Elder-Vass on the Causal Power of Social Structures

Abstract

In this review essay, I examine the central tenets of sociologist Dave Elder-Vass’s recent contribution to social ontology, as put forth in his book The Causal Power of Social Structures: Emergence, Structure and Agency. Elder-Vass takes issue with ontological individualists and maintains that social structures exist and have causal powers in their own right. I argue that he fails to establish his main theses: he shows neither that social structures have causal powers “in their own right” (in any sense of this expression) nor that they exist.

Keywords

causal power critical realism emergence mereology ontological individualism structure

1. Introduction

In his recent, thought-provoking and well-written book The Causal Power of Social Structures: Emergence, Structure and Agency (Elder-Vass 2010), the philosophically informed sociologist Dave Elder-Vass argues that “social structures,” such as states, companies, families, and universities, exist and have causal powers in their own right. The ontological status of such entities is a much-debated topic in analytic social ontology, but the issue of their causal impact is surprisingly little addressed.¹ Thus, by homing in on the causal power of social structures the book highlights a neglected but important issue within the field.² However, I will argue that Elder-Vass does not succeed in establishing his main theses; he shows neither that social structures have causal powers “in their own right” nor that they exist.

2. Principal Objections

When Elder-Vass says that social structures (which I shall refer to as social objects³) have causal powers “in their own right” he means that they have emergent causal powers or properties (causal powers and properties are equated; Elder-Vass 2010, 17).⁴ An emergent property, according to Elder-Vass, is a feature of an object “that is not possessed by any of the parts individually and that would not be possessed by the full set of parts in the absence of a structuring set of relations between them” (ibid.). He mentions the liquidity of water (pp. 17, 56), the power of plants to photosynthesize (p. 51), and the power of stars to cause their particles to emit light (pp. 59-60), as examples of such properties in the natural realm. From the social sphere, he gives the following examples: the power of “norm circles” (e.g., families and religious communities) to affect the normative beliefs and behavior of individual persons (chap. 6); the power of “interaction groups” (e.g., queues), “associations” (e.g., groups of skaters) and “organizations” (e.g., companies) to influence individual behavior in various ways (chap. 7); the power of organizations to produce more (e.g., nails) than individuals working alone (pp. 154-56); and the power of organizations to fire their employees (p. 73). He claims that many of these are cases of downward causation, with higher-level causes having effects at lower levels (e.g., pp. 59-60, 152, 172, 202).⁵ He contrasts emergent properties with “resultant properties,” which are “properties of a whole that are possessed by its parts in isolation, or in an unstructured aggregation” (p. 17). His favorite examples of natural and social properties of this sort are mass (pp. 17, 56) and demographic distributions (p. 84; cf. Elder-Vass 2007), respectively.⁶ According to Elder-Vass, although macroscopic objects are often ascribed both emergent and resultant properties (p. 18), only the former force us to postulate higher-level entities; resultant properties can be analyzed fully in terms of properties of the parts, irrespective of how those parts are organized or interacting (pp. 56-57). Moreover, Elder-Vass wants to distinguish two kinds of emergence: relational (or weak) emergence and strong emergence (chap. 2). Strong emergence cannot be explained, not even in principle (p. 30); the weaker relational emergence can be fully explained by reference to the parts of the whole in question and the way these parts are organized (p. 22). All of the examples of emergent causal powers of social objects that Elder-Vass considers are meant to be of the relational kind (this is true of the physical and biological examples too); he is skeptical about the notion that there actually are any strongly emergent properties.

At this point a fundamental question arises for Elder-Vass (I focus on the social domain): given that the putative emergent social causal powers can be fully explained and accounted for in terms of individual people being organized, or interrelated, in certain ways, why must we also postulate social objects with these causal powers? Why not rest content with the individuals themselves, doing all the causing in virtue of their being interrelated in the relevant ways? Elder-Vass’s answer is that the postulation of social objects does not add anything to this picture, because social objects are nothing but individuals (and possibly other things such as equipment, p. 157) interrelated in the relevant ways: “an emergent entity [i.e., an object with emergent properties, such as a company] is nothing more than its parts and their organisation” (p. 26; see also p. 192). To defend this deflationary stance on social objects (which he also takes on natural objects), Elder-Vass adopts what he calls the redescription principle (pp. 24-25, 57, 81, 155). According to it,

The lower-level account of H’s [H = the higher-level object in question] powers merely redescribes the whole [. . . ]. In other words, “upper- and lower-level accounts refer to the same thing, as a whole and as a set of configured interacting parts.” (pp. 24-25, emphasis added)

It can be seen, then, that social objects are saved from elimination by a simple verbal maneuver. Ontologically speaking, there are only individual people out there, interacting in certain ways, but they and their relations can be collectively referred to and described using the vocabulary of higher-level social objects.⁷

I find it difficult to see how Elder-Vass’s metaphysical position differs from ontological individualism, a view he takes himself to be arguing against—although under the (somewhat inapt) name “methodological individualism” (e.g., pp. 2-4, 8, 54-55).⁸ I presume that no sensible ontological individualist will deny that people can be organized in certain ways, interact, and enter into the kinds of relation with one another with which Elder-Vass is concerned, and thereby cause things they would otherwise not have caused. For example, Jon Elster — arguably both a methodological and an ontological individualist (see e.g., 2007, p. 13)—says the following about game theory: “nothing could be further from the truth, then, than the allegation that game theory portrays the individual as an isolated and egoistic atom […] interaction is of the essence of social life” (Elster, 1982, pp. 465, 477; see also Watkins, 1957/1994, p. 442). Thus, it seems to me that Elder-Vass’s conception of ontological individualists as advocates of the view that all social properties are merely resultant properties (e.g. pp. 143-4), i.e. properties that do not involve relations between individuals, leads him to attack a straw man.⁹

