Kant on Transcendental Illusion and the Argument from Spinozism

2.1 “Spinozism” in the CPrR Passage

The Critique of Practical Reason (henceforth CPrR) passage points to three distinguishable theses that jointly constitute the ‘Spinozism’ to which the transcendental realist view Kant is targeting is supposed to lead, given the various additional assumptions. In what follows, I sometimes use ‘the transcendental realist view’ as shorthand for the conjunction of transcendental realism and the various further assumption about God, space, and so forth, that he takes to entail Spinozism.

(1) Essential Determinations: Space and time are God’s “essential determinations.”

Hence, if this ideality of time and space is not adopted, nothing remains but Spinozism, in which space and time are essential determinations of the original being itself. (KpV, 5:101–102; underlining mine; cf. V-Met-K2/Heinze, 28:803; V-Met-K3E/Arnoldt, 29:977; V-Met-L2/Pölitz, 28:567; V-Met-K3E/Arnoldt, 29:1008–1009; R 6317, 18:626).

To say that space and time are God’s essential determinations means that they are not really distinct from God but rather that they exist as essential attributes or properties of God. At least in one place, Kant formulates this as the claim that “space is the divinity” (V-Met-L2/Pölitz, 28:567).

(2) Monism: God is the only substance and all other putative substances are merely accidents of God.

while the things dependent upon [God] (ourselves, therefore, included) are not substances but merely accidents inhering in it (KpV, 5:102; V-Met/Dohna, 28:666; V-Met-L2/Pölitz, 28:567).

(3) Theological Determinism:^[8] God causally determines the actions of putative substances (including ourselves) in a way that precludes their being free and that makes God the author of their actions.

as soon as one admits that God as universal original being is the cause also of the existence of substance […] one must admit that a human being’s actions have their determining ground in something altogether beyond his control, namely in the causality of a supreme being which is distinct from him and upon which his own existence and the entire determination of his causality absolutely depend […] Therefore I do not see how those who insist on regarding time and space as determinations belonging to the existence of things in themselves would avoid fatalism of actions. (KpV, 5:101; underlining mine, bold Kant’s).

since the creator of substance would also be the author of the entire mechanism in this substance (KpV, 5:102).

for if these things exist merely as its effects in time, which would be the condition of their existence itself, then the actions of these beings would have to be merely its [that is, God’s] actions that it performs in any place and at any time (KpV, 5:101–102; underlining mine, bold Kant’s).

It is not immediately obvious what kind of freedom is at issue in Theological Determinism. Clearly, Kant thinks that transcendental freedom (a species of libertarian free will) is precluded by the view. What I am calling Theological Determinism is arguably meant to rule out not just transcendental freedom but also the kind of compatibilist view accepted by Leibniz, Wolff, and the young Kant in the New Elucidation (PND, 1:400); this position says, roughly, that one’s actions/choices can be deterministically caused and yet free provided they are internally generated. The Argument from Spinozism in CPrR occurs just a few pages after Kant’s famous dismissal of compatibilist (“comparative”) freedom as the “freedom of a turnspit” (KpV, 5:96–97). Significantly, in that discussion Kant explicitly put aside the issue of divine creation, and with it the puzzle of how to reconcile compatibilist freedom with the view that the causal chain containing our actions leads outside of us, back to God. With the Argument from Spinozism, Kant can be seen as bringing the previously bracketed question of the divine creation of putative substances to the fore, and showing how given this claim about God along with other assumptions, even compatibilist freedom, and the compatibilist understanding of what it takes for me to be the “author” of a given action, is ruled out (KpV, 5:102). I return to this interpretive issue later.

These three Spinozistic theses have obvious similarities to claims of the historical Spinoza. Spinoza famously endorses Monism; he also claims that extension is an essential attribute of God.^[9] Finally, Spinoza holds that God is the cause of the existence of everything else, and that God’s nature is the ultimate determining cause of everything that happens. Whether or not Spinoza takes this determinism to rule out all kinds of freedom for humans is up for debate. Certainly he rules out anything like Kant’s transcendental freedom. In any case, during Kant’s time Spinoza was widely associated with fatalism. I am not suggesting that we see Kant’s three Spinozistic claims as in all cases exactly like those meant by the historical Spinoza, just that from Kant’s standpoint it is reasonable for him to regard them as Spinozistic.

We find Kant rehearsing versions of the Argument from Spinozism in various unpublished texts, including transcripts of lectures held after the publication of CPrR. However, as far as I can see, in these other texts we find Kant claiming that transcendental realism leads to Essential Determinations and Monism – without discussing Theological Determinism. For this reason, I hold that the CPrR offers an expanded version of the Argument from Spinozism wherein the transcendental realist view is claimed to lead to all three Spinozistic theses, while the unpublished writings (like transcripts of metaphysics lectures) offer a condensed version showing that the transcendental realist view leads to (just) Essential Determinations and Monism.^[10]

2.2 What’s the Problem with Spinozism?

Assuming I am right about what ‘Spinozism’ for Kant means in this context, what is supposed to be so bad about it? Commentators are not all on the same page here. Paul Franks (2005, 123) thinks that Kant is trying to argue in CPrR that Spinozism is “antinomial,” presumably meaning that it is an internally inconsistent position. I do not see much evidence for this reading. The three theses I have described above are not in any obvious conflict. Moreover, in the CPrR passage, Kant actually emphasizes the “consistency” of Spinozism vis-à-vis what he calls the “creation theory” (by which he means views which take God to be the creator of an ontologically separate world, thus a position that rejects Monism) (KpV, 5:102). While Kant does use the terms “contradiction” and “contradictory” in the CPrR passage (KpV, 5:101), as I argue in detail below, Kant sees the transcendental realist view as implying Essential Determinations because this is the only way to reconcile the various parts of the transcendental realist view with each other, in particular the claims about space, time, and God. Kant thinks that without Essential Determinations the transcendental realist view would be internally contradictory, but with it, it avoids contradiction.

