Kant’s Anthropology and Geography
I begin with Kant because his inspiration for the contemporary approach to cosmopolitanism is impossible to ignore. I cite perhaps the most famous passage from his essay on “Perpetual Peace”: “The peoples of the earth have entered in varying degrees into a universal community, and it is developed to the point where a violation of laws in one part of the world is felt everywhere. The idea of a cosmopolitan law is therefore not fantastic and overstrained; it is a necessary complement to the unwritten code of political and international law, transforming it into a universal law of humanity.”1
Kant’s conception of cosmopolitan law arises in the context of a certain kind of geographical structure. The finite quality of the globe defines limits within which human beings, by virtue of their common possession of the surface of the earth, are forced to accommodate (sometimes violently) with each other. Human beings have the inherent right, if they so desire, to range across the surface of the earth and to associate with each other (through trade and commerce, for example). Means of transport (Kant mentions the ship and the camel) facilitate increasing contacts over space. But in Kant’s schema, the earth’s surface is presumed to be territorially divided into sovereign states. These will tend in the long run to become both democratic and republican. Inhabitants will then possess distinctive rights of citizenship within their states. Relations between states will be regulated by a growing requirement to establish perpetual peace because of increasing interdependence through trade and commerce. War between states becomes less likely for two reasons. First, in a democratic state it will be necessary to gain the consent of a public that would have to bear the brunt of the costs. The habit of sovereigns, emperors, and the nobility of waging war for reasons of personal prestige or aggrandizement will be constrained. Second, trade disruptions from war would inflict greater and greater losses as the levels of economic interdependence between states increased. The cosmopolitan ethic requires that individuals (presumed citizens of one state) would have the right to hospitality when they cross clearly defined borders (particularly for purposes of trade): “Hospitality means the right of a stranger not to be treated as an enemy when he arrives in the land of another. One may refuse to receive him when this can be done without causing his destruction; but so long as he peacefully occupies his place, one may not treat him with hostility. It is not the right to be a permanent visitor that one may demand. A special contract of beneficence would be needed in order to give an outsider a right to become a fellow inhabitant for a certain length of time. It is only the right of temporary sojourn, a right to associate, which all men have. They have it by virtue of their common possession of the surface of the earth, where, as a globe, they cannot infinitely disperse and hence must finally tolerate the presence of each other.”2
Cosmopolitan right is, therefore, circumscribed. “The right of hospitality,” Benhabib notes, “occupies that space between human rights and civil rights, between the right of humanity in our person and the rights that accrue to us insofar as we are members of specific republics.”3 The presumption of a sovereign (preferably democratic and republican) state authority defined by its distinctive territoriality lies at the basis of this formulation. For purposes of citizenship the territoriality of the state is regarded as an absolute space (that is, it is fixed and immovable and has a clear boundary). But it is the universal (that is, deracinated) right to hospitality that opens the absolute spaces of all states to others under very specific conditions.
Kant’s formulation of the cosmopolitan ethic has been the subject of considerable analysis and debate. But no one has cared to explore the implications of Kant’s assumptions about geographical structure for the cosmopolitanism he derives. The only substantive discussion I can find concerns the role that the common possession of a finite globe plays in Kant’s justification of cosmopolitan right. The consensus seems to be that “the spherical surface of the earth constitutes a circumstance of justice but does not function as a moral justificatory premise to ground cosmopolitan right.”4 This conclusion is understandable. To conclude otherwise would be to commit the naturalistic fallacy or, worse still, to fall into a crude environmental determinism (the idea that spatial structure—the sphericity of the globe—has direct causative powers). But relegation of the geographical circumstances to the status of a mere “circumstance of justice” is not the end of the issue. It is as if the nature of the geographical space has no bearing in relation to principles applied to it. Though the material (historical and geographical) circumstances may be contingent, this does not mean that the characterization of those circumstances in the form of anthropological and geographical knowledges is irrelevant to the formulation of a cosmopolitan ethic. Nussbaum and, as we shall see, Kant himself clearly think the circumstances matter. And so, it turns out, does Foucault. So how and why does it matter?
