Kant’s Engagement with Hume’s Enquiry: 1762–1765

1 Introduction

In the late nineteenth century, in later editions of the third volume of his Geschichte der neuern Philosophie, Kuno Fischer argued that Hume’s influence on Kant, to which the latter famously refers in the Prolegomena’s Preface, “first clearly emerges” within the Versuch den Begriff der negative Grössen in die Weltweisheit einzuführen (henceforth: Negative Magnitudes) of 1763.^[1] On his account, Kant’s assertion that Hume “interrupted [his] dogmatic slumber” (4:260)^[2] refers to the intellectual crisis induced by the problem concerning the relation of real grounds and consequences that he had outlined some two decades earlier in the Negative Magnitudes’ General Remark, a crisis that led him to abandon dogmatism and – for a brief period – appropriate Hume’s skepticism.^[3] Accordingly, Fischer claims that the interruption of Kant’s dogmatic slumber occurred in 1762/63. He further claims that the source inspiring that interruption was Hume’s first Enquiry, arguing that Kant found the problem concerning the relation of real grounds and consequences within that text and that he had read Hume’s first Enquiry before composing the Negative Magnitudes.^[4]

While Fischer’s interpretation initially won support from a number of his contemporaries,^[5] it came under attack by scholars such as Paulsen and Erdmann, who believed that Hume’s decisive influence occurred only in – respectively – 1769 or 1772. Of their objections, the following have been identified as the most forcefu1^[6]:

1. In passages that for Fischer evidence Hume’s influence, the difference between Kant ’s terminology is considerable.^[7]

2. Even if Kant’s skeptical reflections concerning causation in the 1760s are analogous to Hume’s, they draw different conclusions from these reflections.^[8]

3. Identifying 1762/63 as the date of the interruption does not sit well with Kant’s own statements about his philosophical development.^[9]

4. If Kant were writing under Hume’s influence in the early 1760s, it is difficult to understand why this was never detected by contemporaries such as Mendelssohn.^[10]

5. If Kant were writing under Hume’s influence in the early 1760s, it is difficult to understand why he never once names Hume in those texts that Fischer considers to evidence his influence.^[11]

In 1990, Kreimendahl claimed that over the previous century neither Fischer himself nor anyone sympathetic towards his hypothesis had succeeded in providing any compelling answer to the above objections.^[12] In the wake of the difficulties attributed to Fischer’s hypothesis, a consensus seemed to emerge that Kant’s interruption must have occurred in either 1769 or 1772 and must have been provoked by German translations of passages from the Treatise, which only became available to Kant some years after the publication of the first German translation of the Enquiry of 1755.^[13] One influential interpretation – pioneered by Vaihinger – claimed that the interruption occurred when Kant read the German translation of Beattie’s Essay on the Nature and Immutability of Truth of 1772, which contained key passages from the Treatise for the very first time. Proponents of this hypothesis argue that it was only by reading this text that Kant could have realized Hume’s skepticism concerning the causal principle in general and that this inspired his answer to the latter within the Kritik’s Transcendental Analytic.^[14] More recent commentators have argued that the decisive interruption occurred when Kant read Hamann’s translation of Treatise 1.4.7, which appeared in 1771.^[15] Kreimendahl argues that Kant read – and was inspired by – this translation some two years earlier in 1769, and that this text gave Kant the “great light” (18:69) he refers to in Reflexion 5037. Kreimendahl wishes to reconcile Kant’s claims about the interruption and his other claims about an awakening from dogmatic slumber and thus argues that Hume’s decisive influence occurred when Kant discovered the antinomy.^[16]

In my view of the matter, however, these later accounts of Hume’s decisive influence are in fact far less convincing than Fischer’s earlier hypothesis. The view that Kant’s dogmatic slumber was interrupted by quotations from the Treatise in which Hume doubts the causal principle in general cannot be squared with Kant’s own account of the interruption within the Prolegomena’s Preface, in which Hume’s very concept of causation, as opposed to the causal principle, is at issue. Kreimendahl’s view that Hamann’s translation of Treatise 1.4.7 awakened Kant to the problem of the antinomy also faces a difficulty, for although Treatise 1.4.7 indeed discusses how a faculty of the mind naturally falls into self-contradiction, the faculty in question is the imagination in its role of generating beliefs about everyday empirical objects, which is very far removed from the view that reason falls into self-contradiction when attempting to cognize the world-whole.^[17] Furthermore, Kreimendahl’s claim that Kant was awakened by Hume in 1769 is at odds with the fact that Kant’s first use of the awakening metaphor occurs in the Dreams of 1765 (see 2:342).

My aim in this article is to defend Fischer’s hypothesis that the interruption occurred in 1762/63 and that the Enquiry was all that was required to occasion it against the aforementioned objections.

In what follows I confront the first objection about the difference between Kant and Hume’s terminology by providing a careful comparison of key passages from Kant’s own writings and large swaths of the German translation of the first Enquiry in Sulzer’s edition of Hume’s Vermischte Schriften of 1755, which Kant probably acquired soon after its publication.^[18] In doing so I intend to show that, both in terms of terminology and subject matter, there are more parallels between this Humean text and Kant’s writings than have previously been assumed.^[19]

In response to the second objection that the conclusions Kant drew from his skeptical reflections concerning causation in the 1760s deviate from those drawn by Hume, I follow Fischer by claiming that Kant’s realization of the “Humean problem” (4:261) in fact drove him – for a period – to embrace the very same conclusions as those drawn by Hume, and thus, within Dreams, to appropriate Hume’s skepticism.^[20]

Commentators who deny this often do so insofar as they are making the very same assumption as those who make the third objection concerning a supposed disparity between the claim that Hume’s decisive impetus occurred in 1762/63 and Kant’s own statements about his philosophical development. This questionable assumption is that, in describing the interruption, Kant is describing an event that directly corresponds with the beginning of his work on the first Kritik. There is, however, no reason to believe that these two events overlapped. I thus support Fischer’s contention that Kant’s own path from dogmatism to criticism passed through an intermediate phase of skepticism,^[21] and thus – as Kienzler has more recently argued – Kant’s own remarks about the ‘history’ of human reason (see A671/B789, A855/B883) can be seen as reflecting the history of his own philosophical development.^[22] Kienzler also convincingly shows how, contrary to those who pose this third objection, Kant’s autobiographical remarks about his development actually fully accord with the hypothesis that the interruption took place in 1762/63.^[23]

The fourth objection can be dismissed by noting that some of Kant’s students had mentioned Hume’s influence in this period (if not earlier).^[24] In raising this objection, Erdmann points out that Mendelssohn, the author of one of the first critical responses to Hume in Germany, seemingly perceived no traces of Hume’s influence when reviewing the Negative Magnitudes in the Briefen, die neuesten Litteratur betreffend. Certainly, Hume is not named in this review. However, we would tentatively suggest here that when Mendelssohn refers to how the essay’s concluding question about the relation between real grounds and consequences might be answered by someone denying that real grounds are logical grounds with the claim that “every cause or power has a certain relation to something else, in which changes thus arise,”^[25] his formulation is so reminiscent of EHU 7.29n’s claims about the relative nature of our idea of power, the importance of which Sulzer emphasized in his commentary on Enquiry Section 7,^[26] that it is possible that Mendelssohn has Hume in mind in writing this. If this is the case, then Mendelssohn would see in Kant’s denial that real grounds are logical grounds the very same causal skepticism that lead Hume to his distinctive conception of power.

