Chapter 4 - The Causality of the Unmoved Mover

Summary

The fourth chapter examines the problem of the causality of the unmoved mover. This issue is central in scholarship on Aristotle and goes back to late antiquity. I argue that here Proclus’ non-harmonist stance towards Aristotle emerges most strongly: not only did Aristotle fail to make the intellect an efficient cause of the cosmos’ being but his metaphysics generally is deficient, since he did not recognise the Platonic One as the highest principle. I contrast Proclus’ view with the position of Ammonius and Simplicius who see a complete agreement between Plato and Aristotle.

4.1 Is the Unmoved Mover a Final Cause, an Efficient Cause, or Both?

While Chapters 1 and 2 have focused on Proclus’ integration of the intellect as prime mover in his system of movers, the question still arises regarding the prime mover’s causality not just in his own philosophy but also in his exegesis of Aristotle. For, as is widely acknowledged, one of the perennial questions of Aristotelian scholarship concerns the type of causal relationship between the prime mover and the universe. It seems well-established in Physics 8 and Metaphysics 12 that the unmoved mover is ultimately responsible for the eternal motion of the universe. Yet, it remains obscure how it causes this motion and whether the two accounts are even compatible. This ambiguity is fundamentally due to the limited description – especially in the Physics – of the unmoved mover’s mode of operation, which has led to fierce debates among scholars. Just to give a brief overview, in recent scholarship Judson (1994; 2019) has maintained that the two accounts are coherent and that Aristotle’s unmoved mover is an efficient cause insofar as it is a final cause – that is, by being an object of desire to the heaven it can be regarded as an efficient cause of the heaven’s desire and, thus, remotely of its motion. This view has been rejected by Gourinat (2012), who – like Manuwald (1989) before him – maintains that the unmoved mover is in both works only a final cause. In contrast to this position, Berti (2007) claims that the unmoved mover is solely an efficient cause of the heaven’s motion. Most importantly, the vast majority of scholars who assume the efficient causality of the prime mover only regard it as an efficient cause of motion, not of being.

The origins of the dispute regarding the causality of the Aristotelian prime mover can be traced back to antiquity. Particularly among late Neoplatonists the problem becomes a central concern in Aristotelian exegesis, arguably due to the need to harmonise Aristotle’s intellect with Plato’s demiurge and to account for the cause of the generation of the cosmos (Simpl. In Phys. 1360.24–31). Crucially, the issue is a major source of contention between Neoplatonists who believe there is an essential agreement between Plato and Aristotle and those who do not endorse this view. Unlike many scholars nowadays, the Neoplatonists ascribe to Aristotle a unitary and systematic theory of the unmoved mover, found not only in Physics 8 and Metaphysics 12 but also in the De caelo.² The most influential interpretation, especially in consideration of its medieval reception, is the one proposed by Ammonius and his pupil Simplicius. Both argue that Aristotle’s unmoved mover is not just a final cause but also an efficient cause of the cosmos’ motion and being – that is, it generates the cosmos. Especially the latter is in sharp contrast to the opinion of most modern scholars. What is precisely meant by this is obscured by the fact that Simplicius dedicates little space to the question and mostly offers us a few testimonies from Ammonius’ – now lost – book on this issue which was central to this debate.

Unlike Ammonius and Simplicius, Proclus criticises Aristotle’s unmoved mover for being exclusively a final cause and not an efficient cause of being as well:

And indeed the inspired Aristotle seems to me for this reason, in preserving his first principle free of multiplicity, to make it only the final cause of all things, lest in granting it to produce (ποιοῦν) all things, he should be forced to grant it activity towards what follows upon it (τὴν πρὸς τὰ μετ’ αὐτὸ ἐνέργειαν); for if it is only the final cause, then everything exercises activity towards it, but it towards nothing.

(In Parm. 5.1169.4–9)

Thus, for Proclus, Aristotle’s intellect is ‘in no way productive’ (In Tim. [1.390.6]: ποιητικὸς δὲ μηδαμῶς). As encountered previously, this attitude is in line with his non-harmonist and more critical approach towards Aristotle. As one of the earliest extensive engagements with the causality of the unmoved mover, Proclus’ critique plays a pivotal role and prefigures many ancient and medieval discussions on this issue. Indeed, as I show, some of the arguments employed by Ammonius and Simplicius in defending the unmoved mover’s efficient causality are found in Proclus in a more elaborate way. The major difference is that Proclus, unlike Ammonius and Simplicius, does not ascribe the results of these arguments to Aristotle. As I emphasise, Proclus’ interpretation is closer to modern views on Aristotle and, indeed, should be preferred to Ammonius’/Simplicius’ reading as it is closer to the meaning of Aristotle’s text. I will also demonstrate that the way these authors interpret Aristotle is grounded in their general views on Aristotle’s relationship with Plato. For this, I offer the first in-depth comparison of these authors on such a challenging issue.

My objective in this chapter is threefold: (1) to set out Proclus’ criticism of Aristotle’s intellect through a detailed analysis of his objections; (2) to compare it with Ammonius’ and Simplicius’ position by focusing on their different strategies in reading Aristotle; and (3) to present Proclus’ own reasons for making the demiurge a final cause and efficient cause of being, which are connected to his critique. The chapter is split into four sections. I first set out briefly Aristotle’s own view on the causality of the intellect (4.2), before I move on to Proclus’ critique (4.3). I elucidate how this specific criticism is part of a general attack on Aristotelian metaphysics which Proclus regards as deficient. In defending his view of the demiurge’s causality, Proclus chides Aristotle numerous times for rejecting the efficient causality of the intellect. I reconstruct two central objections in which Proclus demonstrates that Aristotle’s own principles would have committed him to accept the intellect as efficient cause of being. Aristotle himself, however, did not draw this conclusion, as Proclus makes clear. Then (4.4) I set out the views of Ammonius and Simplicius, who regard Aristotle’s intellect as final cause as well as efficient cause of being. I show that they partly use the same arguments as Proclus with the crucial difference that these Neoplatonists ascribe them completely to Aristotle. As I demonstrate, their strategy of reading Aristotle differs from Proclus’ more critical position because of their commitment to harmonising Aristotle with Plato on fundamental issues which Proclus does not share. This emphasises that Aristotle’s authority is not the same in Proclus as in Ammonius and Simplicius. By reconstructing this late antique debate I render these different approaches to Aristotle among the Neoplatonists more palpable. Finally, (4.5) I discuss Proclus’ positive views on the subject matter. As I show, he backs up his view of the demiurge’s causality not only by his general metaphysical theory of causation found in the Elements of Theology but also by an exegesis of Plato’s Timaeus. The former offers an attractive theoretical solution to why we should assume that the intellect is both a final and an efficient cause.

4.2 Aristotle

I briefly outline in the following my own interpretation of Aristotle’s views. I do not have space to do justice to the complexity of this question nor to the wide variety of interpretations. It remains, nevertheless, necessary to introduce the discussion of the Neoplatonist positions with a treatment of Aristotle as it inevitably influences my analysis of them. Part of my intention is to show that the unmoved mover’s causality is just as controversial nowadays as it was in late antiquity. As it emerges, various points of contention are very similar and centred around the same passages. Since the prime mover’s final causality has rarely been called into question, the focus is on the prime mover’s efficient causality, which has been negated by Aristotle’s commentators since antiquity. The meaning of efficient causality in this context is often obscure in modern scholarship. The majority of scholars understand it as a cause of motion and not of being like some ancient commentators. Yet, whether this causation of motion implies a transmission of force or energy from the unmoved mover to the cosmos is a matter of debate.

Oddly enough for a treatise meant to explain the origin of the cosmos’ motion, Aristotle is surprisingly taciturn in Physics 8 when it comes to how exactly the unmoved mover brings it about.³ Characterisations of the unmoved mover as either final or efficient cause seem vague. This issue becomes even more pressing if we consider Aristotle’s effort in Physics 2 to set out a nuanced theory of causality (which, however, applies primarily to natural substances).⁴ Due to this perceived ambiguity, scholars like Manuwald (1989) and Gourinat (2012) have abandoned the identification of the unmoved mover with efficient causality. Yet, there still remain numerous scholars who take this very position (as we will see in the following sections). The picture differs in Metaphysics 12 where the prime mover is described as an object of desire and thought as well as something for the sake of which (οὗ ἕνεκα)⁵: it moves as a beloved (12.7.1072b3: κινεῖ δὴ ὡς ἐρώμενον). These descriptions have led to the widespread view that the prime mover there is a final cause. How then are we supposed to square this position with the view offered in Physics 8?

There are strong reasons for assuming that both accounts of the unmoved mover in Physics 8 and Metaphysics 12 are essentially in agreement and complement each other, although the contexts and approaches clearly differ.⁶ The argument for the unmoved mover in Metaphysics 12 is highly dependent on Physics 8 and, in fact, to a large degree unintelligible without the latter.⁷ It is thus incorrect to claim that the ‘conceptions of the First Cause developed more or less independently in the Physics and Metaphysics’ (Wardy 1990: 123). Moreover, De motu animalium (1.698a7–11, 6.700b7–9) refers indiscriminately to both works for the underlying theory of motion without a hint of a substantial difference between them.⁸ In the following, I consider three arguments for the efficient causality of the unmoved mover. The first evidence is the way the argument is sustained in Physics 8 (4.2.1). As further proof, I examine the infinite power argument of Physics 8.10 which strongly suggests an efficient causality of the unmoved mover and is, most importantly, also encountered in Metaphysics 12 (4.2.2) where we find further evidence for this type of causality (4.2.3).

Before I examine these two works, I would like to consider the claim that the prime mover cannot be both an efficient and final cause on general grounds. The widespread view that whatever is a final cause cannot be an efficient cause is based on an interpretation of GC 1.7.324b13–15:⁹

Ἔστι δὲ τὸ ποιητικὸν αἴτιον ὡς ὅθεν ἡ ἀρχὴ τῆς κινήσεως. Τὸ δ’ οὗ ἕνεκα οὐ ποιητικόν. Διὸ ἡ ὑγίεια οὐ ποιητικόν, εἰ μὴ κατὰ μεταφοράν.

The thing which is efficient is a cause in the sense of that from which motion originates. The final cause is not efficient. Therefore, health is not efficient, except metaphorically.

(tr. mine)

Proponents of this interpretation are, for instance, Manuwald (1989: 16) and Gourinat (2012: 176), who regard this as evidence that the unmoved mover can be only a final cause.¹⁰ In contrast to these scholars, Sedley (2000: 345) and Judson (2019: 185–6) maintain that the passage does not apply to the unmoved mover. This is either because Metaphysics 12 simply goes beyond the doctrine of De generatione et corruptione¹¹ or because Aristotle refers in the De generatione et corruptione passage to ‘those cases of being active which involve interaction, and by the same token he is thinking of final causes such as health which are clearly not active’ (Judson 2019: 185).

While I sympathise with Sedley’s and Judson’s conclusion, that is, that the unmoved mover can have both types of causality, I do not think they offer strong arguments for rejecting the prima facie reading of 324b13–15. Rather, I take it that the point of the passage is to emphasise that being poiētikon (ποιητικόν) automatically entails being an origin of motion, whereas a final cause – since it is not strictly speaking producing something – does not have to be an origin of motion. According to GC 1.6.322b22–4, to be productive stricto sensu (κυρίως) implies a mutual contact between mover and moved object. This only applies to moved movers but not to unmoved movers who can only have non-reciprocal contact with the moved objects. Nevertheless, in an extended sense¹² a final cause can be productive and thus an origin of motion. A good example for this is the soul which Aristotle characterises as final, efficient and formal cause (DA 2.4.415b8–12). Additionally, in GC 2.9.335a30–2 he admits that there is an efficient cause for eternal beings, that is, the heaven and stars.

4.2.1 The Argument of Physics 8 Requires an Efficient Cause

The line of argumentation developed in Physics 8 generally suggests an investigation into the efficient cause of the cosmos’ motion since Aristotle is looking for the origin of motion and conducts his discussion in efficient terms. The view has been proposed by Broadie and Judson as an evident fact without much further investigation.¹³ Aristotle himself refers in De generatione et corruptione (1.3.318a1–6) to the prime mover of the Physics as an efficient cause. Internal confirmation from Physics 8 for this view can be found in chapter 4. There Aristotle proves that everything in motion is moved by something (256a2–3: ἅπαντα ἂν τὰ κινούμενα ὑπό τινος κινοῖτο) – a phrase clearly indicating that efficient causality is discussed here, that is, the moving cause. More specifically, the preposition hupo (ὑπό) with the genitive indicates agency in this context.¹⁴ At no point in the argument of chapter 4 does Aristotle distinguish between the causation of the unmoved mover and moved movers. Instead, he talks about causes of motion in general. However, elsewhere he entertains the possibility of only one-sided or non-reciprocal contact in the case of unmoved movers, which would imply that they bring about motion differently from moved movers. For instance, at 8.5 258a18–21 the unmoved part in a self-mover is presented as either being in reciprocal contact or only touching the moved thing while not being itself touched by it.¹⁵ This presumably has to do with the unmoved mover’s immateriality. Even if the prime mover causes the cosmos’ motion either without any contact or by non-reciprocal contact, it still acts as an efficient cause of the motion and is, as such, treated together with other moving causes. There is no reason to assume that causing motion without contact or, at least, a non-reciprocal one excludes being an efficient cause. More puzzling is rather Aristotle’s view that motion can be caused with non-reciprocal contact in the first place. This is due to the non-/super-natural origin of motion in the cosmos.

However, Gourinat (2012) has recently rejected this interpretation: while a great deal of the argumentation in Physics 8 seems to be looking for an efficient cause of motion, he argues that the introduction of an unmoved mover changes the type of causation under discussion.¹⁶ According to Gourinat, when Aristotle posits an unmoved mover – either as part of a self-moving animal or as the prime mover itself – he is no longer investigating the efficient cause of motion. He bases his claim on the consideration that unmoved movers cause motion differently than moved movers, which is grounded in a short passage from Physics 7.2:

Τὸ δὲ πρῶτον κινοῦν, μὴ ὡς τὸ οὗ ἕνεκεν, ἀλλ’ ὅθεν ἡ ἀρχὴ τῆς κινήσεως, ἅμα τῷ κινουμένῳ ἐστί (λέγω δὲ τὸ ἅμα, ὅτι οὐδέν ἐστιν αὐτῶν μεταξύ)· τοῦτο γὰρ κοινὸν ἐπὶ παντὸς κινουμένου καὶ κινοῦντός ἐστιν.

