In a famous letter to his student and friend Markus Herz of 21 February 1772, Kant describes two ways in which it is possible—indeed “easy”—to construe “the relation of that in us which we call representation (Vorstellung) to the object” as one of “conformity (Konformität)”: either the things cause the representations in the mind, or the representations in the mind cause the things:
If the representation comprises only the manner in which the subject is affected by the object [hereafter: Way One or the First Way], then it is easy to see how it is in conformity with this object, namely as an effect accords with its cause, and it is easy to see how this modification of the mind can represent something, that is, have an object … Similarly if that in us which we call “representation” were active with regard to the object, that is, if the object itself were created (hervorgebracht) by the representation (as when divine cognitions are conceived as the archetype of things) (wie man sich die göttliche Erkenntnisse als die Urbilder der Sachen vorstelle) [Way Two or the Second Way], the conformity of these representations to their objects could also be understood. (10:130) Footnote 2
The First can be called the Way of Affection, the Second that of Intellectual Intuition. As for the problem to which Kant is drawing Herz’s attention here, it apparently has nothing to do with the nowadays obvious and perhaps insurmountable objections to the naive (causal and representative) realist solution embodied in Way One. Footnote 3 Nor does Kant allude to the misgivings he was to express later regarding the intelligibility of the Second Way. Footnote 4 For upon the quoted passage follows the confident assertion that “the possibility of both an intellectus archetypus (an intellect whose intuition is itself the ground of things) and an intellectus ectypus, an intellect which would derive the data for its logical procedure from the sensuous intuition of things, is at least comprehensible” (ibid.). Since this is as much as to say that Way Two is quite as intelligible as Way One, the problem must lie elsewhere. It stems from the fact that our human intellect is simply not an intellectus archetypus, though it is not simply an intellectus ectypus either, being also a faculty of pure concepts (the categories) not derived by any logical procedure from matter furnished by the senses. As Kant puts it here, human understanding is “neither the cause of the object … nor is the object the cause of our intellectual representations in the real sense (in sensu reale)” (10:130, e.a.). Footnote 5 Yet, apart from these two, there is no other way (or none that Kant can see in 1772) in which the conformity of representations with their objects can be satisfactorily understood.
The key word is ‘satisfactorily.’ For, in the same letter, Kant goes on to describe a Third and even a Fourth Way, both of which he considers unsatisfactory:
Plato assumed a previous intuition of divinity as the primary source of the pure concepts of the understanding and of first principles. Mallebranche [sic] believed in a still-continuing perennial intuition of this primary being [or divinity] … Crusius believed in certain implanted rules for the purpose of forming judgments and ready-made concepts that God implanted in the human soul just as they had to be in order to harmonize with all things. Of these systems, one might call the former [two, i.e., those of Plato and Malebranche] the Hyperphysical Influx Theory [influxum hyperphysicum] and the latter [the system of Crusius] the Pre-established Intellectual Harmony Theory [harmoniam praestabilitam intellectualem]. (10:131) Footnote 6
The Way of Hyperphysical Influx can be called the Platonic Way since, on closer examination, it proves to have more to do with the anamnesis theory than with Malebranchian occasionalism. As for the Fourth Way, it bears, despite its name, a greater resemblance to Descartes’s theory of innate ideas or dispositions implanted in the human mind by a veracious God than to Leibniz’s theory of pre-established harmony. In the First Critique, where the third and fourth are collapsed into a single way, Kant speaks of “a kind of pre-formation system of pure reason” (B167).
This paper attempts to paint a fuller picture of the four ways in which Kant believed the problem first formulated in 1772 to have been solved (to their own satisfaction) by his predecessors and near-contemporaries, with only occasional references to Kant’s own Fifth Way, the new Critical Path of Transcendental Idealism or Copernican Way. Apart from the fact that none of the ways has anything to do with God’s existence (the subject of the famous quinque viae of St. Thomas), what may seem to speak against its title is Kant’s apparent abandonment of this version of the problem in favour of the “General” (Allgemeine) or “proper (eigentliche) problem of pure reason” (B19) in the re-written Introduction to the second edition of Critique of Pure Reason (1787), as indeed already in the Prolegomena (1783), where “How are synthetic propositions (Sätze) a priori possible?” is called the “Main Transcendental Question” (4:280). What this objection overlooks is that Kant continued to speak, now of two, now of three ways, even in the Prolegomena and in the Second Edition of the Critique. The systematic elaboration of what may be called the Fifth or Copernican Way in the works of his maturity is rightly considered the proper task of historical Kant scholarship, but consideration of the other four ways sheds an interesting side-light on the manner in which Kant himself understood the historical bearings of his systematic enterprise well into the critical phase of his thought.
There are, in addition, two quite specific benefits of focusing attention on the rejected four to the neglect of Kant’s own Fifth Way. First, tracing Way Two back to its origins in different medieval and early modern thinkers will go a long way toward explaining the fact that Kant refers to divine cognition, now as an intuitus originarius (B72), now as an intellectus archetypus (A695/B793), in different places within the first Critique. Footnote 7 Admittedly, this requires far more attention to medieval sources than is customary in studies devoted primarily to Kant; but even if Kant did not know these sources at first hand or in detail, an indirect and general acquaintance, via the writings of his predecessors, especially Berkeley, is enough to explain both the fundamental ambiguity in his conception of the divine intellect and the puzzling terminological variation to which it gives rise (see Section II below, in fine). In addition, replacing the usual schematic with an historically detailed picture of Way Two leads to the conclusion that Kant’s Copernican Way involves a new version of the so-called “theocentric model” of human cognition (see the Conclusion on Kant’s alleged anthropocentrism).
