The Roots of Occasionalism? Causation, Metaphysical Dependence, and Soul-Body Relations in Augustine

It has long been thought that Augustine holds that corporeal objects cannot act upon incorporeal souls. However, precisely how and why Augustine imposes limitations upon the causal powers of corporeal objects remains obscure. In this paper, the author clarifies Augustine’s views about the causal and dependence relations between body and soul. He argues that, contrary to what is often thought, Augustine allows that corporeal objects do act upon souls and merely rules out that corporeal objects exercise a particular kind of causal power (that of efficient or sustaining causes). He clarifies how Augustine conceives of the kind of causal influence exercised by souls and bodies.

He [Augustine] did not believe that the body acts upon the soul by making it receive its passions but by offering it an occasion to act at the same time.
It has long been thought that, for Augustine, "lower" entities, such as bodies, may not act upon or causally influence "higher" entities, such as souls. Such readings have long established precedent and have exerted significant historical influence. Thus, for instance, some medieval philosophers, such as John Peckham, took themselves to be following Augustine in claiming that corporeal things do not act upon the soul in perception and are instead merely an occasion for the soul to act upon itself.1 Equally, some early modern occasionalists, such as Louis de la Forge, took Augustine to agree with their own views and found in him an authoritative precedent for thinking that apparent corporeal causes might be demoted to mere occasions without genuine causal efficacy.2 Such readings are also prevalent in more recent studies of Augustine. Thus, for instance, Robert Markus claims: It is axiomatic, for Augustine, that body cannot act on spirit, and that no modification can therefore arise in the mind caused by the bodily senseorgans and their changing states. The reason for this is that body is below mind in the hierarchy of nature, and lower cannot act upon higher on the premises of Augustine's metaphysical scheme.3 On this view, Augustine takes it to be "axiomatic" that "lower" entities, such as bodies, cannot act or exercise any causal influence upon "higher" entities, such as souls. Such a view is frequently asserted in studies of Augustine's account of perception and has also been articulated elsewhere.4 Thus, for instance, 1 John Peckham, Tractatus de anima, 3, 10: "Acquiritur autem animae scientia occasione per sensus prestita." Durand of St.-Pourçain, Robert Kilwardby, Peter John Olivi, and others took themselves to be following Augustine in claiming that bodies do not act upon souls at all, even as remote causes or secondary causes. Cf. Solère, "Sine qua non Causality." 2 E.g. Louis de la Forge, Traitté de l'Esprit de l'Homme, pref. Occasionalism is often characterised as the view that (i) many apparent causes are mere occasions and are not genuine causes, and (ii) God is the only genuine cause (cf. Malebranche,Oeuvres complètes 2: 312). In what follows, I am primarily concerned with (i) insofar as it applies to corporeal items (a species of what is sometimes called "partial occasionalism", e.g. Nadler, Occasionalism). 3 Markus, "Augustine," 376. 4 E.g. Bourke, The Essential Augustine, 68: "Soul can move body, but no body can produce any modification in any soul"; Miles,Augustine,16: "This absolute transcendence of soul over body, axiomatic for Augustine, requires that he describe sensation without allowing any action to be imposed on the mind by the bodily senses"; Miles,Augustine,19: "By definition, in discussing Augustine's views concerning perception and the efficacy of the Christian sacraments, Philip Cary claims that, for Augustine, corporeal things are "devoid of causal power over higher things like the soul,"5 that "the world of bodies has literally no causal effect on the soul,"6 and that: [Augustine] makes a point of giving us a theory of sense perception in which no bodily thing, including words, has a causal effect on the soul … in accord with the Platonist axiom of downward causality, the causal interaction between body and soul runs in only one direction, from inner to outer.7 In this paper, I argue that such views are mistaken and I clarify Augustine's views about the causal powers of corporeal objects and the causal and dependence relations that exist between body and soul.8 To this end, I first examine Augustine's views about the limitations of the causal powers of corporeal objects (Section 2). I show that Augustine maintains that bodies do not bring about anything ( facere aliquid) in the soul, but that there is also evidence that Augustine thinks corporeal things act upon the soul. I then clarify Augustine's views concerning the causal and dependence relations that obtain between bodies and souls (Section 3). Many readers equate Augustine's talk of "bringing about" with acting upon or exercising any causal influence whatsoever and, as a result, they attribute to Augustine the view that corporeal objects do not act upon or exercise any causal influence upon souls. This is a mistake. I argue that "bringing about" denotes a particular kind of causal influence (that exercised by efficient causes, see below) and that, although Augustine does discern an important hierarchy of metaphysical dependence, he does not take it to be true -much less to be axiomatic -that body can have no effect on soul because it is metaphysically inferior"; Caston, "Connecting Traditions," 40: "Augustine firmly denies that any body can act on the soul and so that the visible object can affect the soul in any way"; Silva, "Augustine," 85: "It is impossible for something material to act upon something immaterial, or in other words for something lower on the scale of being to act upon something higher on that same scale." 5 Cary,Outward Signs,105. 6 Cary,Outward Signs,7. 7 Cary,Outward Signs,84. 8 I am not alone in thinking that Augustine allows corporeal things to act upon souls. Thus, for instance, O'Daly thinks there is "reciprocal influence of body and soul" (Augustine's Philosophy of Mind,45;cf. 84,(122)(123) lower entities (such as corporeal things) cannot act upon or exercise any causal power upon higher entities (such as incorporeal souls). Instead, he merely rules out that corporeal items may act per se or function as so-called "efficient causes" while nonetheless recognising that bodily entities may act as causes of certain kinds. Finally, I clarify Augustine's views concerning efficient causes and the (nonefficient) causal powers of corporeal things (Section 4). Augustine's view of efficient causes seems to be strongly informed by a widely prevalent Hellenistic and post-Hellenistic set of causal conceptions (shared by the Stoics, Neoplatonists, various medical thinkers, several Christian thinkers, and others) according to which efficient causes are not merely active, but also productive (or creative) and sustaining. I clarify the relevant causal conceptions and suggest that they also provide a useful framework for understanding Augustine's views of non-efficient causes and what he means when he calls bodily items "proximate" or "antecedent" causes. On my reading, Augustine allows that while corporeal items fall short of being efficient causes (i.e. they are not jointly active, productive, and sustaining), they may nonetheless act upon other things -including the soul -in a fairly "robust" manner which goes beyond constituting mere background conditions or sine quibus non causes.

