Let me begin by expressing my admiration for Kreines’s new book. It offers a wonderfully engaging treatment of fundamental issues in the history of German Idealism, focusing on the relation between Kant and Hegel on the question of metaphysics, and aiming to solve a central puzzle concerning that relation: how could Hegel claim to be a properly post-Kantian thinker on the one hand, while also committing himself to the centrality of metaphysics on the other? Of course, many have tried to solve this problem before Kreines, where on one extreme it has been argued that Hegel in the end is not really post-Kantian but is instead just setting Kant aside and going back to metaphysical business as usual; and on the other it has been argued that Hegel is so fully Kantian that in his hands metaphysics takes no more than an attenuated ‘transcendental’ form of the sort Kant himself wanted to allow. But neither of these options is particularly attractive, the first because it entirely ignores Kant’s impact on Hegel and his time in a way that seems implausible, and the second because it underestimates the critical nature of Hegel’s response to Kant, together with Hegel’s characteristic desire to combine the ancient with the modern, where it is the metaphysical robustness of the former that he seems so much to admire. In this context, therefore, Kreines’s book aims to satisfy a real need in our understanding of Kant’s relation to Hegel.

Another reason for me to be excited about Kreines’s book, I confess, is that I find myself in agreement with many of the views he defends, both interpretatively and philosophically. In particular, we both agree on the significance to Hegel’s metaphysics of his conceptual realism, or what Kreines calls his ‘concept thesis’: ‘the reasons which explain why things are as they are and do what they do are always found in immanent “concepts” (Begriffe), akin to immanent universals or kinds (Gattungen)’ (Kreines Reference Kreines2015: 22; all further unattributed references are to this work). We also agree that it is this conceptual realism that is at the heart of Hegel’s idealism – a claim that will only seem oxymoronic if by ‘idealism’ one thinks only of Berkeley and not of Plato. For, on this account, to be an idealist is to be a realist about universals as opposed to being a nominalist, where the nominalist thinks that such conceptual structures are imposed by the mind on reality and thus depend on us, which the realist qua idealist denies.Footnote ¹ As Kreines notes (22, note 37), this puts both of us (alongside some others, such as Kenneth Westphal and Willem deVries) in broadly the same camp. And I think we would both also agree that this approach leaves Hegel in a philosophically rewarding place, where he can potentially contribute to debates between various forms of contemporary realism and contemporary nominalism when it comes to issues surrounding natural kinds, laws, essences, normativity and much else besides.

Nonetheless, while I think that Kreines and I end up in fundamentally the same position, we choose to get there in rather different ways. I am not sure that this means we disagree exactly, as it seems to me that both routes are compatible with one another, and each may have its own philosophical and interpretative virtues. But I do think they are somewhat different, as does Kreines:Footnote ² so it is that difference I would like to explore in what follows.

The divergence between Kreines and I can be traced back to two different ways in which Kant himself presents his argument for transcendental idealism in relation to metaphysics in the Preface to the second edition of the Critique of Pure Reason, where the first might be called the argument from synthetic a priori knowledge, and the second might be called the argument from the dialectic of reason.

