I. The Systematicity of Reason: The Priority of the Whole of Reason over its Parts

Critical philosophy cannot fulfil its promise to make metaphysics systematic if it is not itself systematic. That is why, Kant insists, “a critique … is never trustworthy unless it is entirely complete down to the least elements of pure reason” (Prol 4:263). Critique must be a “perfect unity” or “unconditioned completeness” (KrV A xiii and A xx). It cannot be complete or systematic to some degree or in some respects. The systematicity of critique, Kant unequivocally states, is a matter of “all or nothing” (Prol 4:263).

To sketch his strong conception of systematicity, Kant often uses an analogy with an organized body. Through this analogy, Kant conceives of reason at two levels: the parts and the whole. The systematicity of reason demands thorough coordination of all parts of reason within the whole of reason. At the level of the parts, systematicity requires that all faculties of reason as well as possible relations among them are coordinated: “pure reason is such an isolated sphere, within itself so thoroughly connected, that no part of it can be encroached upon without disturbing all the rest, nor adjusted without having previously determined for each part its place and its influence on others” (Prol 4:263). At the same time, such tight coordination of the parts of reason entails their prior harmonization at a more fundamental level, namely, the level of the whole: in “pure reason,” “as with the structure of an organized body, the function [Zweck] of any member can be derived only from the complete concept of the whole” (Ibid.). Here, Kant presents the primary systematic principle of pure reason: the whole of reason must be prior to its parts. Like the parts of an organized body, each of which derives its function from, and thus works for, the body as a whole, in a system of reason, each part derives its function from, and thus works for, the whole of reason. In a system of reason, the faculties of reason can acquire their special functions only if they belong to and participate in the same unity or whole of reason. In this sense, the whole of reason is the systematic origin of all functions that the faculties of reason perform.

Despite his insistence on the priority of the whole of reason over its parts, Kant does not tell us much about how to conceive of the whole of reason. But before addressing this question directly, it is useful to 1) qualify Kant’s conception of the whole of reason in two ways, and 2) briefly discuss two major types of interpretation that ignore the priority of the whole of reason over its parts, and understand the unity of reason on the basis of one of its parts or some arrangement among them.

The first qualification is about Kant’s analogy between human reason and an organism. It is important to notice that Kant’s analogy does not make any ontological claim about reason as a whole. In other words, we should not take his analogy literally. Kant uses different metaphors strategically in order to communicate a methodological point about the systematic nature of a critical sense of reason. For example, in the introduction to the Doctrine of Method, he uses the architectonic metaphor of an “edifice” (KrV A707/B735) to articulate the methodological nature of “a complete system of reason” (KrV A708/B736). On this analogy, all parts of the edifice of reason should be designed according to, or derive their functions from, the master plan for the “edifice” as a whole. Similarly, in the first chapter of the Doctrine of Method, Kant uses the juridical metaphor of “legislation” (KrV A711/B739) to explain that all principles of reason with regard to objects must be derived from a single first principle, or law, by which reason disciplines itself. In short, what Kant aims to articulate with the help of different metaphors—whether organic or not—is the methodological nature of critique. Each metaphor is used to express the idea that a critique of reason is a complete system of possible cognitions of reason, which is necessarily derived from a single principle of reason. Kant makes this point in his brief account of “reason in general” at the beginning of the Dialectic, without the aid of metaphors: in its most basic sense, reason “contains the origin of certain concepts and principles, which it derives neither from the senses nor from the understanding” (KrV A299/B355).

The second qualification concerns the two related but distinct senses of the whole of reason in the passages quoted above from Prolegomena. Kant does not sufficiently distinguish these two senses of the whole of reason. In the first sense, the whole is the common systematic origin from which all uses of the faculties of reason and their possible cognitions derive. This is the sense of the whole of reason that is systematically prior to any distinction between the faculties of reason or their cognitions. Kant refers to this sense of the whole as “the complete concept of the whole” from which “the purpose of any member” must be derived (Prol 4:263). In its second sense, the whole of reason concerns the relations between the parts or faculties of reason and their coordination. In this sense, the whole must be “within itself … thoroughly connected” (Prol 4:263). Such thorough connection requires the harmony of the faculties of reason and their object domains within the first sense of the whole of reason. That all faculties systematically derive their possible cognitions from the same source, that all parts are within the same whole, does not mean that they automatically fit together. Their harmony requires not only presupposing the first sense of the whole but also additional principles to organize the parts within that whole. As the passages quoted from Prolegomena imply, the first sense of the whole is about the methodological or systematic derivation or origin of the use (cognitions) of the faculties of reason, and the second sense of the whole is about the internal coordination and organization of these faculties and their cognitions. The first sense of the whole of reason can be characterized as the ‘originary conception’ of the whole. As Kant indicates, this whole is complete in its originary purpose or conception, although its internal organization, which is necessary for multiple ways of pursuing its originary purpose, requires an additional level of organization. The second sense of the whole of reason can be characterized as the ‘internally organized’ whole. This whole is complete not only in its originary purpose and conception but also in the internal organization of the faculties of reason that pursue this purpose in relation to objects. In short, the internally organized whole of reason, which coordinates the uses (cognitions) of the faculties of reason, presupposes the originary conception of the whole of reason. This paper focuses only on the originary conception of the whole of reason.

Though they are implied in his writings, Kant himself does not clearly distinguish these two senses of the whole of reason.

