Working Papers Series #12
ISSN 2045-5763 (online)

What Did Kant Mean by and Why Did He Adopt a Cosmopolitan Point of View in History?

Thomas Sturm
Universitat Autònoma de Barcelona

© 2012 Thomas Sturm
Creative Commons License
Open Anthropology Cooperative Press

Forum discussion in the OAC network.
Download as PDF, EPUB, MOBI.


Forthcoming in: Kant and Philosophy from a Cosmopolitan Point of View: XIth International Kant-Congress. Ed. by Stefano Bacin, Alfredo Ferrarin, Claudio La Rocca & Margit Ruffing. Berlin: De Gruyter.

NOTE: This version represents the pre-print version. This material is intended for purposes of education, research, scholarly communication, or critical commentary, all in conformity with “fair use” and the established practice of authors’ providing single preprints and offprints for noncommercial use. Any other use is unauthorized and may violate copyright.



It is widely held – and not false – that Kant’s philosophy of history expresses the Enlightenment hope for a stepwise progress of humankind towards freedom or morality. However, we are nowadays suspicious of models of a stadial development of human history, especially teleological ones. Furthermore, Kant’s model of historical development is burdened with problems of its own, concerning its epistemic status, and its position within his philosophy in general. To deal with these issues, scholars have mostly focused on connections between Kant’s philosophy of history and his ethics or his views about teleology. They have downplayed or neglected another context, namely, the theories of historiography that he was faced with. I shall show how Kant reacts to debates about a theory and practice of historiography highly influential in his time, especially in his German environment. It was called “pragmatic history”.

In part I, I indicate what major versions there existed of this approach. I then outline three crucial problems that emerged with the requirement, set up by many pragmatic historians, of a stage model of humankind’s development. Among other things, I shall point to how the debate about the meaning of ‘pragmatic history’ became connected to the idea of a ‘cosmopolitan viewpoint’ in history, an issue that was discussed particularly between August Ludwig Schlözer and Johann Gottfried Herder. In part II, I report on Kant’s reception of pragmatic history, and what he found lacking in the most important versions of it – namely, an appropriate understanding of human nature, which he himself developed more fully in his lectures on pragmatic anthropology. I shall thereby try to clarify how his own “cosmopolitan” idea of the development of human nature through history is likewise entangled with the notion of pragmatic history, and that his notion of a cosmopolitan idea itself has three different aspects, responding to the three problems outlined. Thus, relating Kant’s philosophy of history to contemporary debates can make his views more intelligible than merely analyzing their connection to other parts of his critical philosophy.  

I. Pragmatic History and Models of Human Historical Development

1. What Is ‘Pragmatic History’?

By the 18th century the study of history is growing quickly not only in terms of institutions and literary output but also in terms of the level of the debates about its theoretical and methodological presuppositions. In the German countries, this debate takes often shape under the heading of a “pragmatic” orientation. To mention but a few examples, eighteenth-century authors before Kant write pragmatic histories of the Jesuits and Protestants, the rulers of Braunschweig, the school reform in Bavaria, of literature, medicine, the souls of humans and animals, and even of sleep. And many historians at the time have a serious intention with this. As the Göttingen professor Johann Christoph Gatterer, the most influential organizer of historical research in the eighteenth century, writes, in “history, pragmatic is just what in the proper sciences is called systematic”.[1]

But which requirements pragmatic historiography need fulfill becomes controversial. In the debate, the following four requirements become introduced stepwise:

(i)             Most conceptions of pragmatic historiography take it for granted that the object of investigation is human action, particularly in more or less widely conceived areas of social life (at certain times and places).

(ii)           In methodological terms, a history can be pragmatic if it studies the causes, particularly the motives or intentions of human agents.

(iii)          A historical study can be called ‘pragmatic’ if it is tied to a universal history of mankind – either by helping to write that history or by presupposing it. Being “universal” does not necessarily mean that it has to cover all historical details, but at least the major factors and/or stages of human history.

(iv)          Finally, history can be called ‘pragmatic’ if practical consequences or lessons for human (particularly social) action can be derived from it.

