, Volume 22, Issue 3, pp 291–314

An Introduction to Nicolai Hartmann’s Critical Ontology


    • Department of PhilosophyColby College
Original Paper

DOI: 10.1007/s10516-012-9184-1

Cite this article as:
Peterson, K.R. Axiomathes (2012) 22: 291. doi:10.1007/s10516-012-9184-1


Nicolai Hartmann contributed significantly to the revitalization of the discipline of ontology in the early twentieth century. Developing a systematic, post-Kantian critical ontology ‘this side’ of idealism and realism, he subverted the widespread impression that philosophy must either exhaust itself in foundationalist epistemology or engage in system-building metaphysical excess. This essay provides an introduction to Hartmann’s approach in light of the recent translation of his early essay ‘How is Critical Ontology Possible?’ (1923) In it Hartmann criticizes both the pretensions of epistemology as well as the principal errors of classical ontology, and he proposes a series of correctives that lead to his development of a highly original and elaborate stratified categorial ontology. This introduction explains the most important errors of the ‘old’ ontology, his correctives to them, and further fleshes out these correctives with reference to his mature ontological work.


CategoriesCritical ontologyCritiqueEpistemologyIdeal beingLogicismNicolai HartmannOntologyReal beingReductionismSpheres of beingStratificationTeleology

In a recently published book, the late Wolfgang Harich staged a series of dialogues between himself and the fictive interlocutor Paul Forster, in the course of which he attempted to come to terms with the “greatness” (in both extent and value) as well as the “limitations” of Nicolai Hartmann’s philosophy (Harich 2004).1 The eighth dialogue—which concerns the relationship between Heidegger and Hartmann, among other things—begins with the question “who should be credited with the title ‘founder of the new ontology’ in the twentieth century?” Forster remarks that on the basis of the chronology of their publications it looks like Heidegger should get the credit, since Being and Time was published in Husserl’s Jahrbuch in 1927, while Hartmann’s first major ontological text Zur Grundlegung der Ontologie did not appear until (1935) (hereafter GO). Harich points out that this superficial chronology overlooks the fact that the “fundamental ideas for his ontology” already make an appearance in his 1921 Grundzüge einer Metaphysik der Erkenntnis, and are also “the central theme of his contribution to the Festschrift for Paul Natorp of 1923” (Harich 2004, 163).2 The full title of the essay just referred to reads “How is Critical Ontology Possible? Toward the Foundation of the General Theory of the Categories, Part One” (hereafter HCOP), and Harich notes that in the subtitle one can see that Hartmann is already dealing with the essential theme of his 1939 Aufbau der realen Welt, itself subtitled “Outline of General Category Theory.” While Harich does not mention it, we could also add that another essay called “Categorial Laws” and again subtitled “Toward the Foundation of a General Theory of Categories” is published in 1926 in the Philosophischer Anzeiger. Even more than the first essay, whose aim is the largely critical task of revealing and correcting errors, the second essay develops what becomes Hartmann’s most original contribution to the history of ontology, the description of second-order “categorial laws” or “laws of stratification” that display the overall “structure of the real world.” We therefore agree with Harich that Hartmann deserves the credit as leader of the ontological pack.3 Nevertheless, while Hartmann has priority in terms of time there is no doubt whose ontological project became dominant in many circles. Since at least one reason for this has been the lack of translations of Hartmann’s works into English, the accompanying translation makes a small contribution in this regard. We hope that the current century will know more of Hartmann’s work than the last.

Testimony not only to the remarkable consistency of his thought but also to the importance of these two early essays in ontology is the fact that these two pieces are taken up in full, only slightly reorganized and elaborated, a decade and a half later as two of the five major divisions of the 1939 core of his ontological tetralogy, Aufbau.4 Therefore, as the first publication where his distinctive position is clearly and forcefully presented, the text translated here is doubly significant. Joseph Klein has suggested that “from this essay one has the best and easiest access” to Hartmann’s ontology and theory of categories (Klein 1952, 124). It is hoped that a fresh presentation of his views in the present philosophical climate will encourage renewed interest in Hartmann’s philosophy. In addition to the significance that this particular work embodies for his oeuvre, its invigorating tone, clear language, and critical incisiveness are a refreshing change of pace for the contemporary reader.

My aims in the remainder of this introduction are simple. After a few remarks on the various senses of ‘critique’ used here, and a clarification of some of the claims made by Hartmann pertaining to epistemology and ontology in the first section of his essay, I’ll move on to discuss some of the most serious errors he identifies in the history of ontology. I will close with Hartmann’s discussion of the “most important errors of metaphysics” and “most serious obstacle to category research,” the rectification of which may be said to be the guiding aim of much of his ontological labor. Hartmann himself presents the errors of classical ontology by moving from “the historically older and more naïve prejudices and ascend[ing] to the more differentiated and theoretically conditioned ones,” but for the sake of systematically simplifying the presentation I will consider these errors to be clustered around two central terms, ‘category’ and ‘stratification.’ Any new ontology that takes these errors seriously and aims to avoid them has to fulfill a host of “requirements” that are appended to his discussion of each error. I will briefly describe how he himself fulfills these requirements for the two clusters of errors with reference to his mature work, principally Aufbau der realen Welt. Since the first group of errors turns on misconceptions surrounding the notion of an ontological category, I’ll explain Hartmann’s eminently distinctive notion of an ontological principle and of its categorial partner the concretum. The second group of errors involves a cluster related to the central error of Grenzüberschreitung or “boundary crossing,” and I’ll suggest that fulfilling the requirement to avoid this error motivated him to elaborate his original conception of the ontological laws of stratification (already developed in 1926). The laws of stratification are Hartmann’s response to Grenzüberschreitung and related errors, and this is why it is given a prominent place in this introduction.5 Since this piece intends to be introductory, it will not take up larger questions about the place of Hartmann’s work in the history of ontology or its contemporary relevance.6

1 Why ‘Critical’ Ontology?

There are at least three reasons to call the ontological stance developed by Hartmann ‘critical.’ First, as a former student of Cohen and Natorp in Marburg, Hartmann has a complicated relationship with Kant’s critical philosophy. He strongly resists neo-Kantian appropriations of Kant which take him to have put an end to metaphysics altogether and to have developed nothing but a rigorous transcendental foundation for the positive sciences. In contrast, Hartmann argues that Kant opened the path beyond epistemology to a new critically founded ontology. He explains in the essay that, correctly understood, the “transcendental deduction” is ample evidence of this. In asking about our epistemological limitations Kant was also asking a question to which philosophers had always assumed they knew the answer, namely, “How does thought relate to things?” For the most part principles (forms, essences) were taken to relate to things through logical subsumption, as the universal subsumes the particular. Thus the question whether the categories used actually can apply to the objects at all never arose, since if principles are equated with formal structure alone, and formal structure with logical structure, the principles of logic may be directly taken to be ontological principles. In questioning the pretensions of pure reason Kant asked whether the categories could apply to objects at all, and demanded that we justify or legitimize (‘deduce’) our application of concepts or categories to objects. Hartmann saw in Kant’s deduction the recognition that there is a gap between the subject’s thinking and the ontological structure of the world, and that the question of the categories bears this ontological weight. Thus, critique does not put an end to all ontology or metaphysics, it only puts an end to ontologies that take for granted thought’s relationship to reality. Thus a “critical ontology” in this first sense is one that has learned the lesson from Kant that we cannot take thought’s relation to reality for granted.7

This leads directly to the second sense of ‘critique.’ While Hartmann recognizes a debt to Kant for teaching us how to critically pose the ontological question, he cannot accept Kant’s answer to the question of the relation between thought and reality because he claims that it is too thoroughly colored by his idealist “standpoint.” That is, even Kant’s critique was not critical enough of the historically transmitted presuppositions contained in it, so a more radical excavation and critique of these presuppositions is required. Such a critical operation has both negative and positive aspects:

We have to take up the task of a new and more radical critique—not only of ‘pure reason’ insofar as it harbors the a priori presuppositions of positive science—but a critique of the categorial formation of our ontological consciousness and overall consciousness of the world. […] It has always been the meaning of critique […] to recognize prejudices as such, to safeguard the positive ideas hidden behind them, and to disconnect them from these prejudices. [… I]t frees these ideas from the deformity imposed by the prejudices, and gives them back their elemental form in its utmost purity […] the labor of critique is eminently positive (A 64, 67).