The prime mover [of a thing] – which does not supply that for the sake of which but the source of the motion – is always together with the moved object (by ‘together’ I mean that there is nothing between them). This is common to everything moved and moving.

(243a32–35; tr. mine)

Here Aristotle distinguishes between a proximate prime mover, which is moved, and the ultimate prime mover, which is unmoved.¹⁷ Gourinat takes this to be a general distinction between the workings of moved movers and unmoved movers. The former act as efficient causes by transmitting motion via reciprocal contact. However, as outlined, the contact between an unmoved mover and moved thing is only one-sided, that is, the unmoved mover touches the moved thing but is not touched by it in turn. This heterogeneity between unmoved mover and moved thing – to be contrasted with the homogeneity between moved mover and moved thing – indicates to Gourinat a ‘causal heterogeneity’. He thus concludes that, unlike moved movers, unmoved movers do not cause motion as efficient causes but instead only as final causes.¹⁸

I do not find this view convincing since Gourinat works with a very narrow understanding of efficient cause, which seems to imply that a mover is only an efficient cause if a contact on both sides of mover and moved occurs.¹⁹ This is due to a tendentious reading of Physics 7.2 whereby moved movers are the only movers identified with this type of causation. Yet, in this passage Aristotle does not exclude that the prime unmoved mover is an efficient cause but only that the prime moved mover is a final cause. Aristotle’s whole point is to distinguish moved movers from unmoved movers by pointing out the former’s lack of final causality. Consequently, this does not entail that the prime unmoved mover is not an efficient cause.²⁰ More generally, Physics 8 should not be read by automatically importing doctrines from book 7 – whose standing in the Physics is questionable anyway – as book 8 offers a new start in the discussion. Rather, one has to consider his numerous expressions throughout book 8 which indicate that efficient causality is under discussion. A good example for this is found in the next section.

4.2.2 The Unmoved Mover Transmits Power (Physics 8.10 and Metaphysics 12.7)

The so-called infinite power argument in Physics 8.10 implies that the prime mover transmits power (dunamis) to the thing it moves and is thereby an efficient cause. This argument, which is taken up again in Metaphysics 12 has caused great puzzlement especially among scholars who regard the unmoved mover exclusively as a final cause.²¹ As one of the most (in)famous arguments for the causal efficiency of the unmoved mover it has proven to be immensely influential (but also controversial) in late antiquity and the Middle Ages.²² Aristotle sets out to prove through various reductiones ad impossibile the indivisibility of the unmoved mover via its lack of a magnitude:

1. (1) No finite thing can cause motion for an infinite time. (266a12–23)

2. (2) No infinite power can belong to a finite magnitude. (266a24–266b6)

3. (3) No finite power can belong to an infinite magnitude. (266b6–24)

These reductiones lead him to the following conclusion regarding the unmoved mover:

εἰ γὰρ μέγεθος ἔχει, ἀνάγκη ἤτοι πεπερασμένον αὐτὸ εἶναι ἢ ἄπειρον. ἄπειρον μὲν οὖν ὅτι οὐκ ἐνδέχεται μέγεθος εἶναι, δέδεικται πρότερον ἐν τοῖς φυσικοῖς· ὅτι δὲ τὸ πεπερασμένον ἀδύνατον ἔχειν δύναμιν ἄπειρον, καὶ ὅτι ἀδύνατον ὑπὸ πεπερασμένου κινεῖσθαί τι ἄπειρον χρόνον, δέδεικται νῦν. τὸ δέ γε πρῶτον κινοῦν ἀΐδιον κινεῖ κίνησιν καὶ ἄπειρον χρόνον. φανερὸν τοίνυν ὅτι ἀδιαίρετόν ἐστι καὶ ἀμερὲς καὶ οὐδὲν ἔχον μέγεθος.

For if it has magnitude, the magnitude must be either finite or infinite. That there cannot be an infinite magnitude has already been proved in the Physics. That a finite magnitude cannot have infinite power, and that something cannot be moved for an infinite time by a finite magnitude, has just been proved. But the first mover causes everlasting motion for an infinite time. Plainly, then, it is indivisible and without parts, and it has no magnitude.

(267b19–26)

Aristotle deduces that since the prime mover can be neither a finite nor an infinite magnitude it must be without magnitude. He does not attribute infinite power explicitly to the unmoved mover. However, one reason for excluding that the unmoved mover is a finite magnitude is the impossibility of infinite power residing in a finite magnitude. This in turn implies that the unmoved mover must have infinite power and therefore cannot be a finite magnitude. Otherwise, it is impossible to explain why infinite power is even a concern here and part of his argument. Similarly, Aristotle shows that a finite magnitude cannot move something infinitely. Again, here the implication is that the unmoved mover must move something for an infinite time and therefore cannot be a finite magnitude. Thus, both arguments contain attributes of the unmoved mover (i.e., infinite power and capacity to move something for an infinite time) which cannot belong to a finite magnitude. In fact, both are connected: the capacity to move something for an infinite time implies having an infinite power and vice versa.

The same attribution is found in Metaphysics 12.7 whose discussion is doubtless referring back to Physics 8.10:²³

δέδεικται δὲ καὶ ὅτι μέγεθος οὐδὲν ἔχειν ἐνδέχεται ταύτην τὴν οὐσίαν ἀλλ’ ἀμερὴς καὶ ἀδιαίρετός ἐστιν (κινεῖ γὰρ τὸν ἄπειρον χρόνον, οὐδὲν δ’ ἔχει δύναμιν ἄπειρον πεπερασμένον· ἐπεὶ δὲ πᾶν μέγεθος ἢ ἄπειρον ἢ πεπερασμένον, πεπερασμένον μὲν διὰ τοῦτο οὐκ ἂν ἔχοι μέγεθος, ἄπειρον δ’ ὅτι ὅλως οὐκ ἔστιν οὐδὲν ἄπειρον μέγεθος)·

And it has also been proved that this same substance can have no magnitude, but is partless and indivisible. For it causes motion for an infinite time, and nothing finite can have an infinite power. Now every magnitude is either infinite or finite; but it could not have a finite magnitude for this reason, nor an infinite one because there is no infinite magnitude of any sort.

(1073a5–11)²⁴

Here too Aristotle connects moving something for an infinite time with having an infinite power to do so. The argument is used, as in Physics 8, for the purpose of demonstrating the unmoved mover’s lack of spatial extension. Just like there, it seems impossible for the same reasons not to read the passage as ascribing infinite power to the unmoved mover.

Unfortunately, Aristotle fails to explain in both passages how the prime mover uses this power to cause the cosmos’ motion. The discussion in Physics 8.10 seems to make clear that the power is somehow transmitted to an object and allows it to move in a broad sense: Aristotle uses not only the examples of heating, sweetening and throwing but causing motion in general (266a28: ὅλως κινοῦσα). In all of these cases the power or energy of the moving thing is transmitted to the moved object. However, Judson (1994: 165–6) and Laks (2000: 241) point out that the unmoved mover is simply not the type of efficient cause that transmits its own motion or energy like, for example, a human wielding a stick.²⁵ This is because the unmoved mover is not spatially extended and moves the heaven by instilling desire through its own goodness. As such, the modus operandi of an efficient cause like the unmoved mover differs fundamentally from other efficient causes. While this leads Judson to conclude that the infinite power argument is simply incompatible with any account of the unmoved mover’s causation in the Physics and the Metaphysics, Laks only points out that the transmission of δύναμις must have a metaphorical sense here.²⁶ Both of these explanations are far from satisfying.²⁷ As I show below, the Neoplatonists offer an interesting solution to harmonising the infinite power argument with the prime mover’s final causality.

Since the infinite power argument suggests that the unmoved mover is somehow an efficient cause and not just a final cause, it is especially problematic for interpretations of the unmoved mover as an exclusively final cause, such as Gourinat’s (2012), who offers no explanation of how his interpretation relates to this argument.²⁸ Yet, it also seems hardly compatible with current accounts of the unmoved mover’s efficient causality, as proposed by Broadie, Berti or Judson. Broadie (1993), for instance, ignores it altogether, as do also Ross (1924: II, 382) and Fazzo (2014: 341–2) in their comments on Metaphysics 12.7.²⁹ Additionally, the issue is aggravated by the argument’s presence in Physics 8 and Metaphysics 12.7 so that unlike, for instance, the much-disliked passage on the location of the unmoved mover – which only occurs in Physics 8 – this discussion cannot be simply explained away by assuming a development. In this way, both the overall structure of the argument in Physics 8 as well as the discussion of infinite power suggest that the unmoved mover is here conceived as an efficient cause. For Metaphysics 12, however, there is further proof that this type of causality should be attributed to the unmoved mover.

4.2.3 The Unmoved Mover as kinētikon and/or poiētikon (Metaphysics 12.6 and 10)

A crucial passage from Metaphysics 12.6 lends further support for this view:

Ἀλλὰ μὴν εἰ ἔστι κινητικὸν ἢ ποιητικόν, μὴ ἐνεργοῦν δέ τι, οὐκ ἔσται κίνησις· ἐνδέχεται γὰρ τὸ δύναμιν ἔχον μὴ ἐνεργεῖν. οὐθὲν ἄρα ὄφελος οὐδ’ ἐὰν οὐσίας ποιήσωμεν ἀϊδίους, ὥσπερ οἱ τὰ εἴδη, εἰ μή τις δυναμένη ἐνέσται ἀρχὴ μεταβάλλειν· οὐ τοίνυν οὐδ’ αὕτη ἱκανή, οὐδ’ ἄλλη οὐσία παρὰ τὰ εἴδη· εἰ γὰρ μὴ ἐνεργήσει, οὐκ ἔσται κίνησις. ἔτι οὐδ’ εἰ ἐνεργήσει, ἡ δ’ οὐσία αὐτῆς δύναμις· οὐ γὰρ ἔσται κίνησις ἀΐδιος· ἐνδέχεται γὰρ τὸ δυνάμει ὂν μὴ εἶναι. δεῖ ἄρα εἶναι ἀρχὴν τοιαύτην ἧς ἡ οὐσία ἐνέργεια.

Yet if there is something which can cause motion or act upon things, but is not active in some way, there will be no motion; for that which has a potentiality can fail to be active. Nor will it help, then, even if we posit substances which are eternal – as do those who posit the forms – unless there is some principle in them which is able to cause motion. Yet not even this will be sufficient, nor will another substance besides the forms; for unless it is active there will be no motion. Again, it will not be sufficient if it is active but its substance is potentiality; for there will not be eternal motion, since that which is potentially can fail to be. There must, therefore, be a principle of this sort, whose substance is activity.

(1071b12–20)

Aristotle argues here that it is not sufficient for the unmoved mover to be a moving (kinētikon) or producing (poiētikon) cause in potentiality. Rather, it must be so in actuality in order to cause the eternal motion of the cosmos. At any rate, it is clear that the unmoved mover must be an efficient cause, as the expressions κινητικόν and ποιητικόν indicate. This is backed up by his reference to the forms in the next line: insofar as these do not even have potentially a source of motion (δυναμένη … ἀρχὴ μεταβάλλειν), they cannot account for the eternal motion. What Aristotle’s theory requires is thus clearly an efficient cause in actuality, that is, one that has actual infinite power.

The formulations κινητικόν and ποιητικόν recur in chapter 10 but this time without the disjunctive:

ἀλλὰ μὴν οὐδέν γ’ ἔσται τῶν ἐναντίων ὅπερ καὶ ποιητικὸν καὶ κινητικόν; ἐνδέχοιτο γὰρ ἂν μὴ εἶναι. ἀλλὰ μὴν ὕστερόν γε τὸ ποιεῖν δυνάμεως. οὐκ ἄρα ἀΐδια τὰ ὄντα. ἀλλ’ ἔστιν· ἀναιρετέον ἄρα τούτων τι. τοῦτο δ’ εἴρηται πῶς.

In fact, not one of the opposites will also be able to act upon things and able to cause motion; for it would be able not to be. In fact, acting upon things is posterior to potentiality. Therefore, the things which are will not be eternal. But they are. Therefore, one of these must be eliminated: it has been said how this is to be done.

(1075b30–4)

Sedley (2000: 344–6) and Judson (2019: 361–2) rightly see this passage as connected to chapter 6. Unlike there, Aristotle here refers implicitly to the unmoved mover as ποιητικὸν καὶ κινητικόν and not κινητικὸν ἢ ποιητικόν. While it is unclear whether there is a real difference between these formulations, I assume that the conjunction καί at 1075b31 makes clear that, in fact, the ἤ at 1071b12 presents an equivalence, not an alternative.³⁰ That is, the unmoved mover can be described correctly by both terms, κινητικόν and ποιητικόν. The proximity of the two terms is also indicated by a passage from De generatione et corruptione: ἐν ἅπασιν εἰώθαμεν τοῦτο λέγειν τὸ ποιοῦν, ὁμοίως ἔν τε τοῖς φύσει καὶ ἐν τοῖς ἀπὸ τέχνης, ὃ ἂν ᾖ κινητικόν (2.9.335b27–8). Thus, both passages strongly suggest that the unmoved mover is an efficient, that is, moving and producing, cause.

4.2.4 Conclusion

In conclusion, there is significant evidence in Physics 8 and Metaphysics 12 for understanding the prime mover not just as a final cause but also as an efficient one. The general argument and especially the infinite power argument of Physics 8 present the unmoved mover as an efficient cause of the cosmos’ eternal motion – even though the details of the causation remain obscure. This account is then further developed (or at least elaborated) in Metaphysics 12. It thus seems fallacious to view the prime mover as solely an efficient cause (Berti 2007) or solely a final cause (Gourinat 2012).

Yet, the lack of an explicit discussion of the prime mover’s efficient causality, as well as the ambiguity of some of the passages discussed, posed a difficulty for future exegetes. This left Aristotle’s texts susceptible to differing interpretations, as the survey of different positions in scholarship showed. For instance, it remains questionable whether the prime mover is (1) a final cause by being an efficient cause or (2) an efficient cause by being a final cause. Frede (2000: 43–7) and Menn (2012b: 447) opt for (1), while Judson (1994: 164–7) and (2019: 185–6) goes for (2).³¹ As I show, the Neoplatonists who believe that the prime mover has both types of causality believe that one type of cause implies the other and vice versa so that there is no subordination of one to the other. A major issue remains of precisely how we are to understand efficient causality in this context. In the next two sections, I analyse two different reactions to this issue.