These elements are not mutually exclusive. However, some pragmatic historians require only some of these features, while others demand that all be satisfied; furthermore, some authors claim that a certain requirement is more important than others; and, occasionally, some requirements are developed and discussed more closely and thereby become understood in different ways.

For instance, in the early eighteenth century, Johann David Kö(h)ler claims that a historical study is already pragmatic if it treats of public matters, especially the official and social deeds of rulers, and if it offers practical orientation in civil life, having in mind specifically political action and the design of public affairs.[2] But no later than in the 1750s, such a meaning of “pragmatic history” becomes viewed as overly narrow. This is accompanied by a growing awareness that there might be a pragmatic historiography of the “highest level” or in the “truest understanding” of the term, which has to be distinguished from lesser degrees and incorrect meanings.[3] To begin, a number of authors stresses that pragmatic histories must also inform readers of “impelling forces” (Triebfedern)[4], motives, and other causes.[5] Gatterer himself, who voices this point with particular emphasis, does not claim that previous historians had never sought out “causes and effects, means and intentions”.[6] Thucydides and Polybius clearly did. Gatterer’s main criticism is directed at the genres of mere annals, chronicles, and genealogies, and the accompanying conception that history merely records particular facts of the past. The causes behind historical events seldom coincide with periods or commencements of rule, and outcomes often extend beyond the dates covered by annals.

Gatterer moreover argues that the “highest level of what can be considered pragmatic history” can only be achieved by developing a universal history, by embedding historical investigation in “the idea of the overall connection of things in the world (Nexus rerum Universalis)” – that is, causal explanations in history must be embedded in a system of world history:

For no occurrence in the word is – as it were – insular. Everything is connected, is produced, is induced, and in turn produces and induces. The affairs of the noble and the lowly, individual persons and all of them together, private life and the world at large, indeed, even those of reasonless and lifeless entities and humans; all are intertwined and interconnected.[7]

While these requirements are all repeatedly emphasized by the majority of authors, requirement (iv) remains relatively negligible for Gatterer, unlike for others. He hints at it in one of his earlier writings[8], but later on clearly rejects the view that it would be constitutive of the idea of pragmatic history.[9] He also makes fun of the view, held by several authors, that one could derive practically useful conclusions from mere annals, chronicles or genealogies.[10] One might also think here of Lord Bolingbroke’s well-known dictum that “history is philosophy taught by examples”.[11] Obviously, Gatterer denies that such views help to raise the rank of history – to approximate it to the bona fide sciences.

While Gatterer becomes the most influential German historian of his times, his conception of pragmatic history does not remain undisputed. For instance, the Church historian Johann Matthias Schroeckh (1733-1808) favors a combination of all four requirements: A truly pragmatic history should focus upon human actions, provide causal explanations, develop and use a system of universal human history, and attempt to draw practical lessons on the basis of the first three requirements.[12] Also, other authors raise questions about various requirements. Some already discuss the possibility of giving causal explanations in history, while others are concerned about whether pragmatic histories ultimately have to study humankind as a whole, and whether such histories – if they aim at practical conclusions at all – should instruct particular individuals or groups or humankind as a whole.

2. The Requirement of a System of Universal History

Of special relevance here is the call for a system of history as a whole (requirement iii). How should or even could one write “the” complete history of humankind’s development? Most authors agree that it will not suffice to collect and order all existing special studies, and then continue them. That had been tried before. Schroeckh emphasizes that causal explanation demands various kinds of weighting. It is not easy, he writes, to describe the universal historical “Nexus” in a way that gathers and lists all causes and outcomes. It is not necessary, for instance, to note every historical detail or every slight causal connection. On the contrary, it is the difficult task of the historian to select the facts relevant for an adequate explanation of events. As Gatterer remarks, one has to identify and structure the “revolutions” of human history. Only these will help to identify the really important causes of human actions in history.[13]

His colleague at Göttingen, August Ludwig Schlözer (1735-1809) works out this approach in his Vorstellung seiner Universal-Historie (1772-73). He claims that one needs a unifying viewpoint in order to be able to select and order facts and turn them into a system:

World history can be imagined from a double perspective: Either as an aggregate of specialized histories, a collection of which, if it is complete, constitutes a whole in its own way; or as a system, in which the world and humanity constitute the one entity, for which from among all the parts of the aggregate some are preferably selected and ordered purposefully.[14]

 Furthermore, Schlözer demands that for this we need to single out factors that “interest not individual nations or classes of the human race, but that are significant for the cosmopolitan [Weltbürger], for man as such”.[15] More specifically, he claims that Roman history – from the city’s founding, the formation and division of the world empire, to its decline – provides the best focal point:

[Roman history] is the overall guiding thread [Leitfaden] that throughout various concurrent courses of almost innumerous peoples prevents chronological confusion. Rom deserves this honor: For which empire of the world has had greater influence on the fate of the world?[16]

3. Three Problems with the Requirement of Universal History

While the requirement for a structured system of universal history has its attractiveness for authors at the time, it has several problems.

(I) A first problem concerns an assumption about human nature, and it can best be explained by the impact of Hume. He does not, neither in his History of England (1754-62) nor elsewhere, use the term ‘pragmatic history’. Yet, German reviews praise the History as an example of pragmatic work and applaud Hume’s skill at “using his knowledge of human nature to enlighten and promote the usefulness of history”.[17] Two of Hume’s philosophical theses on human nature and history – to be found in the Treatise and the first Enquiry – are of particular importance here. He claims, first, that the historian may and should presume that human nature is constant, or subject to unchangeable causal laws. Second, he advances the methodological claim that by studying history we can discover these laws:

Mankind are so much the same, in all times and places, that history informs us of nothing new or strange in this particular. Its chief use is only to discover the constant and universal principles of human nature, by showing men in all varieties of circumstances and situations.[18]

Pragmatic historians often follow Hume on these points.[19] But this raises problems for their views. Many of these historians also stress that human history includes “revolutions”, necessitating a system of the most important developments. Also, as one reviewer of Hume’s History points out, impartiality is seen as vital to causal explanation: In order to reveal true causes, it is crucial to assess the past not in terms of maxims of the historian’s time, but in terms on those that held in the period and place under investigation[20]. However, these points only make sense given that modes of human conduct change substantially over time. Moreover, if pragmatic history should be used to draw practically relevant conclusions, then such conclusions may repeatedly lead to new principles for conduct – which threatens the Humean claim of the constancy of human nature as well.

(II) Second, how ought one to structure human history as a whole? If you take dominant nations as in Gatterer’s and Schlözer’s proposals: Should universal history first depict their histories and then turn to the subordinate countries? Or should the mutual influence of countries on one another be examined together?[21] Moreover, besides dominant nations, natural, economic, technological and intellectual factors are important too. Schlözer himself stresses that earthquakes, floods and epidemics, or also “the discovery of fire, bread and alcohol, and so on, are facts equally as important as the battles at Arbela, Zama, and Merseburg”.[22] Can all the factors be arranged within a single system of human history? In a review of 1772, Johann Gottfried Herder complains that Schlözer merely presents a plan lacking clear execution. In 1774, Herder furthermore suggests that what one reads “in almost all so-called Pragmatic Histories of the World is nothing but the disgusting tangled mass of ‘the time’s prized ideals’”.[23] In other words, Schlözer’s cosmopolitan orientation may in the best case be useless and in the worst case be the expression of an ideology.

(III) Finally, what is the epistemic role and status of the stage models of human history? The views here are quite divided. The outlines by Gatterer, Schlözer, and others are shaped by tangible tasks of empirical history. Claims about dividing the past into epochs, or questions of chronology are viewed as subject to empirical scrutiny. However, even the very same authors characterize their historical ideas and frameworks as “conjectural” or “philosophical”. This indicates that their function and status is not clear.

To sum up: One can see that the shift towards pragmatic history, reasonable as it was when compared with other traditions of history writing, led into serious new predicaments.