That said, Kant does say of Spinozism in CPrR that there is an “absurdity” in “its fundamental idea” (KpV, 5:102). Moreover, in some lecture manuscripts, he speaks of a contradictory aspect to Spinoza’s Monism. However, here again, close reading reveals that Kant does not take it to be internally contradictory but instead to contradict the intuitive view of ourselves as substances. For an ‘I’ to affirm Spinozistic Monism (which turns me into an accident of God) “contradicts itself” insofar as “the [use of] ‘I’ shows that I cannot inhere in an another” (V-Met/Volckmann, 28:458). Monism “contradicts my concept of my I, in which I think myself as the ultimate subject which is not the predicate of any other thing” (V-Phil-Th/Pölitz, 28:1052–1053; V-Th/Baumbach, 28:1269). It is in this sense that Spinozism is absurd and contradictory: I cannot but think of myself as a substance and, since I know I am not God, Monism is untenable.

Because of these “absurd” consequences, Kant regards Spinozism as unacceptable. He takes the fact that the transcendental realist view leads to it to be grounds to adopt his own “Critical” transcendental idealist position, just as he takes the fact that his position is the only way to avoid accepting the soundness of the arguments of the Antinomies to be grounds for accepting his position. In this way, both arguments are of the reductio ad absurdum variety, though the absurdity in question is different – in one case we have conflicting Theses and Antitheses; in the case of Spinozism we do not.

It is noteworthy, however, that in the CPrR passage, Kant begins by treating Spinozism (especially Theological Determinism and Monism) as a potential difficulty for his own position, since he himself is also committed (albeit on practical grounds) to the existence of God as creator of all other substances. Kant’s strategy for parrying this worry is to first show that the transcendental realist view is saddled with the Spinozistic commitments, while then briefly explaining why his position is immune to these. My focus in this paper is on why Kant thinks that the transcendental realist view leads to Spinozism, not his case for thinking that his own position avoids it. On the latter point, it should be fairly obvious that, since transcendental idealists think that space and time are limited to appearances, and since God is supposed to be a thing-in-itself, idealists cannot be accused of holding that space and time are “essential determinations of God.”^[11]

3.1 Theological Assumptions

Fairly clearly, transcendental realism only entails Spinozism if it is combined with theological assumptions – no God going in, no going out. Kant evidently regards the assumptions in question as “dogmatic ones.” Consider that he sometimes speaks of ‘dogmatism’ or ‘dogmatic metaphysics’ as leading to Spinozism:

One can use Spinozism to overturn dogmatism. (R 6278, 18:546)

Spinozism is the true consequence of dogmatic metaphysics. (R 6050, 18:436)

Now, dogmatism for Kant involves uncritical acceptance of reason’s illusory deliverances about the supersensible. One of Kant’s examples of this phenomenon involves dogmatically-formed views based on reason’s naturally generated Idea of God (KrV, A 581/B 609). Such views include the beliefs that this being necessarily exists and is an ens originarium, ens realissimum, and ens summum (the highest being) (KrV, A 578/B 606). Kant thinks that reason naturally generates (and finds tempting) ‘dialectical’ arguments establishing such conclusions. Those who uncritically follow reason in this regard are, by Kant’s lights, dogmatists – like Mendelssohn. Kant complains, for example, of Mendelssohn’s failure to see how “dogmatism has its wings clipped through strict critique” and describes his Morning Hours (with its naive proofs of God’s existence) as a “masterpiece of the deception of our reason,” where this is a clear reference to the doctrine of transcendental illusion (Br, 10:428). I will have more to say about the details of this doctrine below, but for the present, transcendental illusion can be glossed as the phenomenon whereby human reason naturally views ‘subjective principles’ and Ideas of Reason (like that of God) as objectively real/true and naturally generates arguments that purport to establish the existence and properties of such objects. ‘Dogmatism’ for Kant involves being led astray by transcendental illusions that are natural to reason itself. Among other things, dogmatists like Mendelssohn are deceived into thinking the ens realissimum provably exists. In various places, Kant ties dogmatic reasoning about the ens realissimum to Spinozism (FM, 20:302).

Kant’s association of dogmatism with Spinozism, and his linking of dogmatism with acceptance of dialectical ‘proofs’ of the ens realissimum, suggest that the transcendental realist view he is targeting in CPrR is one that is dogmatically committed to God’s existence and various other claims about God, such as his being the ultimate cause of the existence of putative substances. Kant says of this last claim that it is “a proposition that can never be given up without also giving up the concept of God as the being of all beings and with it his all-sufficiency, on which everything in theology depend” (KpV, 5:100).^[12] Presumably Kant means by theology what he elsewhere calls transcendental theology, which “thinks its object […] merely through reason, by means of sheer transcendental concepts (as an ens originarium, realissimum, ens entium)” (KrV, A 631/B 659).