In any case, the question is whether Elder-Vass can have his cake and eat it too: that is, can the redescription principle be sustained? I doubt that it can, for the following reasons. First of all, predicates such as “is a state” and “is a corporation,” and names such as “France” and “IBM,” are ostensibly expressions that apply to singular entities: names are singular terms and the form of the copula in such predicates is singular. However, the ontological correlates of these expressions are, on Elder-Vass’s view, mere pluralities of individuals in certain kinds of interrelation. Thus, a name such as “IBM” fails to name, and a predicate such as “is a state” fails to be satisfied by, any singular entity. Elder-Vass must therefore be willing to endorse some revisionary semantics according to which ordinary names and predicates refer to, and are true of, respectively, pluralities rather than singular entities—a view which is out of keeping with the semantics presupposed in orthodox philosophy of language.¹⁰ He has not shown that such a revisionary semantics is tenable. One objection to it is its implication that there are true statements of the following form: the xs (suitably interrelated) are identical with a. Here, “the xs” is a plural referring expression (referring, for example, to a plurality of individuals) and “a” is the name of a putative (social) object. (In the formal mode, Elder-Vass’s thesis is that plural and singular expressions can be, and often are, coreferential.) Such identity statements, or rather the sentences that are construed to express them, are arguably ungrammatical: identity predicates can be flanked either by singular terms or by plural terms; they cannot be inserted between one of each sort (see, for example, van Inwagen 1994). Moreover, even if we ignore this first problem, “hybrid” statements of the kind countenanced by Elder-Vass seem to be in conflict with Leibniz’s Law. This principle says that if A and B are identical, then whatever is true of A is also true of B, and vice versa. The problem, of course, is that the xs are many but a is (ostensibly) one (cf. Lewis 1991, 87). One and the same thing cannot be one and many. As Bertrand Russell pointed out already in 1903,

To say that A is one seems to amount to much the same as to say that A is not of the form “A₁ and A₂ and A₃ and . . . ” [. . . ] It is important to realize that a whole is a new single term, distinct from each of its parts and from all of them: it is one, not many, and is related to the parts, but has a being distinct from theirs. (Russell 1903, 70, 141)

If Russell is right, then we must conclude that the relevant xs cannot be identical with object a, because if they were then a would be many and one, which is impossible.¹¹ This latter difficulty stemming from Leibniz’s Law could perhaps be avoided by denying that a is one—Elder-Vass could suggest that we merely happen to speak in this misleading way of certain pluralities.¹² However, Elder-Vass does not deny this. On the contrary, he refers persistently, and in my view incoherently, to social (and natural) objects—the referents of the relevant singular terms—as real wholes, that is, singular units (e.g., Elder-Vass 2010, 49-50, 66-67). For example,

Perhaps the most neglected sense of structure in the sociological literature is structure-as-whole. There are some theorists who have used structure in this sense, such as Nadel and Tönnies (Lopez and Scott 2000, pp. 47, 52). But it is more common to deny the existence or causal value of social wholes, or to assume that it is of little importance to the study of social structure. In the extreme case [. . .] some accounts of social structure “hold that there is no whole or totality separate from the structuring activities and practices that are engaged in by individual actors” (Lopez and Scott 2000, p. 5). Here structure-as-relations is held to exist without its structuring any concrete whole at all. This book [i.e., Elder-Vass’s book] argues that we can only make sense of social structure by reinstating [the] sense of structure-as-whole and identifying the social entities that can possess causal powers. These are the entities that are structured by structure-as-relations, that possess the properties sometimes misleadingly identified as structures themselves and that cause those (partial) empirical regularities sometimes misidentified as structures. [. . .] Their parts, in other words, are human individuals and the resulting assemblages are real entities. (Elder-Vass 2010, 81-82, 145, emphases original)

It appears here that even Elder-Vass himself tacitly assumes that the redescription principle is false: a social object is not a mere plurality of entities, suitably interrelated, but instead a real unit with other entities as proper parts.¹³ (Alternatively, he is reaffirming the ostensibly contradictory thesis that a “structure” is a singular object and a plurality. On either interpretation, the position adopted seems to be inconsistent.)

