I. Spinoza and the problem of systematic argument

Hegel aims to defend a philosophical system. For my purposes here, it is enough to think of two senses of systematicity:

1. a. Hegel aims for a whole philosophy with some unifying organization, rather than just a heap of arguments.

2. b. Hegel aims for some kind of comprehensiveness, so that what he is doing, especially in the Logic, should have implications everywhere in philosophy.

While we have readily accessible models to explain how arguments work, I do not think they can on their own explain how Hegel even hopes to argue for a philosophical system, in these senses.

We can of course spell out arguments in premise-conclusion form. We can explain, in this way, Descartes’s conceivability argument for substance dualism. But this is clearly just one argument, and anything similar to this is not yet really trying to explain how an argument might aim for comprehensiveness. We can of course think of another argument in premise-conclusion form—say, Hume’s argument from perceptual variation against direct realism. Does it make sense to think of aiming to compile a list of such argument explanations so long that it aims for comprehensiveness? It hardly seems to matter, because the farther we go in this direction, the farther we seem to move away from any organizing unity, and so in this respect away from a philosophical system. Perhaps we would do better, with organization, if we restrict ourselves to arguments addressing only one kind of philosophical issue—say, issues concerning perception—but then we clearly seem to do worse with respect to comprehensiveness, resulting in a kind of narrowness. If the argument for a system is still lacking, then we can try to imagine it is just another argument explicable in premise-conclusion form, which we could add to our list; but since the problem concerns in part what organizes or unifies all of the items on the list, it is hard to see how adding another element to our flat list could provide what is needed. And so it seems from the beginning difficult to see how one might hope to explain how argument could support a system.

We can explore the problem further by asking: if there is an argument supporting Hegel’s system, then what would be the specific issues addressed by this argument? For example, would they be the epistemological issues at stake in Descartes’s argument for dreaming scepticism? Or some other set of issues? With respect to the Logic, there is a proposal that is entirely natural, but in fact makes no progress toward resolving our problem. The natural proposal is that the central issues in the systematic argument of the book concern self-determination or freedom, for the transition to the final part of the Logic is clearly a transition to this topic: the conclusion of the Doctrine of Essence is supposed to show that Spinoza’s ‘necessity is elevated to freedom’ (SL: 11:408/504).Footnote ³ And this gives the Subjective Logic or the Doctrine of the Concept its topic: ‘the concept, the realm of subjectivity or of freedom’ (SL: 11:409/505).Footnote ⁴

But what is the argument suggested by the natural proposal? On the face it, the very broad strokes of the argument would be something like this:

1. a. Spinoza’s account of the absolute as substance precludes a sufficient account of freedom.

2. b. Thus, Spinoza’s philosophy is incomplete, and substance must also be self-determining subjectivity.

This too is a natural way of thinking. After all, Spinoza is radical in dismissing at least some senses of freedom, including freedom of the will, even in the case of God (e.g., E: 1P32); Hegel’s interpretation of Spinoza, following Jacobi, is specifically as a determinist and fatalist; and Hegel is certainly eager to contrast his metaphysics with Spinoza’s on this score. But the question here is one of argument, and if we start with this argument structure, then I think that any way of filling in the details will still fall short of supporting Hegel’s system. This is for reasons that Hegel himself highlights in the transition to the Subjective Logic. For here Hegel says that the Spinozist can resist any refutation from outside:

The kind of argument I just sketched cannot meet this standard: for all it makes explicit, it merely assumes that there is freedom in some sense ruled out by Spinoza’s system. And this assumption is what Hegel singles out for complaint, alluding to Fichte:

Since the natural proposal about self-determination does not explain how an argument in Hegel hopes to really engage and impact Spinoza’s system, it certainly does not yet explain how an argument in Hegel hopes to impact comprehensively everything.

The results are similar, I think, if we take the currently popular route of treating the Logic as focused on issues concerning what is now called ‘normativity’. Perhaps there is some sense in which Spinoza leaves no place for some kind of normativity. But if we explain by saying that the basic issue-focus, all the way down, is on that kind of normativity, then it looks as if the Spinozist can declare whatever part of this she cannot reconstruct within her system as something external and indifferent to her system.

I am not sure if anyone would be tempted, in the face of these difficulties, to hold that Hegel is somehow beyond arguments or refutations, but it is important to note that Hegel himself clearly conceives of philosophy as requiring argument. This is clear, for example, in what his criticisms of Schelling—right or wrong about Schelling—tell us about Hegel’s own commitments. Hegel says that ‘[w]hat is lacking in Schelling’s philosophy’ is that its central claims are ‘absolutely presupposed, without any attempt being made at proving that this is the truth’. Schelling is supposed to appeal to ‘intellectual intuition’ in a specific manner amounting to only bare assertion of authority or privilege, so that in response, ‘[o]ne can say nothing else than: you do not have intellectual intuition, if this appears false to you … The proving of anything, making it comprehensible, disappears’ (LHP: 20:435/3:525–26). Hegel, by contrast, says that ‘[w]hen we philosophize, we want to have proven that it is so’ (LHP 20:435/3:525–26).Footnote ⁶ Setting aside the Schelling interpretation, if there are any philosophers claiming to be somehow beyond argument, this seems to me a powerful response, and certainly makes clear Hegel’s view of philosophy as committed to argument. So it is worth focusing interpretive efforts on Hegel’s systematic argument, and working to broaden our appreciation of the forms that arguments can take, until we can make sense of this.Footnote ⁷

II. Transcendental analytic model of systematic argument

My focus on these problems so far is not meant to suggest that we have no model of how Hegel’s systematic argument might work. I think there is one familiar kind of interpretation that can in fact provide us with a powerful model. This will not be my approach to Hegel, and I do not mean here to get bogged down in interpreting interpreters. My aim is to start with some familiar and powerful ideas,Footnote ⁸ and to extrapolate a model of how an argument for a system could work, and draw some positive lessons from this.

The key text here is the praise, in the introduction to the Subjective Logic, of Kant’s discussions of the transcendental unity of apperception, including the account of these in terms of the constitution of ‘objective validity’; this ‘is one of the profoundest and truest insights which constitutes the essence of the concept’ (SL: 12.18/515). Of course, Hegel also here criticizes Kant; for example, Kant is supposed to fail to systematically derive the categories from the fundamental principle of apperception (SL: 12:28/525). But one approach would be to understand this as a basically Kantian attempt to improve Kant’s execution.