3.2 Space/Time Assumptions

Transcendental realism only leads to Spinozism in conjunction with assumptions about space and time that go beyond transcendental realism’s modest claim that they are features of things-in-themselves. For example, Kant says in the CPrR passage that space and time “must be presupposed as necessary a priori condition[s] of the existence of things” (KpV, 5:101). The context suggests that Kant thinks the transcendental realist position he is targeting must “presuppose” such a view. This can seem surprising since the view sounds similar to Kant’s own positive characterization of space and time. In the Transcendental Aesthetic, Kant argues that space and time are necessary a priori representations and subjective, formal conditions of the possibility of appearances. However, the position Kant is targeting with the Argument from Spinozism puts a realist, objectivist spin on the Kantian characterization of space and time, taking them to exist with “absolute necessity” and to be prior to and necessary for the existence of all things-in-themselves:

Things in space are all bound a priori to the conditions of space. Were space something in itself [that is, if space were transcendentally real] it would be necessary, and even God in his being would be bound to it. (R 6285, 18:552–553)

Space and time are such necessary a priori determinations of the existence of things that if they were determinations of things-in-themselves, then not only would they along with all the consequences that pertain to them have to be made into conditions of the existence of divinity, but, on account of their infinitude, absolute necessity, and necessity, they would even to have to be made into divine properties. For if one has once made them into determinations of things-in-themselves, then there would be no reason why they should be limited to finite beings. (R 6317, 18:626; underlining mine)

When Kant speaks of what would be true if space and time were transcendentally real – for example, that they would be absolutely necessary, infinite, and apply to all things-in-themselves including God – I take it he means that these are claims one would need to accept if one endorsed transcendental realism. The basis of this ‘need to’ requires clarification, as these specific claims about space and time are no more logical consequences of transcendental realism than are the theological claims described above.

There is a further substantive assumption concerning objects existing in space and time worth mentioning that figures specifically in the CPrR version of the argument: namely, that whatever arises in space and time (including our actions) is subject to thoroughgoing causal determinism (“natural necessity”) (KpV, 5:95–96). Kant calls this the “mechanism of nature” (KpV, 5:97) and the “natural mechanism of actions,” (KpV, 5:102) and we see this assumption figuring both in the pages immediately before the Argument from Spinozism (where Kant rejects the compatibilist view of freedom) and in the argument itself.

So far, I have spoken of the main assumptions figuring in the Argument from Spinozism. As we will see, there are some additional, as it were ‘connecting assumptions,’ that are used to move from the main assumptions to the conclusion. These include claims about what it takes for something to qualify as an X – for example, to qualify as a substance as opposed to an accident, or as an ultimate determining cause – as well as claims about what follows from the conjunction of the major assumptions. I think Kant regards these connecting assumptions as uncontroversial – indeed as analytic truths about underlying concepts. By contrast (what I am calling) the ‘main assumptions’ are evidently not analytic, insofar as they are quite substantive and have an existential element (space/time exist as things in themselves; the ens realissimum exists, etc.).

4.1 Transcendental Illusion and Ideas of Reason

To see that and how both sets of assumptions – theological and space/time – are connected to transcendental realism via transcendental illusion, we need to unpack Kant’s doctrine of illusion and Ideas of Reason. As noted above, transcendental illusion is Kant’s term for “a natural and unavoidable illusion” by which human reason views what are in fact merely subjectively valid principles as objectively valid principles holding for things-in-themselves:

Transcendental illusion […] does not cease even though it is uncovered and its nullity is clearly seen into by transcendental criticism (e. g., the illusion in the proposition: “the world must have a beginning in time”). The cause of this is that in our reason (considered subjectively as a human faculty of cognition) there lie fundamental rules and maxims for its use, which look entirely like objective principles, and through them it comes about that the subjective necessity of a certain connection of our concepts is taken for an objective necessity, the determination of things-in-themselves. [This is] an illusion that cannot be avoided at all, just as little as we can avoid the illusion that the sea appears higher in the middle than at the shores, since we see the former through higher rays of light than the latter, or even better, just as little as the astronomer can prevent the rising moon from appearing larger to him, even when he is not deceived by this illusion […] [W]hat we have to do with here is a natural and unavoidable illusion which rests on subjective principles and passes them off as objective […] Hence there is a natural and unavoidable dialectic of pure reason. (KrV, A 296–297/B 353–354)

These subjectively necessary principles and maxims that are “passed off” as objective include regulative principles of reason according to which it should always seek for the conditions of any kind of conditioned object (or event). In pursuing the various kinds of conditions, reason naturally forms representations of entities that satisfy its demand for a complete chain of conditions. These are Ideas of Reason, which are representations of unconditioned objects (e. g., God, the World, the Soul, and as we will see below, Absolute Space) that can never be given in experience.^[15] Kant thinks these Ideas of Reason have their proper, justifiable, and productive theoretical use as guides for scientific inquiry. For example, Kant takes the Idea of God to have its legitimate use in the thought of a systematic unity of nature (a view of nature as if ordered by God) that we should strive to realize in our science, by, for example, seeking for maximally unified and explanatory sets of empirical laws and genus/species relationships (KrV, A 619/B 647).