So the second aspect of a critical ontology is an obligatory critique of traditional errors, because “the discovery of every source of error at the same time points the way toward the correction of it” (Ibid.). These corrections are indicated by the requirements that Hartmann appends to the end of each discussion of a criticized error.8
The third sense of ‘critique’ goes to the core of his ontology and attaches to Hartmann’s self-conscious retention of the term “category” to designate ontological principles. He explains that using the expression “category” is itself an instance of critique. He claims that it always remains equivocal whether our attempted formulation of an ontological principle in a category really captures the principle itself. The notion of a category reminds us that the principle and the concept of the principle (the category) are not the same—thus also recalling ‘critique’ in the first sense above. We have to resist the temptation, encouraged by the conventional opposition of idealist and realist standpoints, to create an exclusive disjunction between them. This difference, while “self-evident,” is at the same time “never directly demonstrable in terms of content” (my emphasis, A 13). This means that because the ontological stance is an analysis of contents, “…we cannot decide at first how much of the content found belongs to the existing thing itself, and how much of it belongs to the interpretation,” and any prior decisions about this are illegitimately made on the basis of a “speculative” assumption (i.e., an idealist or realist “standpoint”). Ontology is initially “indifferent” to idealist or realist resolutions to this ambiguity.

With categories we are not concerned with the question of the Dasein of what is, but with the side of its Sosein. This means that we are not concerned with the ways of being of what is—for these are the ways of Dasein—but with configuration, structure, and content. Categories are substantive principles, and therefore it makes no fundamental difference for them whether in their origin they are to be understood as self-existent ontological principles or as principles of the understanding. This difference is the most important that can be imagined for the ontological character of the real world, but not in terms of its substantive structure (my emphasis, A 14).9

This “substantive structure” is precisely what the historical and ongoing discourse of ontology aims to describe and understand. With his typical lack of fanfare leading up to important conclusions, Hartmann states plainly that we only ever have the side of the predicate and concept, not that of the independent principle, so “practically speaking we can leave this distinction behind” (my emphasis, Ibid.). Provided we remain critically aware of the difference between concept and principle, we need not be constantly reminded of it. This is what it means for critical ontology to be ‘this side’ of idealism and realism, since it does not assume that it can settle the question in advance of the categorial analyses themselves.10

2 Priority of Ontology to Epistemology

Hartmann’s stance ‘this side’ of the distinction between idealism and realism not only stems from his resistance to standpoints, but also from the recognition, shared with many thinkers then and since, that the epistemological relation between subject and object does not have an ontologically privileged status. Like the Lebensphilosophen, phenomenologists, historicists, vitalists, and pragmatists, Hartmann strove to reassert the priority of ‘life-world’ experience and argued that the cognitive relation is a derivative and impoverished mode of experience. Expressly repudiating the neo-Kantian form of constructivist idealism, Grundzüge einer Metaphysik der Erkenntnis (1921) announced that epistemology can only be rightly interpreted within this broader metaphysical or ontological framework.11 In the essay translated here we find remarks that communicate the central points of that work:

There is no question of knowledge without the question of being. This is because there is no knowledge whose whole meaning would not consist in knowledge of ‘what is.’ Knowledge is precisely the being-in-relation of a consciousness to something-that-is. The theory may very well show that the independently existing thing really does not exist independently. But the phenomenon of the relation is not disposed of on that account. It endures, outlasts any theory which denies it, and in the end stubbornly returns, unremedied and indeed irremediable (HCOP 269).12

For Hartmann the cognitive relation is only one of a group of “transcendent acts” whereby the human being makes contact with things in the world and reaches out beyond the bounds of consciousness. In loving and hating, willing and acting, experiencing, expecting, fearing, and hoping we intend something real and are directed at the real stream of events. Cognition, ranging from everyday perception to complex research, is just one among these acts, and not the first or most fundamental. In fact, cognition presupposes these other transcendent acts, and, in their service, relates itself to a surrounding world already disclosed by them (Hartmann 1949, 129).13 Epistemology enjoins a “reflective attitude” toward the world, and when this becomes habitual “it cannot find its way back into the natural relationship to the world; it expires in a criticism, logicism, methodologism, or psychologism estranged from the world” (GO 46).14
Secondly, cognition is at bottom an ontological relation between a subject and an object, both of which are defined with reference to the act of cognition, but neither ‘subject’ nor ‘object’ exhaust the ontological reality of either the human being or the concretum known. Being a knowing subject is just one aspect of the human being, which is also a physical thing, a living organism, a psychological self, a moral person and social agent. Likewise, an ‘object’ is the result of a process of selective ‘objectification’ of a fuller pre-reflective ‘transobjective’ being. In HCOP, he ties this distinction back to Kant’s doctrine of the Ding-an-sich:

[O]bjecthood per se is not exhausted in the ‘objects of possible experience’ according to Kant. There is beyond the latter the ‘transcendental object,’ which as such is not cognizable, because the totality of its conditions is not included in the ‘conditions of the possibility of experience.’ It is by means of the doctrine of the ‘thing in itself’ as noumenon that Kant outgrew his own characterization of the problem and prepared the point of departure for a critical ontology in the problem of knowledge itself. At this point his critique has become critical of itself. The doctrine of the thing in itself is its most critical accomplishment (HCOP 299).

This surprising conclusion is incorporated into Hartmann’s own distinction between the object known and ‘transobjective’ Ansichsein or being-in-itself.15 The question remains as to how the human mind can know “independently existing” beings once it is acknowledged that our concepts, images, or representations of objects are only selections from or limitations of this pre-objective being.16 At this point another aspect of Hartmann’s reinterpretation of Kant comes to the fore, namely his alternative reading of Kant’s “supreme principle.” Readers will recall Kant’s principle: “the conditions of the possibility of experience in general must at the same time be the conditions of the possibility of the objects of experience” (Kant 2007, A158/B197). Hartmann claims that Kant’s idealistic answer to the question, which makes these conditions “internal” to the cognizing subject, results from his own “dogmatic prejudice.” He believes a solution that “satisfies the same facts” can be proposed which places the principles or conditions of experience not within the “transcendental subject,” but “outside” of the subject in the wider reality of which both subject and object are parts.17

Evidently, a subject can only know about the determinations of an object heterogeneous to him if the inner principles upon which this knowledge rests correspond with the principles of the object. In the phenomenon of a priori knowledge, therefore, there lies an extraordinarily valuable starting point for the orientation of category research: a certain identity must exist between subject and object, and we cannot be dealing with two completely heterogeneous systems of categories (HCOP 297).

The at least partial identity between subject and object that conditions the possibility of knowledge results from the fact that both subject and object are determined by (some) shared ontological principles structurally superior to both. These principles are what Hartmann calls categories. As explained in the previous section, these categories ought not to be interpreted as mere “concepts of the understanding.” This means, however, that many ontological principles are not known a priori at all, but have to be investigated historically, empirically, and fallibly, as much as any other object of knowledge (Hartmann 1953, 14, 19–20).