I. The Way of Affection or Physical Influx
Aside from explaining their origin positively and attempting to show how “intellectual representations” can (and why they must) be objectively valid, the Critique and Prolegomena are more forthcoming than the letter of 1772 on why the conformity of pure concepts with their objects cannot be explained by the First Way. By contrast, they shed no light at all on how Way One can explain the contingent agreement of empirical concepts with their objects. These two points will be considered in turn.
On what grounds, then, must Way One be ruled out as a possible explanation of the conformity of pure concepts with their objects? A passage from the first, which was retained unaltered in the second, edition of the Critique states, almost as a truism, that the objective validity of “concepts that are to be related to their objects a priori … cannot be established (dargetan) a posteriori, for that would leave that dignity (Dignität) of theirs entirely untouched” (A135f./B175). This is obviously directed against the—to Kant’s mind, totally misguided—efforts of Locke and Hume, who, as he says a few pages earlier, having “encountered pure concepts of the understanding in experience, also derived them from experience” (B127), thus stripping them of their “dignity” as concepts possessing necessary and strictly universal objective validity. The talk was already of dignity in a still earlier passage where Kant pointed out that “to the synthesis of cause and effect there attaches a dignity that can never be expressed empirically, namely, that the effect does not merely come along with the cause, but is posited through it and arises from it. This strict universality of the rule is therefore not any property of empirical rules” (A91/B124). And where there is strict universality, there is also necessity, and vice-versa, for Kant (cf. B4), who, as early as 1772, made necessary agreement the crux of the hitherto neglected problem
as to how my understanding may, completely a priori, form for itself concepts of things (Dingen), with which concepts the things (Sachen) should necessarily agree, and as to how my understanding may formulate real principles (Grundsätze) concerning the possibility of such concepts, with which principles experience must be in exact agreement and which nevertheless are independent of experience. (10:131, e.a.)
Nevertheless, Kant still writes the following, at the very end of the Transcendental Deduction, echoing an almost identically worded passage from the long run-up to the Deduction proper (A92/B124, to be cited presently):
Now there are only two ways in which a necessary agreement of experience with the concepts of its objects can be thought: either the experience makes these concepts possible [Way One] or these concepts make the experience possible [Kant’s new Way Two].” (B166)
True, the passage continues: “the first [Way One] is not the case with the categories (nor with pure sensible intuition)” (B166-167), which is to say, contrary to the opening statement, that this particular way does not issue in necessary and strictly, but at most in contingent and comparatively, universal agreement. An earlier passage, which likewise evokes “two possible cases in which synthetic representations and their objects can come together, necessarily relate to each other, and, as it were, meet one another” (A92/B124, e.a.), continues in the same vein:
If it is the first [i.e., if “the object alone makes the representation possible,” or Way One], this relation is only empirical, and the representation is never possible a priori. And this is the case with appearance in respect of that in it which belongs to sensation. But if it is the second [the new Way Two or Way of Transcendental Idealism], then since the representation in itself (for we are not here talking about causality by means of the will) does not produce its object as far as its existence is concerned, the representation is still determinant of the object a priori if it is possible through it alone to cognize something as an object. (A92/B125)
The parenthesis in this passage is clearly intended to ward off confusion of this new Way Two with that of the letter to Herz. However, the first sentence, like its counterpart in the previously cited passage, is still simply false as it stands, and the whole passage contradictory, since Way One is not so much a possible way of accounting for “necessary agreement” as a logical muddle. Kant himself says as much in the Prolegomena, where he again substitutes the Way of Transcendental Idealism for Way Two of the letter to Herz, but concludes by asserting, correctly this time, that there is only one way in which to understand necessary agreement between representations and their object, his own Copernican Way:
necessary agreement between the principles of possible experience and the laws of the possibility of nature can come about only from one of two causes: either [Way One] these laws are taken from nature by means of experience, or, conversely [the new Way Two], nature is derived from the laws of the possibility of experience in general and is fully identical with the mere universal lawfulness of experience. The first one contradicts itself; for the universal laws of nature can and must be cognized a priori (i.e., independently of all experience) and set at the foundation of all empirical use of the understanding; so only the second remains. (4:319, e.a.)
Since the problem was already stated in terms of necessary agreement in the letter to Herz (10:131), it only stands to reason that Kant already considered Way One self-contradictory in 1772. Question-begging though this dismissal of the First Way may seem from the perspective of his empiricist opponents, how is it that Kant, who takes necessary agreement as a requirement, continues to speak of “two ways” right through to the B-version of the Transcendental Deduction (see B166-167, partially quoted above)? Granted, it is no longer the same dilemma—Kant, as noted above, is at pains to ward off any confusion between the old and the new Way Two; nor is it any longer really a dilemma in the customary sense, since the second of the two ‘horns’ is now perfectly graspable. Still, the First Way is precisely that of 1772, so that the contradiction that Kant imputes to it renders his own unqualified statement that there are “two possible cases” (A92/B124) not just false but contradictory. Only the Prolegomena expressly retracts that statement in a manner that suggests how the earlier passages can be more sympathetically interpreted.
In Kant’s defence, then, the passages from the Transcendental Deduction can be regarded as loosely worded expressions of a point of view that is perfectly consistent and true (as far as it goes). What Kant meant to say already in the letter to Herz is something like this: ‘At first blush, it might seem that there are two ways in which the necessary objective validity and strict universality of certain concepts can be explained, but upon reflection it turns out that the first applies only to the contingent agreement of empirical concepts with their particular objects, while the second (the Way of Intellectual Intuition) is irrelevant to the concepts of our human understanding which, though more than ectypal, is still not archetypal.’ Following the lead of the Prolegomena, one can paraphrase both passages from the Transcendental Deduction in like manner, merely substituting a different ‘while’-clause at the end: ‘while the second (now the Copernican Way of Transcendental Idealism) is in fact the sole possible way in which this task can be accomplished.’ The phrase ‘the first serves to explain only the contingent agreement of empirical concepts with their particular objects’ points to the real dilemma that Way One represents for Kant: we must either sacrifice the ‘dignity’ of the pure concepts of the understanding, thereby simply dispensing ourselves from the task of explaining their necessary (and strictly universal) objective validity, or we must explain objective necessity and strict universality in terms of the affection of the senses and the logical operations of the understanding, which is self-contradictory. On such a reading, Kant only seems to take for granted what his empiricist opponents deny, namely that universally and necessarily objectively valid knowledge through a priori concepts (pure philosophy or metaphysics as a science) is indeed possible, the sole question being how; for he goes on to assume the burden of proof fully, demonstrating that metaphysics is possible precisely by showing in detail just how it is possible. Hence even the charge of begging the question falls away when we read his other statements in the light of the Prolegomena passage.