4.3 Proclus’ Critique of Aristotle’s Intellect

In a number of passages from his commentaries on the Timaeus and the Parmenides, Proclus criticises Aristotle’s intellect as being only a final cause and lacking efficient/productive causality.³² The latter is understood not just as causation of motion, as in most modern scholarship on Aristotle, but also of being. This is a very serious objection given Proclus’ Platonist conception of intellect as creative demiurge: ‘those, then, who make intellect a final but not a demiurgic cause possess only half the truth’ (In Parm. 4.842.20–2). Consequently, Aristotle’s prime mover is ἄγονος (842.26). The fundamentals of his critique are found in Proclus’ teacher Syrianus (see Section 4.3.3.4). However, it is in Proclus that we get the most extensive discussion.

In this section, I argue that

1. (1) Proclus’ critique is part of a more fundamental disagreement with Aristotle’s metaphysics.

2. (2) Consequently, Proclus maintains that Aristotle and Plato have different understandings of efficient causality and that Aristotle’s prime mover is not an efficient cause in the Platonic sense.

3. (3) Yet, Proclus believes that ultimately Aristotle’s arguments for establishing the existence of the prime mover commit him to conclusions more in line with Platonist doctrine. That is, if Aristotle had taken the premises of his arguments seriously, he would have been forced to conclude that the intellect is a cause of the cosmos’ being and not just of its motion.

4. (4) However, unlike Ammonius and Simplicius, Proclus does not believe that Aristotle actually drew these conclusions. Instead, Aristotle has compromised his metaphysics through a deficient understanding of the intellect’s causality. In this way, he is in disagreement with Plato’s concept of the demiurge.

Let us first consider Proclus’ general misgivings about Aristotelian metaphysics.

4.3.1 The Fundamental Deficiency of Aristotelian Metaphysics

In the following, I argue that, according to Proclus, Aristotle’s misunderstanding of the intellect’s causality is part of a general deficiency in Aristotle’s metaphysics. This, Proclus upholds, is caused by his confusion of the nature and the identity of the highest principle: Aristotle denies the existence of the One and instead mistakenly posits the intellect as first principle. Due to the parsimony of Proclus’ remarks,³³ this issue has not been appreciated enough in scholarship: rather, both Steel (1987a) and d’Hoine (2008) have emphasised that Proclus sees an interdependence between denying the intellect’s efficient causality and denying the existence of the paradigm.³⁴ Additionally, Steel suggests that for Proclus Aristotle’s rejection of the forms has ‘the most disastrous consequence’ (225: ‘la consequence la plus désastreuse’) of his inability to posit a higher principle than intellect. While this might be the case in the passage Steel focuses on (In Parm. 4.972.29–973.12; cf. In Tim. 2.91.4 [1.266.30]), I show that elsewhere Proclus presents the causal relationship differently: by denying the existence of the One and instead attributing some of its characteristics to the intellect, Aristotle rejects the intellect’s efficient causality.³⁵ In this way, Aristotle’s other metaphysical shortcomings follow from his rejection of the Platonic One and not vice versa, as in some of the texts on which Steel and d’Hoine base their analysis.

Aristotle’s repudiation of the One emerges more clearly from a passage in the commentary on the Timaeus where Proclus compares Aristotle with Plato and emphasises their differences:³⁶

(1) ὅσα γὰρ τῷ ἑνὶ Πλάτων, ταῦτα τῷ νῷ περιτίθησι, τὸ ἀπλήθυντον, τὸ ἐφετὸν, τὸ μηδὲν νοεῖν τῶν δευτέρων· (2) ὅσα δὲ τῷ δημιουργικῷ νῷ ὁ Πλάτων, ταῦτα τῷ οὐρανῷ καὶ τοῖς οὐρανίοις θεοῖς Ἀριστοτέλης· παρὰ [τούτων] γὰρ εἶναι τὴν δημιουργίαν καὶ τὴν πρόνοιαν· καὶ (3) ὅσα τῇ οὐσίᾳ τοῦ οὐρανοῦ ὁ Πλάτων, ταῦθ’ οὗτος τῇ κυκλοφορίᾳ, τῶν μὲν θεολογικῶν ἀρχῶν ἀφιστάμενος, τοῖς δὲ φυσικοῖς λόγοις πέρα τοῦ δέοντος ἐνδιατρίβων.

(1) What Plato attributes to the One, he ascribes to the intellect, that is, non-multiplicity, being the object of desire and not having any of the secondary things as object of its thought; and (2) what Plato attributes to the demiurgic intellect, Aristotle ascribes to the heaven and the heavenly gods, for it is from them that creativity and providence take place; and (3) what Plato attributes to the essential nature of the heaven, this man ascribes to its circular movement, placing theological principles at a distance and spending more time on physical argumentation than he should.

(In Tim. 2.132.15–133.4 [1.295.20–7])

Proclus describes here Aristotle’s tendency to ‘downgrade’ (metaphysical) attributes: (1) the characteristics of Plato’s One match those of Aristotle’s intellect, (2) those of Plato’s demiurgic intellect those of Aristotle’s heaven/heavenly bodies, and (3) those of Plato’s heaven those of Aristotle’s heavenly circular motion. This effectively leads to a misalignment of Plato’s and Aristotle’s principles (Figure 4.1).³⁷

Figure 4.1

The reason for that is said to be Aristotle’s distance from theological principles (τῶν μὲν θεολογικῶν ἀρχῶν ἀφιστάμενος) and undue focus on physical arguments (τοῖς δὲ φυσικοῖς λόγοις). Aristotle thus focused in his investigations too much on physical explanations instead of considering metaphysical causes. According to Proclus, this procedure led to his fallacious views and must be contrasted with Plato’s more adequate, theological approach to physics (In Tim. 1.3.13–4.14 [1.2.29–3.19], 1.302.7–9 [204.8–10], 2.32.6 [227.2–3]).

A few further remarks on (1) and (2) are necessary here. (1) is especially significant for Proclus’ interpretation of Aristotelian metaphysics, as it illuminates that, according to Proclus, the intellect is the highest principle in Aristotle: καὶ ὅ γε Ἀριστοτέλης – τοῦτο [sc. ὁ νοῦς ὁ ἐγκόσμιος] γὰρ ἀπεφήνατο εἶναι τὸ πρῶτον (In Tim. 2.147.4–5 [1.305.20–2]).³⁸ This is again emphasised in a rarely cited passage from the commentary on the Parmenides where Proclus points out Plato’s superiority in positing the One as first principle:

These doctrines are normally propounded by the majority of [Platonic] commentators (ἐξηγητῶν) about the One, and considering it the first principle, they say that it is not body, as the Stoics maintained, nor incorporeal soul, as Anaxagoras claimed, nor unmoved intellect (νοῦν ἀκίνητον), as Aristotle said later; by this, they claim, the philosophy of Plato differs from the others, in that it rises up to the cause above intellect (ὑπὲρ νοῦν αἴτιον ἀναδραμοῦσαν).

(7.1214.6–12)

In the Platonic Theology, he also emphasises Plato’s uniqueness in this regard (1.3) and calls Aristotle’s view a Peripatetikē kainotomia (PT 2.4.31.21–2: Περιπατητικὴ καινοτομία), whereby the latter term negatively means a ‘departure from established (i.e., Platonico-religious) tradition’ or simply a ‘modernism’.³⁹ For Aristotle theology not only stops at the level of intellect but in fact coincides with its study (PT 1.3.12.23–13.5, esp. 13.4–5: εἰς δὲ ταὐτὸν ἄγουσι θεολογίαν δήπου καὶ τὴν περὶ τῆς νοερᾶς οὐσίας ἐξήγησιν). In the eyes of Proclus, Aristotle thus commits a grave mistake since the first principle is supposed to be the One/Good which transcends being and intellect altogether. The seriousness of this objection should not be downplayed, as for instance Baltes does.⁴⁰ Proclus’ interpretation of Aristotle’s highest principle mirrors Plotinus’⁴¹ and Syrianus’ position,⁴² both of which probably influenced him. At the same time his position must be contrasted with Ammonius’ and Simplicius’ view according to which Aristotle recognises the transcendent One.⁴³

Yet, most interestingly, Proclus claims that Aristotle does not simply reject the One but rather transfers some of its characteristics to the intellect. Accordingly, Aristotle’s intellect is similar to Plato’s One insofar as it is (i) non-multiplied, (ii) desired and (iii) does not think about lower beings. Proclus implies that these three characteristics should be attributed correctly to the One and not to the intellect. Rightly understood the intellect is not (i) non-multiplied but possesses multiplicity since its thinking involves at a minimal level a subject that thinks and an object that is thought.⁴⁴ Instead, (i) must be attributed to the One who is absolute unity.⁴⁵ Proclus’ objection to (iii) implies that Aristotle wrongly conceived the intellect’s thinking as exclusively self-centred and unconcerned with essences (or any other characteristics) of other beings. This brings Proclus’ reading close to many modern interpreters such as Ross (1924: I, cxli–iii), Guthrie (1981: 261–2) and Brunschwig (2000).⁴⁶ Additionally, this objection fits well to Proclus’ observation that Aristotle’s intellect has no activity towards other beings (In Parm. 7.1169.4–9). Proclus believes (iii) must be denied of the intellect and instead applied to the One since the intellect has knowledge of lower beings and is concerned with them due to its providential nature. Indeed, as he argues at In Parm. 3.790.12–791.10 and 4. 964.16–25, if intellect has self-knowledge, as Aristotle holds, it knows itself as a cause, which implies knowing of what it is a cause. What about (ii)? Proclus, of course, holds fast to the idea that the intellect is epheton (ἐφετόν) – he even goes so far to say that it is desired by all beings (ET §34.38.3; discussed in Section 4.5.2). However, what he means here is that the intellect should not be seen as the ultimate object of desire like in Aristotle. This place should be reserved to the One or absolute Good, as he clarifies in In Tim. 1.4.1–2 [1.3.6], ET §8.8.31–2 and PT 2.9.59.13–16. Perhaps, this is why Proclus uses here the term with an article, that is, τὸ ἐφετόν (just as in τὸ ἀπλήθυντον and τὸ μηδὲν νοεῖν τῶν δευτέρων): the intellect clearly is an ἐφετόν but not the ἐφετόν.⁴⁷ Ultimately, this downgrading of attributes also makes it difficult to compare metaphysical principles, as the table reveals: Aristotle’s intellect is not equivalent to Plato’s demiurge, since it embodies certain characteristics of Plato’s One. Crucially, this seems to compromise any project of harmonising them from the beginning.

In the last part of (2), there is a puzzling interpretation of Aristotle who apparently claims that παρὰ [τούτων] [i.e., heaven and heavenly bodies] γὰρ εἶναι τὴν δημιουργίαν καὶ τὴν πρόνοιαν.⁴⁸ According to Proclus, Aristotle ascribes demiurgy and providence only to the heaven and not to the intellect as he should have.⁴⁹ The problem is that Aristotle obviously never refers to demiurgy or providence in explaining the nature or activity of the heaven. By demiurgy Proclus means a specific type of efficient causality, namely the one that brings about what is becoming/generated.⁵⁰ Since in Aristotle generation occurs only in the sublunary realm and is, most importantly, dependent on the circular motion of the heaven,⁵¹ Proclus is able to claim that for Aristotle the heaven is ‘demiurgic’, whereas the intellect is not as it does not cause the cosmos’ being. What about providence? While Aristotle himself did not develop a theory of divine providence, Alexander filled this gap with his treatise On Providence.⁵² There, the same view attributed here by Proclus to Aristotle is encountered: providence is exercised by the heaven over the sublunary realm and consists in safeguarding the regular generation and destruction as well as the eternity of species.⁵³ Given Alexander’s significance as a commentator and Proclus’ frequent references to the ‘Peripatetics’, he presumably has here Alexander’s interpretation of Aristotle in mind.⁵⁴

Proclus has made clear so far that Aristotle’s metaphysics departs in crucial points from Plato’s and has thus significant shortcomings. By denying the existence of the One and wrongly attributing some of its characteristics to the intellect, Aristotle fails to make the intellect an efficient cause of the cosmos’ being. Unlike Baltes (1978: II, 69), who claims that this and the previous passage (2.131.11–132.14 [1.294.28–295.19]) ‘stand out in their effort to harmonise the teachings of both philosophers’ (‘zeichnen sich durch das Bemühen um Harmonisierung der Lehren der beiden Philosophen aus’), I see almost no harmonisation effort on Proclus’ behalf in 2.132.15–133.4 [1.295.20–7]. However, Proclus makes clear how Aristotle went wrong and implicitly offers a solution: if one ‘upgrades’ some of the attributes, for example, by attributing non-multiplicity to the One and so on, an agreement can be established.

He makes this explicit at In Parm. 4.973.6–12, where he states that the Peripatetics

declare that there is one thing only which is non-multiplied and unmoved cause as an object of desire (ἀπλήθυντον καὶ ἀκίνητον αἴτιον ὡς ὀρεκτόν); and they attribute (προσάπτοντες) to the intellect what we say of the cause which is situated above the intellect and intelligible number. Insofar as they consider the first principle in this way they were correct, for the beings must not be governed badly nor should multiplicity be the principle of the beings, but the One; but insofar as they postulate that the intellect and the One are the same thing, they are not correct.

Again, we have here the charge of falsely attributing non-multiplicity to the intellect. Interestingly, Proclus also adds an attribution Aristotle got right, namely the intellect as ἀκίνητον αἴτιον ὡς ὀρεκτόν – an expression which Proclus himself uses (e.g., ET §34.38.3). Proclus regards Aristotle’s intellect as incorporating attributes from both Plato’s One and demiurge. Thus, as Aristotle’s metaphysics presents itself, it is not in agreement with Plato. This explains, for instance, why Proclus elsewhere accuses Aristotle of possessing only half of the truth when denying the efficient causality of the intellect (In Parm. 4.842.20–2). Plato’s views therefore form an indispensable corrective lest the student of metaphysics embraces Aristotelian heterodoxy and καινοτομία. For by studying Plato’s metaphysics Aristotle’s intellect can be ‘purified’ of certain inappropriate attributes, such as non-multiplicity and ultimate final causality, in order to reach a correct conception thereof.

