Self-Explanation as the Basic form of Explanation - Oxford Scholarship Jump to ContentJump to Main Navigation
Freedom and ReflectionHegel and the Logic of Agency$

Christopher Yeomans

Print publication date: 2011

Print ISBN-13: 9780199794522

Published to Oxford Scholarship Online: January 2012

DOI: 10.1093/acprof:oso/9780199794522.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2015. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 27 August 2016

Self-Explanation as the Basic form of Explanation

Self-Explanation as the Basic form of Explanation

Chapter:
(p.75) 4 Self-Explanation as the Basic form of Explanation
Source:
Freedom and Reflection
Author(s):

Christopher Yeomans

Publisher:
Oxford University Press
DOI:10.1093/acprof:oso/9780199794522.003.0004

Abstract and Keywords

This chapter reconstructs Hegel's response to doubts about the reality of free will grounded in the requirements of explanation by exploring his theory of explanation in the Logic's treatment of the category of ground. It shows that Hegel sees the basic form of explanation holistically in such a way that all explanation is partially self-explanation as a three-term relation between explanans, explanandum, and conditions. The way that Hegel develops the role of the conditions within self-explanation is shown to internalize an anarchic externality within the ‘self’ that is self-explanatory. This internalization partially destabilizes the explanandum as an orienting point of reference while simultaneously providing it a content that makes it a true locus of responsibility. Throughout it is shown that Hegel's conception of the grounding relation is essentially one of expression.

Keywords:   Hegel, explanation, erotetic analysis, ground, logic

§4.1: Ground as Expression

As a category of the Doctrine of Essence, ground is a way of understanding what it is to be an essence, namely as constituted by a certain functional role in explanatory or justificatory relations. A thing or fact is explained by its own essence, with which it is identical. As has come out already, Hegel's concept of essence is closely connected with his concept of reflection. If the categories of the Doctrine of Being answer the question about what a thing is in terms of immediate qualitative and quantitative properties, the categories of the Doctrine of Essence answer the same question in terms of higher-order principles that serve to answer why questions about the specific qualitative and quantitative properties.

Here we see Hegel working through certain tendencies in Leibniz's metaphysics. Hegel holds that the principle of sufficient reason just expresses the commitment that a fact is not known in its immediate presentation but in its deeper connections as an expression of logical relationships (WL293/SL446). Leibniz holds that the way to understand the subject-predicate form of judgment is to think of the predicate as being contained within the subject. As a result, the complete notion of a subject contains all of its relations and all of its actions or behaviors in the past, present and the future. This means that, strictly speaking, all relations are internal because all relations constitute the essence of the substance. But if all of these relations are contained eternally within the monad, then it must be windowless and interactions between substances are illusory, the confused perception of what are truly logical and static relations. The important thing for Hegel is the fact that Leibniz's view appears as a potential solution to a problem introduced by the fact that explanatory relations seem just as much internal as external. But rather than privilege either their internality or their externality, (p.76) Hegel tries to do justice to both. He does so, however, in metaphysical terms that have an affinity with Leibniz, and by beginning with something close to Leibniz's conception of essence as the complete notion of a substance.

Hegel develops the notion of ground not only in relation to Leibniz's principle of sufficient reason, but even more so in relation to the formulation Kant gives to it that is responsible for the antinomies into which reason falls: “If the conditioned is given, then the whole sum of conditions, and hence the absolutely unconditioned, is also given, through which alone the conditioned was possible” (KrV A409/B436). Now, as part of generalizing the antinomial conception of reason beyond the strict limits established by Kant, Hegel sees that the conception of the unconditioned as the sum or whole of the series in particular is just one way in which the unconditioned might appear. And Kant himself thinks that the true satisfaction with respect to the unconditioned comes not from taking an infinite series as a whole to be unconditioned, but rather when the unconditioned is one member of the series to which the others are subordinated but is not itself subordinated to or conditioned by any other (KrV A417/B445).

In Hegel's conception, the basic tension is rather between external conditioning (the demand that the explanation give us some new information that supplements or repairs our understanding of the explanandum) and internal conditioning (the demand that the explanans stand on its own as sufficient to supplement our understanding). The latter is a demand for a locus of responsibility, and the former a demand that such a locus have a determinate and informative character. Only the two together give us a sufficient reason or complete explanation. Now, since the demand is for external conditioning of the explanandum and internal conditioning of the explanans, there is at first no necessary conflict between the two. But because this demand of reason is taken to be universally applicable to all contents of thought, the explanans itself seems to be subject to the demand for external conditioning, and the explanandum to that of internal conditioning: the explanans has to have a distinctive content and relation to the explanadum in virtue of which it is explanatory, and the explanadum must be individuable as a distinct element in order for the further explanatory relation to it to be discriminable. For the problem of explanation itself, whether this internal conditioning is understood in terms of the completeness of a series or in terms of some other status of the explanans itself is not relevant, as Kant's own example of noumenal agency (a complete series of one) itself shows. The basic point needed to understand Hegel's conception of ground is just that these twin exigencies of explanation are both complimentary and in tension, and thus that the proper account of explanation must be one in which it can be shown how it is at least possible that their contrary demands could be reconciled.

Kant's most evocative presentation of the significance of these two exigencies comes in the antithesis of the Third Antinomy: (p.77)

Thus nature and transcendental freedom are as different as lawfulness and lawlessness; the former burdens the understanding with the difficulty of seeking the ancestry of occurrences ever higher in the series of causes, because the causality in them is at every time conditioned, but it promises in compensation a thoroughgoing and lawful unity of experience, while the mirage of freedom, on the contrary, though of course offering rest to the inquiring understanding in the chain of causes by leading it to an unconditioned causality that begins to act from itself, since it is itself blind, breaks away from the guidance of those rules by which alone a thoroughly connected experience is possible. (KrV A447/B475)

But Kant's moral theory demands the identity of the noumenal cause with the phenomenal being subject to natural causal determination, so his view already expresses the need for an identity between the internal determination of the former with the external determination of the latter. Because it is the understanding of this problem in terms of a series that ties it to time in particular for Kant (and, through time, to space), the spatio-temporal aspect of the problem drops away for Hegel, and it becomes a more purely conceptual problem about the relation between a ground (explanans) and what it grounds (explanandum). But what remains is Kant's association of self-activity or freedom with an unconditioned element that completes an explanation (KrV A418/B446).

This association is further developed by Fichte, who phrases the problem of practical self-determination in the same terms: “Self-sufficiency, our ultimate goal, as is often recalled, consists in everything being dependent on me, and my not being dependent on anything else … Now this goal is indeed unattainable, but I should nonetheless always advance towards it.”1 But as Frederick Neuhouser notes in his discussion of this idea in Fichte, though the idea of an objective and not merely subjective form of self-determination is an attractive conception, it is much easier to describe the independence from external influences than it is to describe the positive conception of self-sufficiency (one is tempted to say, “self-dependency”).2 The difficulty is in understanding how I could be identical to features on which I am dependent. Here we return, then, to the problem of setting out a conception of self-grounding that will articulate such a positive conception of self-sufficiency without the absurd implication that the agent should attempt to somehow conquer and eliminate all external influence on her will.

As we might expect from the discussion in chapter 2, a ground expresses the unity of external and internal determination as a unity of the twin forms of creation and (p.78) interpretation. First, the same content can be conceived as both non-externally determined (i.e., as simply given to a thinker), and as externally determined by its ground. To take the subjective side, if a fact is explained it is both initially considered independently and then considered as determined by some other fact. So it is taken as a starting point, but then seen as an expression of the ground that it presupposes. Second, this external determination by a ground is just as much an internal determination of the fact when it is taken inclusively. When the given is taken as necessarily or intrinsically determined by something else, that “something else” is a part of the complete description of the given. If one begins from the presupposed ground, and posits on that basis a behavior or phenomenon, one is positing that behavior or phenomenon as an expression of the ground and thus seeing that behavior or phenomenon as inherently characterized by the description provided by the ground. That is, to see the grounded as an expression of the ground requires both thinking of it as created by the ground (in the sense that the ground is responsible for it) and as being interpretable by reference to the ground, so we have here both positing and reflection-into-self. Any given fact requires explanation by another (external determination) if it is to be completely described, but since this entails that the determinate description of the fact includes that relation to the other, this relation is properly an aspect of that fact itself (internal determination). In this sense, the ground counts also as an expression of the grounded, since it provides a fuller picture of the grounded:

In so far as the determination of a first, an immediate, is the starting point of the advance to ground (through the nature of the determination itself which supersedes itself or falls to the ground), ground is, in the first instance, determined by that first. But this determining is, on the one hand, as a superseding of the determining, only the restored, purified, or manifested identity of essence which the reflected determination is in itself. (WL291–2/SL444–5)

The semblance (the “first … immediate”) is internally determined by the ground in such a way that the ground seems to take on the form of the internal determination of the given; it is so closely connected to the nature of the given (the guise or appearance) that we are inclined to include it in our description of the given (what Leibniz might call its “complete description”) and therefore to count its external determination by the ground as a form of its own self-determination. This is what Hegel means in describing the determining as the “manifested identity of essence.” Thus we often include in our determination of what is happening elements of our account of why it is happening, as, for example, I might describe the rabbit's running as “trying to escape the dog.”

To connect this back to the more detailed articulation of reflection as a unity of creation and interpretation, we now need to understand how ground is an initial (p.79) interpretation of that unity. Recall that we said that in reflection-into-self or interpretation, external determination (input) reveals itself to be internally determined (perspective) through external determination (process) and so transformed into internal determination (output), and in positing or creation internal determination (input) reveals itself to be internally determined (perspective) through external determination (process) and so transformed into external determination (output). We begin with the grounding relation as a kind of creation (positing): the initial independence of the guise (a kind of immediate internal determination) is our input, but we are forced by one pole of the explanatory exigency (that of conditioning or external determination, process) to change that internal determination internally (in the sense that this demand is driven from the internal perspective by the nature of the content of the immediate presentation itself as necessarily contrastive) into a form of external determination (output/result: its existent contrasts with other guises). But in doing so we necessarily move to thinking of the grounding relation as a kind of interpretation (reflection-into-self): here we take the existent external relations between the guises (input) and subject them to the demand of the complementary explanatory exigency (that of finding a locus of responsibility, that is, of finding a form of internal determination, process/main relation). This subjection is internal (in perspective) to the contrastive relations that are our object (since Hegel argues in the Doctrine of Being that an infinite regress of conditions renders the content of thought unintelligible), and results in the construction of a ground that functions as an explanans precisely in virtue of its independence (an output of internal determination). Here Hegel pushes past the spatial metaphors that dominate our more abstract thought to present an account in which the boundaries of self- and other-responsibility are not so much permeable as productively interrelated in a complex pattern of thought.

Dieter Henrich is thus right to claim that ground is the first category to thematize the project of essence, which is the attempt to hold together the divergent consequences of the concept of an essence in one construction, while simultaneously validating those consequences as divergent. This is true both of internal and external determination as divergent consequences (or rather opposed sources of divergent consequences), and, at the meta-level, the attempt to find a single construction is the attempt to validate internal determination whereas the frank recognition of divergence retains the element of external determination. That each is validated only through the other is the deepest sign of this attempt on Hegel's part.3 On my view, this is best understood historically as Hegel's response to the twin explanatory exigencies as they develop from Leibniz to (p.80) Kant. In this response, the ground is productively identical to what it grounds, that is, it is self-grounding.