The first argument starts from the well-known problem of synthetic a priori knowledge which Kant raises, particularly as that relates to metaphysics, and argues that it can only be resolved after we have made the Copernican turn and thus within the resources of transcendental idealism (cf. Kant 1781/Reference Kant1787: Bxvii–xviii). But, as Kreines emphasizes (16–17), Kant then remarks that one might reasonably worry that this approach leads to a ‘very strange result’, which is that ‘we can never get beyond the boundaries of possible experience’, where this is ‘strange’ because ‘it appears very disadvantageous to the whole purpose with which the second part of metaphysics concerns itself’ (Kant 1781/Reference Kant1787: Bxix), which is precisely to get beyond such boundaries and seek knowledge of the unconditioned (Kant 1781/Reference Kant1787: Bxx). But, Kant argues, in fact the tables can be turned, and what seems to be a threat to transcendental idealism can be used to provide a ‘cross-check’ (Gegenprobe) or corroboration of this position and thus ‘the truth of the result of that first assessment of our rational cognition a priori, namely that such cognition reaches appearances only, leaving the thing in itself as something actual for itself but uncognized by us’ (Bxx). The ‘cross-check’ seems to proceed as follows: reason ‘necessarily and with every right demands’ that reality contains an unconditioned, as this is required in order to complete the series of conditions in what we find around us; but if we assume that this unconditioned behaves in the manner of things as we experience them and so apply our usual concepts to it, we cannot avoid this unconditioned proving contradictory and thus unreal; but if we conclude that it is unreal then we cannot use it to guide theoretical inquiry in the way that reason requires, so we must instead say that it does not behave as the world of experience does, and so must lie outside or beyond that world and thus beyond our knowledge, which is precisely what the transcendental idealist claims. Thus, ironically enough, what may initially have seemed like a threat to transcendental idealism in fact serves to corroborate it, by enabling Kant to ‘show that what we initially assumed only as an experiment is well grounded’ (Kant 1781/Reference Kant1787: Bxx-xxi). There is therefore no alternative to Kant’s cognitively constrained metaphysics, which allows for the possibility that there are unconditioned objects about which we know nothing further from a theoretical point of viewFootnote ³ and so which confines its synthetic a priori inquiries to the realm of appearances. The only other option is a metaphysics that would undercut itself entirely, as in attempting to do more it would be forced to hold that such unconditioned objects are contradictory and so non-existent, leaving itself nothing outside the conditioned to wax metaphysical about, thereby making itself redundant after all, while also threatening the vital role that the idea of the unconditioned plays within our inquiries.

We thus have two distinct but related arguments for transcendental idealism in Kant: the argument from synthetic a priori knowledge, and the argument from the dialectic of reason. The next question for Kreines and me is how Hegel gets from Kant’s position to his own conceptual realism, and whether this is best seen as responding to the first or second of these arguments.

Now, I confess that when I have thought about the route from Kant to Hegel on these issues, I have largely concentrated on the first argument, where roughly I conceived the route to Hegel as follows: Kant claims that we can only account for synthetic a priori knowledge by taking the Copernican turn and thus endorsing transcendental idealism, by saying that this knowledge is made possible by the conditions that necessarily govern our experience. But Kant then seems to owe us an explanation of how we know about those necessary conditions, which seem equally synthetic a priori. However, it would seem that the transcendental turn cannot provide an explanation at this level, as it is not possible to explain such knowledge by appeal to the necessary structures of our experience, on pain of an infinite regress. Kant therefore can’t appeal to transcendental idealism as offering a uniquely satisfying explanation of synthetic a priori knowledge, as it does not account for the synthetic a priori knowledge it itself relies on to offer the explanation it does. On the other hand, if Kant does offer some sort of different explanation for this knowledge, then the metaphysician can claim that they know their synthetic a priori truths about the world the same way, so rendering transcendental idealism redundant. This sort of objection to Kant was raised fairly early on by his critics (such as Schulze and others), and is also echoed in Hegel’s work, offering him an immanent critique of the Kantian account of synthetic a priori knowledge.Footnote ⁴

Once this account is exploded, however, some other account of how metaphysical knowledge is possible is still required. Now, Kant had himself considered other possibilities in his famous letter to Herz of 1772, where it was precisely dissatisfaction with those possibilities that had in the end led him to take the transcendental turn. One way to see how Hegel comes to adopt conceptual realism, however, is to see him as finding here a further option that Kant had overlooked. In Kant’s letter, Kant raises the problem of how ‘intellectual representations’, which are not themselves caused by objects, nonetheless come to agree with them, ‘since the agreement has not been reached with the aid of experience’, so that ‘this question, of how the faculty of the understanding achieves this conformity with things themselves is still left in a state of obscurity’ (Kant Reference Kant1999: 10: 131). He then considers two traditional answers, both of which appeal to something divine as a deus ex machina to help us out: the first (which Kant associates with Plato and Malebranche, and which he names the Hyperphysical Influx Theory), involves some sort of participation in a divine understanding of the world, while the second (which Kant associates with Crusius, and which he names the Pre-established Intellectual Harmony Theory), involves God implanting these ideas in us and ensuring that they conform with the world around us. Either way, Kant argues, these accounts are hopeless, as ‘the deus ex machina is the greatest absurdity one could hit upon in the determination of the origin and validity of our cognitions’, as it is guilty of both ‘vicious circularity’ and also ‘all sorts of wild notions and every pious and speculative brainstorm’ (Kant Reference Kant1999: 10: 131). Kant then proceeds to sketch the outline of his ‘critique of pure reason’ as a response to this difficulty, although without exactly telling Herz how it might solve the problem he has raised, perhaps because he didn’t yet know himself. Nonetheless, it is easy to see how he might have hoped that the developed position of transcendental idealism could provide an alternative to the two positions he had rejected here.

