable to us - he leaves an indeterminable and yet unmistakeable share to natural mechanism. The ability
of matter in an organized body to take on this organization he calls a formative impulse (Bildungstrieb).
Kant 1793, ¶81. Ak V, p.424
Soon, Kant and Blumenbach formed a mutual admiration society. Blumenbach was impressed enough by Kant’s interpretation of his work to substitute in later editions of his Uber den Bildungstrieb (Blumenbach 1789), and his influential Handbuch der Naturgeschichte (Blumenbach 1802, 1st ed. 1797), Kant’s own formulations of what an organism is - something whose parts are reciprocally ends and means, and so forth - and of how we must investigate such things by combining mechanistic and teleological perspectives. In addition, Kant’s way of distinguishing between, while also coupling together, the notions of Keime and Anlagen was taken up by Blumenbach, and through him became diffused not only among other Gottingen biologists, but among nineteenth-century German embryologists generally.
Yet, from the outset, Kant was also at cross-purposes with Blumenbach and his colleagues (see Richards 2000). Blumenbach and other members of the multi-generation research tradition, which, since Hailer’s day, had been centered at Gottingen - arguably, the premier research university of its time - treated Kant’s notion of a regulative idea as little more than a heuristic method for arriving eventually at determinate explanations and actual causal mechanisms that might, in principle, displace the purposive conception of organisms from which we must admittedly begin. For Kant, on the other hand, we must remain forever wedded to the concept of the concrete “natural purposes” on which we reflect. This insistence is echoed in his resolution of the antinomy of teleological judgment. To attempt to reflect on matters outside the context of inquiry afforded by the concept “natural purpose” - on the origins of species, or on the point at which life meets its non-living substrate - is to lose touch with the very things we are talking about. In this way, Kant wanted to put some fundamental questions about natural history, in his sense of the phrase, off limits. A few years later, Kant’s students heard him remark: “The system of epigenesis does not explain the origin of the human body. Rather it says much more: that about this we know (wissen) nothing” (Kant 1794, Ak XXVIII, p. 761).
The sources of these dissonances between Kant and the Göttingen biologists depend, in the first instance, on the fact that Gottingen ranks high among sites where the modern discipline of biology was being formed. The term “biologia” goes back at least as far as the Wolffian
philosopher Michael Christoph Hanov’s Philosophia Naturalis, the third volume of which, published in 1766, names it as the part of physics that studies living things (Hanov 1766).  Lamarck, too, used the term; in an unpublished manuscript, written in French in 1800, he speculated about a discipline he called biologie, which would find the exact point at which the “corps vivant” was unified with the “corps brats” by “fluides invisibles” (Lamarck 1800, in Grassé 1944; see Caron 1988 for use of the term in eighteenth-century French medical schools, especially Montpellier). But it was Blumenbach’s student, the Gottingen professor Georg Reinhold Treviranus, who, in a six-volume text that began to appear in 1802, gave the term “Biologie” both wider currency and the sense that it has borne ever since. In the title of his Biologie, oder Philosophie der lebenden Natur fur Naturforscher und Artze, Treviranus named a comprehensive science that would gather together information about living things from a variety of special sciences, such as physiology, systematics, and comparative anatomy. Treviranus’s Biologie, it would seem, is no longer part of physics, as it still was for Hanov; physics is now restricted to the study of inanimate, or life-less, objects.
Treviranus’s work implies an optimistic view about the prospects of biological knowledge. This is so, in part, because, although they were influenced deeply by Buffon, the Gottingen biologists, led by Blumenbach and Treviranus, rejected Buffon’s skepticism about classification, and because they did not adopt Linneaus’s “artificial” system, which was open to potential doubts, but instead worked toward a more realistic, or natural, classificatory scheme based on a multiplicity of traits - the so-called complete “habitus” of an organism.  Treviranus hypothesized
19. On Hanov’s early use of the term biologia, we are indebted to the researches of Gary Hatfield (perusal communication). It would appear that Hanov (and perhaps others that we do not know about) was led to posit biologia as a science of living beings because he rejected the traditional Aristotelian way of dividing ensouled (psuchikos) nature into the nutritive-reproductive soul of plants and the sensitive-locomotive soul of animals. For Hanov, only animals have souls. Hence a new term must be found to jointly name the two sorts of living, as opposed to inanimate, beings.