The Limits of the Causal Influence of Bodies
Some of Augustine's more detailed discussions of the causal relations between body and soul occur in De musica and De Genesi ad litteram. In De musica (begun ca. 387, but emended ca. 408),9 a didactic dialogue between a teacher and a student about how music affects listeners, the ears are taken to undergo a certain kind of affection (affectio, passio) when they are touched (tangere) by sounds (Mus. 6.2.3) and this raises a difficulty. When hearers perceive (sentire) a rhythm (numerus), they undergo (pati) something (Mus. 6.4.7). It thereby seems that corporeal things bring about something ( facere aliquid) in the soul. However, it also seems that, if x brings about y or something in or from y, then x is superior to y and prior to y (Mus. 6.4.7). Accordingly, admitting that corpo- real sounds bring about something in the soul seems to have the unattractive consequence that corporeal things are superior and prior to incorporeal souls (Mus. 6.4.7). In response to this difficulty, the teacher claims that the soul should not be thought to be inferior to bodies and neither should it be thought that the body acts upon the soul as an artisan acts upon his raw materials (Mus. 6.5.8).10 Somewhat more starkly, the teacher goes on to claim: I do not think that this body is animated by the soul in any other way than through the intentio of the agent. And I do not believe that the soul is affected by the body; instead, it brings about something from it and in it as if the body were subject to the soul by divine command and sometimes this is done with ease, other times with difficulty … Accordingly, whenever corporeal things are introduced into our body or brought to it from outside, they bring about something not in the soul but in the body itself which either hinders or helps the soul's tasks.11 The human body is thus subject to the soul and is animated by the agent's intentio (a term which is difficult to translate but which here seems to denote a stretching or tensing).12 Corporeal things do not bring about ( facere) anything in the soul but, at most, something in the perceiver's body. When the perceiver's body undergoes some change, this facilitates or impedes the soul's activities (opus) and the soul brings about something in itself from the body. In what follows, the teacher emphasises that sense-perceptions are activities of the soul (operationes animae, e.g. Mus. 6.6.16) while frequently emphasising that the soul acts upon itself in perception (e.g. Mus. 6.5.10-12, 13.39).
In his later De Genesi ad litteram (written between ca. 400 and 415), Augustine puts forward similar claims. He states that, although corporeal things 10 Mus. 6.5.8: "The soul is thus in no way subject to the body as matter is subject to an artisan …" ("Nullo modo igitur anima fabricatori corpori est subiecta materies"). Translations are my own except where noted otherwise. 11 Mus. 6.5.9: "Ego enim ab anima animari hoc non puto corpus nisi intentione facientis. Nec ab isto quidquam illam pati arbitror sed facere de illo et in illo tamquam subiecto divinitus dominationi suae, aliquando tamen cum facilitate, aliquando cum difficultate, operari, quanto pro eius meritis magis minusve illi cedit natura corporea. Corporalia ergo, quaecumque huic corpori ingeruntur aut obiciuntur extrinsecus, non in anima sed in ipso corpore aliquid faciunt, quod operi eius aut adversetur aut congruat." 12 Cf It is not to be thought that the body brings about something in the spirit as though the spirit, like matter, were subjected to the body which made something out of it. That which brings about something is in every way superior to that thing from which something is brought about, but the body is in no way superior to the spirit. On the contrary, spirit is clearly superior to body. Accordingly, we first see a body, which we had not seen previously, and then its representation -by which we remember it when it is absent -begins to exist in our spirit. However, it is not the body that brings about its representation in the spirit. Instead, it is the spirit which brings it about in itself with a remarkable swiftness which is indescribably different from the sluggishness of the body.13 In describing how the soul -or a particular part of the soul -brings about an imago in itself upon seeing some external object,14 Augustine here offers a simple argument, which goes as follows (cf. Mus. 6.5.8): (1) if α brings about β or something from or in β, then α is superior to β; (2) the body is not superior to the soul; ∴ (3) the body does not bring about something from or in the soul.
For now it suffices to note that (1) was a commonly assumed claim concerning creative causal power,15 and that (2) is taken by Augustine (like many 13 Gn. litt. 12.16.33: "Nec sane putandum est facere aliquid corpus in spiritu, tamquam spiritus corpori facienti materiae vice subdatur. Omni enim modo praestantior est qui facit ea re de qua aliquid facit. Neque ullo modo spiritu praestantius est corpus, immo perspicuo modo spiritus corpore. Quamvis ergo prius videamus aliquod corpus, quod antea non videramus, atque inde incipiat imago eius esse in spiritu nostro, quo illud cum absens fuerit recordemur tamen eamdem eius imaginem non corpus in spiritu, sed ipse spiritus in seipso facit celeritate mirabili." 14 For the nature of these imagines, their role in Augustine's epistemology, and Augustine's account of representation, see Nawar, "Augustine on the Varieties of Understanding"; Nawar, "Augustine's Defence"; Nawar, "Augustine on Active Perception." 15 Cf. Plotinus, Enneads 5.5.13.37-38: "What makes is superior to what is made because it is more perfect (Κρεῖττον γὰρ τὸ ποιοῦν τοῦ ποιουμένου· τελειότερον γάρ First, and most simply, even though the soul may act upon itself and bring about things in itself in the relevant cases, it is hard to see how corporeal items could hinder or resist the soul's activities (e.g. Mus. 6.5.9-10; Gn. litt. 7.9.25; 12.20.42) or how the soul could "accept things from the senses" (Gn. litt. 7.14.20) or be "mixed in with the body" (cf. Gn. litt. 3.16.25;8.21.42) without bodies thereby acting upon or exercising some causal influence upon the soul.