Now the idea that Hegel’s systematic arguments might focus on the topic of the transcendental unity of apperception does not automatically resolve the problem noted in the previous section, concerning comprehensiveness and the relation to Spinoza’s system. We can certainly imagine arguing that Spinoza’s system does not leave room for something about a transcendental unity of apperception, or perhaps a kind of spontaneity associated with apperception.Footnote ⁹ But whatever that something is, precisely to the extent that it could not be reconstructed within Spinoza’s system, the way would remain open for the Spinozist to declare it a matter of indifference. So the problem remains that a systematic argument focused all the way down on issues involving apperception would apparently lack an argument for investing such importance in the topic in the first place.

Progress requires instead consideration of the arguments of the Transcendental Analytic of Kant’s first Critique where apperception figures so importantly. These arguments do not assume its importance; they argue for the importance of apperception from consideration of broader issues. What are the broader issues? One way of formulating them comes from Kant’s famous letter to Herz, focused on the question: ‘What is the ground of the relation of that in us which we call ‘representation’ to the object?’ (131/10:130). To go farther into the Analytic itself would mean breaking down the issues in more detail: perhaps the issues concern the possibility of correct representation. Or perhaps there is some complex interplay with issues concerning conditions of the possibility of any representation at all. Either way, we can already see how a focus on these issues from the Transcendental Analytic might be thought to change things with respect to the engagement with Spinoza. The idea would be that the Spinozist himself is purporting to think of or represent or judge about an object—about substance, for example. So, one could argue that the Spinozist owes an account of how such representation or judgement is possible, or an answer to the question of how the concepts he is using designate something in some determinate fashion, and something one might then judge correctly or incorrectly. And the way would be open to argue that the Spinozist must fail to provide what she owes. For example, if the Kantian can show that accounting for how concept-use represents determinate objects requires a Transcendental Unity of Apperception, in some sense, and that this requires in turn some spontaneity of the subject—and if Spinoza’s monism somehow conflicts with our having that sort of spontaneity—then we might argue thereby that there is an inescapable problem already inside Spinozism, preventing the Spinozist from declaring these issues to be external and a matter of indifference.

This approach to Hegel would also give us a nice approach to what it means to transition to a Subjective Logic. The idea would be that an Objective Logic would concern attempts at philosophical theories of objects considered independently of spontaneous and apperceptive contribution by a judging subject. The failure of these attempts would be supposed to demonstrate the need, now in a subjective logic, to construct a theory foregrounding the account of the representation of objects, and specifically in terms of the active contribution of a subject.

And this approach can explain in similar terms Hegel’s criticism of Kant as well. Part of it would run along these lines: consider Kant’s epistemic restriction, leaving us inescapably ignorant of things in themselves. One might argue that, if knowledge of objects requires an active contribution from the subject, and if this would leave things in themselves unknowable, then those very supposed things would not even be representable or thinkable for us.Footnote ¹⁰ Thus the unknowability claims would be eliminated from within.Footnote ¹¹ And Hegel might then seek to argue, against Kant, that properly executed deductions do not just generate conclusions about a priori conditions relative to a restricted form of cognition of ours, but rather a priori conditions that are absolute.Footnote ¹²

This approach to Hegel may be relatively familiar, but I want to argue that we better understand its considerable power if we make more explicit what I call—following early work by Rorty—a metaphilosophical claim: a claim about what is fundamental or inescapable in philosophy.Footnote ¹³ In particular, the claim here would assert the fundamentality or inescapability in philosophy generally of the issues favoured in the sketch above: those concerning the possibility of representation. Note that this is what provides the promising approach to both respects, with which I began, in which Hegel seeks to support a system.

First, the metaphilosophical claim would explain a unifying organization, but without narrowness: A system could address many diverse issues throughout philosophy, but all in a manner organized by approaching them through the lens of our one supposedly most fundamental kind of issue. So our system inspired by the Transcendental Analytic need not just address one kind of issue. If it argues that some form of spontaneous apperception is needed to make sense of representation, then we might also explore its implications concerning ethical issues, for example.Footnote ¹⁴ And the prospects for this approach hinge, to my mind, on the same being true of metaphysical issues: this approach need not banish them, or give a so-called ‘non-metaphysical’ reading of Hegel. Metaphysical issues, if they can be formulated and approached in light of fundamental questions about representation, could be resolved in that same unified and organized fashion. If we prefer to say that the fundamental issues concern ‘intentionality’, then we could call this a ‘metaphysics of intentionality’.Footnote ¹⁵

Second, this would allow a claim to comprehensiveness: we could in this way try to argue that other, rival philosophical projects in general, in so far as they all must make claims about something or purport to represent something, would be impacted by this systematic approach. We could say that rival systems would generally be addressed by means of the kind of argument just-noted with respect to Spinoza, which I would call a ‘track-shifting’ argument: philosophical systems working along a track that fails to foreground issues about the possibility of representation inevitably incur a philosophical debt; a truly systematic and comprehensive philosophy would have to pay that debt, and so would have to shift tracks, to a project committed now to view everything rather through the lens of issues about representation of objects. Meanwhile, Kant would be addressed in a different way. We would read Kant’s own system as based around the argument of the Transcendental Analytic, where transcendental apperception figures so prominently; and then we would give what I would call a ‘track-extending’ response: We would claim our new system follows the track of Kantian deductions, from the Analytic, even farther than Kant managed.Footnote ¹⁶

So here we can make explicit how an argument might support systematicity: it might do so by defending a metaphilosophical commitment that will then organize a system of subsidiary arguments addressing the philosophical comprehensively. This way in which arguments reflect back on the nature of philosophy itself turns out to be what was missing in our initial thoughts about a kind of flat list of arguments in premise-conclusion form, addressing substance dualism, perception, etc.

All this gives us a wonderfully systematic, consistent approach to four important questions considered below:

Transcendental Analytic Model of Hegel’s Systematic Argument

1. 1. What are the fundamental issues at stake in Hegel’s arguments? Issues about the conceptual conditions of the possibility of representation or cognition of objects.

2. 2. What is the central argument strategy? Focus on the case for a necessary spontaneity of the subject in any possible representation of objects.

3. 3. What distinguishes Subjective as opposed to Objective Logic? Subjective Logic has overcome attempts at theories of objects supposedly independent of contribution of or mediation by spontaneous subject.