Kant’s philosophical teaching concentrated on logic, metaphysics, and ethics. But he also taught geography and anthropology on a regular basis. Is there any relation between these teachings? His writings on anthropology and particularly geography have, until very recently, been generally ignored or relegated to a zone of insignificance in relation to his three major critiques. The Anthropology has, however, been translated into several languages and subjected to some commentary. Foucault, for example, translated the Anthropology into French in 1964, promising a deeper analysis of it in a subsequent publication. He never made good on this promise (though he did leave behind an extended commentary that is now finally available to us). Kant’s Geography is known hardly at all (Foucault, interestingly, barely mentions it). Whenever I have in the past questioned Kantian scholars about it, their response has almost always been the same. It is “irrelevant,” “not to be taken seriously,” or “there is nothing of interest in it.” There is as yet no published English edition (though there is a translation of Part I as a master’s thesis by Bolin). A French version finally appeared in 1999, and an English translation is scheduled.5 There is as yet no serious study of Kant’s Geography in the English language other than May’s, coupled with occasional forays by geographers into understanding Kant’s role in the history of geographical thought (without any attempt to link this to his metaphysics or ethics). The introduction to the French edition of the Geography does attempt an evaluation, and a recent English-language conference bringing together philosophers and geographers finally promises serious examination of the problems the Geography poses.6
This historical neglect of the Geography does not accord with Kant’s own assessment. He went out of his way to gain an exemption from university regulations in order to teach geography in place of cosmology. He taught geography forty-nine times, compared to the fifty-four occasions when he taught logic and metaphysics, and the forty-six and twenty-eight times he taught ethics and anthropology, respectively. He explicitly argued that geography and anthropology defined the “conditions of possibility” of all knowledge. He considered these knowledges a necessary preparation—a “propaedeutic” as he termed it—for everything else.7 While, therefore, both anthropology and geography were in a “precritical” or “prescientific” state, their foundational role required that they be paid close attention. How else can we interpret the fact that he taught geography and anthropology so persistently alongside his metaphysics and ethics? Though he signally failed in his mission, he plainly thought it important to bring anthropology and geography into a more critical and scientific condition. The question is: why did he think so?
F. Van de Pitte, in his introduction to the Anthropology, provides one answer to this question. As Kant increasingly recognized that “metaphysics could not follow the method of pure mathematics,” then, as Kant himself put it, “the true method of metaphysics is basically the same as that introduced by Newton into natural science.” Metaphysics must rest, therefore, upon a scientific understanding of human experience. But if metaphysics now was to begin in experience, where would it find the fixed principles in terms of which it could build with assurance? As Kant himself expressed it, the variations in taste and different aspects of man give to the flow of experience an uncertain and delusive character. “Where shall I find fixed points of nature which man can never shift and which can give him indications of the shore on which he must bring himself to rest?”
Kant, according to Van de Pitte, turned to Rousseau’s writings to find an answer. There he discovered that “because man can consider an array of possibilities, and which among them is more desirable, he can strive to make himself and his world in a realization of his ideals.” This could be so because human beings possessed powers of rational thought (though mere possession of these powers did not guarantee their appropriate use). But this meant in turn that metaphysics need no longer be purely speculative. It must proceed “in terms of clearly defined absolute principles derived from man’s potential.”8 By what means, then, could man’s potential be established, if God and traditional cosmology could not provide the answer?
At several points in his articulation of the cosmopolitan ethic, Kant expresses the view that the ethic arises out of nature or out of human nature (he sometimes seems to conflate the two). The cosmopolitan ethic is therefore based on something other than pure speculation or idealism. Kant (unlike President Bush) refuses to invoke any notion of God’s design. The attention Kant pays to both geography and anthropology then makes more sense. If theology and cosmology could no longer provide adequate answers to the question “what is man?” (hence Kant’s determination to eliminate cosmology from the curriculum and replace it with geography), then something more scientific was needed. Where was that “science of man” to come from, if not from anthropology and geography? The distinction between geography and anthropology rested, in Kant’s view, on a difference between the “outer knowledge” given by observation of “man’s” place in nature and the “inner knowledge” of subjectivities (which sometimes comes close to psychology in practice). This dualism bears a heavy burden, for it underpins the supposedly clear distinctions between object and subject, fact and value and, ultimately, science and poetry that have bedeviled Western thought ever since. He began teaching geography first (in 1756), and much of what he there examines concerns the physical processes that affect the earth’s surface and human life upon it. This suggests a certain initial attraction to an underlying theory of environmental determinism as providing a potentially secure scientific basis for metaphysical reflection (and, as we shall see, many of the examples he evokes in his geography reflect that tendency). His later turn to anthropology (which he began teaching in 1772), and the fact that he paid far greater attention to elaborating upon it (even preparing it for publication) in his later years, suggests that he increasingly found the inner knowledge of subjectivities more relevant to his philosophical project. “As a result,” Foucault provocatively suggests, “the notion of a cosmological perspective that would organize geography and anthropology in advance and by rights, serving as a single reference for both the knowledge of nature and the knowledge of man, would have to be put to one side to make room for a cosmopolitical perspective with a programmatic value, in which the world is envisaged more as a republic to be built than a cosmos given in advance.”9 It is significant that the final passages of the Anthropology address the whole question of cosmopolitan law directly, while there is no mention of this topic in the Geography.
