Presentist Social Functionalism—The Foundations

Abstract

This chapter develops a new account of social functionalism: presentist social functionalism. In particular, the chapter argues that social functionalism should be based on the actual bio-cultural selective or sorting pressures on the institution as it is now. The chapter shows how this idea can be deepened and underwritten with recent insights from evolutionary biology and the philosophy of biology. It also considers a number of the key objections to this kind of view that have been or could be proposed, and shows why they fail to be compelling.

3.1 Introduction

As I show in this chapter, it is possible to provide a plausible account of functional ascription in the social sciences that satisfies the desiderata just laid out. This account grounds the function of social institutions in those of their features that, in the current cultural environment, increase the chances of these institutions to survive or reproduce. Such a presentist account is shown to be able to make clear that the function of many social institutions may be more complex than hitherto assumed, and that existing discussions of these functions have often focused on the wrong kinds of empirical data.

As pointed out in the introduction, the best way to see how this account works is by putting it into practice. This will therefore be the target of the next three chapters. However, in order to provide a compelling foundation for these applications, it is necessary to begin by (i) spelling out the basic structure of the account, thus making clear how it meets the desiderata of the previous chapter, and (ii) responding to some possible worries concerning the structure of the account. The goal in what follows is thus to display the promise of account to improve on the existing accounts and make its fundamental machinery clear. The rest of the book will then deliver on this promise and show how this machinery can be put into practice in specific cases.

The chapter is structured as follows. In Sect. 3.2, I develop the novel form of functionalism that is the core of the present book, and show how this account meets the desiderata of the previous chapter. In Sect. 3.3, I respond to some worries for the account. I summarize the discussion in Sect. 3.4.

3.2 Presentist Social Functionalism

The account of social functionalism to be developed here uses some of the same core insights at the heart of the accounts of Bigelow (1998); Kincaid (1990); and Pettit (1996)—and indeed that of Merton (1968)—but spells these insights out in a different way. In particular, the present account shares with structural functionalism and the account of Pettit (1996) the idea that social functionalism should not be grounded in the history of the relevant social institutions (whether that history is design-based or selection-based in one sense or another). However, unlike the account of Pettit (1996), the present account does not look towards how a social institution would react to various counterfactual scenarios to ground functional ascription to it, but towards the actual bio-cultural selective or sorting pressures on the institution as it is now.¹ In this regard, the present account has more in common with that of Bigelow (1998) (and see also Kincaid 1990)—however, the way this idea is spelled out here is very different from the (rather brief) hints in Bigelow (1998).

Specifically, the key idea of the account to be defended here is that a social institution N has function F if it is now selected or sorted for F.² That is, the question at the heart of the account is: does feature F of social institution N make it more likely that N will survive or reproduce in the current socio-cultural environment? If it does, F is (part of) the function of N; if not, it is not. The present account thus makes the function of N dependent on those features of N—if any—that increase the expected survival or reproductive success of N in its current environment. More precisely:

Presentist Social Functionalism: Feature F of social institution N is (part of) the function of N if F makes it more likely that N will survive or reproduce in the current socio-cultural environment.

To understand this better, a number of further points about this account need to be noted.

First, the present account explicitly and intentionally groups together genuine selection—i.e. the heritable differential reproduction of social institutions—and mere sorting—i.e. the differential growth or persistence of social institutions. There are two reasons for this.

On the one hand, as noted in the previous chapter, a persistence-based view of selection may in fact be in line with evolutionary biological theorizing: in particular, there may be good evolutionary biological reasons for seeing mere persistence and differential reproduction as merely different instantiations of the same kind of phenomenon (Doolittle 2014; Price 1995; Fresco et al. 2018).

On the other hand, even to the extent that this is denied—that is, even to the extent that only differential reproduction of entities in a population is seen as truly a case of natural selection—for present purposes, this is not problematic. The goal of the present account is to ground the function of social institutions. As also noted in the previous chapter, while an appeal to the grounding of functions in biology is useful as a starting point, there is no requirement that these two groundings need to be the same. Social science has different (if at least sometimes related) explanatory goals than the biological sciences, and so some divergence in their core notions need not be greatly problematic. In the present context, the divergence results from the fact that it is not generally plausible to see social institutions as reproducing. However, the differential growth or survivorship of different kinds of social institutions is generally plausible. So, when it comes to social functionalism, the focus should be on the latter kind of process (though, as noted in the previous chapter, the former need not be ruled out a priori either): this is what allows us to see certain institutions as being more stable parts of a given society. Hence, the fact that social functions can be grounded in either the sorting or is made explicit on the present account.