Thirdly, the question remains concerning just how much coincidence or overlap there is between cognitive and ontological principles. He distinguishes first between the primary real and the ideal spheres, and within the ideal, the secondary spheres of logic, knowledge, and thinking. As he explains, it is crucial that we keep these spheres apart or we risk repeating the errors of the old ontology, many of which lead directly to idealist, deductivist, and teleological metaphysics. The ideal structures of thinking are evidently not identical with those of logic, since much thinking is illogical, and real structures are not identical with either, not least because they are temporal while ideal (cognitive and logical) structures are atemporal (though dependent on temporal ones). While we cannot go further into this problem here, it has to be kept in mind that cognitive categories of the knowing subject are understood to be recurrences in the cognitive sphere of ontological categories that pre-exist the cognitive relation, in the same way that the human being comes to take its place in an already pre-existing world (A 158). Hartmann’s mature epistemology does not posit a human-world dualism or correlationism, for “[t]he world is no correlate of anything” (GO 222). Hartmann never tires of repeating that ‘what is’ is indifferent to being known by a subject, just as the world is as indifferent to the presence of human beings on the earth now as it was prior to their presence. Hartmann’s late philosophical anthropology makes it clear that human knowers are embodied and embedded in an encompassing world, and that the conditions of human cognition include the ontological features of this world as well as the structure of the human being, and this entails that many cognitive categories are recurrences (in the stratum of spiritual acts) of fundamental and special ontological structures. This will become clearer in the discussion of strata in Sect. 4.2 below.

3 Ontology Between Relativism and Logicism

A final point should be made before reviewing the errors he finds in the old ontology. Ontology for Hartmann is also critical in the sense that it reveals irresolvable metaphysical antinomies in the structure of the world, and even in some of our common categories. This meant that Hartmann defended ontology not only on the basis of its evident priority to epistemology, but also recommended the theory of categories as an approach that would steer us between the Scylla of relativism and the Charybdis of logicism. Both of these views rest on metaphysical misunderstandings that he believed ontology could resolve.

Ontology has become suspect for two main reasons, he claims: its association with teleological metaphysics and with logicism or deductivism. He takes pains to distinguish ontology from both, and sees category theory as a means to resolve the errors that have led to these associations. “Ontology… in no way stands or falls with the teleological character of one’s image of the world” (HCOP 271). This way of putting it shows that Hartmann also sought to legitimate the project of ontology in the face of the kind of historicist relativization of philosophy itself represented by Dilthey. Historicism would reduce philosophies to a set of possible types of Weltanschauungen (naturalist, subjective idealist, or objective idealist), no one of which is more or less true than another. This would make ontologies just as relative to worldview as ethics or aesthetic judgments. Hartmann rejects this relativization of ontology on the grounds that this historicism itself, as a kind of relativism, is not only self-contradictory, it is also based on a typical metaphysical error of its own. Historicist relativism—which says first that all peoples and times express a worldview through which they understand reality to have a meaning, and that the task of the historian is to ‘understand’ this meaning—is itself implicitly teleological (A 88). The idea of a comprehensive ‘worldview’ results from the mistaken subordination of ontological categories to normative elements, namely, ‘meanings.’ This approach defines the real relative to the meanings attributed to things by human beings, collapsing the distinction between concept and thing, or object and transobjective being. If ‘understanding’ is a kind of knowing, and knowing aims at something that is, then it too must be ontologically grounded. If ‘meanings’ are on a par with representations, images, and concepts for Hartmann, then they too are not identical with the ontological contents whose meanings they express. All teleology and normativism inverts what he calls the “law of strength” by claiming that the higher categories (value, meaning, purpose) are more significant than others. Categorial analysis reveals this category mistake in the assumptions of strictly hermeneutic approaches (among other species of teleology), thereby showing that ontology can be and must be pursued free from even the subtlest teleological assumptions.18

Similarly, logicism or deductivism is the other “tantalizing view that has encumbered ontology with odium.” By means of “the presupposed identity of logical form and ontological form,” ontology becomes an a priori deductive procedure (HCOP 272–273). “[T]hinking does not need to follow the troublesome path of experience, but wherever it reaches it immediately grasps the essence of things that are” (HCOP 271). Just as much as—or perhaps even more than—teleological metaphysics, this perspective is a serious threat to ontology: “This standpoint is the root of all evil, it is radically false” (HCOP 274). By means of another categorial analysis Hartmann shows that the logical sphere, as a secondary sphere within the ideal sphere, is in a way subordinate to the real sphere.19 This step alone allows him to argue that much of the logicism and mathematicism of rationalist and strictly analytic approaches to ontology is both misleading and simply too poor in content to describe the structures of the real world. In addition, it does not approach the question of the relation between thinking and being at all ‘critically,’ but instead assumes that “in its structures thinking immediately reveals the structure of the real.” The application of the categories of the logical sphere to the entirety of the real sphere is a different form of the same category mistake identified above. Thus “the most important errors of metaphysics” take the form of Grenzüberschreitungen, whether of ontic strata or spheres.

Before discussing more of these errors in detail, a question arises that permits a comment on Hartmann’s methods. If many of the errors of classical ontology and metaphysics are a matter of illegitimate boundary crossing, then how are the boundaries crossed defined, and how do we know when a boundary has been crossed? For Hartmann the boundaries of ontic domains and spheres are defined in terms of phenomenological description and experience, including historical and scientific research. (I’ll return to this question in Sect. 4.2 below.) Categorial analysis of principles draws on the broadest set of initial data, and a stance of openness characterizes his ontology. As Helmuth Plessner appreciatively described it, Hartmann’s “ontology is only the theoretical expression of his willingness to recognize and take seriously even the things that have been written off, just as it is ontology’s methodological principle to face openly and without preconceptions every domain of the ‘world’ on offer” (Plessner 1952, 102–103). Hartmann places a good deal of weight on accurate phenomenological description, and claims that to a large extent the phenomena themselves suggest methods of approach to us, making us aware that we should not impose the same method on all types of object.20 Our approach is to some extent “predetermined for us by the species of object with which we are presented” (cited in Plessner 1952, 97). We’ll see below how this results in a pluralistic but nonrelativist conception of the structure of the real world. Hartmann defends ontology as a general category theory that eludes the critique of ontology as either a relativistically conceived total world view or as an absolute apriorism which requires no experience of the world.

4 Review of the Errors

The centerpiece of “How is Critical Ontology Possible?” is the treatment of the major errors of the old ontology, to which we now turn. Hartmann’s answer to the title’s question becomes clear in the course of his examination of these errors: critical ontology is possible by correcting or avoiding these pervasive and often hard-to-detect mistakes.

4.1 First Group of Errors: What is a Category?

While the influence of Aristotle and Kant is apparent throughout Hartmann’s work, it should be plain by now that Hartmann will develop neither an Aristotelian nor a Kantian theory of categories, and both come under critical scrutiny. The leading question in the first group of errors that I will deal with is “what is a category?” What Hartmann calls the errors of chorismos, homogeneity (homonymy), formalism, conceptualism, rationalism, and subjectivity belong to this group.