According to the fifth objection Kant would have mentioned Hume’s name if he had been influenced by him in the 1760s. I would suggest that it is Hume’s reputation as a ‘freethinker’ who defended opinions detrimental to natural theology and religion more generally^[27] that explains why Kant did not acknowledge his indebtedness to him. After all, Kant confessed to Mendelssohn in 1766 that “I am absolutely convinced of many things that I shall never have the courage to say” (10:69) and, when finally admitting that Hume interrupted his dogmatic slumber, it seems, as Anderson puts it, “as if his debt to Hume were a crime of which someone had accused him.”^[28]

In thus defending Fischer’s hypothesis, I acknowledge the substantial correspondence between my own interpretation of Kant’s development and his. Given the developments in scholarship within the years that separate us, however, there are also important points of divergence. One significant difference concerns the assumption of nineteenth-century scholarship that the composition of the Negative Magnitudes preceded that of Kant’s Der einzig möglicher Beweisgrund zu einer Demonstration des Daseins Gottes (hereafter: Beweisgrund), whereas it is now almost universally acknowledged that the converse is the case. Accordingly, I regard the Beweisgrund of 1762 as Kant’s last contribution to dogmatic metaphysics, and thus regard Kant’s subsequent realization of the Humean problem as interrupting his dogmatic slumber in the sense that it shattered his conviction in the metaphysical system outlined in the Beweisgrund.^[29]

In order to appreciate the position Kant held immediately prior to the interruption, this article begins by discussing the Beweisgrund. Secondly, it argues that the Negative Magnitudes illustrates a turn away from the latter, and thus represents the first testimony to Kant’s interruption of his dogmatic slumber. Thirdly, it shows that Dreams attests to Kant’s full awakening in arguing for the nullity of any attempts to cognize that which lies beyond experience.^[30] In the final section, I show how my account of Kant’s development in the early 1760s fully accords with his account of Hume’s influence within the Prolegomena’s Preface.

2 The Dogmatic Slumbers of 1762

Although there is much evidence to suggest that Kant was acquainted with Hume’s Enquiry during the composition of the Beweisgrund,^[31] that text grants no hearing to the Enquiry’s principal skeptical reflections concerning causation and rather espouses that which they throw into doubt.^[32] Hume acknowledged that we possess a commonsensical tendency to believe that reason can infer a priori the effects of which things are capable (EHU 4.8; PVME, 70–71), and that this tendency generates the belief among many philosophers that the fundamental laws of nature can be known a priori (EHU 4.9; PVME, 71). Whereas – at least within the Enquiry – he defended the possibility of a priori knowledge in mathematics (EHU 12.27; PVME, 368), Hume argued that truth-claims concerning causal relations conjoin ideas that are “totally different” and that therefore no such truths can be known a priori (EHU 4.9; PVME, 71–72). He thus denies that a priori analysis of our concepts of the most elementary of material things can discover either the effects of which such things are capable or any “supposed tye or connexion between the cause and effect which […] renders it impossible, that any other effect could result from the operation of that cause” (EHU 4.10; PVME, 72). Accordingly, he claims that, without appeal to experience, reason (i) lacks all insight into the powers that we presuppose that the most elementary material things are invested, (ii) is incapable of inferring a priori any effect from a supposed cause, and (iii) only knows laws of nature through experience.

The Beweisgrund’s claims about causation and laws of nature are diametrically opposed to Hume’s position. Thus, whereas Hume claims that our complex idea of body does not contain the idea of “power or energy (Vermögen oder […] Kraft)” (EHU 7.8; PVME, 155), Kant claims that the concept of body consists of the marks ‘extension,’ ‘impenetrability’ and ‘power (Kraft)’ (2:80).

The Monadologia Physica, published six years before the Beweisgrund, contains some of Kant’s justifications for this position. Kant here maintains that bodies are composites consisting of a finite number of simple parts, “occupy a space absolute” (2:287), and are extended in virtue of the external relations between their elements. Each element exerts an ‘orbit’ of repulsive force (1:481) in virtue of which bodies are impenetrable, and an attractive force, which explains cohesion. If this concept of body is accepted, it is possible to derive the force of inertia from the repulsive force that any material element exerts and gravitation from its attractive force. The Beweisgrund declares the latter to be “in all probability, a fundamental power (Grundkraft), which is a property of matter itself” (2:137–138). We can therefore understand the Beweisgrund’s claim that:

The only conditions under which it is possible to conceive what is supposed to fill space and what is supposed to be capable of thrust and pressure must be the very same conditions which necessarily yield the [laws of nature]. On this basis it can be seen that these laws of the motion of matter are absolutely necessary. That is to say: if the possibility of matter is presupposed, it would be self-contradictory to suppose it operating in accordance with other laws. (2:100)

Contra Hume, Kant thus maintains that reason can obtain a priori cognition of the laws of nature by analyzing the concept of body. The fact that material things conform to the laws of Newtonian dynamics is thus a truth determinable by the principle of identity.

However, Kant does not believe that all events are governed by these laws. For the Beweisgrund argues for something else Hume denies, namely, a qualitative distinction between (i) “alterations […] which [take] place mechanically in the corporeal world” and (ii) “actions which issue from freedom” (2:110). In the latter case, Kant maintains, reason lacks insight into the “determining grounds” that explain why certain actions follow from “the powers of freely acting beings (die Kräfte frei handlender Wesen)” (2:111). Thus, in the case of a freely acting agent, we cannot conceptually grasp why she chooses what she chooses, since these actions do not follow with a logical necessity in virtue of “determining grounds” constitutive of the concept of the agent (2:112).

To fully appreciate Kant’s justification for maintaining that the effects of simple material things are a logical consequence of their concept or essence, we must consider its broader context. The Beweisgrund promulgates a reformed Leibnizian-Wolffian metaphysics, which demonstrates the necessary existence of its foundational principle, namely a transcendent, omniscient and omnibenevolent God, in a superior manner to the attempted demonstrations of Leibniz and Wolff themselves. Section 1 argues that the God of Leibnizian-Wolffian metaphysics can only be demonstrated by means of the ‘possibility proof,’ which argues that such a God must exist as the ‘ground’ of all possible concepts or essences. Section 2, however, follows Sulzer’s recommendation in the Preface to his edition of Hume’s Enquiry that philosophers who demonstrate “universally useful truths” should subsequently present them by following the regressive path of the natural scientist so as to provide a presentation more appealing to the “universal common sense” of mankind.^[33] In Section 2, Kant engages in physico-theology to support the claim that the essences of particular things are grounded by the God who is demonstrated to exist by the possibility proof (see 2:91). Kant’s claim that the concept of body can only be consistently thought insofar as it is resolvable into the simpler concepts of ‘extension,’ ‘impenetrability’ and ‘power’ (2:80) occurs in Section 1. Accordingly, the conjunction of these concepts with the concept of body is not only justified by considerations concerning the way in which material elements fill space, but also by the claim that God’s understanding dictates that bodies are only possible insofar as they possess such characteristic marks. Meanwhile, the claim that laws of nature are derivable from the nature of matter itself is made in Section 2. The fact that even the simplest material things should universally possess characteristics from which so many useful consequences necessarily follow justifies the regressive inference that they possess one ground consisting of understanding and will. God’s understanding dictates that matter is only possible insofar as it essentially possesses the aforementioned characteristics, and God’s will actualizes it, because that something from which many useful consequences necessarily follow should exist conforms with God’s desire for the ‘best.’ The full justification for Kant’s claim that the effects of simple material things are a logical consequence of their concept or essence relies, therefore, on the theological foundation of Leibnizian-Wolffian metaphysics.