Second, certain other texts indicate that, for Augustine, corporeal things do in fact act upon and causally influence the soul. Thus, for instance, in De civitate Dei Augustine claims that states like pain are caused by what the flesh undergoes and arise from the flesh (e.g. Civ. Dei 14.15.2; cf. 14.5) and in De Trinitate, a perceived corporeal object is identified as a quasi-parent of perception (Trin. 11.5.9;cf. Trin. 10.2.2,7.10;. Moreover, in De Trinitate Augustine sees dryness or moist humours as a genuine cause (vera causa) of psychosomatic illness (Trin. 3.3.8) (the kind of claim partial occasionalists would typically reject) and claims that -in some cases at least -body and soul are affected by each other in turn (sic ex semetipsis afficiantur vel anima ex corpore vel corpus ex anima, Trin. 3.8.15). More explicitly, in an early letter to Nebridius (written 723a3-10; Posterior Analytics i.31, 88a5; Physics viii.9, 265a22; Plotinus, Enneads 6.1.6.37-39; Proclus, Elements of Theology 7; Augustine,15 between 388 and 391), Augustine notes that, while "every motion of the soul brings about something in bodies" (omnem motum animi aliquid facere in corpore, Ep. 9.3), changes in the body may also influence the soul in turn.17 Thus, for instance, increases in bile may move the soul to anger and "thus what the soul has, through its own motion, brought about in the body is able to stir it up in return" (Ita quod suo motu animus fecit in corpore, ad eum rursus commovendum valebit, Ep. 9.4). Third, those texts which seem to offer the strongest evidence for thinking that bodies do not act upon souls in fact offer evidence that bodies do act upon souls. For instance, in Gn. litt. 12.16.33 (cited above, cf. Mus. 6.5.8), Augustine merely claims that one should not think that "the body brings about something in the spirit as though the spirit, like matter, were subjected to the body which made something out of it." That is to say, Augustine does not claim that bodies cannot act upon souls simpliciter (a common misreading),18 but merely that bodies cannot act upon souls in a certain way, that is, in the way in which an artisan shapes raw material. This leaves open the possibility that bodies can act upon souls in other ways. Moreover, within that very same work, Augustine speaks of souls being affected by the things they perceive (e.g. Omnis enim anima viva … visis movetur, Gn. litt. 9.14.25) and -in the context of discussing Galatians 5: 17 ("For the flesh lusts against the spirit and the spirit against the flesh") -notes that the cause of carnal concupiscence lies not in the soul alone (concupiscentiae carnalis causa non est in anima sola, Gn. litt. 10.12.20) but in the soul and body together. The body must, Augustine thinks, play some causal role since, without it, no carnal pleasure may be experienced by the soul (… ex carne autem, quod sine illa carnalis delectatio non sentitur, Gn. litt. 10.12.20).
Equally, Augustine notes that hallucinations and certain disturbances wherein the soul is more vehemently affected (vehementius movetur) "sometimes occur from the body and sometimes from the spirit" (aliquando a corpore accidit, aliquando a spiritu, Gn. litt. 12.19.41), and he says "the cause of such (hallucinatory) visions is discerned as being from the body" (a corpore causa est ut talia visa cernantur, Gn. litt. 12.20.42), while noting how such causes may be in the eye, the brain (e.g. causa est intus in cerebro, Gn. litt. 12.20.43), or elsewhere. 17 Peter John Olivi, In ii Sent., q. 72, 3-4, also thinks Augustine claims that bodies act upon souls at Contra Iulianum 5.14.51, but Augustine's text is unclear (at least to me). 18 E.g. Silva, "Augustine," 85: "The purpose of the above passage [Gn. litt. 12.16.33] is to deny the possibility of something corporeal acting upon something spiritual. This is the statement of a principle that is at the heart of Augustine's Philosophy, the metaphysical principle (mp) according to which it is impossible for something material to act upon something immaterial, or in other words for something lower on the scale of being to act upon something higher on that same scale." Vivarium (2021)  Thus, while processes such as perception manifest the soul's agency (it is not the body which perceives; instead the soul uses the body in order to perceive, Gn. litt. 10.12.21; cf. Plato, Theaetetus 184c-e),20 corporeal items do seem to play some causal role (and seemingly even an active causal role) in the relevant psychological processes. Such remarks suggest that traditional readings are not correct but they also raise a puzzle. On the one hand, Augustine seems to maintain (no-bodilybringing-about) in earlier and later works. On the other hand, in earlier and later works -and sometimes within the very works wherein he affirms (no-bodily-bringing-about) -Augustine seems to speak as if bodies do exert causal power and act upon souls. No simple developmental hypothesis can account for this and while it might thus seem that Augustine's views are simply inconsistent (a conclusion favoured by some readers),21 in what follows I will argue that Augustine's views are in fact largely consistent and that we may better understand them by distinguishing between different kinds of causes and causal activity and by clarifying precisely what (no-bodily-bringingabout) amounts to.