4.3.2 Aristotelian versus Platonic Efficient Causes

Besides Aristotle’s confusion of theological principles and their characteristics, Proclus also accuses Aristotle of misunderstanding what efficient causality is. Aristotle pays only lip-service to the efficient cause since he does not conceive it as a productive cause that brings about being. That is, Aristotle might attribute it to the intellect (as he does to nature), but his understanding of efficient causality is fundamentally misguided so that he effectively denies it of the intellect. Thus, Aristotle does not have an efficient cause in the sense Proclus has in mind. This critique occurs at the beginning of his commentary on the Timaeus:⁵⁵

1. (1) For although they [Plato’s successors] may perhaps make mention of the productive cause as well, as when they affirm that nature is the origin of motion, they still deprive it of efficacity and productivity in the strict sense (τὸ δραστήριον καὶ τὸ κυρίως ποιητικόν), since they do not agree that this [cause] embraces the reason-principles (λόγους) of those things that are created through it but allow that many things come about spontaneously too.

2. (2) That is in addition to their failure to agree on the priority of a productive cause to explain all physical things at once (πάντων ἁπλῶς τῶν φυσικῶν ποιητικὴν αἰτίαν ὁμολογεῖν προϋφεστάναι), only those that are bundled around in generation. For they openly deny that there is any productive [cause] of things everlasting (τῶν γε ἀϊδίων οὐδὲν ποιητικὸν εἶναί φασι διαρρήδην). Here they fail to notice that they are either attributing the whole complex of the heavens to spontaneous generation, or claiming that something bodily can be self-productive. (1.3.3–13[1.2.15–29])

Proclus puts forward two criticisms: (1) nature conceived only as origin of motion is not productive in the strict sense; (2) there is no single productive cause of physical reality since eternal physical beings lack such a cause. Proclus turns here Aristotle’s well-known criticism of Plato – namely that Plato was unable to make use of the efficient cause (and the final cause) (Met. 1.6; GC 2.9) – against Aristotle himself.⁵⁶

(1) Regardless of the specific discussion of nature here, it is important to note the underlying assumptions Proclus makes about efficient causality which are quite different from Aristotle’s.⁵⁷ What Proclus says here about nature’s efficient causality, applies a fortiori to higher causes.⁵⁸ Productivity or efficiency here means to be creative and to bring something into existence as well as to cause its being, and not merely to move something, as the choice of words such as δραστήριον, ποιητικόν and ποιουμένων indicates. This usage is primarily influenced by the definition of cause in the Philebus.⁵⁹ Moreover, it also implies transmitting certain properties to a lower being – just as nature is supposed to do via its logoi – as well as preserving and completing the effect.⁶⁰ The term ‘efficient cause’ becomes in this way a richer concept which accounts for a thing’s motion and being. Proclus here appears to be well aware that at least Aristotle’s conception of efficient causality is not the same as Plato’s. He thus departs from a widespread ancient and modern interpretation according to which one can find Aristotle’s causes in Plato.⁶¹

(2) In the second part of the passage, Proclus complains of the lack of a productive cause of physical reality.⁶² This includes a relevant claim to my discussion of the unmoved mover’s causality. Proclus maintains that Aristotle limits efficient causality to generated (ἐν γενέσει) beings, that is, to the sublunary realm (1.3.8–13 [1.2.24–9]). Most significantly, he then accuses Aristotle of denying that eternal beings (τῶν ἀϊδίων), that is, the celestial beings and the cosmos itself, have an efficient cause – a claim repeated later on in the commentary.⁶³ The reason, as becomes clear soon, is that Proclus takes Aristotle’s intellect not as an efficient cause of the cosmos’ being but only as an efficient cause of its motion. But the latter, as has been made clear, is not the type of efficient causality Proclus has in mind.

4.3.3 The Critique of Aristotle’s Intellect in the Commentary on the Timaeus 1

After these preliminary remarks on Aristotle’s metaphysics, specifically on the intellect and the nature of the efficient cause, I now turn to Proclus’ main criticism of Aristotle’s unmoved mover, which has to be read in conjunction with these general objections. In his criticism, Proclus shows that Aristotle’s commitment to both

1. (1) the intellect as cause of the cosmos’ essential desire and

2. (2) the intellect as cause of infinite power

leads to the conclusion that

1. (3) the intellect is the efficient cause of the cosmos’ being.

Aristotle’s mistake lies in not endorsing (3) although it necessarily follows from either (1) or (2).

The passage examined here (In Tim. 2.90.17–93.17 [1.266.21–268.23]) has received less attention in scholarship.⁶⁴ By analysing it in greater detail I bring to light Proclus’ lengthy critical engagement with Aristotle and offer insights into his views of Aristotle’s metaphysics. The text starts with a brief doxography (2.90.17–91.8 [1. 266.21–267.4]) and then offers four objections (2.91.8–93.17 [1.267.4–268.22]). The first two are the philosophically most interesting and discussed in detail in Sections 4.3.3.2 and 4.3.3.3.⁶⁵

4.3.3.1 Doxography (In Tim. 2.90.17–91.8 [1.266.21–267.4])

Let us start with the doxography:

ἀποροῦσι δέ τινες, ὅπως ὁ Πλάτων ἔλαβεν ὡς ὁμολογούμενον τὸ δημιουργὸν εἶναι τοῦ παντὸς εἰς παράδειγμα βλέποντα· μὴ γὰρ εἶναι δημιουργὸν εἰς τὸ κατὰ ταὐτὰ ἔχον ὁρῶντα· πολλοὶ γὰρ καὶ τούτου προεστᾶσι τοῦ λόγου τῶν παλαιῶν· οἱ μὲν γὰρ εἶναι δημιουργὸν Ἐπικούρειοι καὶ πάντη τοῦ παντὸς αἴτιον οὐκ εἶναί φασιν, οἱ δὲ ἀπὸ τῆς Στοᾶς εἶναι μέν, ἀχώριστον δὲ ὑφεστάναι τῆς ὕλης, οἱ δὲ Περιπατητικοὶ χωριστὸν μὲν εἶναί τι, ποιητικὸν δὲ οὐκ εἶναι, ἀλλὰ τελικόν· διὸ καὶ τὰ παραδείγματα ἀνεῖλον καὶ νοῦν ἀπλήθυντον προεστήσαντο τῶν ὅλων. Πλάτων δὲ καὶ οἱ Πυθαγόρειοι τὸν δημιουργὸν ὕμνησαν τοῦ παντὸς ὡς χωριστὸν καὶ ἐξῃρημένον καὶ πάντων ὑποστάτην καὶ πρόνοιαν τῶν ὅλων, καὶ μάλιστά γε εἰκότως·

Some people are perplexed about the way that Plato has taken as agreed that there is a demiurge of the universe who looks to a paradigm. For, they think, no demiurge looking to what remains the same exists. In fact many of the ancients were proponents of this argument. The Epicureans deny that a demiurge exists and state that there is no cause of the universe at all. The [philosophers] from the Stoa say he exists, but that he is inseparable from matter. The Peripatetics state that a separated entity exists, but that it is a final rather than an efficient cause. For this reason they have both destroyed the paradigms and placed a non-multiple intellect at the head of the universe. Plato and the Pythagoreans, however, have celebrated the demiurge of the universe as separate and transcendent and founder of all things and providence of the whole. And this is indeed an eminently reasonable view.

(In Tim. 2.90.17–91.8 [1.266.21–267.4])

This doxographical account – which is in many respects representative of Proclus’ views on the history of philosophy⁶⁶ – presents the different opinions on the nature of god and his causation in an ascending order. Proclus does not focus on the divine in general but rather on the equivalent of the demiurge in the five philosophical schools he considers. Thus, the demiurge of the Timaeus, as creator of the universe (τοῦ παντός), separate (χωριστόν), transcendent (ἐξῃρημένον), founder of all things (πάντων ὑποστάτην) and providential towards the whole (πρόνοιαν τῶν ὅλων), is the benchmark for Proclus.⁶⁷ Specifically the last two characteristics, which emphasise the productive activity of the demiurge towards the cosmos,⁶⁸ are decisive in understanding Proclus’ position throughout this passage.

The survey starts with the Epicureans, who are doctrinally the furthest away from the truth espoused by the Plato and Pythagoreans, with whom the account culminates. Most importantly, the Peripatetics – including Aristotle – are presented as closest to Plato and the Pythagoreans, since they maintain that there is an entity which is separate from the cosmos (unlike the Stoics) and also its cause (like the Stoics but unlike the Epicureans). In contrast to In Tim. 2.132.15–16 [1.295.20–1], Proclus emphasises here the characteristics that Aristotle and the Peripatetics attributed correctly to the intellect, namely χωριστός and αἴτιον. These can be added to other correct attributes like ἀκίνητος and ὀρεκτός. Yet, their metaphysics is still deficient, very much along the lines discussed above (In Tim. 2.132.15–133.4 [1.295.20–7]). For they mistakenly attribute to this separate cause only final and not also efficient causality. For Proclus this has the consequence (διὸ) that they abolish the paradigm and posit a ‘non-multiplied intellect (νοῦν ἀπλήθυντον) in front of the whole’. Proclus claims here that, by denying the efficient causality of the intellect, the Peripatetics deny also the existence of the paradigmatic causes and posit the intellect as the first principle.⁶⁹ The latter, as has been seen, is the most serious error in the eyes of a Neoplatonist.

Here again, like in Section 4.3.1, it emerges that Proclus’ objections to the causality of Aristotle’s intellect are part of a general critique of Aristotle’s metaphysics. It is thus after this introductory doxography (In Tim. 2.90.17–91.8 [1.266.21–267.4]) that Proclus proceeds with his specific criticisms. Proclus’ goal in the first (2.91.8–16 [1.267.4–12]) and the second objection (2.91.17–92.9 [1.267.12–24]) is to show that Aristotle’s reasoning actually commits him to accept that the unmoved mover is a final as well as an efficient cause of being:

1. O1 Insofar as the intellect is a final cause, it is necessarily an efficient cause as well. If the intellect causes the cosmos’ essential desire, it also brings about the cosmos’ being.

2. O2 Insofar as the intellect possesses infinite power and transmits it to the universe, it is necessarily an efficient cause as well. If the intellect causes the cosmos’ eternal motion, it causes the cosmos’ eternal being.

4.3.3.2 First Objection (In Tim. 2.91.8–16 [1.267.4–12])

Let us have a closer look at the first objection.

[O1] εἰ γὰρ ἐρᾷ ὁ κόσμος – ὥς φησι καὶ Ἀριστοτέλης – τοῦ νοῦ καὶ κινεῖται πρὸς αὐτόν, πόθεν ἔχει ταύτην τὴν ἔφεσιν; (i) ἀνάγκη γάρ, ἐπεὶ μή ἐστι τὸ πρῶτον ὁ κόσμος, ἀπ’ αἰτίας ἔχειν τὴν ἔφεσιν ταύτην αὐτὸν τῆς εἰς τὸ ἐρᾶν κινούσης· κινητικὸν γὰρ τὸ ὀρεκτὸν τοῦ ὀρεκτικοῦ φησιν εἶναι καὶ αὐτός. (ii) εἰ δὲ τοῦτο ἀληθές, ὀρεκτικὸν δὲ ὁ κόσμος αὐτῷ τῷ εἶναι καὶ κατὰ φύσιν ἐκείνου, δῆλον ὅτι καὶ τὸ εἶναι αὐτοῦ πᾶν ἐκεῖθεν, ἀφ’ οὗ καὶ τὸ εἶναι ὀρεκτικόν ἐστι.

If the cosmos loves the intellect – as Aristotle says – and it comes into motion in relation to the intellect, where does it obtain this desire from? (i) It is necessary, since the cosmos is not that which is first, that it obtain this desire from a cause which moves it towards love. After all, he himself says that it is the object of desire that moves the desiring subject. (ii) If this is true and the cosmos is desiring of the intellect through its [i.e., the cosmos’] very being and in accordance with its nature, it is clear that its entire being comes from there, including also its being the desiring subject.

(In Tim. 2.91.8–16 [1.267.4–12])

This objection is loosely based on Metaphysics 12.7 and repeats the charge made against the Peripatetics in the doxography that the intellect is not ποιητικόν. In brief, Proclus argues that if the intellect is the final cause (i.e., the object of desire) of the cosmos, as Aristotle maintains, it also needs to be the efficient cause of the cosmos’ being.⁷⁰ The argumentation proceeds in two steps. First (i), Proclus claims that the cosmos is not a first or principle (τὸ πρῶτον) – unlike the intellect – and as such is dependent on a cause (ἀπ’ αἰτίας) for having a certain desire. That is, insofar as the cosmos desires the intellect, the intellect must account for or cause that desire in the first place. Moreover, the intellect as cause of the desire moves the cosmos towards love (τῆς εἰς τὸ ἐρᾶν κινούσης). The reason, so Proclus, is that, according to Aristotle, himself the object of desire (ὀρεκτόν) and the cause of motion (κινητικόν), that is, the final cause and moving cause, coincide – at least in the case of the intellect. Qua object of desire the intellect causes the motion of the cosmos. Proclus’ interpretation matches modern accounts: Judson (2019: 185–6), for instance, claims that the unmoved mover is an efficient cause of the heaven’s desire, which, as proximate cause, brings about the heaven’s motion. So far, so Aristotelian, one could say.

Then, in the second step (ii), Proclus’ argument takes a decisively Neoplatonist turn. He states that if the object of desire is the cause of the desire in the desiring subject, and the desire for the intellect⁷¹ in the cosmos is essential/due to its being (αὐτῷ τῷ εἶναι) and according to its nature (κατὰ φύσιν), then the intellect is not just the cause of the cosmos’ being desiderative but of the cosmos’ being (εἶναι) at all. In other words, if x’s desire of y is essential, and if y is the cause of x’s desire, then y is the cause of x’s being. Insofar as y causes not just a desire in x – as numerous other objects of desire would – but rather a desire inseparably linked to the being of and thus constitutive of x, y is also an efficient cause of x’s being. In turn, x only has an essential desire towards y, if y is the cause of x’s being. In any case, Proclus is here not committed to the blatantly false claim that every object of desire is causally responsible for the being of the desiring subject.⁷²

Two issues which are crucial for the success of the argument arise here and merit further investigation. First, it is not straightforward why Proclus assumes that the cosmos’ desire for the intellect is essential, as Aristotle does not express this explicitly. I take it that Proclus’ assumption is based on the view that eternal motion is a sine qua non for the cosmos’ existence, and in order to maintain it the cosmos has to continually desire the intellect. If the cosmos stops desiring the prime mover, it stops moving. In this way, its desire can be rightly regarded by Proclus as ‘essential’.

