In her impressively wide-ranging book For the Love of Metaphysics, Karin Nisenbaum takes on the challenging task of exploring Immanuel Kant's conflict of reason through selected post-Kantian responses to this conflict.1 The result is a highly insightful tour de force through the complex philosophical systems of Friedrich Heinrich Jacobi, Salomon Maimon, Johann Gottlieb Fichte, Friedrich Wilhelm Josef Schelling, and Franz Rosenzweig, driven by a deep systematic interest in the challenge of nihilism and in ways of restoring the idea of God or of the All of reality. The book not only demonstrates in an exemplary way the relevance of the tradition of German idealism for questions of metaphysics and the limits of our intelligibility, but also encourages the appreciation and revaluation of figures who tend to be overlooked in the grand narrative from Kant to Georg Wilhelm Friedrich Hegel.

A particularly rich and illuminating chapter in this respect is Chapter 2, in which Nisenbaum discusses the legacy of Maimon and highlights his decisive role in the emergence and development of post-Kantian German idealism. My comments focus narrowly on her discussion of Maimon, in this chapter, in relation to Kant's account of discursive cognition and the demands of reason for systematic unity. Nisenbaum manages to present Maimon both as a sharp critic of Kant's overall project of transcendental philosophy, as it is presented primarily in Kant's Critique of Pure Reason, and as an original thinker who pushes Kant's project beyond its own inevitable limitations in a highly fruitful way.

In Chapter 2, Nisenbaum examines two demands with which Maimon challenges Kant's transcendental philosophy: first, the demand that all a priori knowledge, that is, all transcendental conditions of experience, including both the forms of sensibility and the forms of the understanding, be systematically derived from a single first principle, namely the principle of determinability, and second, the demand that philosophy may not remain merely an empty formalism, but that it be actualized by tracing real contents or real conditioning relations through the real activity of the thinker or philosopher.

These two demands, as Nisenbaum shows in the chapter under discussion, lead Maimon to develop his own philosophy by going beyond Kant's philosophy in two respects. First, they lead Maimon to overcome Kant's dualism between sensibility and the understanding and thus between the two orders of intelligibility, the order of appearances and that of things in themselves. Second, they lead Maimon to reconceive the discursive nature of human cognition, which for Kant makes it different in kind from divine cognition. Instead, Maimon shows its in-principle-identity, or more precisely, its identity-in-difference, with divine cognition. In consequence of her discussion of both demands, Nisenbaum concludes that Maimon's philosophy opens up the possibility for a Maimonian rereading of the argumentative structure of Kant's Critique of Pure Reason—a rereading that would help Kant to counter Maimon's critique of his philosophy, but at the cost of a radical revision of his discursive account of human cognition. What remains unclear to me, however, is the extent to which her proposed rereading is a radical departure from Kant's original philosophy such that one can claim that

Maimon radicalizes Kant's account of the regulative use of the ideas of pure reason. [LM 98]

With the following comments, I would like to invite Karin Nisenbaum to elaborate on the details of her Maimonian rereading of Kant and to suggest an expansion that shows it to be genuine to Kant's own philosophy without requiring a fundamental revision of his discursive account of human cognition. Rather, in my view, this requires a radical reassessment of the role that ideas of reason play for human cognition. My key thought here is that one needs to reread Kant's Analytic of Concepts through his own Appendix to the Dialectic in the Critique of Pure Reason, in which he develops a positive role of his ideas of reason for empirical concept formation and for the truth-evaluation of empirical cognition. Nisenbaum herself points out the importance of Kant's ideas and their regulative use for her rereading of Kant's Transcendental Deduction. However, I would like to take her proposal further by distinguishing between two regulative functions of ideas: one function that relates to the human intellect in its pursuit of the systematic unity of discursive human cognition and another function that projects as the ultimate goal of human cognition the world as it is in itself and thus as it would be grasped by a divine intellect. Only if one carefully distinguishes between these two fundamental tasks of ideas, I argue, can one see how Kant's philosophy can finally reconcile his initial dualism between sensibility and the understanding, and subsequently explain the transition from the order of appearances to the order of things in themselves. And one can then perhaps recognize a relationship between the human and divine intellect that, while still different in kind, shares fundamental features, in a way that is comparable to Maimon's identity-in-difference account.

Kant argues that his transcendental philosophy addresses quid juris questions, rather than quid facti ones. That is, his arguments show that one has the right to apply certain a priori concepts, especially the categories of the understanding, to sensible intuition to arrive at experience, or empirical cognition, of sensible objects. These arguments do not establish any positive factual knowledge about things in themselves, however, they presuppose as a fact of human consciousness—at least according to a common regressive interpretation—that one can have actual experience of such objects in which such a priori concepts are used. Thus, while one would not have the right to determine things in themselves a priori (although one may be able to think them), which is an aspect of Kant's transcendental idealism, one can rest assured of Kant's empirical realism, which includes the possibility of experience of sensible objects.