Consider first, then, the implications of his Anthropology. The work amounts to a detailed inquiry into our species being (it foreshadows, therefore, Marx’s examination of the concept in The Economic and Philosophic Manuscripts of 1844). The purpose is plain enough. Not only must we understand what we have been about and how we now are as a human species: we must also understand what we can become by virtue of our particular capacities and powers. Human nature is not fixed but evolving, and by studying that evolution we can say something about the destiny of the human race. Foucault, in his commentary, is as profoundly admiring of Kant’s capacity to ask these questions as he is critical of Kant’s actual answers. “Man is not simply ‘what he is,’ but ‘what he makes of himself.’ And is this not precisely the field that Anthropology defines for its investigation?” Foucault asks. The Anthropology is, therefore, in Foucault’s view, a central rather than marginal text in relation to the three major philosophical critiques that Kant contributed. Amy Allen summarizes Foucault’s argument this way:
Thus, Foucault suggests, the Anthropology (perhaps unwittingly) breaks open the framework of the critical philosophy, revealing the historical specificity of our a priori categories, their rootedness in historically variable social and linguistic practices and institutions. Foucault’s reading of Kant’s Anthropology thus suggests that Kant’s system contains the seeds of its own radical transformation, a transformation that Foucault will take up in his own work: namely the transformation from the conception of the a priori as universal and necessary to the historical a priori; and the related transformation from the transcendental subject that serves as the condition of possibility of all experience to the subject that is conditioned by its rootedness in specific historical, social and cultural circumstances.10
This transition in thinking from a disembodied to a rooted human subject is critical, and the vehicle is in the first instance supplied by the Anthropology. Kant’s views on our species being are not confined to his text on Anthropology, so on this point some contextualization is needed. Kant generally rejects any notion of the inherent goodness of humanity. He does not appeal to any figure of the noble savage or of Godly innocence. “Everything,” he says, “is made up of folly and childish vanity, and often of childish malice and destructiveness.”11 Enlightenment, he says in his celebrated essay on that subject, depends upon “man’s emergence from his self-incurred immaturity,” defined as the inability to use understanding “without the guidance of another.” Only a few, Kant suggests, “have succeeded in freeing themselves from immaturity and in continuing boldly on their way,” while all manner of prejudices (even the new ones created in the course of revolution) “will serve as a leash to control the great unthinking mass.” For enlightenment to progress depends on “the most innocuous form” of freedom—“the freedom to make public use of one’s reason in all matters.” While we live in an age of enlightenment, we do not live in an enlightened age. This way of thinking enters into the final passages of the Anthropology. Human beings, he says:
cannot be without peaceful coexistence, and yet they cannot avoid continuous disagreement with one another. Consequently, they feel destined by nature to develop, through mutual compulsion and laws written by them, into a cosmopolitan society which is constantly threatened by dissension but generally progressing toward a coalition. The cosmopolitan society is in itself an unreachable idea, but it is not a constitutive principle. . . . It is only a regulative principle demanding that we yield generously to the cosmopolitan society as the destiny of the human race; and this not without reasonable grounds for supposition that there is a natural inclination in this direction. . . . [W]e tend to present the human species not as evil, but as a species of rational beings, striving among obstacles to advance constantly from the evil to the good. In this respect our intention in general is good, but achievement is difficult because we cannot expect to reach our goal by the free consent of individuals, but only through progressive organization of the citizens of the earth within and toward the species as a system which is united by cosmopolitical bonds.12
The mission of Kant’s anthropology—written, as he insists, from “a pragmatic point of view”—is, therefore, to define “the conditions of possibility” for that “regulative principle” that can lead us from a condition of folly and childish vanity, from violence and crude brutality, to “our destiny” of a peaceful cosmopolitan society. This entails an analysis of our cognitive faculties, of our feelings (of pleasure and displeasure), and of desire (the influence on Foucault’s work is obvious). It also entails reflection on how and why natural endowments (“temperaments”) are transformed by human practices into “character.” Kant writes: “what nature makes of man belongs to temperament (wherein the subject is for the most part passive) and only what man makes of himself reveals whether he has character.”13 While this introduces an unfortunate dichotomy between our “animal” and our “civilized” being, it does open up the possibility for the ongoing work of perpetual transformations of character. Pheng Cheah summarizes Kant’s argument as follows:
As natural creatures with passions and sensuous inclinations, we are, like things and animals, creatures of a world merely given to us and are bound by the same arational mechanical laws of causality governing all natural objects. However, as moral subjects we are self-legislating rational agents. We belong to a transcendent realm of freedom we create for ourselves, a world that encompasses all rational beings governed by universal laws we prescribe through our reason. The moral world is supersensible and infinite because it is not subject to the blind chance of meaningless contingency that characterizes finite human existence . . . culture provides a bridge to the transcendent world of freedom because it minimizes our natural bondage by enhancing the human aptitude for purposive self-determination . . . [it] liberates the human will from the despotism of natural desires and redirects human skill toward rational purposes by forming the will in accordance with a rational image.14