Transcendental illusion consists partly in the fact that the representations human reason naturally forms in accordance with its demand for completeness – the Ideas of Reason – also naturally appear to it to be really existing things-in-themselves.^[16] Principles and Ideas that are only subjectively necessary naturally look to reason like they exist in themselves and definitely possess various features (for example, the soul is viewed by reason as simple). Transcendental illusion also consists in the fact that human reason naturally generates, in accordance with its demand for completeness, ‘dialectical arguments’ that purport to demonstrate the existence, properties, and consequences of the (apparent) objects of its Ideas (e. g., God and the World as a totality).

Kant claims that this is as unavoidable as optical illusion (KrV, A 296–297/B 353–354). But just as with optical illusions one can avoid being taken in when one has a proper understanding of what is going on, Kant thinks that one can avoid being deceived by transcendental illusion when one adopts the Critical philosophy, including transcendental idealism. The Critical perspective allows one to see that reason’s natural view of things (wherein Ideas and principles look like they are objective) is just an illusion, and that the dialectical arguments that purport to establish the existence, properties, and consequences of the objects of the Ideas are dialectical. The Critical philosophy is able “to resolve and forever eliminate the unavoidable dialectic in which pure reason becomes involved and entangled when it is employed dogmatically everywhere” (WDO, 8:144n). While the transcendental illusion remains, it is no longer a threat to the Critical philosopher. By contrast, Kant thinks that human reason, under the spell of transcendental realism, is not able to see through the illusion. Recall Kant’s description of Mendelssohn’s Morning Hours as “a masterpiece of the deception of our reason.” Under the influence of transcendental realism, human reason cannot recognize that when it represents things in accordance with its (natural) demand for completeness, its view is distorted by the funhouse mirror of illusion. When in the grip of transcendental realism, human reason is unable to show, to its own satisfaction, that the claims and arguments that appear from its natural, Idea-mediated perspective to be warranted and sound are in fact illegitimate: for example, arguments that God necessarily exists (in this respect reason is dogmatic), or that the world is infinite, or as we will see, that Spinozism is true.

But why should transcendental realism have this blinding effect – and why does transcendental idealism enable human reason to expose, to its own satisfaction, these apparently well-founded claims and arguments as unfounded? Here is a sketch of a possible answer. If it is assumed that (1) objects given to us are all things-in-themselves (per transcendental realism), and further that (2) if the conditioned is given then the whole series of conditions is given as well (what Kant calls the ‘Supreme Principle of Reason,’ KrV, A 308–309/B 365–366), then human reason would seemingly be fully justified in thinking unconditioned objects are given to it and in turn cognizable by reason. Acceptance of transcendental realism leaves reason without a justified way of denying the Supreme Principle, which in turn motivates and supports the various dialectical inferences to the existence and properties of unconditioned objects like God, the world, the soul (and, as we will see, absolute space and absolute time). By contrast, the transcendental idealist denies that everything given is a thing-in-itself. As for the Supreme Principle, they will hold that it is a way that reason must think and form Ideas about things-in-themselves, but they will deny that we have any (epistemic) entitlement to think that these ways of thinking and forming Ideas must refer to anything actually existing, even though transcendental illusion makes it seem that they do. That is, they will hold that the Supreme Principle governs our thinking about things-in-themselves, but it need not govern thing-in-themselves. (As for appearances, they will simply deny that it is true of them.) As an ontological principle applied to things-in-themselves, the Supreme Principle is epistemically unwarranted and cannot be known. What enables the transcendental idealist to avoid being deceived here is that, as transcendental idealists, they distinguish subjective principles of sensibility and thought from objective principles of things-in-themselves. They also have independent reasons (e. g., in their account of synthetic a priori cognition) to deny our ability to know anything substantive regarding the supersensible. These aspects of their view position them exactly right to recognize as unwarranted this (synthetic) ontological principle, which, if it were theoretically warranted and knowable, would allow us to gain knowledge of the existence and properties of various supersensible objects (like God).

4.2 The Space/Time and Theological Assumptions as Rooted in Transcendental Illusion

We have already seen evidence that Kant views the theological assumptions that figure into the Argument from Spinozism as linked to transcendental illusion. These are commitments about God as ens realissimum (the “Ideal of reason”), for example, that this ens exists with absolute necessity and is the highest being.

I think it should be relatively non-controversial that Kant thinks the theological assumptions are linked to transcendental realism via transcendental illusion. What has not at all been appreciated is that the same plausibly holds for the space/time assumptions. An under-discussed passage in the Ideal of Pure Reason provides strong evidence for this reading. In this passage, Kant indicates that space is subject to a transcendental illusion analogous to that to which the Idea of the ens realissimum is subject:

The ideal of the highest being is, according to these considerations, nothing other than a regulative principle, to regard all combination in the world as if it arose from an all-sufficient necessary cause, so as to ground on that cause the rule of a unity that is systematic and necessary according to universal laws; but it is not an assertion of an existence that is necessary in itself. But at the same time it is unavoidable by means of a transcendental subreption, to represent this formal principle to oneself as constitutive, and to think of this unity hypostatically. For, just as with space since it originally makes possible all forms which are merely limitations of it, even though it is only a principle of sensibility, it is necessarily held to be a Something subsisting in itself with absolutely necessity and an a priori object given in itself, so it also comes about entirely naturally that since the systematic unity of nature cannot be set up as a principle of the empirical use of reason except on the basis of the idea of a most real being as the supreme cause, this idea is represented as an actual object, and this object again, because it is the supreme condition, is represented as necessary, so that a regulative principle is transformed into a constitutive one. (KrV, A 619/B 647; underlining mine, bold Kant’s)

Kant is clearly talking about transcendental illusion in this passage (“entirely naturally”). The passage shows, among other things, that the subjective principles that reason naturally views as constitutive do not only include regulative principles of reason. They also include our subjective forms of sensibility. There is a natural illusion by which space, as our form of sensibility/pure intuition, strikes reason as being a really existing, absolutely necessary thing-in-itself.

In this passage, Kant highlights the fact that reason naturally views the ens realissimum and space as existing with absolute necessity and as unconditioned conditions of possibility. The latter point is also made elsewhere in the Ideal, so that the analogy between the ens realissimum and space can be said to be one of the themes of the section:

All manifoldness of things is only so many different ways of limiting the concept of the highest reality, which is their common substratum, just as all figures are possible only as different ways of limiting infinite space. Hence the object of reason’s ideal, which is to be found only in reason, is also called the original being (ens originarium); because it has nothing above itself it is called the highest being (ens summum), and because everything else, as conditioned, stands under it, it is called the being of all beings (ens entium). (KrV, A 579/B 607; underlining mine)

One might wonder why Kant underscores this analogy between God and space, given that the former is an Idea of reason, while the latter is a form of sensibility. But in fact the analogy is strengthened when we consider that in other texts Kant explicitly claims that reason naturally generates an Idea of Absolute Space. Moreover, Kant is plausibly alluding to the Idea of Absolute Space in the under-discussed passage I began this subsection with.

As with other Ideas of Reason, Kant links the formation of the Idea of Absolute Space to reason’s pursuit of unconditioned conditions. As Kant explains in the Critique of Judgment, reason naturally views the Idea of Absolute Space as among other things an infinite, objectively “given” (that is, objectively existing) totality, and thus as complete in regard to its magnitude:

But now the mind hears in itself the voice of reason, which requires totality for all given magnitudes, even for those that can never be entirely apprehended although they are (in the sensible representation) judged as entirely given, hence comprehension in one intuition, and it demands a presentation for all members of a progressively increasing numerical series, and does not exempt from this requirement even the infinite (space and past time), but rather makes it unavoidable for us to think of it (in the judgment of common reason) as given entirely (in its totality).^[17] (KU, 5:254; cf. KU, 5:267)

(Notice the talk here of the “judgment of common reason,” which is plausibly an allusion to the doctrine of natural transcendental illusion.) In the Metaphysical Foundations, Kant focuses on another aspect of the Idea of Absolute Space: it is a representation of an all-encompassing, ultimate reference-frame for all motion – everything moves or is at rest in relation to it, but it (in contrast to merely “relative spaces”) is itself absolute and unmoving. It is thus conceived of as a kind of unconditioned condition of motion (MAN, 4:559–560; MAN, 4:480). The passage from the Ideal (which, as noted, is plausibly referring to the Idea of Absolute Space without however using that phrase) indicates that reason also conceives of Absolute Space as an unconditioned condition of spatial possibilities: it is a space that grounds the possibility of various kinds of figures/delimited spaces and spatial properties without having its own spatial possibility grounded in anything else (it is “original”).^[18]

Evidently, Kant thinks that, when human reason forms a representation of space in accordance with its demands for various kinds of completeness, it removes any restrictions on the application of our spatial form of intuition and injects the representation of space with the notion of being an ‘unconditioned condition’ for various series of conditions, thereby turning it into an Idea of Reason. This Idea of Absolute Space in turn naturally appears to reason to have as its object a really existing thing-in-itself that is complete in various respects (with respect to possibility, magnitude, and motion), and is unrestricted in its application – it holds of all existing objects. However, this is all an illusion: there is no such unconditioned thing-in-itself. While the Idea of Absolute Space does not correspond to any object in reality, it has (like the Idea of God) a legitimate usage in the investigation of motion as a regulative principle (MAN, 4:559–560; KrV, A 619/B 647).

I have focused here on space, just as Kant does, though I think parallel considerations hold of time. Although he is admittedly not explicit about this (I know of no texts where Kant speaks of the Idea of Time), Kant must hold either that there is a separate Idea of Absolute Time that is formed in a manner analogous to that of space, or he must hold that the Idea of Absolute Space incorporates in itself a conception of absolute time as unconditioned condition, so that what is dubbed the Idea of Absolute Space is really an Idea of Absolute Space and Time. On the latter proposal, the Idea of Absolute Space might include a concept of space as a persisting frame of reference for motion, and/or a concept of an absolute spatio-temporal reference frame. For my purposes, I just need one of these proposals (about how the Idea of Absolute Time enters the picture) to be plausibly ascribed to Kant; I need not determine which.