If Elder-Vass were prepared to admit sets or mereological sums to his “official” ontology, the situation would improve to some extent, because sets and sums are traditionally conceived as singular entities, not pluralities, even though they may have members or proper parts. He does occasionally speak of sets and aggregates (“aggregate” and “mereological sum” are often treated as synonyms in the philosophical literature), but it is unclear what ontological status he attributes to them.¹⁴ Howsoever that may be, if they are to have a chance to do their job as reductive referents of the relevant proper names and satisfiers of the relevant predicates we must construe them as having not just people as their members, or parts, but also the relevant relations these people bear to one another. Otherwise the sets/sums will exist irrespective of whether the people in question are interrelated in the appropriate ways.¹⁵ However, even if we allow that there are such sets or sums, the redescription principle cannot be sustained. (To clarify, the idea under scrutiny here is that terms for social objects, and terms for specific sets/sums of individuals and their relations, have the same referents.) The problem with invoking sets or sums here is that sets and sums are extensionally defined, that is, defined in terms of their members/parts, while the social objects in question typically are not.¹⁶ As Elder-Vass himself repeatedly observes, social objects like organizations are changeable entities: they can change their parts and the way they are structured (e.g., pp. 35-36, 161). We do not, for example, see a new state come into existence every time a citizen dies, or is born. It may even be stipulated in rules, laws, or articles of association, that certain social objects (such as companies and cabinets) can survive certain changes of membership, or reorganization of a circumscribed kind. Sets and sums, however, cannot change their members/parts. It would therefore be a mistake to try to identify social objects with sets or sums of people and their relevant interrelations.¹⁷

It seems to me, then, that if Elder-Vass wishes to avoid eliminating social objects—if he wants to hold that we can quantify over, refer to, and describe those objects truly—he must say that they are either: composed of individuals interrelated in appropriate ways (where the individuals and the relations may differ over time); or, if sets/sums are endorsed, constituted by sets/sums of individuals and their interrelations (with different sets/sums at different times). Composition is a many-one relation holding between a plurality of entities (such as several individuals) and a single thing (such as a social object) at a certain time. Constitution is a one–one relation holding between one object (such as a set/sum) and another (e.g., a social object) at a certain time. Neither is the same thing as identity (Lowe 2009, chap. 6; van Inwagen 1994). It should be noted also that the adoption by Elder-Vass of either of these positions would oblige him to abandon the redescription principle. This is because on these views, social objects are numerically distinct from the pluralities/sets/sums of people that compose/constitute them at certain times, although there is a high degree of overlap, that is, sharing of components or proper parts (cf. Wiggins 1968).

What would it be for a social object construed along these lines to have a causal power of its own at a time t? Something like the following, I suggest: it would have, or could have (in relevantly close possible worlds), a causal impact at t that is not merely the sum (or net effect) of the impacts of its proper parts interrelated as they in fact are interrelated at t. On this conception, we keep the parts and their organization at t fixed and we query whether the putative higher-level entity makes, or could make, an impact distinct from the sum of the impacts of its parts. I take it that something like this is what contemporary metaphysicians understand by the idea that a higher-level object has a causal power in its own right (see, for example, Merricks 2001, 2005). However, as has been thoroughly discussed in the literature on causation (especially in connection with the putative causal impact of the mind (e.g., Kim 2005)), it seems that such higher-level causation can exist, and be nonredundant, only if the causal closure principle fails to hold at all the lower-levels, assuming that higher-level phenomena supervene on lower-level phenomena (no difference at the higher level without a difference at the lower level). Consequently, if one thinks higher-level entities cannot be reduced/identified with lower-level entities but only supervene on them (e.g., are composed of them), and if one also thinks that at least some lower-level realm is causally closed, one is apparently committed to holding that higher-level entities cannot have any nonredundant causal powers of their own: their causal powers must be held to be either mere sums of the causal powers of their proper parts (organized as they fact are organized) or else mere over-determiners.¹⁸

Notice how the view expressed above differs from Elder-Vass’s conception of an object with causal power of its own, distinct from the powers of its parts. That conception involves having “a causal impact that is not just the sum of the impacts its parts would have if they were not organised into this kind of whole” (p. 5, emphasis added). It is somewhat difficult to make full sense of the proposal as stated, but it seems to me Elder-Vass is saying that we are to think of the parts randomly scattered, no longer interrelated as they in fact are interrelated, and see what causal impacts they then have. Such a counterfactual scenario involves a situation in which the higher-level object no longer exists because the required organization of its parts is lacking. Thus we are contrasting an actual scenario in which the object is supposed to exist with a counterfactual scenario in which it definitely does not exist, although its parts do. The idea, then, is that if the actual organization of the parts makes a difference to causation, then the higher-level object makes a causal difference (has a causal power in its own right) because the object is (according to the redescription principle) the parts organized in such a way. (In Elder-Vass’s terminology, the object has “emergent” causal powers.)

I think this view of “causal power in its own right” is diluted and uninteresting for two reasons: first, no one denies that the way entities are arranged and interrelated makes a difference to causation, and to suppose otherwise is to turn one’s attention to a straw man; secondly, if a “higher-level” object is identical with a plurality of “lower-level” entities with a certain organization (a position I have questioned above), the higher-level object no doubt makes a causal difference “too,” but for wholly tautological reasons.¹⁹ Thus, under scrutiny, this conception of an entity having a causal power in its own right turns out to be insubstantial. Elder-Vass’s notion of relational emergence is not as exciting as he takes it to be. Notice, for example, that, given the redescription principle, Elder-Vass’s relation of “downward causation,” obtaining between a higher-level entity and a lower-level entity, involves nothing more than a plurality of entities (e.g., individuals) affecting a single entity (e.g., an individual) at the same level of description or reality²⁰—who could possibly deny that the latter kind of causation regularly occurs (at least at some level of reality)? The interesting question, in my view, is whether, in the natural or social realms, there are any objects with causal powers in their own right in the substantial, first sense described above; but nothing in Elder-Vass’s book suggests that there are.²¹