As a way of bringing out Hegel's conception of the identity involved in explanation, and of avoiding the misleading though understandable impression that Hegel's argument for self-explanation is based on idiosyncratic metaphysical grounds of questionable contemporary interest, a comparison with Wilfrid Sellars's conception of scientific explanation is helpful. Sellars is concerned to understand “theories of the type which, to speak informally, explain the behavior of objects of a certain domain by ‘identifying’ these objects with systems of objects of another domain, and deriving the laws governing the objects of the first domain from the fundamental laws governing the objects of the second domain,” such as the explanation of the properties of water by reference to the identity of water with H2O molecules and the properties of those molecules.4 As Sellars acknowledges, the notion of the identity of non-observable objects with observable objects is perplexing, in particular because it requires the identity not only of objects but also of observable and non-observable properties. Nonetheless, Sellars thinks that the reality of the non-observable theoretical entities is dependent on just such an identification:

In my opinion, the only alternative to this conception [of the identification of theoretical and empirical entities] is the instrumentalist conception of theories as deductive systems, the distinctive vocabulary of which consists of what, in the context of pure geometry, are called uninterpreted expressions, doomed as a matter of principle to remain so … But surely our willingness to use the language of identity in connection with empirical and theoretical objects involves a commitment which goes beyond anything which would be implied by correspondence rules if these were formulated ascetically, in accordance with instrumentalist convictions, as syntactical bridges between a language and a calculus. I do not think that this willingness rests on a mistake.5

Sellars is forced to speak “informally” against the background of the Leibnizian notion of identity as differenceless continuity. But on Hegel's view of identity as essentially involving difference, and as being more significant the greater this difference is, Sellars is actually speaking quite strictly. Sellars proposes a specific model for this identity in terms of correspondence rules, theoretical models, and definitions that would preserve differences between the two sets of objects.

(p.81) Sellars's concern about the reality of theoretical entities parallels Hegel's own interests in explanation.6 Specifically, Hegel's argument for the validity of partial self-explanation is an attempt to understand the status of those essences to which we make appeal in explaining, for example, the camouflage behavior of the cuttlefish.7 On the one hand, this basic nature to which we appeal cannot be understood as just another phenomenal appearance (the nature of the cuttlefish is a different kind of thing than specific instances of behavior). But on the other hand, essences cannot be mere fictions or ens rationis, or their very explanatory power is vitiated.8 Furthermore, Hegel's attempt to determine the status of explanatory grounds takes on a form similar to Sellars's project, with its specific rules connecting observable events with unobservable theoretical entities, as Hegel looks for more complex structures to describe the identity between ground and grounded.9

There is a second, somewhat distinct ground for Hegel's support of this identity, which turns on understanding explanation as a kind of inference or entailment relation. The kinds of relations discussed in the Doctrine of Essence differ from those discussed in the Doctrine of Being by supporting inference. But the relevance of an inferential ground to its conclusion cannot be secured without some sort of shared content between the two, given how tight entailment relations are.10 Interestingly, Hegel is here formulating the problem at the level of Hempel and Oppenheim's seminal discussion of explanation as a pattern of inference. In “Studies in the Logic of Scientific Explanation,” Hempel and Oppenheim note that our intuitions prohibiting self-explanation cease to be informative at a crucial juncture.11 On the one hand, a theory of explanation should obviously rule out vacuous self-explanation in which the singular component of the explanans is identical to the explanandum. The fact that it is raining cannot be explained by the fact that it is raining, even if the latter is coupled with appropriate meteorological laws. On the other hand, to insist that there be no overlapping content between explanans and explanandum is to require that they be irrelevant to each other. If the explanans of the rain cannot even make reference to the (p.82) atmospheric moisture that the rain is, it is hard to see how it could be explanatory. “Therefore,” they conclude, “in every potential explanation in which the singular component of the explanans is not dispensable, the explanandum is partly explained by itself.”12

One final ground for this view comes from Hegel's understanding of Newtonian mechanics, which has seemed to many to provide paradigms of explanation. Hegel rebukes Leibniz for his criticism of Newton's conception of gravity as an “occult quality,” and claims that the truth is precisely the opposite: “gravity” does not add enough to the phenomena observed. So it counts as an explanation in virtue of that identity, but as a somewhat insignificant explanation (WL305/SL459). Leaving Hegel's tendentious consideration of scientific explanation aside, some sense can be made of mechanistic physical explanations as involving identity. Consider the standard philosophical example of the billiard balls. Though we are tempted to say that the cue ball's striking the 8-ball was the cause of the 8-ball going into the corner pocket, and thus to formulate an explanation that does not appear to have any substantial identity but rather trades on an antecedent state, the contemporary physicist is likely to give a somewhat differently framed explanation. The first layer of physical explanation would be in terms of conservation of force. Making the kind of simplifying assumptions that one makes in first-year physics, one can say that the movement of the 8-ball can be described quantitatively as the product of the mass of the 8-ball and its acceleration, and that this product is equal to the force exerted on it externally by the cue ball.13 This force exerted on it is, in turn, equal to the product of the mass and the acceleration of the cue ball. So the basis of the first-level physical description is precisely an equality and the principle of the conservation of force. On Hegel's expanded view of identity, this is to make an identity claim: the force of the 8-ball is the same as the force of the cue ball that hit it. The second-level explanation—answering the question, why are forces conserved?—will be in terms of conservation of momentum (i.e., the sum of the momenta of the two balls before impact must be the same as the sum of the momenta after impact), and the third-level explanation—answering the question, why is momentum conserved?—will be in terms of symmetry and invariance.14 All three of these (p.83) explanations trade on identities in Hegel's sense of the term. This is not a historical point about Newton's mechanics specifically or about Hegel's relation to Newton, but rather in the service of the continuing point that Hegel's view here is not as idiosyncratic as it at first seems from a contemporary perspective.

§4.2: Internal and External Determination in Explanations

So far, we have considered general arguments for considering ground as an articulation of what it means for something to be an essence, where this involves an identity between the ground and the grounded. Now we turn to the specific form of that identity. Hegel's development of the structure of ground has three phases: “absolute ground,” in which he formulates the problem of giving an account of grounding that will coordinate the internal determination of a subsisting essence with the external determination of real differences without collapsing into the rejected, single-level accounts of the Doctrine of Being; “determinate ground,” in which a provisional, three-term account of the ground-relation is given; and “condition” in which this three-term account is given a holistic interpretation. The guiding thread is the concern to make the relation of ground reflective rather than simply reflexive: to show that the ground-relation internalizes a relation to another (but without prejudice to the other doing the same), rather than being a simple relation to self unmediated by a relation to another. Though we are a long way from concrete agency, this is clearly the kind of explanatory schema necessary for understanding agency as an embedded phenomenon.

In his opening discussion of absolute ground, Hegel phrases the requirement of the reality of the explanans as the need for the subsistence of the essence, and like Sellars he understands this in terms of the identity of the essence qua ground (explanans) with the grounded (explanandum). In order not to collapse into uninformative simple identity (a purely reflexive relation), explanation must have a form that distinguishes the explanans from the explanandum—that is, something analogous to Sellars's correspondence rules, models, and definitions. As Hegel puts it, the form is “the positing and determining principle; simple essence, on the other hand, is the indeterminate and inactive substrate in which the form-determinations subsist and are reflected into themselves” (WL296/SL449). Here the problem is formulated as the difficulty of understanding the relation between a determinate form (external determination) and a subsistent essence (internal determination).

Despite the language of substrate and its determinations, Hegel thinks it would be a mistake to understand this substrate as mere matter, unformed and independent of the form it takes in explanatory relations. This substrate is not merely the potential for an explanation—rather it is the essence as the explanans of the explanation, and so (p.84) constrained by the form of explanation in general, and so something with a specific content. Thus the problem is phrased as one of understanding the “reciprocal presupposition” of the form and content of explanation as a way in which “the absolute grounddetermines itself” (WL299/SL452).

What this means is not immediately clear, but may be brought out by connecting this question back to the earlier theme of identity. For Hegel, explanations are a certain complex of identities (including differences). In the earlier development of Hegel's Logic, the problem of identity devolved onto the problem of explanation as a way of specifying which similarities and differences were most relevant to the question of individuation; knowing what a thing is turns out to involve knowing why it is the way it is. Hegel's argument here makes this connection by conceiving of the essence as just those similarities and differences that are explanatorily relevant. For example, it is not just any matter that explains the temperature at which water freezes, but rather the particular nature of H2O. In contrast, if the question is about an explanation for the electrical conductivity of water, then the fact that water usually contains dissolved salts and minerals will be the relevant essence, which is only indirectly related to its chemical composition as H2O. Whether H2O or some other feature is taken as the ground of any particular guise of water will depend on the particular needs of explanation, but clearly not just any content can play the role of the explanans. There must be a basic content that is the subject of the explanation, and which is then subject to the particular constraints of the ground-relation itself. Paradoxically, as John Burbidge puts it, “matter is to be the necessary ground of form.”15 An explanation is then doubly constrained: first, by the available identities included in the content of the subject of explanation (e.g., this water with dissolved salts, with H2O, with liquid phenomena, etc.); second, by the form of explanation itself that will determine when a particular identity can be explanatory (i.e., some structure such that when the identity is placed in the right relation to a certain phenomenon and context, it satisfies the request for an explanation). And yet Hegel argues that the presupposition of content and form is reciprocal. The phrasing above articulates the way in which the form of explanation is constrained by the content, but the kinds of identities contained in the understanding of a thing as having an essence are already constrained by their relevance to interpreting guises of that thing as expressing that essence. So, in the above example, the potential inclusion of the idea that water is a solution with dissolved salts in the understanding of what water is, is guided by the fact that this common (and dangerous and therefore practically significant) feature of water can only be explained in this way, even if I am not explaining that feature at a given time.

(p.85) So far, rather than solving the problem of explanation, Hegel has doubled down on his gambit to make sense of essences by their functional role in explanation, because he has made the case that the very notion of the content of the object is parasitic on ground-relations:

The content is … that which is identical in form and matter, so that these would be only indifferent external determinations … The identity of the content with itself is, therefore, first that identity which is indifferent to the form; but secondly, it is the identity of the ground. The ground has, in the first instance, vanished in the content; but the content is at the same time the negative reflection of the form-determinations into themselves; therefore its unity which is, at first, only the unity that is indifferent to the form, is also the formal unity or the ground relation as such. The content has, therefore, the latter for its essential form, and the ground conversely has a content (WL301/SL455).

On the next page, Hegel summarizes this idea as follows: “The determinateness of the content is … the substrate [Grundlage] for the form … ” (WL302/SL456). His next discussion of “Determinate Ground” tackles the problem of how to differentiate explanatory from non-explanatory identities.

Before moving on to this next stage in Hegel's development of the concept of ground, it is worth noting the way in which form and content fill out the notion of ground as reflection. Initially, form and content appear to play the role of inputs and outputs: form as input of reflection-into-self and output of positing, and the converse for content. That is, interpretive reflection on the specific formal differences between the guises produces a conception of the content that they express, since it traces the creation of those guises through the positing of the content in specific contexts. Since form is differentiation and content identity, this confirms the earlier analysis of the nature of the inputs and outputs of these two movements of reflection, but form and content are of course more than just differentiation and identity; they are particular conceptions of the kind of differentiation and identity relevant to explanation. In the notion of form Hegel plays on connotations of an articulated structure, and in the notion of content or matter Hegel plays on connotations of a substrate that is immediate in contrast to the formal articulation that it is destined to receive. In addition to playing the roles of input/output, form comes to appear to be the determining movement of its own production through positing, and content or matter comes to appear to be the receptive process through which it is generated. Form, that is, appears to be the process of positing, and content that of reflection-into-self. That this latter receptive process is taken to have a perspective internal to the input (which is the differentiated form) is a continuation of the notion of the distinguished guises as self-undermining (p.86) that we saw in the basic conception of reflection, a self-undermining that is expressed in the interpretive movement from the guises to the essence.

Here we start to make out in preliminary and undeveloped form the structure of the developmental recursiveness that Hegel employs to shape his argument here. Specifically, the output of positing at an earlier stage comes to be the perspective of positing at the next stage, and the output of reflection-into-self comes to be the process or main relation of the next stage of reflection-into-self. In this way, the externally determining output of positing comes to constitute the internal perspective of positing at the next stage, and the internally determining output of reflection-into-self comes to constitute the externally determining process of change. The conceptions are too abstract for this process to do much work here, but I mention it not only to prepare the way for further forms, but also to emphasize the complexity of the problem of internal and external determination for Hegel, and the way in which the one turns into the other and vice versa. This is emphatically not an indication that Hegel rejects the terms of the problem; what we see here is not a reductio ad absurdum of thinking in terms of internality and externality. Rather, this sense of their mutual interdependence defines them as categories in Hegel's Doctrine of Essence and indicates that they are being grasped in such a way that will enable Hegel eventually to formulate a conception of embedded agency.