Maimon, by contrast, aims to show that, while one must doubt—as the Humeans do—whether actual experience of objects is possible (for us) or not, one must presuppose, following the principle of determinability (or of sufficient reason)—as the Spinozists or the Leibnizeans do—the complete a priori determinability of transcendental reality and hence the possibility of real concepts based on a real use of the understanding. For Maimon, the determination of this supersensible reality is not a distinct realm of things-in-themselves, but the realm of the rules that govern the sensible, which he then identifies with the ideas of reason (Vernunftideen): for Maimon,

noumena are ideas of reason serving as principles to explain how objects arise according to certain rules of the understanding.2

Maimon then answers the quid juris question, as Nisenbaum shows, by developing a new conception of space and time according to which space and time as forms of sensibility are derived from the same source, or common principle, as the categories, which are the forms of the understanding, and therefore "space and time are the conditions for all real thought" (LM 79). However, one can never be sure whether actual human perception fulfills the rules of real thought; Maimon is thus denying Kant's answer to the quid facti question. Hence, Maimon combines a rational dogmatism with an empirical skepticism, which prima facie seem to be opposed to Kant's transcendental idealism and empirical realism.

In developing her Maimonian rereading, Nisenbaum first follows Béatrice Longuenesse in arguing that, in the Transcendental Deduction, the figurative synthesis of the imagination explains how space and time are unified in accordance with the categories such that whatever is given in space and time will be determinable by the categories, too. Nisenbaum then argues, in a Maimonian spirit, that this synthetic unity of space and time can only be fully understood if the transcendental ideal of pure reason, as it is introduced in the Transcendental Dialectic, is also applied, at least in a regulative way: that is, the assumption that the manifold of empirical cognitions will form a system, or systematic whole, and that each individual spatiotemporal object can be thoroughly determined through real concepts. The complete determination of objects that Maimon envisions thus becomes a regulative goal for Kantian cognition, as well.

While I agree in general with Nisenbaum's strategy to reread the Transcendental Deduction through the Transcendental Dialectic, I would like to hear more about the details of the regulative use of ideas of reason. Expanding on Nisenbaum's reading, I would like to suggest that one can give a genuine interpretation of this regulative use in Kant that bears strong resemblances with Maimon's conception of ideas. Such a reading would show why the regulative use of ideas is "indispensably necessary" in defining the goals of human cognition,3 while responding to a skeptical residue that still plagues Kant's theoretical philosophy.

Going back to the quid facti question, one can see that Kant and Maimon have two different issues in mind. Kant's transcendental arguments presuppose the possibility of actual experience, say of a causal series, yet it does not presuppose the actual truth of a particular cognition of an empirical causal law. That is, for Kant, falsehood of empirical cognition is possible, as one can be mistaken concerning the point whether the sequence of observed states really follows a particular causal rule.

Maimon, however, rejects Kant's answer to the quid facti question as for him the falsehood of real thought seems impossible, that is, one is in the possession of actual experience or cognition only if one has the true determination of an object, namely, if one cognizes the correct causal law that governs the observed event. With his critique, Maimon indeed points to a skeptical residue in Kant's account of empirical cognition: How can one ever be certain that one's actual empirical cognition is in fact true of the perceived object?

Kant rarely talks directly about truth in the Critique of Pure Reason, but the topic makes a central appearance in the Appendix to the Dialectic: reason's demand of systematic unity offers humans a "sufficient mark of empirical truth" (CPuR 595, A651/B679) or "touchstone of truth" (CPuR 593, A647/B675). While Kant presupposes a correspondence theory of truth, as he confirms in several places (for instance, CPuR 685, A820/B848), he acknowledges that humans lack sufficient criteria for assessing truth in particular cases (for instance, CPuR 197, A58/B82). If one seeks only the correspondence between empirical cognition and the object of experience, that is, empirical truth in the order of appearances, then the question arises as to how one acquires the real content of the empirical concepts one uses and ensures that they trace real conditional relations in the objects to which they are applied.

I have argued elsewhere, that one cannot read off the marks of an empirical concept from a particular experience; the senses alone cannot provide one with the real content of concepts, nor can the understanding do so.4 Rather, as Kant argues in the Appendix to the Dialectic, the formation of empirical concepts requires reason's demands of systematicity, such as the principles of homogeneity, specification, and continuity. These principles guide the complex activity of concept-formation that involves not only sensibility and the understanding, but presumably also the imagination in filling in the gaps of one's actual perception and producing the content of empirical concepts according to the rules of the understanding, but also most importantly in line with reason's demand to seek the totality of conditions. Empirical concepts are thus only meaningfully defined within a conceptual system and through their relationship to superordinate, subordinate, and neighboring concepts. Only if empirical concept-formation follows the rational standard of systematicity can one hope to track down real conditioning relations and gain true empirical cognition. Therefore, the demand for systematicity is a mark of empirical truth.