Second, in virtue of the non-historical nature of the present account, stating that social institution N has function F should not and cannot be taken as an explanation of why the institution is in existence now. Rather, N’s having F is forward looking: it expresses why we should become more confident that N will continue to exist in the future. However, this is again in line with what a compelling version of social functionalism should look like (Bigelow 1998). As noted in the previous two chapters, one of the core aims of social functionalism is to explain why some social institutions are more stable parts of a given society than others. This is precisely what the present account can do. By showing that certain social institutions have features (their function) that make their survival or reproduction more likely, we are put in a position to hone in on the institutions that are more likely to be around in the future, too.³

Third, the role of counterfactuals is quite different on the present account as compared to that of Pettit (1996). On the present account, the only counterfactual that matters is whether the expected reproductive or persistence-focused success of social institution N would decrease if it did not have F.⁴ We do not need to consider whether N with F would continue to exist in all nearby possible worlds. We thus do not need to imagine all sorts of scenarios that may threaten N’s existence. The question is just whether feature F contributes to N’s expected survival or reproductive success as it is now. Typically—though, as will also be made clearer in the next section, not necessarily—this can be much more easily assessed, both theoretically (e.g. using background knowledge about the causal structure of the relevant economic system) and empirically (e.g. using carefully designed randomized controlled trials). In this way, we can sidestep the key problems that befall the account of Pettit (1996).

Fourth, the present account also sidesteps the problems of the historically-focused versions of social functionalism. Since the present account is not historical, it does not matter that many social institutions do not emerge after being “tried out,” in different versions, in one or several cultures. Similarly, it does not matter that social institutions are often relatively young and thus lack the kind of history that would make a purely persistence-based view of selection (as in Doolittle 2014) plausible to ground their function, or that their history would make such functional ascriptions ambiguous and unclear. Here, the past selective or other history of the institution does not matter at all to its functional ascription. In this way, the present account can take the best from historical selectionist accounts—namely their focus on actuality, rather than counterfactuals—but does not face their worries. In particular, it does not matter for the present account if there is not institutional variation in the current social environment: the present account asks just if feature F increases the expected survival or reproductive success of institution N. The comparison here is not with other existing social institutions or different time-slices of the current institution, but with a (possibly) counterfactual version of the present social institution that lacks F. This appeal to the present thus avoids the ambiguity problems plaguing some of the historical version of social functionalism.

For the same reason, the present account is also entirely consistent with the fact that the evolution of many social institutions is heavily dependent on chance. The core of the present account is just that N’s having F increases the expected survival or reproductive success of N. The present account does not claim that N’s having F fully determines the survival or reproductive success of N. It is not a claim about the actual evolution of N—just about one of the factors that influences this evolution in the current bio-cultural environment.

The fifth point worth noting here is that the present account can still allow for malfunction—a major requirement of any plausible account of what grounds functional ascription (Millikan 1984, 1990; Neander 2006; Fodor 1990). It is not like anything that N does is part of its function. Rather, only those features that contribute to its expected reproductive or survival success are part of this function. So, to see a social institution N (ritualistic pig slaughter, say) as having function G (to entertain children, say) might turn out to just be wrong: while G may indeed be a feature of N (children may be entertained by the ritual), unless G increases N’s expected reproductive or survival success, it is not its function. In this way, the present account can be seen to be a defensible way to ground functions.⁵

Sixth and finally, according to the present account, a social institution N could have more than one function.⁶ So, it may be that there is a set of features F₁ to F_n that, individually, increase the expected reproductive or survival success of N. In that case, all of F₁ to F_n would be the function of N. Also, it may be that some feature F increases the expected reproductive or survival success of N but only in the right circumstances—e.g. if there are not too many other tokens of N with F in the population (as in situations of frequency-dependent selection in biology: see e.g. Gillespie 1998; Futuyma 2009). These points will become important again momentarily. For present purposes, though, it is sufficient to note that the present version of social functionalism allows for—and indeed invites—complex, multi-faceted and dynamic functional ascription in the social sciences. However, importantly, it does so in a non-ambiguous way: it is not that it is not clear whether F₁ to F₂ should be taken to be the function of N, as was true on persistence-based versions of historical functionalism. That is, the issue is not that at some point in the past, F₁ increased the survival chances of N, and at some point in the past, F₂ did, thus making it ambiguous which of these is the current function of N. Rather, the claim is that both F₁ and F₂ increase N’s expected reproduction or survival success now. Despite its pluralism, therefore, the present form of social functionalism is non-ambiguous.

