Causation in the Social Domain
It seems to be a truism that there is causation in the social domain. For example, if a person takes on a socially defined role or is ascribed a "status function" (becomes a prime minister, a judge, or a managing director) then it appears evident that the person acquires new causal abilities in virtue of the powers associated with the role. It also seems obvious that social groups and structures causally influence individuals. But how is social causation possible and how is it to be conceptualized? Social causation does not appear to comply with prevalent philosophical theorizing on causation. For example, causal abilities are often construed as intrinsic properties or powers in current philosophy. But properties such as being prime minister, judge and managing director are socially grounded, and hence are extrinsic. Moreover, modern philosophers are often skeptical about the notion that there is downward causation. They argue that the fundamental physical level is causally closed and that all higher levels "supervene" on, or are grounded in, the physical level - a fact that leaves no room for any causation going from macro to micro. However, group influence is prima facie an instance of downward causation. The crucial question, then, is whether the modern philosophical theories and arguments show that putative social causation is in fact unreal, or whether the evidence for social causation is so solid that it is the modern theories of causation that need to be revised.
Final report
PURPOSE AND IMPLEMENTATION
The purpose of the project was to investigate in what sense, if any, there is causation in the social domain. Specifically, two overarching research questions were posed:
(A) Are so-called status functions – and related phenomena such as social roles and social positions – to be understood as being partly causal? If yes, in what sense?
(B) Is there downward causation from groups to individuals? If yes, in what sense?
These two questions formed the basis of the project’s two sub-projects, (A) and (B), which were conducted in parallel. The main orientations of these sub-projects did not change in any substantive way during the project period (even though some new issues arose, see below); thus, the subprojects were implemented in accordance with the project plan. The research questions were addressed in a series of seven single-authored papers, two of which are still in preparation ([6] and [7]). As of now, four of the papers ([1], [2], [3], and [5]) have been published in international peer-reviewed journal, and one paper ([4]) has been published in an anthology.
THE THREE MOST IMPORTANT RESULTS OF THE PROJECT
It is difficult to maintain that results, in the sense of finality, are established in philosophy (formal logic possibly exempted). But if “result” is understood more weakly as substantiated thesis or outcome, then I would highlight the following results of the project:
1) Social groups, as they are characterized in current social sciences, are not reducible to individualistic entities. That is to say, social groups cannot be identified with pluralities, sets or mereological sums of individuals (se article [1] for detailed discussion). There are several reasons for this: one is that social groups and individualistic entities have distinct persistence conditions; another is that social groups are non-extensional entities, meaning that two distinct groups can have, or consist, of the same members. This non-reducibility result opens up for the possibility of there being non-reducible causation from groups to individuals. However, the result does not show that there actually is such non-reducible causation; nor does it show the manner in which such causation (if it exists) is realised.
2) Putative causation in the social domain cannot be understood in Aristotelian terms, i.e., in terms of productive “powers” (for detailed discussion, see article [3]). However, in the Critical Realism movement, which is a current influential philosophical school in the social sciences (the sociologist Dave Elder-Vass and the economist Tony Lawson are key figures in this school), social causation is conceptualised in precisely such terms. This is a mistake for the following reasons. First, powers are intrinsic properties, while status functions, social roles and social positions are extrinsically grounded features. Thus, status functions and similar phenomena cannot be categorised as powers; their (putative) causal abilities must be understood in other terms. Secondly, alleged intrinsic powers at the level of groups (which consequently cannot be status functions), which are thought to exert causal influence downward over individuals, are in all likelihood redundant, given the causal, productive processes that go on at the individualistic and physical levels (cf. Jeagewon Kim’s so-called “exclusion argument” in the philosophy of mind). Arguably, then, downward causation, from the level of groups to the level of individuals, must be conceptualised in non-Aristotelian terms, if the notion that there is such causation is to be defended.