On the one hand, there are places where Kant presupposes the unity or wholeness of reason. In these places, he seems to use the whole of reason in its first sense, as the originary conception of the whole of reason. For instance, in Groundwork, Kant writes: “there can be only one and the same reason, which must differ merely in its application” (GMS 4:391). See also (MS 6:207). In the second Critique, Kant speaks of “one and the same reason which, whether from a theoretical or a practical perspective, judges according to a priori principles” (KpV 5:121). In the third Critique, Kant presents reason as “our cognitive faculty as a whole” (KU 5:174) or “the pure faculty of cognition in general” (KU 5:176) prior to the division of this unitary faculty into three faculties of understanding, the reflecting power of judgement, and practical reason (KU 5:198).

On the other hand, Kant often seems to refer to the whole of reason in its second sense, with regard to the internal coordination of the parts. In this sense, the wholeness or unity is not presupposed but is to be attained. For example, Kant’s account of “the primacy of pure practical reason in its connection with speculative reason” in the second Critique, i.e., the “necessary … subordination” of theoretical reason to practical reason, is concerned with the internal organization of theoretical and practical cognitions within “one and the same reason” (KpV 5:121). Likewise, the very idea of a third Critique rests on the assumption that the critical system is not yet a completely internally organized whole. Using “a critique of pure reason” in a broad and unitary sense, as the critique of the “faculty of judging in accordance with a priori principles,” Kant emphasizes the need for making its internal organization complete:

The absence of a clear account of the unity of reason in Kant’s corpus has led some interpreters to presume that Kant’s critical philosophy is not or cannot be completely systematic.^{Footnote 8} Perhaps more importantly, Kant’s unclear use of the two senses of the whole or unity of reason is not sorted out in the secondary literature on the unity of reason, especially in the prevalent approach to understand the whole or unity of reason on the basis of one of its faculties or some arrangement among them.^{Footnote 9} Two major types of interpretations can be isolated within this general approach. The first type usually seeks the unity of reason in practical reason or in some kind of arrangement among the faculties under the guidance of practical reason. The second type takes the whole of reason to be the unification of the theoretical and practical faculties through the reflecting power of judgement.^{Footnote 10}

An example of the first type of interpretation is Onora O’Neill’s. She writes:

O’Neill formulates her conclusion (number 3) more specifically and in reference to Kant’s comparison of the natural dialectic of reason to the Tower of Babel: “in thought as well as in action we must, if we are to evade the threat of Babel, act only on the maxim through which we can at the same time will that it be a universal law.”^{Footnote 12} O’Neill’s claim that the supreme principle of practical reason is the supreme principle of the whole or unity of reason lacks philosophical and textual support. First, that the practical use of reason is “more fundamental” (italics added) than its theoretical or speculative use, or that Kant speaks of the primacy of practical reason over theoretical reason, does not necessarily mean that practical reason is the supreme principle of (the unity of) reason. In short, a higher principle is not necessarily the highest principle. Second, in the second quote above, where O’Neill extends the principle of practical reason to all forms of reasoning, she does not accurately represent Kant’s statement in the opening paragraph of the Doctrine of Method of the first Critique. In this paragraph, Kant emphasizes the need for a methodological or systematic approach in order to avoid the threat of Babel (the natural dialectic of reason) and to build the edifice of reason on a solid foundation. But O’Neil presents the methodological necessity of the whole or unity of reason—i.e., the systematic necessity of the oneness of reason prior to its applications—as the practical necessity or the necessity of its moral part (the categorical imperative). Contrary to O’Neill’s claim, nowhere does Kant identify the transcendental method of reason with the categorical imperative. The only way to avoid the fate of Babel, Kant says, is to develop a “plan” or “design” for the edifice of reason, which does not “exceed our capacity” and yet is “suited to our needs” (KrV A707/B735). At no point and in no way in the opening paragraph of the Doctrine of Method, does Kant imply that the supreme principle of this plan or design is the categorical imperative. In fact, Kant explicitly states the four key components of the plan or design for the edifice of reason or “a complete system of pure reason:”

Kant’s presentation of the discipline of pure reason as the first formal condition of a complete system of pure reason is not accidental. As will be discussed below, this priority indicates that the first principle of the unity of reason is the discipline of pure reason. Kant discusses the systematic condition of the practical (positive) use of reason in the Canon of Pure Reason, which, as Kant states, presupposes the discipline (negative use) of pure reason (KrV A796/A824). Third, as the supreme principle of morality, the categorical imperative by itself cannot even ground the systematic use of practical reason, let alone the entirety of the critical system. The categorical imperative cannot determine the relation between practical reason and empirical nature in a thoroughgoing and systematic manner. The determination of empirical nature requires a level of specification that neither theoretical reason proper nor practical reason proper can supply.^{Footnote 13} This issue, with which O’Neill does not engage, is one of the major motivations behind the Appendix to the Transcendental Dialectic and the Canon of Pure Reason in the first Critique, even before Kant addresses it more thoroughly in his revised doctrine of faculties in the third Critique.^{Footnote 14}