However, once Hegel had seen that Kant’s transcendental idealism also fails, it is then natural to view him as looking for an alternative account, where conceptual realism appears to be a viable option. Put very roughly: Kant had rightly rejected the pre-established harmony view, because it has introduced God as an external and arbitrary ‘third thing’ to ensure that the world and our concepts operated successfully together. But if we think of the world itself as conceptually structured, then we can explain how mind and world function together, but without appeal to this deus ex machina: reality is conceptually structured, while the mind thinks in terms of concepts, so it is hardly surprising that the latter gives us insight into the former.

To be sure, Kant or a Kantian might object to this view on grounds of circularity, just as Kant objects to the views he criticizes. He isn’t explicit about this, but presumably the circularity Kant has in mind is that the explanations themselves both involve ‘intellectual representations’, namely ideas about God, so how can they successfully use an appeal to God to explain the possibility and validity of such intellectual representations? Unless we already know how a priori metaphysical knowledge is possible, how can we legitimately appeal to claims based on such knowledge to explain this possibility?

However, Hegel’s well-known response to this kind of Kantian doubt in the Introduction to the Phenomenology and elsewhere would seem to be relevant at this point: we don’t have to have some prior knowledge that our cognitive capacities are successful before deploying them, any more than we have to first learn to swim before we get into the water. Of course, if we have some reason to think those capacities are problematic (as Kant does in his letter to Herz), then we should question them; but Hegel’s conceptual realism is meant to dispel such worries, and to make our knowledge here unmysterious. Beyond that, the fact it does so by relying on such knowledge is not relevant.

More potentially troubling, therefore, is Kant’s objection that prior accounts give rise to ‘speculative brainstorms’, namely unconstrained and implausible forms of metaphysical theorizing which lack any justification, where this objection might then also be applied to Hegel’s position. But of course, Hegel devotes much of his time to showing why his theory of conceptual realism is plausible and not a mere ‘speculative brainstorm’, on grounds that we cannot explore in any detail here.

We have seen, therefore, how by taking Kant’s argument from synthetic a priori knowledge, and offering an internal critique of the way in which Kant tries to argue for transcendental idealism on the basis of that argument, we can arrive at Hegel’s conceptual realism as a plausible result. We have thus found one route from Kant to Hegel, which takes the former’s position seriously, but goes beyond it in a metaphysical direction that seems both defensible and well-motivated.

But I now want to consider Kreines’s story, which I think is different, and which begins from a different place—namely Kant’s second argument for transcendental idealism, the argument from the dialectic of reason.

Kreines presents that argument as follows (148):

Kreines presents Hegel as responding to this Kantian position as a mixture of what he calls inflationary and deflationary elements, where Hegel here contrasts to a pure inflationist or a pure deflationist:

The pure inflationist would accept E1 and E2, and reject E3; but Kreines thinks this would commit Hegel to a kind of hyper-rationalism, able to deduce the existence of Krug’s pen, which is something Hegel himself rejects. A pure deflationism would reject E1 and the call for explanation that it embodies, thus saving our epistemic blushes; but Kreines argues that Hegel is clearly more committed to the idea of explanation than that kind of approach would allow.

Instead, Kreines suggests that Hegel takes a middle way, of endorsing E1 and the idea of explanatory completeness (hence moving towards inflationism), but challenging the conception of explanatory completeness as requiring some sort of substance metaphysics (hence moving towards deflationism), where once the goal of explanation is no longer seen in these terms it promises to escape the antinomies and thus bring metaphysics as the search for explanations back within our epistemic capacities.