The general proposition that “man makes himself” carries over very strongly, of course, into the Marxist tradition. Echoes of Kant’s transcendent definition of freedom can also be heard in Marx’s pronouncement that “the realm of freedom actually begins only where labour which is determined by necessity and of mundane considerations ceases” and that this “lies beyond the realm of material production.”15 Kant reflects on how far we have progressed in reshaping temperaments into character through the making of culture by examining differentials in national character and cultures. The text is lighthearted, anecdotal, and on occasion deliberately amusing in the national stereotypes it evokes. But this should not detract from the seriousness of Kant’s purpose. Human beings have made themselves differently in different places and produced different cultures. Our task—and on this point Kant’s arguments are surely powerfully to the point—is to exercise both judgment and intelligence with respect to this process: “Just as the faculty of discovering the particular for the universal [the rule] is called judgment, so the faculty of discovering the universal for the particular is called intelligence. Judgment concentrates on detecting the differences within the manifold as to partial identities; intelligence concentrates on marking the identity within the manifold as to partial differences. The superior talent of both lies in noticing either the smallest similarity or dissimilarity. The faculty to do this is acuteness, and observations of this sort are called subtleties, which, if they do not advance knowledge, are either called empty sophistries or conceited prattlings.”16
Judged against this high-sounding standard, Kant’s own formulations often appear unduly crude (if not as empty sophistries and conceited prattlings). Throughout the Anthropology we are assaulted by all manner of seemingly prejudicial statements about race, class, gender, and nation. His statements on the nature of woman and the feminine character will likely outrage even the mildest feminist (though they will probably delight some ardent evangelicals). In seeking to understand the differentiations that plainly occurred within our species being, Kant initiated the idea (which later had a very unfortunate history) that the question of race should be put upon a purely scientific footing. And his consideration of the roots of national identity is problematic: “By the word people (populus) we mean the number of inhabitants living together in a certain district, so far as these inhabitants constitute a unit. Those inhabitants, or even a part of them, which recognize themselves as being united into a civil society through common descent, are called a nation (gens); the part which segregates itself from these laws (the unruly group among these people) is called rabble (vulgus), and their illegal union is called a mob (agere per turbas), a behavior which excludes them from the privileges of citizen.”17
Not all residents, by this account, qualify as citizens. It all depends, according to Kant, upon the “maturity” of the individual, a normative concept of rational behavior that Foucault challenges head on in his essay “What Is Enlightenment?”18 Is a character like Baudelaire to be excluded from citizenship, Foucault asks, by virtue of what Kant would almost certainly consider his irrational immaturity? And what happens when—as Kant, along with most other Enlightenment thinkers, holds—women are by definition considered immature and therefore incapable of participating in public life?
The definition and significance to humanity of nationhood by common descent, however, leads Kant to one of his most important conclusions. A singular world government could only exist as monarchical despotism because it would have to erase and suppress national differences based upon common descent within territorial configurations. A world government of that sort would, in short, go against nature and human nature. The only form of cosmopolitan government that will work is one based on a federation of independent (preferably democratic and republican) nation-states. This may or may not be a good idea, but it is important to recognize that Kant’s derivation of it arises out of his highly questionable anthropological conception of the nation-state as a civil society based on common descent (to say nothing of the exclusion of “troublesome elements”—however defined—from rights of citizenship). This presumption also helps explain why the cosmopolitan right to cross borders is so circumscribed and why the right to hospitality must be temporary. Permanent residence for foreigners is inconsistent with the requirement of common descent. Those, like Benhabib, who want to extend the rights of migrants in meaningful ways have therefore to struggle mightily with the restrictions of the Kantian cosmopolitan frame. The real problem lies, however, in the questionable anthropological foundations for Kant’s arguments. But if Kant’s specific anthropological foundation is rejected, as I think it must be, then the question arises as to what is or what might be an adequate anthropological foundation for understanding the territorial structures of human association. Indeed, questions might reasonably be asked as to whether the cosmopolitan ethic requires any kind of anthropological foundation whatsoever. Plainly, both Nussbaum and Foucault, as well as Kant, believe it does.
An examination of Kant’s Geography raises even deeper problems. The lack of interest in it on the part of Kant scholars is understandable since its content is nothing short of an intellectual and political embarrassment. As R.-P. Droit remarks, reading it “comes as a real shock” because it appears as “an unbelievable hodge-podge of heterogeneous remarks, of knowledges without system, of disconnected curiosities.”19 The thought that this might provide a secure foundation for metaphysical reflection is just absurd. To be sure, Kant seeks to sift the sillier and obviously false tales from those that have some factual credibility, but we are still left with a mix of materials more likely to generate hilarity than scientific credibility. But there is a more sinister side to it. While most of the text is given over to often bizarre facts of physical geography (indeed, that was the title of his lectures), his remarks on “man” within the system of nature are deeply troubling. Kant repeats without critical examination all manner of prejudicial remarks concerning the customs and habits of different populations. Thus we find:
In hot countries men mature more quickly in every respect but they do not attain the perfection of the temperate zones. Humanity achieves its greatest perfection with the White race. The yellow Indians have somewhat less talent. The Negroes are much inferior and some of the peoples of the Americas are well below them.