§4.3: The Argument for Explanation as a Three-Term Relation

But to come back from the brink to the more concrete significance of Hegel's argument, one obvious way to formulate this criteria to distinguish explanatory from non-explanatory identities would be to attribute the identity to content and the differentiation to form, as we see Hegel doing in the longer passage from WL301 just quoted.16 On this proposal, one would say that the explanation consists of a shared content, and counts as informative in virtue of the different forms taken on by the explanandum and explanans and thus the potential presence of one (or recognition of one) in the absence of the other. Given the metaphysical terms in which Hegel phrased the question, this would be to take the form of ground to be a relatively superficial property of the underlying identity at issue, and yet one that could obscure understanding of that underlying identity. In more intuitive terms, this is the idea that (p.87) explanation responds to a misunderstanding of the phenomenon at issue that can be repaired with a relatively short reply. So perhaps I am surprised that my toddler falls asleep so quickly on a given night, and express my surprise to my wife in the form of a request for an explanation. She might reply with, “he has been very tired all day.” This kind of an explanation exploits the near identity between being tired and sleeping, where this identity only becomes explanatory as a result of my own lack of awareness of the fact that the former characterized my child. These kinds of explanations are closest in erotetic form to questions about identity, which similarly seem (on the standard account) to be motivated primarily by incomplete understanding.17

Such formal disguises of shared content might be deeper or more widely spread than one might think. Hempel and Oppenheim point out that contentful identity of the explanans and explanandum is not always immediately obvious from the form of the propositions in an explanation, whereas restatement in terms of logically equivalent propositions may make such identity of content apparent. So in the following example, Explanation 1 meets their criteria for explanation and seems intuitively valid, and yet it can be rephrased in Explanation 2 in such a way that the partial identity of content between the explanans and the explanandum is made apparent:

Explanation 1:

T 2 = ( x ) [ P ( x ) Q ( x ) ]
C 2 = P ( a )
E 2 = Q ( a )

Explanation 2:

T 2 = ( x ) [ ~ P ( x ) vQ ( x ) ]
C 2 = [ P ( a ) vQ ( a ) ] & [ ~ P ( a ) v ~ Q ( a ) ]
E 2 = [ P ( a ) vQ ( a ) ] & [ ~ P ( a ) vQ ( a ) ]

At this point they remark that there may be no clear dividing line between acceptable and unacceptable cases of the self-explanation of an explanandum.18 The purely formal (p.88) understanding of the differentiation at issue does not seem to give us a criterion for distinguishing between such cases.

Along these same lines, Hegel claims that mere formal differences between the explanans and explanandum create explanations in name only that lack the fundamental human significance of the true concept or practice of explanation: “To answer the question, why is this person going to town, with the ground, that it is because there is an attractive force in the town which urges him in that direction, is to give the kind of reply that is sanctioned in the sciences but outside them is counted absurd” (WL304–5/SL458–9). On Hegel's view, this difficulty is a symptom of the fact that such forms of explanation (what Hegel calls “formal grounds,” the first form of determinate ground) fail to unify creative positing (the movement in the direction from ground to grounded) and interpretive reflection-into-self (the movement from grounded to ground) precisely because they fail to distinguish between the two. Hegel reasons as follows: In formal explanations, quite often the phenomenon being explained is the reason that we believe in the explanans. This is the simplest form of inference to the best explanation. But if this is true, then the grounded is the ground for the ground, and Hegel worries that we are likely to think of the grounds as having the same status as the phenomena that they ground simply because the two have the same formal relation to each other. To go back to the discussion of Sellars, Hegel thinks that in chemistry, we will confuse the unobservable molecules with the observable substances of which they are the essence because they have a symmetrical relation to each other: we take the molecules to ground the observable behavior, but that behavior is precisely our evidence for the molecules themselves (WL306–7/SL460–1).19 Because the form is external to the content (as is necessary to safeguard the internal determination or independence of the content on this conception), its application to the content in one way rather than another is arbitrary:

What is present in the determinate ground is this: first, a determinate content is considered from two sides in so far as it is posited first as ground and again as the grounded. The content itself is indifferent to this form; in both it is simply one determination only. Secondly, the ground itself is just as much a moment of the form as that which is posited by it; this is its identity in respect of the form. It does not matter which of the two determinations is made the first, whether the transition is made from the posited to the other as ground, or from the one as ground to the other as posited (WL303/SL457).

(p.89) To put this point in terms more suggestive for agency, if we do not have this account of the difference between the creative positing of the grounded by the ground and the interpretive reflection-into-self that discriminates the ground as compared with the grounded, then we do not yet have an understanding of the way in which the essence of the phenomenon counts as the locus of responsibility that makes the external relations of its guises into a process of self-expression, and thus we still lack a basic conceptual model for embedded agency. One way to understand this failure in the specific context of explanation is to see it as the lack of an account of the directionality or asymmetry of explanation, for example, of why the cue ball's force explains the 8-ball's force rather than the other way around. The cue ball's force creates that of the 8-ball, but we interpret the latter in terms of the former.

If the problem is the insufficiency of merely formal difference, then the natural solution is to insist that the restriction on self-explanation rest on the demand for difference of content between the explanans and the explanandum, which Hegel discusses under the heading of “real ground,” the second form of determinate ground:

The side of the ground has shown that it is itself a posited, and the side of the grounded that it is itself ground; each is in itself this identity of the whole. But because they belong at the same time to the form and constitute the form's determinate difference each is, in its determinateness, the identity of the whole with itself. Consequently each has a distinctive content of its own. Or considered from the side of the content, because this is the identity of the ground-relation with itself, it essentially possesses this difference of the form within itself, and then is, as ground, other than what it is as grounded. (WL307/SL461)

The question then becomes, how to articulate this difference in content. One constraint, of course, is that there be some shared content—this is the necessary identity between the explanans and the explanandum which first gives the explanatory relation a content in Hegel's technical sense—so the options appear to be limited to additional content on one side or the other of the relationship. Hegel thinks that adding to the explanans won’t help because this would still constitute complete self-explanation as far as the explanandum is concerned ((A,B) explaining A), and thus does nothing to address his concerns about the insignificance of tautological explanations. Instead he adds to the explanandum, so that one part of the explanandum explains the explanandum as a whole (A explaining (A,B)): “[The grounded], as the posited, has its self-identity and subsistence only in the ground. But apart from this content of the ground, the grounded now also has its own distinctive content and is accordingly the unity of a twofold content” (WL308/SL462).

(p.90) A further complication: The relation between A and B in the grounded phenomenon is not itself an explanatory relation, even though A is taken to explain A,B considered together. So there are two kinds of relations here: the essential relation between A and A,B that provides the heart of the explanation, and the secondary relation between A and B that is not itself explanatory (and does not directly get explained), yet is somehow necessary to making the explanation what it is. Hegel characterizes this additional content B as only indifferently connected with the ground A, in the same way that different properties might be added arbitrarily to a substrate. In this way Hegel argues that “the ground relation has thus become external to itself” (WL308/SL462–3).

Before considering the difficulties of this proposal, an example may help to bring out Hegel's meaning here. The event of lightning striking a house might be explained by appeal to properties and laws concerning the electricity that lightning is. The fact that the strike hit one particular house rather than another might be explained by the fact that the house had an iron weathervane coupled with the tendency of electricity to discharge itself at the nearest conductor. Even though the fundamental explanation will be in terms of laws describing the nature of electricity, the conjunction of the discharge and the particular location will be partially explained by something quite distinct from those laws, namely the fact that one house had a weathervane and another did not. Thus in usual contexts, the nature of electricity should count as the locus of the activity of the lightning strike, with the additional fact of the weathervane as a secondary explanation or condition. Structurally speaking, we have a hierarchical combination of grounds that come together to explain the phenomenon.

More metaphysically, on this picture one can understand how the explanans or essence is “realized” in the explanandum, since the explanandum would be the content of the explanans connected with further characteristics and thus made more determinate. This I take to be the import of Hegel's use of the notion of a substrate to which unessential characteristics are added: the substrate itself does not exist in its pure form, but is related to the more particular characteristics of the individual lightning strike, such that the essential nature of electricity in general was expressed in this particular strike. And yet the essence of electricity could be expressed in many different ways. In action, the goal, intention, or agent's understanding is often considered the essence of an action because they explain the action. My goal not only explains that I do what I do, but often how I do it as well (i.e., my particular choice of means). If you ask me why I took the freeway instead of the coast highway, I might explain myself by appealing to my goal of not being late for the concert together with the fact that the freeway is quicker, if not quite so scenic. The action itself is the realization of that essence because it makes the goal, intention, or understanding objective and concrete in relation to other actions and events. And though we can understand how the agent's goals remain the locus of activity here, the fact that the freeway is quicker is not normally going to (p.91) be under the agent's control, nor is its status as a condition for the action, and thus neither will be directly explained by the agent's goal.

More methodologically, we can understand this Hegelian claim as providing a basis for the common assumption that an explanation that explains more phenomena with fewer concepts is a deeper or more satisfying explanation than one that needs more concepts to explain fewer phenomena. The former are more satisfying precisely because of the greater number and variety of features that can be brought into play as mediating between the fundamental terms of the explanation and their realization in observable phenomena.

But even more important that this insight about explanation is the way in which this notion of ground clarifies the notion of expression. This model of realization through explanatory relation sheds some light on the account of the prioritized unity of internal and external determination. Here Hegel suggests that the locus of activity is an essence that expresses itself in external relations that are not grounded in it itself, and yet serve as resources for expression. If I express an idea, for example, there is more in the expression than there was in the original idea, because the expression brings the idea into contact with other ideas and with the internal differentiations required by the language used to express it. Recall from chapter 1 that this idea of external relations as resources for self-expression is precisely the way in which Hegel understands agents’ motivations. If this model can be made to work, then, Hegel will have the resources to construct action explanations on the basis of his expressivist account of self-determination.

That which will express itself is necessarily exposed to failure and chance; this is a central theme in Hegel's philosophy of action. Sometimes this is put as Hegel's acceptance of an element of moral luck, for example, in relation to Hegel's admiration for the saying that “the stone belongs to the devil when it leaves the hand that threw it” (PR§119Z). The point here is just that the externality of the circumstances with respect to the ground is itself internally determined (i.e., necessitated by the nature of the ground itself). There is an element of potentially alienating loss of control in externalization that must be accepted at the very heart of every kind of ground.

At this level, the response to this potential through reflection-into-self is rather insufficient, since it attempts just to recoup the initial identical content of the ground A, and thus in an important sense to deny the contingent articulation of the ground A its proper due. In some respects, standard belief-desire models of action are also forms of such inadequate recuperation of the essence, since the contingent features of the action are chalked up to external circumstances registered in the belief that in the situation in which the agent finds herself, action X will satisfy desire Y. The desire itself remains pure of these external circumstances and represents the simple identity of the agent in her action. Though not generally acknowledged, it is precisely this purity or (p.92) abstraction of the desire that leads to the well-known problem for such accounts that they are insufficient to secure agency in virtue of the possibility that the agent does not identify with the desire. It is precisely because of this abstractness of the identity of the ground-desire that real flesh-and-reason agents can fail to see it as an expression of who they are in the concrete context of their action. This empiricist view then actually shares the same problem as the Kantian view: its basic ground-principles of action—desire for the empiricist, the moral law for Kant—are posited as being only in an external and arbitrary relation to the concrete circumstances and thus to the specific form of the agent's action and behavior.