Contrary to O’Neill, Pauline Kleingeld presents a version of the first type of interpretation that appreciates the limits of practical reason proper when it comes to the question of the systematic unity of reason. Invoking Kant’s argument for the primacy of practical reason over theoretical reason, Kleingeld locates the unity of reason in the belief that harmonizes theoretical reason with practical reason: “In sum, the two uses of reason are not only not in conflict, but they both lead to the (thin) view of nature as divinely designed. Theoretical and practical reason are united or combined through the belief that nature harmonizes with morality.”^{Footnote 15} Kleingeld suggests that the unity of reason rests on assuming the existence of God as the creator of a teleological world that is in harmony with morality. And it is through the systematic needs of theoretical reason and ultimately practical reason that we can justifiably make such an assumption.^{Footnote 16} Kleingeld’s account has the advantage of taking seriously the systematic needs of the use of practical reason and acknowledging that practical reason proper cannot satisfy these needs by itself. That said, Kleingeld’s interpretation does not conceive of the unity of reason in a way that is systematically prior to theoretical reason and practical reason. The whole is taken to be the result of a systematic arrangement between the two parts: the subordination of theoretical reason to practical reason within the belief in God. This approach conflates the two senses of the whole, as they were outlined above, and thus does not help us conceive of the originary conception of the whole of reason, i.e., the most basic sense of the whole of reason prior to its parts. Without presupposing the originary conception of the whole, what Kant characterizes as “the complete concept of the whole,” the parts cannot be systematically arranged and harmonized. The limitation of Kleingeld’s approach becomes more evident if we notice that, contrary to her claim, “the belief in the existence of a God who has purposively ordered nature”^{Footnote 17} cannot be the ultimate basis for the unity of theoretical reason and practical reason. For Kant, any assumption regarding the supersensible realm can be legitimate only if it rests on a critique—or more precisely, a disciplining—of theoretical reason. This means that the belief in the existence of God requires presupposing the discipline of pure reason. As “a system of caution and self-examination” (KrV A711/B739), this discipline tests all claims of reason regarding the supersensible realm.^{Footnote 18} As Kant notes in his essay on the so-called ‘pantheism controversy,’ a justifiable assumption about the existence of God first needs to be subject to the discipline of the critical method of reason:

To be sure, Kleingeld does not argue dogmatically for the existence of God. In her account, the belief in the existence of God performs a regulative function. Kleingeld, nevertheless, overlooks the foundational role of the discipline of the critical method, which determines “the boundary at which it [reason] must stop.” She neglects that, in the critical system, the right to assume the existence of God presupposes the right of reason to prohibit its theoretical application beyond the sensible realm. This prohibition, which is the disciplinary determination of boundaries of reason, is necessary for systematizing all uses of reason. Kant calls this prohibition “a negative principle in the use of one’s faculty of cognition” (WDO 8:146). It is only on the basis of this prohibition or negative principle that reason can legitimately assume the existence of God. To put it more generally, all regulative legislations of reason must presuppose the “negative legislation” of the discipline of pure reason, “before which no sophistical illusion can stand but must immediately betray itself” (KrV A710/B739).^{Footnote 19} This methodological prohibition or discipline, which also functions as the systematic condition of “the belief” in a “divinely designed” nature, or reason’s faith (Vernunftglaube) in God, is missing in Kleingeld’s account.^{Footnote 20}

The second type of interpretation that understands the unity of reason on the basis of one of its faculties relies on the faculty of the reflecting power of judgement in the third Critique. An example of this type of interpretation is Angelica Nuzzo’s:

Nuzzo presents the unity of reason as the unification of theoretical reason and practical reason through the reflecting power of judgement. She draws on Kant’s statement that the power of judgement bridges the “incalculable gulf” between nature and morality (KU 5:175). This interpretation ignores that the role that Kant assigns to the reflecting power of judgement concerns the systematic arrangement among the faculties of reason. Such arrangement presupposes the most basic sense of the whole of reason, i.e., the originary conception of the whole that is complete in its purpose. In the third Critique, Kant characterizes this sense of the whole of reason as “a critique of pure reason” (KU 5:168), of “our cognitive faculty as a whole” (KU 5:174), or of “the pure faculty of cognition in general” (KU 5:176) prior to its division into three faculties of cognition. Nuzzo conflates the conception of the whole of reason that is the systematic condition for the internal organization of the faculties of reason with the conception of the whole that directly concerns such internal organization in its strict sense. She does so in a way that also goes against Kant’s unequivocal claim that the unity of critical reason must be conceived “independently of all experience” (KrV A xii). As is evident in her interchangeable use of human “reason” and the “human being,” Nuzzo does not conceive of the unification of theoretical reason and practical reason in a strictly transcendental sense. She assumes that the unity of reason cannot be “independent” of or in “separation from … sensibility,” and thus collapses into each other the transcendental and anthropological levels of analysis in Kant’s philosophy.^{Footnote 22}

Despite significant differences between and within the two types of interpretation of the unity of reason, neither distinguishes the two senses of whole that Kant implies in his writing. Both types of interpretation neglect the originary conception of the whole of reason. In doing so, they ignore that the whole of reason, from which all possible cognitions of faculties of reason are supposed to be derived, cannot be based on these faculties or some arrangement among them.

The next section begins to outline a sense of the whole of reason in the first Critique that is prior to the parts or faculties of reason and their possible cognitions. This sense of the whole of reason remains central to the entirety of the critical system, including to the second and third Critiques, although at the time of the publication of the first Critique Kant had no intention of writing the last two Critiques and had not developed his tripartite doctrine of faculties.

II. The Whole of Reason: The Apodictic Modality or Necessary Standard of Reason

If we take seriously Kant’s view that the systematicity of reason requires the whole or unity of reason to be prior to its parts or faculties, and if we also accept that the existing interpretations understand the unity of reason in a way that is not prior to its faculties, we need to determine how Kant might conceive of the whole or unity of reason prior to its parts. To do so, it is helpful to start with an outline of how, in Kant’s account, the priority of the whole of reason over its parts is supposed to facilitate the completeness or systematicity of reason. I will suggest that the way in which this priority makes reason complete or systematic indicates that we should understand the whole of reason in transcendental-modal terms, as the apodictic modality or necessary mode of reason, i.e., as the standard that determines what can systematically belong to reason and thus determines what can be possible content of the faculties of reason.