All inhabitants of hot lands are exceptionally lazy; they are also timid and the same two traits characterize also folk living in the far north. Timidity engenders superstition and in lands ruled by Kings leads to slavery. Ostoyaks, Samoyeds, Lapps, Greenlanders, etc. resemble people of hot lands in their timidity, laziness, superstition and desire for strong drink, but lack the jealousy characteristic of the latter since their climate does not stimulate their passion greatly
Too little and also too much perspiration makes the blood thick and viscous. . . . In mountain lands men are persevering, merry, brave, lovers of freedom and of their country. Animals and men which migrate to another country are gradually changed by their environment. . . . The northern folk who moved southward to Spain have left progeny neither so big nor so strong as they, and which is also dissimilar to Norwegians and Danes in temperament.20
Burmese women wear indecent clothing and take pride in getting pregnant by Europeans, the Hottentots are dirty and you can smell them from far away, the Javanese are thieving, conniving and servile, sometimes full of rage and at other times craven with fear. It is difficult to attribute any notion of rationality or maturity to such populations.
This, surely, cannot be the kind of geography that Nussbaum has in mind. When projected into a world of sovereign democratic and republic states, it conjures up a threatening image of unwashed Hottentots, drunken Samoyeds, conniving and thieving Javanese, and hordes of Burmese women lusting to become pregnant by Europeans, all clamoring for the right to cross borders and not be treated with hostility. It is precisely in such geographical “circumstances” that we can better understand why Kant included in his cosmopolitan ethic and in his notion of justice the right to refuse entry (provided it does not result in the destruction of the other), the temporary nature of the right to hospitality (provided the entrant does not create any trouble), and the condition that permanent residency depends entirely on an act of beneficence on the part of a sovereign state that in any case always has the right to deny rights of citizenship to those who create trouble. Only those who exhibit maturity, presumably, will be granted the right to stay permanently. Again, those like Benhabib who struggle mightily to loosen the constraints of Kantian cosmopolitan law as it relates to the rights of migrants in effect have to undo the hidden trace of these geographical preconceptions upon Kant’s formulation of cosmopolitan law.
None of this has gone away. Acrimonious debates about the rights of minorities and of migrants to be received “without hostility” even on a temporary basis abound in our contemporary world. All manner of prejudicial and stereotypical conceptions about “others” and “strangers” exist, even among highly educated political elites. Denial of rights of citizenship to strangers on the grounds that they are immature and not like us is all too familiar. And while Kant’s excursus into the idea of national character may be barely acceptable, the long tradition of writing on the “peculiarities” of the English, the particularities of the French, the Spanish, or the Italians, and the like by eminent and much respected writers (such as P. Anderson, T. Zeldin, and L. Barzini, with tacit support from epic works such as that of E. P. Thompson on The Making of the English Working Class)21 suggests that Kant was onto something. Furthermore, when political philosophers of the stature of John Rawls (most particularly in The Law of Peoples) and Michael Walzer (particularly in Spheres of Justice) ground their arguments in something akin to Kant’s original idea with respect to national character and culture, then we have to take the whole question of the anthropological and geographical rootedness of political philosophy, if not of politics itself, far more seriously than is our wont22
Kant’s geographical depictions can, of course, be excused as mere quotations from or echoes of Montesquieu and other scholars with environmental determinist and racist leanings, such as Hume and Buffon (to say nothing of the lore that Kant picked up from merchants, missionaries, and sailors passing through Königsburg). Many of the fervent defenders of universal reason and of universal rights at that time, Droit notes, cheerfully peddled all manner of similarly prejudicial materials, making it seem as if racial superiorities and ethnic cleansings might easily be reconciled with universal rights and ethics (though Kant, to his credit, did go out of his way to condemn colonialism on the grounds that this was occupation without permission and therefore a violation of cosmopolitan law).23 And all manner of other extenuating circumstances can be evoked: Kant’s geographical information was limited; his course in geography was introductory, meant to inform and raise issues rather than solve them; and Kant never revised the materials for publication (the text that comes down to us was compiled from Kant’s early notes—around 1759—supplemented by those of students, and there is controversy over how corrupted the text is relative to other, later versions given during the 1790s).24 And his later shift in emphasis toward the Anthropology in any case suggests a gradual progress away from some of the grosser forms of environmental determinism that are featured in his Geography.
But that Kant’s Geography is of such uncertain genealogy and an embarrassment to boot is no justification for ignoring it. Indeed, this is precisely what makes it so interesting, particularly when set against his much-vaunted universal ethics and cosmopolitanism. Dismissal of his Geography does not accord with Kant’s own positioning of it as a “condition of possibility” and as a “propaedeutic” for all other forms of reasoning (including his metaphysics and his ethics). The problem is that Kant failed entirely to bring geographical knowledge out of its “precritical state” and place it on a rational, scientific basis. He later hinted as to why. He simply could not make his ideas about final causes work on the terrain of geographical knowledge. “Strictly speaking,” he wrote (in a passage that Clarence Glacken regards as key), “the organization of nature has nothing analogous to any causality known to us,”25 and this problem blocked his ambition to construct geographical understandings in a style akin to Newtonian natural science. If this is so, then his metaphysics and his ethics lack the solid scientific foundations he considered essential to their formulation. They revert to the sphere of mere speculation as to “man’s species being.”