Secondly, what does Proclus mean by the term einai (εἶναι) – as in the intellect is the cause of the cosmos’ εἶναι – in this context? Does it denote existence, essence or being (as translated here) – or somehow all three? Although the meaning of this ambiguous term is crucial in understanding this and the following objection, scholarship is silent on this issue. Steel (1987a) in his discussion of this text chooses the translation ‘existence’, as do also Sorabji (1988: 252) and d’Hoine (2016: 390). The problem is that usually the technical term for ‘existence’ is huparxis (ὕπαρξις) in Proclus, as when he discusses the ὕπαρξις τῶν εἰδῶν at In Parm. 4.880.19.⁷³ Steel is indeed aware of this and, thus, when he cites Proclus’ claim that sensibles get their desire ‘from the source of their ὕπαρξις and εἶναι’ (In Parm. 4.842.25), he renders ὕπαρξις as ‘existence’ and εἶναι as ‘being’. I assume that εἶναι does not refer here just to factual existence, whereby the attribution of εἶναι to cosmos simply means that the cosmos exists. Instead, it is a richer notion that includes the mode of existence as well as certain essential attributes, as the expression τὸ εἶναι αὐτοῦ πᾶν and τὸ εἶναι ὀρεκτικόν at 91.15–16 [267.11–12] seem to indicate. Parallel evidence from ET suggests the same (e.g., ET §28.32.29, §31.34.35, §34.36.24).⁷⁴ To put it in contemporary terms, εἶναι here has an existential and predicational dimension: due to the causation of the intellect the cosmos exists and does so in a certain way.⁷⁵ The best translation therefore seems to be ‘being’, as it is able to render the term’s ambiguity also in Proclus.

What do we make of Proclus’ objection here? Proclus might be right in claiming that Aristotle cannot regard the unmoved mover exclusively as a final cause, since causing the desire in the desiring subject can be considered as being an efficient cause. Indeed, as pointed out, this is the interpretation of the intellect endorsed by Judson (1994: 164–5) and (2019: 185–6.) Yet, Proclus goes further than this by concluding that something that causes the cosmos’ essential desire also causes the cosmos’ being. If εἶναι here meant ‘existence’, the move would be warranted insofar as the unmoved mover would be the remote efficient cause of the cosmos’ existence by bringing about its essential desire and, thus, its eternal motion. But for Proclus, εἶναι seems to mean here more than factual existence. Thus the unmoved mover would not just be the reason why the cosmos exists full stop but rather why it exists in a certain way. Even here, however, Aristotle could agree: the desire-induced motion makes the cosmos what it is – namely a complex system of spheres that ultimately influence the sublunary realm.⁷⁶ In this way, the intellect would be the cause of the cosmos being in a certain way.⁷⁷

Yet, while this might be the case in the way the argument is presented here, it becomes clear from other passages that Proclus has a distinctly Neoplatonist conception of the intellect’s causality. In ET §34, he explains that ‘everything proceeds (πρόεισι) from intellect’ (38.3–4), including the cosmos. This procession has to be understood of course by considering Proclus’ understanding of the constitution of being, which is characterised by the triad μονή – πρόοδος – ἐπιστροφή.⁷⁸ According to this, an effect proceeds from its cause, which already contains it in a superior way (ET §7), in order to differentiate itself from it.⁷⁹ While I do not see any evidence that Proclus would believe that Aristotle agrees to this, Proclus’ specific objection in 2.91.8–16 [1.267.4–12] still stands as a line of argument that could be accepted by an Aristotelian.

4.3.3.3 Second Objection (In Tim. 2.91.17–92.9 [1.267.12–24])

From the fact that the intellect causes the cosmos’ essential desire the last objection concluded that it causes the cosmos’ being as well. The second objection reaches the same conclusion, that is, that the intellect is the cause of the cosmos’ being, by starting from the intellect’s causation of the cosmos’ eternal motion. Proclus’ reasoning here is based on the ‘infinite power argument’, where δύναμις is understood as a power to do something not as a potentiality to undergo something. Although we find a brief version of the argument in Syrianus (In Met. 117.25–118.11), Proclus seems to be the first to make extensive use of it by not only summarising the argument itself and its background, in EP⁸⁰ and elsewhere, but also by using it against Aristotle.⁸¹

[2. Objection] πόθεν δὲ τὸ κινεῖσθαι ἐπ’ ἄπειρον πεπερασμένον ὄντα; πᾶν γὰρ σῶμα πεπερασμένην ἔχει δύναμιν, ὥς φησι. πόθεν οὖν τὴν ἄπειρον ἔσχε ταύτην τοῦ εἶναι δύναμιν τὸ πᾶν, εἴπερ μὴ ἐκ ταὐτομάτου κατὰ τὸν Ἐπίκουρον; ὅλως δέ, εἰ τῆς κινήσεως αἴτιος ὁ νοῦς τῆς ἀπείρου καὶ ἀδιακόπου καὶ μιᾶς, ἔστι τι τοῦ ἀιδίου ποιητικόν· εἰ δὲ τοῦτο, τί κωλύει καὶ ἀίδιον εἶναι τὸν κόσμον καὶ ἀπ’ αἰτίας εἶναι πατρικῆς; καὶ γὰρ ὡς τοῦ κινεῖσθαι δύναμιν ἄπειρον ἐκ τοῦ ὀρεκτοῦ λαμβάνει, δι’ ἣν ἐπ’ ἄπειρον κινεῖται, οὕτω καὶ τὴν τοῦ εἶναι δύναμιν ἄπειρον ἐκεῖθεν πάντως λήψεται διὰ τὸν λόγον ὅς φησιν ἐν πεπερασμένῳ σώματι μὴ εἶναί ποτε δύναμιν ἄπειρον.

From where, moreover, does the cosmos, though itself finite, derive its infinite motion? After all, as he [Aristotle] says, every body has a power that is finite. From where, then, does the universe derive this infinite power to exist, if it does not obtain it spontaneously in accordance with [the doctrine of] Epicurus? In general, if the intellect is cause of the infinite and uninterrupted and single motion, there exists an entity which is the efficient cause of that which is everlasting. If this is the case, what prevents the cosmos from being both everlasting and derived from the paternal cause? For just as it obtains from the object of desire an infinite power of motion, through which it moves to infinity, so it will certainly obtain the infinite power of being from there in virtue of the argument which states that there can never be an infinite power in a finite body.

(In Tim. 2.91.17–92.9 [1.267.12–24])

In brief, Proclus again objects to reducing the unmoved mover to a final cause. Instead, it has to be an efficient cause as well, since it must cause the infinite being of the cosmos.

The argument compressed in the first three lines is:

1. (1) A finite magnitude has a finite power.

2. (2) Moving for an infinite period of time requires an infinite power.

3. (3) The cosmos is a finite magnitude and moves for an infinite period of time.

4. (4) Infinite power is either intrinsic (in certain unextended entities) or extrinsic (in magnitudes).

5. (5) Given (3), the cosmos’ infinite power is extrinsic.

In establishing that the cosmos’ eternal motion requires an external infinite power, the question poses itself as to the origin (πόθεν) of this infinite power. Before Proclus considers the two options, he claims that moving for an infinite period of time (τὸ κινεῖσθαι ἐπ’ ἄπειρον) implies being for an infinite period of time (τὴν ἄπειρον … ταύτην τοῦ εἶναι δύναμιν). This implication is absolutely crucial for Proclus, as it transforms the proof from an argument about motion to one about being.⁸² Again, the same ambiguity concerning εἶναι arises. If it means ‘existence’ here, Proclus’ identification of moving for an infinite period of time with existing for an infinite period of time is warranted insofar as the cosmos cannot exist if it does not move continuously. A stand-still means, in fact, the end of the cosmos’ existence. Yet, considering the previous passage (In Tim. 2.91.8–16 [1.267.4–12]) as well as other related texts such as ET §31 and §34, εἶναι seems to have a broader meaning.

The background of the argument is Aristotelian and found in Physics 8.10 and Metaphysics 12.7.1073a5–11 which refers back to the Physics. As discussed in Section 4.2.3, in these passages Aristotle sets out to demonstrate the indivisibility of the prime mover. According to Proclus’ interpretation, Aristotle (1) attributes in these lines infinite power to the unmoved mover, which (2) is transmitted to the cosmos. While some commentators have questioned either (1) or (2), or both, since these claims are not mentioned explicitly by Aristotle, I believe Proclus’ interpretation is correct and a majority of modern scholars, for example, Judson (1994; 2019), Laks (2000) and Touzzo (2011), essentially concur. In short, Aristotle wants to show that the prime mover must be without magnitude, since due to its lack of infinite power a (finite) magnitude is unable to cause an infinite motion. This, however, implies that the prime mover possesses infinite power. For how – on this reasoning – could it otherwise cause an infinite motion? Moreover, the causation of the cosmos’ infinite motion can be considered as a transmission of power, since Aristotle describes how a mover with its power acts on something in order to change it (Phys. 8.10 266a24–30). This description clearly implies also the workings of the prime mover.

Given the accuracy of Proclus’ reading, I claim that his ensuing objection is well-founded: as shown, many modern scholars have struggled to understand how the idea of the unmoved mover transmitting its infinite power to the universe can be squared with Aristotle’s view of the unmoved mover’s presumed mode of operation, that is, as an object of desire. Proclus, I argue, rightly recognises that this argument offers a strong foundation for assuming the intellect’s efficient causality. He is thus right in his objection: Insofar as we take Aristotle on his word and understand the unmoved mover as transmitting power to the universe – and there are, as I argued, strong textual reasons for assuming that –, the unmoved mover cannot be simply a final cause and also not just a moving cause. Instead, the argument requires a metaphysically richer notion of efficient causality – which Proclus and later commentators readily provide.

4.3.3.4 Conclusion

In order to assess Proclus’ approach, I have to consider first how much of the interpretative strategy and arguments are genuinely Proclean. As often with Proclus’ philosophy, including his criticisms of Aristotle, a strong influence by his teacher Syrianus is detectable.⁸³ After all, Proclus himself claims after presenting his objections of Aristotle’s intellect: ‘In relation to Aristotle, then, many refutations have been made by many people’ (In Tim. 2. 93.18–19 [1.268.23]).⁸⁴ As mentioned, Syrianus holds a very similar view of Aristotle’s intellect (e.g., In Met. 10.33–11.5⁸⁵; 175.21–23) and we find evidence for both of Proclus’ objections, O1 and O2.⁸⁶ Syrianus, like Proclus, claims that Aristotle failed to draw explicitly the conclusion from these two arguments that the intellect is an efficient cause of being: ‘to this extent he falls short of his father’s philosophy’ (tr. Dillon and O’Meara; 10.37: τοσοῦτον ἀπολείπεται τῆς πατρίου φιλοσοφίας).⁸⁷ Yet, since this conclusion follows from his own principles (118.27: ἐξ ὧν δίδωσιν), Aristotle is ‘forced to accept the same doctrine whether or not he wants’ (ibid.: εἰς ταὐτὸν ἐκείνῳ δόγμα καὶ ἑκὼν καὶ ἄκων καταναγκάζεται). Thus, based on the necessary implications of his arguments, Syrianus claims that Aristotle in this respect ‘says the same things as Plato in another way’ (27–8: τὰ αὐτὰ τρόπον ἕτερον ἐκείνῳ φθέγγεσθαι). Like Proclus and in contrast to Ammonius and Simplicius, Syrianus believes that, although Aristotle is committed through his own postulates to view the intellect as an efficient cause of being, he fails to take this position himself.⁸⁸ Syrianus states clearly that, once the conclusion has been drawn from Aristotle’s arguments, there is no doctrinal disagreement between Plato and Aristotle on the causality of the intellect and the intelligibles.⁸⁹ Thus, in contrast to Proclus, Syrianus emphasises the resulting agreement between Plato and Aristotle in this respect. At the same time, Syrianus makes clear that this agreement was not Aristotle’s intention. Instead, Aristotle has to be forced (καταναγκάζεται) to accept it.

It is likely that Proclus goes further in his criticism than Syrianus – although this cannot be conclusively determined given our limited access to Syrianus’ works.⁹⁰ At In Tim. 2.91.4–5 [1.266.30–267.1] and 2.132.15–133.4 [1.295.20–7] Proclus clearly presents Aristotle’s metaphysics as deficient for rejecting the One and the paradigm as well as attributing characteristics to the intellect which actually belong to the One. Some of the objections have no correspondent in Syrianus, although he also maintains that Aristotle rejects the One (In Met. 118.21–2). However, in his critique of the causality of Aristotle’s intellect Proclus greatly resembles Syrianus. Similarly to his teacher, Proclus criticises Aristotle by starting from Aristotle’s own premises. Proclus’ view is that by following Aristotle’s own reasoning – especially his infinite power argument – Aristotle should have committed himself to the position that the intellect is an efficient cause of the cosmos’ being. This is the main difference from modern versions of this interpretation. In both (a) and (b), as set out in Sections 4.3.3.2 and 4.3.3.3, Proclus reaches from unquestionably Aristotelian premises – the unmoved mover causes (a) the desire and (b) the eternal motion of the cosmos – the (questionably Aristotelian) conclusion that the unmoved mover is the cause of the cosmos’ being. Aristotle failed to reach this conclusion due to a limited understanding of efficient causality, as seen in the discussion of In Tim. 1.2.21–3.13 [1.2.15–29]. His understanding of the efficient cause primarily as a moving cause effectively denies the type of causality Proclus has in mind for the unmoved mover.

In my opinion, Proclus’ observation that Aristotle’s view of the first principles differs from Plato’s metaphysics just as the Aristotelian type of efficient causality differs from the Platonic one makes his exegesis of Aristotle more nuanced and closer to the original than the interpretations of Ammonius and Simplicius (especially, if one considers their shared Platonist commitments) who attribute a Platonic type of efficient cause to Aristotle’s intellect. By contrast, Proclus shows clearly that Aristotle’s intellect does not share the same characteristics as Plato’s demiurge. Yet, at the same time, he paradoxically contributes to the dissemination of this Platonising-creationist reading of Aristotle’s intellect, since his arguments are picked up by his pupil Ammonius – but with a different intention.