Here Hegel has quarried down to a substratum common to both views, a fundamental conception that brings arbitrariness in the relation of ground-principle to the specific form of action, regardless of whether some specific capacity for choosing arbitrarily (Willkür) is posited. The arbitrariness is in the abstractness of the identity of the principle, so from Hegel's perspective, a libertarian like Kant is actually more intellectually honest in frankly acknowledging the element of indeterminacy that is inherent to this form of the connection between ground-principle and circumstance. That capacity doesn’t add anything new to the picture of agency so much as it simply acknowledges the difference between the abstractness of the principles and the specificity of action, and notes that, in fact, something mediates between them that we call the will. As Hegel's acceptance of moral luck suggests, he rejects this simple view of agency. On Hegel's view of ground, arguing about the intelligibility of the capacity for arbitrary choice while accepting the basic (abstract) framework is to stay at the level of the symptoms instead of diagnosing the underlying pathology. But in the Logic, Hegel argues more abstractly that the arbitrariness of the response of “real” reflection-into-self is disastrous for the project of identifying the locus of responsibility on which every explanation depends for its sufficiency. The dialectic of the argument thus proceeds through Hegel's frank recognition of the inadequacy of the simple model of self-grounding that many commentators attribute to him.

We can see this difficulty if we return to the details of explanation proper, and notice that the devil is in the details of connecting the two sets of contents in the explanandum (A and B). What is needed is an account of this additional content and of its relation to the first content of the original ground, since this relation now is a relation that partially constitutes explanation and serves as the crucial feature for distinguishing significant from insignificant explanations. This account must also be an account of which content is which, and the need for such an account is made all the more pressing by Hegel's recognition that in principle any of the varied aspects of the explanandum may be taken as essential and explanatory: (p.93)

An official has an aptitude for his office, as an individual has relationships with others, has a circle of acquaintances, a particular character, made an appearance in such and such circumstances and on such and such occasions, and so on. Each of these attributes can be, or can be regarded as, the ground for holding his office; they are a diverse content which is joined together in a third; the form, in which they are determined as being either essential or posited in relation to one other, is external to the content. (WL311/SL465)

There is nothing yet in the bare relation of explanatory relevance to isolate any particular attribute as explanatory.

Hegel has yet to meet the criterion of being able to distinguish explanatory from non-explanatory identities, though he has already specified one necessary condition: at least one of the differences involved in the identity relation between explanans and explanandum is that the explanandum contain additional content not found in the explanans. But without a further criterion for determining which additional contents in the explanandum make the relation explanatory, there will be no way of saying, for any particular action or event, whether the relevant explanatory essence is something that defines it as a locus of activity. As Hegel puts it,

Real [as opposed to merely formal] ground contains a diversified content; but this brings with it the contingency (Zufälligkeit) and externality of the ground relation. On the one hand, that which is considered as the essential and therefore as the fundamental determination, is not the ground of the other determinations connected with it. On the other hand, it is also undetermined which of the several determinations of the content of a concrete thing ought to be taken as essential and as ground; hence the choice between them is free. (WL309/SL463)

The externality of the new content is crucial to its positive value, since it breaks ground out of the immediate and unproductive circle of the simple and traditional metaphysical notion of something responsible for its own being, but as a matter of the theory of explanation, we still lack a criterion for distinguishing between explanatory and non-explanatory identities. The externality of the new content B that has been added to the grounded, in virtue of which we have both informative explanations and the metaphysical realization of the ground, is grounded in the ground, but what that means is precisely that the specific character of B is not grounded in the ground, and thus that the ground-relation itself does not distinguish between A or B, or determine whether A or B is the essential element, and thus does not say whether A or B is the ground of (A,B). Paradoxically, the contingency of the additional content has infected the ground in such a way that the ground is unable to incorporate that additional content into (p.94) itself. The best it can do is to return back into itself as presupposed (reflected into itself), but without registering the changes brought about by its embedding in its context. This inadequacy then makes itself apparent as the arbitrariness of selecting particular contents in the explanandum to serve as the explanans.

Although Hegel's tone in discussing this possibility of a variety of explanatory grounds is understandably quite negative, it nonetheless points to an aspect of explanation that is important for its significance and fruitfulness in ordinary reasoning, namely that different contexts of inquiry allow isolation of different ground-relations in a single phenomenon. The explanation for the rust on my car may be the water in the rain (for me) or the iron in the car (for the chemist), or a poor paint job (for the repair shop) or inadequate quality control at the factory (for the manufacturer) or lax enforcement of automotive manufacturing regulations (for the investigating congressional committee). Although this certainly opens the door for sophistry as Hegel claims, it is also important that our basic conception of the role of explanation in the world be sufficiently complex to allow for these different possibilities.20 Determination of the sufficient ground requires more structure and information than the basic two-place conception of the ground-relation can provide. As Longuenesse nicely puts it, “the essential determination (defined in ‘real ground’) does not suffice to ground its own unity with that for which it is essential.”21

The two sides of “determinate ground” represent the failures of theories of explanation that attend to only one pole of the explanatory exigency. If, in formal ground, expression and explanation failed in virtue of the excessive role of the internal determination of the content, real ground fails in virtue of the excessive role of the external determination of the multitude of external circumstances involved in any grounded. Hegel is sensitive, therefore, to the ways in which essence's divergent tendencies can move it out of balance. It becomes quite clear that Hegel's interest in ground is closely related to the issue of how to connect general principles taken to control a certain phenomenon with that phenomenon in its specificity. As mentioned above, this takes on a certain form in conceptions of agency, as philosophers search for a philosophy of action that will helpfully articulate the basic structure of action in a limited number of basic principles or processes and which can naturally take on a concrete phenomenal form in the experience of individual agents in a way that is driven by the nature of the basic principles themselves. Otherwise we will be unable to (p.95) interpret behavior as agency except arbitrarily—we would employ the schema of agency as one among many schemas that we had at our disposal—and thus we would be unable to understand behavior as an expression of agency. But Hegel's discussion of ground cuts to the logical core of this issue, since this is a problem for any theory that attempts to explain in terms of basic principles.22

In fact, Clark Butler sees Hegel's discussion of ground here as a “self-refutation of determinism” accomplished precisely through this tension between the greater generality required of explanatory grounds if they are to constitute explanations rather than mere descriptions, and the infinite specificity of any event qua explanandum. On Butler's view, Hegel's argument against determinism in the discussion of ground is connected with the principle of determinability when connected with the abstraction characteristic of thinking in terms of essences: (1) Given the non-identity of discernibles, every concrete event must have infinite properties in order to exclude the possibility of multiple instantiation. But (2) the state of the world together with causal laws can only entail the occurrence of a general type of event indicated by a finite description. (3) Since what occurs is necessarily specific and thus infinite, “there is an unbridgeable gap between the fact that an explanatory ground grounds a future event under an abstract description and the occurrence of that concrete event.”23 In the terms introduced earlier, we might think of A as the abstract description that is then coupled with the infinite multiplicity of properties B.

The questionable part of the argument is clearly (2). It may certainly be granted if meant in an epistemic sense (i.e., it is impossible for we limited knowers to cobble together more than a finite description of the world and its laws which is thus at a distance from the infinitely specific event that description is to ground). But that interpretation would make thought in the Logic contrastively subjective in the Kantian way that Hegel rejects. On this interpretation, Hegel's argument regarding grounds would only produce the conclusion that human action is unpredictable, but as is widely recognized, that is not the same as to say that it is ungrounded or undetermined, and it is this latter status that is relevant to free agency. Furthermore, as Hume and others have pointed out, if action is ungrounded or undetermined, this is not necessarily any better for the prospects of free will, since it appears to make action random.24 In the contemporary idiom, there is an inherent tension between alternate possibilities and (p.96) control in the concept of free will. Unless that indeterminacy can be further determined by the agent, we will have no conception of self-grounding and therefore no conception of free will.

I suspect that whatever validity the fundamental claim that the ground must be finite may have as a metaphysical thesis counseling parsimony in our fundamental picture of reality is in fact parasitic on our conception of explanation. Hegel's arguments here regarding the dilemma between formal and real ground can be seen as analogous to the more specific arguments about explanatory laws presented in the Force and Understanding chapter of the Phenomenology. There Hegel can be seen as grounding precisely this premise in the requirement of explanation through development of the tension inherent in explanation itself between the generality of principles (which is required in order to say that an event has been explained rather than merely described) and the specificity of events (the multiplicity of which must be subsumed under those general principles of the ground). Here in the Logic there is not yet any explicit construction of this argument in terms of laws, though Butler is certainly right to see in this section Hegel's attempts to come to grips with the fundamental tension in the notion of explanation between elegance and completeness.25 But because of both the uncertainty surrounding premise (2) and the manifold meanings of “determinism” as a threat to free will, we should not yet see Hegel's argument here as a decisive refutation of determinism in that sense.

To return to the development of Hegel's theory of explanation, he makes a virtue of necessity in officially introducing a third term, an “external ground,” to distinguish between the essential and the inessential aspects of the explanandum: “Hence the real ground-relation is rather the ground as superseded; consequently it constitutes rather the side of the grounded or of positedness. But as positedness, the ground itself has now withdrawn into its ground; it is now a grounded and this has another ground” (WL312/SL467). This recursive move of grounding the ground-relation in another ground raises an obvious question as to the nature of the new ground: what could it be like that would allow it to avoid the very problems which necessitate it? Hegel argues that this external ground must contain the two sets of contents contained in the ground- (p.97) relation and explanandum, but stripped of the form of the ground-relation itself. Because the very independence of the ground A is parasitic on the performance of a reflection-into-self that, if not arbitrary, must be determined by an external reflection in order to pick out A rather than B as the ground of A,B, the best way to understand this within the framework of ground is precisely to think of the ground as having to be grounded in this external perspective.

In developing this new conception in which A and B are connected by some non-explanatory relation, Hegel is almost certainly oriented by the possibility of a double-aspect style regress: if this new ground contains a ground-relation it too will have to be split into two aspects, a determinate ground and an external ground, and then the latter will have to be so split, and so on to infinity.26 The second “ground” must somehow represent the unity of the combined characteristics in the explanandum without the kind of articulation or positing expression that characterizes the ground-relation as such. Thus Hegel characterizes this additional feature as an implicit relation: that is, a relation which has being in itself (“an sich seiende Beziehung,” WL313/SL468) or an “absolute relation” (WL312/SL468). Hegel calls this “complete ground,” since it contains both the identity of formal ground and the specific, differentiated content of real ground. On this view, (A,B) in absolute relation ground or explain (A&B) in their structured, grounding relation. This seems at first to be the unhelpful suggestion that the aggregation of the characteristics in the explanandum is explained by the aggregation of those same characteristics, but Hegel's conception is rather the following: that A grounds A&B is itself grounded in the absolute relation between A and B outside of the grounding relation.

Thus Hegel claims that the mere aggregation of characteristics in one instance or respect grounds the positing of one of those characteristics given the presence of the others in another instance or respect:

The inference is as follows: in one something, the determination B is implicitly connected with determination A; therefore, in the second something to which only the one determination A immediately belongs, B is also linked with A. In the second (p.98) something, not only is this second determination a mediated one, but the fact that its immediate determination is ground is also mediated, namely by its original connection with B in the first something. (WL313/SL468)

A is taken to explain B in one instance only because A and B are found together in other instances: not only is the first instance mediated in the sense that there is a kind of productive or controlling priority to A as ground of B, but that mediation or relationality is itself parasitic on the non-mediated co-occurrence of A,B in the second instance.

