1. Schulbegriff and Weltbegriff of philosophy

It has become clearer in relatively recent times that the Doctrine of Method is not to be considered merely a kind of appendix to the Critique of Pure Reason. Otfried Höffe had even spoken of a Nichtlektüre,Footnote ¹ and it was possible to cultivate doubt about the patience and attentiveness of readers of the Critique regarding this text.Footnote ² However, its role may even be reversed into that of a ‘gateway’Footnote ³ to the Critique, if its reading and interpretation is combined with some of Kant’s isolated but important statements for which the entire Critique would be ‘a treatise on method’ (B xxxi) or ‘a doctrine of method’ (A82-3/B108-9). Gabriele GavaFootnote ⁴ moves from these ‘often completely overlooked’ (p. 41) assertions by taking them very seriously: he resolutely chooses the path of using them for a comprehensive reinterpretation of the aims of the Critique, offering perspectives of great interest. There is no need to emphasize how important and convincing I find this approach. The book offers a plurality of relevant themes and analyses, largely made possible and nourished by the general framework chosen. I will limit myself here to some observations on aspects more directly related to the basic thesis presented: the idea of the Critique of Pure Reason as a doctrine of the method of metaphysics.

The idea of criticism as a doctrine of method naturally leads to assigning importance to the part of the Critique that bears this title. The theoretical centrepiece of the Transcendental Doctrine of Method is undoubtedly the Architectonic, in which Kant attempts to outline a general project of philosophy as such, and of metaphysics in particular. The key idea of this general project is what is indicated by Kant with the Weltbegriff of philosophy, as distinct from its Schulbegriff.

Gava frames this concept in his reading of the transcendental doctrine of method. His thesis is that the principal task of a doctrine of method is not to identify appropriate procedures of investigation or argument but that rather its purpose is to show that a body of cognitions can be considered a science if it forms a ‘system’ with an ‘architectonic’ unity (p. 2). Considering this goal, and the way it is conceived and then carried out, would also allow for some important clarifications on much-discussed and certainly not marginal issues such as that of the relationship between the critique of reason and transcendental philosophy, and between these and metaphysics.

Gava is right to point out that the Kantian idea of a doctrine of method is rather far from the idea that is natural to us of a methodology understood as a set of procedural rules. This idea – which Kant links to that of a ‘general’ doctrine of method – is viewed negatively by Kant: procedural rules without reference to a specific science turn out to be essentially tautological, and ‘instead of giving means for the solution of questions, they give the same questions again’.Footnote ⁵ The Kantian idea of a doctrine of method arises – with a wholly original transformation in a radically different theoretical framework – from a particular intersection between the ideas of logica practica or utens, ‘particular logic’, and the section on Lehrart (methodus), which in Meier was the second part of his Vernunftlehre.Footnote ⁶ In the latter, the aspect that we today would call methodological was referred to as the order and organization of cognitions; this involves a closer consideration of their whole, a Ganzes, dominated by a Baumeister, an architect.Footnote ⁷ Already very early outlining the idea of system, connected to that of totality, Kant therefore develops primarily – but not only – this motif he finds in the tradition. The notion of system is then decisive for the interpretation of the meaning of a (transcendental) doctrine of method, and thus of the very idea of the critique of reason.

Kant defines the Architectonic precisely in relation to the notion of system, conceived as a condition of scientificity: ‘Since systematic unity is that which first makes cognition into science, i.e., makes a system out of a mere aggregate of it, architectonic is the doctrine of that which is scientific in our cognition in general’ (A832/B860). Gava’s interpretative keys on this point can be sketched as follows. He articulates the concept of systematic unity into two aspects that he believes do not coincide: (1) mere systematicity; (2) architectonic unity. The second is a narrower and more stringent condition: ‘Unlike Kant’s general demand for systematicity, which can be met in different ways, there is only one ordering of the cognitions belonging to a science that can grant architectonic unity’ (p. 18); ‘different systems of cognitions can equally satisfy that demand’ (p. 22). While systematic unity would basically concern principles such as those of homogeneity, specification, and continuity, which guarantee ‘inferential relationships among cognitions’ (p. 22), architectonic unity presupposes an idea, as a rational concept of the form of a whole, which functions as an end and allows the inference of parts of the system from other parts (p. 23). Despite the difficulties in identifying with certainty whether a body of cognitions is the realization of an idea, two minimum criteria are possible – according to Gava – in order to be able to attribute architectonic unity, namely, (1) systematic coherence and (2) the possibility ‘to describe the cognitions belonging to a science as dependent on an idea’ (p. 29). He distinguishes furthermore two senses in which architectonic unity can be conceived: in a first sense, architectonic unity simply requires that a body of cognitions be organized according to the idea of the whole of a science describing the parts-whole relationships among its doctrines; in a second sense, architectonic unity requires ‘that reason “provides the ends a priori” … and that the cognitions are organized around “a single supreme and inner end”’ (p. 34).