4.4 Ammonius and Simplicius on the Causality of Aristotle’s Unmoved Mover

In the following, I contrast Proclus’ interpretation with Ammonius’ and Simplicius’. Ammonius wrote a treatise on this issue, excerpts of which are preserved – and endorsed – by Simplicius’ commentary on the Physics. Since there is no in-depth analysis of Ammonius’ work, I first offer a reconstruction of its content, in which I also consider evidence from other commentaries by Ammonius’ pupils (4.4.1). Additionally, I set out Simplicius’ reasons for Aristotle’s reticence regarding the intellect’s causality, which, again, possibly mirror Ammonius’. This analysis allows me to situate the treatise within Ammonius’ intellectual climate (4.4.2). As I show, Ammonius’ main motivation for writing it was his desire to refute the interpretations of some Peripatetics, represented by Alexander, and some Neoplatonists, such as Syrianus and Proclus, which prevented the harmonisation of Plato and Aristotle on this issue. Finally, I reach a more general conclusion about the distinct approaches to Aristotle by Proclus and Ammonius/Simplicius (4.4.3).

4.4.1 Ammonius’ Treatise

In regard to their interpretation of Aristotle’s intellect, Syrianus and Proclus remained in opposition to other Neoplatonists. Those associated with the school of Alexandria took a different stance, which was strongly propagated by Ammonius, son of Hermias, in a treatise on this issue whose precise title is unknown.⁹¹ Ammonius himself studied in Athens under Proclus – whom he greatly revered (In DI 1.7–11) – and had a personal connection to the Athenian school, since his father Hermias was a student of Syrianus and his mother Aedesia a relative of Syrianus. After his education in Athens, Ammonius left (around 470/5) for Alexandria where he had a rich teaching activity, especially on Aristotle (Phot. Bibl. §242.341b24: μᾶλλον δὲ τὰ Ἀριστοτέλους ἐξήσκητο), and counted among his pupils Simplicius, Philoponus, Olympiodorus and Asclepius.⁹² While Ammonius’ commitment to Syrianus’ and Proclus’ type of Neoplatonism is debated, he undoubtedly broke with their anti-harmonist stance and (re-)established a more thorough harmony between Plato and Aristotle which is reflected in the writings of his students.⁹³ It is possible that Ammonius achieved this by simply returning to a position prevalent in the Athenian school under Plutarch of Athens until Syrianus became its head in AD 431/2. While I focus in this section mostly on Ammonius and Simplicius, I also refer to the writings of Ammonius’ other students, insofar as they are useful in reconstructing their teacher’s arguments or exegesis of a specific passage.⁹⁴ These philosophers too regard the Aristotelian god as a final cause and an efficient cause of being.⁹⁵

The most extensive evidence for Ammonius’ interpretation is preserved in a well-known passage at the end of Simplicius’ commentary on the Physics (1360.24–1363.24). The text can be divided in five parts. After briefly (i) introducing the problem and the goal of his discussion (1360.24–31), Simplicius (ii) underlines the final and efficient causality of the Platonic demiurge by referring to various passages (1360.31–1361.11). He then (iii) turns to Aristotle and demonstrates the efficient causality of the unmoved mover (1361.11–1362.10). Since this does not suffice, he shows in the next step (iv) that it is an efficient cause of the cosmos’ being (1362.11–1363.8). He (v) concludes with some final remarks on Ammonius’ book and the reasons for Aristotle’s reticence in calling the unmoved mover an efficient cause (1363.8–24). While the whole passage has attracted a certain attention in scholarship,⁹⁶ a close analysis of the procedure and the arguments is still outstanding as is also a discussion of its intellectual context. Such an analysis will help us in comparing the views of Ammonius/Simplicius with Proclus’.

Before I proceed, it is necessary to discuss to what extent this material is directly excerpted from Ammonius’ treatise. Given that much of its content as well as its overarching goal are obscure, much depends on how we understand the following lines:

My teacher Ammonius has written an entire book (βιβλίον ὅλον) that provides many proofs (πολλὰς πίστεις) of the fact that Aristotle considers god to be also the efficient cause (ποιητικὸν αἴτιον) of the entire world (τοῦ παντὸς κόσμου), and I have here taken over (μεταγαγών) some points sufficiently for my present purposes. His more complete instruction on this topic (τελειοτέραν περὶ τούτου διδασκαλίαν) can be found in that book.

(1363.8–12)⁹⁷

Simplicius refers here to the ‘many proofs’ (πολλὰς πίστεις) in favour of the efficient causality of the unmoved mover towards the whole cosmos which Ammonius brought forward in his book. Indeed, in a parallel passage, Simplicius states that Ammonius there demonstrates that ‘Aristotle recognises that the god is not only a final (τελικόν) but also a productive cause (ποιητικὸν αἴτιον) of the cosmos’ (In DC 271.19–21).⁹⁸ Simplicius admits to using some of these πίστεις freely in this passage (ἐγώ τινα μεταγαγὼν ἐνταῦθα τοῖς προκειμένοις ἀρκούντως). This clearly refers to (iii) and (iv) which are mostly interpretations of various passages. Whether Ammonius’ treatise included a short section on Plato’s views on the demiurge, as Simplicius does in (ii), cannot be excluded. At any rate, it seems clear that Ammonius’ book was primarily exegetical and consisted in a wide-ranging collection of passages which were then interpreted to yield a certain result. Such a type of work seems to be the exception in Aristotelian exegesis, as few known treatises on Aristotle from late antiquity deal exclusively with a single interpretative question.⁹⁹ There is, however, a rich tradition among Neoplatonists of writing μονοβίβλια on specific topics, of which Ammonius himself published a few.¹⁰⁰ This emphasises the importance of the problem for Ammonius and his desire to create an agreement in this respect by counteracting dissenting views of certain Peripatetics and Neoplatonists.

Simplicius starts (i) the discussion by addressing other exegetes and stating the goal of his endeavour:

Ἐπεὶ δέ τινες οἴονται τὸν Ἀριστοτέλη τὸ πρώτως κινοῦν, ὅπερ καὶ νοῦν καὶ αἰῶνα καὶ θεὸν ἀνυμνεῖ, τελικὸν μόνον, ἀλλ’ οὐχὶ καὶ ποιητικὸν αἴτιον λέγειν τοῦ κόσμου καὶ μάλιστα τοῦ οὐρανοῦ ὡς ἀιδίου ὄντος καὶ διὰ τοῦτο ἀγενήτου, ἀκούοντες αὐτοῦ πολλάκις λέγοντος, καὶ ὅτι κινεῖ ὡς ἐρώμενον, καὶ πολλάκις ὡς τελικὸν αἴτιον ἀνευφημοῦντος, καλῶς ἔχει κἀν τούτῳ δεῖξαι συμφώνως αὐτὸν τῷ σφετέρῳ καθηγεμόνι μὴ τελικὸν μόνον, ἀλλὰ καὶ ποιητικὸν αἴτιον τὸν θεὸν λέγοντα, τοῦ τε κόσμου παντὸς καὶ τοῦ οὐρανοῦ.

Some think that Aristotle says the prime mover – which he hymns as intellect, eternity and god – is only a final cause and not also an efficient cause of the cosmos and in particular of the heaven, since it is eternal and consequently ungenerated. They think this because they hear him often saying that it causes motion as the object of love, and often proclaiming it as a final cause. It is a good idea, then, to prove that here too he is in agreement with his teacher in calling god not only a final cause but also an efficient cause both of the entire cosmos and of the heaven.

(In Phys. 1360.24–31)

Casting aside for a moment the question of who Simplicius’ addressees are, Simplicius intends – in his typical manner – to demonstrate even in this respect the agreement between Aristotle and his teacher Plato.

In order to do this, he first (ii) sets out Plato’s own position: ‘from what he says in the Timaeus …, Plato clearly calls god the final and efficient cause of the cosmos’ (1360.31–4). Simplicius refers to various passages from this dialogue which he takes to be descriptions of the demiurge’s goodness as well as productive activity.¹⁰¹ He also mentions that the demiurge himself ‘looks at the Good’ (1360.36–7) which I take to mean that the final causality of the demiurge is ultimately dependent on the One/Good.¹⁰² Moreover, he specifies that while the demiurge himself creates (ὑφίστησι) the heavenly gods, that is, the heaven itself and the planets, they in turn create the sublunary realm (1360.37–1361.1). The demiurge’s creation of the whole cosmos is thus mediated through proximate causes, the heavenly gods.¹⁰³ Generally, this preliminary discussion of Plato emphasises Simplicius’ allegiance to Plato, which guides his specific reading of Aristotle.

Turning then to the latter, Simplicius claims that he only needs to ‘defend’ (ἀρκεῖν) the efficient causality of Aristotle’s unmoved mover, since no one ‘disputes’ (ἀμφισβητεῖ) its final causality (1361.11–12). In order to do so Simplicius (iii) establishes that the unmoved mover is an efficient cause – presumably copying here Ammonius’ πίστεις. He does so by listing five passages from four different works where Aristotle supposedly refers to the intellect as an efficient cause. These are the following (1361.12–1362.10):

1. (1) Phys. 2.3.194b29–31. This is the definition of the efficient cause as first origin of motion. It is important for Ammonius, I take it, that the section includes as example the producer of the produced object (τὸ ποιοῦν τοῦ ποιουμένου).

2. (2) DC 1.4.271a33. Ammonius quotes (approximately) the phrase that ‘neither god nor nature do anything in vain’, where ‘god’ is presumably interpreted as the prime mover.¹⁰⁴

3. (3) DC 1.9.279a27–30. Ammonius refers here to Aristotle’s description of the αἰών (eternity/everlastingness), on which the other beings’ existence and life depend (1.9.279a29–30: ὅθεν καὶ τοῖς ἄλλοις ἐξήρτηται … τὸ εἶναί τε καὶ ζῆν). He seems to take here αἰών as a reference to the intellect, which – in his view – is responsible as an efficient cause for the being and life of other entities.¹⁰⁵ Regardless of the identification of αἰών with intellect, it is above all problematic to construe αἰών as subject of ἐξήρτηται, which is rather governed by οὐρανός.¹⁰⁶

4. (4) GC 1.3.318a1–5. Aristotle here sets out two causes responsible for the being of perpetual generation (τοῦ γένεσιν ἀεὶ εἶναι): efficient and material. Regarding the former Aristotle states that he has treated it in his work on motion, that is, Phys. 8, where he discussed the unmoved mover and the ever-moving heaven. Ammonius takes this to mean that Aristotle understands both unmoved mover and heaven as efficient causes: the one of all things, the other only of sublunary beings.¹⁰⁷ Although he does not mention it, Ammonius probably also favours this passage because of its portrayal of the unmoved mover as cause of the being of generation.

5. (5) Met. 1.3.984b15–22. Here, Aristotle lauds Anaxagoras and Hermotimus for having attributed efficient and final causality to the intellect. Ammonius regards this as evidence for Aristotle’s own position.¹⁰⁸

Except (3) and (4), these passages are not conclusive, as they can be understood as referring to the intellect as an efficient cause of motion, that is, a motive/kinetic cause, but not of being, that is, a ‘Platonic’ efficient cause – like Ammonius and Simplicius intend. The former is precisely the way most modern scholars understand the efficient causality of the unmoved mover – insofar as they attribute it to the unmoved mover in the first place.¹⁰⁹

The authors realise this and briefly interrupt their exposition of arguments (iv):

Alexander and some other Peripatetics hold that Aristotle believes in a final and motive cause (τελικὸν αἴτιον καὶ κινητικόν) of the heaven, but not an efficient cause (ποιητικόν) – as indeed the passage of Alexander cited shortly above revealed, which says, ‘[t]he prime mover is the efficient <cause> of the motion of the divine body (τῆς τοῦ θείου σώματος κινήσεως ποιητικόν), which is ungenerated.’

(1362.11–15)

Simplicius clearly states that Alexander conceived the unmoved mover as final cause and efficient cause of motion.¹¹⁰ Since being an efficient cause of motion is not enough to create an agreement with Plato, Ammonius’ remaining discussion serves to show that the intellect is an efficient cause of being (1362.20–1363.8):¹¹¹

1. (6) Phys. 2.6.198a2–13.¹¹² Here Aristotle claims that chance and luck as efficient causes are posterior to intellect and nature: ‘so however much chance may be the cause of the heaven, intellect and nature are necessarily prior causes both of many other things and of this universe (τοῦδε τοῦ παντός)’ (198a11–13; tr. Charlton, modified). Since Ammonius admits that Aristotle’s argument could be purely hypothetical, that is, ‘if someone were to take chance and luck as efficient causes, then etc.’, he follows up – unconnected to the passage discussed – with a general argument: whatever is moved by something else must have its ὑπόστασις from something else ‘if οὐσία is superior to motion’. The idea is that if y receives a lower-order characteristic, such as motion, from cause x, y needs to receive a higher-order characteristic, such as substance, from x as well.

2. (7) Infinite power argument (based on Phys. 8.10.266a10–b27, 267b17–26 and Met. 12.7.1073a5–11). This is a shortened version of the same argument we encountered in Proclus – with the important difference that its result (i.e., the intellect as an efficient cause of the cosmos’ being) is here attributed to Aristotle.

In summary, Ammonius and Simplicius believe that an exegesis of these passages as well as the infinite power argument shows that in Aristotle the unmoved mover causes the being of the cosmos. This causation is not temporal, since the being of the cosmos is eternal (1363.7: τὴν ἀίδιον σωματικὴν οὐσίαν). An assessment of the persuasiveness of the reasons given varies: from the inconclusive (e.g., the apparent figure of speech that ‘god makes nothing in vain’) to convincing arguments (e.g., the infinite power argument).¹¹³ Strikingly, Metaphysics 12 does not occupy a more central role in the discussion, as in modern scholarship, although it could offer evidence for Ammonius’ position.¹¹⁴ At any rate, the interpretation here of Aristotle’s intellect makes it possible for Ammonius and Simplicius to establish an agreement with Plato’s demiurge.

Simplicius’ discussion, however, does not end here. For there remains at least another pressing question: if Aristotle had this view in mind, why was he not more explicit in his writings? Simplicius gives an answer in the final part of his exposition (v) which again possibly derives from Ammonius:

If someone inquires why in the world Aristotle does not say that god is an efficient (ποιητικόν) as evidently (φανερῶς) as <he said that he is> a final cause, I will now again state the account I gave earlier about what is subject to generation (περὶ τοῦ γενητοῦ). For since what works as an efficient cause produces something that is generated (τὸ ποιοῦν γινόμενον ποιεῖ), and what is generated seems to bring with it a temporal origin (χρονικὴν ἀρχὴν) of its generation, this is why he refuses to speak of eternal bodies as coming to be and to identify their cause frequently and evidently as efficient.