In advancing an innovative a priori demonstration of an omniscient and omnibenevolent God, which obviates criticisms of previous proofs, Kant thus believed he could justify the veracity of our commonsensical convictions concerning causality that Hume’s Enquiry threw into doubt. Indeed, he may be alluding to Hume when he remarks how those who deny that the very possibility of contingent things depend upon God in the way he has described would be forced to concede that the world might be “an extensive manifold, in which each individual thing [has] its own independent nature,” so that the ‘harmony’ and ‘unity’ that we find within it could only arise from “an amazing accident” (2:99) – a position which Hume could be said to adopt in EHU 5.21, and one which Kant, at least at this point in his career, regards as absurd.

In the Beweisgrund, therefore, it is Kant’s innovative a priori demonstration of an omniscient and omnibenevolent God that grounds his claim that matter is only possible insofar as it is invested with causal powers, which, in turn, grants the Newtonian laws of nature their logical necessity. As we have seen, he does not claim such certainty – as he also admits in the Prize-Essay of 1792/94 – “when it comes to forming a judgement about [God’s] free actions” (2:297), recognizing that our cognition of the effects which follow from God’s will – and indeed our own will – are more ‘obscure.’ As I will argue in the next section, Kant in the Negative Magnitudes extends this obscurity to his conception of causal interactions in general.

3 Negative Magnitudes: The Interruption of 1763

The Negative Magnitudes defends the thesis that – within all changes within the natural world – every coming-to-be entails a passing-away of the same magnitude in order to metaphysically ground Newton’s law of inertia and his doctrine of the conservation of energy. Within the bulk of the text Kant continues to presuppose that matter is invested with attractive and repulsive forces (see 2:179–180, 2:185, 2:191). There is, however, a marked half-heartedness in Kant’s defense of the essay’s claims, which he describes as “incomplete experiments” presented merely “in a problematic fashion” (2:197). In the concluding General Remark, Kant takes this half-heartedness one step further by offering reflections which undermine the very thesis of the essay.

Kant begins the General Remark by confessing the “frailty of my insight (der Schwäche meiner Einsicht),” which makes him “grasp (begreife) least of all that which all mankind believe they understand easily” (2:201); and he concludes it by contrasting himself with “those philosophers whose self-assumed insight (Einsicht) knows no limitations (Schranken)” (2:204). These remarks echo Hume’s claim in the Enquiry that, in spite of the fact that “the generality of mankind […] suppose (setzen voraus), that […] they perceive (begreifen) the very force or energy (Vermögen und die Kraft)” by means of which a cause is necessarily connected with its effect (EHU 7.21; PVME, 165) they are in fact deluding themselves, something which Hume claims illustrates “the weakness (Schwachheit) and narrow limits of human reason” (EHU 7.28; PVME, 177).^[34]

This echo is significant insofar as what Kant denies he lacks insight into is the very same thing which Hume denies we can grasp in the Enquiry. Before turning to this limitation, however, Kant discusses that of which he does possess insight, namely (a) relations of concepts established by analysis.^[35] He writes:

I fully understand how a consequence is posited by a ground in accordance with the rule of identity: analysis of the concepts shows that the consequence is contained in the ground. It is in this way that necessity is a ground of immutability, that composition is a ground of divisibility; that infinity is a ground of omniscience, etc., etc. And I clearly understand this connection (Verknüpfung) of the ground with the consequence, for the consequence is really identical (wirklich einerlei) with a part of the concept of ground. (2:202)

It is unsurprising that Kant asserts that our understanding has insight into relations of concepts that conform to the principle of identity. In the same vein, Kant claims that he has insight into (b) logical opposition, that is the ‘repugnancy’ of concepts determined by the principle of contradiction, from which nothing follows as a consequence.

What Kant says he lacks insight into are (c) the grounds that account for the real connection between heterogenous concepts. He writes:

What I should clearly like to have distinctly explained to me […] is how one thing issues from another thing, though not by means of the law of identity. […] [This] kind of ground […] I call the real ground, for while this relation (Beziehung) certainly belongs to my true concepts, the manner of this relating can in no way be judged. As for the real ground and its relation (Beziehung) to its consequence, my question presents itself in the following simple form: How am I to understand the fact that, because something is, something else is? (2:202)

As we have seen, Kant used similar terms in the Beweisgrund and prize-essay to describe the relation between free will and its consequences. He there described how we lack insight into the grounds for free actions insofar as, by definition, they are not necessitating grounds that essentially belong to the concept of the agent. Accordingly, we cannot form a ‘judgement’ about the grounds for why particular arrangements and events follow from God’s will (see 2:110–111, 2:297). It is therefore quite appropriate that Kant in the Negative Magnitudes returns to God’s free will to illustrate the more fundamental question concerning the relation between heterogenous real grounds and consequences. He thus tells us:

The will of God contains the real ground of the existence of the world. The will of God is something. The world which exists is something completely different. Nonetheless, the one is posited by the other. […] [Y]ou may subject the concept of divine willing to as much analysis as you please: you will never encounter in that concept an existent world as something which is contained within the concept of God’s willing, or something posited through that concept through identity. (2:202)

The problem is clear: The concepts of God’s will and an existing world are heterogeneous, and, as such, no analysis of the former will discover the latter, or indeed any mark convincing us that the former must invariably ‘posit’ the latter.^[36] This echoes Hume’s claims about how “the weakness (Schwachheit) of human reason” (EHU 7.24; PVME, 170) dictates that we cannot “comprehend (begreifen)” (EHU 7.25; PVME, 171) the “universal energy and operation (Kraft und Wirkung) of the Supreme Being” (EHU 7.24; PVME, 169). This problem was already implicit within the Beweisgrund. What is new in the Negative Magnitudes, however, is Kant’s extension of the same problem to the mechanical interactions of corporeal things, which means that he endorses Hume’s view that we are just as ignorant of why “motions may arise from impulse” as we are of the ‘operations’ of the ‘volition’ of “the Supreme Being” (EHU 7.25; PVME, 171).^[37] Kant thus writes:

A body A is in motion; another body B, lying in the direct path of A, is at rest. The motion of A is something; the motion of B is something else. (2:202)

He then makes it clear that what he asserted of God’s will is likewise true of body A: an analysis of the concept of body discovers neither the effects of which it is capable nor any mark which serves to explain why those effects invariably follow from its essence.