The problem that Nussbaum poses of how anthropological and geographical knowledges might be better constructed and positioned in relation to the “proper” formulation of a cosmopolitan ethic is left open by Kant, and hardly anyone has cared to investigate it since. Most contemporary commentators either ignore this question or, as does Seyla Benhabib or Tim Brennan, attempt to deal with some of the issues that Kant left dangling through ad hoc adjustments to his concept of cosmopolitan law.26 Laudable though such adjustments may be, these writers deal with symptoms rather than underlying structural problems, not only with Kant’s original formulations but with almost all subsequent work on the subject. So what would it take, then, to reconstruct anthropological and geographical knowledges in a way that could better inform struggles over the proper conception of cosmopolitan law? How, in short, should we attempt to answer Nussbaum’s foundational demand for a proper set of geographical and anthropological understandings? And what, under contemporary conditions, could “proper” possibly mean? While such questions may appear daunting, if not unanswerable in any simple sense, this should not deter us from investigating them.
Foucault’s position on this is interesting. He seems to have been profoundly affected by his reading of Kant’s Anthropology and clearly saw this as a propaedeutic to Kant’s ethics. And his own writings bear the trace of that influence throughout. But Foucault apparently never read Kant’s texts on geography. He did, however, make frequent use of spatial concepts. This was particularly evident in the relatively early and long-unpublished essay on heterotopia—about which more anon—and in his careful delineation of spatial forms (such as his celebrated use of the panopticon) in his inquiries into prisons and hospitals in texts like Madness and Civilization and Discipline and Punish. He accepted that spatiality was a key concept and initially at least seems to have accepted Kant’s views on how space should be understood. But later in life, he also openly worried, perhaps with a critique of Kant as well as his own formulations of heterotopia in mind, at the way “space was treated as the dead, the fixed, the undialectical, the immobile,” while “time, on the contrary, was richness, fecundity, life, dialectic.”27 If “space is fundamental in any form of communal life,” then space must also be “fundamental in any exercise of power,” he argued. More surprisingly, when asked in 1976 by the editors of the newly founded radical geography journal Hérodote to clarify his arguments on space and geography, Foucault gave evasive and seemingly incomprehending answers to what, on the whole, were quite reasonable probing questions. By refusing again and again to elaborate or even speculate on the material grounding for his vast arsenal of spatial metaphors, he evaded the issue of a geographical knowledge proper to his or anyone else’s understandings (even in the face of his use of actual spatial forms such as the panopticon to establish his themes). He failed, furthermore, to give tangible material meaning to the way space is “fundamental to the exercise of power.” Yet this is what he eventually did say by way of conclusion: “I have enjoyed this discussion with you because I’ve changed my mind since we started. Now I can see that the problems you put to me about geography are crucial ones for me. Geography acted as the support, the condition of possibility for the passage between a series of factors I tried to relate. Where geography itself was concerned, I either left the question hanging or established a series of arbitrary connections. . . . Geography must indeed necessarily lie at the heart of my concerns.”28
Foucault here accords, albeit somewhat reluctantly, a parallel status of “condition of possibility” for geography to that earlier assigned to Kant’s anthropology. It is, therefore, the geographical rootedness in specific historical, social, and cultural circumstances—the historical and geographical a priori if you will—that now must be taken into account. So how then are we to understand this relation between the geographical as opposed to the historical a priori, and what role does this play in relation to Foucauldian ethics (to say nothing of politics)? In an attempt to find some answers to this question, Foucault subsequently submitted a series of questions to the editors of Hérodote as to how he should properly understand what geography was about.29 But his questions indicated that he equated geographical knowledge with the study of spatiality and spatial order. Why did he take such a limited and undialectical view of what geography might be about?
Critical engagement with Kant here provides some useful pointers, for, as May argues, it is possible to reconstruct some of Kant’s putative principles of geographical knowledge from the general corpus of his writings. Geography was not only a precursor, as we have seen, but also, together with anthropology, a synthetic end-point of all of our knowledge of the world (understood as the surface of the earth as “man’s” habitation). And Kant saw this end-point as more than simply a posteriori knowledge of the world. It is in some sense constructive of our “destiny” for actually living in the world. In other words, we need not only to examine what our geography and anthropology have been and are, but consider what they might become.