With this interpretative toolkit, summarily sketched here, Gava addresses the fundamental distinction between the scholastic and the worldly concept of philosophy. The thesis he arrives at is that ‘it is only when metaphysics is understood according to the worldly concept of philosophy that it is possible to regard it as capable of architectonic unity’ (p. 37), understood, one may suppose, in the second, more radical sense.

The observations I intend to propose mainly concern the following points: (1) the way in which it is possible to understand the notion of philosophy in a cosmic sense; (2) the meaning of the concept of method and its connection with this notion of philosophy, a connection which can be understood in perhaps an even closer sense than Gava’s reading.

Gava’s analysis is very clear and convincing, and gives the prominence it deserves to the worldly idea of philosophy, reading it as Kant’s insistence that metaphysics must ultimately give an answer to the basic question concerning man’s destination. This answer would be possible through the ‘epistemological’ reformulation, so to speak, of its status, realized by the doctrine of postulates and their relation to that particular theoretical ‘commitment’ represented by Glauben. In other words, this idea would clarify the practical purpose of metaphysics and the possibility of its being fulfilled in new forms by that kind of Fürwahrhalten that allows adherence to postulates.

Gava’s approach tries to shed light on some aspects and to overcome some problems in Kant’s text. I am not entirely sure that the threefold distinction among systematic unity, architectonic unity in the first sense, which I would call weak (idea-part relation), and architectonic unity in a second, strong sense (organization of science in relation to a single idea prescribed by reason) is an easily applicable tool for the different cases that may arise, although perhaps it turns out to be clearer than the Kantian account in its various texts.Footnote ⁸ Certainly, distinctions are necessary, since Kant speaks of systems in the plural, that is, of systems with reference to multiple cases (e.g., experience as a system, the system of the faculties of the soul, etc.). What is surely clear is that there is a more radical and ‘authentic’ sense of systematicity that Kant ascribes to philosophy in its worldly sense, which at some point (e.g., in V-Met-L2/Pölitz, 28: 533; Log, 9: 24) leads to ascribing systematicity only to philosophy, which however is not consistent with other assertions by Kant and even with his explicit distance from the tradition,Footnote ⁹ but which certainly signals that the systematic unity proper to philosophy has its own peculiarity and relevance that must be taken into account.

The distinction between Schulbegriff and Weltbegriff fits into this context. In first discussing the former concept, it seems to me that Gava radicalizes his view, as he does not ascribe to philosophy, according to its school concept, even the weak sense of architectonic unity: it ‘can at best reach technical unity’, he writes, but adds: since it ‘according to the school concept does not care about setting limits to rational cognition, however, it cannot even achieve technical unity’ (p. 36). This thesis involves, in my opinion, an excessive devaluation of the notion of philosophy in the school sense, which perhaps also has consequences for the way philosophy in the worldly sense is conceived. On the one hand, there are texts in which to philosophize in the scholastic sense is clearly attributed a systematic unity.Footnote ¹⁰ On the other hand, the way the two concepts are introduced may suggest a different reading of both.