Possibly the most common error to which ontologists of all eras are susceptible is the notion that categories are simply concepts. This can be interpreted in an Aristotelian or a Kantian manner, both of which are devastating for category theory according to Hartmann. Under the heading of the Aristotelian Error (formalism and conceptualism), Hartmann describes how Aristotle’s understanding of beings as formed matter entailed the privileging of form as the active, determining principle, leaving matter the indeterminate. In turn, form was understood as essence, substance captured in a definition or concept. “This is the root of the attitude that has held ontology in the chains of logic for so long, as well as made ontology and logic ambiguous. Principles are concepts of principles, the forma substantiales is the essentia logica” (HCOP 285). Thus began the identification “Principle = Form = Concept” that constitutes the Aristotelian “double error.” The result is that ontology becomes a game of concepts rather than an investigation into being qua being, or “what is insofar as it is.” Although categories are understood as formal and conceptual, Aristotle does not take the further step of making the forms subjective, but considers them to be “in” the formed substance itself. He thereby avoids what Hartmann calls the error of chorismos (where principles, like Plato’s forms, are entirely separated from concreta), but he does again risk, as Plato did, the error of homogeneity or homonymy (where the principle resembles the thing to such an extent that they are hardly distinguishable).21 Hartmann claims that principles have to be different from their concreta if they are to have any explanatory power. If principles and concepts of principles are identical, and if the sphere of principles is further identified with the logical sphere, then individuals are subsumed under concepts that explain nothing. Deductivism in ontology makes it impossible to grasp the ontological structure of the real world. In order to rectify these errors first one has to insist on the modal disconnection of concept of principle from the principle itself, of thinking and being. Secondly, one has to overcome the bias that claims only formal elements (forms, relations, laws) can be principles or determinative factors in the world. Hartmann therefore argues for the surprising claim that many principles must contain a “factor of substrate,” a “material” side that determines the substantive character of concreta as much as their formal factors.

Categories that contain nothing substrate-like (something which cannot be reduced to form, law, or relation) will never be in a position to ground the entities, whose principles they are, in their full concreteness. […] We can only escape the tiresome dualism of form and matter once and for all if we incorporate the material factor into the principles (HCOP 286–287).

These two correctives, supplemented by thinking of principles as immanent to their concreta and yet distinct from them in their way of being, will start us on the path toward a new ontology.

The other fork down the road of the identification of principle and concept is the Kantian Error. Kant, like Aristotle, makes the mistake of identifying concepts of principles with the principles themselves. Given his peculiar kind of idealist bias, Hartmann argues, Kant also made the fatal mistake of believing these concepts to be the exclusive possessions of knowing subjects. Kant’s supreme principle stated that the principles of the experience of objects must be the same as the principles of the objects themselves, and this identity is assured if all of these principles inhere in the subject itself and are subsequently applied to the objects. Hartmann notes that while Kant is correct in arguing that principles of the subject and of objects experienced must be shared in order for there to be some a priori knowledge of objects, he is wrong to conclude that these principles must be entirely subjective. “It is a theoretical prejudice that the object itself (and not merely the cognition of the object) also has its principles in the subject. In fact, the object quite evidently has principles of its own prior to all cognition, and does not need to receive them first from somewhere else” (my emphasis, HCOP 288). The result of this bias is that the pair subject-object becomes identical with the pair principle-concretum. Above we showed that for Hartmann both subject and object are viewed as concreta determined by ontological principles that partially overlap and partially diverge. Thus, to correct this aspect of the Kantian error the proper orthogonal relation between the two pairs principle-concretum (“vertical”) and subject-object (“horizontal”) must be reestablished.

Linked to the errors of formalism, conceptualism, and subjectivity is the notion that principles must be known intuitively or a priori. Hartmann dubs this the Cartesian Error. With Descartes, principles or categories are not only formal and conceptual, they are held to be both simple and immediately intuitable. First, Hartmann borrows Aristotle’s notion that ultimate explanatory categories are not themselves explicable, and explains that

ultimate categorial elements are really something very dubious, something that cannot be directly grasped in any principle that can be exhibited, or in any group of such principles. It is only the more complex categories that can be grasped in an approximate way, but as soon as we precipitate the elements out of their complex forms they become ungraspable. What we take to be ultimate, still-conceivable elements are not simples (HCOP 294).

In other words, it does not follow that the most fundamental principles are also the simplest. That is likely a rationalist assumption. Take causality as an example. For Hartmann it is just one form of determination among others and has various component factors, including its linear series character, real temporality, process, and productiveness (to name a few), and the productive factor of causality—why anything is produced at all—is for us completely unintelligible.22 It is an explanatory concept, but is not itself internally simple. Second, the alleged simplicity of principles is also said to entail their a priori self-evidence, but here too there is a misunderstanding. “Principles of cognition do not at all need to be cognitions of principles. What is known through a principle is not the principle itself. They are […] the first grounds of cognition, but they are far from being the first things to be known” (Ibid., 295). Once one has recognized these unjustified biases here, and acknowledges the potential complexity of principles and difficulty in apprehending them, it becomes evident that “there is absolutely no reason to treat the principles of the object as more knowable than the concrete object itself” (Ibid.). This is another reason why every flavor of armchair, a priori ontology has to be abandoned, and a return made to the broadest possible basis in experience and “to the things themselves.” Hartmann argues that the method of the new ontology will resemble knowledge from experience, but will not amount to generalization or inference from particular cases. It will be what he calls “ex posteriori” (in contrast to both ‘a priori’ and ‘a posteriori’).

Principles are the structural elements of the concrete and can only be apprehended as such, and only in the concrete—or, more exactly, can only be gleaned from it by means of analysis and reflection, where an hypothetical element always clings to what is so ascertained (Ibid.).

Once this path of analysis has been followed to the principle, the principle has to be made self-evident itself, and this glimpse of principles will be everywhere mediated by the concreta and by our situated knowledge. This approach takes inspiration from Aristotle’s notion that what is immediately known ‘for us’ may be in itself highly derivative, and that we have to work up to discover what is first in the order of being. This whole process of apprehending principles is what Hartmann calls “categorial analysis.”

The positive claims resulting from his examination of these common errors can be summarized as follows: categories are not merely formal (logical), but may also contain factors of substrate. Categories are not mere concepts from which the real can be deduced, since concepts are attempts at grasping something (potentially unintelligible) beyond them. It follows that categories are not merely subjective either, and the relation between principle and concretum does not overlap with that of subject and object. Subjects and objects are both concreta in the real world determined by their own, and many shared, ontological principles. From what has been said it follows that categories are not known intuitively or a priori, but ex posteriori, from the concreta. They are immanent in them although differing from them. Categories are not necessarily simple, but may have their own internal complexity of factors, some of which may be entirely unintelligible. In the last part of this section I’ll expand on Hartmann’s conception of ontological categories with reference to later work.

First, ‘principle and concretum’ is a fundamental pair of contraries that are mutually implicating. ‘Fundamental’ means that they are categories that run through all the strata and spheres of the world, and that they are ‘contraries’ designates their reciprocal implication as well as distinguishes them from Hartmann’s two other types of fundamental categories, namely modal categories and categorial laws. Principles are immanent to their concreta, “inside” them, though he claims this image is not quite adequate (A 387). He explains that the relation between principle and concretum is a “primal being-in-one-another and only-existing-with-one-another” (A 71). The term “concretum” itself expresses “the binding together of many principles, or as a more precise etymology has it, their ‘having grown into one another’” (A 247). Principles are only ‘for’ concreta and do not exist independently of them. In a paradoxical formulation, he says that principles are “independent only in dependence upon the dependent [concretum]” (A 385). Perhaps more helpful than these descriptions are examples of the fundamental pairs of categories that Hartmann finds in the structure of the world. In addition to Principle-Concretum, other fundamental categories include Structure-Mode, Form-Matter, Determination-Dependence, Quality-Quantity, Unity-Multiplicity, Contrariety-Dimension, Substrate-Relation, and Element-Complex. As you can see from the discussion of the Principle-Concretum pair itself, the pair Determination-Dependence is implicated in it (concreta are determined by principles, the latter depend on concreta), as well as Unity-Multiplicity (the same one principle determines many concreta of the same type) and Contrariety-Dimension (the principle-concretum pair are contraries spanning a vertical dimension of determination). Discussion of the other fundamental pairs proceeds similarly, by elucidating their reciprocal implications and relations. Hartmann thinks that our minimal set of fundamental categories is not small, but relatively large and highly complex, a multidimensional “coarse-meshed net” of mutually implicating pairs that run through all of the strata and spheres, structuring contents.23 All principles are considered to be that upon which the concretum “rests,” and this is the basic relation of ontological dependence. Calling principles “conditions” of being and thinking has the advantage of indicating that they are each only partial determinants of concreta, and that principles never determine alone but only in concert, “grown together” in the concretum. They are valid without exception for all cases that fall under their domain. Principles are that by means of which we understand the concretum, even if only a determinate side of it, though this is a secondary (gnoseological) relation (A 247–248). Principles have to be different from concreta to be explanatory at all, and this is captured not only by their peculiar kind of relation to concreta but also by their complex relations to each other (explained in the next section).