4. 4. What is the root reason why pre-Kantian metaphysics is supposed to fail? On the grounds of broadly epistemological problems about the possibility of representation and/or knowledge of its supposed objects, like Spinoza’s one substance, God or nature.

This seems to me a powerful model of everything a systematic interpretation of the Logic should be. But I should say that it is not my approach; my aim is to abstract the model and discard the specifics focused on Kant’s Transcendental Analytic. To explain why I do not rather just stop here and declare success, I will mention my worries about this Transcendental Analytic approach to the Logic.

I have discussed in detail one worry previously: I see Hegel as considering carefully this kind of philosophical program, oriented around worries about the possibility of cognition of objects. Part of his gloss of such programs is this: ‘prior to setting about to acquire cognition of God, the essence of things, etc., the faculty of cognition itself would have to be examined’. But what I call Hegel’s ‘swimming argument’ makes the case that the argument for such programs is ‘as incoherent as the Scholastic’s wise resolution to learn to swim, before he ventured into the water …’ (EL: §10 An). Since Hegel’s argument is quite general, I have argued that it is some reason, at least, to read him as pursuing a different kind of program.Footnote ¹⁷

A second worry is more connected to my topic in this paper, involving the discussion of Spinoza noted above. Part of this is a philosophical worry: it seems to me that the Spinozist can answer that the track-shifting argument above begs the question. Say we argue for the priority of issues about the possibility of cognition of objects by focusing, through and through, on worries about the possibility of cognition of objects. If so, then we are really just assuming the priority of such issues from the start. Someone like a Spinozist could then, with equal right, adopt an opposed metaphilosophy, making the case in his terms that his issues are more important, fundamental, and so on. The Spinozist might take the success of her arguments (as she sees it) as all the reason we need to accept the possibility of cognition of objects, including those with which she is concerned.

Rorty’s wonderful early work on metaphilosophy generalizes this kind of worry. Any revolutionary will want to found philosophy on a new method, with its own metaphilosophy. But the old guard will have its old metaphilosophy, and ‘[s]ince philosophical method is in itself a philosophical topic … every philosophical revolutionary is open to the charge of circularity’ (1967: 2).

But aside from this philosophical concern and Rorty’s general worry, to which I return below, consider the interpretive issues regarding Hegel’s discussion, above, of the refutation of Spinoza. Hegel seems not to argue that Spinoza’s system-building prioritizes the wrong issues, and requires a track-shift—for instance, towards building around issues about the possibility of cognition of objects. On the contrary, Hegel says that what is needed, with respect to Spinozism, is ‘acknowledging its standpoint as essential and necessary and then raising it to a higher standpoint on the strength of its own resources’ (SL: 12:15/512). I do not see how this promise to advance by means of Spinoza’s own resources could fit the idea that Hegel’s Subjective Logic is really switching tracks, away from Spinoza’s issues, to build on something more like a deduction of conceptual conditions of cognition, from Kant’s Transcendental Analytic.

The third worry, again central to my topic today, is this: I think that understanding Hegel’s response to Kant in this way, as an attempt to engage internally and overcome Kant’s arguments for epistemic restriction, would miss something important in Kant. For I do not think that Kant rests his restriction, or his denial of the knowability of things in themselves, on considerations from the Transcendental Analytic. So, we cannot in this way show that considerations from a Kantian argument for the restriction actually, when taken further, support eliminating the restriction. The restriction thesis is part of the package of Transcendental Idealism. Kant does not think that either idealism, or any limitation of our knowledge, follows just from the need for our spontaneity to play a role in any knowledge of objects; he does not give a so-called ‘short argument’ for idealism, but rests his arguments for restriction, and transcendental idealism more generally, on details concerning the pure forms of our sensible intuition: space and time.Footnote ¹⁸ So if one wants to turn Kant’s own argument for the restriction against Kant—as it is pretty widely agreed that Hegel does—then this seems to me to require according rather the central role to either the Aesthetic or (as I will argue) the Dialectic of Kant’s first Critique.

III. Kant’s Dialectic as infiltration of the opponent’s territory

I seek, then, to borrow the above model of a systematic interpretation, but to fill it in with a very different approach to Hegel. Recall again, from my central Hegel text, the problem Hegel raises about there being, with respect to a pre-critical metaphysician like Spinoza, ‘no point in attacking him outside his territory’. This raises an interesting question: does Kant have a critique of pre-critical metaphysics that might address this kind of worry? I think that he recognizes the need, and supplies one. More specifically, my view is that Kant’s Transcendental Dialectic is a solution to just this kind of problem. Consider the way in which the B-Preface distinguishes two main lines of argument in the Critique. The first, epistemological line, addresses issues about the conditions of the possibility of experience, defending synthetic a priori knowledge; the second, centered on the Dialectic, addresses the topics of pre-critical metaphysics. From the first line, Kant says, ‘there emerges a very strange result and one that appears very disadvantageous to the whole purpose with which the second part of metaphysics concerns itself’: ‘we can never get beyond the boundaries of possible experience’ (B: xix). This gets at just how a pre-critical metaphysician, concerned with objects beyond experience, should respond to the first line of argument: given the metaphilosophical commitment built into the kind of project she pursues, it would be strange to recognize the authority of the topic of Kant’s epistemological reflection over her project, bringing as it does such disadvantages concerning the issues and concern that (on her view) are central. But this is the occasion for Kant’s second overarching line of argument, which presumably will not have this limitation relative to pre-critical metaphysics: it will, in my Hegelian terms here, ‘infiltrate the opponent’s stronghold and meet him on his own ground’. So Kant immediately proceeds to say that ‘herein lies just the experiment providing a checkup on the truth’ of his result, or on the position ‘leaving the thing in itself as something actual for itself but uncognized by us’ (B: xx)

I take this to mean that the issue-focus of Kant’s Dialectic criticism cannot be distinguished in broadly epistemological terms. The basic focus cannot be on the possibility of representation, and not on a priority or a priori objective validity, and so on. Nor can the focus be any claim that accounting for representation requires attention not just to judgement but to a broader context of rational inferences, or a space of reasons or the like. That thought may be in Hegel, and it may be in Kant’s Dialectic. But, even if so, it cannot be what drives Kant’s criticism of pre-critical metaphysics, and (I will argue) it cannot be the way that this material from Kant influences Hegel’s systematic argument.