2. Bis dahin

The discourse through which the Schulbegriff/Weltbegriff distinction is introduced takes its starting point from an ‘objective’ idea of philosophy, understood as ‘the archetype for the assessment of all attempts to philosophize, which should serve to assess each subjective philosophy, the structure (Gebäude) of which is often so manifold and variable’ (A838/B866). It is against this background that the distinction – at least in the first Critique – is introduced, and I believe it can be used to guide its interpretation. Philosophy in the objective sense is presented, precisely, as ‘archetype’ (Urbild), as ‘a mere idea of a possible science, which is nowhere given in concreto, but which one seeks to approach in various ways until the only footpath, much overgrown by sensibility, is discovered, and the hitherto unsuccessful ectype, so far as it has been granted to humans, is made equal to the archetype’ (A838/B886). This aspect should not be lost sight of: the authentic idea of philosophy is precisely an idea, a model that is in principle unrealizable. Nor should we lose sight of another element connected with this, and constantly associated with the idea of philosophy in the worldly sense (and, we shall see, with the notion of method): the thesis of the impossibility of learning philosophy, as an accomplished science, and of only being able to learn to philosophize, to exercise philosophical reason. Using the expression that I think is decisive in this context, ‘bis dahin’, Kant precisely introduces this thesis: ‘Until then (bis dahin) one cannot learn any philosophy; for where is it, who has possession of it, and by what can it be recognized? One can only learn to philosophize.’ In fact, the ‘objective’ idea of philosophy as an archetype is related to the primacy of philosophizing: to refer to it means ‘to exercise the talent of reason in prosecuting its general principles in relation to certain present attempts’ (A838/B866),Footnote ¹¹ i.e., precisely to philosophize: ‘to consider all the systems of philosophy only as the history of the use of our reason, and as the object of the exercise of our critical capacities’ (V-Met-L2/Pölitz, 28: 534).Footnote ¹²

The concept of ‘school philosophy’ and then, by contrast, that of ‘worldly philosophy’ are introduced on the basis of these premises: an ‘objective’ idea of philosophy that does not allow any present attemptFootnote ¹³ to be regarded as realized philosophyFootnote ¹⁴ and therefore as something that can be learned. Using again the expression bis dahin, Kant introduces the Schulbegriff: ‘Until then, however, the concept of philosophy is only a scholastic concept, that is, that of a system of cognitions which is sought only as a science without having as an end anything more than the systematic unity of these cognitions, thus the logical perfection of cognition.’Footnote ¹⁵ In contrast to this, it is then said that ‘there is also’ another concept of philosophy, the worldly concept, which conceives philosophy as ‘the science of the relation of all cognition to the essential ends of human reason’ (A839/B867).

According to the most usual reading, the Schulbegriff represents a stage that would be surpassed by the critique itself (when ‘the only footpath … is discovered’) by then proceeding to the terrain of the Weltbegriff. Accordingly, at least the second bis dahin could be understood as ‘until now’: the critique would produce the establishment – now – of a more adequate concept of philosophy. Certainly, the critique grounds in Kant’s eyes a new and more adequate view of philosophy. This does not mean, however, that a clearer view can lead now to a complete realization of the idea. Criticism, on the one hand, discovers or rediscovers (it was an ‘overgrown’ path for Kant) a notion of philosophy that can positively orient the work of reason; but, on the other hand, it does not fully realize a ‘definitive’ philosophy. We should try to read the two sentences together: (1) ‘Until then one cannot learn philosophy’; (2) ‘Until then, however, the concept of philosophy is only a scholastic concept’. While it would perhaps be excessive to read the second assertion as the thesis that philosophy will always remain a Schulbegriff – because it will never be adequate to the Urbild – it would surely be wrong to think that the discovery of the critical path enables one to learn philosophy from then on, and that therefore the worldly concept would represent the philosophy that can be realized and finally learned. If one does not want to attribute the latter idea to Kant – something made difficult by the centrality and pervasiveness that the thesis of the primacy of philosophieren over philosophy retainsFootnote ¹⁶ – the meaning of the contrast of the two concepts should be read as follows.

The notion of philosophy as an objective archetype remains firm, and its nature as a regulative idea, we might say, remains firm. Hence, it is impossible to conform to that idea, and thus there is the impossibility of teaching a philosophy. One concept of philosophy, however, the scholastic one, interprets the relation to the model in a predominantly or exclusively cognitive sense, which also leads to underestimating the radical issue of teaching/learning. If the archetype is a model of knowledge, it is possible in principle – or at least easier – to think that one can conform to it. Kant, on the other hand, in presenting the worldly concept, revives, as it were, and emphasizes its nature as an idea. Even more, he refers to a personified Urbild, thus to an Ideal, which more radically excludes the possibility of conforming to it.Footnote ¹⁷ The personified ideal of the philosopher is not incidentally presented as a Lehrer im Ideal (A866/B838) or Lehrer der Weisheit (9: 24), that is, as the unattainable possibility of lehren, teaching a philosophy that can be learned.