On this account, then, the key issue for Kreines is the question of explanation and whether Hegel can plausibly show that Kant’s conception of fundamental explanation as turning on some sort of unconditioned substance is misplaced, or whether we can satisfy reason’s explanatory needs in another way. If so, Kant can be allowed to reject the former model as unknowable on the one hand, without impugning our ability to grasp the latter model, and so pursuing metaphysics on a different basis, as a search for explanatory satisfaction that takes us in a new direction and away from the antinomies that Kant puts up as a block on our inquiries. On this view, then, Kant’s critique of metaphysics is right as far as it goes, but is strictly limited: namely, that the search by rationalists for an ultimate explanatory ground as they understood what this required is hopeless; but this is not because our cognitive capacity is limited, but because they misunderstood the nature of what the search for explanation requires, and once we get this right a more successful form of metaphysical enterprise can be revived.

How can Hegel show that the traditional rationalist’s pursuit of explanation is misconceived, and so can be abandoned to Kant’s critique without any genuine cognitive price being paid? On Kreines’s account, Hegel’s method here is to show that the rationalist’s explanatory search is governed by the understanding’s commitment to subject-predicate judgements, such that the unconditioned it is searching for is thus an underlying subject or substratum – but precisely because of this, it is a kind of indifferent and propertyless surd which can’t explain anything. It is thus unproblematic to abandon this to the realm of unknowingness, as what is being given up is in fact something that (like Kant’s thing-in-itself) can’t be used to explain anything at all, and so is of no genuine interest to reason after all. For Kreines, this is where the understanding for Hegel does its pernicious work: in committing us to a subject-predicate model on which the search for the unconditioned takes on this form of an underlying substance or substratum, but where the understanding does not notice it has thereby pushed us into a conception of no explanatory value. Kant had an inkling of the problem here, but misdiagnosed the disease and so applied the wrong cure: instead of limiting knowledge, he should have seen through the project altogether, but because he was so keen to stress our cognitive limitations he was cut off from seeing another way in which the interest in explanation could be pursued.

But how does this get us to Hegel’s conceptual realism? The key is explanation once again. For once we have seen through the explanatory project of traditional rationalism—namely the search for the unconditioned as substratum—we can see that there is another way in which we search for and provide explanations that are satisfying to reason, namely in terms of universals relating to individuals, such as laws and kinds, which according to Hegel play a vital role in understanding ourselves and the world around us. In some ways, of course, the kinds of explanations provided in this manner are less glamorous than the kinds of explanation sought for in traditional rationalist metaphysics, such as self-causing causes (or a turtle with a jet pack, as Kreines nicely puts it (43)) that have been espoused by some. But Kreines takes Hegel to have shown that these sorts of explanatory projects are a busted flush, as in the end they don’t explain anything. So we should be completely content with the explanations that a metaphysics of conceptual realism gives us instead, as this too will bring order to what otherwise may seem a chaotic universe, but in a way that also takes in its stride that not everything in that universe needs some further grounding to be fully satisfactory to reason. Of course, some have equally challenged the ability of appeals to laws and kinds to do any real explanatory work, and so sought to replace conceptual realism with conceptual nominalism. But Kreines works through in commendable detail Hegel’s efforts to show that this is a mistake, in a way that shows how he might successfully contribute to various debates on this issue that are being carried on today.

We have thus seen, in outline, how Kreines traces an argument in Hegel to conceptual realism that starts not from Kant’s problem of synthetic a priori knowledge, but from Kant’s dialectical challenge to metaphysics. In what remains I now want to consider the relation between these two arguments, from the perspective of someone who agrees with Kreines about where we end up, but who has tended to favour the first route over the second. Are these two arguments compatible, and even if they are, is there some reason to prefer Kreines’s approach to the other one or vice versa, or should we friends of conceptual realism just embrace both?

On the issue of compatibility, as far as I can tell there isn’t a problem here: the conclusions of both arguments seem to be the same, and no premises of one seem to contradict the premises of the other. But still, even if they are compatible, there might be reasons to prefer the one argument over the other, as being more interpretatively or philosophically satisfying in some way. Judging by some comments by Kreines in his book, I think this is his view, although he doesn’t go into the question in any great detail. Here are some relevant considerations that might be worth discussing further, starting with interpretative issues and moving on to more philosophical ones:

(i) For an Hegelian, one important consideration would of course be which approach is more ‘internal’ or ‘immanent’ than the other, as this is meant to be a virtue in any Hegelian argument. But on this criterion, it would seem that both approaches pass the test, as both offer what can be seen as internal critiques of the Kantian position.