Kant argues for the priority of the whole of reason over the parts of reason in order to ensure the completeness of critique as the science or system of pure reason. In Kant’s account, the priority of the whole of reason over its parts requires that we think of reason in a way that systematically includes all the possible functions (cognitions) of its parts. This systematic inclusion is the nature of what Kant calls “the complete concept of the whole” of reason: if “the function [Zweck] of any member [part or faculty of reason] can be derived only from the complete concept of the whole” of reason, there will be no function “in the domain of this faculty” which is not completely determined and settled (Prol 4:263).

If, as Kant writes, the priority of the whole of reason over the parts implies that the whole of reason means reason that systematically includes all possible functions of its parts or faculties, the question becomes how reason can be such a whole, i.e., how all possible functions of the faculties of reason can be systematically included in “the complete concept of the whole” of reason. Can this systematic inclusion, or the wholeness of reason, be possible through gathering, enumerating, and harmonizing different functions of theoretical, practical, and reflecting parts of reason? If, as Kant suggests, the systematic inclusion of all possible functions of the parts of reason, or the formation of “the complete conception of the whole” of reason, cannot be attained through incremental addition and incorporation of the possible functions of each part into a whole, how and in what sense can reason be a whole independently of its parts?

Reason can systematically include all possible functions of its faculties, i.e., be a whole in its originary conception, if and only if it systematically separates or distinguishes what can be included in reason as possible functions (cognitions) of its faculties from what cannot. Any systematic inclusion requires a systematic separation of the inside from the outside. Without separating what can systematically belong to reason from what cannot, reason cannot be a whole. Nor can the differentiation of the parts of reason and their possible functions in that whole be systematic and complete. Such separation requires a single standard that determines what can systematically belong to reason. This means that the whole of reason is the standard that separates and thus determines what can systematically belong to reason. The standard is not a doctrinal principle that can be imposed on human reason from the outside. As I discuss below, the standard is reason’s constant disciplinary self-enactment, systematically underlying all acts (uses) of reason in relation to objects.

Kant understands the systematicity of reason, including the standard that determines reason as a whole, to be of a modal-apodictic rather than an objective nature. In his account, reason can be systematic or scientific if it is conceived independently of the objective content of its cognitions and on the basis of what is absolutely necessary in human reason. Kant sometimes uses the concept of ‘knowing’ (Wissen) to refer to these requirements. In his account of “Logical completeness of cognition as to modality” in Jäsche Logic, Kant writes: “in a science we often know [wissen] only the cognitions but not the thing presented through them” (Log 9:72). It is the modal-apodictic nature of knowing (Wissen)—i.e., reason’s knowing of its own absolute necessity independent of the objective content of its cognitions—that makes it foundational for a system of reason, including critique as the science of pure reason:

Knowing is foundational to the system of reason because “what I know [wissen] … I hold to be apodictically certain, i.e., to be universally and objectively necessary (holding for all)” (Log 9:66).^{Footnote 23} Knowing is universally necessary in the sense that it concerns only what is absolutely necessary in human reason. Given that this necessity is universal to reason, knowing is also objectively necessary in the sense that its necessity applies to objects “even granted that the object to which this certain holding-to-be-true [Fürwahrhalten] relates should be a merely empirical truth” (Ibid.).

Although in the first Critique Kant does not extensively discuss that, in the science of critique, knowing is first and foremost knowing (or the apodictic certainty of) the standard of reason, he is unequivocal that critique contains only what reason knows, or is apodictically certain of, that is, “all pure cognitions” of reason and most importantly the standard that underlies the universal necessity and objective necessity of these cognitions. Kant writes:

Kant indicates that the determination of the standard for all pure cognitions makes reason certain of all pure cognitions that can belong to it. That is to say, the wholeness of reason consists in the single necessary standard that determines what falls within the purview of reason and thus legitimately governs all possible uses or cognitions of the faculties of reason. In establishing the necessary standard or apodictic modality of its possible cognitions, reason determines all possible cognitions that its faculties can rightfully pursue.

Kant builds this apodictic philosophical certainty or knowing into his foundational concept of “transcendental.” He defines the concept of “transcendental” first and foremost in methodological terms: “I call all cognition transcendental that is occupied not so much with objects but with our mode of cognition [Erkenntnisart] of objects, insofar as that should be possible a priori” (KrV B25, bold added). Kant understands the modality or “mode [Art]” of cognition of objects as methodological and defines it in contrast to the “manner” of cognition that is “free” (arbitrary): “All cognition, and a whole of cognition, must be in conformity with a rule. (Absence of rules is at the same time unreason.) But this rule is either that of manner (free) or that of method (inner necessitation [Zwang])” (Log 9:139). The “cognition” that Kant calls transcendental is not defined in terms of any form of relation to objects. It is neither theoretical, nor practical, nor empirical. Transcendental cognition is reason’s methodological cognition of cognitions, or what Kant more precisely calls reason’s ‘knowing’ (Wissen) of its own “mode of cognition of objects” (KrV A11/B25, bold added), that is, of the modality or standard that is necessary for its systematic cognition of objects.^{Footnote 24} In the preface to the first edition of the first Critique, Kant calls reason’s knowing of its own apodictic modality of cognitions of objects, “critique,” which he says is reason’s “self-cognition” (KrV A xi). Kant takes transcendental critique, or reason’s knowing of itself, to be foundational to the use of the faculties of reason because, as mentioned above, in knowing its own apodictic modality, reason also delimits the possible content or objects of its faculties. Reason’s self-knowing, or more specifically reason’s determination of its standard for all its cognitions, is the highest form of reason’s self-determination. It is precisely in this methodological or modal-apodictic sense that Kant speaks of the genetic nature of the concept of the transcendental, as “the origin of our cognitions of objects insofar as that cannot be ascribed to the objects” (KrV A55-6/B80). This means that reason functions as the systematic origin of our cognitions of objects in the sense that, in knowing itself, reason enacts the standard according to which all possible cognitions of objects belong to it.