The foregoing may suffice to show that Kant retained at least the dilemma form of the letter of 1772, repeatedly reverting to it even though (1) one of the horns had been tacitly replaced by a viable alternative, so that (2) the choice was no longer strictly a dilemma at all, and despite the fact that (3) retaining it creates at least the appearance of falsehood, self-contradiction, and petitio principii. All this surely bespeaks a deep attachment to the talk of “two ways,” despite having hit upon a better way of presenting his core problem in the Prolegomena and the Second Edition Introduction to the Critique. If one were to hazard a guess as to why Kant clings so to the dilemma form, one might plausibly conjecture that the following four-part analogy seemed particularly instructive to him: as (a) Way One is to (b) empirical knowledge, so (c) the Copernican Way of Transcendental Idealism is to (d) synthetic a priori knowledge. In the Conclusion, there will be occasion to examine another four-part analogy that almost certainly exercised a great hold on Kant’s mind.
The other question posed earlier was how causal and representational realism à la Locke can account at least for contingent agreement between objects and the empirical representations derived from them and made general by logical reflection. Even before Locke, Descartes (to say nothing of the Greek sceptics) advanced ample grounds for doubt about such agreement. On this further question, neither the Critique nor the Prolegomena sheds any new light. “Many empirical concepts are employed without question from anyone,” writes Kant in a typical passage from the Critique. “We make use of a multitude of empirical concepts without objection from anyone, and take ourselves to be justified in granting them a sense and a supposed signification even without any deduction, because we always have experience ready at hand to prove their objective reality” (A84/B117). This leaves one none the wiser as to how Kant would have responded to the well-known sceptical jibes against such naive causal and representative realism. Footnote 8 In the end, it is difficult to say which is more surprising: Kant’s repeated assertions that there are two possible ways, when Way One, on his own showing, is impossible; or his blithe assumption that Way One provides a satisfactory account of the contingent agreement of empirical concepts with their objects. Perhaps both are owing to the same deep attachment to the four-part analogy described above.
II. The Way of Intellectual Intuition or Intuitive Understanding
Returning now to the original Second Way, it will be helpful to sketch rapidly the two philosophical creation stories that form its remoter (medieval) and nearer (early modern) historical background. While he clearly drew on both, it is perhaps the Cartesian creation story, anchored in the Augustinian tradition, that represents Kant’s dominant conception of the divine intellect, that evoked in the letter to Herz. Onto it he grafted elements of the Thomistic concept, stemming from Greek sources, so that the term intuitus originarius is frequently supplanted by intellectus archetypus in his writings, notably in the letter to Herz. This terminological variation reflects Kant’s fundamental (i) epistemological distinction between intuition and understanding; but it reflects equally the (ii) essence-existence dichotomy of medieval metaphysics, and the (iii) theological distinction between God’s intellect, or exemplar causality, as certain Scholastics and neo-Scholastics (not to mention Leibniz) called it, and God’s will or efficient causality, the divine fiat. Depending on whether he has in mind (1) our finite, sensible faculty of intuiting actually existing things or appearances or, alternatively, (2) our human incapacity to think the natures of such things otherwise than discursively, that is, through general concepts, and hence in abstracto, Kant selects as a foil to our human way of knowing either (a) the efficient causality of a divine faculty of intuition “through whose representations the objects of the representation should at the same time exist” (B138), or (b) the exemplar causality of an infinite understanding which grasps the natures of all finite things right down to the last contingent detail—their conceptus singulares, as Kant says (cf. 2:396 and 397), or their complete concepts, in Leibniz’s terminology—in a reflexive act of intuiting its own nature which contains the essences of all things, not just formally but eminently. Footnote 9 Unfortunately, he sows confusion by using intellectus archetypus for both; he would have done better to make a consistent use of the freshly coined dichotomy intuitus originarius-intuitus derivatus (cf. B72) for his dominant Augustino-Cartesian conception of a divine intuition that confers existence on its objects, reserving the intellectus archetypus-intellectus ectypus dichotomy for the Scholastic-Thomistic concept of a divine intellect that knows the natures of all things by knowing itself (cf. A695/B723, 5:408). Footnote 10
According to what may be termed the orthodox Christian philosophical creation story, hailing from St. Thomas, God’s intellect or understanding first conceives the natures of all possible things (possibilia) by reflecting on his own infinite nature which, as absolute perfection, contains them all within itself in the manner of an imitable supreme good of which an infinite number of finite natures fall more or less short. By addressing the possibilia as ideae in the divine mind, Aquinas means to convey that all such finite natures are conceived in the very same reflexive act by which God understands his own infinite nature; that is the root meaning of idea in St. Thomas: a representation of an object in which the subject is concomitantly aware of itself. Footnote 11 Next, the divine will selects from among the imperfect natures existing from eternity in the divine intellect those that are to exist outside it in space and time. Initially used interchangeably with idea, the word exemplar comes to be employed more narrowly by St. Thomas for those possibles which have been, are, or will be made actual by the divine will. Finally, God exercises his creative power or efficient causality in order to bring into existence just those entities that his infinite intellect conceives as possible and his perfectly good will selects as most worthy of being made actual; and he continues to exercise his power in order to maintain them in existence from moment to moment for as long as he has decreed they shall exist, without thereby exhausting his infinite power. Since we understand that he can always cause more things to exist than he ever has or ever will make actual, God’s intellect, will, and power are three distinct faculties for us, in our manner of conceiving them, although in God himself they are one and identical. According to the doctrine of divine simplicity, God does not just have intellect, will, and power in the way in which certain of his creatures possess these attributes, any more than he merely has existence or being; rather, God is his intellect, will, and power, just as he is being or existence itself. And things identical with the same thing are identical with each other, though from our limited, human perspective, God’s intellect, will, and power all differ.