Geography, however, looks at “man” as a “natural object within the system of nature.” In the eighteenth century this meant that geographical knowledge was prone to those forms of environmental determinism that could all too easily lurch over (as we have already seen in Kant’s case) into blatant racism. But the general question Kant poses, of how to understand the metabolic relation between human evolution and environmental transformations, is as vital now as ever. Just because Kant plainly got it wrong is no excuse for ignoring the question. And environmentalism has by no means disappeared. In the contemporary work of Jared Diamond and Jeffrey Sachs, as we will later see, it even acquires a seeming scientific respectability. How to conceptualize “man” as a “natural object within the system of nature” remains a core question, and how we answer it will affect not only the technicalities of application of cosmopolitan law but the whole destiny of humanity.
But Kant excluded environmental history from the definition of geography per se. Geographical knowledge concerns the study of spatial order alone, he argued. History is considered distinctive because it provides a narration in time. These two synthetic forms of knowledge—in turn quite distinct from analytic sciences such as physics and biology—should not be confused with each other. Geography (along with other spatial sciences such as archaeology and astronomy) synthesizes analytical findings in terms of space, and history does so with respect to time. This separability of space and time, particularly with respect to the organization of knowledge, positions Kant in the Newtonian tradition with regard to the nature of space and time. But Kant also recognized a problem with the Newtonian adaptation of the Cartesian theory of space and time. If space and time are considered infinite, absolute, and empty of all matter, then they are unavailable to our understandings through direct perceptual experience. Kant’s answer was that space and time are accessible to our intuitions; our knowledge of them is synthetic a priori. This opens up all sorts of possibilities in principle for non-absolute definitions of space and time, but there is no sign in Kant’s work that human intuition would uncover any other scientific truth save that of the Newtonian absolute system with respect to the world of experience. Kant in effect uses Leibniz to seal in an absolute Newtonian view of a separable space and time.30 This, as we shall see later, seriously inhibits the Kantian perspective. Kant’s geography is then defined as an empirical form of knowledge about spatial ordering and spatial structures alone, and this definition dominates Foucault’s perspective in his questions to the editors of Hérodote. Kant’s definition of the Geography as a “synoptic discipline synthesizing findings of other sciences through the concept of Raum [area or space]” has been influential in the history of German and US Geography with unfortunate results.31 R. Hartshorne, in The Nature of Geography, published in 1939 under the auspices of the Association of American Geographers, used Kant’s ideas to dismiss entirely the possibility of a field called historical geography (except as comparative statics) much to the umbrage of Carl Sauer, the main practitioner of historical and cultural geography in the United States from the 1920s onward.32 The particularity of spatial positioning, furthermore, is marked, Hartshorne argued, by contingency, and under this restrictive definition geography can only be concerned with the unique and the particular. This contrasts radically with the universality that attaches to the concept of a unidirectional time that might point us teleologically toward our destiny of cosmopolitan governance. F. Schaeffer was later to dispute Hartshorne’s interpretation of Kant, provoking a series of vitriolic exchanges, by arguing that it was perfectly feasible to determine universal laws of spatial order (applying geometric models to settlement patterns, as in the central-place theories of Christaller and Lösch, for example). Under another interpretation of Kant’s scheme of things, favored by May, spatial ordering produces regional and local truths and laws, as opposed to universals. These local laws are derived territorially by way of the specific rules of citizenship within the history of nation-states defined in terms of common descent. This “absolute” Newtonian conception of space (and of time) then frames Kant’s territorial anthropological approach to cosmopolitan law, much as it also frames Rawls’s and Walzer’s approach to questions of local justice. Kant’s map of the world is equivalent to a Mercator projection with absolute borders of nation-states clearly defined. This perspective of absolute entities in space underpins Foucault’s concept of heterotopia, and this in turn may have influenced Foucault’s view that metatheory is inadmissible and that the politics of the contingent and the local (including local knowledges) is all that matters.33
May does not tell us how Kant proposed to relate local truths and laws (such as national character) to the universals of reason (humanity in the abstract). But if May’s account is right, then Kant’s geographical and anthropological knowledges appear potentially in conflict with his universal ethics. What happens, for example, when universal ethical ideals are applied to issues of global governance in a world in which nation-states set up their own distinctive rules consistent with their national character? Worse still, how do we apply a universal ethic to a world in which some people are considered immature or inferior and others are thought indolent, smelly, or just plain untrustworthy? Either the smelly Hottentots, the lazy Samoyeds, the thieving Javanese, and the indecent Burmese women have to reform themselves for consideration under the universal ethical code (thereby flattening out all kinds of geographical and cultural differences in favor of some normative definition of maturity), or the universal principles operate across different geographical conditions as an intensely discriminatory code masquerading as the universal good. There are reasonable grounds for inferring that Kant actually thought the former, since in his famous essay “What is Enlightenment?” he made much ado about human “maturity” as a necessary condition for proper engagement in a public realm where certain freedoms were institutionally established and politically guaranteed. His rules for the exclusion of troublesome elements from citizenship within a sovereign democratic republic support that view. But, as we have also already seen, the supposedly universal principles laid out in his specification of cosmopolitan law entail all manner of hidden concessions to a certain version of anthropological and geographical realities. One suspects that it is precisely the attraction of Kant’s cosmopolitanism that it can somehow sustain a veneer of attachment to some theory of universal goodness while allowing, even justifying, innumerable concessions to prejudicial exclusions on the ground. From this perspective it may even legitimately be claimed that Bush is a true Kantian.
