Grounding and Metaphysical Explanation

This article has been accepted for publication in Proceedings of the Aristotelian Society 116(3): 395-402, Published by Oxford University Press. Grounding and Metaphysical Explanation ABSTRACT: Attempts to elucidate grounding are often made by connecting grounding to metaphysical explanation, but the notion of metaphysical explanation is itself opaque, and has received little attention in the literature. We can appeal to theories of explanation in the philosophy of science to give us a characterisation of metaphysical explanation, but this reveals a tension between three theses: that grounding relations are objective and mind-independent; that there are pragmatic elements to metaphysical explanation; and that grounding and metaphysical explanation share a close connection. Holding fixed the mind-independence of grounding, I show that neither horn of the resultant dilemma can be blunted. Consequently, we should reject the assumption that grounding relations are mind-independent. This paper is about the grounding relation and the connection between grounding and explanation. Grounding is a relation of non-causal ontological dependence; a metaphysical determination relation which obtains between entities of various ontological categories including facts, properties, states of affairs, and actual concrete objects. Though discussion of grounding has become widespread, much of the literature about grounding is devoted to arriving at a proper characterisation of the notion. Difficulties arise because grounding is taken to be a metaphysical primitive – it resists reductive analysis. This leaves grounding vulnerable to the charge that it has no distinctive content, or no useful role to play. In response (as well as pre-emptively) grounding has been defended by appeal to intuitive examples of grounding; by highlighting the usefulness of a grounding relation; and by connecting grounding to other more familiar metaphysical notions. In particular, grounding has been connected to the notion of explanation (e.g. Kim, 1994; Fine, 2001; 2012; Schaffer, 2009; deRosset, 2010; Audi, 2012). To say that x grounds y is to provide some kind of explanation of y in terms of x. Defenders of grounding usually insist that the explanation on offer is somehow ‘metaphysical’. For example, Fine (2012, p. 37) introduces the idea of ontological ground as ‘a distinctive kind of metaphysical explanation, in which explanans and explanandum are connected...through some constitutive form of determination’. Unfortunately, the precise sense in which the relevant kind of explanation is distinctive remains somewhat opaque. I will argue that attempts to specify a notion of metaphysical explanation useful for elucidating grounding are bound to fail. I. Articulating the problem Here are three independently plausible, but jointly problematic theses about grounding and explanation: (i) (ii) (iii) There is a tight connection between grounding and metaphysical explanation Metaphysical explanation has agent-relative features Grounding is objective and mind-independent If grounding is objective and mind-independent, but metaphysical explanation has agent-relative features, then it seems the connection between grounding and explanation can’t be as close as the literature suggests. If the connection is to be maintained along with the objectivity of grounding, then metaphysical explanation must be similarly objective and mind-independent. Alternatively, grounding, like metaphysical explanation, must have mind-dependent features. Call this tension ‘the grounding-explanation problem’. Page | 1 This article has been accepted for publication in Proceedings of the Aristotelian Society 116(3): 395-402, Published by Oxford University Press. Why believe the three theses of the triad? I have said something about (i) already; it is a cornerstone of the grounding literature. The same is true of (iii). Grounding is supposed to be a paradigmatically worldly relation; grounding relations limn the structure of reality. The only controversial claim amongst the three theses is the second; that metaphysical explanation has pragmatic or agent-relative features. Let’s consider the problem in the form of a dilemma. Assume (iii). We must now either deny (i), or deny (ii). Since we’ve noted that (ii) is the more controversial of the two, we’ll start there. II. First Horn: metaphysical explanation Why accept that explanation, and in particular metaphysical explanation, has pragmatic features? A quick survey of the literature on ordinary explanation (which I do not have the space to rehearse here) reveals that all the major theories of explanation build features of the explanation-seeker into their account of successful explanation. Explanations are successful when they move the explanation-seeker to expect or understand the explanandum on the basis of the explanans. Expectation and understanding are agent-relative notions. Two things remain thus far unexplained. First, why is it successful explanation, rather than mere explanation that we are interested in? Second, why should we think that metaphysical explanation has pragmatic features, even granting that ordinary explanation does? To keep us on our toes, I’ll discuss the second of these first. Any defender of the connection between grounding and metaphysical explanation must have one of two things in mind. First, they might think that metaphysical explanation is like ordinary explanation, but in some kind of metaphysical context. Thus, the better understood notion of explanation helps elucidate the more problematic notion of grounding. Alternatively, they might think that metaphysical explanation is a special kind of explanation distinct from the ordinary conception. The problem with this second view is that it is the aspect of metaphysical explanation which is akin to ordinary explanation (e.g. intuitive understanding of terms