As mentioned above, the pairs of contraries form only one group of fundamental categories. The modal categories (possibility, necessity, actuality) thoroughly investigated in Possibility and Actuality (1938) make up another group indispensably necessary for differentiating between the primary spheres of the real and ideal, as well as the secondary spheres. What he calls “categorial laws” are the last group of fundamental categories, and these are most important for understanding the stratified structure of the real world. These three groups of fundamental categories have to be supplemented by whole ranges of “special” categories that pertain to only some or even one stratum of the real world and its concreta, such as cosmological categories of temporality, spatiality, causality, process, and individuality, or organic categories such as adaptability, purposiveness, metabolism, self-restoration, and hereditary constancy. While we may continue to conceive of principles as being “in” their concreta, we must also contrast the vertical dimension of this relationship with the horizontal relationships between series of concreta with their own relations and serial forms of determination among one another. Only the combination of these dimensions of determination brings us to the conception of singular, fully determinate, unique individuals, the concreta that populate the real world. I’ll discuss the categorial relationships that form the elaborate structure of the real world in the context of the second group of errors.

4.2 Second Group of Errors: How do Categories Relate to Each Other?

Early in the twentieth century Bergson argued that organisms were so unlike nonliving physical things that one could not use the categories of matter and material relations to understand and explain them. Around the same time Husserl argued that ideal logical structures have their own distinctive nature, and that these structures could not be understood in terms of psychological laws. Likewise, Durkheim argued that ‘social facts’ such as social institutions existed that were not explicable by being reduced to individual psychological or biological principles. Hartmann would see in these very different projects different modes of responding to what he recognized as a serious and all-pervasive error. Attempts to reduce one type of structure to another were rampant in philosophy and the natural sciences. Hartmann calls such mistakes Grenzüberschreitungen, “boundary crossing” applications of a category that originates in a specific domain to a target domain where it is no longer legitimately applicable or explanatory. He saw that this widespread error could only be addressed by an equally generalized response, and this response is his fully developed theory of ontological stratification. Hartmann was highly sensitive to debates regarding pluralism and reductionism and created an ontological framework that aimed to be maximally inclusive and critical of reductionist programs of all kinds, whether materialist or idealist. Dealing with the problems of reductionism and heterogeneous categories is still at the forefront of philosophical labor today, and this makes a fresh look at Hartmann’s massive oeuvre all the more enlightening. Today, philosophers are frequently accused of having committed ‘category mistakes,’ and although he does not use the exact phrase, Hartmann may be credited with recognizing category mistakes as a ubiquitous philosophical affliction.

In New Ways of Ontology he singles out this error as one of the most frequent and devastating in the history of thought. Elsewhere he calls it “the most serious substantive obstacle to category research” (A 83). In another late essay he glosses it this way:

The most important errors of metaphysics consist in overstepping the limits of the validity of individual categories, i.e., in their ‘application’ to objects for which they are not appropriate. Categories are simply the tacit presuppositions which we make in our conception, interpretation, and judgment about the given, and every blunder which occurs here has immeasurable repercussions (Hartmann 1947, 500–501).

All of these texts recapitulate what he first argued in HCOP. I quote at length:

The most well-known example of this error occurs in so-called materialism, and is taken to a grotesque extreme. Physico-naturalistic categories are alleged to be sufficient to explain spiritual life, the phenomena of consciousness, of thinking, willing, etc. The same holds for every kind of biologism or evolutionism, where the same phenomena are to be explained using the categories of the organic realm. Throughout these accounts, the inadequacy of the categories lies in the fact that they are too poor in content, and do not touch the structural level of the concrete forms. This is not always the only reason for the inadequacy. In psychologism, for example, it is rather the converse, as far as in it things like the structures of knowledge or of thought (whose basic structure is objective) should be explained based on psychological elements. The situation is again different with logicism, which indiscriminately stamps all phenomena with the forms of the logical-ideal sphere. In a broader sense systems such as pantheism belong here too, which imposes teleological principles on nature; idealism of all kinds, which ascribes subjective categories to objects; personalism, which tries to understand all regions of phenomena on analogy with personal beings, and innumerable other standpoints. All philosophical approaches that are known by their very names as ‘-isms’ commit the same error in principle, no matter how much they may differ from one another in other ways. A kernel of truth lies in all of them, and the principles with which they operate legitimately hold for a small region of reality, but become illegitimate when applied to the whole (284).

In contrast to Bergson, for instance, who staged a dualistic conflict between matter and life and ultimately subsumed the former beneath the latter, Hartmann aimed to create a more inclusive ontology that embraces all of the phenomena of the real world without reducing one type to another, from quantum particles to galaxies, bacteria to natural selection, the unconscious and drives, labor, social life, history, the sciences, and the arts. The categories that we discover as we strive to understand all of these types of phenomena are to be regarded as distinctive categories that belong to each set of concreta and perhaps to them alone, and we must be cautious in applying them beyond their home domain until we have performed a categorial analysis that would legitimate this application. The significance of this general principle for Hartmann’s philosophy cannot be underestimated. In every case Hartmann refuses every attempt at reduction of complexity and aims to preserve as much as possible, while at the same time placing strict limits on the use of categories in target domains outside their source domain. In addition, although he tends to identify such errors as violations of boundaries between strata, his reference to logicism and idealism in the passage above shows that we can consider attempts to think of ontological categories as subjective concepts, for example, to be a species of boundary crossing. In this case it is the crossing of a sphere-boundary rather than an ontic one (NW 56). This is why I will discuss the errors of categorial identity, total identity, normativism (teleology), and monism under this heading as well.

The distinction between the ideal sphere and the real sphere has to be maintained just as much as the distinctions between the domains of physical, organic, mental, and socio-cultural reality. First of all, the error of ‘systematic monism’ can easily be recognized as a rationalistic bias, since the world presents us with a radical diversity rather than any clear indication of homogeneity and unity. Hartmann says here that “the demand for a primary unity, graspable in a central point, is a purely subjective postulate, a rationalistic atavism of human thinking” (HCOP 305).24 Secondly, normativism is the mistake of taking ontological principles to be normative or value-laden ones, and is linked to a persistent teleologism in metaphysics. This can be conceived both as an application of a uniquely human category (purposive action in light of values) to the whole of reality, and as the illicit conflation of ideal value structures with real ontological structures. Thirdly, the error of total identity makes the Eleatic mistake of completely identifying the domain of thinking and the domain of being, an identification that is as clearly mistaken as it is pervasive. We would never ask ourselves if our concepts corresponded to the object or principle if there was such an identity to begin with. The error of the a priori categorial identity of cognitive and ontological categories pertains directly to the relations between primary and secondary spheres. Dealing with errors relating to the partial coincidence of cognitive and ontological categories is more challenging.