Kant’s example here could well reference the Enquiry’s frequent appeal to the interactions of billiard balls, an example that Hume uses twice within the conclusion of Enquiry Section 7 alone (see EHU 7.28, 7.30) to illustrate that conceptual analysis cannot discover circumstances within bodies from which their effects could be derived a priori (see EHU 7.28, 7.29).^[38] In the same vein, Kant maintains that our belief that body B is in motion because of the collision of body A is merely based upon a conjunction of objects, and because we cannot discover marks within body A from which the motion of body B could be derived a priori, we do not recognize any real necessary connexion amongst things themselves. As the discussion proceeds, Kant notes that terms such as ‘causation’ or ‘power’ already presuppose the very relation between real ground and consequence which is at issue:

[I am not] willing to be fobbed off by the words ‘cause’ and ‘effect,’ ‘power’ and ‘action’ (Kraft und Handlung). For if I already regard something as a cause of something else, or if I attach the concept of power (Kraft) to it, then I am already thinking of the cause as containing the relation (Beziehung) of the real ground to its consequence, and then it is easy to understand that the consequence is posited in accordance with the rule of identity. (2:203)

This view parallels Hume’s claim in Enquiry Section 7 that it is only insofar as we find one object constantly conjoined with another that we come to conceive the former as a ‘cause’ and the latter as an ‘effect.’ The concept of ‘power,’ meanwhile, Hume finds to be just as ‘relative’ as the concept of cause, since “both have a reference to an effect” (EHU 7.29n; PVME, 180n). Hume argues that it is only insofar as constantly conjoined objects are associated within our imagination that the concept of ‘power’ is generated. Accordingly, all talk of ‘cause and effect’ and ‘power’ presupposes that two objects are related in such a way that they are constantly conjoined, and thus these concepts cannot, without circularity, be employed to explain why these objects are related in the first place – which is exactly Kant’s point.

Interestingly, in his commentary to Enquiry Section 7, Sulzer identified the proposition “that we cannot think of a power (Kraft) if one does not at the same time clearly represent the manner of its effect”^[39] as the ‘first principle’ of the whole of Section 7’s argument. Whereas Sulzer argues against the legitimacy of such a first principle, however, Kant – at least at this point in his career – embraces it. Kant thus argues that it is only insofar as we are already thinking a relation between real ground and consequence that we come to regard the consequence as an effect of a cause and the cause as invested with power, that is, that the concept of power cannot explain why there is such a relation between ground and consequence in the first place. Kant returns to God’s creation of the universe to illustrate this:

‘[P]ower’ (Macht) signifies something in God, in virtue of which other things are posited. But this word already designates the relation of a real ground to its consequence, but it is this relation (Beziehung) which I wish to have explained. (2:203)

Kant thus tells us that we cannot without circularity employ the concept of God’s power to explain why and how he creates the world, just as we cannot without circularity employ the concept of force to explain why body A posits motion in body B.

Clearly, in appropriating Hume’s view about the relativity of the concept of power, Kant denies that we can have rational insight into the interactions between objects. This is further made clear by Kant’s discussion of (d) ‘real opposition’ that concludes the General Remark. Kant here claims that the cancellation of a motion within body A when it collides with body B is not a logical cancellation, and that it is not contradictory to assume that when body B is set in motion no motion is cancelled in body A. Since Kant insists that this “distinction between logical opposition and real opposition […] is parallel to the distinction between […] logical ground and real ground,” it follows that the cancellation of motion in body A cannot be derived from a priori conceptual analysis. As he puts it:

If I presuppose (voraussetzen) impenetrability, which stands in real opposition to any power (Kraft) which strives to penetrate the space occupied by a given body, I can already understand why the motions are cancelled. But in making that presupposition, I have reduced one real opposition to another. (2:203)

Kant thus tells us that seeking to explain the cancellation of motion in terms of an interaction of repulsive forces does not ultimately explain why motion in body B cancels motion in body A, and he also claims that the possession of such powers by these bodies is something we merely might ‘presuppose,’ just as he had done within the main body of the Negative Magnitudes. But by claiming that we can merely ‘presuppose’ that objects are invested with such powers, Kant clearly denies that the concept of ‘power’ can really be attributed to the concept of an object and, hence, be used to explain why one thing posits another or why the positing of one thing involves the cancellation of another.

The General Remark thus reveals that, at that point in his career, Kant has reached a full understanding of the force of Hume’s problem. He concludes it by confessing that he has “reflected upon the nature of our cognition with respect to our judgements concerning grounds and consequences,” and promises that “one day I shall present a detailed account of the fruits of my reflections” (2:203–204), which suggests that he was not entirely satisfied with the answer Hume himself had given to his question and was still searching for a better answer.

The Negative Magnitudes, therefore, marks a distinct shift from Kant’s position in the Beweisgrund, since it shows him relinquishing his view that causal relations – and the laws of nature – are cognized a priori. The text suggests that Kant not merely abandoned the view that metaphysics can justify such a conception of causality, but rather that, as we shall now see, his new insight into the nature of causality led him to lose his conviction in such metaphysics.

One moment of tension, already implicit within the Beweisgrund, was briefly alluded to in the prize-essay, before becoming corrosive within the Negative Magnitudes. In the Beweisgrund, Kant argued that the logical necessity of causal relations and laws of nature is grounded by God’s necessary existence. However, Kant here attributed the creation of the universe not to God’s necessary nature but to his free will. On this account, the actual existence of the universe is ultimately contingent, and Kant’s counterfactual claim that any ‘harmony’ and ‘unity’ we attribute to it is merely the product of “an amazing accident” (2:99) cannot be decisively ruled out. In 1762, Kant would have responded to such an objection by appealing to the “certainty which is moral” (2:297) that God wills the best, but one could still ask whether our moral certainty is enough to ground the conviction that the laws of nature are derivable from the essential nature of the elements of material things. Once Kant adopts Hume’s principle that all concepts of power are relative, the problem becomes more severe. For if we lack a priori insight into the power by which such a will might operate, we are deprived of all certainty about the extent of its abilities to actualize that which is best, which further weakens our conviction that the ‘harmony’ and ‘unity’ within the material universe is not the product of “an amazing accident” (2:99).

Another problem resulting from Kant’s adoption of Hume’s principle concerning the relativity of the concept of power is that it deprives us of a priori insight into all ‘real opposition’ within the material world, the possibility of which Kant had appealed to within his demonstration of God’s existence within the Beweisgrund (see 2:86). Whereas the possibility of such real opposition may be confirmed by experience, the Negative Magnitudes denies that it can be cognized a priori in accordance with the principle of contradiction, which makes it questionable whether it can play a role within any a priori demonstration of God’s existence. The role that it played for Kant within the latter, however, is far from insignificant, since it was employed to ascertain which properties of things were merely consequences of God and which properties belonged to His determinations and, thus, to affirm the transcendent and spiritual nature of the Deity (see 2:85–88). In this sense, the Negative Magnitudes’ denial of our ability to cognize ‘forces and powers’ through a priori analysis also undermines Kant’s earlier conviction in our ability to cognize the nature of God. Accordingly, in a marked change of tone from that of the Beweisgrund, the Negative Magnitudes describes “the cognition which we have of the Infinite Divinity” as “a cognition which is so fragile,” and approvingly quotes Simonides’ claim that “the more I reflect on God, the less able I am to understand Him” (2:200).^[40]

This is not the only place within the Negative Magnitudes where Kant distinguishes himself from other philosophers. At the beginning of the General Remark, he takes issue with “self-styled ‘thorough’ philosophers (gründliche Philosophen)” who profess to possess such a degree of insight into things that “nothing remains hidden from them which they cannot explain or understand (erklären und begreifen)” (2:201). In the conclusion of the General Remark, he challenges these thinkers to “test the methods of their philosophy” to see if they can give a more adequate answer to the question concerning relations between real grounds and consequences than his own, which is that such relations cannot be “expressed by a judgment” but only by a concept reducible to simple unanalyzable concepts (2:204).