(1363.12–18)

Simplicius’ explanation is based on Aristotle’s use of the term ποιοῦν/ποιητικὸν.¹¹⁵ Since in Aristotle the product of a ποιοῦν or ποιητικὸν αἴτιον is something generated (γινόμενον)¹¹⁶ and everything generated has a temporal origin, he – so Simplicius – shuns from using the term γινόμενον for eternal bodies and, likewise, ποιητικὸν for describing their cause. This accounts for the lack of references in Aristotle to the unmoved mover as ποιητικόν. Is Simplicius right in assuming that in Aristotle something generated necessarily has an origin in time? Regarding the latter, Aristotle lists three possible meanings of γενητόν (which I take to be synonymous here with γινόμενον) at DC 1.11.280b14–20: (1) something which is at some time and is not at another; (2) something which is capable of generation; (3) something which is subject to generation, leading it from non-existence to existence. All three strongly suggest a temporal occurrence, making Simplicius’ interpretation very probable.¹¹⁷

Additionally, Simplicius had earlier differentiated between Plato’s and Aristotle’s use of the terms γένεσις and κίνησις (1359.30–40).¹¹⁸ Simplicius holds that Plato’s γένεσις covers a similar semantic range as Aristotle’s κίνησις, insofar as both refer to μεταβολή (change). Hence, Plato’s γινόμενον is conceptually equivalent to Aristotle’s κινούμενον: everything changeable is described as ‘moved’ in Aristotle but ‘generated’ in Plato. However, in Aristotle a γινόμενον covers only a restricted range of κινούμενα, namely those which have a temporal origin. In contrast, Plato applies the term more generally to all changing and moving beings, including all eternal, corporeal beings.¹¹⁹ Aristotle restricts the application of γινόμενον to perishable beings ‘because the imagination easily suggests a temporal origin for things that are said to be generated’ (1359.39–40).¹²⁰ Simplicius further emphasises that in both philosophers the changeable (μεταβαλλόμενον) depends on the unchangeable (ἀμετάβλητον) (1360.17–18): just like in Plato γινόμενα are caused by ἀγένητα, that is, the demiurge, so in Aristotle κινούμενα are brought about by ἀκίνητα, that is, unmoved mover(s).

Thus, according to Simplicius, Aristotle shies away from calling the unmoved mover an efficient cause of the cosmos’ being in order to prevent his readers from attributing a temporal generation to the cosmos – something that Aristotle, like Plato before him, strictly denies according to Simplicius. This, however, does not exclude a non-temporal generation of the cosmos (like in Plato). Thus, both agree that the cosmos’ being is brought about by god. Aristotle himself is aware of this according to Simplicius.¹²¹ Fundamentally, there is only a difference in vocabulary between Plato and Aristotle – who have different linguistic preferences – but not in the matter itself. The disagreement is over words (ὀνόματα), not reality (πράγματα).¹²²

4.4.2 The Context of Ammonius’ and Simplicius’ Discussion

Was Ammonius the originator of the interpretation of the Aristotelian unmoved mover sketched in the previous section and, thus, the reconciliation of Aristotle’s intellect with Plato’s demiurge? There is considerable uncertainty about this issue which – given the significance of this interpretation of the Aristotelian god – is crucial for the history of late antique and medieval philosophy. For instance, Hadot (2015: 28) chides Verrycken (1990) for wrongly regarding Ammonius as originator of this interpretation. However, I find no evidence that Verrycken actually claims this; he merely points out the importance of Ammonius in establishing this view. While it seems unlikely that Ammonius was the first to propose this interpretation, I argue that his crucial – albeit not pioneering – role should still be emphasised. For Ammonius is the first to offer an interpretation of Aristotle’s intellect as a final cause and efficient cause of being in a separate treatise where he closely analyses relevant passages and actively seeks to refute divergent interpretations. As part of that, he makes use of arguments already employed by Syrianus and Proclus – yet with a different intention. In consequence, this interpretation then allowed him to harmonise the Aristotelian intellect with the Platonic demiurge. This harmonisation was meant to counteract Christian objections to the disunity of Pagan philosophy.¹²³ His reading greatly influenced his students and found its way into medieval philosophy via Al-Farabi who refers to the treatise and presupposes its notoriety among his readers.¹²⁴

Yet, there have been suggestions that also earlier authors regarded the Aristotelian unmoved mover as a final cause and an efficient cause of being. For instance, the fourth century philosopher Themistius describes the Aristotelian intellect as ‘craftsman’ (צורף) and ‘creator’ (בורא) – at least in the extant Hebrew source (e.g., In Met. 5.20–1). Both terms, I assume, could stand for the Timaean expressions δημιουργός and ποιητής, suggesting that the Aristotelian intellect is a creative cause of being according to Themistius.¹²⁵

Hadot (2015: 100) speculates that Hierocles of Alexandria, a pupil of Plutarch of Athens, not only had a similar view but also intended – based on this reading – to reconcile Aristotle with Plato. Similarly, Sorabji (2004: III, 37) claims (without a reference) that ‘Hierocles of Alexandria … made Plato and Aristotle agree on God’s causal responsibility for the cosmos’. While this cannot be excluded due to Hierocles’ strong harmonist tendencies in his work On Providence, there is no explicit evidence in our extant testimonies in Photius. Photius only states in his report that ‘[Hierocles] wants to connect the thoughts of these men [sc. Plato and Aristotle] not only in their accounts of providence, but also in all those in which they consider the soul to be immortal and wherever they have philosophised about heaven and earth’ (ap. Phot. Bibl. §214.171b35–38; tr. Schibli, modified). A reference to god is here conspicuously absent, although one could argue that his intent to prove the agreement ‘in the important and most necessary dogmas of Plato and Aristotle’ (172a7–8: ἐν τοῖς ἐπικαίροις τε καὶ ἀναγκαιοτάτοις τῶν δογμάτων Πλάτωνός τε καὶ Ἀριστοτέλους) implies also god’s nature.¹²⁶ Hadot’s – self-admitted – speculation (e.g., 2015: 153) that this view goes back even further to Porphyry and Iamblichus seems baseless without any explicit proof; a simple nod to their general harmonising tendency is insufficient.¹²⁷ I can only find evidence for Porphyry’s view that god is an efficient and final cause¹²⁸ – which, however, does not mean that he regards Aristotle’s god in the same way. Most importantly, we do not find a systematic and argumentative engagement with Aristotle’s intellect and its relationship to the Platonic demiurge like in Ammonius. In this way, Ammonius clearly stands out from previous commentators.

What can be ascertained with some certainty is Ammonius’ motivation for writing the treatise. At the start of the discussion (1360.24–8; see Section 4.4.1), it is claimed that ‘some’ (τινες) exegetes take Aristotle’s prime mover to be only a final cause. Who are these τινες? Alexander and ‘some other Peripatetics’ (οἱ ἄλλοι τινὲς τῶν Περιπατητικῶν) must be among them, as they are mentioned later on (1362.11).¹²⁹ The reason for their rejection of the intellect as efficient cause of the cosmos’ being is apparently based on the cosmos’ eternity and ungeneratedness (ὡς ἀιδίου ὄντος καὶ διὰ τοῦτο ἀγενήτου). That is, because the cosmos is eternal and ungenerated, it cannot have been brought into being by a cause – so the argument of Alexander according to Simplicius.¹³⁰ Another reason for this misunderstanding is Aristotle’s regular insistence on the unmoved mover’s final causality (1360.27–8)¹³¹ and, implicitly, his reticence to state its efficient causality, which has prompted these one-sided and fallacious interpretations.

However, based on my previous discussion, I submit that Proclus must be also among the addressees.¹³² Proclus, like Alexander, maintained that the Aristotelian unmoved mover is only a cause of the cosmos’ motion and not of its being. As pupil of Proclus, Ammonius had a first-hand acquaintance of his master’s views on this intricate issue. Presumably out of his dissatisfaction with Proclus’ interpretation, Ammonius wrote a treatise and adopted more general harmonist views, which departed from Syrianus’ and Proclus’ position on Aristotle. These, then, he transmitted to his pupils.

Simplicius, who held Proclus in great esteem, likewise disagreed with his views on Aristotle and took him to be generally prejudiced against Aristotle (In DC 297.1–5).¹³³ Simplicius’ – implicit or explicit – rebukes of Proclus’ objections to Aristotle are evidence of this stance.¹³⁴ I limit myself to the most prominent examples. (1) Proclus and Simplicius have diverging views on the history of natural philosophy and Aristotle’s place in it, which are found in their respective prologues to the commentaries on the Timaeus (1.2.14–4.14 [1.2.9–3.19]; 1.9.14–10.18 [1.6.21–7.16]) and the Physics (6.31–8.15).¹³⁵ While Proclus emphasises the inferiority of Aristotle’s natural philosophy vis–à–vis Plato’s and criticises him for ignoring the whole array of causes, as well as unduly focusing on matter in his study of nature, Simplicius takes a different view which should be rightly regarded as a response to Proclus’ portrayal of Aristotle.¹³⁶ According to Simplicius, Aristotle stands out even before Plato in investigating all parts of physics. (2) Also, Simplicius refers to Proclus’ refutation of Aristotle’s objections to the Timaeus (In DC 640.21–32) – a work which is noticeably critical in its attitude to Aristotle – before referring again to his own harmonistic views.¹³⁷ Clearly, this adjacent exposition of the harmony-doctrine is meant to contrast with Proclus’ approach to Aristotle. (3) Lastly, Simplicius criticises Proclus in his Corollaries on Place and Time (In Phys. 601.1–645.19; 773.8–800.25) when he departs from Aristotle’s view of these two notions.

4.4.3 Comparison with Proclus

If one were to ask Proclus and Ammonius the question ‘Are Plato and Aristotle in agreement in regard to the causality of the intellect?’ their replies would be obvious. While Ammonius – and by implication Simplicius and his other pupils – seems to clearly think so, Proclus would give a clear negative response: ‘the one [Plato] has posited an efficient cause from which the universe derives its existence [as being] prior to the universe; the other [Aristotle] does not teach an efficient cause for any of the everlasting beings.’ (In Tim. 2.132.10–12 [1.295.15–17]). Yet, by making the right assumptions and positing correct premises Aristotle’s argumentation is still useful according to Proclus. It just requires a Platonist corrective, otherwise the student of Aristotle is led astray and denies the intellect’s productivity.¹³⁸

Generally, the views of Syrianus and Proclus on Aristotle’s metaphysics vary greatly from those of Ammonius and Simplicius. The latter two are for instance able to find the highest Neoplatonist principle, the One, in Aristotle.¹³⁹ Simplicius claims at In DC 485.19–22 (= fr. 49 Rose³): ‘that Aristotle has a conception of something above intellect and substance is clear at the end of the book on prayer where he says clearly that god is intellect or even something which transcends intellect (ἢ νοῦς ἐστιν ἢ καὶ ἐπέκεινά τι τοῦ νοῦ)’ (tr. Mueller).¹⁴⁰ Statements like that put Gerson’s (2005) claim that ‘Neoplatonists generally recognized that Aristotle’s account of the first principle of all was defective’ (10–11, n. 32) into serious doubt. Instead, Aristotle’s and Plato’s principles are perfectly aligned for these philosophers:

Pl.       ἕν       δημιουργός       οὐρανός

Ar.      ἕν       νοῦς                 οὐρανός

This must be contrasted with Syrianus and Proclus who outright deny that Aristotle recognised the One, positing the intellect as Aristotle’s highest principle.¹⁴¹ More specifically, as we have seen, Proclus claims that Aristotle unduly assimilated the intellect to the One. This, however, goes to the heart of Ammonius’ project of harmonising Plato’s demiurge with Aristotle’s intellect: if these principles differ so much and do not serve the same function in their respective philosophical systems – as Proclus claims – they cannot be reconciled with each other.

The causality of Aristotle’s intellect is thus clearly only one issue where Syrianus and Proclus had diverging views from Ammonius and his students.¹⁴² This brings me back to a more general point of my study: unlike Ammonius and Simplicius who wanted to establish the wide-ranging agreement within Greek philosophy with particular focus on its most significant exponents, Plato and Aristotle, Syrianus and Proclus are not guided by this harmonising spirit towards Aristotle.¹⁴³ While Proclus denies that Aristotle regards the intellect as cause of the cosmos’ being or essence, Simplicius asserts exactly this: ‘just as [the cosmos] has its eternal motion from the unmoved cause, so also it receives its eternal corporeal essence (οὐσίαν) from the incorporeal cause’ (1363.7).

Yet, the accounts of Proclus and Ammonius/Simplicius differ not just in the result of their interpretation but also in the way they present the arguments. This has been ignored in scholarship so far, since important discussions such as Verrycken (1990) and d’Hoine (2016: 392) point out only the interpretative differences. Ammonius and Simplicius focus closely on the textual evidence and quote or paraphrase passages which support their interpretation.¹⁴⁴ In contrast, Proclus’ exposition is less text-based and exegetical but much more argumentative.¹⁴⁵ When Proclus discusses the infinite power argument, he does so at greater length than Ammonius/Simplicius. While Ammonius and Simplicius set out their views in the context of Aristotelian exegesis, Proclus discusses Aristotle’s god as part of his Platonic exegesis. Additionally, one of the main reasons for this different approach is of course that Simplicius presents here in summary-form a version of the more detailed investigation of Ammonius’ book (In Phys. 1363.12). It must be assumed that this treatise included much more elaborate interpretations of the passages quoted and also focused more on expounding the arguments for understanding Aristotle’s intellect as an efficient cause of the cosmos’ being. Also, based on the lecture notes of his students, it is very likely that his treatise included more passages than cited here by Simplicius. For instance, Philoponus uses also GC 1.6.323a15 (τὸ κινοῦν ποιεῖν τί) to establish the efficient causality of the unmoved mover (In GC 136.6–137.3).

It must be stressed that, in their project of establishing a harmony between Plato and Aristotle, Ammonius and Simplicius fight a battle on two fronts. On one hand, there is the general accusation for a lack of unity among philosophers made by Christian intellectuals. On the other, there is the threat posed by certain Peripatetics and Platonists, which is more imminent and internal to their discussions.¹⁴⁶ Hence, they not only have to refute Platonists like Syrianus and Proclus who regard some of their doctrines as incompatible but also a Peripatetic like Alexander.¹⁴⁷ Ammonius and Simplicius aim at creating an agreement in Greek philosophy by disagreeing with philosophers like Alexander and Proclus, who rejected such a fundamental harmony in their respective interpretations of Plato and Aristotle. This slightly paradoxical situation makes their project not only stand out but also shows that there was no universal spirit to harmonise Aristotle with Plato in late Neoplatonism, and certainly no unified approach in doing so. It also points towards the different kind of authority that Ammonius and Simplicius were willing to attribute to Plato and Aristotle as compared to Alexander and Proclus.