If, therefore, the school concept is presented as something that tends to be surpassed (that applies ‘until then’), it is not because it is later replaced by a philosophy that adequately realizes the model by making itself ‘cosmic’. This way of presenting it underscores that conceiving philosophy as a rational and purely speculative enterprise overshadows, to a greater extent, the function that should direct it and the ideal toward which it should move, namely, the fact that ‘the practical philosopher is authentically philosophical’, and that therefore ‘philosophy is the idea of a perfect wisdom, which shows me the ultimate ends of humanity’ (V-Met-L2/Pölitz, 28: 533; my italics). The reference to wisdom, we shall see, must also prevent this task from being understood in an overly theoretical way, as an – albeit transformed – matter of knowledge.Footnote ¹⁸

As mentioned, the devaluation of scholastic philosophy is not important in itself, since it is not the concept of philosophy that matters to Kant; it can nevertheless be relevant, because it makes less visible a central feature of worldly philosophy, the reference to wisdom in its connection with philosophieren. In other words, to put it this way, the question is not to replace one unsystematic conception of philosophy with a systematic or even fully systematic one; but rather to rethink the notion of philosophy in relation to an ideal, that of wisdom, which necessarily transcends any definite form of it, any ‘present attempt’.

3. Way of thinking or theory

I cannot carry out these considerations further here with the detail that would be necessary. I will only try to mention some consequences of the remarks made so far for what is the focus of Gava’s book: the notion of method. Gava rightly emphasizes the peculiarity of the Transcendental Doctrine of Method, especially the Architectonic, which places the notion of the system as a ‘doctrine of scientificity’Footnote ¹⁹ at the centre. It distinguishes method from a ‘procedural’ methodology in a sense closer to the one customary today, as a set of suggestions on argumentative techniques or procedures of investigation. While it is important to keep this in mind, the notion of method that emerges in the doctrine of method and in the critical philosophy in general is not reducible to a theory of systematicity, to a kind of sui generis normative epistemology that describes the conditions under which a body of cognitions can form a system and, thus, a science. An important point of the Kantian notion of method is its connection, not only with the establishment of conditions for systematicity, but also with a Denkungsart, a way of thinking: this aspect too constitutes one of the keys to the idea of method.

Method has essentially to do not with an organization of given knowledge, but with its production. It is from the very beginning systematic if it proceeds from the ‘first sources of reason’: ‘knowledge itself must arise systematically’ (V-Lo/Pölitz, 24: 531), and we have ‘an obligation to proceed systematically’ (A856/B884). The sources are not principles understood in the sense of propositions from which others can be derived, with a ‘vertical’ organization of knowledge, so to speak, but precisely a way of thinking that from the beginning refers at the same time to a whole, a ‘horizontal’ organization, and to a teleological consideration of knowledge.Footnote ²⁰ Kant writes in a letter: ‘I trust that I shall be able to prove formally that not even a truthfully metaphysical proposition can be proved without connection with the whole, but must always be derived only from the relation which it has to the sources of all our rational knowledge in general, thus from the concept of the possible whole of such knowledge’ (Br, 10: 341). One should not confuse ‘maxims of reason’ with ‘axioms’, ‘rules of method’ with ‘principles of theory (doctrine)’ (Refl 4971, 18: 45). ‘The legislation of reason is not based on thoughts, but on the way of thinking’ (Refl 1508, 15: 821). This sort of primacy of method over contentFootnote ²¹ is related to the two elements I tried to emphasize earlier: the characterization of the worldly concept as essentially related to a philosophy as ‘idea’ in which the reference to an unrealizable wisdom that goes beyond the ‘doctrine of wisdom’ is decisive, and the related conception of the primacy of philosophieren over all ‘historical’ philosophy.Footnote ²² This, in turn, is connected with the idea of method.Footnote ²³ It is not easy to develop here all the necessary connections, but I shall try to summarize the prospect that seems to me to emerge. A philosophy that follows the systematic procedure and understands itself according to the worldly concept excludes, on the one hand, all constructive and deductive procedures, always moving from a ‘horizontal’ exercise of reason; on the other, it organizes knowledge – the theoretical knowledge of the ‘technicians of reason’, aimed at Geschicklichkeit – in relation to what necessarily interests everyone, that is, to the ultimate end of humanity. This ends up giving a very special meaning to the ‘doctrine of wisdom’ precisely as knowledge of ends. Its detachment from wisdomFootnote ²⁴ shows, however, that its task is not to point to definite ends, through a kind of ‘knowledge’ of them, and therefore not even to reach a knowledge of the highest good and its conditions. To put it as clearly as possible, the point is not to achieve the same goals as rationalist metaphysics by different (practical instead of theoretical) argumentative means, but to achieve radically different goals.