That Kant’s target is the Leibnizian-Wolffian school is suggested by the Negative Magnitudes’ short Introduction, where he uses similar language to describe how a “metaphysical intelligentsia possessing perfect insight” are reluctant to test their doctrines against those that result from the application of mathematics within Newtonian physics, for the reason that in such a “situation […] learned nonsense cannot create the illusion of thoroughness (Gründlichkeit) as easily as it can elsewhere.” This Leibnizian-Wolffian metaphysics is further described as ‘false’ (2:170) – a term echoing Hume’s description of a “false and adulterate” metaphysics that should be destroyed and replaced (EHU 1.12; PVME, 14).

This contention that Kant is thinking of Leibnizian-Wolffian metaphysics is also suggested by his brief allusion to those whose ‘method’ is to philosophize more geometrio within the first paragraph of the Preface of the Negative Magnitudes, a method that Kant dismisses on the basis of the “refusal of the importunate non liquet to yield to all this pomp” (2:167). These remarks indicate that by the time of the Negative Magnitudes Kant has turned against his earlier position to express a marked skepticism about reason’s ability to demonstrate God’s existence more geometrico, and thus a marked skepticism about whether it is possible to rationally justify the conviction that any effects can be inferred from any causes in an a priori manner. This turn to skepticism is heralded by Kant’s use of Pyrrho’s non liquet against the dogmatists’ attempts to philosophize more geometrico. The skepticism he professes is primarily that of the skeptic described in EHU 12.22 who “justly insists” that “we have no other idea (Begriff) of [the relation (Beziehung) of cause and effect] than that of two objects (Gegenständen), which have been frequently conjoined” (EHU 12.22; PVME, 361) – that is, the kind of ‘skepticism’ professed by Hume himself.

The Negative Magnitudes thus evidences the interruption of dogmatic slumber. If, as I hope to have shown, Kant in 1763 embraced Hume’s skepticism concerning our ability to know any real connexion between cause and effect, then it makes sense that the Prolegomena credits Hume with inspiring precisely this turn. This account is further evidenced by Kant’s output over the next two years, which is addressed in the next section.

4 Dreams of a Spirit-Seer: The Awakening of 1765

Soon after the Negative Magnitudes, Kant published the Dreams of a Spirit-Seer. Whereas he describes the latter as an investigation into spirit-seeing (see 2:359), Kant’s letter to Mendelssohn of April 8, 1766, reveals that he took it to investigate the metaphysical problem concerning “how is the soul present in the world, both in material and in non-material things” (10:71). Kant also notes that this investigation “resolves itself” into a more fundamental investigation into “whether one can by means of rational inferences discover a primitive power (primitive Kraft); that is the primary, fundamental relationship (die erste Grundverhältnis) of cause and effect” (10:72). This formulation clearly testifies to the Negative Magnitudes’ question concerning reason’s insight into the relation of real grounds and consequences and the later formulation of “Hume’s question” as one that asks “whether [causality] is thought through reason a priori” (4:258–259).

In Dreams, Kant explicitly relinquishes claims he had made in the period between the Monadologia Physica and the prize-essay. The relation between the concepts of extended substance and impenetrability is thus no longer a relation into which reason has a priori insight in virtue of the former being the logical ground of the latter. Kant views the concepts of extended substance and impenetrability as heterogeneous, so that the relation between them is one that can only be established by experience (see 2:322).^[41] Likewise, Kant maintains that an extended substance cannot be the logical ground of attractive force, which means that reason has no a priori insight into the possibility of the latter. He thus views the repulsive and attractive forces as ‘fundamental powers’ (see 2:371), and the concepts of such powers as concepts of irreducibly simple relations existing between the elements of material things – the very kind of relations Kant had described at the end of the Negative Magnitudes (see 2:204).

In similar language to that used in the Negative Magnitudes, Kant tells us that the possibility of “the fundamental relations of causes and effects” in general can never “be rendered more distinct” (2:323). While Kant still maintains that the elements of material things must possess “some kind of inner activity” which produces these relations, he denies that reason can “specify in what that inner activity consists” (2:328). Thus, Kant’s ‘answer’ to ‘Hume’s question’ concerning reason’s insight into the relation of real grounds and consequences is now decidedly Humean insofar as, because of the heterogeneity of causes and effects, he explicitly denies that we possess any a priori insight into such relations and explicitly affirms that all our cognition of the most fundamental relations of cause and effect can only result from experience, to thus conclude that any claims about causes and causal powers not derived from experience “must be entirely arbitrary (gänzlich willkürlich)” (EHU 4.9; PVME, 71, EHU 4.11; PVME, 73). He thus writes:

It is impossible for reason (Vernunft) ever to understand (einzusehen) how something can be a cause, or have a power (Kraft), such relations can only be derived from experience. For our rule of reason only governs the drawing of comparisons in respect of identity and contradiction. If something is a cause, then something is posited by something else; there is not, however, any connection between the two things here which is based on agreement. Similarly, if I refuse to regard that same something as a cause, no contradiction will ever arise, for there is no contradiction in supposing that, if something is posited, something else will be cancelled. It follows from this that if the fundamental concepts of things as causes, of powers and actions (Kräfte und Handlungen), are not derived from experience, then they are entirely arbitrary (gänzlich willkürlich) and they admit of neither proof nor refutation. (2:370)

This passage harks back to the Negative Magnitudes’ distinction between logical grounds and real grounds and Kant’s claim that reason lacks all insight into the relation between the latter and their consequences. What is new here is that Kant now explicitly claims that the nature of the latter kind of relations can only be derived from experience, which is to say he has now also adopted Hume’s ‘answer’ to his ‘question.’

When Kant discusses the concept of ‘spirit’ it becomes clear how the adoption of Hume’s answer to his question informs his discussion of how the soul could be present in the world. He declares the concept of ‘spirit’ a “surreptitious concept,” that is, a concept which is “the product of covert and obscure inferences made in the course of experience. Some of these “surreptitious concepts […] are nothing but delusions of the imagination, whereas others are true” (2:320n). When we apply the concept of spirit to beings “which belong to the universe as constituents of it” (2:321n), we are assigning an “immaterial nature” to something (2:327n) which occupies space without filling it. Such a thing could therefore “be present in a space which was already occupied by matter” (2:321). We cognize that material things are related according to attractive and repulsive forces because this is constantly evidenced by experience; but we have no direct experience of a cancellation of the former kinds of relation and thus do not cognize a reciprocal connexion between spirit and a physical being constituting the communion between soul and body (see 2:321n). As he puts it in the aforementioned letter to Mendelssohn, the former kind of relations disclose “a relation of one external activity to another external activity”; whereas the latter relation discloses only “the reciprocal relationship of the inner condition (thinking or willing) of the soul to the outer condition of the material body” (10:72). If one assumes that the soul is a spirit, Kant says that there results “a kind of unthinkability” (2:323), for, if our concepts of the relations between things are derived from experience, they are concepts of relations between impenetrable things, whereas spirits are by definition not impenetrable.

However, Kant does not infer from this that spirits do not exist. Since the relation between ‘impenetrability’ and the concept of something occupying a space is not a logical one, reason lacks insight into whether the mark of ‘not-filling space’ is repugnant to the latter concept. Kant thus tells us that, “there is […] no demonstrable contradiction […] even though the thing itself remains unintelligible (unbegreiflich), if I assert that a spirit-substance […] occupies a space […] without filling it” (2:323). Given our lack of insight into the real grounds of the relations between material things, we cannot assert that the existence of spirits is impossible.