3. Looking at the Examples

I want to give further credence to the last claim of the foregoing section by briefly discussing some of Elder-Vass’s examples from the social sphere. Naturally, I cannot go into great detail here, but consider, first, the case of norm circles. Elder-Vass (2010) writes,

The parts of the circle, I suggest, are the individuals who are its members. But what is the mechanism by which the circle (as opposed to simply the individuals [understood as unrelated]) generates this causal power [to influence individuals’ normative beliefs and behavior]? Although institutions depend on the members of the norm circle sharing a similar understanding of the norm concerned, emergent or collective properties cannot be produced by such formal similarities. It is the commitment that they have to endorse and enforce the practice with each other that makes a norm circle more effective than the sum of its members would be if they were not part of it. [. . . ] It is this commitment to endorse and enforce the norm that is the characteristic relation between the members of a norm circle. As a consequence of being members of a norm circle, then, these individuals act differently than they would do otherwise. Even if they held the same normative belief, they would not necessarily act in the same ways regarding it (either endorsing it so strongly or enacting it so frequently) if they were not part of a circle that shares a commitment to endorse and observe the norm. These relations, then, when combined with these sorts of parts, provide a generative mechanism that gives the norm circle an emergent property or causal power: the tendency to increase conformity by its members to the norm. (pp. 123-24)

I think it is plausible to assume, however, that the effect will occur as long as the individuals concerned have the beliefs and commitments mentioned, that is, are suitably interrelated, and are acting accordingly. The norm circle as such need not do anything in addition; indeed, it need not exist—taken, that is, as an entity “over and above” the mere plurality of individuals standing in these relations. What is important is that the individuals share these beliefs, attitudes, and commitments, and think they are parts of the same norm circle, although they might not in fact pick out a real entity when thinking this. If I am correct about this, it is false to say that the individuals would not act in the same way if they were not part of the circle. The relevant beliefs and commitments are psychological states of individuals (although plausibly relational ones generated by earlier interactions between individuals),²² and they do not give rise by any obvious route to a genuine whole with these individuals as its parts. At the very least, it has not been shown that such a norm circle would have causal powers in its own right in the substantial sense described above.

Let us look next at the queue example.

What causal role does the queue itself [an interaction group] play, beyond the causal contribution of the institution of queuing and the contributions of the participants as individual agents? One contribution would seem to be towards our understanding of the situation. When we see a queue, this makes a contribution to our realisation that this is a queuing situation [. . . ]. Furthermore, the existence of the queue contributes causally to our decision as to where to queue. [. . . ] As a consequence, the queue also has the more interesting emergent causal power to serialise access to the service concerned in a manner that substantially reduces the potential conflict and stress that might otherwise be generated in such situations. (Elder-Vass 2010, 148)

All the relevant causation in this situation appears to be exerted by individuals standing next to each other. The queue itself need not do anything extra. Indeed, there need not even be a queue out there, taken as an entity over and above of the mere plurality of persons interrelated in these ways. It is enough that the interrelated persons induce us to think that this is a queuing situation.

About individuals in organizations Elder-Vass (2010) says the following:

When a role incumbent does act in the role, she adopts behaviours that have been specified by the organisation, as a result of acquiring a normative belief or disposition: the belief that role incumbents ought to act as specified by the norms that make up their role. The role incumbent, that is, is altered as an individual as a result of accepting the role in the organisation and thus we have a case of [. . . ] intrastructuration. In such cases, we can say that the higher-level entity, in this case the organisation, acts through the individual; those properties that the individual acquires by occupying their role are essentially properties of the organisation localized in the individual. And in such cases, the behaviour of the role incumbent in the role is part of the behaviour of the organisation. [. . . ] Now, a methodological individualist might argue that this reduces the behaviour of the organisation to that of the individuals and there is no need for the organisation at all in this explanation. However, the role incumbents have the effects that they do when acting in these roles only because they are organised into this organisation. If there were no organisation there would be no such roles and the people would behave differently. Hence the emergent causal powers of the organisation cannot be eliminated from the explanation of this behaviour. (p. 158)

Again, the coordinated, norm-governed behavior of the interrelated individuals seems to take care of all of the output effects of the putative organization. (Incidentally, in my view coordination effects, together with task specialization, also take care of the nail production example mentioned in Section 2; cf. King 1999b, 212-13.) As to the alleged internal “intrastructuration” effect, this phenomenon can be explained by the fact that incumbents think they are parts of an organization and have certain roles to fill within that organization, with the result that their behaviors become altered and coordinated in certain ways. The organization itself, conceived as an entity over and above the plurality of role incumbents, need not exist; it can be treated as a fictitious entity (persona ficta), as indeed is often implied in legal contexts (cf. Searle 2010, 97-100).