Now talk of an intuitus originarius, first introduced in the second-edition Transcendental Aesthetic (B72), evokes not so much this Scholastic creation story, as a competing, unorthodox notion of God’s efficient causality by means of a creative act of intellectual intuition. To the extent that the divine understanding brings its various objects into existence in the very act of conceiving/intuiting their natures, intellect and will are simply one and the same in God himself, though not in our conception of them, owing to the distorting influence of our immediate awareness of their real distinctness in us. Footnote 12 This particular account of divine simplicity, that is, of the identity of intellect and will in God, has its nearer origins in Descartes’s theological voluntarism, the doctrine of divine creation of the eternal truths, as expounded (for reasons of prudence) mainly in his private correspondence; Spinoza boldly extended it to include God’s power in a context that was no longer that of a creation story at all (cf. IEP17S[II] and IIEP7C). Kant evokes this account of a divine intellect-cum-will, not just under the title intuitus originarius, but also under other names and descriptions, including (regrettably) intellectus archetypus, the term used in the letter to Herz, which is better suited as a designation for God’s reflexive grasp of the exemplars of created things (as distinct from the ideas of never-to-be-created creatables) existing from eternity in the divine mind. The upshot is a mixed or hybrid conception of the divine intellect, in which there is a certain ascendancy of the Augustinian doctrine from which those of Descartes and Spinoza derive.
Whatever may have been the nearer, neo-Augustinian sources on which the early moderns drew, the remote inspiration comes from certain well-known passages of St. Augustine’s own writings. In Confessions, VII, 4, for example, God is addressed in these terms: “Your will is no greater than your power … For the will and power of God are God himself. And no nature is or exists unless you know it.” The concluding phrase anticipates the famous doctrine of XIII, 38: “We see the things that you have made because they are; but as for you, because you see them, they are.” In Summa Theologica I, 14, 8, Aquinas seems at first to side with Augustine on the question that gives the article its title, “Whether God’s Knowledge is the Cause of Things.” After citing three objections to an affirmative answer, Thomas immediately counters (Sed contra) with a quotation from Augustine’s De Trinitate, XV: “Not because they are, does God know all creatures spiritual and temporal, but because he knows them, therefore they are.” And in the corpus articuli he makes an initial determination apparently in Augustine’s favour (“I answer that the knowledge of God is the cause of things …”). Finally, in the reply to the third of the three initial objections (viz. “The thing known is prior to knowledge and is its measure”), Thomas appears to concede at least something to the wording of the Confessions and De Trinitate when he writes: “we receive knowledge from natural things, of which God is the cause by his knowledge. Hence, as the natural objects of knowledge are prior to our knowledge, and are its measure, so the knowledge of God is prior to natural things, and is the measure of them.”
But what Aquinas appears to grant Augustine in these passages, the full corpus articuli takes back by placing the emphasis squarely upon the divine will. The opening clause of the key passage again looks like a conciliatory nod in Augustine’s direction: “Now it is manifest that God causes things by his intellect, since his being is his act of understanding”; but Aquinas continues: “and hence his knowledge must be the cause of things in so far as His will is joined to it. Hence the knowledge of God as the cause of things is usually called the knowledge of approbation (scientia approbationis)” (ibid., e.a.). Already in De Veritate (cf. 3, 3 and 3, 6), Thomas had argued that God’s ideas of actually created or to-be-created things—exemplars in the narrow sense indicated above—have a certain determination that the mere ideas of never-to-be-created creatables lack: God’s considers them with respect to the conditions of their actual existence, such “virtually practical knowledge” (cognitio practica virtute, non actu) being roughly analogous to that of a craftsman who, considering the form of an artefact, merely envisages what is necessary to make it actual without intending to do so. The “knowledge of approbation” mentioned in the Summa Theologica passage refers to God’s will insofar as he wills himself to have this sort of knowledge of what is involved in the production of things; it must be carefully distinguished from that further stage of God’s willing which is analogous to the “actually practical knowledge” (actu practica cognitio) of the artisan who proceeds to fashion an artefact, and which is alone a causal or creative willing, a divine choice or fiat through which the object exists.