On my view, then, we must keep the critical argument in mind when thinking of what Kant means in highlighting the focus of the Dialectic on conditions and ‘the unconditioned’ (B: xx), which are supposed to be of direct interest to our faculty of reason. In the argument critical of prior metaphysics, I think the key sense of condition is: something worldly that is such as can be appealed to in explaining. The corresponding meaning of Kant’s term ‘unconditioned’, I think, would be something worldly such that appeal to it might explain completely some particular kind of regress of conditions.Footnote ¹⁹ These issues are designed by Kant to capture the interests of the pre-critical rationalists, who are his central targets. Take Leibniz’s theory of monads. Monads would be supposed to explain how there could be composites. The topic here is not at base epistemological. It is not about what we can infer from what. Indeed, the epistemology, the inference, goes in the other direction: we infer the existence of monads from the existence of composites. But the metaphysical supporting or grounding or conditioning would flow from the monads, explaining how there can be composites. The existence of composites does not similarly explain how there can be monads.

So the issue-focus here is on what I will call the metaphysics of conditions and the unconditioned; or, I would also say, the metaphysics of explanation. Kant holds that the faculty of reason itself requires that philosophy directly raise and pursue those issues—that we occupy this territory, to use Hegel’s metaphor. But the way Kant argues in the Dialectic is to demonstrate that our pursuing these issues, or occupation of this territory, necessarily leads to the contradictions of the Antinomy (e.g., B: xx). The contradictions, then, must be demonstrated without appeal to Kant’s epistemology, and without appeal to transcendental idealism, because the Antinomy is supposed give independent support for all that. In Kant’s terms, ‘the antinomy of pure reason leads inevitably back to that limiting of our cognition’ (Ak. 20:290–91). That is, we are supposed to conclude that our knowledge is limited, and that our pursuit of theoretical philosophy can never really answer the nonetheless necessary and inescapable questions raised by reason itself. Kant wants to conclude that progress in philosophy requires transforming metaphysics and everything else—it requires re-orienting everything around reflection on broadly epistemological issues concerning conditions of the possibility of representation of objects by a subject.Footnote ²⁰

Now, recall the problem about refutation of a system on its own territory, or Rorty’s worry that targets of refutation can always escape by contesting on a metaphilosophical level. Kant’s Dialectic seems to me a powerful solution to this kind of problem: we can assume the opponents’ metaphilosophy, or the priority of her favoured issues, but for the purposes of showing that the opponents’ project, systematically pursued, goes astray from within—and in some way requiring a solution available only if we shift tracks, to a different and favoured kind of philosophical program.

Hegel, on my view, hijacks, for his own purposes, the Dialectic model, so that the prominence he gives its terminology—‘dialectic’, ‘reason’, and so on—is no surprise. Hegel would thus be agreeing with Kant’s Dialectic in seeing the fundamental issues as those concerning the metaphysics of conditioning and the unconditioned, in Kantian terms. And Hegel would agree that a philosophy pursuing this focus necessarily generates contradictions. But Hegel would be taking the contradictions to teach a very different lesson. They would not show any need for philosophy to shift tracks, making another set of issues fundamental rather than the metaphysics of conditions and the unconditioned. Rather, they would show how to reform and fix such a metaphysics, or respond to these issues in a way that is, for the first time, truly systematic because it is uniformly built around these contradictions. So Hegel aims to show that it never was necessary to change tracks, in the manner Kant thinks is required—that we can and should better extend the track yet farther along its course in the Dialectic. The Logic puts the point like this:

Furthermore, Hegel says that Antinomy contradictions are not limited, as in Kant’s account, but occur everywhere (EL: §48R)—which is what makes for a Hegelian or a dialectical Logic.

I conclude that the considerations about refutations of systems, in my focus text at the transition to the Subjective Logic, suggest further support for my thesis, that the primary issues to be built around in Hegel’s systematic argument come from Kant’s Dialectic, rather than Analytic. I now turn to testing this result against the challenges involved in making sense of what Hegel says at the transition about Spinoza’s and Kant’s systems.

IV. Rising higher with Spinoza’s own resources

The first step, then, is to explain how reading Hegel as extending the considerations of the Dialectic makes sense of his claim to surpass Spinoza’s system building from Spinoza’s own resources. Here I want to highlight two roots or elements which support Spinoza’s argument for monism, and are highlighted in Hegel’s lectures’ explanation of it. The first element is a principle requiring that everything be explicable, which Spinoza interpreters (although not Hegel) now tend to call the principle of sufficient reason (PSR).Footnote ²¹ Spinoza says for example: ‘For each thing there must be assigned a cause, or reason’ (E: 1P11D2).Footnote ²² I see Hegel as thinking of this in terms of explanation, in so far as Hegel views it all through the lens of Jacobi’s interpretation. Jacobi suggests that the Spinozist ‘wants to explain all things absolutely … and will not otherwise let anything stand’ (30/194). Jacobi—and we will see that Hegel follows—thinks of this as the principle that nothing comes from nothing: ‘the spirit of Spinozism … certainly nothing other than the ancient a nihilo nihil fit’ (14/187). Hegel follows. The second element would be the Cartesian ontology of substance, attribute and mode.

Note, about these two elements, that the issues they foreground are those in the metaphysics of explanation or conditions, made central in the Dialectic. In the case of the PSR, this is the demand that everything depend on some condition that explains it. In the case of the ontology, think for example of the way in which ‘mode’ is defined as something conditioned by something that explains it, or as ‘that which is in another through which it is also conceived’ (E: 1D5).Footnote ²³ So, on my approach, Hegel would take Spinoza to be broadly correct concerning which philosophical issues are most fundamental.

Further, I think that Hegel takes Spinoza’s way of pursuing these issues to be natural, even if Hegel thinks this will generate contradictions and need to be overcome.

The key for Hegel is specifically Spinoza’s pursuit of those issues in his proof of monism. The proof is worth attention in its own right, but here I will just note the role of the two elements I have highlighted, focusing on steps highlighted in Hegel’s lectures.

An important step, emphasized in Hegel’s lectures, is Spinoza’s argument for ‘P5: … there cannot be two or more substances of the same … attribute’ (E: 1P5; LHP: 20: 173/265).