Like Hume, Kant therefore maintains (a) an agnosticism in regard to whether the soul be material or immaterial and (b) our inability to explain the communion of soul and body. Hume had argued that the fact that “the motion of our body follows upon the command of our will” is “a fact, which […] can be known only by experience” since “we are so far from being immediately conscious” of “the energy, by which the will performs so extraordinary an operation,” that is, “any […] power in the cause, which connects it with the effect” (EHU 7.10; PVME, 157–158). This lack of consciousness of the will’s ‘power’ to move the body means that we are ignorant of (a) “the nature of the human soul” (EHU 7.17; PVME, 163) and (b) “the union of the soul with the body” (EHU 7.11; PVME, 158). To emphasize this, Hume writes:

Were we empowered, by a secret wish, to remove mountains, or control the planets in their orbit; this extensive authority would not be more extraordinary, nor more beyond our comprehension (Begreifungskraft). (EHU 7.11; PVME, 158)

Hume’s point is that our cognition that our will moves our body but not the planets rests entirely on experience and is not based on a priori insight into circumstances within the will which would explain why it can move the one and not the other. For Hume denies that we possess any a priori insight into how any cause produces certain effects, which means that all causal inferences must either be based on experience or be “entirely arbitrary (gänzlich willkürlich)” (EHU 4.9; PVME 71, EHU 4.11; PVME, 73). At the end of the Enquiry, he uses the same example:

If we reason a priori, any thing may appear able to produce any thing. The falling of a pebble may, for aught we know, extinguish the sun; or the wish of a man control the planets in their orbits. (EHU 12.29)

These passages are referenced in Dreams, where Kant likewise asserts that no amount of analysis of the human will discovers circumstances explaining how and why it controls our body:

I know, of course, that thinking and willing move my body, but I can never reduce this phenomenon, as a simple experience, to another phenomenon by means of analysis; hence, I can recognize the phenomenon, but I cannot understand it (erkennen, aber nicht einsehen). That my will moves my arm is no more intelligible to me than someone’s claiming that my will could halt the moon in its orbit. The only difference between the two cases is this: I experience the former, whereas my senses have never encountered the latter. (2:370)

Like Hume, therefore, Kant claims that while we experience that volition moves our bodies, analysis of the concept of will grants us no insight into how and why it does so.

He also claims that the relation between volition and bodily movement cannot be explained in terms of the relation between an extended thing and features such as impenetrability and gravitation. In the letter to Mendelssohn, Kant writes that we experience “the reciprocal relationship of the inner condition […] of the soul to the outer condition of the material body,” but cannot further explain it because we cannot experience “the subject’s external power or capacity (Kraft oder Fähigkeit)” (10:71–72). Any positive assertion about the latter can thus “only be the product of poetic invention (nur erdichtet)” and “a heuristic fiction or hypothesis” (10:72).

In Dreams itself, Kant asserts that such a ‘hypothesis’ cannot possess the same status as a hypothesis within natural science. For the latter kind of hypothesis seeks to explain phenomena in terms of the attractive and repulsive forces that one knows through experience. Attempts to explain the union of the soul and body within rational psychology, by contrast, “invent fundamental powers (Grundkräften)” (2:371), the very possibility of which cannot even be proven. For this reason, the Dreams advocates abandoning all investigation into “alleged experiences” that “cannot be brought under any law of sensation” (2:372).

Kant’s discussion reminds us of Hume’s discussion of ‘miracles’, that is, events which violate laws of nature through either “a particular volition of the Deity” or “the interposition of some invisible agent” (EHU 10.12n). Bodily action occasioned by spontaneous voluntary activity on the part of an immaterial soul would indeed be a ‘miracle’ in Hume’s sense. As well as investigations into the union of soul and body, Kant therefore proscribes investigations into a “liberty […] opposed to necessity” (EHU 8.25) and a future state. He writes:

Questions concerning the spirit-nature, freedom, predestination, the future state […] initially activate all the powers of the understanding. […] But […] if […] philosophy should subject its own procedure to judgement, and if it should have knowledge not only of the objects themselves but also of their relation to the human understanding, its frontiers (Grenzen) will contract in size and its boundary-stones will be securely fixed. […] We found that some philosophy was necessary if we were to know the difficulties surrounding a concept [i. e., the concept of spirit]. […] Somewhat more philosophy removes this phantom of knowledge (entfernt dieses Schattenbild der Einsicht) still further away, convincing us that it lies wholly beyond the horizon of man. (2:369–370)

The passage reminds us of Hume’s claims that reconciling “the indifference and contingency of human actions with prescience” exceeds “all the power of philosophy” since such investigations take it outside of “her true and proper province” (EHU 8.36; PVME, 232). This province is determined by a self-reflexive enquiry into the “powers and capacity” of the human understanding that reveals that “it is by no means fitted for such remote (entfernete) and abstruse subjects” (EHU 1.12; PVME, 14).

Kant’s explicit undermining of rational psychology (and implicit undermining of other branches of special metaphysics) in Dreams is thus the result of his appropriation of Hume’s ‘answer’ to his ‘question.’ In the concluding paragraph of Dreams, Kant notes in similarly Humean terms that we should not fear that undermining rational psychology will have significant practical implications. As he puts it, “science in its vanity (Eitelkeit)” pretends that “a rational understanding (Vernunfteinsicht) into the spirit-nature of the soul” is required to support belief in a future state, which in turn is an important incentive to morality, but in this it is deluding itself since “the heart of man [contains] within itself immediate moral prescriptions” (2:372). Hume had likewise criticized traditional rationalist metaphysics for its ‘vanity (Eitelkeit)’ as regards its attempt to “penetrate subjects utterly inaccessible to the understanding” (EHU 1.1; PVME, 12), and attacked the notion that “any new principles of conduct and behaviour” follow from belief in a ‘future state’ (EHU 11.23; PVME, 321).

Dreams also evidences a marked skepticism regarding the rationalist method of philosophizing more geometrico. We already saw how the Negative Magnitudes raised Pyrrho’s non liquet against this method; within the Dreams, however, skepticism about this method is more pronounced. The procedure of philosophizing more geometrico is described as dogmatic and criticized for involving the ‘difficulty’ that “one starts I know not whence, and arrives, I know not where: the advance of the arguments refuses to correspond to experience.” Kant further states that “the most general and most abstract concepts” with which those who philosophize more geometrico begin can never ‘explain’ the actual world (2:358).