Metaphysical Hierarchies, Dependence, and Causal Relations
As noted above, many readers of Augustine take him to maintain that it is "axiomatic" that higher entities cannot be affected or acted upon by lower entities and that corporeal things may not act upon or exercise any causal influence whatsoever upon the soul.22 Call these kinds of views "Hierarchical Axiom readings". Such readings of Augustine -according to which he accepts what is 19 Despite his earlier remarks (Mus. 6.5.9-13, discussed above), Augustine later recognises that bodies do act upon souls at least insofar as souls do act (operari) upon souls through bodies (per corpus) (Mus. 6.13.41). Equally, even in Augustine's early De immortalitate animae (where he aims to safeguard the soul from certain kinds of changes so as to demonstrate its immortality), Augustine notes that the soul is said to be changed according to the affections of the body or its own affections (secundum corporis passiones, aut secundum suas, anima dicitur immutari, Imm. an. 5.7; cf. Imm. an. 14.23). 20 For detailed discussion of perception as an active process and precisely what the soul does, see Nawar, "Augustine on Active Perception." 21 E.g. Miles, Augustine, 9; Silva and Toivanen, "The Active Nature," 248; Toivanen, Perception, 136; Silva, "Augustine," 79. Cf. Kalderon sometimes called "downwards only causation" -have venerable precedent,23 but as things stand, there are at least two significant problems with existing discussions of these issues. First, to claim that Augustine holds that corporeal items cannot act upon incorporeal items because of certain "hierarchical causal axioms" (according to which lower entities cannot act upon or exert any causal influence upon higher entities) at best merely describes rather than explains Augustine's views. Second, although it is often claimed that Augustine subscribes to the relevant "hierarchical causal axioms," little in the way of textual evidence has been provided for attributing such views to Augustine.24 Some Hierarchical Axiom readings (e.g. by Markus, Miles, and Cary) take themselves to be supported by Augustine's remarks about a metaphysically important hierarchy with at least three levels (e.g. De quantitate animae 36.80; Ep. 18; Gn. litt. 8.20.39). However, levels of this hierarchy are characterised not by their ability or inability to act upon entities in other levels but by their ability or inability to be affected with respect to time and place (cf. Mus. 6.14.44). Thus, at the lowest level, the body is characterised as an entity which is mutable (mutabilis) with respect to time and place. At the intermediary level, the soul is an incorporeal entity which is mutable with respect to time (but not place). At the highest level, God is immutable (immutabilis, Ep. 18.2; incommutabilis, De natura boni 1, 19; Gn. litt. 8.20.39) in all respects.
In these contexts, Augustine does not claim that "lower" entities cannot exercise any causal influence upon "higher" entities (as is sometimes suggested).25 Instead, what Augustine says is that items at higher levels of the hierarchy, such as souls, are prior to items at the lower levels, such as bodies: Just as the substance which is altered only with respect to time is prior to the substance which is altered with respect to time and place, so too the substance which is not altered with respect to place or time is prior to the substance that is altered with respect to time.26 Thus the soul is prior to the body, and God is prior to the soul and also to the body. We may suppose that -in line with other ancient accounts of prioritypriority is asymmetric (such that if A is prior to B, then B is not prior to A) and transitive (such that if A is prior to B and B is prior to C, then A is prior to C) and that, roughly speaking, A is prior to B iff B could not exist or be the way it is without A existing or being the way it is.27 Some of Augustine's remarks suggest that the relation between prior and posterior also implies a certain kind of causal dependence. Thus, in discussing God's creation of the world, Augustine remarks that "every cause (efficiens) is greater than that which it brings about" (Omne autem efficiens maius est quam id quod efficitur, De diversis quaestionibus octoginta tribus 28), i.e. that if A is an efficient cause of B, then A is greater than B (cf. Gn. litt. 12.16.33), and he elsewhere suggests that, in such cases, A is prior to B (Civ. Dei 7.9.1). As Augustine explains elsewhere, things possess whatever substantial form (species, forma) they have because of "higher" things. Thus, for instance, the body "subsists through the soul" (per animam ergo corpus subsistit, Imm. an. 15.24; cf. Imm. an. 7.12). That is to say, the soul holds the whole (body) together through a present power (praesente potentia tenet universum, Imm. an. 11.14). Its continued presence is required to provide form, life, and unity to the human body and to endow it with its real qualities.28 Just as bodies require souls to sustain them (Imm. an. 7.12-8.13), so too souls, in turn, require God to sustain them (De natura boni 10; Mus. 6.13.40; Civ. Dei 8.6; 12.26). There is thus a chain of metaphysical dependence and a certain kind of power (potestas) which stems ultimately from God (De natura boni 11; Imm. an. 15.24). As Augustine puts it in De civitate Dei: In De immortalitate animae, Augustine offers an example of how a higher item, such as the intentio of an artisan (i.e. something at the level of soul), may change the corporeal things the artisan is working upon while itself remaining unchanged (Imm. an. 3.4). One might think this gives support to Hierarchical Axiom Readings, but Augustine (Imm. an. 3.4) merely says that superior things may act upon inferior things without changing: "Thus, it should not be thought that, if that by which a bodily change occurs is the soul changing [it] then -even if it might be intent toward it -the soul is thereby necessarily changed and on account of this might die" ("Non igitur si qua mutatio corporum movente animo fit quamvis in eam sit intentus hinc eum necessario mutari, et ob hoc etiam mori arbitrandum est"). Equally, when Augustine briefly argues against the possibility that bodies bring about bodies (Imm. an. 11.14), he focuses on the impossibility of self-causation for the entirety of body (i.e. the body of everything, as it were, bringing about the body of everything).
First, as we have seen, corporeal things receive their form from higher things and thereby depend upon higher things for their nature. Second, corporeal things are not efficient causes because they do not in themselves bring about things (either in higher items or other bodies). They lack internal or intrinsic motion and are thus not self-movers (Gn. litt. 7.16.22). Third, there is a hierarchy of metaphysical dependence such that lower items depend upon higher items and higher items do not depend upon lower items. Higher items bring about but are not brought about by lower items and higher items are thus prior to lower items.