Finally, what are we to say about the case in which an organization dismisses an employee (p. 73)? Well, it does not seem to be a case of causation at all. To begin with, such an act is typically executed by a role incumbent—an individual manager, say (as Elder-Vass acknowledges). And when the manager exclaims “You’re fired!” or signs a certain document on behalf of the company, she does not cause the employee to be dismissed. Rather the act constitutes, or is, the dismissal.²³

4. Conclusion

I conclude that Elder-Vass has failed to demonstrate that social structures have causal powers in their own right: he has shown neither that they have causal powers in their own right in the substantial, interesting sense discussed by contemporary metaphysicians, nor that they have causal powers in their own right in the diluted and misleading sense defended by Elder-Vass. The latter sense requires the truth of the redescription principle, but I have argued that this principle is utterly problematic and I have indicated that Elder-Vass himself at times appears to implicitly reject it. Consequently, Elder-Vass has failed to demonstrate that social structures exist: he has shown neither that they are entities over and above pluralities of suitably interrelated individuals (an ontological view which the substantial sense of “causal powers in their own right” arguably entails), nor that they can be identified with such pluralities (a position which Elder-Vass’s diluted sense of “causal powers in their own right” involves). However, in spite of these deficiencies I find Elder-Vass’s book stimulating and rewarding: analytic philosophers interested in social ontology—especially those who, like myself and the regular critical realist, find the Eleatic Principle (roughly: to be is to make a causal difference) plausible²⁴—will have to address in depth precisely the kinds of issue discussed by Elder-Vass in The Causal Power of Social Structures.

Footnotes

Acknowledgements

Thanks to an anonymous reviewer for many helpful comments and suggestions.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interests with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The writing of this paper was funded by The Swedish Research Council (Research Grant 421-2011-1694).

1

I am thinking of prominent publications in traditional analytic social ontology such as Quinton (1975-1976), Ruben (1985), Gilbert (1989), Searle (1995, 2010), Thomasson (2003), Tuomela (2007), and List and Pettit (2011), in which causal considerations are fairly absent. John Searle, for example, has a lot to say about deontic powers (rights and obligations), but says almost nothing about causal powers. Brief discussions of the topic appear, though, in Gilbert (1989, 274-88, 304-306) and List and Pettit (2011, 161-63). It should be noted that it is fairly recently that analytic philosophers have taken a genuine interest in the metaphysics of the social domain, as opposed to the epistemology and methodology of the social sciences (e.g., Popper, 1957/2002; Nagel, 1961/1979) and the justification of various institutions (e.g., Rawls, 1971; Nozick, 1974).

2

Of course, the topic of the causal impact of social structures has been debated for ages in the literature of the social sciences, a discussion that dates back at least to Émile Durkheim’s (1895/1982) and Max Weber′s (1921/1968). Elder-Vass (chap. 1) says that he is primarily influenced by thinkers in the critical realism tradition, such as Roy Bhaskar (1975/2008; 1979/1998), Margaret Archer (1995), and Tony Lawson (1997). Critical realists tend to hold, following Bhaskar (1979/1998), that social structures are causally efficacious and therefore real. As Elder-Vass points out, Bhaskar, in his turn, was influenced by Rom Harré’s non-Humean theory of causal powers (e.g., Harré and Madden 1975). Interestingly, Harré is a harsh critic of the notion that social structures are causally efficacious (e.g., Harré 2002; Varela and Harré 1996)—a fact insufficiently flagged in my opinion by Elder-Vass. For a historical survey of the concept of causal power within the critical realism tradition, see Kaidesoja (2007). For further criticism of the critical realism movement, see King (1999a, 1999b) and Sawyer (2001). For criticism of Elder-Vass, see King (2007), Porpora (2007), Varela (2007), Sawyer (2011), and .

3

I prefer this term because it indicates that we are talking about the (putative) social entities having the causal powers rather than the powers themselves. Elder-Vass uses the term “structure” somewhat ambiguously: sometimes it seems to stand for social entities or objects (e.g., chap. 1, pp. 81-82, chap. 9)—that is, groups of people, in Elder-Vass’s metaphysical scheme (p. 7); sometimes it appears to denote the causal power of social objects: “the concept of social structure refers to the causal powers of specific social groups” (p. 66; see also p. 86). On the second understanding, the title of the book becomes a bit odd: is the book really about second-order causal powers of first-order causal powers? I suppose not; consequently, I read “structure” in the first sense. The ambiguity is ironic given that Elder-Vass criticizes other sociologists for not using the concept in a clear and univocal way (chap. 4).

4

For historical and systematic surveys of the notion(s) of emergence, see Stephan (1992), and O’Connor and Yu Wong (2012). I will not try to relate Elder-Vass’s conception of emergent properties to historical formulations of the notion, but note merely that Elder-Vass takes himself to be providing a refined and synchronic version of Margaret Archer’s so-called morphogenetic theory of structural emergence (Archer 1995; Elder-Vass 2007; cf. King, 1999b; ). I will argue, however, that the rather thin concept of emergent properties that Elder-Vass defends fits badly with how contemporary metaphysicians construe emergent causal powers, at least when emergent powers are understood as causal powers that macroscopic objects have in their own right.

5

Along with many others writing on emergence (see, for example, ), Elder-Vass seems to restrict downward causation to causal relations holding between wholes and their parts (p. 59). For my part, I fail to see why this restriction is imposed on the concept. The crucial idea behind “downward causation” is that causation can go from macro to micro, and this notion does not by itself exclude the possibility that the effects in question involve micro entities that are not parts of the macro object doing the causing.

6

However, so-called “system mass” is in fact not additive, pace Elder-Vass. The mass of a bathtub filled with hot water exceeds the sum of the individual masses of the particles constituting it, owing to the relative motions of the particles (, 222-28). Thus the relations between the parts are important in the case of mass too. Elder-Vass seems to be operating with an outdated Newtonian conception of mass.