such as ‘because’ and ‘in virtue of’; formal features such as asymmetry, transitivity, nonmonotonicity and hyperintensionality; and a sense of what it is to convey answers to why-questions) that allows us our intuitive grasp on metaphysical explanation in the first place. Success in arguing that metaphysical explanation is distinct from ordinary forms of explanation means that we lose our grasp on what metaphysical explanation actually is, and so it is very hard to see how the connection between ground and metaphysical explanation could shed light on grounding. The connection between grounding and explanation is only illuminating in so far as the relevant form of explanation is similar enough to our ordinary conception for us to understand it. If metaphysical explanation is like ordinary explanation but in a metaphysical context, then (assuming we can meet the challenge of specifying what this context is) the problem is that metaphysical explanation, like ordinary explanation, will have pragmatic features. What makes for successful metaphysical explanation will depend (to an extent) on features of agents. If the connection between grounding and explanation is as close as (i) in our original triad suggests, then grounding too has pragmatic features. But that straightforwardly contradicts (iii) – grounding relations are supposed to be entirely objective and mind-independent. So if metaphysical explanation is to elucidate grounding, then it can’t be a form of explanation entirely distinct from our ordinary conception. But if it’s like our ordinary conception, then it seems that metaphysical explanation and hence grounding has agent-relative features. Perhaps we can escape this troubling consequence by pressing the second of the points with which we began our discussion of metaphysical explanation; a distinction between explanation and successful explanation. Defenders of the causal theory of explanation (e.g. Lewis, 1986) claim that explaining some event involves citing a portion of its causal history. Successfully explaining some event involves citing a relevant portion of that history, where relevance varies with context. Can we not then say that the entire grounding chain of some entity metaphysically explains it, and avoid having to say that metaphysical explanation has Page | 2 This article has been accepted for publication in Proceedings of the Aristotelian Society 116(3): 395-402, Published by Oxford University Press. any pragmatic features? Perhaps, but this proposal won’t give us what we need to maintain that we can look to explanation to tell us about grounding. Defenders of the grounding-explanation link think that by reflecting on what metaphysically explains what, we can find out what grounds what. But all that is apparent to us is successful explanation. If we are to appeal to features of explanation to elucidate grounding, it must be the features of successful explanation that we appeal to (since these are the ones we are aware of). But as we have seen, what it is for an explanation to be successful varies with context, and so any useful connection between grounding and metaphysical explanation is a connection between grounding and an agent-relative notion of explanation. Attempts to reject the first horn of our dilemma are unsuccessful; metaphysical explanation does have agent-relative features. III. Second horn: the grounding-explanation connection It is unappealing to deny that metaphysical explanation has some agent-relative features, and so to resolve the grounding-explanation problem we must now consider re-evaluating the purported link between grounding and metaphysical explanation. I take it that there are broadly three ways to understand the connection. First, we can think of metaphysical explanations are tracking grounding relations. Second, we might think that grounding relations have a closer connection to explanation such that the grounding relation just is an explanatory relation. Both accounts of the groundingexplanation connection can find support in the literature. Alternatively, we can deny that grounding and explanation are closely connected at all. I’ll return to this third option at the end of the section. Take the first view. Audi (2012) insists that ‘grounding is not a form of explanation, even though it is intimately connected with explanation (p. 119)... an explanation...is something you can literally know; a grounding relation is something you can merely know about (p. 120)’. A metaphysical explanation might be a proposition expressing the grounding relation, but is not the relation itself. We rely on intuitions about what explains what in order to tell us what grounds what. Furthermore, we rely on what we know about explanation (formal features and so on) to give us a characterisation of grounding. If metaphysical explanations merely track grounding relations, then our reliance on features of explanation for our understanding of grounding is illegitimate. There is far more scope for error in our judgements about grounding than we ought to consider acceptable. There is an epistemic gap between judgements of ground and judgements of explanation that is not present if grounding just is a form of explanation, and it is especially troubling given the limited supply of alternative methods for discovering facts about grounding. Tracking conceptions of metaphysical explanation rob us of the special insight into grounding that metaphysical explanation has been supposed to provide. In response to such concerns, a defender of the tracking conception of the link might stress that the grounding-explanation connection is very tight. Perhaps there is something like a necessary connection between grounding and metaphysical explanation, so that whenever there is a grounding relation there will a metaphysical explanation that corresponds to it. But in so tightening the link, other difficulties arise. These are held in common