However, this leaves open the possibility that bodies do exercise causal influences of certain kinds (whether upon other bodies or upon souls) and even that bodies may act upon souls. In fact, Augustine explicitly recognises that, although corporeal items are not efficient causes, they can rightly be regarded as causes of some significant kind. Thus, for instance, in De Genesi ad litteram, Augustine recognises corporeal items as inferior causes (Gn. litt. 6.17.28) (i.e. inferior when compared to God's will)33 and in De Trinitate 3, Augustine suggests that, although God's will is the cause of all things and is superior to all other causes (superiorem ceteris omnibus causam, id est voluntatem Dei, Trin. 3.2.7, cf. 3.4.9; Civ. Dei 5.9), other kinds of items -such as corporeal entitiesmay be regarded as proximate causes (proximae causae, Trin. 3.2.7; cf. praecedentes causae, Gn. litt. 6.9.14). To elucidate what he has in mind in the case of "proximate causes", Augustine goes on to suggest that, although a thing at the level of soul, e.g. a psychological item such as a person's decision to work too hard, may be regarded as a higher cause (superior causa) of a person's illness than the unbalanced humours or excessive dryness of their body, the latter corporeal items (e.g. humours, dryness) may nonetheless rightly be regarded as certain kinds of causes -e.g. proximate causes -and Augustine identifies these as corporeal causes (de proximis causis, id est corporalibus, Trin. 3.3.8).34 Such remarks, especially when read alongside the numerous instances where Augustine speaks of corporeal items as exerting causal influence upon the soul (e.g. Civ. Dei 14.5, 15; cf. Trin. 10.2.2, 7.10; 11.5.9; Ep. 7.2.3-4) or of psychosomatic interaction (e.g. Ep. 9.3-4; Trin. 3.8.15; see Section 2 above), indicate 33 Augustine's example concerns Hezekiah, who, according to the natural order of things and inferior causes, should have died at a certain time but to whom God gave an additional fifteen years of life (2Kings 20: 4-6). 34 God's will is a higher cause than the person's will to work too hard (Trin. that Augustine does not maintain that lower items (such as bodies) do not act upon or exercise any causal influence upon higher items (such as souls). Instead, Augustine merely claims that corporeal things do not exercise any causal influence of a particular kind -i.e. the kind of influence exercised by efficient causes -upon anything whatsoever while leaving open the possibility, and sometimes explicitly recognising, that bodies may act upon and causally influence the soul in certain (seemingly non-efficient) ways. If one is willing to attribute to Augustine even a minimal form of consistency (e.g. within a single work), then it seems that although (no-bodilybringing-about) rules out that bodies bring about things ( facere aliquid) in souls, it does not rule out that bodies act upon or causally influence souls. Accordingly, bringing about should not be taken to denote acting upon, exercise causal influence upon, or any similarly broad category of causal activity. Instead, it denotes a particular kind of causal activity, i.e. that exercised by what Augustine calls "efficient causes". In order to see what (no-bodilybringing-about) amounts to, then, one must clarify how Augustine conceives of efficient causes and of the difference between efficient and nonefficient causes.

Efficient Causation, Non-efficient Causation, and the Causal Powers of Bodies
We have seen that Augustine uses causa efficiens and cognate terms to designate the kind of causes which souls may be and which bodies are not (e.g. Civ. Dei 5.9; De diversis quaestionibus octoginta tribus 28, see above), and terms such as proxima causa (e.g. Trin. 3.3.8) and praecedens causa (e.g. Gn. litt. 6.8.13-9.14) to designate the kind of causes which bodies may be. What remains, then, is to clarify the different kinds of causal power and the kind of non-efficient causal power exercised by bodies. According to these causal conceptions (which owe much to the Stoics and find parallel in Seneca, e.g. Epistulae 87.31-40, and elsewhere),37 so-called "efficient causes" "contain within themselves a natural effectiveness" (cohibentes in se efficientiam naturalem, Cicero, De fato 9.19; cf. De fato 14.32; Topica 58; Seneca, 36 Cicero, Topica 58 (trans. Reinhardt, adapted): "Causarum enim genera duo sunt; unum quod vi sua id quod sub eam vim subiectum est certe efficit, ut: Ignis accendit; alterum, quod naturam efficiendi non habet sed sine quo effici non possit, ut si quis aes statuae causam velit dicere, quod sine eo non possit effici." 37 Thus, for instance, in discussing the extent to which riches should be considered a cause of evil, Seneca remarks: "Posidonius, who said that riches are causes of evil not because they themselves bring about something but because they provoke those who do bring about things (non quia ipsae faciunt aliquid, sed quia facturos inritant), is, I think, superior. The efficient cause, which necessarily immediately produces harm, is one thing; the antecedent cause is another (Alia est enim causa efficiens, quae protinus necesse est noceat, alia praecedens  4, 87.31) and, in addition to being active (ἐνεργοῦν, Sextus Empiricus, Outlines of Pyrrhonism 3.14), efficient causes were deemed to be distinctively sustaining and productive (ποιητική) (e.g. Clement, Stromata 8.9.25.5.1-5, 27.6; cf. Sextus Empiricus, Outlines of Pyrrhonism 3.14-15). That is to say, efficient causes themselves do things and are responsible not only for bringing their effects into being but also for sustaining their effects in being (and on this account such causes were also called "sustaining" or "cohesive" or "containing" causes, αἴτια συνεκτικά).38 This notion of efficient causation, which owes much to the Stoics but which was widespread in Hellenistic and post-Hellenistic antiquity,39 differs from Aristotelian accounts of efficient causes and most modern accounts of causation (since at least Mackie's "Causes and Conditions" and Lewis's "Causation") in several respects. Notably, according to this conception, if something is an efficient cause, then its continued presence and activity is required to sustain what it has brought about in existence (Sextus Empiricus, Outlines of Pyrrhonism ph 3.14-15; Clement Stromata 8.9.25.1 ff.; 8.9.33.1-9). Thus, a soul is the efficient cause of life and a fire is the efficient cause of heat.40 However, in contrast with Aristotelian causal accounts, a father would not be regarded as the efficient cause of a child's existence; instead he would be regarded as, e.g., an antecedent cause.