Here the conceptual structure is so intricately turned in on itself that any attempt at adequate formalization is futile. Nonetheless, our earlier dual formula for reflection can be used to capture some of what Hegel means here. In fact, in summarizing this conception Hegel gets very close to the formulaic structure of inputs, outputs, perspectives, and processes:

First, ground as the original relation, is the relation of immediate content-determinations. The ground-relation, being essential form, its sides are determined as superseded or as moments. Therefore, as form of immediate determinations, it is self-identical relation at the same time that it is the relation of its negation; hence it is ground, not in and for itself, but as relation to the superseded ground-relation. Secondly, the superseded relation or the immediate which, in the original and the posited relation, is the identical substrate, is likewise not in and for itself real ground; on the contrary, it is posited as being ground through that original connection. The ground-relation in its totality is therefore essentially presupposing reflection. (WL314/SL469)

Here, we start with the positing side: the input is the “original” (internally determining) relation, the structured relation of A and B where A grounds the two together. This is understood from its own perspective of being form as therefore subject to the process of transformation of its constitutive elements into dependent (externally determined) outputs. But since the dominant element of this original relation is precisely the prioritized ground-relation, this itself must be superseded and understood as dependent on its opposite. Therefore the output is in fact A and B in the immediate or absolute relation, which then serves as the input of reflection-into-self. That process takes up the perspective of that immediate relation as a substrate which is therefore intrinsically subject to the process of receiving form and thus self-producing the output which is the prioritized ground-relation. The output of positing and the input of reflection-into-self—this immediate or absolute relation—is the content of real ground, whereas the input of positing and the output of reflection-into-self—the structured ground-relation—is the form of formal ground. But because each input (the priori (p.99) tized ground-relation and the given multiplicity of content) can be understood as a kind of internal determination (the output of reflection-into-self), Hegel claims that the ground-relation as a whole can be understood as essentially presupposing (i.e., as interpretation or reflection-into-self). And yet it is worth emphasizing, so as to combat the criticism that Hegel's view here is excessively “unitarian,” that one of these forms of internal determination just is the existent multiplicity of circumstances in their bare differences from each other. Though they are not grounds in a contrastive sense, Hegel thinks he has isolated a meta-level sense in which these manifold differences count as a diffuse but necessary locus of responsibility upon which the more specific, prioritized ground necessarily depends. To use a vocabulary that suggests Heidegger as much as Hegel, Hegel thinks that the very non-thematized or unarticulated form of the background serves as a secondary locus of responsibility in virtue of its role in making possible the foregrounding of the theme through the articulating activity of the ground. That background is more than a formally necessary condition; it is rather an enabling condition or resource.

Though the conceptual structure of this proposal is very complicated, it actually describes a very common view about explanation, namely that which results from the Humean constant conjunction model of causal regularity coupled with Hempel and Oppenheim's notion of the isomorphism of explanation and prediction. According to this common position, the regular co-occurrence of two events justifies positing or expecting the second event when the first is given (or isolating it as the causal explanation when the second event occurs; it is immaterial whether, within grounding, we focus on positing or on reflection-into-self). It is only the regular co-occurrence of cue balls hitting 8-balls and 8-balls going into pockets that licenses (or at any rate habituates) my prediction that the 8-ball will go into the pocket if hit in the right way. But Hegel's argument in the Doctrine of Being dismisses the picture of atomistic determinate contents that supports the Humean view. That is, the Humean view presupposes that the two events could initially be given independently (entirely internally determined), and only later related as ground and consequent (externally determined). Hegel, however, continues the Kantian critique in arguing that the possibility of the events themselves is parasitic on their participation in and constitution by ground-relations. If Hegel's argument works in showing that facts as such are parasitic on determinate ground-relations, then the simple conjunction of characteristics must conceptually (though not necessarily temporally) presuppose a determining ground-relation. Yet on the suggestion Hegel is here considering, this is just the type of ground-relation that the simple conjunction is supposed to ground. If one is to avoid vicious circularity, this picture must be developed into a more complex model of the way in which the two aggregates of content—implicit co-occurrence and explicit explanatory (p.100) structure—are related. That is just to say that the only way for Hegel to avoid vicious circularity at this point is to develop an adequate model of expression.

§4.4: The Role of Conditions as the Third Term in Explanation

Hegel proposes that one consider the given aggregate to be the condition of the ground-relation, so that the condition contains the same content, but stripped of the dependency and entailment relations between parts of that content as present in the explanandum. Here he is essentially foregrounding the problem discussed above, in that we appear to have regressed to a view of the content of thought that Hegel has already rejected in the earlier Doctrine of Being, only now even this regress is described as part of the very activity of grounding. As Hegel puts it,

In this determination [the condition] is the form relation to the ground, withdrawn into identity with itself, and is consequently the content of the ground. But the content as such is only the indifferent unity of the ground, as in form—without form there is no content … Something is not through its condition; its condition is not its ground. Condition is the moment of unconditioned immediacy of the ground, but it is not itself the movement and the positing that is negatively self-related and that makes itself into positedness. (WL315–6/SL470–1)

To think of the content of the grounded phenomenon as condition is precisely to acknowledge its given character as a material on which the activity of grounding works. And since any material can be worked in a variety of different ways (though of course constrained by its form as well), the element of alternate possibility is now internalized in the ground-relation itself.

The difficult thing to see here is the way in which the whole relation—and not just the relata A and A,B—is taken up as the content and thus as the output of the interpretation or reflection-into-self of grounding. An example may help: though the presence of oxygen is generally only a standing condition in the explanation of house fires, it might become explanatory in the context of arson at the new lunar housing development. Here its presence in the former case is unremarkable and so non-explanatory, but because it is unexpected it becomes explanatory in the latter case. But even in the normal case, the presence of oxygen mediates between the ground of the fire (e.g., the candle tipped against the drape) and the whole fire both by making the connection possible and by leading to characteristics of the actual fire that are not determined directly by the tipping candle itself. As a result, the greater amount or flow of oxygen in one room rather than in another would help to determine why the fire moved in one (p.101) direction rather than another. Thus the conditions include not just background contextual factors, but determinative elements as well, and together these are the content of the ground-relation.

In agency, this would mean that explanatory characteristics of the action (e.g., goals, intentions, or consequences) could also be considered conditions of the action in another context. For example, ends that are explanatory in one context may be merely standing values that do not contrastively pick out the action against the relevant alternative in another context. For a very young child first developing altruistic ends, the intention to please her mother would explain her production of a birthday drawing. This would be a context in which we think it likely that the child would not on her own think to do anything for her mother. For the spouse, however, this intention is usually a standing project, such that the choice of gift will be better explained by a belief about the mother's tastes. In this case, the intention to make the same person happy is a mere background condition of action. But it is nonetheless true that the regular correlation between that intention and birthday gifts helps to support the ground-relation between the child's same intention and her gift. In both cases, though we speak easily of just one element (the spouse's intention or the presence of action) as a background condition, implicit in the significance of each element as a condition is the specific relational structure that comprehends both it and the other elements of the circumstances, and which implies the determining function of the background condition as well. This implicit nature is sometimes obliquely expressed in discussions of causation in the distinction between regularity as opposed to productivity. The form of the ground is an explicitly productive structure, but the implicit relations in the conditions are best understood as regularities. But Hegel's Kantianism is to understand these regularities as having themselves some determinative function.

Thus, Hegel continues the appeal to regularities in experience from the Humean view, but this appeal is understood more indirectly. It is precisely the fact that the appeal to the intention to please the spouse's wife is not explanatory in the normal context that makes the experience of such actions the proper conditions for explaining the young child's actions by reference to that intention, and the fact that the presence of oxygen is not explanatory in the terrestrial house fire that makes it explanatory in the lunar case. This is what Hegel means when he claims that the condition is a condition not for its own ground, but for the ground of some other explanandum (WL315/SL470). Unlike the Humean view, in which the same feature is explanatory both in the past contexts of co-occurrence and in the future context of prediction (because causation simply reduces to regularity), on the Hegelian view precisely the non-explanatory status of a content or phenomenon in another context licenses its explanatory status in the given context in which our inference to the explanans is an interpretation of the explanandum as an expression of the explanans. So the ordered but not productive or (p.102) prioritized regularities of the flow of experience are just as important to the Hegelian view as to the Humean, but their import is of a different sort.

We can go further into this distinctive import if we understand it in terms of the way the condition is both given to and yet constituted by the ground-relation. Hegel's attempt to make out this double nature of the condition follows on a similar problem in his earlier discussion of reflection, a problem that is the subject of an extended discussion in Henrich's analytical commentary. Just as the problem here is to understand the essential yet apparently derivative externality of the condition to the ground, Henrich argues that the earlier problematic externality of reflection can only be overcome if the object of that reflection can achieve the status of the concept of essence as a whole. This tracks the development we have just seen, in which the condition is turned into the whole of the ground-relation. According to Henrich, Hegel's way forward is to bring reflection itself under the dominance of immediacy, and then to invoke the preceding development that all reflection-into-self is also positing. This Hegel tries to think in the idea of essential immediacy, which prevails in the idea of essence. The task here is to think of reflection itself as externality in which externality is not reflection's relation to its presupposition (e.g., to some necessary but ultimately fortuitously given context), but rather its internal condition.27

Though Henrich does not himself apply this pattern of reasoning to the discussion on ground, it is clearly both applicable and in fact motivated by that later argument. The basic move is twofold: First, to understand that which is the object of reflection precisely as a form of immediacy—that is, to think of it as given and not inherently conditioned by anything outside of it. However, this is not just an object that is found in that way, but one that has been formulated with just this character of being sufficient and unconditioned in mind. Thus the second move is to suggest that precisely in so far as the object is presupposed by reflection, it is in fact constituted by it. All presupposition is revealed to be reflection-into-self, as the interpretation of that which is presupposed leads back to the idea for which that which is presupposed serves as a resource for expression. Precisely because the presence of oxygen is presupposed by the explanation of the fire it is constituted by the explanans: the candle makes the oxygen relevant. And further this positing is an act of self-constitution or self-positing, so that the ground produces itself in its own conditions. Somehow in constructing the presence of oxygen or the intention of the spouse, the tipped candle and the child's intention must constitute them as reflecting their own nature, that is, as resources for their own expression. Let us take these moves in order, and try to make some sense of them.

(p.103) The condition is a presupposition: this first Hegelian move is easy to understand intuitively, since if we think of the ground loosely as a kind of cause, then we can recognize that no cause is necessary and/or sufficient on its own, but only in the context of conditions that are therefore presupposed (in the sense that the reference to them is implicit rather than explicit) by the ground-relation between cause and effect. But our intuitive understanding gives out in the face of Hegel's emphasis on the identity of content between the condition and the ground, since we generally think of conditions and grounds as separate facts that at most conspire together towards the production of effects. In contrast, Hegel's emphasis is on the way in which any condition can be a ground, and vice versa. This possibility previously had to be understood in terms of an external perspective that would be brought in to explain why A was the ground and B the condition, rather than vice versa, but now Hegel again tries to internalize that perspective in the ground-relation itself, while nonetheless acknowledging that different elements of the content may be picked out as explanatory. Hegel must somehow show that the prioritized, vertical, asymmetrical structure of the ground—which is essential to it as a locus of responsibility—is maintained in the anarchic, horizontal and symmetrical relation between conditions that come together as a heap. The ground must be self-active even as it necessarily generates the conditions that allow for the possibility that some other element may be the true locus of responsibility, and furthermore it must do this in such a way that its own status as a locus of responsibility is conditioned and thus constrained by these other loci. This is the second move: in so far as the object is presupposed by reflection, it is in fact constituted by it.

As abstractly considered, this goes far beyond the counterintuitive to bring us to the very precipice of unintelligibility. But as is frequently the case, here Hegel has the phenomenal tiger by the logical tail. Very generally, this doctrine receives a concrete formulation in Hegel's insistence that great expression requires self-limitation (e.g., PR§15Z); one must mark out a domain in which other agents will take the lead in order to create something meaningful within one's own domain. More specifically concerning explanation, we can make some sense of Hegel's claims by returning to the example of the child's birthday drawing for her mother. The co-occurrence of the intention and the action in the spouse's case is in one sense independent of its status as licensing the explanation in the case of the young child (it is presupposed rather than an expression of the nature of that grounding relation in the latter case). In this sense, its presence is a supporting condition for the explanation of the child by providing a way in which the child might have come to understand the value or even possibility of altruistic intentions, and even more generally as part of the general background understanding we all share about the emotional dynamics of human families. On the other hand, it can become a resource for the child's self-expression (and for our understanding of the child), since it is something in the child's environment that she can pick up and make her own through her specific action. And yet to do so is in fact both (p.104) to articulate and isolate the altruistic intention as a relevant background condition, and to form that condition into something specific by her own action on it in producing the drawing. This process of articulation is perhaps difficult at first to see, but is in fact presupposed by our judgment that the child's intention counts as a locus of explanatory responsibility only if it is something that the child models rather than having the intention thrust upon her. Otherwise, the proper explanation will be in terms of the spouse's intentions rather than the child's. Here we see the way in which grounding requires something more than mere presence; it requires the identification of a locus of responsibility in such a way that certain background conditions become relevant. It makes an important human difference whether the spouse or the child's intention is the ground of the drawing presented to her, and both explanations are compatible with the child having the intention as she was making the drawing. In terms of the condition itself, the child's actions may in fact give the spouse's earlier actions a determinate meaning, in that they now become the circumstances in which the child learned of the availability of altruistic intentions. Since we have no direct access to people's intentions, but only through the interpretation of their actions, even speaking of the spouse's intention as a background condition is to construe certain prior events in a certain way, as having a certain essence or significance. In a sense made relevant by the centrality of explanations to the intelligibility of the world, the child's action on the ground of the altruistic motives modifies the spouse's earlier actions (i.e., it posits them in the sense of specifying them in external relations). Further chapters will try to spell out in some detail what this means in terms of the production of action, but they will follow this basic pattern that depends on the identity of grounding qua positing of the grounded with grounding qua presupposition of a condition.