I think it is plausible that Kant takes the Idea of Absolute Space/Time to also include the thought of absolute space and time as the ground of a thoroughgoing causal determinism to which all items that exist and change in space and time are subject. (Recall that causal determinism was one of the assumptions in the Argument from Spinozism.) The position in question is not that absolute space/time is itself a causal determiner but that it is in virtue of its nature that items existing and coming to exist within it belong to a deterministic system. Though Kant is not as explicit as one might wish, in various texts he describes commitment to absolute space as entailing that space “determines” the relations and connections of things existing in it to one another (so that, for example, they belong to the same world or nature) (KrV, A 431/B 459; MSI, 2:406–407). And he speaks similarly of absolute space as the source of a kind of necessity (MSI, 2:391). One might take him to be saying here (among other things) that space/time is that in virtue of which items existing in it must stand in thoroughgoing causal connection according to laws of nature.

Even if one is not prepared to accept my specific proposal that the causal determinism assumption is due to a transcendental illusion connected to the Ideas of Space and Time, this is compatible with the more general, plausible proposal that Kant regards the assumption that objects in space are ipso facto subject to causal determinism as being in some sense ‘natural to reason.’ It would be a principle – of the sort Kant describes at the very beginning of the A-edition of the first Critique – “whose use is unavoidable in the course of experience,” with which reason proceeds “as its nature also requires,” and which “even ordinary commonsense agrees with” (KrV, A vii–viii).

This brief tour through Kant’s general account of transcendental illusion and Ideas provides everything we need for a satisfying solution to the questions we started this section with. Kant thinks that Spinozism follows from transcendental realism together with some further assumptions. The main assumptions (in contrast to what I called above the connecting assumptions) are the following:

• Theological assumptions: God is a thing-in-itself who exists with absolute necessity; God is an ens realissimum; ens originarium; ens summum; infinite; and an unconditioned condition of existence and possibility.

• Space/Time assumptions: space and time exist as things-in-themselves; they are a priori necessary for the existence of all objects/things-in-themselves; they are unconditioned grounds of all spatial and temporal possibility; they are infinite totalities and absolutely necessary; everything that exists and changes in space and time is, because of the nature of space and time, subject to thoroughgoing causal determinism.

It was initially unclear what sort of link Kant sees between these claims and transcendental realism (since they do not follow logically from it), but now we have seen that he thinks that these are claims about parallel and analogous Ideas of Reason, ones which human reason (1) naturally generates (given its need for representations that satisfy its demand for completeness), and (2) whose (apparent) objects naturally strike reason as being really existing things-in-themselves with various properties. Moreover, (3) human reason, under the spell of transcendental realism, is unable to see through the transcendental illusion and expose to its own satisfaction these natural claims as unwarranted. However, (4) human reason can do this with the aid of the Critical philosophy (which rejects transcendental realism). The sense in which the transcendental realist ‘must accept’ these claims is that they reflect a natural illusion, so that human reason will not be able to satisfy itself that these claims – which appear warranted to it – are in fact unwarranted, so long as reason is under the spell of transcendental realism.

It was also initially puzzling why the space/time commitments should sound so similar to Kant’s own positive account of space and time (but for their subjectivity), but now we have seen why this would be: Kant thinks that reason develops its Idea of Absolute Space/Time from the space and time given to us as forms of sensibility. Kant thinks that human reason naturally objectifies these, conceiving of them as unconditioned objects and things-in-themselves (but otherwise retaining the properties they have as forms of sensibility). That Kant sees transcendental illusion – whereby subjective principles viewed in the rearview mirror of reason appear more ontologically robust than they really are – as behind the extra space/time assumptions is strongly supported by the passage from the Ideal quoted above.

4.3 The Argument from Spinozism as a Natural Dialectical Argument

The phenomenon of transcendental illusion is not simply a matter of its striking reason that there exist unconditioned objects corresponding to its Ideas, but also reason’s coming up with dialectical arguments for their existence, properties, and consequences. In the case of the Antinomies, for example, reason (under the spell of transcendental realism) generates dialectical arguments establishing contradictory properties of the world as a totality. When reason assumes transcendental realism, then it not only takes itself to be warranted in viewing its Idea of the World as actually realized (it thinks objects of experience really constitute a world as a totality); it also produces arguments showing for example that this world is both finite in time/space and infinite in time/space. It is only by abandoning transcendental realism and instead accepting transcendental idealism that reason is able to avoid being deceived by the absurd antinomial conflict, even while the illusion that there is a world of appearances remains.

On my reading, Kant sees the Argument from Spinozism as like the Antinomies in this respect: it is a problem of reason in the form of a natural dialectical argument, one concerned not with the Idea of the World, but instead with the Ideas of Absolute Space, Time and God. He thinks human reason (under the spell of transcendental realism) naturally forms the space/time and God assumptions, and reasons from these plus transcendental realism to Spinozism. Consider one line of reasoning from transcendental realism and the assumptions to Essential Determinations. According to the theological assumptions, God is (among other things) a supreme, unlimited being, and unconditioned condition of all existence and possibility. But if absolute space and time exist as things in their own right (not as attributes of God, but rather as infinite, necessarily existing, self-subsisting containers for all existents) then various features of God are undermined: God’s existence and activity are limited and conditioned by space and time, so God is not truly supreme, unlimited, and unconditioned. (Another problem is that God would not in fact be the condition of all existence/possibility since space and time would be existents that are not conditioned by God.) Reconciling the theological and space/time assumptions requires viewing space and time as attributes of God.^[19] This reasoning incorporates the main assumptions regarding the Ideas of Space/Time and God, as well as what I have been calling connecting assumptions regarding what follows from the main assumptions. Kant plausibly regards these too as natural (and naturally compelling) to reason, though he need not take them to be illusory or false – indeed, I think he regards them as analytic truths.^[20] The Antinomies are similar in this regard, incorporating both illusory assumptions about the Idea of the World and connecting assumptions, for example, about the necessary conditions on transcendental freedom in the third antinomy.