3) Ascriptions of status functions via Searlean Declarations should be understood in terms of so-called “Mere Cambridge Change”, a notion picked up from contemporary metaphysics (for detailed discussion, see [5]). In other words, the fact that a thing, person or group acquires a status function does not, as such, involve any creation of a new “entity” (such as an institutional feature or fact), as the philosopher John Searle maintains. Rather, what happens is simply that a predicate begins to be true of the thing/person/group in question, because of the collectively accepted declaration (a version of Mere Cambridge Change), and that persons in the relevant society begin to act differently vis-à-vis the thing/person/group in question because of the Mere Cambridge Change (and, in some cases, that the person himself, or the members of the relevant group, start to act differently due to the Mere Cambridge Change). Status functions are thus not themselves causal. The causation in question is grounded in our representations of status functions. However, the fact that status functions are not themselves causal does not entail that status function cannot occur in true causal statements; they can, somewhat analogously to how an “absence” can figure as a “cause” or an “effect” (so-called negative causation) in a true causal statement without thereby being a cause or an effect in the world (an absence is obviously not a form of entity in the world). In other words, it is imperative to distinguish between what I call “sparse causation” and “abundant causation”: the former causation involves genuine relata, and a causal relation, that are located in the world; the latter “causation” involves true causal statements (that arguably should be analysed in terms of sets of counterfactual statements, in line with David Lewis’s and James Woodward’s theories of causation) where at least one of the putative relata is not a genuine entity in the world. (See [5] for an initial description of the sparse/abundant distinction. The distinction is developed further in [6] and is applied to putative downward causation from groups to individuals in [7].)
NEW RESEARCH QUESTIONS GENERATED BY THE PROJECT
One research question generated by the project, which was not anticipated in the project description, is whether putative social causation can involve simultaneous causation. Causes are generally thought to occur before their effects. But some philosophers of late (e.g. Stephen Mumford and Rani Lill Anjum) have defended the notion that causes and effects can occur simultaneously. Also, some of social ontologist Johns Searle’s work seems to be implicitly premised on there being simultaneous causation: if a group or a plurality of individuals ascribe a status function to some entity X, X will acquire this putative property causally and simultaneously with the ascription.
In the project (see article [2]), I show that Mumford and Anjum’s reasoning is, in several respects, deeply problematic: their theorising is, among other things, incompatible with the special theory of relativity (STR), according to which no causal process can propagate faster than the speed of light (which is finite). In article ([5]), I go on to show in detail that, given STR, Searle’s simultaneous causation assumption entails that entities that are ascribed status functions acquire these backward in time relative to reference systems that are moving at very high speed relative to the reference system in which the status function ascription occurs. Hence, Searle’s theory does not only involve simultaneous causation, it entails backward causation in some reference systems (a patently absurd consequence).
An additional issue raised by the project concerns the question whether Aristotelian causation must be understood in terms of interactions between active and passive powers – as has been assumed by this school of thought since the days of Aristotle. This issue is not essential to the project, since I reject the idea that social causation should be understood in Aristotelian terms (although I am sympathetic towards the view that causation at lower levels is Aristotelian). The question is rather an interesting spin-off issue generated by, but investigated outside of, the project. These investigations have resulted in article ([8]), in which I argue that Aristotelian causation is to be understood in terms of active powers only; postulating passive powers is redundant.
THE INTERNATIONAL DIMENSIONS OF THE PROJECT, AND COOPERATION
May 2017, I was a guest researcher at Institute de Philosophie in Neuchâtel, Switzerland; April 2018, I was a guest researcher at the Centre for Philosophy of Science in Pittsburgh, USA. The purpose of the first visit was to establish contact with Professor Fabrice Correia (now professor in Geneva) and the research group “eidos” which among other things investigates the concept of grounding and the notion that reality is stratified into levels. The main purpose of the second visit was to establish contact with Professor James Woodward who is a very influential philosopher of causation. During both visits I attended various seminars and conferences. In Lund, I co-organised the international conference ENSO V, The Fifth Conference of the European Network on Social Ontology (Aug 2017), with over sixty speakers, where I also presented material from the project. Papers from the project were also presented at international conferences in Tampere, Uppsala and Umeå, and at seminars with international participants in Linköping and Lund. Between Sep 2018 and Aug 2019 our department hosted a guest visitor from Turkey, Abrek Canbolat, who did research on social ontology with me as his contact person and advisor. During the project I have also been a member of the research group Metaphysics and Collectivity which consists of theoretical and practical philosophers from Swedish universities doing research on social ontology. (The group, e.g., co-organised ENSO V, with the Lund members being the main organisers.) The papers of the project have been written in English and have been published in international journals, except for [4] which is published in an anthology with international contributors.