Note here that the first element (PSR) requires that, if there are distinct substances, then there would have to be a cause explaining this. The second element (the ontology) dictates what kinds of causes would be available. Modes are one possibility, but modes are supposed to be too dependent to explain this; so Spinoza concludes that only attributes can serve as a cause here, and that distinct substances would require distinct attributes.

Another key step is the ban on cross-attribute causal relations (E: 1P2–3). What Hegel emphasizes, in this connection, is P10: ‘Each attribute of a substance must be conceived through itself’ (E: 1P10; Hegel 20: 173/265).

Together these steps can support a big conclusion, by this route: if there is any substance, then by the PSR it requires a cause of its existence. Ruling out cross-attribute causality, and attribute sharing, rules out the cause being another substance. So any substance would have to be the cause of itself, or a necessary being in this sense (E: 1P7). And Spinoza argues this requires any substance to be infinite in its kind (E: 1P8).

But Hegel sees that this falls short of monism; it would, so far, still allow multiple substances, each a necessary being and infinite within its kind or attribute. And Hegel sees that Spinoza needs to rule this out by arguing that, should there be any substance other than God—defined as a substance with all attributes—this would be unexplainable. Hegel highlights this idea, translating Spinoza’s argument in E: 1P14: ‘if there were a substance other than God, it must be explained by means of an attribute of God’ (E: 1P14D; Hegel 20: 174/266).Footnote ²⁴ Since that would require two substances of the same attribute, the existence a substance other than God would be unexplainable, and the only scenario satisfying the principle requiring that everything be explained is supposed to be the existence of God alone, and so substance monism.

We can give in these terms a Hegel-inspired interpretation of Spinoza’s way of arguing for a philosophical system, along these lines:

Spinozist Systematic Argument

1. 1. Fundamental Issues: Issues about the metaphysics of explanation /of conditions and conditioning.

2. 2. Argument Strategy: Draw consequences from (a) demand that everything has an explanation, and (b) the ontology of substance.

Some may expect Hegel’s own systematic argument to fit the Spinoza model, but I think that we should evaluate this question by looking to Hegel’s response to Spinoza in the Logic.

Of particular interest are the discussions of Spinoza in ‘Actuality’, the final section of the second book or division of the Logic, prior to the transition to the culminating Subjective Logic. For here Hegel levels a powerful criticism: Spinoza’s monism involves a denial of all determinacy and finitude. For example, ‘Spinozism is a deficient philosophy’ because, with respect to the one substance, ‘there is no determinateness which would not be … dissolved into it’ (SL: 6:195/472; cf. E: §151Zu).Footnote ²⁵ I do not think that Hegel is saying that Spinoza nowhere contradicts that conclusion. The point is that Spinoza’s argument strategy should force him to it. We can see this, for example, where Hegel says that those who employ the principle that nothing comes from nothing (PSR), whether they know it or not, are forced to Eleatic monism, denying the reality of change, determinacy or anything but the abstract ‘one’:

I think that Hegel has multiple lines of attack here, each worthy of more discussion: I focus here on the simplest and most ambitious, arguing that Spinoza’s own argument should commit him to the denial of the existence of determinate attributes, and so of all the modes dependent on them, leaving nothing but indeterminate substance.

I would defend Hegel’s case by showing how it can rest on a re-working of Spinoza’s argument in P5, about the possibility of distinct substances. The argument rests on the need, if there is distinction, for a cause to explain it. It should be in this spirit to require that, if there are distinct attributes, then there would have to be a cause to explain this as well. And yet Spinoza’s version of the Cartesian ontology would offer nothing that can do this work. Modes cannot serve; for, Spinoza argues, in considering modes as explaining the distinctness of substances, that their dependence renders them incapable of explaining the distinctness of that which they depend upon (E: 1P5D). The alternative allowed by Spinoza’s own demonstration of P5 is attributes themselves as causes. What would it be for the distinctness of attributes to explain the distinctness of attributes? This would have to be the case: to be attribute X is to not be attribute Y, and so on for all attributes. But this is what Spinoza rules out, as in Hegel’s citation of P10: ‘Each attribute of a substance must be conceived through itself’. In an earlier letter, Spinoza had tried defining attributes by ruling this out: ‘By attribute I understand whatever is conceived through itself and in itself, so that its concept does not involve the concept of another thing’ (E: 67).Footnote ²⁷

In sum, Spinoza’s own way of arguing leaves no room for a cause to explain the distinctness of distinct attributes. Hegel’s Logic charges that Spinoza just cheats here: ‘Differentiation occurs with Spinoza quite empirically—attributes (thought and extension) and then modes, affects, and all the remaining’ (21:381/333). Spinoza cannot allow empirical evidence to settle the matter of distinctness, since his anti-monist opponents would happily adduce what they take to be equally strong empirical evidence for distinct substances. So Spinoza cannot allow any case of the distinctness of distinct attributes, and consequently his own principles require him to deny the existence of distinct attributes at all, leaving only the Eleatic one, with no determinate differences.

One sense in which this Hegelian criticism uses Spinoza’s ‘own resources’ (as Hegel requires) should be clear: the point is that the very elements supporting Spinoza’s argument for monism in fact force the elimination of determinacy.

Granted, a certain kind of Eleatic monist might try to retrench here, in just the manner I have worried about all along: she might embrace the conclusion that there is nothing determinate, and reject Hegel’s worries by declaring determinacy itself to be external and indifferent to her Eleatic system. But consider this possible retrenchment from the point of view of Hegel, understood as following Kant’s Dialectic. Part of what makes Hegel think that Spinoza’s steps are natural ones for philosophy is that Hegel takes philosophy to be an expression of reason’s interest in the metaphysics of explicability and complete explicability. Philosophy, for example, exists to ‘procure for reason’s urges the satisfaction it deserves’ (EL: 8:38/26). But Spinoza pursues this same interest in a very specific manner, and one that leads to a destination that can seem perverse given the starting point: it turns out that if we pursue explicability in this way, via the Cartesian ontology and the demand that everything real would have to be explicable, then we end up having to conclude that there can be nothing determinate about which explanatory questions could arise. So, from this point of view, the source of appeal of Spinoza’s starting point—the interest in the metaphysics of explicability—would be itself contradicted by the conclusion to which Spinoza’s way of pursuing that interest forces him.