Kant puts the originary character of the apodictic modality, or absolutely necessary standard, of reason at the centre of the metaphysical deduction of the categories, where he attempts to conceive of the whole of reason as the a priori source or “origin of the a priori categories in general” (KrV B159). Kant’s account of how the whole of reason functions as the a priori origin of the table of judgements and thereby the table of categories can be outlined in three major steps.

In the first step, Kant introduces “an idea of the whole [Ganzen] of the a priori cognitions of the understanding” (KrV A64/B89). This whole is an “absolute unity” or “systematic unity” (KrV A67/B92), from which the table of judgements and thereby the table of categories are to be derived. To perform such an a priori originary function, the whole must be “a unity that subsists on its own” and in no need “to be supplemented by any external additions” (KrV A65/B89-90, italics added). The “whole [Inbegriff]” (KrV A65/B90) of the a priori cognitions of the understanding must be conceived in separation “not only from everything empirical, but even from all sensibility” (KrV A65/B89, italics added).^{Footnote 25} This stipulation implies that the whole of the a priori cognitions of the understanding cannot belong to either the understanding proper, or the power of judgement, or pure imagination. These faculties presuppose the possible givenness of objects in sensibility. Therefore, they cannot abstract from sensibility as such and be able to function as the originary source or self-standing whole of a priori cognitions of the understanding.

In the second step, Kant indicates that the whole of a priori cognitions of the understanding can be thought only in modal terms, as the necessary touchstone or standard for the systematic organization of all cognitions of reason. Kant writes: “the whole of its [understanding’s] cognitions will constitute a system that is to be grasped and determined under one idea, the completeness and articulation of which can at the same time yield a touchstone of the correctness and genuineness of all the pieces of cognition fitting into it” (KrV A65/B89-90, italics added). In short, the whole that systematically grounds all cognitions of objects is also the touchstone or standard for relation to and cognition of objects without itself being in any sense object-related.

In the final step, Kant implies that the whole of the a priori cognitions of the understanding is the modality of reason in its apodictic judgements. For Kant, the modality of judgements “contributes nothing to the content of the judgement … but rather concerns only the value of the copula in relation to thinking in general” (KrV A74/B100, italics added). In Jäsche Logic, Kant explains modal judgements in a way that brings out the holistic nature of “the value of the copula in relation to thinking in general.” He states that the modality of judgements deals with “the relation of the whole of judgement to the faculty of cognition” (Log 9:109, italics added). The three modal judgements, however, do not determine the value of the copula, or the whole of judgement, in relation to thinking in general in the same way. Problematic judgements only verify the merely logical integrity or wholeness of judgements, and therefore are in an arbitrary relation to the objective content of judgements. Assertoric judgements can verify the existential integrity or wholeness of judgements—i.e., the existence or non-existence of the objective content of judgements—but they cannot establish the integrity or wholeness of all possible objective content of judgements. Doing so requires apodictic judgements that establish an absolute necessity with regard to all possible content of judgements. Apodictic judgements enact the necessary value or standard by which all possible content of judgements can be systematically organized in thinking in general. To repeat Kant, they “yield a touchstone of the correctness and genuineness of all the pieces of cognition fitting into it” (KrV A65/B89-90, italics added). In other words, apodictic judgements establish the systematic integrity or wholeness, from which the table of judgements and ultimately the table of the categories can be derived. Referring to problematic, assertoric, and apodictic judgements in a footnote in the metaphysical deduction, Kant states that apodictic judgements of modality belong to reason: “It is just as if in the first case thought were a function of the understanding, in the second of the power of judgement, and in the third of reason. This is a remark the elucidation of which can be expected only in the sequel” (KrV A75/B100).^{Footnote 26} It is only the apodictic modality of reason—the necessary touchstone of all acts of the faculties of reason—that can establish the self-standing whole of a priori cognitions of the understanding and thus function as the “absolute” or “systematic unity” of reason (KrV A67/B92).

It is noteworthy that Kant formulates in various ways the modal-apodictic or holistic sense of reason as the necessary standard for all possible cognitions. For example, Kant sometimes speaks of reason’s modal determination of its systematic integrity or wholeness as the ‘correction’ of all cognitions of reason. Transcendental critique, Kant writes, “does not aim at the amplification of the cognitions themselves, but only at their correction [Berichtigung]” (KrV A12/B26, italics added). Such correction, which is also called “the purification of our reason” (KrV A11/B25), aims to establish the systematic integrity or wholeness of a priori cognitions. The systematic correction of all cognitions is not the correction of an error resulting from an application of improper rules or a faulty application of proper rules. Kant’s conception of systematic correction methodologically precedes all rules and their applications in cognitions. Employing language similar to that which he used in the metaphysical deduction Kant says that the systematic correction of cognitions aims “to supply the touchstone of the worth or worthlessness of all cognitions a priori” (KrV A12/B26, italics added). Systematic correction (Berichtigung) consists in the invention of reason’s absolutely necessary standard (Richtmass) or apodictic modality. This necessary standard is the immanent principle of all rules, the methodological necessity underlying the logic of a priori object-relatedness, transcendental logic in its strict sense.^{Footnote 27} The invention of reason’s necessary standard functions as the prevention of the systematic errors of reason’s convention (natural dialectic), i.e., the determination of reason’s standard that sorts out what qualifies as a possible cognition.