7

In his (2007), commenting critically on Elder-Vass’s (2007), Anthony King appears to have missed this verbal maneuver of Elder-Vass’s; at least, King does not highlight it (even though the redescription principle figures prominently in Elder-Vass’s 2007), and he suggests that Elder-Vass holds that ”structure” denotes a phenomenon separate from individuals interrelated in certain ways (, 214-15). However, as we shall see below, some passages in The Causal Power of Social Structures can be read as vindicating King’s understanding of Elder-Vass, because these passages appear to commit Elder-Vass to structures being real singular entities distinct from, or over and above, pluralities of interrelated individuals—a view which is incompatible with the redescription principle.

8

See Sawyer (2011) for a similar complaint; cf. also Sawyer (2001, 560-65) and Porpora (2007, 198). For the need to distinguish between methodological and ontological individualism, see Nagel (1961/1979, 542), Quinton (1975-1976), Sawyer (2001, 572), and , 2, 350-51); for example, an ontological individualist can very well be a methodological collectivist/holist. In fact, it seems to me that Elder-Vass, superficially, could be characterized as an ontological collectivist (“there are structures”) who is a methodological individualist (“structures and their properties can ultimately be explained in terms of individuals and their interrelations”). But as I indicate in the text, in the end I think his ontological position boils down to ontological individualism, because this is what the redescription principle and his “nothing more” assertions appear to entail. (But then again, as we shall see below, some passages in Elder-Vass’s book appear to commit him to genuine ontological collectivism, making Elder-Vass’s ontological position highly unstable.)

9

If one looks at what general eliminativists (i.e., those who deny that here are any composite objects, apart perhaps from humans and other animate entities) have written, one finds they do affirm that the relations holding among their mereological atoms are of importance in causation. Trenton Merricks (2001), for example, holds that particles may be arranged in certain ways (e.g., tablewise) and have causal effects, collectively or in concert, in virtue of these arrangements. For example, they may cause us to have certain experiences (e.g., “of” a table), which may induce us to say (falsely, on Merricks’s view) things such as “There is a table here” (cf. Eddington 1929, ix-xii; ). I can see no reason why an ontological individualist could not hold an analogous view of pluralities of persons. That is, it seems to me that an ontological individualist can coherently maintain that individuals can be interrelated “groupwise” without thereby composing a higher-level object, that is, a group. Of course, an ontological individualist who maintains this is implicitly rejecting the redescription principle.

10

I have in mind traditional philosophy of language as it is displayed in the works by Russell, Tarski, Strawson, Searle, Kripke etcetera; see for a collection of classical essays in this tradition.

11

For further critical discussion of composition as identity, see van Inwagen (1994) and Byeong-Uk (1999). For a controversial defense of the notion, see .

12

Notice, though, that this escape route is not open if the redescription principle is meant to entail that statements of the form “a is one” are literally true, when “a” refers to many objects suitably interrelated.

13

14

For example, “upper- and lower-level accounts refer to the same thing, as a whole and as a set of configured interacting parts” (pp. 24-5, emphasis added).

15

Moreover, to avoid the outcome that the sets/sums exist when the putative social entities do not we will probably also have to regard the relations as tropes rather than as universals. Tropes are often conceived as nontransferable, that is, as being necessarily instantiated by the individuals they in fact are instantiated by, but universals are not tied to specific individuals in this way (, chap. 8).

16

A further difficulty with sets is that they are generelly understood as abstract, platonic entities that do not exist in space and time. Mereological sums, on the other hand, are supposed to be concrete entities (in effect, concrete analogs of sets) with spatio-temporal existence (Goodman 1966, chap. II; , chap. 7).

17

One could try to deflect this objection by adopting stage theory, because stage theorists allow single entities to have several distinct persistence conditions (see Sider 2001, 200-201). However, stage theory is a revisionary theory of persistence with problems of its own; for a critical discussion of it, see Hansson Wahlberg (2008). In any case, in the present paper I am presuming the ordinary endurance theory of persistence—along with most sociologists, including Elder-Vass, I take it, even if this assumption generally is not articulated—according to which objects persist by being “wholly present” at distinct times as numerically the same three-dimensional entity (Johnston 1987). I should perhaps stress that Leibniz’s Law is compatible with enduring objects changing their properties and parts over time (see ). The problem posed in the text has nothing to do with that issue. Rather, the difficulty has to do with how a changing enduring object (such as a corporation) could be identified with an unchanging enduring entity (such as a set or sum of individuals and their interrelations). Answer: it couldn’t. Moreover, even if enduring social objects per impossible could be identified with such sets/sums it could still be questioned that social objects have causal powers in their own right; this is because it can be questioned that sets or sums have causal powers in their own right distinct from the powers of their parts/members (in this case, individuals suitably interrelated).

18

It is sometimes suggested that the causal closure principle (regarded as applying to some lower level, for example, the micro-physical level) can be coherently combined with the idea that supervenient higher-level objects have nonredundant causal powers of their own if we take the latter causation to be formal and the former to be efficient in Aristotelian senses of these terms (e.g., Emmeche, Køppe and Stjernfelt 2000); thanks to an anonymous reviewer for pointing this out.) For my part, I fail to see how such distinctions are supposed to help; for criticism, see . In any case, it should be noted that this kind of escape route is not open to Elder-Vass (were he to adopt the above suggested sense of macro objects having causal powers in their own right) because he apparently holds that power causation is of the same kind (presumably, of the efficient sort) all the way “up” (see chap. 3).