Hegel seeks to draw a conclusion from this. But it is not the conclusion that philosophy need shift tracks, to make some other issues fundamental, such as broadly epistemological issues about representation and its objects. Rather, we must extend the same tracks further, or to better pursue the same issues within the metaphysics of explanation. It is better not to pursue those same issues by following Spinoza’s specific argument-strategy, drawing consequences from intuitive principles and ontology. It is better rather to attend to the way in which such a natural pursuit generates contradictions, and to draw whatever metaphysical conclusions follow from those contradictions, even if these are initially less intuitive than Spinoza’s principles and ontology.

Part of Hegel’s point is negative: the rejection of the ontology in which substance is basic. More generally, the treatment of Spinoza at the end of the Objective Logic is supposed to bring to a conclusion Hegel’s case for the abandonment of all forms of what Hegel calls ‘the metaphysics of the understanding’. We could also call this a thing metaphysics, in the sense that it makes basic what Hegel calls substance as substratum, supposed to correspond with the subject-place in judgement (see Kreines Reference Kreines2015: chapter 6). Natural as that ontology is, Hegel argues that it leads to unacceptable dead ends. The last of these is supposed to come after investment in supposedly independent things as causes, including Spinoza’s conception of substance as cause of itself (§153An), and of everything else. Hegel argues that it would have to be in the nature of the cause to bring about the effect. What it is to be the cause, then, would end up depending on the effect. The result is ‘reciprocity’ as a ‘vacuousness’ (§156): everything would be equally cause and effect of everything else, in infinite regress, and metaphysics would be left with nothing which is such that appeal to it truly explains anything else.

We can see in these terms why Hegel would take Spinoza as an essential and necessary step, while still pressing a far-reaching disagreement. Spinozism is ‘the foundation of all true further development, but it is not possible to stand pat with it’ (§151Zu). The well-known expression of the point in the lectures on Spinoza is that ‘[w]hen one begins to philosophize one must be first a Spinozist. The soul must bathe itself in the aether of this single substance, in which everything one believed true has perished’ (LHP: 20:165/3:257, my translation). This cannot be an embrace, on Hegel’s part, of a one indeterminate substance—given the criticism of Spinoza. The point is rather that philosophy must focus on Spinoza’s issues, and must think through his powerful argument for monism, in order to free itself from the intuitive thing-metaphysics of the understanding—‘everything one believed true’—and in order move on to better resolve those fundamental issues in very different ways.

But attention to contradiction is supposed to support not just the abstract negation of subtracting intuitive content; rather, specific contradictions are supposed to support determinate negation and so specific positive lessons. The specific defect in Spinozism teaches the need to try to replace Spinoza’s conception of explanatory completeness in terms of the idea of a necessary being as the self-causing ground of reality with something else—with some one something that nonetheless generates from itself difference, or with a metaphysics of self-determination, or freedom, or ‘self-negating negation’ (11:376/472). Metaphysics was left with nothing such that appeal to it could explain anything; it must turn to something self-determining enough to be responsible for something, such that appeal to it can explain.

This last point provides us with a very different account of the systematic relevance of the transition from an Objective to a Subjective Logic. What distinguishes the Objective Logic is a focus on the intuitive thing-metaphysics of the understanding, on substance as substratum, and so on. The Logic aims to get beyond this. So:

And the transition away from the intuitive thing metaphysics still dominant in the Doctrine of Essence—leaving us with the ‘conceptless’ (§156Zu) relation of cause and effect and reciprocity—must give way to the Doctrine of the Concept or the Subjective Logic. Or, ‘This truth of necessity is thus freedom, and the truth of substance is the concept’ (§158). I take this to mean that the Subjective Logic retains the same issue-focus, on the metaphysics of explanation; but it now pursues this focus freed from the old intuitive constraint, free to develop towards a metaphysics of complete explanation in terms not of a necessary being but a kind of free self-determination—ultimately, what Hegel will call ‘the idea’.

It is now possible to fill our model from above in a new way:

Transcendental Dialectic Model of Systematic Argument

1. 1. Fundamental issues: metaphysics of explanation.

2. 2. Argument strategy: expose contradictions, determinate negation produces better theories of the fundamental issues.

3. 3. What distinguishes Subjective from Objective Logic? Subjective Logic pursues the metaphysics of conditions and the unconditioned, but no longer constrained by intuitive commitments like that to the substrata of the metaphysics of the understanding, generating an account of the unconditioned as a kind of free self-determination.

4. 4. What is the root reason why pre-Kantian metaphysics is supposed to fail? Because it generates, within metaphysics, Antinomy contradictions that it cannot resolve.

Now I have focused here on the transition to the Subjective Logic, and it is important to mention that there are crucial further steps to come, closer to the end of the Subjective Logic. The view I have defended elsewhere (Kreines Reference Kreines2015) argues that the end makes clear the focus on the metaphysics of explanation, and turns out even more anti-Spinozist than is yet clear at the transition discussed here. I think the key argument will be this: out of contradictions in mechanism and chemism, Hegel draws the conclusion that an adequate metaphysics of explanation must take teleology as more complete, less conditioned. But the greater explanatory completeness of teleological life turns out to require realization in something non-teleological, something it can use for its purposes. Without the non-teleological realizer, life could not act without getting sucked into the mere conditionedness of mechanism. So this will mean denying also Spinoza’s principle. In order for there to be something completely explicable—the idea, including in the first case life—there must also be things not completely explicable: non-teleological things. So it cannot be the case that everything is explicable. The less explicable must be neither completely a form of, nor in, the idea—they are rather insubstantial, not entirely actual or wirklich, and external. So I think that Hegel’s criticism of Spinoza also leads away from metaphysical monism in important respects.

But here I will not pursue those later parts of the Logic. What is crucial here is that it is thinking in terms of the Dialectic that lets us see how Hegel uses a resource from Spinoza himself, but to lift philosophy higher, into a different system.