with those who think grounding just is an explanatory relation. One of the main problems we encounter was discussed in the previous section; the agent-relativity of metaphysical explanation infects grounding, contradicting its assumed mind-independence and objectivity. A second problem is epistemic, and concerns the nature of the grounding relation. In the grounding literature, appeals to the grounding-explanation connection are often made in order to specify formal features of the grounding relation (e.g. irreflexivity, asymmetry, transitivity, hyperintensionality and non-monotonicity) (see e.g. Rosen, 2010). Our understanding of the nature of the grounding relation is coloured by our understanding of the nature of metaphysical explanation. A tight conception of the grounding-explanation connection means we cannot untangle agentrelative features from the rest. Page | 3 This article has been accepted for publication in Proceedings of the Aristotelian Society 116(3): 395-402, Published by Oxford University Press. Consider the common contention that grounding is asymmetric because explanation is asymmetric (e.g Audi, 2012, p. 102; Lowe, 2010). Symmetric explanations are considered bad explanations because they fail to provide us with any new information. It might be, however, that some cases of symmetrical grounding ought to be considered quite plausible (e.g. Thompson, 2016). If it were true that symmetries are ruled out of explanations on epistemic grounds, but there are cases of symmetrical grounding, then the epistemic features of explanation would be influencing our theorising about grounding in a way that led us to believe false things about the grounding relation (i.e. that it is symmetric rather than nonsymmetric). All of this suggests the possibility of severing the link between grounding and explanation, and maintaining that the two concepts are independent and distinct (though they may of course be similar in various respects). The upshot of this would be that it would not be safe to infer anything about the character of grounding relations from that of explanation, and that any appeal to our intuitive sense of what is explanatory in a given context ought to lend at most little credence to our judgements about grounding. The major problem with this strategy is that it robs us of one of the most prominent approaches to making sense of the grounding relation. As we are now well aware, a crucial way to understand grounding is via its connection to explanation. If the connection is severed, the notion of grounding is once more rendered obscure. Without the grounding-explanation link, our understanding of grounding must come from other sources that do not make appeal to the explanatory character of grounding.1 IV. Conclusion Not much attention has been paid to the nature of metaphysical explanation in the literature. Appeals to metaphysical explanation are made in order to shed light on the notion of grounding, but investigation into the nature of metaphysical explanation uncovers the grounding-explanation problem – a tension between our attitude to explanation, realism about grounding, and the purported link between grounding and explanation. To attempt to solve the problem, we held fixed the thesis that grounding is mind-independent and grappled with the resultant dilemma. Since we found no way to blunt either horn, the only thing left to do is to reject the assumption of the mindindependency of grounding with which we began.2 Naomi Thompson University of Southampton Avenue Campus, Highfield, Southampton SO17 1BF [email protected] 1 Such attempts have been made, but they are beyond the scope of this paper. In any case, without the grounding-explanation connection, the project of making grounding intelligible is severely damaged. 2 Thanks to Darragh Byrne, Nicholas K. Jones, Penelope Mackie, Alexander Skiles, Pekka Väyrynen and Alastair Wilson, audiences in Barcelona, Birmingham, and Hamburg, and at the 2015 SWIP workshop in York, and the 2015 Joint Session in Warwick. Page | 4 This article has been accepted for publication in Proceedings of the Aristotelian Society 116(3): 395-402, Published by Oxford University Press. References Audi, Paul. 2012: ‘A Clarification and Defense of the Notion of Grounding’. In Correia, Fabrice, and Benjamin Schnieder (eds.) Metaphysical Grounding: Understanding the Structure of Reality. Cambridge: Cambridge University Press. Clark, Michael, and David Liggins 2012: ‘Recent Work on Grounding’. Analysis, 72 (4), pp. 812-823. Kim, Jaegon. (1994) ‘Explanatory Knowledge and Metaphysical Dependence’. Philosophical Issues, 5, pp. 51-69. deRosset, Louis 2010: ‘Getting Priority Straight’. Philosophical Studies, 149(1), pp. 7397. Fine, Kit 2001: ‘The Question of Realism’. Philosopher’s Imprint, 1(1), pp. 1-30. Fine, Kit 2012: ‘A Guide to Ground’. In Correia, F. and Schnieder, B. Metaphysical Grounding: understanding the structure of reality. Cambridge: Cambridge University Press, pp. 37-80. Kim, Jaegwon 1994: ‘Explanatory Knowledge and Metaphysical Dependence’ Philosophical Issues 5, pp. 51-69. Lewis, David 1986: ‘Causal Explanation’ in his Philosophical Papers, Volume II. Oxford: Oxford University Press. Lowe, E. J. 2010: ‘Ontological Dependence’ in E. Zalta (ed.) The Stanford Encyclopedia of Philosophy [online]. URL = <http://plato.stanford.edu/archives/spr2010/entries/dependence-ontological/>. Rosen, Gideon 2010: ‘Metaphysical dependence: Grounding and reduction’. In Bob Hale and Avid Hoffman (eds.), Modality: Metaphysics, Logic, and Epistemology. Oxford: Oxford University Press, pp. 109-136. Schaffer, Jonathan 2009: ‘On What Grounds What’. In David Chalmers et al (eds.) Metametaphysics: New Essays on the Foundations of Ontology. Oxford: Oxford University Press, pp. 347-83. Thompson, Naomi 2016: ‘Metaphysical Interdependence’. In Mark Jago (ed.) Reality Making. Oxford: Oxford University Press, pp. 39-56. Page | 5