All in all: according to the present account, a social institution N has function F if F increases the expected reproductive or survival success of N in the present cultural environment. Functional ascription in the social sciences is thus shown to be ahistorical and to just be based on comparing the expected reproductive or survival success, in the current bio-cultural environment, of N with F to that of N without F. In this way, the present account is well placed to satisfy the four desiderata revealed by the discussion of the existing accounts. Of course, as noted before, the best way to really bring out the power of the account is by applying to some concrete questions in the social sciences. Before doing this, though, it is useful to note that all the pieces are in place here for a successful treatment of social functionalism.

3.2.1 The Account Should Be Focused on the Present

This is a defining feature of the present account. As just noted, this allows it to make sense of the fact that social institutions tend to be too recently created and lack the kinds of features—variation, heritability, and reproduction—to make appeal to historical forms of selection compelling. Apart from the fact that appealing to features of the present avoids the difficulties of justifying historical claims (Turner 2007) (as noted in the previous chapter), there is another benefit of this present-focus that is worth making explicit here. This concerns the fact the present account can avoid “swampman-”type worries (Millikan 1996; Neander 1996): a social institution N may be structured identically and be identically situated in its social system compared to another social institution N’, but lack the requisite history to have a function. On historical versions of functionalism, only N’ would have function—even though N is identical to it. The present account, though, is immune to this kind of worry: functional ascription independent of historical information, and we thus do not need to worry about the history of the institution—or what we know of it.

This matters, for while “swampman”-type worries may not be particularly pressing in biology (Neander 1996), they are not so easy to brush aside in the social realm. As also noted in the last chapter, social institutions can arise quickly, and it is entirely possible than an institution N₁ (making music with string instruments, say) appears in a society S₁ that has much of the same features F as a much older social institution N₂ in society S₂, which has furthermore been (culturally) selected for having F (producing continuously manipulable notes at a volume conducive to smaller gatherings). In those cases, it can be theoretically and empirically problematic (and not merely “counterintuitive”) if N₂ is said to have function F, but N₁ is thought to be functionless. From a social scientific standpoint, there can be good reason to treat these two cases similarly—especially if F makes both N₁ and N₂ more likely to be able to reproduce or survive, and thus be a continued element of their respective societies.

3.2.2 The Account Should Be Non-arbitrary

The selection and sorting pressures on a social institution N—i.e. the set of features F of N that increase the expected reproductive or survival success of N—are not observer-dependent in an arbitrary way. Of course, the grain of detail with which we describe N or F is dependent on our theoretical interests, moral, political, and aesthetic interests (a point to which I return in the next section). We can describe the function of ritualist pig slaughter as food security maintenance, agricultural yield maximization, or soil preparation. Similarly, which social institutions we choose to analyze—whether it is Pokémon trading card exchange or kawaii culture—depends on our personal, social, and theoretical commitments. However, the functional ascriptions we make are constrained by the nature of the social institutions in question, and cannot be freely chosen. They are thus non-arbitrary in the desired way.⁷

3.2.3 The Account Should Be Actualist, Not Counterfactual

By basing functional ascription on those features F of a social institution N that increase its expected reproductive or survival success of N, presentist social functionalism is inherently actualist. Of course, as also noted earlier, there are some implicit counterfactual commitments in the account: to say that F increases the expected reproductive or survival success of N, we are saying that if N had lacked F, its expected reproductive or survival success would have been lower. However, this does not take away from the fact that the present account is strongly focused on the actual world: we are not considering what would happen to N just in other possible worlds—we are just interested in what N does here, in the actual world. It is just that the latter has to be underwritten with some counterfactuals. However, it is the claim about the actual world that matters, not the features of N in various worlds that are independent from the actual world. I return to this point in the next section.

3.2.4 The Account Should Be General

There are no content-restrictions built into presentist social functionalism. It can be applied to all social institutions—even ones in the past. (Indeed, as will be made clearer in Chap. 6, the account can be extended to functional ascriptions to artifacts.) The point is just that functional ascription is relativized to a given time: the function of N at t is whatever set of features F of N increases the expected reproductive or survival success of N at t.