So much for the philosophical creation stories that form the historical backcloth of Kant’s various descriptions of the divine mind. Footnote 13 Apart from B72 of the Transcendental Aesthetic, where the terms intuitus originarius and intuitus derivatus are first introduced, the Critique of Pure Reason contains relatively few straightforward references to an intellect that brings its objects into existence in the very act of intuiting them. There are, however, two unambiguous passages in the B-edition of the Transcendental Deduction. One is that quoted earlier from B138 (“through whose representations the objects of the representation should at the same time exist” e.a.); the other is this striking passage:
For if I wanted to think of an understanding that itself intuited (as, say, a divine understanding, which would not represent given objects, but through whose representation the objects would themselves at the same time be given, or produced), then the categories would have no significance at all with regard to such a cognition. (B145)
Of the two, only the Scholastic-Thomistic notion of an intellect whose contents are archetypes or exemplars is clearly discernible in the Metaphysica of Baumgarten, whose treatment of the topics Intellectus Dei (§§ 863-889) and Voluntas Dei (§§ 890-925) is typical of the German Scholastisticism of the 18th century. Leibniz himself at least alludes to both at certain places in the correspondence with Clarke, distinguishing expressis verbis between God’s exemplar and efficient causality (cf. Leibniz’s Fifth Letter, paragraph 87; G VII: 411). Berkeley, for his part, develops an idiosyncratic version of both conceptions of the divine intellect in the Three Dialogues between Hylas and Philonous, a work which we know Kant knew (in Latin translation). Footnote 14 “Do I not acknowledge a two-fold state of things, the one ectypal or natural, the other archetypal and eternal?” says Berkeley’s spokesman, Philonous. “The former was created in time, the latter existed from everlasting in the mind of God.” Footnote 15 According to the creation story being evoked here, “[a]ll objects are eternally known by God, or, which is the same thing, have an eternal existence in His mind; but when things, before imperceptible to creatures, are, by decree of God, made perceptible to them, then are they said to begin a relative existence with respect to created minds” (ibid., 252). In this immaterialist version of God’s exemplar and efficient causality, the creation of an object or thing in space and time is just the decision on God’s part to make it perceptible to us here and now, it having existed in, and been perceived by, the divine mind from eternity. It seems likely that Kant’s manner of thinking about Way Two was decisively shaped by Berkeley who, like Kant in the letter to Herz, speaks of an intellectus archetypus, but understands by it what Kant later came to designate at times more aptly as an intuitus originarius, at times (following Berkeley) as an intellectus archetypus. But this, again, belongs to the realm of historical conjecture. Footnote 16
III. The Platonic Way of Hyper-Physical Influx
According to Calvin Normore Footnote 17 , it was Philo of Alexandria who first removed the Forms or Ideas from the Platonic Heaven and placed them in the mind of God; instead of looking away (apoblepein) to a world of independently existing Forms as models for the transformation of a pre-existing matter, the Platonic-demiurge-become-Christian-creator-God looks back (reflectere), or looks within, at the Ideas which constitute his own infinite intellect, thus intuiting the models or paradigms for his creation of the world ex nihilo. When Kant speaks of Plato in the context of the five ways, it is probably this Neo-Platonic and Christianized Platonism of his 17th-century predecessors and certain contemporaries that is uppermost in his mind. Thus, he mistakenly ascribes to Plato himself the view
that we men possess intuitions a priori, which would, however, have their first origin, not in our understanding (for the latter is not a faculty of intuition, but only a discursive or thinking faculty), but rather in one [i.e., in an understanding other than our own] that was simultaneously the ultimate ground of all things, i.e., the divine understanding, whose intuitions direct would then deserve to be called archetypes (Ideas). (8:391)
The passage continues:
But our intuiting of these divine Ideas … would to us have been given only indirectly, at our birth, as an intuiting of copies (ectypa), as it were shadow-images of all things … though that birth has simultaneously brought with it a darkening of these Ideas, through forgetfulness of their origin, as a consequence of the fact that our mind (now called soul) has been thrust into a body, from whose fetters it would now have to be the noble task of philosophy to release us. (ibid.)
In the letter to Herz, Kant attributes an even more radical version of the same doctrine to Malebranche, whom he surprisingly presents as a disciple of Plato rather than the historically nearer Descartes or even Augustine. The whole passage, which describes Kant’s Third and Fourth Ways, is worth quoting again:
Plato assumed a previous intuition (ehemaliges Anschauen) of divinity as the primary source of the pure concepts of the understanding and of first principles. Mallebranche [sic] believed in a still-continuing perennial intuition (noch dauerndes immerwährendes Anschauen) of this primary being [or divinity] … Crusius believed in certain implanted rules for the purpose of forming judgments and ready-made concepts that God implanted in the human soul just as they had to be in order to harmonize with all things. Of these systems, one might call the former [two, i.e., those of Plato and Malebranche] the Hyperphysical Influx Theory [influxum hyperphysicum] and the latter [i.e., the system of Crusius] the Pre-established Intellectual Harmony Theory [harmoniam praestabilitam intellectualem]. (10:131)
From this, together with another, related passage, to be quoted presently, it is reasonably clear how the Third or Platonic Way is to be understood. Formerly, in its pre-natal existence in the immediate presence of the divine Forms, the human mind, itself then still a divinity, possessed a faculty of direct intellectual apprehension of the divine archetypes or Forms; this it lost upon its incarnation as a soul in a human body, so that now, in this world, all that it can do is “remember” (see next quotation) or as Kant puts it elsewhere (see the long passage quoted below), “unfold” the traces or copies (ectypa) left in the embodied soul by its former intellectual intuitions of the eternal archetypes. Kant has no sympathy whatever with this very strong rationalism, which he disparagingly calls “enthusiasm”: “The principle of Plato, namely, that by virtue of their previously possessed faculty of an intuitive understanding, human beings would now still have the power to remember by their understanding back to previously held concepts, rests clearly on a mistake …” (29:954, e.a.). Thus far, it might seem that Kant’s interpretation has at least struck a genuinely Platonic chord. That impression is dispelled by the other passage mentioned above, which reveals what he understands by an “intuitive understanding”: not the discarnate soul’s direct apprehension of Ideas or Forms existing ‘in their own right’—genuinely Platonic universales ante res such as those to which the demiurge of the Timaeus ‘looks away’—but rather archetypes existing in the infinite mind of God, as in Christian Platonism:
[Plato] believed that all a priori cognitions are cognitions of things in themselves, and because we participate in the former [viz., a priori cognitions], we also participate in the latter [viz., cognitions of things in themselves], and among those he included mathematics. But we could not participate in those [a priori cognitions of pure mathematical objects] on our own [or ‘directly,’ as Kant puts it in the passage cited at the outset of this section], consequently only [indirectly] through the communication of divine ideas. But since we are not conscious of them as having been imparted and transmitted merely historically [i.e., as something only learned through teaching], but rather as being immediately understood [i.e., as objectively necessary truths], they cannot be inborn concepts that are believed [as a matter of subjective necessity], but immediate intuitions that we have of the archetypes in the divine understanding. But we can unfold these only with difficulty. Thus they are mere recollections of old ideas from communion with God. (18:434-435, e.a.)