In contrast to the Antinomies, the conclusion of the argument from Spinozism is not antinomial. Instead, as I argued in section 2, Spinozism is internally consistent, though as we saw, Kant does regard it as absurd. One might wonder how reason could be naturally led to move from transcendental realism to two very different conclusions: both to Antimonies and Spinozism. However, we have to keep in mind that different Ideas of Reason are in play in the Antinomies as opposed to the Argument from Spinozism. The fact that different conclusions are reached is no more surprising than the fact that, say, the arguments in the Ideal of Reason (claim to) establish the necessary existence of God, while the paralogistic arguments (claim to) establish the existence and various properties of the soul, like its simplicity. Transcendental realism when combined with different Ideas (and different natural commitments) leads to different conclusions.

In the case of the Argument from Spinozism, I think that Kant holds that the various assumptions are natural to reason, as is the reasoning that leads from them and transcendental realism to Spinozism. Kant holds further that Spinozism really does follow just as he holds that the conclusions of the thesis and antithesis arguments really do follow on the assumption of transcendental realism. In this way, the Antinomies are a reductio ad absurdum of transcendental realism (and in turn an indirect argument for the truth of transcendental idealism). So too is the Argument from Spinozism.

To be clear, I am not claiming that Kant explicitly formulates the Argument from Spinozism in the Transcendental Dialectic of the Critique of the Pure Reason. Kant first explicitly articulated the argument in the CPrR, doing so in the wake of the Pantheism Controversy. However, I think that when Kant, inspired by the controversy, formulates the argument in CPrR (and reiterates a condensed version of it over and over in later writings), he is consciously drawing on his account of reason and transcendental illusion in the Dialectic, especially the Ideal of Pure Reason. Consider in this regard that Kant discusses the phenomenon of transcendental illusion (and reason’s search for the unconditioned condition) in CPrR just a few pages after formulating the Argument from Spinozism (KpV, 5:107).

Evidently, the Pantheism Controversy led Kant to believe that the Argument from Spinozism is a dialectical argument natural to reason. At the same time, he saw that his account of transcendental illusion and Ideas of Reason (including the analogy between the Idea of Absolute Space and ens realissimum) could explain this. He also saw that, as with other natural dialectical arguments, one can avoid being deceived by them only by means of the Critical philosophy. Jacobi had claimed that reason leads to Spinozism. From Kant’s standpoint, Jacobi is in a sense right about that (there is an illusion and a natural dialectic there), but Jacobi errs when he suggests that reason has no way to expose this view as unwarranted and false. Jacobi, from Kant’s standpoint, missed the salvific force of the Critical philosophy.

I call my interpretation the Natural to Reason Reading. It is supported by the texts we have considered so far, and it makes good sense of the source and content of the assumptions that figure in the Argument from Spinozism. Kant’s talk of Spinozism “arising in an instant” (V-Met-K2/Heinze, 28:803) may be taken to mean that Spinozism arises for reason naturally – automatically and inevitably, just as the Antinomies do.

Step 1: Essential Determinations

Essential Determinations is the only way of reconciling the theological and space/time assumptions. For example, the theological assumptions include God’s being independent, unconditioned, infinite, and the highest and most real being. But it follows from the space/time assumptions that God must exist in space and time. That is because absolute space and time are unconditioned entities and are unrestricted in their application. Moreover, God’s production of things must take place in space and time (since these are conditions of all existence). Now, if Essential Determinations is false, and God is something really distinct from space and time, then God’s independence (and status as unconditioned condition) is compromised, since God is subject to separately existing entities (absolute space and time) that condition his existence and his causality.^[26] God’s status as ‘highest being’ is also compromised, since all reality is supposed to be in or grounded in God, but if space and time are entities separate from God and not grounded in God (they are unconditioned), then in fact God is not really the ground of all reality. Moreover, space and time are infinite realities. This is incompatible with God’s infinity if it is assumed that an infinite being is not limited or conditioned by anything else. I think this is what Kant is getting at when he says that on the transcendental realist view “[God’s] causality with respect to the existence of these things must be conditioned and even temporally conditioned; this would unavoidably have to bring in all that is contradictory to the concept of his infinity and independence” (KpV, 5:101).^[27] Essential Determinations removes the contradictions that would otherwise result, and it does so in a way that preserves the “absolute necessity” of space and time (V-Met-K2/Heinze, 28:803). Notice that Essential Determinations is not saying that God grounds space and time. It is saying that they are essential properties of God. As such, they are unconditioned in the way that Spinoza’s attributes are (they are the divinity).