In all therefore: presentist social functionalism—the view that the function(s) of social institution N consist(s) in the set of features F of N that increase its expected reproductive or survival success—is well-placed to improve on the other accounts of social functionalism in the literature: it is non-historical, non-arbitrary, actual, and general.

While the remaining chapters of the book will illustrate this in more depth, it is at this point useful to show how the present account improves our understanding of a classic problem issue in the social sciences (especially anthropology): the function (or lack thereof) of Chinese foot-binding. The literature on foot-binding practices is very large and variegated (see e.g. Brown 2016; Ko 2007; Blake 1994), but for present purposes, it is sufficient to note that there is much controversy over whether foot-binding has a function at all, and, if it does, what its function actually is. Here, the presentist social functionalist treatment is helpful, as it can correct, clarify, and advance the existing discussion.

So, some authors have suggested that foot-binding is a “culturally arbitrary fashion” (Shepherd 2018; see also Ko 1997). Here, what the presentist account makes clear is that just because something is a “fashion” does not mean that it is functionless: wearing white on one’s wedding day (say) could spread due to its signaling commitment to traditional social values—and thus have this function—even though it is otherwise “culturally arbitrary.” Similarly, presentist social functionalism shows that the function of foot-binding can be complex and go beyond single-factor explanations—whether based on the pressures exerted by a mating market (Mackie 1996) or other social, ethnographic, economic pressures (see Shepherd 2018, for an overview). This is helpful not only because multi-factor explanations of the emergence and distribution of foot-binding are increasingly recognized as important (Tran 2020), but also because it makes it possible to see the function foot-binding as a combination of all of these different factors, and as something that can differ in different geographic regions or historic periods. This will become important again in the next chapter.⁸

3.3 Objections and Responses

So that presentist social functionalism can be seen to be truly credible contender in the discussion of social functionalism, it is necessary to respond to a set of objections that may come to mind. These objections fall into two categories. In the first group are objections that target the basic, defining features of the account. These objections tend to derive from similar worries for related views in other contexts, such as evolutionary biology or cognitive science. In the second group of objections are those that target specific features of the account, as it is spelled out here. These objections are thus unique to this context; while they might still have mirror images in other contexts (such as evolutionary biology and cognitive science), this is not their origin. While there are of course many possible objections of either kind that could be raised here, the following discussion will consider two of the most central objections falling under each of the above two classes. This ensures that this discussion remains manageable without sacrificing too much in the way of depth or breadth.

Before beginning this discussion, though, it is important to emphasize that aspects of the objections and replies will also be picked up again in the ensuing chapters. As noted above, the goal in this chapter is just to show that presentist social functionalism can get off the ground in the first place. In this sense, the discussion of the objection and replies in this chapter also helps situate and the discussion in the chapters to follow.

3.3.1 General Objections

The first foundational objection to consider here concerns the related claims that (a) presentist social functionalism has counterintuitive consequences and (b) presentist social functionalism answers the wrong sorts of questions. In particular, given the presentist focus of the account, it would seem that the account implies that the function of various social institutions is or could be very different from what it is commonly taken to be. For example, consider the police force again. If, in some particular social context, it turns out that the police force of a given state is persisting because it is being used to suppress dissent, then, on the present account, that would now be seen to be its function. This may seem odd, since we may also want to say that the function of the police really is to uphold law and order (say)—especially if that is what it was founded or culturally selected for in that past. It is just that (we may be tempted to go on to say), in the (supposed) case in question, the police force is doing something that is not in line with what it is for. Indeed, we may want to use the fact that the police force is failing to act in ways that match its historical “function” (i.e. that it suppresses dissent rather than upholding the law) as grounds for criticizing these actions. However, it may appear that this is not something that we can do with presentist social functionalism in the background: if suppressing dissent is what increases the chances of survival and reproduction of the police force, then that is its function. (As should be clear, this worry is related to the worries discussed in the context of structural functionalism in the previous chapter.)

Relatedly, as far as claim (b) above is concerned, with presentist social functionalism, we are unable to explain why a social institution operates in the ways it does. Since this account is not historical, it cannot answer the question of the origin of the institution. However, explaining why a social institution does what it does—e.g. how a given police force came to suppress dissent—would seem to be something we want to do.