In his lectures on logic, Kant formulates the critical-methodological correction of reason as the transformation of the natural life of reason (natural logic) into critical reason (artificial or systematic logic). Natural logic or “logica naturalis” deals with “rules for how we think” whereas artificial logic or “logica artificialis” concerns “rules for how we ought to think” (V-Lo/Wiener 24:791). While he emphasizes the systematic primacy of artificial logic over natural logic, Kant acknowledges that rules of artificial logic can be invented “only through observation of that natural use” (Log 9:17). In other words, as logic, critical reason is inevitably retrospective. In Kant’s account, the ought of artificial or systematic logic is not a mere recommendation. It is primarily the methodological must, the systematic corrective or self-discipline of reason at the core of artificial logic, which is “a science of the correct [richtigen] use of the understanding and of reason in general” (Log 9:16). This “correct” use is primarily the systematic correction, or methodological prevention, of dialectical errors. The correction supplies the standard of reason, which functions as the absolutely necessary standard for all uses of reason. It is thus not surprising that Kant speaks of logica artificialis in a manner that emphasizes the identity of logic (in its transcendental-methodological sense) and critique regarding their modi, or systematic forms. In light of this identity of the transcendental-methodological sense of logic with critique, Kant uses his description of critique, as discussed above, to explain the science of logic: the science of logic is the “universal propaedeutic to all use of the understanding and reason in general,” which is necessary “for correcting our cognition, but not for amplifying it” (Log 9:13). Similarly, Kant describes the science of logic primarily in terms of reason’s necessary (systematic) form or apodictic modality, in the same way that he defines the concept of the transcendental in the introduction to the second edition of the first Critique: “Logic is thus a self-cognition of the understanding and of reason, not as to their faculties in regard to objects, however, but merely as to form” (Log 9:14).^{Footnote 28}

In his introduction to the third Critique, Kant succinctly formulates this primary, disciplinary sense of critique as the systematic correction or determination of the boundaries of all possible uses of reason in relation to objects. He states that the critique of the faculties of cognition should not be understood in terms of any object domain or the sum total of object domains:

Having outlined some of Kant’s key insights into the methodological or modal-apodictic nature of the whole of reason and the systematic priority of this whole over the three faculties of reason, we should now turn to the question of how reason determines itself as a whole. The single most important obstacle to appreciating reason’s methodological self-determination is the widespread neglect of the Doctrine of Method (of the first Critique) as a whole.^{Footnote 29} As Kant states in a note in Opus Postumum, the whole of reason can be conceived only methodologically, in a Doctrine of Method:

Kant’s reference in this passage to a doctrine of method as the locus of a scientific, or systematic, whole of reason applies exclusively to the first Critique.^{Footnote 31} It is only in the first Critique that the Doctrine of Method performs its strictly systematic function to found itself as a whole of all cognitions of reason. In the second Critique, Kant remarks, “the doctrine of the method of pure practical reason cannot be understood as the mode [Art] to proceed … with pure practical principles with a view to scientific cognition of them, which alone is properly called method elsewhere, in the theoretical” (KpV 5:151). The methodological appendices in the third Critique are not concerned with founding the whole of reason either, since no species of reflecting judgements is determinable by principles (KU 5:355 and 5:417). It is thus only in the Doctrine of Method of the first Critique that we can focus on reason’s strictly systematic function and examine how reason enacts itself as a whole or necessary standard for all faculties of reason. The critical method of reason in the first Critique, particularly its primary component the discipline of pure reason, is systematically foundational for the entirety of critical philosophy and not merely the operation of theoretical reason. The next section will discuss how reason enacts its negative standard, or its discipline, as the primary systematic condition for the possible (positive) uses or cognitions of all faculties of reason in the critical system.

III. The Discipline of Pure Reason: Reason’s Enactment of Itself as a Whole

In his introduction to the Doctrine of Method, Kant refers to the Doctrine of Method as the place where “the whole [Inbegriff] of all cognition of pure and speculative reason” (KrV A707/B735) is to be established. Comparing the system of reason to an edifice, Kant states that the task of the Doctrine of Method is to supply the “design” of the edifice the “building materials” of which were surveyed in the Doctrine of Elements (Ibid.). The design functions to make the building materials reason’s own, to organize the separate elements of reason surveyed in the Doctrine of Elements into a unified system of reason. The first step in this systematic organization is to separate what can systematically belong to reason from what cannot, to delimit the sphere of the legitimate power of reason. As Kant puts it, critique follows the advice of Persius (Satires, 4:52): “Dwell in your own house, and you will know how simple your possessions are” (KrV A xx). The establishment of what systematically belongs to reason requires the determination of the standard for what can systematically belong to reason, which is the determination of the boundaries of reason.^{Footnote 32}