Hartmann has argued that a nonpartisan phenomenology of cognition shows that a priori knowledge runs into distinct limits, and ignoring these limits leads to an “absolute apriorism,” which he calls the Leibnizian error. The fact that there is a limit to a priori knowledge means for Hartmann that while there must be a partial coincidence of cognitive and ontological categories, the domain of ontological categories is broader and richer than that of cognitive categories (HCOP 299). Although there must be some coincidence of the two series of categories,

individual categories—certainly the higher ones at least—are, in their role as cognitive categories, not the same as they are [in their role] as ontological categories. They bear the same names in both domains (such as space, time, substance, causality) and rightfully so, for there must be some underlying identical element in them, but their categorial content is in many ways different in the two cases (300).

For instance, if categories can be internally complex, then a category of the real may recur as a category of cognition with only one of its elements or factors.25 While the causal nexus does not operate in the process of knowing (since knowing is not spatial), both knowing and causal processes are nevertheless temporal, and this is one example of a partial identity or recurrence of categories.26 He claims that each category has to be considered in its roles as cognitive and ontological, and this is one of the tasks of categorial analysis. The task is complicated by the fact that we are actually dealing with four (or five) different spheres, not merely two: the ideal, gnoseological or cognitive, logical, psychological, and (four different strata within) the real sphere. Categorial analysis has to work out the multiple threads tying individual categories to each of these various spheres, in order to discern where, for what, and how far any given category is valid. This is not merely an epistemological affair, but lies at the heart of a critical categorial ontology. While these relations are some of the most challenging to understand throughout Hartmann’s work, it should at least be clear that we cannot be satisfied with any view that baldly asserts the absolute identity between cognitive and ontological principles.

As mentioned above, the error of boundary crossing proper pertains to violation of ontic boundaries. It is no secret that Hartmann conceives of these ontic domains of physical, organic, mental, and socio-cultural reality in terms of the vertical metaphor of stratification. Within the framework of these four strata he can thus call teleologism the error of taking categories from the highest stratum and applying them to lower strata, borrowing principles “from above” to explain the lower strata rather than understanding them in terms of their own principles. Vulgar materialism does the same thing in reverse, where categories “from below” are indiscriminately applied to all phenomena, right up to consciousness and social life. Vitalism does both: it generalizes the categories of life to all strata, from above in relation to the underlying material stratum and from below in the case of the higher psychological and socio-cultural strata. Thus, any attempt to privilege one set of categories over others and generalize them beyond their proper domain is a form of the error of heterogeneity. It is important to note that he applies this observation symmetrically, and it is not merely one more critique of materialist reductionism alone. As a more contemporary example, Hartmann would no doubt consider structuralism and (some forms of) poststructuralism to have made genuine discoveries in the linguistic domain, but then, as many other enthusiastic discoverers have done, generalized them to all domains, from kinship relations and mythology (Levi-Strauss) to the unconscious (Lacan) and society (Foucault). Blanket statements like “nothing is outside the text” (Derrida) are typical signs of this error of generalization. The generalization of the categories of systems theory to society by Luhmann would be just as open to Hartmann’s criticism. With reference to popular metaphysical views, Hartmann would likely complain that Whitehead’s process philosophy commits a series of category mistakes by generalizing what are fundamentally psychological categories (e.g., prehension) to all domains of the real.27

In order to avoid such errors there is a demand to respect the phenomena and “preserve categorial integrity” (A 83). What it will take to enable the preservation of this integrity is not only the maintenance of the distinction between real and ideal spheres (and the secondary spheres of thinking and logic), but also an elaborate theory of second-order categorial laws of stratification. In the following discussion it is crucial to bear in mind that the metaphor of strata applies to the relations between categories, not to the relations between concreta themselves. Let me briefly describe these laws here and his notion of ontological strata, as these are still poorly understood and often conflated with hierarchical, developmental, and other forms of multileveled ontology.28

For Hartmann categories may belong to all the strata (fundamental), some of them (e.g., spatiality), or just one (e.g., purpose). Above I discussed one group of fundamental categories that belong to all of the strata and spheres (pairs of contraries). Since Hartmann is determined to preserve as much categorial diversity as possible in his stratified scheme, he also needs to consider how the strata of categories relate to one another. This requires him to develop a theory of second-order principles to range over the categorial strata, namely principles that take the whole stratified structure of the real world as their complex concretum, and these are what he calls the categorial laws. Already almost completely worked out in 1925–1926, the laws of category relationships and stratification are what provide a comprehensive vision of the order and structure of the real world, and also provide clear limits to the application of categories to the concreta within it. The first two groups of laws govern the internal relations between categories themselves within strata (laws of validity and coherence), and the second two pertain to relations between whole strata of categories (laws of stratification and dependence). I’ll describe the first two briefly and expand more on the latter two. Given the limits of this introduction I cannot explain in detail the reasoning behind the formulation of these principles.

The laws of validity include and expand on the idea that the being of a principle consists in its being valid only for some concretum, as discussed in the previous section (4.1). The rest concern the scope of validity of the principles. The three further aspects of the laws of validity can be summarized this way: the determining power of a principle extends without exception throughout the whole stratum; this uninterrupted sway of a category only determines concreta at the level to which they belong; and a concretum of a given stratum is entirely determined by the principles of its stratum, it is “saturated” by them (A 382–383). Taking an organism as object of inquiry, for example, demands reference to biological categories (rather than physical or mental ones), and these saturate the organism with specifically organological forms of determination (nexus organicus), no more (such as teleology) and no less (such as physical causality). The laws of coherence concern the holistic character of each stratum. All categories of a given stratum determine the concretum in combination, never alone; all categories of a stratum are reciprocally implicated, forming a unity; all categories of a stratum form a complex whole that has a priority over the parts; and all of the categories of the stratum are implicated in every single category and can be discovered there (A 394). If we claim that metabolism belongs to a proper understanding of the organism, then aspects of every other category of organic life is entailed in it as well, codetermining and ingredient in it. With these two sets of principles he has covered the internal coherence of a stratum and determination between principle and concretum. These laws together already imply a high degree of incommensurability of categorial domains to one another, but given the all-pervasiveness of the fundamental categories this substantive incommensurability is never total. Relations between different strata are captured in the last two sets of structural laws.

The laws of stratification can be summed up in four key terms: recurrence, modification, novelty, and distance. Some lower categories recur in higher strata as partial aspects of higher categories, and every recurring category is modified in its recurrence (due to the laws of coherence). Whenever a lower element is taken into the higher it is affected by its new place in relation to others at the new stratum. These two principles constitute a vertical interconnectedness of the strata among themselves. While the categories of causality and substance, for example, appear to us initially in discussion of physical things, they recur in the domain of the organic. Because categories are not simples but complexes of factors, some of those factors may remain stable while others are modified, constituting the recurrence of the same but non-identical category. These recurrences have to be shown in each case, and one of the tasks of categorial analysis is tracing the modifications of a single category throughout the strata.29 Next, every stratum contains is own unique and novel categories that are not present in the lower stratum, nor are they a sum of them. Finally, recurrence, modification, and novelty imply that there is not a continuous series of levels but gaps, leaps, or incisions between them. The last two laws of novelty and distance are what give the impression of the ontological irreducibility of the strata (A 432). For instance, while the category of metabolism in the organic may necessarily incorporate some aspects of linear causal process, it is itself a distinctive kind of process that is irreducible to them. Thus recurrence and novelty respectively reflect the aspects of continuity and discontinuity among the strata. Without novel principles, there would be a categorial monism with only differences of degree of complexity in the world, and no room for difference of kind. Categorial novelty inserts an incision or cut into the apparent continuum of categories, creating a distance or gap between strata.