This is, as many have recognized, a fundamental and unresolved difficulty in Kant’s whole approach to knowledge. Hannah Arendt puts the dilemma this way: “The chief difficulty in judgment is that it is ‘the faculty of thinking the particular’; but to think means to generalize, hence it is the faculty of mysteriously combining the particular and the general. This is reasonably easy if the general is given—as a rule, a principle, a law—so that the judgment merely subsumes the particular under it. The difficulty becomes great if only the particular be given for which the general has to be found. For the standard cannot be borrowed from experience and cannot be derived from outside. I cannot judge one particular by another particular; in order to determine its worth.”34
Kant’s answer, as we have seen, is to invoke that acute intelligence that acknowledges subtleties. But then the danger is the production of “empty sophistries and conceited prattlings.” The version of Kant’s geography that has come down to us amounts to an incoherent bunch of anecdotic particulars for which the general has yet to be found. To this day, geographical knowledge continues to lie very much in this state, in spite of some of the best efforts of geographers and others to reform its ways. Kant’s anthropology, though more systematic, is also deeply flawed. And contemporary anthropology, in spite of the efforts of its best practitioners, has hardly eliminated “empty sophistries and conceited prattlings” (particularly in its so-called “postmodern” guise). All of this would not be a problem were it not for the fact that the political consequences (as in Iraq, Rwanda, Palestine, and Darfur) can on occasion be nothing short of catastrophic.
What appears so dramatically with Kant has widespread ramifications for politics. Popular geographical and anthropological knowledges in the public domain (in the United States in particular) are either entirely lacking or of a similar prejudicial quality to that which Kant portrayed. Stereotypes about geographical “others” abound, and prejudicial commentary can be heard daily in casual conversations even in elite circles (listen in to any conversation about Mexicans, sub-Saharan Africans or Arabs in university common rooms, let alone upon the street, and see how quickly stereotypes are invoked and pass unchallenged).
It then becomes all too easy for the U.S. government (or any other government for that matter) to portray itself as the bearer of universal principles of justice, democracy, liberty, freedom, and goodness, while in practice operating in an intensely discriminatory way against others judged different, unfamiliar, or in some sense lacking in proper qualifications or human qualities. Bush, for example, propounds his version of the Kantian cosmopolitan ethic while shock and awe over Baghdad and the horrors of thousands of Iraqi deaths, plus the sordid sights of Abu Ghraib, bring us back to what in the technical language of remote sensing is referred to as the “ground-truthing” of abstract concepts in relation to anthropological and geographical realities.
This contrast between the universality of Kant’s cosmopolitanism and his ethics, and the awkward and intractable particularities of his anthropology and geography is therefore of critical importance. If knowledge of the latter defines (as Kant himself held) the “conditions of possibility” of all other forms of practical knowledge of the world, then on what grounds can we trust Kant’s cosmopolitanism if his anthropological and geographical groundings are so suspect? Yet there is a way to see this as a fruitful starting point for discussion. For while it is possible to complain endlessly about “the damage done by faction and intense local loyalties to our political lives,”35 it is also important to recognize how “human passions” (which Kant believed to be inherently aggressive and capable of evil) so often acquire a local and disruptive expression. In the face of this, it will take a tremendous effort to even approach that cosmopolitan state of which we are, at least in Kant’s judgment, potentially capable. Is there, then, some way in which we can facilitate that effort by answering Nussbaum’s call for a radical overhaul in our curricula for the teaching of geography, ecology, and anthropology? Kant identifies the questions but fails to provide adequate answers. He may have lived in an age of Enlightenment but it was most certainly not, as he himself understood, an enlightened age. We are then faced with an interesting choice. We can either reject the whole Enlightenment project, along with all of Bush’s rhetoric about freedom and liberty, as a sordid and hypocritical justification for imperial rule and global domination or accept the basic thrust of what the Enlightenment (and its U.S. off-shoot) was about, with the clear understanding that that particular stab at enlightenment was not enlightened enough. And one of the prime areas of knowledge that remains to be reconstructed is that of “appropriate” anthropological and geographical understandings that can illuminate the way to a genuinely cosmopolitan future. But behind this there lies a certain imperative that pushed the whole question of cosmopolitanism and a federated republicanism into the forefront of Kant’s concerns. Why, for example, do we need borders, and why do we need to cross them, anyway? On this point it is clear that the needs of trade and private property are paramount, and that implies that the legal requirements of merchant and landed capital in particular and perhaps even of capital in general play a highly significant closet role in Kant’s formulations.