Note the way in which Kant’s presentation of the Third Way simply substitutes “the communication of divine ideas” or the “communion with God” for the pre-natal existence of the soul in the realm of the independently existing Platonic Forms. This obviously has little to do with the historical Plato or, for that matter, with Leibniz who, Kant claims, “seems, with Plato, to attribute to the human mind an original, though by now dim, intellectual intuition of these supersensible beings” (8:248). But, if Leibniz followed, Malebranche went decidedly beyond Plato, in Kant’s estimation, treating the human intellect or soul as still even now possessed of a faculty of intellectual intuition of the divine archetypes. It can only be Malebranche’s famous dictum that “we can see all things in God” that provides the—on the whole, rather slender—basis for this transmutation of Descartes’s disciple into a follower of Plato in Kant’s description of the Third Way. Footnote 18
The solution that the Platonic-Malebranchian Way of “Hyperphysical Influx” provides to the problem of the agreement between “synthetic representations” and their objects can be summed up this way: just as the divine intellectus archetypus grasps the essences or natures of all things reflexively, by turning its reflective gaze back upon the Forms that exist in it from eternity, so (mutatis mutandis) the intellectual faculty of intuition that Plato and Malebranche impute to the human soul (the latter, as actual even now) permits an intuition of the archetypal Forms of all things in God’s intellect itself. Nevertheless, there is this significant difference between Plato and Malebranche:
Plato, through no fault of his own (for he used his intellectual intuitions only backwards, to explain the possibility of synthetic knowledge a priori [in mathematics], not forwards, to extend it through those Ideas that were legible in the divine understanding), became the father of all enthusiasm. (8:398)
It was Malebranche who, according to Kant, first used intellectual intuitions “forwards,” thus inaugurating “the second stage” of “enthusiasm” or “mysticism”:
The origin of all philosophical enthusiasm lies in Plato’s original divine intuitions of all possible objects, i.e., in the ideas … Now on this is grounded, first, Plato’s opinion that all of our a priori cognition (mathematics), especially that of perfections, stems from the recollection of these prior intuitions and that we must now only seek to unfold them ever more; from this, however, arises the second stage of mysticism, that of even now intuiting everything in God, which then makes all research into synthetic a priori cognition unnecessary, insofar as we read it in God.” (18:437, e.a.)
The second is, however, not the “highest degree” of enthusiasm; that dubious honour belongs rather to Spinozism:
The highest degree of enthusiasm is that we are ourselves in God [i.e., modes of the divine substance] and feel or intuit our existence in Him. The second: that we intuit all things in accordance with their true nature only in God as their cause and in his ideas as archetypes.” (18:437)
Although neither Spinoza nor Malebranche is mentioned here by name, the talk of ‘being in God’ and of ‘intuiting everything in God’ leaves little doubt about whom Kant has in mind. In On a Superior Tone, Kant surmises that what led Plato to “put the torch to enthusiasm” was his inability to conceive that a priori knowledge might yet be sensible, not intellectual:
Could he [Plato] have guessed … that there are indeed intuitions a priori, but not of the human understanding, since (under the name of space and time) they are actually sensuous … he would not then have looked for pure intuition … in the divine understanding and its archetypes of all things, as independent objects; or thereby put the torch to enthusiasm. (8:391n.)
This is no mere obiter dictum. The Prolegomena (cf. 4:375n.) allude, in a similar vein, to “visionary idealism … which (as was already to be seen with Plato) always inferred, from our cognitions a priori (even those of geometry) to another sort of intuition (namely intellectual) than that of the senses …” And that, in Kant’s eyes, is the end of all sound philosophizing. Only Way Four, to which we turn next, poses an even greater threat (leaving aside, for the moment, Spinozism).
IV. The Way of Pre-established Harmony or Rational Pre-formation
In the letter to Herz, the Third is clearly distinguished from a Fourth Way, the “Pre-established Intellectual Harmony Theory (influxum hyperphysicum)” (10:131) ascribed to Crusius, who “believed in certain implanted rules for the purpose of forming judgments and ready-made concepts that God implanted in the human soul just as they had to be in order to harmonize with things” (ibid.). In the Prolegomena and the Critique, by contrast, there is but a single “middle way (Mittelweg),” as Kant calls it (cf. 4:319n. and B167), which in fact coincides with Way Four alone and is again expressly associated (in the Prolegomena at least) with the name of Crusius:
Crusius alone knew of a middle way, namely that a spirit who can neither err nor deceive originally implanted these natural laws in us. But since false principles are often mixed in as well—of which this man’s system itself provides not a few examples—then, with lack of sure criteria for distinguishing an authentic origin from a spurious one, the use of such a principle looks very precarious, since one can never know for sure what the spirit of truth or the father of lies may have put into us. (4:319n)
At the very end of the Transcendental Deduction, Kant offers a similar description of this “middle way,”
namely, that the categories were neither self-thought a priori first principles of our cognition nor drawn from experience, but were rather subjective predispositions for thinking, implanted in us along with our existence by our author in such a way that their use would agree exactly with the laws of nature along which experience runs (a kind of preformation-system of pure reason). (B167)
This is just “the Pre-established Intellectual Harmony Theory” fittingly renamed in terms less strongly reminiscent of Leibniz. Here, however, Kant raises a “decisive” (ibid.) objection against this middle way:
in such a case the categories would lack the necessity that is essential to their concept. For, e.g., the concept of cause, which asserts the necessity of a consequent under a presupposed condition, would be false if it rested only on a subjective necessity, arbitrarily implanted in us, of combining certain empirical representations according to such a rule of relation. I would not be able to say that the effect is combined with the cause in the object (i.e. necessarily), but only that I am so constituted that I cannot think of this representation otherwise than as so connected; which is precisely what the skeptic wishes most …. (B168)
The sceptic whom Kant has in mind here is, of course, Hume, for whom the relation of cause and effect reflects a merely subjective or psychological necessity based on the association of ideas. Not surprisingly, this objection to the Fourth Way sounds very much like that raised against Way One, which was after all Hume’s as well as Locke’s way: both the First and Fourth Ways explain a necessary relation in a manner that entails its contingency, which is self-contradictory. It is an objection to which Kant believed his own doctrine of the transcendental subjectivity of the categories immune; the latter provides a correct account of what the talk of natural or objective necessity truly means, or rather of the only thing it can mean if the cavils of the sceptics regarding de re necessity are to be avoided.