19

also finds Elder-Vass’s characterization of the notion of having a causal power of its own empty. Although I am with Tsilipakos on this, I disagree with much in his Wittgensteinian criticism of Elder-Vass and social ontology in general.

20

It is crucial to realize, then, that Elder-Vass’s theory in the end is reductive (so long as the redescription principle is assumed), although it is formulated (rather seductively) in terms of “emergence” and “downward causation.”

21

As far as I can see, such causal powers, if they exist, need not be unexplainable (“strongly” emergent). Here is a speculative example. System mass (see footnote 6 above) can be derived from the masses of the constituents of the system and their relative movements. If system mass (exceeding the sum of the masses of the constituents) has effects (e.g., gravitational ones) that are not merely effects of the masses of the particles and their movements, the system should have causal powers of its own in the substantial sense. Whether this is actually the case turns on how the features postulated in special relativity (mass, kinetic energy, total energy, etc.) enter into the equations of the mathematically complex general theory of relativity (see , for some discussion).

22

For discussion of related issues and mechanisms, see , chaps. 22-24).

23

As John Searle would put it: the act counts as a dismissal. That is, the example seems to fit his “X counts as Y” formula (Searle 1995, 40), as does the fact that a certain individual is a manager— that is, being a manager is simply a status function, involving certain deontic powers, that is assigned to certain individuals by individuals (cf. , 214-15).

24

This often-adopted principle is named after the Eleatic Stranger in Plato’s Sophist; for some discussion of its scope, see , 41-43).

Author Biography

Tobias Hansson Wahlberg is a senior lecturer and researcher in theoretical philosophy at Lund University, Sweden. His past research has focused on the philosophy of time and persistence. Currently, he is working on a project titled “Social Ontology and Theories of Persistence” in which he endeavors to show how theories of persistence bear on the issue of how social objects are related to their constitution bases.

References

Archer

1995. Realist Social Theory: The Morphogenetic Approach. Cambridge: Cambridge University Press.

Armstrong

D. M.

1997. A World of States of Affairs. Cambridge: Cambridge University Press.

Baxter

D. L. M.

1988. “Many-One Identity.” Philosophical Papers 17 (3): 193-215.

Bhaskar

1975/2008. A Realist Theory of Science. London: Verso.

Bhaskar

1979/1998. The Possibility of Naturalism: A Philosophical Critique of the Contemporary Human Sciences. London: Routledge.

Byeong-Uk

1999. “Is Mereology Ontologically Innocent?” Philosophical Studies 93 (2): 141-60.

Clayton

Davies

, eds. 2006. The Re-Emergence of Emergence. Oxford: Oxford University Press.

Durkheim

É.

1895/1982. The Rules of Sociological Method and Selected Texts on Sociology and its Method, edited by Lukes

New York: The Free Press.

Eddington

A. S.

, 1929. The Nature of the Physical World. New York: The MacMillan Company.

10.

Elder-Vass

2007. “For Emergence: Refining Archer’s Account of Social Structure.” Journal for the Theory of Social Behaviour 37 (1): 25-44.

11.

Elder-Vass

2010. The Causal Power of Social Structures: Emergence, Structure and Agency. Cambridge: Cambridge University Press.

12.

Elster

1982. “The Case for Methodological Individualism.” Theory and Society 11 (4): 453-82.

13.

Elster

2007. Explaining Social Behavior: More Nuts and Bolts for the Social Sciences. Cambridge: Cambridge University Press.

14.

Emmeche

Køppe

Stjernfelt

2000. “Levels, Emergence, and Three Versions of Downward Causation.” In Downward Causation, edited by Andersen

P. B.

Emmeche

Finnemann

N. O.

Christiansen

P. V.

, 13-34. Aarhus, Denmark: Aarhus University Press.

15.

Gilbert

1989. On Social Facts. Princeton: Princeton University Press.

16.

Goldman

A. I.

1987. “Cognitive Science and Metaphysics.” Journal of Philosophy 84 (10): 537-44.

17.

Goodman

1966. The Structure of Appearance. 2nd ed. Indianapolis, IN: The Bobbs-Merrill Company.

18.

Hansson

2007. “The Problem(s) of Change Revisited.” Dialectica 61 (2): 265-74.

19.

Hansson Wahlberg

2008. “Can I be an Instantaneous Stage and yet Persist Through Time?” Metaphysica 9 (2): 235-39.

20.

Harré

2002. “Social Reality and the Myth of Social Structure.” European Journal of Social Theory 5 (1): 111-23.

21.

Harré

Madden

E. H.

1975. Causal Powers: A Theory of Natural Necessity. Oxford: Basil Blackwell.

22.

Hulswit

2006. “How Causal is Downward Causation?” Journal for General Philosophy of Science 36 (2): 261-87.

23.

Johnston

1987. “Is There a Problem about Persistence?” Aristotelian Society Supplementary Volumes 61:107-35.

24.

Kaidesoja

2007. “Exploring the Concept of Causal Power in a Critical Realist Tradition.” Journal for the Theory of Social Behaviour 37(1): 63-87.