Kant introduces the determination of the boundaries of reason, or the first principle or necessary standard of reason as a system, in the first chapter of the Doctrine of Method, the Discipline of Pure Reason. This determination takes place through what Kant calls “negative judgements” (KrV A708/B736). Through these judgements, reason disciplines and thus determines itself as a whole, i.e., as a standard that a priori applies to all uses of the faculties of reason. Kant’s use of the language of “humiliation” (KrV A710/B738) indicates the way in which reason enacts its integrity and wholeness through negative judgements of discipline. In its transcendental-methodological sense, humiliation is reason’s realization of its proper boundaries; it is the self-correction through which reason enables itself to avoid its natural-dialectical transgressions.^{Footnote 33}

In the opening sentence of the Discipline of Pure Reason, Kant tells us that these “negative judgements” “do not stand in high regard … and it almost takes an apology to earn toleration for them, let alone favor and esteem” (KrV A708/B736). Nevertheless, Kant emphasizes that the negative use of reason in negative judgements of discipline is of the highest significance for the system of critical philosophy as a whole: “the entire philosophy of pure reason is concerned merely with this negative use” (KrV A711/B739, italics added). In a similar statement, Kant writes: “The greatest and perhaps only utility of all philosophy of pure reason is thus only negative, namely that it does not serve for expansion, as an organon, but rather as a discipline, serves for the determination of boundaries, and instead of discovering truth it has only the silent merit of guarding against errors” (KrV A795/B823, italics added). Kant makes such strong statements because the negative use of reason in negative judgements of discipline is the act of reason determining its own boundaries—i.e., determining the standard that separates what can systematically belong to reason from what cannot. Such determination is the first step in forming the system of reason as a system of possible cognitions.

Despite his emphatic statement that pure reason is concerned only with negative judgements of discipline, Kant discusses them only briefly. These judgements do not belong to either general logic or transcendental logic proper. In fact, they are modal, apodictic, and determine the wholeness or systematic integrity of reason. I will discuss each of these aspects of negative judgements in turn.

To see how negative judgements of discipline do not belong to general logic, we must first notice how they differ from negative judgements that are placed under the title of quality in the table of judgements. Negative judgements of quality are governed by the merely logical principle of contradiction, and abstract from all content of the predicate-concept. They show the merely logical impossibility of attributing the predicate-concept to the subject-concept. In these judgements, “negation always affects the copula” (Log 9:104), and “the subject is posited outside the sphere of the predicate” (Log 9:103). Negative judgements of quality assert that ‘S is not P’ regardless of what P is. By contrast, negative judgements of discipline “are negative not merely on the basis of logical form but also on the basis of their content” (KrV A708/B736, italics added). Here “content” should be understood as possible content rather than as an immediately given sensible content. Excluding that which cannot be acquired as a possible content of theoretical cognitions of reason, negative judgements of discipline demarcate the whole of possible cognitions of reason. They demarcate the boundary between what is objectively possible (the theoretically cognizable) and what is objectively impossible (the theoretically un-cognizable). The self-determination of reason as a whole of possible cognitions in negative judgements of discipline is the transcendental-methodological acquisition or formation of content in general, prior to the distinction between the logical form of understanding and the empirical content of sensibility. In this self-determination, reason forms itself as a whole of possible cognitions, regardless of the specificity of its objective content. That is, negative judgements of discipline at once determine the transcendental-systematic form and thus the possible content of cognition in general, prior to any relation to objects in theoretical, practical, and reflecting uses of reason.

Negative judgements of discipline are not infinite judgements, i.e., negative judgements of transcendental logic proper. To be sure, infinite judgements are not concerned with the merely logical negation of the copula. In them, “the predicate … is affected” and “the sphere of the predicate” is limited (Log 9:104). Infinite judgements take the content of the predicate-concept into account, and affirm that ‘S is not-P.’ These judgements, nevertheless, indefinitely delimit the content of a predicate-concept through presupposing the possible givenness (existence) of objects. Determining the content primarily through a relation to possible objects, infinite judgements are not entirely independent of sensible objects. By contrast, negative judgements of discipline form reason itself, and thus determine the possible “content of pure cognition in general” (KrV A709/B737), regardless of the existence or non-existence of objects. They determine the methodological (systematic) form of possible cognitions, before objects are given in intuition. Prior to the possible (relational) givenness of objects, negative judgements of discipline enact the negative methodological (non-relational or absolute) law of reason’s self-givenness. The positive givenness of objects in sense must presuppose reason’s negative (disciplinary) self-givenness. Otherwise, cognitions of objects cannot be systematic.

Negative judgements of discipline are modal. As such, they have “a quite special function … which is distinctive in that it contributes nothing to the content of the judgement (for besides quantity, quality, and relation there is nothing more that constitutes the content of a judgement), but rather concerns only the value of the copula in relation to thinking in general” (KrV A74/B99-100).^{Footnote 34} These judgements perform a genetic function: they enact “the method of cognition from pure reason” (KrV A712/B740, italics added), i.e., the corrective (critical) method of reason: they “have the special job of solely preventing error” (KrV A709/B737), prior to any relation to objects. As seen in the Dialectic, speculative reason as applied to the supersensible is naturally and inevitably plagued with “an entire system of delusions and deceptions.” Hence, reason needs a discipline, i.e., “a quite special and indeed negative legislation … a system of caution and self-examination out of the nature of reason and the objects of its pure use, before which no false sophistical illusion can stand up.” Unlike “individual errors” which “can be remedied through censure” (i.e., the local use of the sceptical method of doubt), systematic errors in the Dialectic can be corrected only in a global way, transcendentally-methodologically, “through critique” (KrV A711/B739). Rather than dealing with dialectical errors of reason one by one, negative judgements of discipline correct these errors systematically, i.e., through reason’s methodological determination of its boundaries, the self-enactment of the apodictic modality or systematic unity of reason:

In short, negative judgements of discipline invent the necessary standard of reason that prevents dialectical errors rather than extend metaphysical cognitions in merely logical or actual terms. These modal judgements are neither problematic nor assertoric. They only preserve what is absolutely indispensable for reason to be a self-standing system, independent of objects.

