This is a memo for an upcoming working conference on social ontology from a powers-based perspective. We were charged with setting out something that vexes us, that we are trying to sort out. All thoughts welcome! Sorry about the formatting glitches; I can’t seem to fix them.
- The problem
1a. I’m interested in whether or not social structures can be causes, in keeping with an anti-passivist account of causation, in which causing is either akin to doing or derivative of doing.
1b. This question, in turn, raises the issue of what a social structure actually is. I need to know what something is before I can think about whether or not it is the type of thing that can do anything – and if so what the sense of “doing” is.
Note: since I am a person who thinks that causation is causation is causation, I am going to have to say the same about “doing.” That is, concretely there are all sorts of qualitatively different kinds of doing – and in fact there is no such concrete thing as generic doing (except maybe for commodified labor-power in its value-form); nonetheless, “doing” will have to be understood in such a way that it can apply to all instances of it. There won’t be an easy “pluralism about doing” move to be made, just as (in my view) there isn’t one to be made when one is using the term “causation” rather than the term “doing.”
2. Formal Cause
2a. One way to at least think that you’ve answered both questions – what structures are and in what sense they can be causes – is to say that structures are formal causes, as per Aristotle. I have suggested in the past that they are this, though I don’t think that I was clear enough at the time about what they were the formal causes of. I am not sure now if structures are formal causes, though they might be, and I am even less sure that being a formal cause amounts to doing.
2b. What is a formal cause?
Aristotle tells us in the Physics that when you ask what caused x there are really 4 different questions that you might be asking, with four very different kinds of answer.
- “What is x made of?” – Answer gives you the material cause of x
- “What is it to be an x?” – Answer gives you the formal cause of x
- “Who or what made it be that x came into being; who or what was the source of the change?” – Answer gives you the efficient cause of
- “What is the purpose of things that are x’s? I.e., what is involved in being afully actualized x?” – Answer gives you the final cause of x
2c. The formal cause of x, then, is the “what it is,” as Aristotle says, in virtue of which it is x and not y; the formal cause of x is its essence, its x-ness. Instantiated x-ness is material, of course, which is to say that concrete x’s are not exhausted by their x-ness; they also are their materiality – though it is x-materiality all the way down.
- What are social structures?
3a. Not sure, though I don’t think that they aren’t what people such as Giddens and Searle say they are, and they might be what Doug Porpora says they are.
3b. It’s easy to think of a social structure in spatial terms, as an arrangement of persons, as in a human pyramid, or an arrangement of steel, as in a bridge. It might be that the spatial metaphor of isn’t quite right, though.
Synchronically, it may be better to think of a structure as a (relatively fixed, diachronically) distribution of various things (“things” as a count noun): (i)material things, (ii) psychological things and (iii) power (i.e. efficacy), chief amongst them. The impulse is to say that the structure is that which accounts for a given stable distribution, and I think that there is something right about that, but I’m not sure how to think of it, and it seems to me that a synchronic configuration of goods could be diachronically self-reproducing.
Such a configuration would, in the case of some structures at least (e.g. patriarchy) be an essential feature of the social whole of which the structure is a structure. Synchronically, it seems to me, a distribution being as it is could give rise to emergent relational properties, and to irreducibly sociological positions, including ones that we could call “internal relations,” in which each position requires the other. Diachronically, it would be reproduced via activity mediated by these same conditions.
I’m not at all sure about this. I think that a society is a relational entity made out of individuals (and other things, too, no doubt); I just don’t know that that’s what a structure is.
3c. Whatever exactly a structure is, though, if a social whole is organized in a patriarchal rather than non-patriarchal way, then the society will indeed be an x the formal cause of which is patriarchy. Patriarchy will be the “what it is,” or essence, of such a social whole. (Or at least one aspect of what it is, though that is a further complication.)
- We can see two things right away.
First, in thinking about whether or not structures can do anything, I wasn’t asking if structures are the “what it is” of societies of given kinds. I was asking if structures can bring about change. I was asking if they can be what Aristotle calls efficient causes. On the face of it, at least, it won’t do, therefore, to say: “They are efficient causes in the sense that they are formal causes.” If nothing else, it’s not an answer. It would still be an open question whether or not the “what it is” of a thing can do what efficient causes can do.
Second, it seems as though things that are efficient causes can do what they can do in virtue of having the properties that they do. If structures can do things, and structures are essences (formal causes), we will next have to ask if essences are propertied things. Or if they are properties. And if they are neither, but we still wanted to say that structures are not just essences but efficient causes in virtue of being essences, then we would need to know how essences qua efficient causes compare to regular efficient causes, i.e., ones that aren’t essences but have essences. More generally, it will be necessary to be very clear about the relationship between the essences of things and their powers.
- Aristotle does give us a model of essences being efficient causes, at least sort of, in the case of the growth of natural things. This might be the place to look, if one wanted to say that patriarchy is both (a) the “what it is” of certain kinds of social wholes and, precisely qua the “what it is,” (b) the efficient cause of the internal development of such wholes.
- At a minimum, though, the question about whether or not structures can be causes has to be understood to be the question of whether or not social structures can bring about change. So the answer can’t be simply that they can do so because they are formal causes, i.e., essences. Perhaps they are essences, but that fact just pushes the question back a frame.
- In keeping with point #3 above, I might think that it is social wholes that are powerful particulars, not their structures. I could be talked out of this, for sure. But since I think that it’s propertied things and not their properties (let alone their essences) that do the doing, when causation occurs, it follows that if it’s a patriarchal society and not patriarchy that is the propertied thing, then in saying that patriarchy is a cause what we’d actually be saying – or, what we should be saying – is that a society that is organized in a certain way brings about certain internal outcomes in virtue of being the kind of society that it is. This would make it be that it is society, not its essence, that is the efficient cause.
- We’d still want to know, though, what the sense is in which a society can do things. And also what to say about the relationship between how a society is configured (what its characteristic internal distribution pattern is) and what the powers are that are had by the whole. The powers had by the whole are properties, and the structure of the whole is certainly intimately connected to its identity (or essence). But I’m not sure that structures just are essences. If they aren’t, and it’s in virtue of its structure rather than its essence that a whole has the powers that it does, then seems as though structures might be complex properties (or collections of properties), but not be dispositional ones.
- I’m dissatisfied, though, after all of this. The point that I want to go back to is at 3b, where I said that a configuration or distribution pattern of a whole could “mediate” activity. It’s that mediation that is the thing that I wanted to know about in the first place, I think. What kind of doing is that? Can it pass for efficient causation? I think that what I want to say is yes: it is a second-order enabling or constraining of agency. Given what I said above, I will have to say that it is actually the so-configured whole that is doing the mediating, not the configuration itself; I will have to think about that. And I’m still going to want to know how to categorize the configuration, the pattern of distribution: is it a power (of the whole) too? That doesn’t seem right. Perhaps it just is the concrete whole of a given kind. Though that doesn’t seem right either. The other two options would seem to be either that the configuration, or distribution, or structure of a social whole is a categorical property (or set of properties), rather than a power; or that it is the essence of the whole. I don’t have a settled view about which of these four options is right.
- If nothing else, I think that there is a better answer to be had to the question of whether or not structures can be productive causes than saying “Structures are formal causes.”