However, a deeper question arises for Kant: empirical truth, defined as the correspondence between empirical cognition and sensible objects, remains fundamentally constrained by the ways in which humans conceptualize the world of appearances and is therefore bound to a conceptual system that one presupposes in the assessment of empirical truth. It seems that humans can never step outside their own systems and see whether they are indeed on the right track with the whole system of concepts and the corresponding system of cognitions that they have developed in a particular science, for example. This means that if one can only decide whether a newly found law of physics, for example, is true by presupposing the common system of physical concepts and the system of already established laws of physics, scientists will never be able to find out whether the entire science of physics is in fact corresponding to something that is ultimately independent of the human mind and human conceptualization. Hence, the question arises as to whether there can be a correspondence between the systems of human cognition and something unconditioned, that is, whether there can be truth in an absolute sense or what one may call "transcendent truth." This is a common problem for proponents of coherence theories of truth, which rely on criteria of coherence (or systematicity) to define what is true at all, and therefore ultimately deny absolute truth. And this problem points to the skeptical residue in Kant's account of human cognition.

Now, the second function of ideas of reason comes into play: these ideas not only provide one with coherence criteria for the assessment of empirical truth, but also project the ultimate goals for the entire systems of concepts and systems of cognition. They direct "the understanding to a certain goal" by projecting a focus imaginarius (CPuR 591, A644/B672). On my view, this can be understood as the projection of placeholders for the unconditioned, for things-in-themselves or the world as it is in-itself independent of the human mind. This is the world that learned people should ultimately aim to describe by way of physics and other sciences, and which all must hope to aspire toward asymptotically, for the ideas of reason impose a priori constraints on those worlds that are ultimately intelligible to the faculty of human reason.

It seems plausible to me that Kant assumes that the human and divine intellect share the same generic faculty of reason, even though it plays different roles in human and divine cognition. Since human cognition depends on sensible intuitions, reason primarily plays the role of setting rational standards for the activities of the understanding and the imagination, whereas in divine cognition reason is presumably capable of the intellectual intuition of real things (in accordance with its concepts or ideas). This would mean that the world humans project by means of their ideas of reason could very well be the world that a divine intellect would cognize by means of intellectual intuitions.

Thus, by distinguishing between two functions of ideas of reason one can see that Kant's faculty of reason, on the one hand, helps to reconcile the dualism between sensibility and the understanding, and, on the other hand, explains the transition from human to divine cognition. First, in their semantic function, ideas of reason are involved in a creative activity of producing the real content of empirical concepts, which must be guided by reason's demand for systematicity. Second, in their epistemic function, they project the world as it is in-itself as the ultimate goal of human cognition, against which one would ultimately have to measure the truth of the entire systems of cognition, at least with respect to scientific systems, even if humans can only ever approximate this goal.

If my reconstruction is plausible, then it seems that both Kant's and Maimon's account of human cognition appear to be even closer in spirit than Nisenbaum is willing to admit, even if not in letter. In letters, they seem to place their emphasis on different aspects of human cognition—Kant on the interplay of sensibility, imagination, and the understanding in the figurative synthesis in the Transcendental Deduction, and Maimon on the creative and conditioning power of reason, which for Kant appears only in the Appendix to the Dialectic. In both views, the divine and the human intellect can be recognized as identical in their dependence on ideas of reason, but as different in their use of these ideas in relation to intuition.

Does the distinction I propose for Kant between two functions of ideas of reason match onto the two different kinds of ideas in Maimon—ideas of the understanding (Verstandesideen) that function like fictions (Erdichtungen) for the creative formation of actual content of experience and ideas of reason (Vernunftideen), which serve as substitutes for real objects and are identified as noumena?

In their semantic function, Kant's ideas of reason seem to be very similar to Maimon's fictions insofar as they provide necessary regulative (normative) guidelines for our human creative activity in the formation of real concepts and the progressive determination of real content regarding an individual object. In their epistemic function, Kant's ideas seem to share similarities with Maimon's ideas of reason, insofar as both can be understood as a projection of a goal or target world as it would be known by a divine intellect, which the human intellect, however, can never reach. In turn, it seems that even Maimon cannot avoid a certain dualism between the human and the divine order of cognition. Despite Maimon's rational dogmatism, the infinite order of intelligibility remains a rational fiction from the human standpoint, albeit one that is presupposed as a transcendental reality.

In conclusion, Nisenbaum's highly insightful discussion of Maimon in relation to Kant not only opens up a revaluation and reinterpretation of Kant's philosophy, but also sheds light on the deep connections that run from Kant to Maimon and beyond, revealing a movement of thought whose internal dynamics emanates from Kant's thinking and yet necessarily transcends it. In this sense, Nisenbaum's book is a vivid exemplar of Maimon's concluding words in his Essay on Transcendental Philosophy, in which he quotes a Talmudic Saying: "the students of wisdom find no rest, neither in this life nor yet in the life to come" (TP 227).