This objection to the Fourth or “middle way” is new vis-à-vis that raised in the latter part of the letter to Herz. In 1772 Kant had objected only that
the deus ex machina is the greatest absurdity one could hit upon in the determination of the origin and validity of our cognitions. It has—besides its vicious circularity in drawing conclusions concerning our cognitions—also this additional disadvantage: it encourages all sorts of wild notions and every pious and speculative brainstorm. (10:131)
In other words, invoke God and a priori principles divinely implanted in the human mind at the moment of creation, and the divine origin of those principles must be proved using those very principles, which is manifestly circular; moreover, the gates of the citadel of reason are thrown open to arbitrariness and dogmatism, since anything of the truth of which one is blindly convinced may be claimed to be known with perfect certainty thanks to divine illumination. As Kant reformulates this latter objection in the Critique, again singling out the Fourth Way: “on such an hypothesis we can set no limit to the assumption of predetermined dispositions to future judgments” (B167). In the Metaphysical Foundations of Natural Science, he describes this general type of solution to the problem of the agreement between representations and things as “much worse than the evil it is supposed to cure” (4:476). And so it is, since it manifestly comes down to explaining that which we understand only very imperfectly (the grounds of the necessary relation between “intellectual representations” and their objects) by means of something that we understand not at all (God’s will and purposes): obscurum per obscurius, to give the fallacy its Latin name.
To conclude this discussion of the Fourth Way, it is worth remarking that Kant probably regarded Leibniz’s system of pre-established harmony of the body (and, via the body, of the whole material universe) with the human soul as a variant of the Cartesian solution he attributed to Crusius, the agreement between representations and things being, for Leibniz, decreed by God from eternity rather than instituted from moment to moment, as in Malebranchian Occasionalism, or at the moment of creation of each human soul, as in the case both of Descartes’s Divine Guarantor and of the theory ascribed to Crusius. As for the other solution that commonly figures in lists of the different ways of guaranteeing agreement between ideas and things, the Spinozan mind-body parallelism and mind-body identity theses do not figure in Kant’s tally at all. The reason may be that neither Leibniz nor Spinoza is uniquely concerned with the necessary agreement of a priori concepts with their objects, which was Kant’s exclusive concern already in 1772.
V. Conclusion
Some 10 years prior to the first Critique, then, Kant can only think of four ways in which to explain the conformity of “intellectual representations” to their objects, none of which he deems at all satisfactory. Unable to envisage a viable solution, he merely congratulates himself on having discovered, and succeeded in formulating in all its generality, the problem on which the future of metaphysics depends. The situation is very different 10 years after publication of the work, when Kant adverts to his own solution in a letter of 20 January 1792 to another former pupil and friend, J. S. Beck: “You put the matter very well when you say: ‘the whole complex (Inbegriff) of representations is itself the object of the mind, and the action of the mind, whereby the whole complex of representations is represented, means: relating them to an object’” (11:314, e.a.). Here the relation of representations to an object is described as the ‘handiwork’ of the mind (Handlung des Gemüts); the object, moreover, is just an Inbegriff of representations. The striking thing about this formulation, apart from the tacit substitution of appearances for things in themselves, is its deliberate reversal of the common sense realism of Way One: the mind makes the object possible by objectifying its representations. Even if the talk is no longer of “two ways,” there is a discernible continuity in Kant’s manner of conceiving the problem between 1772 and 1792.
Does the “Main Transcendental Question” first formulated in the Prolegomena and taken over in the second edition of the Critique simply supersede the consideration of multiple “ways”? Not if the many passages cited above are any indication. That is the reply to the second of the two objections adduced earlier against “Kant’s ‘Five Ways’”: given that it was only modified, never abandoned, the straightforward, completely non-technical way in which Kant persistently formulated his problem, even after having hit upon, and worked out in detail, the “Main Transcendental Question,” deserves at least the degree of attention accorded those doctrines of pre-Critical works that survive relatively intact in his mature philosophy. Granted that the later question—How are synthetic judgements a priori possible?—is the most promising point of entry into Kant’s transcendental idealism, consideration of the other four ways sheds an interesting light on how Kant himself viewed his achievement in the context of the history of metaphysics.
As for the first objection to the title, though they have indeed nothing to do with the existence of God, the four ways considered here do have an indirect bearing on the solutions to the mind-body problem evoked in Leibniz’s equally famous Three Ways (of Influence, of Assistance, and of Concomitance or Pre-established Harmony), from which two of the names are taken over (cf. G IV:498f.). To dismiss the talk of ‘ways’ as a primitive, later superseded manner of formulating the central problem of the Critique is therefore not just to neglect Kant’s deep and lasting attachment to this manner of posing his problem; it is also opens a greater gulf than actually exists between Kant’s “problem of pure reason” (B19) and those 17th-century philosophical postures to which Leibniz adverts. Even if Kant’s sporadic talk of “ways” pales to insignificance beside the quinque viae of St. Thomas, the historical precedent set by Leibniz is reason enough to regard it as much more than just a façon de parler.