The causal taxonomy described above seems to portray non-efficient causes as being fairly variegated in nature. In Cicero's Topica, some non-efficient causes are described as passive or inert. Such inert causes seemingly include For discussion of the Stoic account of the soul and the causal relations between (corporeal) souls and bodies, see Nawar, "The Stoic Theory." For discussion of the Stoic account of the soul and identity and its influence on some early Christian thinkers, such as Origen of Alexandria and Basil of Caesarea, see Nawar, "The Stoics." items which would intuitively be classed as background conditions for certain causal processes to take place, i.e. some items which are counterfactually necessary for the cause to be able to do its work as a cause but which don't seem to actively do anything. Thus, for instance the time during which a process occurs or the place in which a process occurs might be regarded as inert non-efficient causes of this kind (Cicero, Topica 59; Clement Stromata 8.9.25.4.4). Equally, depending upon the particular details of one's physics, one might suppose that (e.g.) air is a background condition or sine qua non cause of fire burning or that the presence of a medium, such as air, is a background condition or sine qua non cause of the propagation of sound.41 However, it is crucial to appreciate that, according to the ancient causal taxonomy being discussed here, not all non-efficient causes are inert.42 In fact, several kinds of non-efficient causes are active while nonetheless differing from efficient causes in that they are not in themselves active (e.g. they do not "contain within themselves a natural effectiveness"), or are not themselves creative or productive, or are not in themselves necessary and sufficient for the continued existence of their effects (Cicero, Topica 59). Thus, for instance, an appearance or the thing seen is a "proximate but not principal [i.e. efficient] cause" (proximam causam … non principalem, Cicero, De fato 18.42) of some agent's mental state or action (cf. Seneca, Epistulae 87.31). Equally, as per the relevant ancient discussions, one of the two oxen drawing a cart is a cooperative (συναίτιον, cf. adiuvans) cause of the cart being drawn while intense exposure to the sun or drinking too much are the antecedent (προκαταρκτικά, cf. antecedentes, praecedentes) causes of sunburn or illness.43 Despite not being necessary and sufficient for the (continued) existence of their effects and not being in themselves active, these non-efficient causes are saliently different from mere background conditions, sine quibus non causes, or inert causes. While it is hard to spell out what "activity" amounts to in each of these contexts, it is clear that in the examples just mentioned the father, 41 Later medieval accounts of sine quibus non causes, such as that of Durand of St.-Pourçain (cf. Solère, "Sine qua non Causality"), also typically viewed them as inert. Since antiquity, several thinkers (including at least some Stoics) have contested that such inert items should be regarded as any kind of cause. the ox, and the sun all do something. Such items are plausibly viewed as being active causes in some fairly robust sense (let "robust" here signify whatever is exemplified in the cases just described) and yet, according to the relevant causal conceptions, none of these should be considered efficient causes of the relevant effects but should instead be regarded as proximate, cooperative, or antecedent causes.44 While the kind of causal taxonomy just described owes much to the Stoics, it was widely prevalent in later antiquity (as noted in Sextus Empiricus, Outlines of Pyrrhonism 3.14-16) and was shared by many Hellenistic and post-Hellenistic thinkers, including medical writers (such as Galen, notably in De causis continentibus and De causis procatarcticis), Christian theologians (such as Clement of Alexandria, Stromata 8.9.25.1ff., 33.1-9, and Augustine's contemporary and opponent, Julian of Eclanum, cf. Augustine, Contra Iulianum opus imperfectum 5.41-42), and later Platonists.45 Such thinkers may have strongly disagreed about whether the soul was incorporeal, whether incorporeal items had robust causal powers, or whether a particular item might be regarded as a particular kind of cause.46 However, they nonetheless shared a similar conception of efficient causes as being not merely active but as being active, productive, and sustaining (cf. Galen, De causis continentibus 1.1-2.4). They also recognised similar varieties of non-efficient causes, including various active non-efficient causes, in virtue of the kind of causal contributions they made (e.g. Clement, Stromata 8.9.26.1.1-7; Sextus Empiricus, Outlines of Pyrrhonism 3.15).47 Most saliently, although later Platonists -much like Augustine -have frequently been misunderstood on these issues, they do not seem to have claimed that bodies cannot act upon souls.48 Instead, they merely claimed that bod- 44 The causal taxonomy and nomenclature is complex and precisely how each of the relevant causes should be characterised in a given text is controversial and merits its own discussion. Here I merely rely on the kinds of examples invoked in the ancient discussions to give a seemingly paradigmatic example of each kind of cause or causal contribution. 45 Moreover, as can be observed in writers such as Seneca (e.g. Epistulae 65.8-11), the causal taxonomy described could interact in complex ways with an expanded Aristotelian causal taxonomy. 46 Augustine was aware of many of the relevant debates and indeed exploits them when he argues for the incorporeality of the mind in De Trinitate. See Nawar, "Augustine's Master Argument." 47 Cf. Plotinus, Enneads 5.  3.8-10). That is to say, although bodies might act only by participating in incorporeal items (e.g. Proclus, Elements of Theology 80), even Plotinus -whose remarks on the issue are occasionally stark and liable to produce misunderstandings -recognises that bodies do nonetheless act (e.g. Enneads 4.7.8a.1 ff.).49 It thus seems that later Platonists merely ruled out that alterations in bodies might produce certain kinds of alterations (e.g. self-moved alterations) in souls (Enneads 3.6.1.12-17, 4.34-43) while recognising that bodies exercise certain kinds of causal influence upon the soul (e.g. Proclus, In Platonis Timaeum 1.2. 