II. Kant on Pragmatic History and the Development of Humankind

4. Kant’s Reaction to Pragmatic History

Now to Kant. First, a bit about his standpoint towards pragmatic history. Since the mid-1770s, he presents his views especially in his annual lectures on anthropology. Here, he praises Hume’s History for not confining itself to chronicles of wars and rulers, but relating to humanity in general.[24] Also, Kant is familiar with the Enquiries. And in his early statements, one can see Kant as understanding and sharing the idea of pragmatic history along Humean lines: as a study of individual and social intentions causing actions, ideally useful for a practical instruction of agents in the social sphere. At least until 1775-76, he also accepts the ontological thesis that human nature is constant, linking it even to his own conception of anthropology. At the same time Kant becomes also interested in the genre of histories of the stadial development of humankind, including the idea of genuine change in human history.

In the 1780s, he suddenly scathes pragmatic historians for lacking the knowledge of human nature they pretend to have:

… since the authors of many history books have little knowledge of human nature, they have no idea of pragmatic history and much less of how to write it.[25]

I will explain in a moment what he means. Before this, I need to briefly comment on a related passage in the Groundwork. Here, Kant first distinguishes between pragmatic principles as leading to prudence, and notes that there are two different notions of prudence: Weltklugheit and Privatklugheit. The first is the competence to use other human beings for one’s purpose, the second is the competence to order one’s purposes such that one approximates one’s own happiness. He also says that Weltklugheit should serve Privatklugheit, because knowing how to manipulate other persons but not doing so for furthering one’s own well-reflected purposes isn’t very bright. But all this expresses not his fully considered opinion on what ‘pragmatic’ means but, rather, a report on widely held views. Just one page later he gives his own viewpoint:

It seems to me that the proper meaning of the word pragmatic could be determined thus most precisely. Pragmatic are called the sanctions which do not properly follow from the law of states as necessary laws, but from the precaution for general welfare. Pragmatically written is a history if it makes prudent, that is if it instructs the world how to reach its advantage better, or at least as well as its preceding world.[26]

So what he wants pragmatic history to do is not to teach us how to use other human beings simply for our personal purposes. But what would be wrong with that (leaving moral concerns aside here)? And what does he really have in mind with the „general welfare“? His answers stem from the background of his then developing anthropological views about what it means to be a citizen of the world. This then leads him to a specific notion of a cosmopolitan standpoint in history.

5. Kant’s Response to the Three Problems of Universal History

Let me explain this by reference to Kant’s response to the three problems of the various approaches to universal history described earlier on (section 3 above).

(I) First, Kant comes to reject a naïve view of the constancy of human nature. He does so by means of assumptions concerning basic factors of the dynamics of social interaction developed in his anthropology lectures. Six basic claims are necessary here.[27]

(1) Human dependency upon society. Human beings need education, and later on other forms of social cooperation to achieve our goals, to improve action possibilities and to uphold our self-regard.

(2) Human egoism. At the same time, unfortunately, human beings are mostly driven by self-interested inclinations. We do not trust each other; we are jealous; we try to manipulate and exploit one another. The conjunction of (1) and (2) Kant famously calls the “unsocial sociability” of humankind.

(3) The first-person point of view. That such things are possible is rooted in other, basic human facts. There is an important difference between our having of mental states and our having of physical states. Not only can we note that we are in such-and-such a mental state – say, that we feel a pain or have a desire. Unlike mere animals, we can be happy or sad about that, or we can view these states – and those of other persons as well – with a critical eye, reflect upon and change them. This requires a first-person point of view upon first-order mental states: To know that one is unhappy about a certain pain, and that one wishes that the pain goes away, requires knowing whose pain it is. Also, egoism and self-regard as well would be impossible without such a first-person point of view.

(4) Prudence and learning to adopt the third-person point of view. But what can we do about the dilemma of our unsocial sociability? Kant’s answer: If I want to act prudently, I have to learn that others have that egoism as well, and that it can be useful to take into account their first-person point of view.

(5) Invention of new social roles and rules. Thereby, however, social interaction becomes easily extremely complex. Not only do I perceive others as having egoistic motives and as having abilities for hiding such motives; they perceive me in the same way. Hence, our basic purposes of receiving respect and support must not be exerted too obviously, and we must be able to find new ways by which to pursue our goals prudently. This leads to iterated forms of role-playing in society, to a concealing and dissembling of egoistic intentions before others.