Negative judgements of discipline are apodictic. That is, they respond to and satisfy the most basic transcendental need of reason to protect itself against errors. Without satisfying this need, reason cannot be systematic in any of its uses. Unlike problematic judgements (‘It is possible that ‘S is P’’) and assertoric judgements (‘It is true that ‘S is P’’), negative judgements of discipline enact a transcendental apodicticity (‘It is necessary that ‘S is P’’). The apodictic identification of the subject-concept and the predicate-concept is prior to and therefore is not concerned with any particular relation between a subject and a predicate. This identification necessitates the non-relational or absolute ground, the whole (Inbegriff) or standard, in terms of which all concepts (Begriffe)—subject-concepts and predicate-concepts—can relate to each other in the system of reason. The apodicticity or necessity of negative judgements of discipline is the systematic ground that enables the possibility of relation between subject-concepts and predicate-concepts in general. In other words, these judgements enact the discipline of pure reason, i.e., “the inner necessitation [Zwang] through which the constant propensity [of reason] to stray from certain rules is limited and finally eradicated” (KrV A709/B737).^{Footnote 35} This inner necessitation or the self-preserving inner force or impulsion of reason is distinct from the empirical (psychological or anthropological) necessity of skills in culture (Kultur) and the dogmatic necessity of doctrine (Doktrin) in metaphysics proper. Culture “would merely produce a skill without first cancelling out another one that is already present” (KrV A709/B737). Doctrine would only present a dogmatic rule and leaves no room for transcendental negativity. Culture and doctrine both lack the singular inner necessitation of a self-enacting negativity (self-discipline).^{Footnote 36}

Negative judgements of discipline enact reason’s whole or integrity, because they establish the standard by which reason determines what can systematically belong to it from what cannot. This methodological determination of the boundary between the inside and the outside of reason systematically forms all possible cognitions of reason. It is primarily in this sense that Kant considers critique to be propaedeutic to metaphysics; by determining or knowing (Wissen) itself systematically, reason makes its possible cognitions systematic. As Kant notes in a section titled “Form of science—Method” in Jäsche Logic, the systematic form or whole of cognitions of reason is enacted through the methodological reflection on all its elements:

Negative judgements of disicipline determine the boundaries of pure reason. They enact reason’s own inner necessity rather than eacting the empirical necessity of culture or the dogmatic necessity of doctrine. The inner necessitation or discipline of reason functions as a correction or cultivation (Zucht) prior to any culture and doctrine. Thus, through negative judgements of discipline, reason is at once corrective and self-enacting. Transcendental correction delimits or forms the boundaries of pure reason and in this way it is transcendentally self-enacting. The denial of cognitive claims of speculative reason about the supersensible in negative judgements of discipline is at the same time the formative ground—the apodictic modality or necessary standard—for the systematicity of all cognitive claims of reason about the sensible. Reason’s methodlogical self-negation is also a systematic object-determination, prior to the determination of the existence or non-existence of objects.

Kant explains the “negative contribution” of the discipline of pure reason by distinguishing it from the scholastic sense of discipline as instruction:

The self-limitation or self-determination of reason in negative judgements of discipline functions as the drive that negates reason’s “lust for knowledge” of the supersensible and thus forms the “unremitting” natural spontaneity or “cognitive drive [Erkenntnistrieb]” of reason (KrV A708/B716). Rather than following the empirical or dogmatic teaching of culture or doctrine, in its discipline, human reason becomes its own disciple, and thinks for itself. Preparing its own material, it forms, or cultivates/corrects (züchten), itself as a whole of possible cognitions. It is this self-generative character of reason in its self-discipline that Kant considers thinking for oneself not being primarily a matter of instruction and teaching.

In negative judgements of discipline, the cognitive drive that unremittingly extends theoretical reason into the supersensible is disciplined by reason’s own inner necessitation (Zwang) that protects reason against its own systemic errors. This self-enacting negativity of reason gives birth to reason as a whole: it separates what can systematically belong to reason from what cannot. This self-enacting negativity of reason in the discipline of pure reason is different from the negativity of dogmatism that retains “traces of ancient barbarism” and the negativity of scepticism, i.e., “a kind of nomads who abhor all permanent cultivation of the soil” and shatter “civil unity” (KrV A ix). Critique is the negative force, or self-instituting discipline of reason, which systematically necessitates all possible cognitions of reason. As the simultaneously self-dividing and self-cultivating act of reason, critique seeks the enemy within itself, and directs its negative force against this enemy, which naturally inhabits reason. Kant makes this point in the first Critique:

Thus, critique is primarily the immanent negativity of reason towards itself, in order to determine its boundaries (the whole or standard), and sort out what can systematically belong to it. This self-determining negativity or self-discipline is an inner transformation through which reason (as natural logic) is transformed into critical reason (as artificial logic), being formed as a whole of possible cognitions, demarcated from the impossible cognitions whose pursuit can take the form of dialectical errors. Reason needs to undergo such a transformation in order to watch over its transcendental use and thus be a whole and protect itself against its own dialectical errors. This wholeness of reason is the standard against which “no false sophistical illusion can stand up but must rather immediately betray itself” (KrV A711/B739).^{Footnote 37}