Gava very effectively points out how the fulfilment of metaphysics in Kant, at which the critique as its doctrine of method aims, must consider ‘the systematic relevance of Kant’s practical argument’ regarding God and immortality ‘to the project of the critique of pure reason’ (p. 211). This argument is interpreted in a sense that, regardless of discussions about its validity, can go beyond a notion of hope, and can ‘justify a stronger attitude’ (p. 218). The impossibility of drawing on wisdom but also the idea of a philosophy that promotes essentially, first and foremost, a method of referring to the ‘sources’ of reason, a way of thinking, can go instead, in my view, in a direction of a decisive weakening of the relation to theoretical assumptions. This means that the direction, or at least the trend, of Kant’s argumentation moves precisely toward a notion of hope very different from a ‘commitment’ to truths, albeit differently argued. In the end, what enables ‘a free holding to be true’ (ein freies Fürwahrhalten) (KU, 5: 472; Fort, 20: 298) toward God and immortality is precisely only a hope – a basic attitude more than a commitment to a proposition. In the third Critique, hope is clearly identified with a Habitus, a way to act, a stance, and then with a Vertrauen, an overall attitude of trust toward the possibility that the pursuit of happiness compatible with morality is not contradictory for those acting on the basis of moral law.Footnote ²⁵ The meaning of this holding to be true is ultimately centred on the structure of acting oriented by the ‘as if’, the weakest structure from the point of view of theoretical ‘commitment’: it results in a ‘teaching’ ‘so to act, as though we knew that these objects were real’ (Fort, 20: 298). What is outlined here is a type of ‘teaching’ (Belehrung) that is basically about a possible and free way of acting and thinking: Kant returns to the idea of ‘a free holding to be true’, emphasizing that this freedom is what gives it ‘moral worth’, and that, on the other hand, produces what is as far as possible from a generally shareable theoretical stance: ‘It does not, therefore, confer an imperative (crede), and the ground of proof of its correctness is not a proof of the truth of these propositions, considered theoretically, and thus no objective teaching (Belehrung) as to the reality of their objects’ (20: 298).

I do not think it is a coincidence that the theme of the Lehren returns here. The centrality of method for Kant emerges as the centrality of a habitus, and also as the centrality of the spirit over the letter of science:Footnote ²⁶ method is more than a tool for theory-building.

Although Kant is not always clear on this, he basically oscillates between two tendencies: (1) toward a more radically new vision and (2) toward a position in which he seems to re-propose a teleological system as a ‘given’ theoretical structure. The prevailing and most interesting tendency seems to me to be the former, the one whereby the ‘maxims of reason’ are not to be understood as ‘axioms’, and ‘the rules of method’ are not to be considered ‘principles of theory (doctrine)’.Footnote ²⁷ From this perspective, theoretical construction serves the exercise of a rationality that intends to move in the horizon of wisdom without being able to draw on it. As they are conceived in Kant, neither virtue (based on a formal law) nor happiness (which everyone can conceive of differently), nor their union (based on a free holding to be true) – and thus neither the resulting ‘articles of faith’ – indicate any kind of content that philosophy allows one to affirm, and obliges one to share, but only a way of thinking, and the establishment and promotion of a ‘life-structuring’ rationality.Footnote ²⁸ Even the later Kant takes up and emphasizes this perspective: ‘Transcendental philosophy is not a type of knowledge of some object of philosophy, but only a certain method or (formal) principle of philosophizing; a type of discursive knowledge of creating for oneself the object of reason in the concepts of God, freedom and totality’ (OP, 21: 85).