It remains, in closing, to cast a rapid glance at a leading contemporary interpretation of Kant’s transcendental idealism which, for all its merits, misleads owing to a faulty (even incoherent) conception of divine cognition. Invoking the sound exegetical principle that “often the best way to understand a philosophical doctrine is to see what it denies,” Henry Allison highlights three inter-connected doctrines implicit in Kant’s “rather cryptic characterizations” of his “meta-philosophical or meta-epistemological” arch-enemy, transcendental realism. Footnote 19 First, transcendental realism adopts (a) an anti-Copernican model on which human cognition conforms to its objects, not vice versa: “what makes transcendental realism a form of realism is that, implicitly at least, it regards the conditions of human cognition as determined by the nature of a pre-given reality.” Footnote 20 Hence not only empiricists, like Locke, but even rationalists who posit a human faculty of intellectual intuition “of a pre-given reality,” like Spinoza, for example, or Leibniz, as Kant understands him (see above), subscribe to an anti-Copernican model. Footnote 21 Behind this almost universal adherence to some form of Way One (broadly understood as any view which posits a pre-given reality as the object of cognition) lies (b) a “theocentric paradigm or model” of human cognition, on which human knowledge conforms to its objects precisely by conforming to that paradigm of perfect knowledge which is divine cognition; as with divine cognition, then, the objects to which it conforms are the things as they are in themselves. Footnote 22 Implicit in (a) and (b) is the third characteristic of transcendental realism: (c) conflation of the transcendental distinction between appearances and things in themselves. In his ensuing point-by-point contrast with transcendental idealism, Allison stresses that Kant’s (a’) Copernicanism and (c’) transcendental distinction both spring from (b’) his abandonment of the traditional theocentric in favour of an anthropocentric model of cognition. Thus, notwithstanding the fact that Kant himself invariably stresses (c’), Allison gives special weight to (b’), something Kant never mentions at all.
Without wishing to question the idea of an historically dominant theocentric model of cognition, one may legitimately wonder how it can also be an anti-Copernican model, when the ‘way’ of divine cognition is precisely not the anti-Copernican Way One, but the (so to speak) ultra-Copernican Way Two. As for Kant’s having been the first to break with the theocentric paradigm, thus introducing a completely new meta-epistemological standpoint, things can hardly be so simple if the theocentric paradigm is precisely ‘Copernican.’
It is easy to see why Allison does not hesitate to conjoin anti-Copernicanism and theocentrism in his account of transcendental realism: regarding “the conditions of human cognition as determined by the nature of a pre-given reality” is, for him, “equivalent to assuming that they reflect the ideal model of God’s way of knowing.” Footnote 23 The implication of this is that God’s way of knowing is itself “determined by a pre-given reality.” Since Allison is well aware that God’s intuitus orginiarius or intellectus archetypus is conceived, not as conforming to a pre-given object, but as bringing its object into existence, there is a certain incoherence in his account of transcendental realism.
However, the anti-Copernican and theocentric character of transcendental realism can be reconciled in another way. Instead of saying, with Allison, that (1) human knowledge conforms imperfectly to things in themselves (Way One, broadly understood) insofar as it approximates to divine knowledge which conforms perfectly to things in themselves (Way One again), all one need do is say that (2) human knowledge conforms imperfectly to things in themselves (Way One) insofar as it approximates to divine knowledge to which the things themselves conform perfectly (Way Two). With (2), the “structural similarity” (Allison’s term) between human and divine cognition is lost, but the anti-Copernicanism of transcendental realism is preserved along with the ultra-Copernicanism of divine cognition without sacrificing the historically important idea of an assimilation of human to divine knowing. Footnote 24
As for Kant’s alleged anthropocentrism, the problem here is that the structural similarity that disappears on substituting (2) for (1) reappears in the account of synthetic a priori knowledge of transcendental idealism: (3) objects as appearances (the phenomenal world) must conform to our synthetic a priori intuitions, concepts, and principles (Way Five, or the new Way Two) insofar as this sort of objectifying representation conforms to the model of divine intuitus originarius, i.e., a cognition to which its objects (things in themselves, the noumenal world) must conform since it brings them into existence (the original Way Two). Kant’s Copernican reversal and transcendental distinction, far from abandoning the model of divine cognition, adapt it to our human, finite, essentially discursive way of knowing things as appearances. Why else does Way Five, as the new Way Two, come to occupy the place of the Second Way of the letter to Herz, if not because it is modelled on it in the manner made explicit in (3)? The ancient idea of an homoiōsis theō finds new expression in another four-part analogy, the last two parts of which are just “finitized” versions of the first two: as (i) God’s creative intellectual intuition (the original Way Two) is to (ii) things in themselves (which must conform to it since it bestows noumenal existence on them), so (iii) synthetic a priori human knowledge (the new Way Two) is to (iv) appearances which must conform to it to the extent that it bestows empirical reality or phenomenal existence on them. Footnote 25 If this is indeed Kant’s model of synthetic a priori cognition, then the Copernican Way, while no doubt anthropocentric in some respects, is anything but a decisive break with the dominant theocentric model of cognition. Footnote 26 Kant can insist all he likes that we humans possess no faculty of intellectual intuition, that even our synthetic a priori knowledge does not create its object ex nihilo, but must receive the material for its discursive objectifying operations from the faculty of sensibility; all this is so far from signifying the end of theocentrism that it rests squarely on the novel idea of a four-part structural analogy between human and divine cognition. If correct, this only illustrates further the value of considering transcendental idealism in the wider historical context examined here.