In response, three points need to be noted. First, the counterintuitive implications of the account are in fact to be embraced. The account is revisionist in structure, and is not trying to capture “our” pre-theoretic intuitions about the functions of social institutions. This would anyways be difficult, since it is far from clear what these intuitions are, nor is it clear who the “we” is that is meant to share them (All humans? All citizens of a given state? All political scientists?). Rather, the presumption behind presentist social functionalism is that the function of a social institution is a technical, social scientific concept that we can choose to structure in ways conducive to our social scientific investigations. Indeed, it is precisely this which has led to the formulation of the four desiderata from the previous chapter: we want an account of social functions that is presentist, actualist, non-arbitrary, and general for the reasons laid out there. We are not strictly beholden to whatever we or someone else may have thought about the function of a given social institution. Of course, the latter is a useful data point to consider, but it should not be seen as a non-negotiable constraint. In other words, the goal of the project is not a summarizing description of the foundations of our social functionalist intuitions, but a theoretical account to be used in the social sciences and philosophy. The former may of course also be an interesting project; it is just not the one at stake here.

Second, this does not mean that it is not possible to criticize the way a social institution operates. It is just that the nature of this criticism changes. On the one hand, it is still possible to criticize a social institution for not fulfilling its function: we are just spelling out this function in a different way here. This is an important point to which the next three chapters return (especially Chap. 5). For now, though, it is sufficient to note that presentist social functionalism does not mean that social institutions cannot be criticized. It is true that we cannot criticize the above police force for not fulfilling its function by suppressing dissent. However, we can criticize the police force for doing other things—taking bribes (assuming to does so), say.

On the other hand and very importantly, instead of noting that a social institution does not fulfill its function, we can also criticize a social institution for having the function it does. So, if a police force persists because it is being used to suppress dissent, then we can say that we are now living in a society where the police force has a function it should not have.⁹

This can be illustrated well by the discussion surrounding human rights. Instead of starting with a fixed conception of what humans rights are and of who has them, we could say: if a given set of laws surrounding human rights is problematic—e.g. because it affords certain members of the species homo sapiens (African Americans, say, or females) fewer protections than others—then that is a reason for criticizing these laws. This would not be because the laws do not match existing human rights, but because the laws set up the wrong kinds of “human rights” (Darby 2003, 2004; Martin 1993). To be sure, the latter treatment of human rights is very controversial (see e.g. Feinberg 1992; Dworkin 1977; Nickel 1987), but there is no need to engage with this discussion here. Whatever may be the case when it comes to human rights—an issue that heavily depends on the nature of legal and moral systems—what matters for present purposes is just that these discussions illustrate the fact that a revisionist account of social functions is not required to be evaluatively quietist. A parallel treatment may be plausible when it comes to human rights, but it is certainly compelling when it comes to the much less morally and legally loaded question of the function of social institutions. We can criticize social arrangements for making it the case that certain institutions have certain functions just as much as criticizing them for making institutions act in ways that fail to match their function. Both forms of criticisms are legitimate—and both can be accounted for on the presentist social functionalist viewpoint.

Third, it is true that presentist social functionalist theorizing cannot be used to explain where a given institution is coming from. However, this is not greatly problematic. It is of course still true that we can inquire into the historical presuppositions and development of a given social institution. It is just that this is not built into the functional analysis of the institution per se, but treated as a separate inquiry. Importantly, it is far from clear that this division is at all problematic. As also noted by Pettit (1996), functionalist analysis is typically used to identify the stable social institutions—the ones that are likely to persist into the future. It is not at all obvious that what we also want to use social functionalist analysis to identify why a social institution came to be structured the way it is. Of course, we may want to know the answer to the latter question, but, at the very least, it is far from obvious that an account of social functionalism needs to provide it. On the contrary, not using social functionalist analysis to answer the latter question is not a major departure from the major forms of functionalist theorizing in the social sciences.

The second general objection to consider here concerns the fact that, on presentist social functionalism, determining the function of a social institution is very difficult. After all, it often seems very unclear which features of a social institution increase its expected reproductive or persistence success—and even where we have evidence about this set of features, it would seem to be subject to frequent change. A small change in the social environment can make it the case that what was once reproductive or persistence success-enhancing is no longer so, and vice versa. For example, consider a state where the police force is used to suppress dissent as well as to uphold law and order. What is the function of the police force? How can we tell which of these features increases its expected chances of reproduction or persistence? Moreover, it would seem that a small change—the promotion of a new police chief, say—can change this function quickly. In short: the worry is that, in virtue of its fundamental structure, presentist social functionalism is hard to apply to actual cases: due to the difficulties with ascertaining the often changing features of a social institution that increase its expected chances of reproduction or persistence in the present cultural climate, functional ascription appears to become difficult or impossible.