1-11;In Platonis Parmenidem 1059.11-19). Accordingly, although later Platonists thought that bodies were not efficient causes, they nonetheless recognised bodies as, e.g., auxiliary or cooperative or proximate (e.g. προσεχὲς) causes which are active while not being in themselves active.50 On Augustine's view, the soul acts upon the body as an efficient cause and he seems to conceive of efficient causes in a manner very much in line with the widely shared Hellenistic and post-Hellenistic causal conceptions described above. Thus, for instance, Augustine calls the soul an efficient cause (e.g. Civ. Dei 5.9 [cited above]) and he follows the Neoplatonists in thinking that the soul exercises a kind of creative or productive causal power. More concretely, the soul continuously provides form and unity to the body and the body thus "subsists through the soul" (per animam ergo corpus subsistit, Imm. an. 15.24;cf. Imm. an. 7.12). Without the soul being present to it, the body ceases not only to be alive, but even to persist as a unified thing (Imm. an. 11.14; De quantitate ani- 49 At one point, Plotinus suggests that when the body is affected, the soul does not undergo any qualitative alterations (ἀλλοιώσεις) (Enneads 3.6.2.32-41, 4.34-43), but his more considered view seems to be that it does not undergo qualitative alterations of a certain kind (Enneads 3.6.  Mus. 6.5.9ff.;Gn. litt. 7.18.24). In line with the relevant conceptions of efficient causes, Augustine sees the soul as an efficient cause which is active, productive, and sustaining. What then of the causal powers of bodies? Augustine calls bodies "proximate" causes (e.g. Trin. 3.2.7-3.8) and also explicitly speaks of bodies being "antecedent" causes (praecedentae causae, Gn. litt. 6.9.14;cf. Gn. litt. 4.32.49), saying that a person's father, grandfather, or great-grandfather may be regarded as their antecedent cause (an example he illustrates with reference to Levi and Abraham). Augustine thus uses terms which Cicero and Seneca had used to speak of antecedent causes and cooperative causes (Cicero, De fato 18.41; Seneca, Epistulae 87.31), i.e. non-efficient causes which are active while nonetheless not acting per se, and whose continued presence and activity is seemingly not necessary and sufficient for the continued existence of their effects.51 I know of no definitive proof that Augustine himself endorsed the relevant conceptions of non-efficient causes described above (i.e. as items which may be active while not being active, productive, and sustaining). However, the relevant causal conceptions were widely prevalent in later antiquity, Augustine and at least some of his contemporary interlocutors, such as Julian of Eclanum, were clearly familiar with the relevant non-efficient causal conceptions (e.g. of antecedent causes, proximate causes, and so on, cf. Civ. Dei 5.2-10; 7.28; Contra Iulianum opus imperfectum 5.41-42), Augustine's own causal vocabulary often mirrors what we find in Cicero and Seneca,52 and Augustine was deeply indebted to numerous adjacent Stoic and later Platonist views and discussions (including Platonist views concerning how the incorporeal soul acts upon the body as an efficient -i.e. active, productive, sustaining -cause Augustine's debt to the Platonists, and especially some of Plotinus' discussions (e.g. Enneads 1. 2, 6; 5.1; 6.4-5, 9), is well known and touched upon above. Augustine's engagement with and debt to Stoic views is also deep. Thus, for instance, Augustine was clearly influenced by: the Stoic notion of apprehension (κατάληψις, perceptio, comprehensio) and knowledge (ἐπιστήμη, scientia) (e.g. Contra Academicos 2.5.11, 6.14; 3.9.18, 21; cf. Nawar, "The these facts, something like the causal taxonomy described above seems to constitute a promising framework for understanding Augustine's remarks about non-efficient causes and the causal powers of bodies. Moreover, it deserves attention that Augustine's own characterisation of bodies does not provide any reason to deny them the relevant kind of active non-efficient causal efficacy. In contrast to some later occasionalist thinkers, such as Nicolas Malebranche, who viewed bodies as purely passive,54 Augustine does not characterise bodies in terms of passivity or as lacking any active causal powers whatsoever. In fact, Augustine does not even characterise bodies by their ability to be acted upon (contrast, for instance, Proclus, Elements of Theology 80; as was noted above, this characterisation only rules out per se activity while allowing that bodies may act). Instead, throughout his works, Augustine characterises body as that which is extended in three dimensions and which occupies space or place in such a way that its proper parts occupy less space than it does.55 Although Augustine thinks that incorporeal souls have certain causal powers which bodies lack and the precise nature of the causal power exerted by non-efficient causes is not made clear by Augustine, his remarks on the limited causal powers of bodies (e.g. that they are weak in comparison with many incorporeal items, Imm. an. 6.11, 13.22) seem to assume that they possess causal powers of some sort or other,56 and this is something that Augustine explicitly Stoic Account"; Nawar, "Augustine's Defence," 222-223); Stoic views of mixture (e.g. Ep. 137; Trin. 9.4.7); Stoic views of seminal reasons (e.g. Gn. litt. 9.17.32; 10.20.35 ff.); Stoic views of voluntas (i.e. ὁρμή [impulse]; cf. Civ. Dei 5.9; Brittain, "Non-Rational Perception," 285; Byers, Perception); Stoic views about the content of perception (Brittain, "Non-Rational Perception"); and Stoic views of language (cf. Nawar, "Every Word is a Name"). Augustine's debt to Stoic views of causation is briefly suggested, but not substantiated, by Stead, "Causa." The present paper offers, among other things, a development of that suggestion. 54 Cf .  