20. Others who worked at or studied at Göttingen included Abraham Gotthelf Kaestner, who translated Buffon’s Histoire naturelle I-III and wrote notes on it that formed the basis of many of the questions pursued by Göttingen biologists David Sigmund Buttner, Christian Wilhelm Büttner (no relation), and Christian Gottlof Heyne, who did some of the work on polyps on which Blumenbach depended. Blumenbach’s students included some of the most illustrious biologists of their times. In addition to Treviranus, he trained the great Alexander von Humboldt, Carol Friedrich Kielmayer, Henrich Friedrich Link, and Johann Friedrich Menkel. See Lenoir 1981.
that questions falling within the scope of biology could be answered by using weighted statistical reports about the values of a number of variables that attend species, each defined by its total habitus: irritability (the capacity that, ever since Haller’s idea of how the sperm awakened the potentials in the egg, had, at Göttingen, marked the line between living and non-living), sensibility, secretion, propulsion, and reproductive prowess. (Statistics was a specialty of Göttingen research into government.) The underlying hope was that by following how individual elements of the complete portrait of a species change with changing environments, definitive answers would eventually be found to questions that Kant would have classified as natural history, even those bearing on the origin of kinds and the material conditions for ontogeny. A fundamental basis of this program was Blumenbach’s explicit rejection of the Great Chain of Being, and of classificatory practices dependent on it, such as those of Linnaeus. Close inspection of the habitus of species will reveal that there are in fact plenty of gaps in nature. “It is a great weakness;” Blumenbach wrote, “to see in such pictures [as the Great Chain] the Plan of Creation... on the ground that ‘nature makes no leap,’ as natural theologians are wont to say” (Blumenbach 1802, p. 8; quoted in Lenoir 1981, p. 132).
Given the nature and ambitions of their project, the Göttingen biologists, Blumenbach among them, could not help failing to walk steadily along the tightrope that Kant had strung between the determinate and regulative uses of teleology. Their project explicitly eschewed Kant’s distinction between natural history and description (as well as his retention of Linnaean classification in a descriptive context). Regulative ideas became methods for acquiring positive science, thus leaving open the possibility that somewhere down the road, the physical and chemical bases for the operation of the Bildungstrieb might be known.
These issues were brought out into the open when Georg Forster, who had studied at Göttingen, taught natural history at Kassel, and accompanied his naturalist father Johann on Captain Cook’s voyages to the South Seas, attacked Kant in Teutsche Merkur (which at this time was still hostile to Kant’s philosophy) (Forster 1786).  Natural history of the sort that Kant envisioned, Forster claimed, “would be a science for gods alone, not for men.” By this standard, “we would be incapable of demonstrating the genealogical tree of so much as one single
21. On this episode, see Sloan 1979, pp 131-137; Zammito 1992, pp. 207-213; Richards, 2000.
variety up to its genus if it does not arise nearly in front of our eyes” (Forster 1786; quoted by Kant in Kant 1788, p. 161). Chafing at this limitation on biological knowledge, Forster went on not only to deny Kant’s distinction between natural history and natural description, but to suggest (much to Kant’s annoyance) that Kant had made the former so difficult precisely because he was an obscurantist, who was more interested in theology than in biology.
Kant’s reply came in his 1788 essay On the Use of Teleological Principles in Philosophy, the text that forms the link between his writings on race and the “Critique of Teleological Judgment.” Kant complained that Forster simply had not understood his principles (Kant 1788. AK VIII, pp. 160-1). In response, Kant reiterated his distinction between natural description and natural history, and reinforced the limits he had drawn on scientific knowledge of the latter by insisting that teleological ideas (of a rather external sort) are indispensable in natural history and never rise to the level of determinate, scientific knowledge. It seems to us more likely, however, that Forster understood all too well what Kant was saying. Indeed, Kant’s relegation of classification conceived along Linnaean lines to mere natural description was reinforced, as he moved from this encounter to the Critique of Judgment, by a desire to damp down the expectation that much determinate knowledge would ever be found within the deeper recesses of natural history. Kant’s reiterated insistence that there will never be a “Newton of a blade of grass” should be read in the light of the controversy we have just reviewed.