(6) New roles and rules become “another nature”. In this interaction, humans therefore develop new rules of interaction, or “another nature”.[28] But that means that our actions do not simply fall under rules as if they were natural laws; rather, we follow certain rules with a greater or lesser amount of rational deliberation. We can thus be producers instead of being mere products of our development.

From all this derives a first sense of cosmopolitanism in Kant’s, which is related to human nature: We are citizens of the world in the sense that our nature is partly plastic, and more specifically that we ourselves produce our rules of action and, thereby, our social world. This is a fact that holds, in principle, for each of us, and which each of us better recognizes in social interaction – instead of expecting to extract more superficial kinds of egoistic prudence from history.

(II) How does this notion of cosmopolitanism relate to the project of universal history? Kant – like Schlözer – claims that the historian needs a guiding “idea”, and again characterizes this idea by claiming that it centers on the human being as a “citizen of the world”. But, unlike Schlözer, Kant gives this notion a distinctive and not implausible meaning: the knowledge about the plasticity of human nature and its conditions is the knowledge he finds lacking in many pragmatic historians.

In Idea Kant then first outlines basic features of human social dynamics and explains afterwards how an adequate universal history would have to look like. It should start with ancient Greek history, for the contingent reason that only here a real source-based historiography could start. But the further steps should not look at dominant people and then wonder how to include other important factors; they should focus upon the development of forms of society that reduced aggression and war (such as the introduction of international commerce), introduced different elements of a republican constitution (the French Revolution becoming later on the outstanding “sign” of such a history), and that may lead to the establishment of a league of nations. This is obviously a second, richer notion of cosmopolitanism, but one presupposing the first. It flows from the former in the sense that such institutions help us to realize more fully the possibilities inherent in our nature, and to cope with our unsocial sociability.

(III) Finally, what about the epistemic role and status of this cosmopolitan idea? The answer is not surprising. No universal history should or even could aim a sum-total of all past events. Instead, by using the idea as guiding thread – another notion already to be found in Schlözer, as cited above, but not clarified by him – helps to find concepts and principles for selecting, linking and organizing historical knowledge in a certain way. The idea thus has a regulative function. Still, history seen from that perspective can be connected to empirically discoverable occurrences and developments.

There might be other perspectives, of course; but these have to be brought to the fore first. Kant emphasizes the sketchy nature of the Idea essay, it being “only one of the thoughts that a philosophical mind (that incidentally must be well-versed in history) might also toy with from a different standpoint”.[29] Kant does not claim that the propositions he sets forth about the development of human capacities, the mechanism of unsocial sociability, and the resulting sequence of forms of social or political order of humankind are already to be taken as full-blown developmental principles of history. Rather, he explicitly aims to provoke contemporary historians to develop better ideas and frameworks. This is further evidence that his views should be seen as responding to contemporary debates rather than internal problems of his own philosophy only.



It would be a misunderstanding to view my foregoing considerations as a complete defense of Kant’s views. I tried to add an important facet to the existing interpretations. What this contextualization cannot explain (and, a forteriori, defend) are the strongly teleological claims of his views on human history, or their exact relation to his ethical theory. Even then, critics might either reject the very demand for a grand-scale model of human history, or at least claim that Kant’s sketch is useless for, say, current historical research.  But note that I have tried to reduce his claims about human social dynamics to their most simple, largely innocent basic points. Given this, and given the epistemological modesty of his claims about human development, perhaps things look better for a kind of reflection about the question of how we could give meaning to the fragmented masses of historical knowledge.



[1] Gatterer, Johann Christoph: “Vorrede von der Evidenz in der Geschichtkunde.” In: Die Allgemeine Welthistorie die in England durch eine Gesellschaft von Gelehrten ausgefertiget worden. Ed. by F. E. Boysen. Halle. 1767, vol. I, 1-38, here 12.

[2] Kö(h)ler, Johann David: Lectorem benevolum programmate de historia pragmatica. [Altdorf.] 1714.