V. Kant in the Doctrine of the Concept

The initial problem included how to make sense of discussions of both Spinoza and Kant at the transition point to the Subjective Logic. And the remaining challenge here is significant. Recall Hegel’s complaint that Spinoza fails to generate attributes out of his one substance. In that case, I concluded that Hegel is promising to pursue similar issues, and yet in a different way that does better with respect to those issues. But now come back to Hegel’s comparable complaint, noted above, about Kant: he fails to generate his categories out of the ‘I’ of apperception. Parallel reasoning would seem to suggest the reading that Hegel is promising to use the conclusion of the Logic to argue in a way that pursues fundamentally the issues from Kant’s Analytic, about the possibility of cognition of objects, and yet to execute in a way that does better in this same respect. But then we have all of the problems encountered along the way: if we now say there are two equally fundamental tracks of argument in the Logic pursuing distinct kinds of philosophical issue, then this would be to say that there is no organizing unity, and no systematic argument in this respect. If those tracks are non-fundamental, and organized or given unity by something else, then none of the ways of reading Hegel so far considered has made any progress toward explaining this something else, or how Hegel hopes to argue for a system. And if the Spinoza-inspired track is supposed to be non-fundamental and superseded by an Analytic-inspired track, then we have lost our account of the comprehensiveness of Hegel’s systematic argument, in losing track of a convincing refutation of Spinoza, given the problem of contesting metaphilosophy—and more specifically, of a sense in which Hegel commits to raise Spinozism higher by means of its own resources.

I would argue that the natural, parallel reasoning causing trouble in the last paragraph is mistaken: that the Logic is not, in the end, fundamentally pursuing the issues of Kant’s Analytic, concerning the possibility of representation, and that they are not what unifies its argument into defence of one system. And that means providing a different way to understand Hegel’s complaint about Kant’s execution—one that fits better, I think, with Hegel’s response to Spinoza and the aim of systematic argument. We need not see the complaint against Kant as a promise to do better in the very same respect. The point could instead be this: consideration of the potential for a systematic philosophy based on the issues from the Transcendental Analytic shows this to not work out, given the problem about derivation of the categories from transcendental apperception. What this shows is not that this kind of system can be better executed. It shows rather that a system of philosophy is better built by making fundamental instead the distinct issues within the metaphysics of explanation. The point is then that consideration of a Transcendental Analytic-based system eventually shows this to go astray, and in need of digging down below itself, as it were, and out of itself into another kind of project or system. For such a project cannot rest content with something like formal conditions of the possibility of the representation of objects. Ultimately, it would require also a metaphysical account of how some one, unified, subject might generate diversity out of itself.Footnote ²⁸ This would be not a track-extending but a track-shifting argument: even if we do try to pursue a positive program modelled on deductions from Kant’s Analytic consideration of the possibility of representation, and pursue it systematically, ultimately this forces us to shift, so that we instead recognize as more fundamental the kind of metaphysical issues that Hegel and Spinoza orient their systems around.

Consider in this light a passage in the introduction to the Subjective Logic that takes an interesting turn. The well-known passage begins like this:

So far, this might seem to be Hegel proposing to argue or justifying conclusions in exactly the manner of Kant’s Analytic. But Hegel immediately continues with a familiar complaint about Kant:

So the complaint is that Kant’s way of arguing, by pursuing conditions of the possibility of representation or cognition, prevents philosophy from doing what it needs to do, namely, of revising the store of metaphysical ideas, or revising the metaphysics of the ‘I’. Kant’s ideas remain, Hegel often charges, shaped by the metaphysics of the understanding, and substrata in which properties inhere. Kant’s denying knowledge does nothing to help reconceive the metaphysical ideas.Footnote ²⁹ So I take Hegel’s conclusion to be that proper pursuit of Kant’s own principle should have pushed him to shift tracks and alter his positive project, so that it came to cede the priority of different issues: not issues about deducing pure categories of the understanding, but about the metaphysics of the unconditioned.

Now perhaps some would make a new proposal here, saying that this kind result is just a way of better executing Kant’s Analytic project. The issue-focus in such a program would remain on the possibility of the representation of objects. The conclusions would include the claim that among the conditions of this is a metaphysically self-determining or spontaneous subject. So the program would be organized by issues concerning representation, and the metaphysical conclusions would rest on those underlying it, or more basic considerations.

But I think that, from Hegel’s perspective, this would be more ambivalent than systematic. To see why, think of the full shape of the resulting philosophical program. And think in terms of what I called Hegel’s swimming argument above. So, the first step of such a program would be to reject direct engagements with the metaphysics of explanation, conditions or grounds. The reason we would be supposed to not immediately think about essence or God, and the way these are supposed to have kinds of explanatory priority, would be because (so the story would go) there is something unsafe or worrisome there. The worry might be scepticism about knowledge, or the worry that we might not be able to know God or essence. But the worry could also be more like a kind of semantic scepticism, or a worry that we might not be able to represent at all these objects from pre-critical metaphysics. In either case, the worry would be supposed to force philosophy to change tracks and systematically pursue issues about the possibility of such representation as basic—deriving conceptual conditions of the possibility of this, and so on. But now the new proposal we are considering would add that it turns out that pursuit of that Analytic-based program cannot be complete, and the needed reassurance for the worry cannot be provided, until and unless we step right back into the metaphysics of explanation or reason. If so, then this seems to be a spinning of the methodological wheels, rather than settling on a systematic argument. And then there really was not any reason to abandon direct pursuit of such metaphysical issues in the first place, and no reason to privilege questions about representation of objects, so that the whole Analytic-inspired track drops out as unnecessary. The apparent reason rested on what is now conceded to be an illusion, namely: that we could pursue something else, something meta-theoretical, in some pure form, and get reassurance about the possibility and limits of representation of objects first. I do not think that Hegel’s Logic is ambivalent or confused here: the consistent position is that such hope is just an illusion, like the hope to learn to swim before getting in the water. If so, then the systematic way of making positive progress is to face the inescapable issues about the metaphysics of explanation or the unconditioned, in their full breadth and generality, wherever they arise—whether these concern the subject, or essence or God—and develop these by consideration of internal contradictions into Hegel’s metaphysics of self-determination. Since it is now conceded that systematic philosophy must eventually jump in those metaphysical waters, it should now also be conceded that a unified, systematic approach would be all-in.