Kant also takes issue with the opposed ‘historical’ procedure. Echoing EHU 4.12, Kant describes how the ascent from “empirical cognitions” to “higher concepts” is “far from sufficiently learned or philosophical” since it “soon leads to a Why? to which no answer can be given” (2:358). He devotes more attention, however, to satirizing philosophers who begin by employing the dogmatic procedure and seek to confirm their results by the historical one; a method recommended by Sulzer in the Preface to his edition of Hume’s Enquiry.^[42] In showing how Leibnizian metaphysics could be employed to rationally justify Swedenborg’s delusions, Kant suggests that the entirety of Dreams could be read as a parody of philosophical texts that currently employ such a method. He writes:

I cannot blame the […] reader […] if […] he has begun to feel reservations about the method which the author has thought proper to follow. For by placing the dogmatic part of the work before the historical part […] I must have created the suspicion that I was proceeding in a cunning fashion […] my purpose being to end by surprising the completely unsuspecting reader with a welcome confirmation derived from experience. And, indeed, this is a stratagem which philosophers have very successfully deployed on a number of occasions. […] [Such a] philosopher […] recognized that his rational arguments […] and experience […] would probably, like two parallel lines, continue to run side by side. […] Our philosopher thus reached an agreement with his fellow philosophers […]: each would adopt his own starting point […]; after that, rather than follow the straight line of reasoning, they would rather import to their arguments an imperceptible clinamen by stealthily squinting at the target of certain experiences or testimonies, they would thus steer reason in such a fashion that it would be bound to arrive at precisely that point which would surprise the unsuspecting student; they could prove, namely, what they all along knew was going to be proved. […] Adopting this ingenious method, various men of merit have even suddenly come upon mysteries of religion on the bare path or reason. (2:358–359)

In view of Kant’s own earnest employment of precisely this method in the Beweisgrund, this passage is especially striking.^[43] It provides us with decisive evidence that, by 1765, Kant’s conviction in the system he sketched in 1762 has entirely disintegrated, and that he no longer believed that one can appeal to the Leibnizian-Wolffian God to answer ‘Hume’s question.’

The text of the Dreams only mentions God once, when Kant remarks that:

The concept of the spirit-nature of [the Infinite Spirit] is easy, for it is merely negative and consists in denying that the properties of matter belong to it, for they are incompatible with an absolutely necessary substance. (2:322n)

Later on, however, Kant notes that “the possibility of [the] negations” which “one has to make do with […] if one is to think something which differs so much from anything of a sensible character” can be ‘based’ only “on a fiction, in which reason stripped of all assistance whatever, seeks its refuge” (2:352). Kant thus implicitly asserts the problematic status of claims about God in much the same way that he explicitly rejected claims about an immaterial soul-substance. One wonders whether this is one of the “many things” of which Kant was convinced, but, as he mentioned in his letter to Mendelssohn of 1766, he “should never have the courage to say” (10:69).

This letter also refers to methodological issues. Kant thus takes “the methods now in vogue” to “infinitely increase the amount of folly and error in the world.” Regarding the ‘knowledge’ produced by rationalist metaphysicians who philosophize more geometrico, Kant says that it is “best to pull off its dogmatic dress and treat its pretended insights skeptically” (10:70). Kant thus describes how the dogmata of rationalist metaphysicians, which cannot be justified by experience, can, through the skeptical procedure of positing equipollent counterclaims, be exposed as ‘entirely arbitrary’ – and thus possess the same ‘arbitrariness’ that Hume identified as pertaining to any attempt to prove something through causal inferences a priori (see EHU 4.9; PVME, 71, EHU 4.11; PVME, 73, 2:370).^[44] Kant indeed adopts such a skeptical procedure within the Dreams when he claims that any explanations of the union of soul and body “admit only of a very superficial proof or no proof at all,” and yet can only be “refuted in a correspondingly weak fashion” (2:326); the mutual destruction of arguments proving that this union “far transcends [our] powers of understanding (Einsicht)” (2:328). Likewise, within the second chapter of Dreams Part 1, Kant appeals to Leibnizian metaphysics to attempt a rational justification of the claims of the spirit-seer, while the very next chapter offers an equipollent natural-scientific explanation of such phenomena, which exposes both types of arguments to be mere delusions.^[45] Furthermore, at the beginning of the third chapter, Kant compares the private nature of spirit-seer’s visions with the lack of scientific consensus within metaphysics, and thus shows how whole systems of metaphysics may be treated skeptically by counterpoising them against those of their antagonists. He writes:

Aristotle somewhere says: When we are awake (Wenn wir wachen) we share a common world, but when we dream each has a world of his own. It seems to me that one ought, perhaps, to reverse the final clause and be able to say: if different people have each of them their own world, then we may suppose that they are dreaming. On this basis, if we consider those who build castles in the sky in their various imaginary worlds […] – if we consider, for example, the person who dwells in the world known as The Order of Things, a world tinkered together by Wolff […] or the person who inhabits the world which was conjured out of nothing by Crusius – if we consider these people, we shall be patient with their contradictory visions, until these gentlemen have finished dreaming their dreams. For if they should eventually […] awake completely (völlig wachen), that is to say, if they should eventually open their eyes to a view which does not exclude agreement with the understanding of other human beings, then none of them would see anything which did not, in the light of their proofs, appear obvious and certain to everybody else as well. And the philosophers will all inhabit a common world together at the same time, such as the mathematicians have long possessed. And this important event must now be imminent, if we are able to believe certain signs and portents which made their appearance some while ago above the horizon of all the sciences. (2:342)

Here we find Kant’s first use of an ‘awakening’ metaphor. Just like Kant’s admission within the Prolegomena’s Preface, he is assuredly referencing Sulzer’s wish that his edition of Hume’s Enquiry might “awaken” German philosophers “a little from their rest (aus ihrer Ruhe ein wenig aufwecken).”^[46] Sulzer was hoping, in other words, that the philosophers of the Leibnizian-Wolffian school would become more self-critical about the extent to which their proofs addressed all possible objections. The kind of ‘awakening’ Kant had undergone by 1765 is, however, more far-reaching: Kant sees that each philosopher who employs the ‘dogmatic’ procedure inhabits “his own world” (2:342), a ‘world’ which can be counterpoised against the equipollent ‘world’ of a metaphysical antagonist, to reveal the impossibility of scientific consensus regarding such claims. Kant’s awakening thus signifies the rejection of the metaphysics of the Leibnizian-Wolffian school per se.^[47]

5 The Prolegomena’s Preface: Confirmation

I will now consider how my account of Kant’s output between 1762 and 1765 sheds light on his discussion within the Prolegomena’s Preface. The Preface begins, much like the Enquiry, with outlining the need for a reform of metaphysics. Kant writes that “there is as yet no metaphysics at all” (4:257), which parallels remarks he made two decades earlier, when he first turned against philosophizing more geometrico (2:170, 2:283). He also writes that “a rebirth of metaphysics […] is inevitably approaching” (4:257), a remark which likewise parallels those in Dreams about the ‘imminent’ arrival of a truly scientific metaphysics (2:342). In Dreams, Kant may well have been thinking of the “true metaphysics” Hume described in the Enquiry, whereas the Prolegomena evidently alludes to the scientific metaphysics for which it is a preparation. Since Hume took himself to destroy “false and adulterate” (EHU 1.12; PVME, 14) metaphysics so as to cultivate “true metaphysics,” it is unsurprising that Kant introduces Hume as someone attacking the previous ‘unreformed’ metaphysics.

Hume, Kant says, “brought no light to [metaphysical] knowledge, but he certainly struck a spark from which a light could well have been kindled” (4:257). As becomes clear from the subsequent uses of this metaphor, ‘to bring light to metaphysical knowledge’ means to give a positive account of how a truly scientific metaphysics is possible; ‘striking a spark,’ however, means to engage in the primarily negative enterprise of demonstrating the inadequacy of unreformed metaphysics. We can thus see that the critical Kant took Hume to have succeeded in pointing out the inadequacies of previous unreformed metaphysics, but not in replacing it with a truly scientific version of the same.