(ii) But perhaps one advantage of Kreines’s approach is that it also incorporates a critique of traditional rationalist metaphysics, by building on Kant’s position in the Dialectic but developing this in an Hegelian way, as a critique of the notion of substance that such metaphysical views deploy. It is this element of Kreines’s position that he uses to distance himself from various Spinozistic readings of Hegel proposed by Beiser and others, where Kreines argues that to read Hegel in this way is to fail to see how much he wanted to overcome this kind of substance metaphysics (cf. 211). At one point, Kreines seems to contrast this critique of Spinozism with the alternative approach, where he sees the attack on the unconditioned being based more on epistemological considerations rather than explanatory ones, that here we would overstep our cognitive capacities, where he rejects such concerns as the fundamental ones (172). This relates to a general theme of Kreines’s book: that we should be wary of ‘epistemology first’ arguments in Hegel, that first make claims about our epistemic capacities and then try to tailor our metaphysical views accordingly, which is what seems to be happening here (cf. §0.3). Kreines’s concern seems to be this: by focusing on the epistemological worries raised by metaphysics, and the Copernican revolution’s response to them, we run the risk of occluding what might be called the ‘explanation first’ approach that he favours, where it is then the explanatory rather than epistemic deficit of traditional rationalist metaphysics that comes to the fore.

Now, I am certainly sympathetic to some of Kreines’s worries about ‘epistemology first’ strategies, and for what I take to be good Hegelian reasons: namely, those Cartesian strategies which trade on what Peirce would call ‘empty doubts’ and which Hegel criticizes in similar terms in the Introduction to the Phenomenology and elsewhere. But I am not sure that Hegel would put epistemological concerns raised by metaphysics in this boat,Footnote ⁵ as he seems perfectly prepared to concede that Kant was right to see something genuinely puzzling about the sort of synthetic a priori knowledge that metaphysics appears to involve. Of course, ultimately Hegel wants to dismiss those Kantian concerns as groundless and based on various questionable epistemological and metaphysical presuppositions; but I would argue that he partly wants to introduce his option of conceptual realism in order to make these claims, so that it is wrong to set epistemological issues aside altogether here.

And conversely, I think that someone who sees Hegel as introducing his conceptual realism to resolve Kant’s problems in the letter to Herz can still agree with Kreines that Hegel’s objection to a metaphysics of substance of a rationalist kind need not only be an epistemological one. This suggests, then, a kind of division of labour between the two approaches: when it comes to metaphysica specialis, the fundamental route to conceptual realism is via the ‘explanation first’ argument, showing how conceptual realism can replace the search for explanation via an underlying substance as the unconditioned without falling into the snares of Kant’s dialectic. But when it comes to metaphysica generalis, and its search for a science of ‘being qua being’, the epistemological questions that Kant raises against such a science need to be dealt with by an appeal to conceptual realism, so here epistemological considerations do play a role.

(iii) However, Kreines may still reply that I am missing something of vital importance here, which is that on his account of metaphysics, this conception of metaphysica generalis also drops out of the picture to a significant degree, where this is again a consequence of focusing more on Kant’s second argument in the B Preface, rather than the first.

The issue here comes back to what is of course the elephant in the room: namely, what is metaphysics, and, relatedly, how does it differ from other forms of inquiry, if at all? On one approach, which Kant arguably inherits and which drives his ‘how is synthetic a priori knowledge possible?’ question, while metaphysica specialis deals with certain peculiar kinds of metaphysical objects (such as God and the soul), metaphysica generalis deals with the nature of being, and in particular with the necessary conditions for being and thus what can and cannot exist in this sense, so that this branch of metaphysics becomes ontology.Footnote ⁶ As traditionally conceived in this way, it is easy to see how metaphysics differs from the empirical sciences, because the latter just concerns the conditions of the world around us, whereas metaphysics concerns the conditions for there being any world at all. And of course it is precisely the epistemological immodesty of this metaphysical project that troubles Kant, and leads him to urge on us the transcendental turn whereby ‘the proud name of an ontology, which presumes to offer synthetic a priori cognitions of things in general in a systematic manner (e.g., the principle of causality), must give way to the more modest one of a mere analytic of the pure understanding’, whose ‘principles are merely principles of the exposition of appearances’ (B303). Understood in this way, metaphysics pretty quickly generates epistemological concerns which then need to be addressed accordingly.