[3] See Anonymous [Abbt, Thomas]: “Hundert und ein und fünfzigster Brief. Anmerkungen über den wahren Begrif einer pragmatischen Geschichte.” In: Briefe, die neueste Literatur betreffend 9 (No. 151, 1761), 118-125, here 119; Gatterer, Johann Christoph: “Vom historischen Plan und der sich darauf gründenden Zusammenfügung der Erzählungen.” In: Allgemeine historische Bibliothek 1 (1767), 15-89, here 84; Anonymous: “J. M. Schröckh, Christliche Kirchengeschichte.” In: Königsbergische Gelehrte und Politische Zeitungen (No. 79, September 30, 1768), 315f.

[4] Anonymous: “Hundert und ein und fünfzigster Brief”, 118f.

[5] See e.g.: Köster, Henrich Martin Georg: Über die Philosophie der Historie. Giessen. 1775, 9 and 14.

[6] Gatterer, “Plan”, 79f.

[7] Gatterer, “Plan”, 84f.

[8] Gatterer, “Plan”, 27.

[9] Gatterer, Johann Christoph: “Ueber die Philosophie der Historie, von H. M. G. Köster.” In: Historisches Journal 6 (1776), 164-166.

[10] Gatterer, “Plan”, 77f.

[11] Bolingbroke, Henry St. John: “Letters on the Study and Use of History” (1735). In: The Works of Lord Bolingbroke. Ed. by H. G. Bohn. 4 vols. London. 1844, vol. II, 173-334, here 177.

[12] Schroeckh, Johann Martin: Christliche Kirchengeschichte. Frankfurt a.M. 1768, vol. I, 251-278.

[13] Schroeckh, Kirchengeschichte, 264-275; Gatterer, “Plan”, 86-88.

[14] Schlözer, August Ludwig: Vorstellung seiner Universal-Historie. 2 vols. Göttingen. 1772-73, vol. I, 14. – Schlözer rejects to characterize his approach to universal history as a pragmatic one, at least in the sense of giving practical lessons to the reader – these, the reader should draw himself (ibid., vol. I, 26).

[15] Schlözer, Vorstellung, vol. I, 30. Similarly Schroeckh, Johann Martin: Lehrbuch der allgemeinen Weltgeschichte. Berlin. 1774, 24f.

[16] Schlözer, Vorstellung, vol. I, 80f.

[17] Anonymous: “D. Hume, Geschichte von Großbritannien. Dt. Übers.” In: Neue Zeitungen von gelehrten Sachen 59 (July 23, 1764), 467f.

[18] Hume, David: Enquiries Concerning Human Understanding and Concerning the Principles of Morals (1748-51). Ed. by L. A. Selby-Bigge (3rd ed. by P. H. Nidditch). Oxford. 1975, 83f.

[19] See, e.g., Gatterer, “Plan”, 84f.; Schlözer, Vorstellung, vol. I, 15 and 19; Schroeckh, Kirchengeschichte, 275-278.

[20] Anonymous: “D. Hume, History of Great Britain, Vol. 1.” In: Göttingische Anzeigen von gelehrten Sachen 147 (December 8, 1755), 1350-1354, here 1350f.

[21] Köster, Philosophie der Historie, 55-62.

[22] Schlözer, Vorstellung, vol. I, 29f.

[23] Herder, Johann Gottfried: “A. L. Schözers Vorstellung seiner Universal-Historie.” In: Idem: Sämtliche Werke. Ed. by B. Suphan. Berlin. 1877ff., vol. V, 436-440. – Idem: Auch eine Philosophie zur Geschichte zur Bildung der Menschheit (1774). In: Ibid., vol. V, 555.

[24] V-Anth/Fried, AA 25: 472. These references point to the German original in the Academy edition (Kant, Immanuel: Gesammelte Schriften. Ed.: Vol. 1-22 Preussische Akademie der Wissenschaften, vol. 23 Deutsche Akademie der Wissenschaften zu Berlin, vol. 24ff. Akademie der Wissenschaften zu Göttingen. Berlin 1900ff.).

[25] V-Anth/Mron, AA 25: 1212; see also V-Menschenkunde, AA 25: 857f. – As to how far Kant knew the works of relevant historians, see Sturm, Kant, 332-338.

[26] GMS, AA 04: 417, footnote.

[27] For detailed textual evidence for the following points, see Sturm, Kant, 429-446.

[28] Anth, AA 07: 121.

[29] IaG, AA 08: 30.