The purpose of the project was to investigate in what sense, if any, there is causation in the social domain. Specifically, two overarching research questions were posed:
(A) Are so-called status functions – and related phenomena such as social roles and social positions – to be understood as being partly causal? If yes, in what sense?
(B) Is there downward causation from groups to individuals? If yes, in what sense?
These two questions formed the basis of the project’s two sub-projects, (A) and (B), which were conducted in parallel. The main orientations of these sub-projects did not change in any substantive way during the project period (even though some new issues arose, see below); thus, the subprojects were implemented in accordance with the project plan. The research questions were addressed in a series of seven single-authored papers, two of which are still in preparation ([6] and [7]). As of now, four of the papers ([1], [2], [3], and [5]) have been published in international peer-reviewed journal, and one paper ([4]) has been published in an anthology.
THE THREE MOST IMPORTANT RESULTS OF THE PROJECT
It is difficult to maintain that results, in the sense of finality, are established in philosophy (formal logic possibly exempted). But if “result” is understood more weakly as substantiated thesis or outcome, then I would highlight the following results of the project:
1) Social groups, as they are characterized in current social sciences, are not reducible to individualistic entities. That is to say, social groups cannot be identified with pluralities, sets or mereological sums of individuals (se article [1] for detailed discussion). There are several reasons for this: one is that social groups and individualistic entities have distinct persistence conditions; another is that social groups are non-extensional entities, meaning that two distinct groups can have, or consist, of the same members. This non-reducibility result opens up for the possibility of there being non-reducible causation from groups to individuals. However, the result does not show that there actually is such non-reducible causation; nor does it show the manner in which such causation (if it exists) is realised.
2) Putative causation in the social domain cannot be understood in Aristotelian terms, i.e., in terms of productive “powers” (for detailed discussion, see article [3]). However, in the Critical Realism movement, which is a current influential philosophical school in the social sciences (the sociologist Dave Elder-Vass and the economist Tony Lawson are key figures in this school), social causation is conceptualised in precisely such terms. This is a mistake for the following reasons. First, powers are intrinsic properties, while status functions, social roles and social positions are extrinsically grounded features. Thus, status functions and similar phenomena cannot be categorised as powers; their (putative) causal abilities must be understood in other terms. Secondly, alleged intrinsic powers at the level of groups (which consequently cannot be status functions), which are thought to exert causal influence downward over individuals, are in all likelihood redundant, given the causal, productive processes that go on at the individualistic and physical levels (cf. Jeagewon Kim’s so-called “exclusion argument” in the philosophy of mind). Arguably, then, downward causation, from the level of groups to the level of individuals, must be conceptualised in non-Aristotelian terms, if the notion that there is such causation is to be defended.