To come back to the more general train of Hegel's argument, the distinction between the ground (explanans) and grounded (explanandum), on the one hand, and between the ground and the condition, on the other, is crucial for avoiding the double-aspect style regress, since it is supposed to give teeth to the notion that this additional feature is not understood as a ground but in a fundamentally different way. Because of this missing structure, the condition is not itself a ground at all, because it is not a “determining” ground. When Hegel writes that “something is not through its condition; its condition is not its ground,” he means that the condition is not an account of the being of the explanandum in the sense that the ground is (WL315/SL471).

This is perhaps easiest to see in the case of causal explanation, where the explanans might be split into a set of unnecessary but sufficient conditions and the cause itself whose effectiveness is understood in terms of being a necessary but insufficient member of that set.28 In the cases of the young child and the spouse, we pick out a different (p.105) element of that set of conditions as explanatory, but that explanatory status depends indirectly on the other elements of the set which are, in the ordinary cases, not explanatory at all. Hegel, however, makes an even stronger claim, which is that if the set of conditions is exhaustively described it must include the explanandum as well; that is, it must include the entire content of the ground-relation. Once again, this claim depends for its justification on the idea that explanation is a kind of entailment relation, and thus that the only way such a tight connection can be drawn between the premises and the conclusion is in terms of an exhaustive set of conditions for the activity of the explanans. In the example of causal explanation, this claim is supported by the notion that it is impossible to specify the set of sufficient background conditions for a cause without making reference to the effect itself, that is, without including something like “and nothing happens which prevents the cause from causing the effect.” This part of the explanans is only justified by the obtaining of the explanandum. This is to say that the relevant conditions are in fact posited by the cause (ground) itself in having produced the effect (grounded). But this form of argument appeals to the subjective limitations of finite human subjects engaged in the practice of explanation; though perfectly proper to the philosophy of science, it is unavailable to Hegel in the Logic. The official argument of the Logic brings us back to the form of argument reconstructed by Henrich, namely to argue that the condition is only relatively immediate, that is, it is posited as immediate in relation to ground (WL317/SL472). It is only relatively immediate, and this makes the whole ground-relation absolutely unconditioned, because now that which seemed to be an externality is internalized. In an important but deeply obscure passage, Hegel tries to clarify this in terms of the relation between positing and presupposing:

In the conditioned ground, the in-itself is not only the appearing [Scheinen] of an other in it. The ground is the self-subsistent, i.e., the self-relating reflection of positing, and therefore the self-identical; or, it is in its own self its in-itself and its content. But it is also presupposing reflection; it is negatively related to itself and opposes its in-itself to itself as to an other, and condition, according to its moment of in-itself as well as according to its moments of immediate determinate being, is the ground-relation's own moment; the immediate determinate being is essentially only through its ground and is moment of itself as a presupposing. (WL317/SL473)

In this very abstract characterization, the absolute unconditioned—Hegel's term for what becomes of the explanatory relation when the condition is internalized within it—is analyzed in terms of absolute ground. In doing so, Hegel implicitly claims that the very structured and recursively formed relation between ground and conditioned is in fact a conception of that more basic concept of absolute (p.106) ground which was not much more than the simple sense that some locus of responsibility must realize itself in determinate ways in order underwrite the intelligibility of the world. But the contrasts between the two formulations are also striking, and reveal the extent of the recursiveness involved in this notion of reflection, and thus the way in which the later notion of the absolute unconditioned represents changes at the level of the very concept of an essence itself. In the initial notion of absolute ground, the idea was of course that in positing, some ground externalizes itself into the complicated world of external relations, but because the principle of that externalization was internal to the ground we could somehow see in the web of determinate facts a sign of the ground's priority and so interpret the former in terms of the latter. Here the presupposition involved is interpretation only in the sense of returning us to the ground in its original form. Rather than being recursive, this procedure might better be described as circular. But in the absolutely unconditioned, interpretation in terms of the presuppositions of the ground's manifestation primarily leads back to a set of conditions that qualify and constrain the ground just as much as they enable and promote it. That the condition can nonetheless be seen as internal to the ground requires that we shift our very sense of self-subsistence (Selbständigkeit) from that of an isolated essence at some remove from the external relations in which it manifests itself to one which maintains itself essentially in interaction with those relations. Since I suspect that in reality our sense of basic conceptual independence is fundamentally derived from our sense of the independence of human beings, it is fair to say that this sense of independence has been chided here. The external relations have been internalized, to be sure; but at some cost.29

This gives us the key for understanding how Hegel's view here is consistent with his rejection of the atomistic picture of constant conjunction in favor of the view that all content is only content in virtue of its structuring by ground-relations: even those contents that originate outside the expressive form of the ground-relation (1) are only isolated and made relevant in virtue of the ground-relations and (2) only find their determinate or explicit expression in the ground-relation. Though conditions must be presupposed to have a being outside of the ground-relation, both that being and their significance depend on their functional role in enabling ground-relations; it just turns out that in order to enable ground-relations they must themselves be ungrounded. Outside of that relation, they are the inarticulate background of both activity and (p.107) explanatory practice.30 Here Hegel has taken the two explanatory exigencies: external conditioning and sufficiency of the explanans, and turned them inside out. On the one hand, it turns out that the very externality of the condition requires the sufficiency of the explanans or ground at such a deep level that the condition itself takes on that sufficiency. On the other hand, it turns out that the sufficiency of the explanans requires not the elimination of the externality of the condition but rather its apotheosis; yet the only account to be given of that elevation is in terms of the activity of the ground in distinguishing between background and foreground.

An example may help to leaven the abstraction here: Consider the state of the lower atmosphere as the condition for rain explaining the wetness of the wheelbarrow.31 The conditions will have to include the moisture in the atmosphere, the location of the wheelbarrow, and the direction and force of the wind at that point. There is an important sense in which these conditions contain all the elements for the explanatory story, but without the narrative structure of movement, expression, or entailment that allows us to identify the relevant elements in the first place. But once we start to tell that explanatory story, the relevant regularities in the atmosphere come into focus. On Hegel's view, this is the way in which the earlier categories of the Doctrine of Being have their determinate application in our thinking—that is, the way in which later categories of the Logic stabilize earlier categories by providing contexts for their use. Hegel had argued that qualities and quantities cannot be independently identified, and in explanation he provides the contexts of inquiry in which such identification is possible. Hegel writes that, “accordingly the truth of determinate being [Dasein] is to be condition; its immediacy is, solely through the reflection of the (p.108) ground-relation which posits itself as superseded … the immediacy of being is accordingly essentially only a moment of the form” (WL320/SL476). Ground-relations are the proper form of the facts themselves, and the becoming of the fact is the way in which the indeterminate multiplicity of conditions is transformed into a unitary fact. There is also a much larger point here about reciprocity: because Hegel thinks that the conditions are the way in which the characteristics of being present themselves in reflective thinking and existence, and he holds that those characteristics require understanding in terms of the essences that form the organizational principles of reflective thinking, it is the condition's own act to make itself into a resource for the fact in its aspect as the active ground (WL320/SL475). It is interesting that the basis for the intelligibility of mutual interaction is already being laid at this very abstract level. The quantities and qualities of the world are seen to have their significance in virtue of the reflection on them that turns them into conditions for explanatory and expressive relations. When they are explicitly made into conditions, that happens by placing them in an explanatory or expressive story that isolates and articulates them. Yet qua conditions they lack that structure even though their general status cannot be understood except by reference to such stories.

This is analogous to the way that for Kant, sensory experience forms the material for judgment in the sense that it is pre-constituted in such a way as to provide objects for the conceptual discrimination and relation that are accomplished in self-reflective judgment. This pre-constitution takes place at the unreflective level of the imagination, as an anticipation of explicit judgments. It is in many respects not surprising that Hegel would produce a conception of explanation more closely aligned with Kant's doctrine of the imagination, as opposed to Kant's own solution in the Third Antinomy. But the formulation of the problem of ground is clearly influenced by the latter discussion as well, and in particular the way in which it presents a need to understand the way in which self-activity is connected with a realm defined by external conditioning. For Kant, of course, the key to solving this problem is distinguishing between the phenomenal level of external conditioning through efficient causation and the noumenal level of autonomy. Thus he can say that the effect of the noumenal free cause takes its place in the temporal series without being conditioned by it (KrV A450/B478). But Hegel was not the first to find Kant's solution here untenable, and the heart of the untenability is precisely the attempt to maintain the absolute nature of both self-activity and external causation, because this is what makes talk of their interaction (i.e., of that self-activity having effects in the series of conditions) so puzzling.32 Hegel's discussion (p.109) here of the way that the ground posits itself with respect to conditions and therefore enters into relations that it both initiates and by which it is influenced is a way to understand what must be the case with respect to free will if the Kantian solution is rejected and the difficulty that it attempts to dodge is faced clearly. Hegel's solution is rather to see the very external condition as itself constituted by its role in enabling the activity of the ground.

§4.5: The Infection of Internality by the Conditions

At the end of the first book of Alfred Döblin's novel Berlin Alexanderplatz, just as the main character Franz Biberkopf is getting his feet under him after having been released from prison, he is told the following story to temper his new enthusiasm for the possibility of self-directed agency:

A man once had a ball, you know, the kind children have, but not made out of rubber, of celluloid, transparent, and inside there are little lead shots. Children can rattle it and throw it. Then the man took the ball and threw it and he thought: there are lead shots in it, so I can throw it, and the ball won’t run any farther, it’ll stand still right on the spot I intend it to. But when he threw the ball, it didn’t go the way he had intended, it made one more jump and then it rolled a bit, about two hands sideways … You see, the ball don’t go the way you throw it and the way you want it to, it goes about that way, but then goes a little bit farther and perhaps a great bit, you see, and a little bit sideways, too.33

The important thing about this story with respect to Hegel's reflective conception of self-grounding is that the same lead shot that provides for the stability of the ball also introduces an anarchic element that makes the ball resist complete direction by the thrower of the ball. This is the best image I know for the idea that the internal determination of the ground is itself necessarily infected with the externality of the conditions in virtue of the reflective operation of grounding itself. If the product of reflective grounding is reflective, then it has an independence that can resist control by the ground.

Thus Hegel ends his discussion of ground in the greater Logic with an extended meditation on this fundamental change in the nature of both conditioning and explanatory sufficiency that is required to maintain the coherence of ground as a fundamental category of intelligibility. Once the radical nature of the change is understood, (p.110) Hegel suggests that “ground” and “condition” no longer capture the foundational quality of the category at issue; instead Hegel opts for the term “Sache,” which for our purposes we can translate as “the point” of expression, or “what matters” to its expressor.