In response, there are again three things to note. First, since the proof of the pudding is in the eating, the best way to show the applicability of presentist social functionalism is by actually applying it. This is therefore one of the main goals of the next three chapters. As will become clear there, it turns out that the applicability of presentist social functionalism is actually quite high, and far from an impossibility.

The second point to note in response to this objection is that this concern is not specific to presentist social functionalism, but in fact affects all versions of social functionalism. As noted, earlier, much the same questions (concerning the features that increase an institution’s expected reproductive or persistence success) need to be answered for the historical accounts—only with the added difficulty that obtaining evidence of the past is often tricky. In the present, we can design experiments (at least sometimes) and generally access social institutions in ways we cannot when it comes to the past. For structuralist functionalism, the problem is slightly different, but still related: we need to determine the contribution an institution makes to a given social arrangement. This also need not be obvious (and, as noted earlier, it faces the arbitrariness worry that we can choose which social arrangement to focus on). Finally—and as also noted earlier—for Pettit’s counterfactual account, determining which features of an institution would make it the case that the institution would persist in cases where its existence were threatened is very difficult, too: assessing such counterfactuals is far from easy. Putting this the other way around: far from it being a problem for presentist social functionalism, the applicability of the approach is, if anything, a benefit of the latter when compared to the alternative approaches.

That said, the objection is correct in pointing out that determining the features that increase and institution’s expected chances of reproduction or persistence need not be easy. However, this is a challenge to overcome, not a deep objection to the account. In fact, the same can be said for most other sciences: determining the nature of dark matter or the genetic bases of complex traits is not easy either. This, though, does not mean it is impossible either. The same is true here: while it may be a challenge to determine the features that increase and institution’s expected chances of reproduction or persistence, doing so need not be impossible. For example, while it may be hard to determine whether a specific police force’s chances of reproduction or persistence are increased by its being used to suppress dissent or the fact that it upholds law and order, there may well be ways to assess this: we may be able to consider whether there are other institutions that uphold law and order (the armed forces, say), or whether the suppression of dissent in fact increases the pressure on the police force to be abandoned.¹⁰ The following chapters further underwrite this point; but for now, it is just important to note that the application of presentist social functionalism, while non-trivial, is not impossibly difficult either.

The third point to note in response to this objection is that it is not a vice of presentist social functionalism that the function of an institution can change quickly—it is a virtue. Given that the aim of functionalist analysis is the establishment of which social institutions are stable features of a society, and which merely transitory, being attuned to the dynamic interactions between an institution and its wider social environment is crucial: it helps determine which of these two sides particular institution falls on. Again, this is something that future chapters will come back to and make more precise. For now, though, it is just important to note that it is not problematic that presentist social functionalism sees the function of social institutions as quickly changing, but something that can just be accepted as a positive feature of the account.

All in all, therefore: there are no general reasons for thinking that presentist social functionalism cannot get off the ground as a compelling approach towards functionalist analysis in the social sciences. It is a revisionist proposal that does not try to capture “our” intuitions about social functions (whatever these may be) and which does not try to explain the origin of a given social institution (though it is consistent with such explanations). It also is no worse off in terms of its ability to ascertain the function of a given social institution (vis-à-vis rival accounts) and can just accept the fact that the function of a social institution can change quickly. With this in mind, consider some more specific objections to the details of presentist social functionalism, as it was spelled out in the previous section.

3.3.2 Specific Objections

The first of the specific objections returns to the fact that one of the key advantages of presentist social functionalism, when compared to Pettit’s counterfactual alternative, is that it is based on what is actually the case. However, as also noted earlier, it does not seem that the former approach can really avoid all counterfactuals: it seems we still need to assess what would have happened to the relevant institution if it did not have features F (which are meant to be its function). In what sense, therefore, does presentist social functionalism really improve over the counterfactual alternative? More generally: exactly what is the role of counterfactuals in this proposal?