Such remarks seem to confirm that Augustine does not deny fairly "robust" causal powers to bodies and instead sees corporeal items as having genuine causal efficacy. Although the precise manner in which corporeal items and incorporeal souls causally interact is difficult to positively characterise (it is clearly not, for instance, through regular contact given that souls lack extension or resistance) and may admittedly seem somewhat mysterious,59 there is ing from their union'? I understand nothing by these general terms … that bodies should receive in themselves a certain power, by the efficacy of which they could act on the mind -this I do not understand. For what would this power be? Would it be a substance, or a modality? If a substance, then bodies will not act, but rather this substance in bodies. If this power is a modality, then there will be a modality in bodies which will be neither motion nor figure. Extension will be capable of having modalities other than relations of distance." 57 Augustine elsewhere claims that the seeds (semina, cf. Here it suffices to note that, even if such causal powers and "seeds" depend upon God (as the causal powers of souls also do, cf. Trin. 3.9.16), corporeal things nonetheless have genuine causal efficacy. 58 Gn. Litt. 9.17.32: "Et elementa mundi huius corporei habent definitam vim qualitatemque suam, quid unumquodque valeat vel non valeat, quid de quo fieri possit vel non possit." 59 There are two relevant issues worth noting and distinguishing: (a) the causal contributions of souls and bodies; and (b) how these causal contributions might occur. The issue of (b), i.e. the precise 'mechanics' of how bodies and incorporeal entities may causally interact, posed significant difficulties for several ancient thinkers but is generally not addressed in detail in extant texts. Notably, the Stoics, who were corporealists and characterised bodies by their ability to act and be acted upon (e.g. Cicero, Academica 1.39; Sextus Empiricus, Adversus Mathematicos 8.263; 9.211; Seneca, Epistulae 106.4-5; 117.2) (and also as being extended in three dimensions and perhaps possessing resistance, e.g. Diogenes Laertius, no reason to think that incorporeal souls are somehow isolated from the genuine causal efficacy of bodily entities. In sum, while the soul continuously acts upon the body and is an efficient (i.e. sustaining) cause of the body's life and unity, corporeal items may nonetheless causally influence the soul, and Augustine regards bodies as "proximate" and "antecedent" causes. Augustine does not make the nature of such causal activity clear but according to a widely prevalent Hellenistic and post-Hellenistic set of causal conceptions, which Augustine certainly knew and which probably informed his thought, bodies which are proximate or antecedent causes might be active (e.g. as a father is a cause of his offspring or one ox in a team of two is a cause of a cart being drawn) while nonetheless falling short of being sustaining and thus falling short of being efficient causes. Although bodies do not act per se and are causally weaker than souls, they nonetheless contain genuine causal power and efficacy and Augustine speaks of the soul as being affected by the body. In light of these facts, Augustine is best understood as claiming that, while bodies are not efficient causes (i.e. causes which are jointly active, productive, and sustaining and which are per se active), they may nonetheless be active and may act upon souls.

Conclusion
In this paper, I have sought to clarify Augustine's views about causation, metaphysical dependence, and soul-body causal relations. I hope to have shown that while Augustine discerns an important hierarchy of metaphysical dependence according to which efficient causes are prior to their effects, in asserting (no-bodily-bringing-about), i.e. that corporeal things do not bring about anything in the soul, Augustine does not maintain that corporeal things do Lives of the Philosophers 7.134-135; Galen, Quod qualitates incorporeae sint 19.483, 13-16 = Long and Sedley 45F), seem to have struggled to find some causal role for incorporeal entities (e.g. Sextus Empiricus, Adversus Mathematicos 8.397-410, especially 409: ἐπ' αὐτοῖς … καὶ οὐχ ὑπ' αὐτῶν). Augustine, who does not characterise bodies in terms of their ability to act and be acted upon does not face the same challenges, but is also not especially forthcoming on (b). However, he does discuss (a), most notably when speaking of perception and how the soul may control, enable, and respond to certain things within the body (an issue which he often discusses by invoking the soul's intentio, cf. De quantitate animae 33.70; Mus. 6.5.9-12; Gn. litt. 7.19.25; Trin. 11.2.5). For detailed discussion of the causal contributions of souls and bodies in the case of perception, see Nawar, "Augustine on Active Perception." not act upon or exercise any causal influence upon the soul. Instead, "bringing about" denotes a particular kind of causal influence: that exercised by efficient causes. Thus, (no-bodily-bringing-about) amounts to the claim that corporeal items may not act as efficient causes upon the soul. However, this leaves open that corporeal items may act as other kinds of causes. As we have seen, Augustine claims that corporeal things may be "proximate" or "antecedent" causes. In light of Augustine's own remarks and the ancient philosophical causal conceptions which he himself discusses and which most likely informed his own views, I suggested that Augustine is best understood as thinking of efficient causes as items which are active, productive (or creative), and sustaining while thinking that at least some non-efficient causes, such as antecedent and proximate causes, may nonetheless be active and may act upon other items (including souls). This reading, I argued, is supported by the fact that Augustine does not characterise bodies as being purely passive and instead seems to allow them genuine causal efficacy which goes beyond that afforded to mere occasions, sine quibus non causes, or background conditions. Thus, on the reading offered here, (no-bodily-bringing-about) is consistent with psychosomatic interaction and with bodies acting upon souls, and Augustine's views concerning the causal relations between body and soul turn out to be largely consistent. If the reading offered here is right, then we can better understand not only Augustine's views of causation and metaphysical dependence and what Augustine means when he says bodies do not "bring about" anything in souls, but also how certain readers of Augustine might verbally agree with some of his claims while going significantly further than he himself does in limiting the causal powers of corporeal things.