Eventually, some support for Kant was proffered by a former Göttingen student, Christoph Girtanner, who had become acquainted with Kant’s philosophy through the work of Karl Reinhold, as well as from Girtanner’s personal friend Johann Jachmann, who had served as Kant’s secretary. It was Reinhold who first popularizied Kant’s philosophy in the suddenly pro-Kantian Teutsche Merkur. In an essay in that journal, Girtanner proposed to mediate the quarrel between Kant and the Gottingen biologists (Girtanner 1796). He agreed with Kant and Buffon on the species concept, as well as with Kant’s refinement of it, according to which a species consists of a number of distinctive germs with the potential for multiple realizations in different environments. But Girtanner concurred with his Göttingen teachers on classification by multiple characteristics. He proposed that the two ideas be put together in such a way that comparative knowledge of traits could lead to knowledge of dispositions, which in turn could lead to knowledge of
underlying species-specific germs.  But this interesting suggestion did little to bridge the gap. In some ways, it even combined the worst, not the best, of both sides. From the Göttingen perspective, it appeared to make positive biological knowledge harder to acquire than they would wish. For it would require them to find species-specific germs as grounds for their classifications. From the Königsberg perspective, Girtanner’s proposal flirted with promising more than human beings could ever deliver. Kant made no response.
Kant’s insistence on the limits of biological knowledge could only have become more pronounced if he had lived to see the full revolt against these limits by a group of young philosophers who had been brought to the University of Jena, in part through the influence of Goethe (who was looking for philosophical allies in his anti-Newtonian program for science). While still at the Tubingen seminary, Friedrich Schelling, the leader of the group, had made common cause with his classmates Georg W. F. Hegel and the poet Friedrich Hölderlin against Kant’s restrictions of the knowable to mere appearance. They wished to turn what Kant had consigned to regulative ideas into constitutive knowledge, dreaming of a physics that “takes wings” by means of an aesthetic intuition, which, in true Romantic style, makes the poet the true philosopher and legislator, and that can be converted into knowledge by means of projecting one’s imagination, feelings, and hypotheses onto a world presumed to be fitted to stimulating and satisfying all the human faculties.  With his careful delimitation of the spheres of aesthetic judgment from both moral duty and physical inquiry, this reinstallation of full-scale intellectual intuitionism, developed ironically enough out of his own aesthetics, was precisely what Kant had wanted to prevent in his opposition to Herder. It was, in some sense, the return of the repressed.
Schelling’s pitch was based on the proposed advantages of his “system” for unifying the unruly array of forces that, as we noted at the outset, had emerged in the final decades of the eighteenth century. Kant himself was preoccupied with these issues in his Opus Postumum. For his part, Schelling, having broken with Kant’s theory of matter as inert stuff by redefining that concept in terms of a dynamic polarity between
22. For an account of Girtanner’s intervention, see Sloan 1979, pp. 137-143.
23. This sentence summarizes the contents of the so-called “Oldest System-Program of German Idealism,” which was jointly written by Schelling, Hegel, and Holderlin in 1797. For an English translation, see Krell 1985.
attraction and repulsion, proposed to unify magnetism, electricity, and the vital force in terms of various proportions and disproportions between these dynamic poles (Schelling 1797/1988). This produced a marked shift toward Herder’s hylozooism and toward his idea that the unity and diversity of nature’s overall plan can be displayed as if all species sprang from a common, original matrix. This notion would have been disdained at Göttingen. It might even have constituted further evidence that Kant’s concept of organisms as natural purposes and his insistence on retaining Linnaean classification merely served as a screen for recovering the universal teleology of the old hierarchical scala naturae under the guise of a regulative idea. Exactly the opposite reaction, however, took place at Jena. A number of comparative anatomists at that university, especially Lorenz Oken and Carl Gustav Carus, were swept up by Schelling’s identity philosophy into enthusiasm for treating Herder’s thought experiment as what Kant would have classified as determinate claims about the world - not, to be sure, as an evolutionary hypothesis, or even a devolutionist one along lines proposed by Buffon, but as a restatement of the classical scala naturae. For Schelling and his idealist followers, the restoration of the ancient and medieval world-view through aesthetic, rather than intellectual, intuition would provide an essential foundation for a chain of argument establishing that human beings are at home in the world (and shouldn’t pine for another) because the world is perfectly fitted to the development and expression of their superior capacities. We may assume that Kant, who had set out to limit reason to make room for faith, would have disagreed.