In response, two points can be noted. In one sense, the objection is exactly right: presentist social functionalism does not forgo all reliance on counterfactuals. However—as also noted earlier—this is not surprising, since this would not be possible. Many claims about the actual world imply claims about counterfactual worlds (Lewis 1973; Stalnaker 1968; Kusch 2005; Fodor 1990): for example, causal claims and conditional claims—indeed, many relational claims. This includes claims about the impact a set of features has on the expected reproductive or survival success of a given social institution. The issue is not whether all counterfactuals can be avoided, but what the role is of the counterfactuals in the account. On Pettit’s approach, the function of a social institution is made dependent on the persistence of the institution in counterfactual—non-actual circumstances. On presentist social functionalism, the function of a social institution is made dependent on the persistence or survival of the institution in the actual world—it is just that this is being assessed by comparing it to the actual world.

This leads to the second key point to note here. While presentist social functionalism thus cannot avoid appealing to counterfactuals altogether, the kinds of counterfactuals it appeals to are different. In particular, they are different in two dimensions. On the one hand, they are different in content. On Pettit’s counterfactually-driven functionalism, all sorts of counterfactual scenarios would need to be considered. Would gold clubs persist in worlds with urban hiking clubs? What about in worlds with frolf (frisbee golf) clubs? What about in worlds with virtual reality metaverse golf clubs? By contrast, on presentist social functionalism, the only situation that needs to be considered is what the expected reproductive or survival success of the institution is in the nearest world where it does not have feature set F. There is thus less of a range of scenarios to consider here. This makes presentist social functionalism narrower in its counterfactual content.¹¹

On the other hand, the counterfactuals that underlie presentist social functionalism are easier to assess than those underlying Pettit’s alternative proposal. As noted in the previous chapter, it is often far from clear how a given social institution would respond to a specific counterfactual scenario. However, it is at least sometimes easy to determine whether the expected reproductive or survival success of a given institution would decrease without feature set F. We can compare the institution to relevantly similar institutions (including past versions of the institution in question) that lacked the feature. We may also be able to do a controlled experiment—or rely on a natural experiment—to assess the effect of removing feature set F from the institution. These are all familiar scientific inferences, and as such do not pose major methodological challenges (Sober 2008). This is especially so, since we do not need a fully precise estimate of the exact the expected reproductive or survival success of the institution with and without F; all we need is a comparative assessment of whether it is higher with or without F. This is easier to do than to assess the survival of golf clubs in a whole host of extraordinary counterfactual scenarios. Of course, it is true that assessing the expected reproductive or survival success of the institution with and without F need not always be straightforward. As noted earlier, there is no guarantee that this will not, at least at times, be a challenge that needs to be overcome. The point is just that, all in all, the counterfactuals are likely to be easier to assess on presentist social functionalism when compared to the fully counterfactual approach.

The final objection to be considered here picks up another point that was mentioned earlier: the individuation of institutions and features. Many things can impact the likelihood with which a social institution survives or reproduces, including the presence of other social institutions and various external features of the bio-social environment.¹² However, these do not necessarily become part of the function of a given social institution. Only if they are features of the institution could they be part of its function.¹³ In other words, it is built into the statement of presentist social functionalism that there is an existing universe of discourse that includes the relevant social institutions and their features. What presentist social functionalism allows us to determine is which (if any) of the given features of a given social institution is its function. This may be thought to raise two concerns, though: first, where are these individuation schemata coming from? Is it so clear that it is more useful to theorize about the function of “golf clubs” than about that of “sports clubs” or of “private golf clubs with membership fees that are greater than $100,000 per year?” Similarly, how do we know whether the features we should consider include “providing business networking opportunities,” “spending time with like-minded people from a similar cultural background,” or “learning about industry gossip?” Second and relatedly, it may be thought that it is one of the main jobs of social functionalism to help us determine what institutions to focus on in our social analysis, and what features of these institutions are particularly important. If—as seems to be the case—this is being presupposed by presentist social functionalism, this would seem to be a problem for the approach.

In response to this worry, there are three points that need to be noted. First, while the objection is correct in noting that presentist social functionalism depends on an individuation schema for social institutions and their features, it is not the case that it is the only account for which this is true. On the contrary: similar points hold for all forms of social functionalism (and indeed biological and social theorizing quite generally: see e.g. Baum 2013; Bertrand 2013; Laland et al. 2005; Odling-Smee et al. 2003; Dawkins 1982, 2004; Griffiths 2005; Griffiths and Gray 1994). In this regard, presentist social functionalism is no worse off than the other approaches mentioned in the previous chapter: it needs to be combined with some reasonable approach towards individuating society into different institutions with various features.

