https://orcid.org/0000-0003-0634-1976
![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
ABSTRACT
The Causal Exclusion Problem is raised in many domains, including the metaphysics of macroscopic objects. If there is a complete explanation of macroscopic effects in terms of the microscopic entities that compose macroscopic objects, then the causal efficacy of the macroscopic will be threatened with exclusion. I argue that we can avoid the problem if we accept that macroscopic objects are ontically vague. Then it is indeterminate which collection of microscopic entities compose them, and so information about microscopic entities is insufficient to provide a complete explanation of certain properties of macroscopic objects. After outlining this solution, I consider several objections.
Disclosure Statement
No potential conflict of interest was reported by the author.
Notes
1 The billiard ball is identified with a collection of molecules, but how we understand this depends on what we think about composition. We may take our references to a singular billiard ball to refer to a plurality of distinct molecules (if we take composition to never occur), or we may think that there is a singular billiard ball that is identical with this plurality (if we take composition to be identity). We are not committed here to a particular reductive thesis. For the sake of ease, I will assume the latter reductive thesis in what follows, and hence that objects are threatened with exclusion by the entities that compose them.
2 Attributing this to Jonathan Edwards, Kim introduces ‘Edwards’s dictum’ as follows: ‘There is a tension between “vertical” determination and “horizontal” causation. In fact, vertical determination excludes horizontal causation’ [Citation2005: 36]. Much of Kim’s discussion of the exclusion problem is governed by this, that there is a competition for what determines a given higher-level effect, and that the supervenience base that vertically determines it wins.
3 Framing exclusion arguments in this way allows us to use the idea that causal overdetermination is impossible (or rare), and it allows us to use a closure principle whereby physical effects have sufficient physical causes.
4 It is not as straightforward to show how this kind of overdetermination is problematic as it is in the case of direct causal overdetermination. For the latter, we can appeal to the intuition that effects almost never have two sufficient causes. Although this case is not direct causal overdetermination, it does involve two distinct explanations, and both explanations are at least partly causal. Exactly how to motivate why this is intolerable is an interesting question, but here I will take for granted that it is a problem, since both Jaegwon Kim and Trenton Merricks frame cases like this as leading to an exclusion problem.
5 Carmichael [Citation2011: 316] discusses a world with items dispersed in such a way that it is vague whether anything is composed from them. In such a case, it is equally true, for each of the dispersed items, that it is indeterminate whether it in part composes some further item. This is a case of what he terms a ‘homogenous case of borderline composition’, which is in contrast to the ‘heterogeneous’ cases with which we are more familiar, where objects have certain determinate parts and then candidate indeterminate parts in the penumbra. It is hard to think of homogeneous cases in the actual world, as most of our cases involve vagueness at the boundary (although perhaps a particularly wispy cirrus cloud might qualify, as one reviewer suggests).
6 Merricks claims that denying this premise entails emergentism, as the baseball must cause something that ‘its parts, working in concert, do not’ [Citation2001: 62], and he thinks that any such causation over-and-above something’s parts entails some emergent property. This may be so, and I will return to this in note 9. However, vagueness alone does not entail emergence, strictly given what Merricks says. The baseball does not necessarily cause anything that its parts, working in concert, do not cause. It may be perfectly acceptable to say that the parts of the baseball cause the window to shatter. But we still will not have said that a particular collection of atoms is capable of causing the window to shatter. After all, it is indeterminate which atoms are parts of the baseball.
7 If the speed of Ball2 is vague, then there is no fact of its exact speed to be determined. Then it seems that the molecules determine all of the facts to be determined, and so there’s nothing for Ball1 to determine. However, it’s a fact that Ball2 has a vague speed; no set of molecules is in a position to explain this fact; but the vague speed of Ball1 as it collides with Ball2 is in a position to explain this.
8 This may be stating her view in a way that is unfair to both of us. It is more accurate to say that the precisifications pick out possible worlds, and that it is unsettled which one is actualised. This at least makes it less clear that Ball2 would be strictly identified with one of its precisifications.
9 This may compel us to say that being vague is a property of macroscopic objects, and that this property is not vertically determined by any microscopic entities. I cannot see why proponents of the modal views would not accept this. If so, this may involve accepting emergentism, after all.
10 This involves appealing to a difference between the causation relation and the relation of causal explanation. Davidson [Citation1967], Lewis [Citation1986], and Beebee [Citation2004] take there to be a difference, where certain items (such as facts) may feature in causal explanations even if they do not count among the causes of the effect. For some, such as Mellor [Citation1995], however, causal explanations between facts are the central causal claims.
11 Related here is Kim’s endorsement of what he calls the ‘causal inheritance principle’, which is, roughly, the idea that if a property is realised by some physical base then the causal powers of that property are identical to those of its physical base [Citation1992: 18]. There is precedent for rejecting the principle outright (e.g. Pereboom [Citation2002]; Gibbons [Citation2006]), but I do not discuss this at length because, if there is no physical base of molecules to which the ball bears a determinate relation, the antecedent of the principle is not satisfied anyway.
12 For Thomasson, there’s no tension, as causation and upward determination are different and non-competing determination relations. Gibbons claims that if there is a tension then surely horizontal determination wins.
13 One way to secure the generative yet non-competitive nature of horizontal and vertical determination would be to fall back on answers to the exclusion problem that rely on the close connection between the excluded and the excluder. Such answers may claim that there is no overdetermination [Bennett Citation2003, Citation2008] or that the overdetermination is acceptable [Paul Citation2007]. These might be successful ways of answering the causal exclusion problem in certain contexts, but they are not necessary when the entity is vague.
14 The cases for which the vagueness solution is viable do not require (and perhaps must not involve) downward causation. However, my solution is compatible with the possibility of downward causation occurring elsewhere.
15 The question can be raised for artifacts generally. Billiard balls are also artifacts, which invites confusion, but I have been concerned with whether billiard balls qua macroscopic objects are excluded by microscopic objects. Another question is that of whether billiard balls qua artifacts are excluded by the acrylic balls that constitute them.
16 I would like to thank Shieva Kleinschmidt for encouragement, Rebekah Rice for commentary, and a colloquium audience at the 2018 Pacific APA meeting. Thanks also to Dan Pallies for helpful discussion.