One way to get at the nature of this change is to pull on the thread of mutuality that first makes its appearance here. Just as the more concrete mutuality of recognition in Hegel's Phenomenology is introduced by the notion that self-consciousness requires an object that can deny itself while maintaining itself, or turn itself into an object while remaining a subject, the point of expression derives in part from the fact that the condition, in taking up the independence required of it as an object of grounding, must take up the form of that reflection itself. And yet the form of that reflection is a kind of self-supersession including self-denial. This brings us back to Henrich's earlier point from chapter 2, section 2.4 that essential to reflection is the identification of that which is posited in reflection as reflection; reflection requires independence and stability in its objects, but the best account of that independence that Hegel has thus far is precisely reflection itself. For this reason, Hegel claims that

This immediacy [the condition] as superseded reflection is reflection in the element of being, which thus develops itself as such into a whole; the form, as a determinateness of being, goes on to multiply itself and thus appears as a manifold content distinct from and indifferent to the determination of reflection … The determinate being [Dasein], therefore, that constitutes the conditions is, in truth, not determined as condition by something else and used by it as material; on the contrary, it is through its own act that it makes itself into a moment of another. (WL319–20/SL475)

Here we have the extension of self-creation qua self-denial as transferred to the conditions out of the ground, so Hegel first presents the point of expression in terms of this mutual self-limitation that is oriented not by dependence on the other but in fact precisely by independence from the other. Here we can see Hegel as generalizing an idea from Kant, who argues that the category of community, having as its structure reciprocity (Wechselwirkung), does not generate a problematic, antinomial series precisely because its terms “are not subordinated to one another as conditions of their possibility” (KrV A414/B441). But here, Hegel wants to make this very point about conditioning itself in terms of the reciprocity between ground and condition. The conditions do not transform themselves into material for the ground in order to submit themselves to it as to a dominating power, but rather because that is the only form in which they can express their own nature; they become what they are by submitting to processes of transformation that they only partially control. In our earlier formula of reflection, the perspectives of both positing and reflection-into-self were decidedly internal: though (p.111) each input was transformed into something different, and was thus subject to external determination in the simplest sense of that term, the coherence of the conception of reflection in its objectivity was secured by the grounding of that process of external determination in the inner nature of the input itself. Then interpretation merely had to retrace these steps, seeing in the externally determined output the sign of the internal activity of the ground and recovering that ground from its external entanglements. But through the discussion of real ground that externality has grown in significance to the point at which the internality of the perspectives is necessarily compromised.

Hegel indicates the damage done to the self-activity of ground quite clearly in his use of the term “becoming” (Werden) to describe what happens to this activity once we come to terms with the necessity of the conditions. This choice of terms is significant given the way that Hegel originally distinguished reflection (as the mode of transformation in the Doctrine of Essence) from becoming (as the mode of transformation in the Doctrine of Being). Recall that at the beginning of the Doctrine of Essence, Hegel writes that

Essence is reflection, the movement of becoming and transition that remains internal to it, in which the differentiated moment is determined simply as that which in itself is only negative, as semblance. At the base of becoming in the sphere of being, there lies the determinateness of being, and this is relation to other. The movement of reflection, on the other hand, is the other as the negation in itself, which has a being only as self-related negation. (WL249/SL399)

In becoming, the input necessarily enters into the process that will change it, but it is changed beyond its control into something that is fundamentally other than it. Here is Hegel's description of the point of expression in terms of becoming:

The reflection of the ground supersedes the immediacy of the conditions and relates them, so making them moments in the unity of the point [Sache]; but the conditions are presupposed by the fact itself, which thus supersedes its own positing, or its positing immediately converts itself equally into a becoming … In this, the mediation as a return-to-self has vanished; it is the simple reflection that appears in itself [einfache in sich scheinende Reflexion], and groundless absolute becoming. (WL320/SL476)

Here, Hegel demonstrates a frank recognition of the way in which otherness that is necessarily contingent and even arbitrary with respect to the positing ground is internalized in the actual operation of that ground as a matter of course. Inherent in the ground's own self-expression is a loss of control and a release to the contingency of external determination. As a result, the possibility of interpretation as a simple (p.112) “return-to-self” is gone for good, reduced to the procedure of an uninformative tautology (formal ground). Interpretation now leads beyond the original input instead of back to it. Thus Lakebrink's portrait of ground as a kind of circular return to self in which beginning and end coincide is fundamentally mistaken; Hegel does not conclude from the non-being of the conditions that their externality simply vanishes, leaving a pure self-communion.34 Rather, the contingency of appearing (or “seeming,” Scheinen) comes to define the internal relation of the point of expression. In this respect, Miller's translation of “einfache in sich scheinende Reflexion” as “the simple, internal movement of reflection” leaves the crucial element out, which is precisely the awkwardness of “seeming” as a form of self-relation. The fact appears to itself, but is forced to interpret itself against the background of conditions and principles that it does not entirely control. Here one can see the logical form of Kant's claim that in empirical self-consciousness the human being relates to herself as appearance rather than as she is in herself; even our own self-expression to ourselves is mediated by externality to our own purposes.

When Hegel writes that “this reflection is accordingly the mediation of the unconditioned point [Sache] with itself through its negation” (WL320/SL476), it is tempting to assimilate this simple phrasing to the simple introduction of ground as self-subsistent reflection. But in the course of the development of ground, the conception of the negation has been developed to include this contingency and externality in a more significant way. As Marcuse puts it, essence becomes posited “by proceeding from the determinateness which it finds given and by taking this ‘negativity’ into itself, but only to let go of it and to let it happen.”35 The ground is like the parent who enters her children's hyperkinetic and chaotic play in order to orient and constrain it; the price of success will be her hair tussled and her own focus waning. Rather than dominating the externality of the other by a return to self, as many would have it, the self is in fact infected by that externality, which becomes a kind of internal cost attached to all forms of self-expression.

As Hegel continually emphasizes, reflection is not something that happens within essence as in a substrate or at a location; essence just is the movement of reflection: “Being is only as the movement of nothing to nothing, and in this way it is essence; and essence does not have this movement within it, but is rather that movement as the absolute semblance [Schein] itself” (WL250/SL400). Essence just is the process of its own problematization, the process of raising questions about its own expression of itself, (p.113) and the relation between ground and condition is Hegel's most developed general statement of the terms in which this problem of expression is posed.

One more attempt to make this point, by returning briefly to the four-fold schema for reflection: the form of externality is no longer merely constrained to the process of reflection, but has come to be a part of the perspective of reflection as well. Here we see why the problem of mechanistic causation for freedom can still arise for Hegel long after the discussion of the finite and the infinite in the Doctrine of Being since that problem involves precisely the external perspective of the principles of causation (necessary laws) to those entities subject to them. Since the transformations of the self as ground are themselves not entirely grounded in the self, the self-activity here is essentially awkward and difficult—though it is not ipso facto alienating or submissive (any more than it is dominating or perfectly active); self-determination is work at its logical core, a talent at which one can be better or worse rather than a status that is achieved. Before turning to application of this model of explanation to the issue of action explanation, let me pause to consider Hegel's relation to contemporary erotetic theorists of explanation.

§4.6: Holism About Explanation

Hegel understands this coordination of internal and external determination through the notion of a totality that includes as its subordinate aspects both ground (explanans) and condition. The true “fact” is then identified with that totality. As a point about explanatory practice, this can be understood as the view that almost everything relevant about a subject must be known before a tractable request for an explanation can make sense—that is, a request such that the requestor has a basic sense of what kind of thing is required to fill out her understanding of the subject, and where that piece of information is small enough to be given as an explanation rather than as a longer process of instruction. So it is really the totality of understanding that enables and licenses both the explanatory request and its response; the (missing) part is parasitic on the whole. Hegel understands this totality as a truly unconditioned structure, “das wahrhaft Unbedingte; die Sache an sich selbst” (“the truly unconditioned; the fact in its own self”) (WL318/SL474).36 Here we have a holism with a particularly idealist spin. The (p.114) whole fact, including the explanandum, has a kind of identificatory and ontological independence. This ontological thesis is analogous to the contemporary erotetic analysis of explanation, which also endorses a kind of holism. Since the contemporary analysis here is erotetic—that is, informed by the logic of questions and answers—and among later philosophers influenced by Hegel Hans-Georg Gadamer has perhaps most prominently emphasized the importance of questions for understanding, I will also briefly note points of comparison between Gadamer, Hegel, and the erotetic analysts.37

On the erotetic analysis offered by Bas van Fraassen and Larry Wright, explanations are essentially answers to certain kinds of why-questions. On this view, explanatory relevance is not a two-term relation between explanans and explanandum, but rather a three-term relation that includes the context of the question. The nature of the context can be further specified by analyzing the presuppositions of a question, such as the obtaining of the explanandum itself, background understanding, and contrasts to the explanandum.38 There are at least four relevant points of contact between Hegel, the erotetic analysts, and Gadamer.

First, for both Hegel and the erotetic analysts, the difference between an explanatory cause and a mere condition does not lie in the fact itself. As Hegel puts it, the condition is indifferent to its status as a condition, and the same content is always involved in both conditions and grounds. The erotetic analysts go further than Hegel's discussion of ground by attributing the deciding factor to the context of the request for explanation as articulated through the notion of a contrast class to the explanandum.39 As Wright in particular points out, explanation works because out of an infinite number of potentially relevant conditions we are able to pick out a few that serve as the explanans. This is possible because while all of the conditions are potentially relevant to the explanandum simpliciter, not all are relevant to the difference between the explanandum and the contrast class. Now it is certainly true that the explanandum must be understood in a contrastive sense on Hegel's view, but the form of the context that Hegel thinks is appropriate to support explanations has not yet been developed at this point in the Logic. Though Hegel doesn’t have these explicitly contextual resources of the erotetic analysts, the fundamental trajectory of Hegel's thinking about explanation is, to use a phrase from Gadamer, that the meaning of the claim that something (p.115) is a ground “necessarily exceeds what is said in it,” and it is clear in Gadamer that this excess involves something similar to the contrast class implied by the question: “We understand the sense of the text only by acquiring the horizon of the question—a horizon that, as such, necessarily includes other possible answers. Thus the meaning of a sentence is relative to the question to which it is a reply, but that implies that its meaning necessarily exceeds what is said in it.”40

Second, van Fraassen emphasizes that questions also request an explanans that has a particular relevance relation to the explanandum and contrast class. Both Wright and van Fraassen argue that this relevance relation cannot be understood on only one model (e.g., on the basis of a counterfactual or modal contribution to the explanation, or on the basis of forming a deductive argument for the explanandum). Instead, relevance relations are particularized by the why-question that requests an explanation. For example, a question might request a goal, or an intention, or a cause. At first this seems to be an important difference with respect to Hegel's view, since in these terms Hegel's category of ground attempts to articulate the basic nature of the relevance relation: what it means for something to make a difference to another in such a way that it could be explanatory. But I think that appearances are deceiving here, and that the semblance of difference arises more from the difference in scope between the two projects and less from a difference in the particular conception of explanation. One of the general aims of Hegel's Logic is to differentiate and understand different forms of dependence, which we could understand in this context as different ways that one fact can make a difference to another. The implicit contrast class for the ground-relation of explanatory relevance are the relations spelled out in other categories of the Logic. For example, Hegel thinks that qualities depend for their determinate nature on contrasts with other qualities, which contrasts make the other qualities relevant. But these contrasts are primarily valuable for answering questions about what a thing is, and not why it is. Causal concepts are more appropriate for uncovering the kinds of differences relevant to the why-questions, and as such come after ground in the exposition of the Logic. But the form of the ground-relation is sufficient, Hegel thinks, to discriminate explanatory relevance relations from descriptive qualitative relations. In this way the ground-relation is an expression of the minimal structure required for the genus of explanatory relevance, whatever differences in species there might be. The rest of the Logic canvasses other possible models. But Hegel would agree, I think, with the erotetic analysts in holding that instantiation of the abstract form of ground is not by itself sufficient to account for the explanatory significance of a certain relation. In this respect Gadamer appears quite close to both the erotetic analysts and Hegel: “The essence (p.116) of the question is to have sense. Now sense involves a sense of direction. Hence the sense of the question is the only direction from which the answer can be given if it is to make sense. A question places what is questioned in a particular perspective.”41 For Gadamer, the question reflects the concrete hermeneutical situation—that is, our being affected by history—since that being affected gives us the specific questions we ask. The historical effect is part of how the object in question presents itself, and thus what is questionable about it.42