Second, the need for an individuation schema for social institutions and their features can be separately addressed from the question of the function and stability of social institutions. In fact, this is a very important point that returns to the objection concerning the explanation of why social institutions have the functions that they do. Social functionalism should not be seen as a panacea and thus not expected to be able to address all questions in the social sciences. The fact that presentist social functionalism needs to be combined with an individuation schema for social institutions and their features is not a problem for the account—it just means that this is one approach among others that can help us answer some of the questions we are looking to address. There is a tendency for functionalist approaches towards the social sciences to be seen as grand theories of everything that can “solve” the social sciences once and for all. This, though, is a tendency that ought to be resisted. Presentist social functionalism is designed to answer a specific set of questions: namely what the function of social institutions, so that we can determine which institutions are likely to be persistent elements of society in the future as well. The fact it is unable to answer every other question we might have—including the questions of why there are institutions of type X in society, why these institutions have features X, Y, and Z, etc.—is not problematic. It just means that presentist social functionalism needs to take its place among other theoretical approaches towards the social sciences.

Third, that said, this does not mean that presentist social functionalism is completely irrelevant for the determination of the right individuation schema for social institutions and their features. Rather, in line with work in non-foundationalist (social) sciences in general, presentist social functionalism can be used to help bootstrap such an individuation schema. We may start out with only a weakly supported such individuation schema, about which we are not sure. However, depending on the success of functionalist analysis with this schema in the background we may become surer of the latter. This is similar to what is happening in other sciences: we may hypothesize that a disease is transmitted through airborne droplets, but not be sure that this is the case. We can then model the spread of the disease based on this assumption. If these models then turn out to be predictively or explanatorily successful, this provides us with a reason to think that the assumption they are built on—concerning the mechanisms underlying the spread of the disease—is reasonable (Levins 1966; Fuller and Schulz 2021). Of course, there are many details to this sort of inference (Orzack and Sober 1993; Weisberg 2006; Odenbaugh 2006; Odenbaugh and Alexandrova 2011), but for now, it is sufficient to note that, exactly how it is to be spelled out, it is a familiar feature of science.

The same is true here. We may hypothesize that a given culture contains an institution of private property with a specific set of features (regulated rights of access and use of resources; right to transfer these rights to others); based on this, we can then ask what the function of this institution is (maybe to regulate access and use of resources). If the latter turns out to be a compelling analysis—maybe because we correctly predict that the institution of private property will persist in the future—then this might provide us with a reason for thinking our initial individuation schema was reasonable. Now, none of this is to say that providing such an individuation schema is not a separate project from (presentist) social functionalist analysis. The point is just that separate projects can still inform each other—and this is likely to be the case here also. In this way, it is not only the case that the objection here may not be much of an objection at all (even if were to grant that social functionalist analysis should help us provide an individuation schema for social institutions and their features.

3.4 Upshot

This chapter presented and defended a novel account of social functionalism: presentist social functionalism. According to this account, the functions of a social institution are those of its features that increase the probability that the social institution will survive or reproduce. This account of social functionalism is shown to improve on the major alternatives in the recent philosophical literature: structuralist functionalism, historical selection-based accounts, and the counterfactually-based account of Pettit (1996). In particular, by being grounded in the actual circumstances prevailing in society, the account defended in this chapter satisfies all the desiderata laid out in the previous chapter: it is non-historical, non-arbitrary, and non-counterfactual.

The account is also shown to be able to deal with the key objections that can and have been raised against this kind of view. In particular: (i) the account is revisionary and thus not answerable to our intuitions about specific cases (whatever these are), nor is it designed to explain the origin of social institutions. (ii) The account is no harder to apply than rival accounts. (iii) While reliance on counterfactuals cannot be fully avoided on this account, this reliance is both narrower in content and easier to justify. (iv) While the account needs to be amended with an individuation schema for social institutions and their features, the account is consistent with many such schemata, and should be seen to be required to itself provide (other than implicitly through standard inter-theoretic scientific bootstrapping).

For these reasons, presentist social functionalism can be seen to provide a compelling and fruitful characterization of institutional purpose, at least on the face of it. The next three chapters make clear that this impression is justified by considering three particular applications of the account. Apart from their inherent interest, these applications are also interesting, as they further spell out the responses to the four objections laid out here.