25.

Kim

2005. Physicalism, or Something Near Enough. Princeton: Princeton University Press.

26.

King

1999a. “The Impossibility of Naturalism: The Antinomies of Bhaskar’s Realism.” Journal for the Theory of Social Behaviour 29 (3): 268-88.

27.

King

1999b. “Against Structure: A Critique of Morphogenetic Social Theory.” The Sociological Review. 47 (2): 199-227.

28.

King

2007. “Why I Am Not an Individualist.” Journal for the Theory of Social Behaviour 37 (2): 211-19.

29.

Lawson

1997. Economics and Reality. New York: Routledge.

30.

Lewis

1991. Parts of Classes. Oxford: Basil Blackwell.

31.

List

Pettit

2011. Group Agency: The Possibility, Design, and Status of Corporate Agents. Oxford: Oxford University Press.

32.

Lopez

Scott

2000. Social Structure. Buckingham, UK: Open University Press.

33.

Lowe

E. J.

2009. More Kinds of Being: A Further Study of Individuation, Identity, and the Logic of Sortal Terms. Oxford: Wiley-Blackwell.

34.

Martinich

A. P.

, ed. 2001. The Philosophy of Language. Oxford: Oxford University Press.

35.

Merricks

2001. Objects and Persons. Oxford: Oxford University Press.

36.

Merricks

2005. “Composition and Vagueness.” Mind 114 (455): 615-37.

37.

Nagel

1961/1979. The Structure of Science: Problems in the Logic of Scientific Explanation. Indianapolis, IN: Hackett Publishing Company.

38.

Nozick

1974. Anarchy, State and Utopia. New York: Basic Books.

39.

O’Connor

Yu Wong

2012. “Emergent Properties.” In Stanford Encyclopedia of Philosophy, edited by Zalta

E. N.

http://plato.stanford.edu/entries/properties-emergent/

40.

Okun

L. B.

1989. “The Concept of Mass.” Physics Today 42 (6): 31-36.

41.

Popper

1957/2002. The Poverty of Historicism. London: Routledge.

42.

Porpora

D. V.

2007. “On Elder-Vass: Refining a Refinement.” Journal for the Theory of Social Behaviour 29 (3): 195-200.

43.

Quinton

1975-76. “Social Objects.” Proceedings of the Aristotelian Society 76:1-27+viii.

44.

Rawls

1971. A Theory of Justice. Cambridge: Belknap Press of the Harvard University Press.

45.

Ruben

D-H.

1985. The Metaphysics of the Social World. London: Routledge.

46.

Russell

1903. The Principle of Mathematics. Cambridge: At the University Press.

47.

Sawyer

R. K.

2001. “Emergence in Sociology: Contemporary Philosophy of Mind and Some Implications for Sociological Theory.” American Journal of Sociology 07 (3): 551-85.

48.

Sawyer

R. K.

2011, “Review of Elder-Vass, Dave: The Causal Power of Social Structures: Emergence, Structure and Agency.” JASSS 14 (3). http://jasss.soc.surrey.ac.uk/JASSS.html.

49.

Searle

J. R.

1995. The Construction of Social Reality. London: Penguin Books.

50.

Searle

J. R.

2010. Making the Social World: The Structure of Human Civilization. Oxford: Oxford University Press.

51.

Sider

2001. Four-Dimensionalism: An Ontology of Persistence and Time. Oxford: Oxford University Press.

52.

Stephan

1992. “Emergence—A systematic View on its Historical Facets.” In Emergence or Reduction? Essays on the Prospects of Nonreductive Physicalism, edited by Beckermann

, 25-48. Berlin: de Gruyter.

53.

Taylor

E. F.

Wheeler

J. A.

1992. Spacetime Physics: Introduction to Special Relativity. 2nd ed. New York: W. H. Freeman and Company.

54.

Thomasson

2003. “Foundations for a Social Ontology.” Proto-Sociology: An International Journal of Interdisciplinary Research 18–19: 269-90.

55.

Tsilipakos

2014. “Theoretical Procedures and Elder-Vass’s Critical Realist Ontology.” Philosophy of the Social Sciences 44 (6): 752–73.

56.

Tuomela

2007. The Philosophy of Sociality: The Shared Point of View. Oxford: Oxford University Press.

57.

Udehn

2001. Methodological Individualism: Background, History and Meaning. London. Routledge.

58.

van Inwagen

1994. “Composition as Identity.” Philosophical Perspectives 8:207-20.

59.

Varela

C. R.

2007. “Elder-Vass’s Move and Gidden’s Call.” Journal for the Theory of Social Behaviour 37 (2): 202-10.

60.

Varela

C. R.

Harré

1996. “Conflicting Varieties of Realism: Causal Powers and the Problems of Social Structure.” Journal for the Theory of Social Behaviour 26 (3): 313-25.

61.

Watkins

J. W. N.

1957/1994. “Historical Explanations in the Social Sciences.” In Readings in the Philosophy of Social Science, edited by Martin

McIntyre

L. C.

, 441-450 Cambridge: The MIT Press.

62.

Weber

1921/1968. Economy and Society. Totowa, NJ: Bedminster Press.

63.

Wiggins

1968. “On Being in the Same Place at the Same Time.” Philosophical Review 77 (1): 90-95.