3) Ascriptions of status functions via Searlean Declarations should be understood in terms of so-called “Mere Cambridge Change”, a notion picked up from contemporary metaphysics (for detailed discussion, see [5]). In other words, the fact that a thing, person or group acquires a status function does not, as such, involve any creation of a new “entity” (such as an institutional feature or fact), as the philosopher John Searle maintains. Rather, what happens is simply that a predicate begins to be true of the thing/person/group in question, because of the collectively accepted declaration (a version of Mere Cambridge Change), and that persons in the relevant society begin to act differently vis-à-vis the thing/person/group in question because of the Mere Cambridge Change (and, in some cases, that the person himself, or the members of the relevant group, start to act differently due to the Mere Cambridge Change). Status functions are thus not themselves causal. The causation in question is grounded in our representations of status functions. However, the fact that status functions are not themselves causal does not entail that status function cannot occur in true causal statements; they can, somewhat analogously to how an “absence” can figure as a “cause” or an “effect” (so-called negative causation) in a true causal statement without thereby being a cause or an effect in the world (an absence is obviously not a form of entity in the world). In other words, it is imperative to distinguish between what I call “sparse causation” and “abundant causation”: the former causation involves genuine relata, and a causal relation, that are located in the world; the latter “causation” involves true causal statements (that arguably should be analysed in terms of sets of counterfactual statements, in line with David Lewis’s and James Woodward’s theories of causation) where at least one of the putative relata is not a genuine entity in the world. (See [5] for an initial description of the sparse/abundant distinction. The distinction is developed further in [6] and is applied to putative downward causation from groups to individuals in [7].)
NEW RESEARCH QUESTIONS GENERATED BY THE PROJECT
One research question generated by the project, which was not anticipated in the project description, is whether putative social causation can involve simultaneous causation. Causes are generally thought to occur before their effects. But some philosophers of late (e.g. Stephen Mumford and Rani Lill Anjum) have defended the notion that causes and effects can occur simultaneously. Also, some of social ontologist Johns Searle’s work seems to be implicitly premised on there being simultaneous causation: if a group or a plurality of individuals ascribe a status function to some entity X, X will acquire this putative property causally and simultaneously with the ascription.
In the project (see article [2]), I show that Mumford and Anjum’s reasoning is, in several respects, deeply problematic: their theorising is, among other things, incompatible with the special theory of relativity (STR), according to which no causal process can propagate faster than the speed of light (which is finite). In article ([5]), I go on to show in detail that, given STR, Searle’s simultaneous causation assumption entails that entities that are ascribed status functions acquire these backward in time relative to reference systems that are moving at very high speed relative to the reference system in which the status function ascription occurs. Hence, Searle’s theory does not only involve simultaneous causation, it entails backward causation in some reference systems (a patently absurd consequence).
An additional issue raised by the project concerns the question whether Aristotelian causation must be understood in terms of interactions between active and passive powers – as has been assumed by this school of thought since the days of Aristotle. This issue is not essential to the project, since I reject the idea that social causation should be understood in Aristotelian terms (although I am sympathetic towards the view that causation at lower levels is Aristotelian). The question is rather an interesting spin-off issue generated by, but investigated outside of, the project. These investigations have resulted in article ([8]), in which I argue that Aristotelian causation is to be understood in terms of active powers only; postulating passive powers is redundant.
THE INTERNATIONAL DIMENSIONS OF THE PROJECT, AND COOPERATION
May 2017, I was a guest researcher at Institute de Philosophie in Neuchâtel, Switzerland; April 2018, I was a guest researcher at the Centre for Philosophy of Science in Pittsburgh, USA. The purpose of the first visit was to establish contact with Professor Fabrice Correia (now professor in Geneva) and the research group “eidos” which among other things investigates the concept of grounding and the notion that reality is stratified into levels. The main purpose of the second visit was to establish contact with Professor James Woodward who is a very influential philosopher of causation. During both visits I attended various seminars and conferences. In Lund, I co-organised the international conference ENSO V, The Fifth Conference of the European Network on Social Ontology (Aug 2017), with over sixty speakers, where I also presented material from the project. Papers from the project were also presented at international conferences in Tampere, Uppsala and Umeå, and at seminars with international participants in Linköping and Lund. Between Sep 2018 and Aug 2019 our department hosted a guest visitor from Turkey, Abrek Canbolat, who did research on social ontology with me as his contact person and advisor. During the project I have also been a member of the research group Metaphysics and Collectivity which consists of theoretical and practical philosophers from Swedish universities doing research on social ontology. (The group, e.g., co-organised ENSO V, with the Lund members being the main organisers.) The papers of the project have been written in English and have been published in international journals, except for [4] which is published in an anthology with international contributors.