The CPrR passage is complicated by the fact that Kant also considers a broadly transcendental realist position that attempts to avoid Essential Determinations and in turn Monism, by (1) maintaining space and time are confined to finite things-in-themselves, and/or (2) maintaining that God grounds space and time. Such a position attempts to uphold the ‘Creation Theory’ (the view that God creates a world that he is ontologically distinct from him). Kant mentions Mendelssohn as an example of (1):

if they [transcendental realist proponents of the creation theory] flatly allow both [space and time] to be conditions necessarily belonging only to the existence of finite and derived beings but not to that of the infinite original being, I do not see how they would justify themselves in making such a distinction, whence they get a warrant to do so, or even how they would avoid the contradiction they encounter when they regard existence in time as a determination attaching necessarily to finite things-in-themselves, while God is the cause of this existence but cannot be the cause of time (or space) itself (because this must be presupposed as a necessary a priori condition of the existence of things); and consequently his causality with respect to the existence of these things must be conditioned and even temporary conditioned; this would unavoidably have to bring in all that is contradictory to the concept of his infinity and independence. (KpV, 5:101)

He responds to this position by first noting that the restriction is unjustified. As I indicated above, I take this to be a way of saying that human reason, under the guidance of transcendental realism naturally takes to be warranted the assumption that space and time are conditions of all existents. (Elsewhere, Kant says there is “no reason”; R 6317, 18:626) for the restriction to only finite existents. On my view, Kant is tacitly invoking his theory of reason, Ideas, and illusion here.) This is related to the fact that reason naturally views space and time as unconditioned conditions, and the fact that it arrives at the Ideas of absolute space and time by removing restrictions on space and time as subjective forms of intuition. In addition, this restriction does not avoid the contradictions that would also result from a non-spatial and temporal God’s creating things in space and time. Nor will it do to try to avoid this consequence by (option 2) claiming that absolute space and time are entities external to God that depend on God for their existence, for this conflicts with their being unconditioned conditions.^[28]

Step 2: Monism

The reasoning for Essential Determinations does not proceed via either of the other two theses – those theses do not provide it with logical support. On the contrary, Kant plausibly takes Monism to logically depend on Essential Determinations, at least in the condensed version of the argument. What is less clear is what, if any, sort of logical dependence there might be between Theological Determinism and Monism in the CPrR (expanded) version of the argument. According to one possible interpretation, the argument is that because our actions are determined by God in such a way that we are not free (Theological Determinism), Monism is true (or, alternatively, because of Theological Determinism and because of Essential Determinations, Monism is true). Such an interpretation would have to explain what the reasoning is for Theological Determinism, and why Monism is supposed to be a consequence of it.

I will not try to claim that such interpretations will not work, nor other possible interpretations in logical space. However, for various reasons, I find particularly attractive an interpretation whereby Theological Determinism is logically dependent on Monism. First, recall that the condensed Argument from Spinozism leaves out Theological Determinism, and so it cannot be playing any role in the reasoning to Monism. Second, as we will see, there is a clear and direct path leading from Essential Determinations and various assumptions, to Monism (without needing to go through Theological Determinism). Third, I think it is challenging to make sense of the reasoning for Theological Determinism without taking it to involve first establishing Monism, in which case it becomes significantly easier to understand. This is particularly so if we are inclined (as I am) to interpret Theological Determinism as ruling out not just transcendental freedom but also Leibnizian-style compatibilist freedom. That we do not have the latter kind of freedom, and that God is the author of our actions, would seem to require showing first that we are not substances.

With that said, here is how I understand the argument for Monism. If space and time are God’s essential determinations (as was previously established), and if space and time apply to all things-in-themselves (per transcendental realism), then everything besides God that exists (like me) exists in God. But how could these things be substances if they exist ‘in’ another substance – indeed, if they exist in its essential determinations? Spatial containment ends up being inherence in a substance – and so, voilà, we have Monism. (Kant is here relying on an auxiliary assumption about the conditions on being a substance and an accident of a substance.) In fact, Kant’s case for the non-substantiality of items in space and time, given Essential Determinations, is arguably strengthened by the fact that those items are subject to causal determinism because of the nature of absolute space and time (this was one of the space/time assumptions). Here one might plausibly hold that for substances to be substances and subject to natural necessity, this must be due to their own natures and causal powers, as opposed to being necessitated by the nature of something that is ontologically prior to them and able to exist without them, namely absolute space and time.

Step 3: Theological Determinism

According to the space/time assumptions, everything that exists in time and space is part of a deterministic causal series. Plausibly, if something exists and acts as part of deterministic causal series and there is some ultimate cause of that thing’s existence and the causal series it belongs to, then that ultimate cause is itself the ultimate determining cause of the thing’s actions.^[29] (This is, again, what I am calling a connecting assumption.) According to the theological assumptions, God is such an ultimate cause. Thus, God is the ultimate determining cause of my actions. This rules out transcendental freedom for my actions. Now, if we add that the things determined to exist, to act, and change by God are not substances (per Monism, previously derived), it plausibly follows that they are not in any way free or the authors of their actions. Leibnizian-style compatibilist freedom has substantiality as a condition – it is only substances that can be authors of their actions. Instead, actions like lifting my hand are ultimately to be imputed to a substance, and since God is the only substance, “then the actions of these beings would have to be merely its actions that it performs in any place and at any time” (KpV, 5:102). In this way, we arrive at Theological Determinism.