Kant proceeds to specify the nature of Hume’s ‘attack’ on metaphysics. He writes:

Hume started mainly from a single but important concept (Begriffe) in metaphysics, namely, that of the connection of cause and effect (Verknüpfung der Ursache und Wirkung) (and of course also its derivative concepts of power and action (Kraft und Handlung) etc.); and called upon reason, which pretends to have generated this concept in her womb, to give him an account of by what right she thinks: that something could be so constituted (beschaffen) that, if it is posited, something else necessarily must thereby be posited as well; for that is what the concept of cause says. (4:257)

Kant thus discusses Hume’s concept of the “supposed […] connexion between cause and effect (Verknüpfung zwischen der Ursache und der Wirkung) which […] renders it impossible, that any other effect could result” (EHU 4.10; PVME, 72). This was the explicit theme of Enquiry Section 7, which specifically describes the “ideas (Begriffe) […] of power (Kraft) […] or necessary connexion (notwendigen Verknüpfung)” as “ideas (Begriffe), which occur in metaphysics” (EHU 7.3; PVME, 152). Kant is referencing how Hume denies that such concepts are concepts of “sensible qualities (Beschaffenheiten)” contained within our complex concept of body (EHU 7.8; PVME, 155) while also claiming that it is impossible to “define a cause, without comprehending, as part of the definition, a necessary connexion with its effect” (EHU 8.25; PVME, 218). Kant’s description of causation as the positing of something heterogenous following upon the positing of something else harks back to a similar formulation in Dreams (2:370) and his discussion of the relation between real grounds and consequences at 2:202 and 28:12. Kant therefore recalls how Hume – on the assumption that reason only has insight into real relations that conform to the laws of identity and contradiction – questioned reason’s insight into causal relations. He thus continues:

He undisputedly proved that it is wholly impossible for reason to think such a connection a priori and from concepts (a priori und aus Begriffen eine solche Verbindung zu denken), because this connection (Verbindung) contains necessity; and it is simply not to be seen how it could be, that because something is, something else necessarily must also be, and therefore how the concept of such a connection (Verknüpfung) could be introduced a priori. (4:257)

In this regard, as we have seen, the Enquiry claims indeed that the causal inferences a priori that reason attempts are “entirely arbitrary (gänzlich willkürlich)” (EHU 4.9; PVME, 71, EHU 4.11; PVME, 73, 2:370), which means that the concept of necessary connexion which is a constituent of the concept of causation has not been generated in reason’s womb. Kant thus continues:

From this he concluded that reason completely and fully deceives herself with this concept, falsely taking it for her own child, when it is really nothing but a bastard of the imagination, which, impregnated by experience, and having brought certain representations under the law of association, passes off the resulting subjective necessity (i. e., habit) for an objective necessity (from insight (Einsicht)). (4:257–258)

Thus, according to Kant, Hume argued that anyone who believes they possess ‘insight’ into the circumstances within certain objects which explain why they produce certain effects are deceiving themselves. It is the imagination’s association of concepts of objects that experience has found to be constantly conjoined that generates the concept of necessary connexion (see EHU 7.28). As we saw, Kant denied that he possessed insight into such circumstances within the Negative Magnitudes (see 2:201–202), at the very same time in which he exhibited a turn towards skepticism. The reason for why the one would have led to the other is stated in the subsequent passage:

[Hume] concluded that reason has no power (Vermögen) at all to think such [necessary] connections [of cause and effect], not even merely in general, because its concepts would then be bare fictions (Erdichtungen), and all of its cognitions allegedly established a priori would be nothing but falsely marked ordinary experiences; which is as much to say that there is no metaphysics at all, and cannot be any. (4:258)

Kant thus takes Hume’s ‘answer’ to his ‘question’ to consist in the claim that, since reason is deprived of all a priori insight into the dynamical relationships between objects, the latter can only be cognized through experience. As we saw, Kant embraced this position within Dreams, arguing that any metaphysical claims about the “powers (Kräfte) of spiritual substances” could be nothing other than the kind of ‘bare fictions’ here described (10:72). Kant thus sees Hume’s conception of “the connection of cause and effect” (4:257) as an ‘attack’ upon metaphysics, since, once accepted, it proscribes any rational inference beyond the sphere of experience.^[48]

The critical Kant thus took Hume to question the very possibility of metaphysics, a question which in his view heralds “a complete reform of the science” (4:258). The alternative answer defends the possibility of metaphysics by arguing, against Hume, that “the concept of cause […] has an inner truth independent of all experience, and therefore also a much widely extended use which is not limited to objects of experience” (4:258–259). Thus, Kant defends at least reason’s right to think beyond the sphere of experience. It is within the context of a discussion of this point that Kant then introduces his famous admission that:

I freely admit that the objection (Erinnerung) of David Hume was the very thing that many years ago first interrupted my dogmatic slumber and gave a completely different direction to my researches in the field of speculative philosophy. (4:260)

The case for reading ‘Erinnerung’ as ‘objection’ (and not ‘remembrance’) has already been convincingly justified by others.^[49] On this account, Kant tells us that it was Hume’s claim that reason can have no a priori insight into causal relations that ‘interrupted’ his conviction in the dogmatic method of philosophizing more geometrico, made him lose faith in the dogmatic metaphysical system he sketched within the Beweisgrund, and provoked a turn towards skepticism. The interruption Kant describes thus took place in 1763. However, the “completely different direction” Kant mentions in the passage cited above is not one upon which Kant embarked within the same year.

Kant continues by noting that “I was very far from listening to [Hume] with respect to his conclusions” (4:260). This claim may be a little disingenuous, for, although at the end of the Negative Magnitudes he implies that he is searching for an alternative answer to Hume’s question (see 2:204), Dreams shows him appropriating Hume’s own answer – and thus Hume’s conclusions. Fischer likewise detects a certain disingenuousness about Kant’s claim here. Expressing sentiments with which I fully concur, Fischer writes that:

These words no longer deceive us, since we have shown that, in the period in question [i. e., in the period in which Dreams was composed], Kant also listened to his predecessor with respect to his conclusions. […] [This was a period] in which he, like Hume, identified the problem of cognition with the question concerning cognition of the real ground; in which the latter was for him the most important concept; in which he, like Hume, wanted to know how to derive it from experience; in which he, like his sagacious predecessor, held the deduction of these concepts from mere reason to be impossible and therefore held metaphysical systems to be “dreams of reason.”^[50]

Kant, however, was a dissatisfied sceptic, and so his output subsequent to Dreams shows him searching for an alternative answer to Hume’s question, namely, one that defends the possibility of a reformed metaphysics. However, Kant remained indebted to Hume for “the first spark of this light” (4:260). Since Kant remarks in an unpublished note that “the year ’69 gave me a great light” (18:69), we may suppose that it was only in 1769 – the year that Kant discovers the ideality of space and time – that he realized that an alternative answer to Hume’s question that defended the possibility of metaphysics might be possible at all. This discovery enabled Kant’s realization that pure reason not only proceeds by means of identity and contradiction, but is also involved in the synthesis of a priori given manifolds, of which the relation between real grounds and consequences is only a particular mode. By expounding the science of transcendental logic, which explains the extent of pure reason’s capacity to cognize the relations of objects, the Kritik thus serves as “the elaboration of the Humean problem in its greatest possible amplification” (4:261).