Two key concepts that complement his conception of strata laws are those of Überformung and Überbauung, superformation and superposition respectively. The metaphor of stratification does not on its own specify the exact type of relationship between strata. Superformation is not only reflected in some relationships between concreta (e.g., abiotic molecules become matter for superformation by the organism), but also in the relation between the whole stratum of the physical and of the organic.30 The simpler and more widespread form of relation is called superposition. Here the higher categories are in most cases existentially dependent on the lower as condition for their existence (e.g., we don’t know of any minds without bodies), but not for their content (bodies don’t simply determine what is in the mind), which in the case of the higher level has an independence from the lower (e.g., ideal value contents are in some ways independent of their bearers). These terms begin to describe the relationships of dependence of the strata on one another, the relationships that are most important for understanding the structure of the world (A viii). These laws of dependence can also be summed up in four terms: strength, indifference, matter, and freedom. The “fundamental categorial law” says that the lower categories on which higher strata depend are the “stronger,” they are conditions or fundaments, while the higher are “weaker,” in inverse relation. The lower are indifferent to whether anything higher ever comes along or not, their vocation is not to serve the higher. As “matter,” the lower categories, if incorporated into higher levels, constrain what the higher may do with them but don’t determine it (or they simply remain a fundament). Lastly, the higher always has leeway despite its weakness and dependence on the lower (A 471–472). Let me expand on these before concluding this section.

The relationship of supporting and supported can be found universally throughout the real world and is the core of the relation of dependence. The law of strength holds that the categories of the lower strata are stronger than those of the higher strata, their strength consisting in their greater “determinative power.” This does not mean that they are more likely to recur in higher strata, since even where they do not substantively penetrate into the higher strata they remain the foundation and basis for them (NW 88). This fact is reflected in the idea of the indifference of the lower strata to that which they support. The lower is never for the sake of the higher, even in the case where the higher superforms the level below. The physical world does not exist for the purpose of being taken up into the organic, just as both do not exist to service human beings.31 Each stratum exists for its own sake and is indifferent to what it is used for or what gets built on top of it (NW 90). The physical world is indifferent to the fact that life somehow emerged from it, just as the bulk of living nature could happily continue on without the existence of human beings. The question then arises how much the lower as matter can determine the higher. He uses the classical metaphor of the builder here to describe how the nature of the materials might constrain, but not positively determine, the form of the finished structure (NW 92). One can build quiet, well insulated homes from straw bales, but not towering skyscrapers. The builder’s genius can only do so much to transform the basic building blocks, it always runs into the constraints or intransigence of the material.

Finally, the law of freedom or autonomy means that regardless of the higher’s dependence on the lower it maintains its independence in relation to it, since at most only the matter of the higher could be determined by the lower, never its formal structure. Lower categories recur in the higher, but are then modified by them. This is what constitutes the superiority of the higher strata above the lower, their “priority of height.” Thus, there is simultaneously a cross-dimensional superiority of strength (though poverty of content) and of height (richness despite weakness) in the real world (NW 95). We can now see how the error of Grenzüberschreitung often involves a violation of these categorial laws and relations as well as simple category mistakes. For example,

[M]aterialist metaphysics…ascribes to the lowest ontological forms the power of producing the highest. It ignores not only the novelty of the higher ontological strata and their freedom but also the limitations of recurrence. It also ignores the limitation of the determining power of the lower categories in furnishing a ‘basis’ or, at most, a ‘matter’ (NW 98).

Materialism not only uses categories too poor in content to explain ontologically higher phenomena, but it also exaggerates the basic categorial law of strength and overlooks the gaps, novelty, and independence of the higher strata. This elaborate theory of second-order principles governing coherent strata of categories and their relationships of dependence and independence is one of Hartmann’s chief contributions to ontology and his response to the error of boundary crossing.

5 Conclusion

Many readers will undoubtedly have questions for Hartmann. Are categories really principles? Is an ontological reading of Kant possible? Aren’t categories historically and culturally relative? Isn’t this just one more attempt to achieve a “view from nowhere”? Doesn’t the nature and order of his strata merely reflect the state of the sciences at the time? Can the strata be reduced to a smaller number given recent scientific developments? Do higher strata “emerge” from lower strata or not? Aren’t stratification, boundary crossing, higher and lower, weaker and stronger, vertical and horizontal just metaphors? I think that although these are questions worth answering, there are few to which Hartmann has not prepared an explicit or implicit response. I hope that this introduction provides some of the basic concepts needed to discover or reconstruct the answers to such questions. I hope it also persuasively shows that Nicolai Hartmann was not only the first to develop (and not merely propose) a new ontology for the twentieth century, but also one of the most comprehensive and imaginative. “How is Critical Ontology Possible?” was the first in a series of publications that will assure Hartmann a place in the history of ontology.


The book is subtitled “Attempt at a Marxist Self-Understanding,” or, a bit more expansively, “An Attempt to Reconcile Hartmann with my version of Marxism.” Harich was, perhaps paradoxically, a life-long follower and friend of Georg Lukács, as well as a great admirer and former student of Hartmann in the 1940s. As the subtitle hints, he always hoped to be able to show how Hartmann’s work could provide a sound ontological framework for Marxism. The book was composed in the 1980s and was never published, and after 1992 the manuscript was given to Martin Morgenstern. For a description of the intriguing history and vicissitudes of the two volumes by Harich recently edited by Morgenstern, see the editor’s forewords to both: Harich (2000, pp. vii–xix); and Harich (2004, pp. vii–xxi).


In this dialogue Harich goes on to explain that Heidegger was familiar with these two early works, referring to them in a footnote in Being and Time. He goes further to claim that Being and Time had Hartmann’s approach to ontology as one of its major targets, and that opposition to Hartmann forms the subtext of many of the discussions throughout the book, in footnotes and other passages where Hartmann is not mentioned by name. He also remarks that Hartmann, after moving on to Köln, read Being and Time in full awareness that he was one of the targets of the critique leveled in it, and that in GO (and other works) he “struck back” against Heidegger (Harich 2004, 166).


It is also necessary to take into account the other contributors to the revivification of the ontological tradition at the time, and to consider Hartmann’s relation to them. While others “announced” the coming of a new ontology, in his own estimation Hartmann was the only one to have actually “carried it out.” There is some justification for this belief. His oeuvre is imposing, covers all major thematic areas of philosophy, and his four volumes of ontology alone span more than 2,000 pages. In this relation he mentions Hedwig Conrad-Martius, Günther Jakoby, Alexius Meinong, Max Scheler, and Heidegger (GO VII), and elsewhere Hans Pichler and Emil Lask. Hans Pichler is singled out as one who “strengthened [Hartmann’s] conviction that [he] was on the right track,” but he too did not develop the new ontology.


Corresponding pages in the Aufbau version of each text are pp. 61–156, and pp. 375–521. References to Der Aufbau der realen Welt: Grundriss der allgemeinen Kategorienlehre (1940) hereafter ‘A’ followed by page numbers. For an early discussion of “categorial laws” in English, see pp. 395–402 of the 1926 Ethics (Hartmann 1932, Vol. II), which he must have been working on concurrently with the 1925 essay “Categorial Laws.” With the exception of New Ways of Ontology (1943, hereafter NW) passages, all translations of Hartmann are my own.


Hartmann himself recognizes the central significance of this error when he devotes a whole section to it alone in NW 54–62. Obviously there may be many other historical and contextual reasons for the development of his theory of stratification aside from this systematic one, only some aspects of which are reflected in the brief treatment of this error in the essay. Harich (2004) is currently the best resource for placing Hartmann in the context of early 20th century debates over positivism, materialism, vitalism, psychologism, historicism, existentialism, Marxism, neo-Thomism, etc.


I have discussed some aspects of his contemporary relevance in Peterson (2010) and Peterson (2012). The person who has contributed most to current renewed interest in Hartmann’s work and argues for its significant contemporary relevance is the Italian philosopher Roberto Poli. Many of his writings on Hartmann can be found on his website,


For his more nuanced interpretation of Kant, see Hartmann (1924).