But Kreines has a rather different view of metaphysics, which is not a search for the necessary conditions of being, but an attempt to address the ‘why’ of things, and thus a search for explanations (cf. 3). As Kreines immediately notes, however, this may seem to render metaphysics redundant: don’t the natural sciences also seek for explanations, and arguably do so in a better way than any armchair metaphysician? Of course, when science was in its infancy, this was something metaphysicians did, in some cases well and in some cases badly: but surely now the whole business can be left to the professionals, and we can pack up and go home?

Kreines, however, offers a response to this worry on Hegel’s behalf immediately after raising it:

Kreines argues, therefore, that we can see Hegel’s project as one of this kind, which again reinforces the centrality of Kant’s second approach in the B Preface, as it is on explanatory rather than epistemological issues that it turns. Indeed, one attractive aspect of this approach is that nothing particularly fancy by way of epistemology seems to be required of it: the metaphysician is simply someone who concerns herself with questions of explanation—of what it is to explain something, how such explanatory questions arise, what the limits to explanation might be, and so on—in a way that operates at a meta-level that may not be the explicit concern of the scientist, but just appeals to ordinary epistemological resources to do so, such as conceptual analysis, appeals to common sense, paradigm cases, the history of science, and so on.

Now, I certainly would not want to deny that this issue of explanation is a vitally important part of Hegel’s project. Indeed, I have argued that it underlies Hegel’s striking claim that ‘The idealism of philosophy consists in nothing else than in recognizing that the finite has no veritable being’, and in this sense ‘Every philosophy is an idealism’, as it always looks for some further way to explain the finite, from Thales’s appeal to water onwards.Footnote ⁷ Nonetheless, Kreines might say, this project can remain relatively modest, in so far as it just seeks for explanations for how things are round here: so, our world contains planetary systems, and chemical processes, and living beings, and to properly understand and explain these Hegel might rightly claim that we need laws, kinds and other immanent universals, for which an inference to the best explanation provides sufficient warrant even if it goes beyond what is given to us in mere experience. But none of this raises the problem of metaphysical knowledge as raised by Kant, as none of this commits us to any claims about being as such, as it is clear that there could have been worlds without plants, chemical processes or living beings, and hence without the explanatory structures that enable us to make sense of them. On this basis, then, we can defend the reality of laws and natural kinds as actually pertaining to our world, in the face of metaphysical and scientific attempts to explain and understand that world without positing them, but without needing to make the grander and more ambitious claims that such laws and kinds are necessary conditions for all possible worlds, and thus without raising or needing to address Kantian epistemological concerns about metaphysics qua ontology at all.

However, while this answer certainly has its attractions, not least in the modesty it seems to imply, I feel it doesn’t capture the scale of Hegel’s ambitions. One worry is interpretative, and may be put in terms of the notoriously tricky structure of The Science of Logic. For, while this modest approach may seem well-suited to the ‘Objectivity’ section of volume two, which contains ‘Mechanism’, ‘Chemism’, ‘Teleology’ and ‘Life’, of which Kreines provides excellent readings, it may seem to fit less well the earlier sections of Being, Essence and the Concept, which arguably do offer more by way of a traditional ontological project, concerned with being qua being.

Moreover, there are two philosophical worries that can be raised. The first is that it seems to me that the focus on explanation which Kreines emphasizes actually dovetails naturally with the ontological project of metaphysica generalis, precisely because in the end ‘why’ questions are best answered by the kind of modal claims it embodies: namely, why are there objects with properties (for example)?; answer, because otherwise there could not be anything at all, as this is the only stable form for being to take. And that leads us back to the problem of synthetic a priori knowledge, given the universality and necessity of this kind of claim.