Third, Wright draws out the consequences of the necessity for a contrast class to the explanandum as follows: “Since indefinitely many items could have made a difference to any A [explanandum], making sense of choosing one requires an actual difference, that is, A itself must be a difference.”43 Since it makes sense to think of the explanans as a difference as well, we have the notion of one difference making a difference to another difference. To use Wright's example, the difference that my house had a tipped candle and not others makes the difference that my house burned and not others. At first this appears not to be Hegel's view, since it looks like a matter of one difference explaining the other, rather than Hegel's view that one part of the explanandum explains the whole. But these differences are both characterizations of the whole contrast that might be taken to be the basic explanandum (which contrast might be expressed in the case of my burned house by the existential “Why me?”). In this sense, the explanans is a partial description or characterization of the explanandum that clarifies the contrast in the explanandum between the topic and the contrast class, and in this sense the explanans is internal to the contrast class more completely specified. By more fully characterizing the contrast—by identifying one difference with another—we come to see why the obtaining of the contrast was not in fact the departure from normality that we had originally thought it to be.44 Gadamer also ties the arising of the question to the experience of surprise: “Logically considered, the negativity of experience implies a question. In fact we have experiences when we are shocked by things that do not accord with our expectations. Thus questioning too is more a passion than an action.”45 This is why for Gadamer, the concept of the question is key to what he calls “the logical structure of openness.”46

Fourth, on Wright's view, the vast majority of information relevant to the explanation must be presupposed, and the explanans plugs a small gap in an otherwise (p.117) sufficient understanding. Here the similarity between Gadamer and the erotetic analysts is perhaps most striking: “The openness of a question is not boundless. It is limited by the horizon of the question … Posing a question … implies the explicit establishing of presuppositions, in terms of which can be seen what still remains open.”47 On the erotetic analysis, in order for there to be such a gap the explanandum must represent a departure from expected normality, and it must depend on some other facts that could potentially serve as an explanation. Van Fraassen also emphasizes the fact that explanation takes place against a background of knowledge about the kind of phenomena that are to be explained. Hegel puts a similar point in a slightly derogative way when he allows that ground-style explanations are useful so long as they are only applied to the “immediate housekeeping needs of cognition.”48 The difference in tone should not obscure a similarity of conception between Hegel and the erotetic analysts, but it does point to an important difference between the projects in which those conceptions are embedded. Instead of proceeding from a desire to understand finite human activities as such, Hegel's Logic is the search for a category sufficient to express absolute, unconditional truth—ultimately, the structure of the form of life itself in which local explanations have their significance. On this axis, Gadamer's analysis of hermeneutic understanding is a project lying between those of the erotetic analysts and Hegel. In any event, this erotetic notion is a way of making sense of Hegel's paradoxical claim that on the one hand, the condition is the “simple self-identity of the ground” and yet the condition is not an account of the existence of the grounded. (WL315–6/SL471). This background understanding must remain inarticulate in order to support the practice of giving explanations at all. It must, therefore, remain present in an enabling role and yet take on an active and determinate role only in virtue of its use in explanatory relations.

To return to the question of the question of the criterion which distinguishes legitimate from illegitimate self-explanation, it should be clear that Hegel agrees with Hempel and Oppenheim that a general yet sufficient criterion is unavailable (though some necessary conditions can be specified). Since Hegel's discussion of ground does not directly compensate for this lack with the contextual or historical resources invoked by the erotetic analysts or Gadamer, one might think that the category of ground is of little philosophical import. But the constraints on explanation that Hegel does provide do rule out some initially plausible models of explanation. In the next chapter I discuss the way in which Hegel's argument rules out the view of action explanation common to Hobbes and Galen Strawson, and with it their regress argument against self-determination and Fichte's worries about naturalism.

Notes:

(1.) System der Sittenlehre nach den Principien der Wissenschaftslehre, SW IV, 229.

(2.) Fichte's Theory of Subjectivity, 141–42.

(3.) See Henrich, “Hegels Logik der Reflexion,” 298, and Angehrn, Freiheit und System, 50–51.

(4.) “Theoretical Explanation,” 156.

(5.) “Theoretical Explanation,” 165–66.

(6.) It is important to note that Hegel's view is broader: Sellars is concerned with understanding both the nature of scientific explanation and the ontology of scientific reasoning, but Hegel is concerned with understanding both the nature of explanation as such and the ontology of reason in general. See also WL304/SL458.

(7.) Cf. Pinkard's account of the role of “self-subsuming explanations” in the argument of the Doctrine of Essence (Hegel's Dialectic, 57–60).

(8.) This is to appeal to the objectivity of explanation, a point emphasized by Kreines (“Hegel's Critique of Pure Mechanism,” 38–43).

(9.) See, e.g., the beginning discussion of Absolute Ground (WL294–7/SL447–50).

(10.) Christopher Pincock helped me see this point.

(11.) “Studies in the Logic of Scientific Explanation,” 29–32.

(12.) “Studies in the Logic of Scientific Explanation,” 29.

(13.) Here I leave out the direction of the force, but this factor can be given the same treatment in terms of identity. Hegel's version of this example in his treatment of cause similarly emphasizes the quantitative identity that underlies the causal relation (WL399/SL561).

(14.) Demian Cho is responsible for anything that is correct about physical explanation in this paragraph. Again I emphasize that these are the terms of contemporary physical explanation in a broadly Newtonian style; the point here is to show the operation of identity within explanation in contemporary practice as a way of trying to understand the contemporary relevance of Hegel on this point, not to try to salvage Hegel's relation to Newton. I thank Patricia Curd for impressing on me the need to make this distinction clearer.

(15.) On Hegel's Logic, 87.

(16.) The project of distinguishing explanatory from non-explanatory characterizations is also central to Kreines's reading of Hegel's conception of ground, which he puts as the need to distinguish between explanation and description. An insightful defense of the necessity of a criterion to distinguish between these two both in general and for Hegel's idealism is in “Hegel's Critique of Pure Mechanism,” 52–54.

(17.) For an erotetic interpretation of Hegel's theory of identity, see Yeomans, “Identity as a Process of Self-Determination in Hegel's Logic.”

(18.) Hempel and Oppenheim, “Studies in the Logic of Scientific Explanation,” 29–30.

(19.) Longuenesse's discussion is quite helpful here, and interesting in its suggestion that the basic form of later Marxist critiques of Hegel is to be found here (Hegel's Critique of Metaphysics, 95–97).

(20.) Of course, Hegel is not accusing all explainers of sophistry. The standpoint of sophistry takes the ground-relation to be the last word on the nature of rationality. Sophistry comes from over-estimating the significance of the basic form of explanation.

(21.) Hegel's Critique of Metaphysics, 99.

(22.) Clark Butler insightfully discusses this issue under the heading of “The Principle of Determinability as the Main Positive Lesson of the Logic” as it relates to the nature of reason in Hegel's conception (Hegel's Logic, 130–32). See also Longuenesse, Hegel's Critique of Metaphysics, 98–99.

(23.) Hegel's Logic, 152. See also Longuenesse, Hegel's Critique of Metaphysics, 98, where this theme is connected with Kant's distinction between the categories as grounds of nature in general without specifying particular empirical laws.

(24.) For a discussion of this so-called “luck” objection, see Robert Kane, “Responsibility, Luck, and Chance.”

(25.) Butler sees this in the logic of essence as an expression of a reductionist determinist theology, and so sees Hegel's argument here as having an implicit historical reference that Butler discusses at some length, rather than seeing it developed in detail in the text of the Logic itself. I would be more inclined to supplement the argument in the Logic with that of the Phenomenology, but one interesting thing about Butler's procedure is it constructs a bridge between the traditional metaphysical interpretation of the Logic as the freedom of God and the more finite interpretation I am offering in this work. This includes not only his discussion of determinism but also his interesting suggestion that as models of will, “the overarching transition from the logic of being to that of essence is from the will to self-release (release from attachment to the finite self) to the will to infinite extension of the finite self's power” (Hegel's Logic, 160).

(26.) This is the “process of fission” that Bradley found in Russell's ontology; the label “double-aspect” comes from Nicholas Griffin, “Terms, Relations, Complexes,” 167. I admit that it is unusual that Hegel never explicitly appeals to such a regress, given his propensity for finding “bad infinites” in the most unlikely of places. At best it seems implied in his criticism of explanations from finite grounds (see WL318/SL474). Nonetheless, it seems like an obvious hazard lurking in the background, and the best philosophical justification for Hegel's next step in the argument. Pinkard appears to see the issue similarly (though phrased at a higher level of generality as an issue for essence itself, rather than specifically as an issue for determinate and external ground); see Hegel's Dialectic, 58.

(27.) Henrich, “Hegels Logik der Reflexion,” 300–02.

(28.) For the classic presentation of this view, see J. L. Mackie, “Causes and Conditions.”

(29.) Burbidge hints at this cost to the “primordial” self-activity of grounding as follows: “This immediate content has lost the arbitrariness that it had in real ground because it has been posited through the process in which the primordial relation dissolved itself” (On Hegel's Logic, 97).

(30.) Pirmin Stekeler-Weithofer grasps this necessity for placing the ground with respect to a background of acquaintance with its context as explanation through placing the explanandum with respect to an articulated system or model of knowledge or practical convictions. Explanation is therefore inherently systematic (Hegels Analytische Philosophie, 253–55). Pippin makes a similar claim in holding that in formulating the doctrine of essence as a problem of the ground of qualitative identity, Hegel wants “to show the dependence of the identification he is interested in (again the equivalent of what Kant would call the a priori synthesis) on some comprehensive, developing theory, one that can ground and defend the way in which this potential relativity is overcome” (Hegel's Idealism, 222). But understanding the systematicity at issue here in terms of a theoretical model strikes the wrong note, since it is essential to the system as condition that it be unarticulated. As the greater Logic's argument makes clear on this point, otherwise one encounters insoluble problems with regresses—not at the level of specific explanations in practice (this is harmless), but rather at the level of the conceptual articulation of explanation (this is fatal). For the same reason, Stekeler-Weithofer's grasp of grounds in terms of types or kinds is too concrete to illuminate Hegel's argument here. Butler's use of the notion of a “general description” does perhaps the same work without as many of the systematic commitments (Hegel's Logic, 151). Longuenesse comes closest to the interpretation offered here. At just that point at which she notes that complete ground involves taking the relation as a whole as the ground and therefore the totality of the circumstances as the ground for the essence itself, she notes that we are accepting the tautology of formal ground “because now the tautology is inseparable from a heterology” (Hegel's Critique of Metaphysics, 100).

(31.) Hegel's brief version of this example can be found at EL§153R.

(32.) For a historical discussion of this problem, see di Giovanni, Freedom and Religion in Kant and His Immediate Successors, 154–56. Longuenesse also sees Hegel's absolutely unconditioned as the successor to Kant's conception of reason's demand for the unconditioned as expressed in the Antinomy (Hegel's Critique of Metaphysics, 107–108).

(33.) Berlin Alexanderplatz 35–36, trans. 28–29 (translation modified).

(34.) Die Europäische Idee der Freiheit, 317–18. On the other hand, Carlson goes to the opposite extreme in characterizing ground as “self-erasure” and attributing to speculative reason the view that “there is no meaningful distinction between Conditions and Ground” (Commentary, 333 and 337).

(35.) Hegel's Ontology and Theory of Historicity, 77–78.

(36.) Although the conditions for any particular explanandum may be merely finite facts, and thus dependent on other grounds, this is something above and beyond the essential nature of the ground-relation. Some actions, for example, are certainly alienated. But alienation does not generalize as a result of the basic conditions for explanation, as the skeptic holds. Conceptually, the only condition of the condition is the ground for which it is the condition. See also EL§121Z.

(37.) For a more thorough discussion of the relation between Hegel and Gadamer, see Redding, Hegel's Hermeneutics, particularly chapter 2.

(38.) Bas van Fraassen, The Scientific Image; and Larry Wright, The Concept of a Reason and “Explanation, Contrast, and the Primacy of Practice.”

(39.) See also Michael Scriven, “The Logic of Cause,” 50.

(40.) Truth and Method, 370.

(41.) Truth and Method, 362.

(42.) Truth and Method, 300–01.

(43.) Wright, The Concept of a Reason, 24.

(44.) This is similar to Hempel's idea that to explain something is to respond to our surprise at its obtaining in such a way that in retrospect we could have predicted it.

(45.) Truth and Method, 366.

(46.) Truth and Method, 362–63.

(47.) Truth and Method, 363.

(48.) EL§121Z.