These are usually elaborated in entirely separate sections in the A version.


Dasein means “existence” for Hartmann, “that something is at all.” (He deliberately uses the term in a more traditional way in contrast to Heidegger’s idiosyncratic usage.) The “ways of being” [Seinsweise] are two: ideal being and real being, and he also calls them the two “primary spheres” of being. (The “secondary spheres” include those of logic, cognition [or knowledge], and thinking.) The two primary spheres are distinguishable principally through their different ‘modalities’ of being, which Hartmann exhaustively investigates in his Möglichkeit und Wirklichkeit (1938). There is more discussion of the spheres above. Sosein has been translated by some as “specific character,” and indicates the “what” of something. He glosses it here with “configuration, structure, and content.” It is important to note that he does not use Sosein as a synonym for “essence,” “form,” or “nature” in any traditional sense. It should also be noted that Sosein is not merely ideal while Dasein is real being, for ideal and real concreta each have a Dasein and Sosein of their own. Contemporaries like Meinong and Scheler also used the term Sosein, but each with their own meaning. For Hartmann’s thorough discussion of Sosein and Dasein, see GO 81–138. I leave the term Sosein untranslated.


Even if ontology will eventually have to come down on the side of realism, it is unlike any conventional type. “The expression ‘realism’ is not at all fitting for the position of ontology, which is also why none of the conventional types of realist classification is in accord with it” (GO 140).


“…[O]ntology is superior to epistemology [Gnoseologie] in that it takes up the whole breadth of the cognitive relation within itself and recognizes in it just one among other ontological relations” (Hartmann 1921, 185).


Throughout I will use the Kleinere Schriften page numbers, included in brackets in the translation.


“There is no such thing as ‘knowing’ isolated by itself. It appears only in the complex of other transcendent acts, in which the coherence of life consists” (GO 222).


Like Husserl, Hartmann distinguishes between a “natural attitude” toward the world and the “reflective attitude” adopted by logic and epistemology. These he calls intentio recta and intentio obliqua respectively. Unlike Husserl he thinks that the reflective attitude is not a sound basis for ontology, and takes ontology, as he does science, to be an extension of the natural attitude. See GO 46 and note.


He discusses this term further in GO. “‘Ansichsein’ is not a strictly ontological concept. It is only a defense and a line of demarcation against merely objective being. Epistemology needs this line of demarcation, it has to draw it for its own sake. For it and its reflective stance what is as such becomes apparent only in exceeding the object-relation. ‘Ansichsein’ is and remains a gnoseological concept. Ontology can do without it. Ontology has reverted to intentio recta from intentio obliqua. This defense is not needed for it; the correlativist prejudice does not belong to it. In its concept of ‘what is insofar as it is’ the concept of Ansichsein is already surpassed.” (GO 78–79).


“The greatest and most difficult of all metaphysical questions is precisely this: whether and how far thinking, with its own lawfulness, can even strike upon the essence of being” (HCOP 274).


See Hartmann (1921, 351–353), and Hartmann (1924), for his reasoning regarding this principle.


For a thorough critique of teleology in all its forms, see Hartmann (1951).


For a full treatment of the distinction between primary and secondary spheres, see Hartmann (1938).


Hartmann’s relationship to the phenomenologists is complex. For one view, see Spiegelberg (1960, 358–389). While he learned much from the early Husserl, Scheler, and the Munich circle, and recognized in phenomenology a valuable descriptive method that would help prevent precipitous category errors, he did not believe that phenomenology could on its own solve long-standing philosophical problems. See his critical remarks in, for example, A 535 note, and 536–537.


There is a much expanded discussion of this error in A 72–78.


For his analysis of the category of causality see Hartmann (1950, 318–330).


This position is already clearly stated in HCOP: “The only evident fact is that within the ultimate intelligible stratum all members are reciprocally conditioned by one another, such that in a certain sense each is the supreme principle and yet each is dependent upon all the others” (304), so that we have “a plurality of ultimate conceivable elements.” A quick review of his “rhapsodic” table of fundamental pairs of contraries reveals 36 altogether, and if we add modal categories (possibility, actuality, necessity) and their negative counterparts (negative possibility, inactuality, impossibility), as well as categorial laws (four groups of four), we end up with no less than 58 fundamental ontological categories. While it may be possible to describe the ideal sphere with fewer categories, we cannot understand the real world with a smaller number. In fact, the special categories stemming from specific domains of physical, biological, psychological, and spiritual phenomena must be added to the list. We may safely conclude that Hartmann’s ontological taxonomy is one of the most elaborate ever created.


He repeated this sentence with a few alterations in A: “The postulate of a punctual unity is a human-subjective rationalistic atavism of immature thinking” (143). He adds that ‘dualism’ is another version of the error of heterogeneity, an exaggeration of the significance of one categorial pair over all others (145–146).


“Each category that falls at all within the realm of identity simultaneously belongs to both spheres, the ontically real and the gnoseological actual sphere, but it spans this double allocation with only one part of its nature, in the other part it is split or torn apart by it. Obviously, the division for each category is also substantively different, such that an unlimited multiplicity of gradations between the extremes of full identity and of complete nonidentity is possible. Here is a new, as yet completely fallow field for research, undoubtedly rich with consequences, with whose disclosure and fruitful treatment the task of a critical ontology can first genuinely begin.” (300) To further clarify these relations would require a more detailed look at his epistemology than can be pursued here.


Hartmann does not always make a clear distinction between categories understood as principles that 1) determine both ontologically and cognitively, structuring both spheres while not being thematized by cognition (partially identical categories); 2) those that are used involuntarily or ‘intuitively’ by thinkers in order to understand objects; and 3) those that are more deliberately ‘applied’ by cognition to objects. The distinction between ‘intuitive’ and ‘applied’ categories is discussed in Aufbau, pp. 98, 114–115.


The explicit argument for doing so comes out clearly in Hartshorne (1969). He argues that we have to choose psychological over physical categories as basic principles because as fundamental categories the values of physical categories could never cover the range necessary to explain sentient beings. See pp. 111–123. Hartmann would approve of the fact that Hartshorne recognizes that two sets of categories apply to different domains of phenomena, but would charge him with Grenzüberschreitung when he generalizes the psychological set to cover the whole of reality. The same charge would likely be made against Whitehead.


This is one of the most frequent mistakes in interpretations of Hartmann’s philosophy. Real concreta are on a flat plane of tangled spatial and temporal determination and interconnection where their ontological principles are immanent to them. These concreta might be conceived, as in ecology, to occupy levels of a scalar hierarchy where every higher level incorporates elements from the lower. These relations between concreta are orthogonal to the conception of categorial stratification and do not coincide with it in any respect. I’ll discuss this further below. Even a reader as sympathetic as Peruzzi makes claims about “genesis” between levels that indicate an unfortunate misunderstanding of Hartmann’s theory of strata (2001).


For further discussion of strata laws see NW 73–83 and A 429–464.


This is the only “pure case of superformation” (NW 82) between strata, where the totality of the higher stratum takes up all of the categories of the lower stratum within itself.


This simple ontological observation is one of the roots of Hartmann’s generalized non-anthropocentrism. Hartmann uses it, and the previous law, to guard against teleological interpretations of the world that are explicitly or implicitly anthropocentric. “All teleology of forms […] makes the mistake of inverting the law of strength. It makes the higher categories the stronger ones. This corresponds to a certain dream image of the world, fondly framed by man at all times. It permits him to consider himself, in his capacity as a spiritual being, the crowning achievement of the world.” (NW 89–90). In contrast, he claims that our task is “to come to terms with a world not made for” us.



I would like to thank Frederic Tremblay and two anonymous reviewers for helpful comments on an earlier draft.

Copyright information

© Springer Science+Business Media B.V. 2012