A second related philosophical worry is this: On Kreines’s account, the heart of Hegel’s argument for laws and kinds is that without them the world around us would be less readily comprehensible to reason, so we are entitled to posit them to render the world around us more intelligible. But that raises the question: why think the world is rationally ordered?; why not think it works in a way that is opaque to reason? Kreines might well dismiss the question as a kind of empty scepticism. But it could also be raised as a request for explanation of its own (which in a way takes us back to the letter to Herz): given that the world is such as to be intelligible reason (so that scepticism can be dismissed), how come this is so? The ontologist can provide an answer to this, if conceptual realism is treated as an ontological position; but if it is treated in Kreines’s more modest way, it is not clear that it can be addressed in a way that is itself explanatorily satisfying. And if conceptual realism does provide an explanation at this level, then what is to prevent us inferring to its truth, as we do with explanatory arguments at other levels? Of course, as with these other arguments, conceptual realism is vulnerable to the charge that the explanation is bogus or empty, or that it is a mere contingency lacking in any explanation, but this would need to be argued out and there doesn’t seem to be any such argument for it in Hegel, as there are in other cases where he wants to check our explanatory ambitions.

Kreines might reply, however, that the joy of Hegel’s position is precisely the way in which he shows that the search for explanations need not push us in the metaphysically ambitious direction, that nothing is missing from our understanding of the world if we leave room in it for greater contingency than this seems to require, so there is no ‘explanatory gap’ if we just show that there are objects with properties (or whatever), not that there somehow had to be in a metaphysically heavy sense?

However, there are two difficulties with this response. The first is that Hegel often does seem to be tempted into making the more ambitious claim. Thus, for example, his handling of all the principle ‘all determination is negation’ seems to introduce the principle of negation as a principle of being: that there could not be pure being, as an undifferentiated Parmenidean one. Likewise, his treatment of the Concept seems to be ontologically ambitious, in claiming that there must be universals as well as individuality and particularity. The second difficulty is that when it comes to contingency itself, his claim seems to be (as Henrich and others have emphasized) that such contingency is necessary, so there is ultimately no metaphysical modesty here. It is because of this, therefore, that Hegel accepts the prima facie force of Kant’s epistemological concerns, and which he takes his own conceptual realism to address.

Now, at this point it may seem that there really must be a parting of the ways between Kreines’s approach and my own: for it may seem that this approach I am offering is committed to just the kind of foundationalism that Kreines thinks Kant’s dialectic taught Hegel to reject, namely the view that ‘there is something—whether transcendent or immanent—that depends on nothing while being the reason for itself and for everything real’ (5). I have argued that ontology and inquiry into the necessary conditions for being are precisely a search for a satisfying explanation of this sort, as it then no longer makes sense to ask why the world has certain features that it does, as it could not be otherwise. In some sense, then, this looks like a foundationalist project, which Kreines thinks Hegel escapes.

However, perhaps again an irenic response is possible. For, of course, this kind of foundationalism found in metaphysica generalis is not the same sort of foundationalism one finds in metaphysica specialis, with a special class of self-grounding objects, such as God or the world-whole. Indeed, this is precisely why critics such as Spinoza thought that metaphysica generalis could not answer the ultimate explanatory question, namely why is there something rather than nothing, because all that it can explain is if there is something, then it must be this way rather than that. But I have argued elsewhere that Hegel saw this explanatory question as empty,Footnote ⁸ hence hanging on to foundationalism of a more modest form, which I think Kreines need not object to when (rightly) objecting to the foundationalism of metaphysica specialis.

To conclude: My suggestion here is that Kreines’s approach and the approach that focuses more on the problem of synthetic a priori knowledge (which I have generally adopted) are compatible with one another, and in fact complement each other. The approach that Kreines offers which focuses on explanation can be used to undercut the need for any metaphysica specialis of transcendent objects, in a way that marks the influence of Kantian modesty on Hegel’s engagement with metaphysics. But the danger is that this ignores the way that this very commitment to explanation leads Hegel to offer a metaphysica generalis, which raises Kantian epistemological issues which Kreines tends to downplay. But the good news is that we can both agree that where Hegel ends up is with a distinctive kind of conceptual realism, for which there are now not just one but two complementary arguments to explore.