Some who are attracted to Analytic-like programs in philosophy will worry that jumping in the water sounds like a reversion to the pre-critical. But this will only seem so if one merely assumes that the divide between the pre-critical and the critical is drawn in terms of neglect and attention to issues about the conditions of the possibility of the representation of objects. I have provided here what seems to me both a better and a more Hegelian way of classifying the possibilities here: Hegel does not understand the modern spirit of critical philosophy, which he intends to develop, as defined by reflection on issues concerning the conditions of representation. He defines in terms of attention to Antinomy-style contradictions, necessarily arising from consideration of issues within the metaphysics of explanation. So, despite what can seem to be (from Hegel’s perspective) a ‘retrograde step’ of Kant’s restriction of our knowledge, still:

Another way to put the point is to say that Hegel advocates jumping in the water, but not swimming the same old stroke. That is, he prioritizes the old issues, but addresses them differently than pre-critical metaphysicians. Instead of a pre-critical drawing of consequences from intuitive principles, as in Spinoza’s use of Cartesian ontology, Hegel advocates attending to contradictions in those old views, and learning from them that the old, pre-critical ways of thinking of substance and the like must be revised. The procedure here is Kantian, but not at all in the Analytic sense of building around consideration of the conditions of the possibility of representation—it is Kantian in the Dialectic sense of reasoning from contradictions within the metaphysics of conditions.

One more reason to think that this is what is going on in the Doctrine of the Concept is to think of the ‘Mechanism’ chapter, for example. I do not think Hegel asks whether the absolute mechanism considered there fails to provide the resources necessary to account for the possibility of representation of objects by a subject. That argument could have been very short. But this is not what Hegel argues. He is here still, and finally fundamentally, pursuing the old issues, like Spinoza’s. Hegel takes absolute mechanism seriously as a treatment of such issues, concerning the metaphysics of explanation. And he seeks to refute it in its own terms—to meet it, on its own grounds. What he is doing is working through these issues, trying to show that any take on them generates contradictions ultimately resolved only with a metaphysics of self-determination. Of course, this eventually leads into a penultimate ‘Cognition’ chapter, but I would argue that this pursues metaphysical issues about greater self-determination than covered in the ‘Life’ chapter: the experiment considered turns out to rely on a metaphysics of ‘the good’ and ‘the true’ (§233), and in ways requiring revision in the final chapter on ‘The Absolute Idea’.

Now some might lodge a final objection. Gabriel has advanced it in a useful form, raising the worry that my Dialectic-focused ‘reading ignores the fact that Hegel does not avoid epistemology’ (Gabriel Reference Gabriel2016: 204, fn 26). Here I have tried to argue that consideration of the problem of arguing for a system highlights not questions about which issues are included and which avoided, but rather questions about which kind of issues is prior, or taken as fundamental. My proposal is that Hegel bases his project on the metaphysics of the Dialectic, without needing to ignore epistemology. There are two senses in which it is included:

The first sense is that Hegel pursues broadly epistemological issues in a track-shifting argument that has been my focus in this section. At some points, like the beginning of the subjective logic, Hegel considers the possibility of a systematic philosophy focused on such broadly epistemological issues. But he does so in order to show that such a system is eventually forced out of itself and into a project organized by the issues in the metaphysics of explanation.

To see a second sense, consider the very simple idea of organizing a system around something like evil demon scepticism from the Meditations. We might first reject competing views on grounds that they cannot resolve this scepticism without begging the question, by appealing to something placed in doubt by thoughts about an evil demon. We would have to conclude with a better answer to scepticism, meeting these high standards.

But of course many other kinds of philosophical programs will reject the standard of the evil demon. They will not take this issue as fundamental, but will have some other metaphilosophy. For example, say we instead hold that the problem of representing objects at all, let alone correctly, is more fundamental. We could argue that evil demon scepticism about knowledge founders in so far as it assumes but cannot explain the possibility of representing the objects of which it questions the possibility knowledge. This rival kind of system is not thereby precluded from addressing issues about knowledge. But the standards it faces would be very different: it would not have to justify the possibility of knowledge without presuming anything the demon places in doubt. The high standards here would come with the alternative issues taken as fundamental: after rejecting evil demon scepticism for failing to explain representation without begging the question, by assuming representational notions in explaining this, this new system will have to do better on that score—better in accounting for the possibility of the representation of objects. This is part of why it is so important to make explicit what metaphilosophy is driving the systematic argument, because this sets the standard for what does and does not beg the question.

Similarly, I think that Hegel can argue that both kinds of broadly epistemological programs just mentioned miss the fundamentality of yet other issues still, about the metaphysics of conditioning and the unconditioned. To say this is not to say that Hegel must ignore knowledge and representation. It is just to say that he is not building or organizing his system in those terms. So he can address knowledge, for example, without having to accept the terms of evil demon scepticism and somehow resolve it. Similarly, consider the Kantian question about representation:

Hegel can treat this as important, not because of any specialness of the topic of representation. Rather, just in so far as it is another question about grounds or conditions, and a natural topic in a metaphysics of explanation. And this would make a big difference in the standards he would have to meet. There is, after all, in the Subjective Logic, an initial orientation around reflection on forms of judgement (§166ff.). It is right to say that Hegel gives epistemological issues from the Transcendental Analytic, such as those concerning ‘objective validity’ (SL: 12:18/518)Footnote ³⁰ , a Kantian answer in terms of the transcendental unity of apperception—so long as it is added that Hegel takes this answer not to be able to stand on its own, but to raise metaphysical issues, or to reveal a priority of metaphysical issues, the resolving of which is what unifies the broader systematic argument.Footnote ³¹

Finally, I can bring this to a conclusion in these very terms. The problem at issue has been how to explain the aim for a systematic argument. The discussions of Spinoza and Kant at the transition to the Subjective Logic can make this harder still. But thinking of Kant’s Dialectic as Hegel’s model offers what I think is the best way forward: This will mean that Hegel sees Spinoza’s defence of monism as focused on the right issues, concerning what I call the metaphysics of explanation. But, also, Hegel can argue that Spinoza’s case goes awry from within, and the track that considers these issues needs to be extended into a Subjective Logic focused on the metaphysics of self-determination. With respect to Kant, this will mean that the main systematic relevance of Hegel’s drawing concepts from Kant’s Analytic is what I call a ‘track-shifting’ argument: showing that pursuit of the Analytic program as a system would eventually have to shift tracks to something more forced on the metaphysical issues highlighted in Kant’s own Dialectic. Finally, settling in this way the metaphilosophy or fundamental issues organizing Hegel’s systematic argument promises to make clearer the stakes he imposes on himself: The basic challenge is not to accept the terms of evil demon scepticism and reply to it. The basic challenge is not to account for the possibility of representation without employing representational notions. In my view, the basic challenge would be addressing the broadest issues about grounds or conditions and the metaphysics of explanation, in a manner that—while it must raise dialectical contradictions—can eventually give a more satisfying resolution of those contradictions in a metaphysics of self-determination.Footnote ³²