Lastly, there remains no doubt as to whose interpretation was more successful and influential. By appropriating his teacher’s infinite power argument and by expounding some rather doubtful passages, Ammonius is able to establish an interpretation which is eagerly picked up by philosophers and theologians adhering to the creationist God of the Abrahamic religions. In this way, Ammonius and his students have contributed to the success of Aristotelian theology in the Middle Ages.¹⁴⁸

4.5 Proclus on the Causality of the Demiurge

In Section 4.3, I have delineated the main reasons why Proclus holds that Aristotle wrongly attributed only final causality to the intellect and, thus, fatally diminished the value of his metaphysics. Based on Aristotle’s own premises, he has shown that accepting the intellect’s causation of the cosmos’ desire as well as infinite power in fact amounts to accepting the intellect – or its equivalent, the demiurge – as efficient cause of the cosmos’ being. While these reasons, especially the infinite power argument, are also brought up by Proclus for his views on the demiurge (and other unmoved movers), he has also Platonist reasons for conceiving the demiurge as such a cause. In order to explore these, I now turn to Proclus’ theory of the intellect’s causality in various passages from ET as well as the commentaries on the Timaeus and the Parmenides. In these works, he lays out general, metaphysical grounds as well as exegetical motivations for conceptualising the demiurge’s causality in a specific way.¹⁴⁹ Before I analyse these texts, I would like to briefly consider EP, as this treatise will naturally lead us to Proclus’ own views.

4.5.1 The Unmoved Mover’s Causality in EP

As already extensively discussed, in EP Proclus establishes an unmoved mover as origin of motion by rehashing passages from Physics 8.10:

The prime mover of the circular motion is indivisible.

Let A be the mover of the primary motion. For there must be such a thing since everything in motion is moved by something. If A is the prime mover, it will be unmoved. For the unmoved is prior to the things in motion. And since A causes an eternal motion, it possesses an infinite power to move. For finite powers have also finite activities because the activity depends on the power, so that, if the activity is infinite, so is also the power.¹⁵⁰ It is then necessary that the prime mover of the circular motion is either a body or incorporeal. If it is a body, either finite or infinite. But there is no infinite body (§2.15), and if there were one, it could not move the finite, as has been demonstrated (§2.12). But if the first mover is a finite body, it would not have an infinite power. For finite magnitudes have finite powers, as has been demonstrated (§2.8). Thus, the prime mover of the circular motion is not a body. It is then incorporeal and possesses infinite power, QED.

(§2.21.58.11–27)

Does Proclus endorse here Aristotle’s view that the unmoved mover is an efficient cause of motion, considering that he reaffirms the claims that everything in motion is moved by something (πᾶν τὸ κινούμενον ὑπό τινος κινεῖται) and that the prime mover must possess infinite power to cause the cosmos’ motion, which in Aristotle suggested efficient causality? Although this has been claimed by Opsomer (2009: 198), I do not think this needs to be assumed here, since Proclus has a different understanding of efficient causality, which goes beyond just causing the motion of something, as I have shown above at Section 4.3.2. In fact, he criticises Aristotle in his commentary on the Timaeus for not attributing this type of causality to the unmoved mover. I thus believe that Proclus is simply content in EP to point out that the unmoved mover is the cause of motion in the cosmos through its infinite power.

Proclus says almost nothing about how the unmoved mover causes the cosmos’ motion – which is in line with the reticence of Physics 8. Indeed, besides indivisibility and lack of motion, he barely specifies its characteristics. I propose that another reason for why Proclus does not go further here are his aforementioned critical views of Aristotle on this issue. As I have demonstrated in Chapter 1, Proclus presents in EP Aristotelian doctrines in such a way as to fit a Platonist framework. This in turn implies that contentious issues are either excluded (e.g., self-motion) or only superficially treated (e.g., nature of the heaven, generation of the cosmos). The causality of the unmoved mover belongs to the latter group of issues and like those it is discussed at length elsewhere. EP thus serves a preparatory function appropriate to its place in the curriculum of establishing that the unmoved mover is causally responsible for the eternal motion of the cosmos. Yet, the mode of this causal interaction is discussed in a more advanced work where Platonic doctrine can be considered as well. Precisely such works are ET and his commentary on the Timaeus. These are discussed in the next two sections.

4.5.2 The General Metaphysical Background of ET

In ET, Proclus provides us a metaphysical theory for why efficient and final causality coincide in certain higher beings, among which we must consider also the demiurge. Proclus’ solution to this puzzle is thus grounded in his elaborate metaphysics. This theory is based on his triadic conception of causation as μονή (remaining) – πρόοδος (procession) – ἐπιστροφή (reversion), which I have briefly set out in Section 3.4.4.1. Proclus argues at length in ET §§31–4 that a being reverts to the cause from which it proceeds and likewise proceeds from the cause to which it reverts – so that a being’s ἀρχή and τέλος coincide. The reversion of the effect, moreover, occurs through as many causes as the effect proceeds through. Procession and reversion can thus include a number of intermediate efficient and final causes. Proclus emphasises that while an effect has its being through procession, it gets its well-being only through reversion. It is noteworthy that a similar terminology is also found in Asclepius’ discussion of the causality of the Aristotelian god (e.g., In Met. 28.28–32).

Since an analytical discussion of Proclus’ crucial argument is outstanding, I now offer a close analysis of §31 and §34. In §31, Proclus explains that a being reverts ‘according to its essence’ (κατ’ οὐσίαν) to that from which it proceeds – that is, final cause and efficient cause are identical. As Proclus makes clear, both reversion and procession occur through a number of proximate causes:

Πᾶν τὸ προϊὸν ἀπό τινος κατ’ οὐσίαν ἐπιστρέφεται πρὸς ἐκεῖνο ἀφ’ οὗ πρόεισιν.

εἰ γὰρ προέρχοιτο μέν, μὴ ἐπιστρέφοι δὲ πρὸς τὸ αἴτιον τῆς προόδου ταύτης, οὐκ ἂν ὀρέγοιτο τῆς αἰτίας· πᾶν γὰρ τὸ ὀρεγόμενον ἐπέστραπται πρὸς τὸ ὀρεκτόν. ἀλλὰ μὴν πᾶν τοῦ ἀγαθοῦ ἐφίεται, καὶ ἡ ἐκείνου τεῦξις διὰ τῆς προσεχοῦς αἰτίας ἑκάστοις· ὀρέγεται ἄρα καὶ τῆς ἑαυτῶν αἰτίας ἕκαστα. δι’ οὗ γὰρ τὸ εἶναι ἑκάστῳ, διὰ τούτου καὶ τὸ εὖ· δι’ οὗ δὲ τὸ εὖ, πρὸς τοῦτο ἡ ὄρεξις πρῶτον· πρὸς ὃ δὲ πρῶτον ἡ ὄρεξις, πρὸς τοῦτο ἡ ἐπιστροφή.

All that proceeds from any principle reverts according to its essence upon that from which it proceeds.

For if it should proceed yet not revert upon the cause of this procession, it must be without desire of that cause, since all that has desire is turned towards the object of its desire. But all things desire the Good, and each attains it through the mediation of its own proximate cause: therefore, each has desire of its own cause also. Through that which gives it being it attains its well-being; the source of its well-being is the primary object of its desire; and the primary object of its desire is that upon which it reverts.

(34.28–36.2)

Proclus reaches his conclusion through a reductio ad impossibile. He thus assumes the opposite of what he wants to prove: (T) effects do not revert to their cause and, hence, do not desire their cause (whereby desire implies a turning towards the desired objects). This assumption is absurd, if we consider that (1) all things desire and hence turn towards the Good – here conceived as the metaphysical principle¹⁵¹ – and (2) the Good is only acquired by the effect ‘through the mediation of its proximate cause’ (διὰ τῆς προσεχοῦς αἰτίας). The impossibility is guaranteed by considering (1) and (2) in conjunction. On its own (1) is not sufficient, since one could object, for instance, that soul’s desire of the Good does not imply its desire of one of its proximate causes, such as intellect. And this in turn would mean that (T) is correct. That is why Proclus makes the crucial addition (2): in order for soul to obtain the Good it needs to revert through (and, hence, desire) its proximate causes as well. Just as soul proceeds ultimately from the One/Good through all the proximate causes, so it returns through these preceding causes to the ultimate principle. This means that soul’s desire of the Good implies its desire of its proximate cause, intellect. Proclus concludes (34.34) by stating that the effect reverts to its primary object of desire, that is, the Good.¹⁵²

This result is of course significant for reconciling intellect’s efficient and final causality with the Good’s. Although Proclus often does not make this explicit, the intellect is only a proximate cause in his system. This has significant implications for his view of the Aristotelian intellect and the reason(s) why he cannot accept it as the first cause, as outlined above. However, insofar as all the beings caused by intellect desire the Good, they must also desire intellect. At the end of the proposition (34.34–36.2), Proclus further emphasises that the origin of the effect’s being (τὸ εἶναι) is also the origin of its well-being (τὸ εὖ). Hence, efficient archē (ἀρχή) and final telos (τέλος) coincide in caused beings (i.e., all beings except the One/Good). Through procession an effect acquires its being and through reversion its well-being.

After two intervening propositions,¹⁵³ he proves in §34 the reverse of §31: ‘everything whose nature it is to revert reverts upon that from which it derived the procession of its own substance’, that is, a final cause is also an efficient cause.

Πᾶν τὸ κατὰ φύσιν ἐπιστρεφόμενον πρὸς ἐκεῖνο ποιεῖται τὴν ἐπιστροφήν, ἀφ’ οὗ καὶ τὴν πρόοδον ἔσχε τῆς οἰκείας ὑποστάσεως.

εἰ γὰρ κατὰ φύσιν ἐπιστρέφεται, τὴν κατ’ οὐσίαν ὄρεξιν πρὸς ἐκεῖνο κέκτηται, πρὸς ὃ ἐπιστρέφεται. εἰ δὲ τοῦτο, καὶ τὸ εἶναι αὐτοῦ πᾶν εἰς ἐκεῖνο ἀνήρτηται, πρὸς ὃ τὴν οὐσιώδη ποιεῖται ἐπιστροφήν, καὶ ὅμοιόν ἐστιν ἐκείνῳ κατ’ οὐσίαν· διὸ καὶ συμπαθὲς ἐκείνῳ κατὰ φύσιν, ὡς τῇ οὐσίᾳ συγγενές. εἰ δὲ τοῦτο, ἢ ταὐτόν ἐστι τὸ εἶναι ἀμφοτέρων ἢ ἐκ θατέρου θάτερον ἢ ἄμφω ἐξ ἑνὸς ἄλλου τὸ ὅμοιον ἔλαχεν. ἀλλ’ εἰ μὲν ταὐτὸν τὸ εἶναι ἀμφοτέρων, πῶς κατὰ φύσιν θάτερον πρὸς θάτερον ἐπέστραπται; εἰ δὲ ἐξ ἑνὸς ἄμφω, πρὸς ἐκεῖνο ἂν εἴη τὸ κατὰ φύσιν ἐπιστρέφειν ἀμφοτέροις. λείπεται ἄρα ἐκ θατέρου θάτερον τὸ εἶναι ἔχειν. εἰ δὲ τοῦτο, καὶ ἡ πρόοδος ἀπ’ ἐκείνου, πρὸς ὃ ἡ κατὰ φύσιν ἐπιστροφή.

Everything whose nature it is to revert reverts upon that from which it derived the procession of its own substance.

(1) For if it reverts by nature, it has essential desire of that upon which it reverts. (2) And if so, its being also is wholly dependent on the principle upon which it reverts essentially and (3) in its essence it resembles this latter: hence it is naturally sympathetic with this principle, since it is akin to it in essence. (4) If so, either the being of the two is identical, or one is derived from the other, or else both have received their like character from a single third principle. But if they be identical, how comes it that one is by nature reverted upon the other? And if the two be from one source, that source must be the goal of natural reversion for both. It remains, therefore, that one has its being from the other. (5) And if so, its procession is from that upon which it naturally reverts.

(36.20–38.2)

Proclus makes a number of interconnected claims. He proceeds in a hypothetical manner that differs from his argumentative strategy in §31. The first condition is: (1) if x reverts naturally (κατὰ φύσιν) to y, then x has an essential (κατ’οὐσίαν) desire of y. What does it mean for x to revert according to its nature? Presumably the addition κατὰ φύσιν distinguishes this type of reversion from other types – such as Socrates desiring to eat an apple – by being in some way more fundamental. In this sense, Proclus uses it at ET §7.8.23–4: ‘all things desire the good by nature (κατὰ φύσιν)’. This then allows Proclus to conclude that an effect reverting in this way has a desire κατ’ οὐσίαν of that to which it reverts. It thus seems that reverting naturally implies reverting essentially, since they are inextricably linked to an entity’s being. Based on (1), Proclus then establishes (2): if x has an essential desire of y, x’s being is completely dependent on y (τὸ εἶναι αὐτοῦ πᾶν εἰς ἐκεῖνο ἀνήρτηται¹⁵⁴). That is, x’s being – understood here as factual existence as well as essential features – is inextricably linked with and, possibly, derived from y. Insofar as x’s desire for y is essential and constitutive of x’s being, Proclus can claim that at least a significant part of x’s being is dependent on y. Since, if y were non-existent, x would have no essential desire. This presumably would have the consequence that x does not exist. (3) Granting that x has this relation to y, x must essentially resemble y. Proclus here deduces from the ontological dependency of x on y an essential similarity between them. Establishing this resemblance between x and y is important for Proclus in order to specify their type of relationship, which in §31 has been presented as a causal one. (4) If x and y are like each other, there are three possible reasons for their likeness: either (i) they are identical or (ii) one derives from the other (iii) both derive from a third, higher principle. Proclus excludes (i) and (iii), settling for option (ii): x derives its being from y. This, however, means nothing else than: (5) x proceeds from y, that is, the object of x’s reversion. In this way, ἀρχή and τέλος coincide again, as in §31.

In a corollary to §34, this line of thought is applied specifically to the intellect:

ἐκ δὴ τούτων φανερὸν ὅτι καὶ ὀρεκτὸν πᾶσι νοῦς, καὶ πρόεισι πάντα ἀπὸ νοῦ, καὶ πᾶς ὁ κόσμος ἀπὸ νοῦ τὴν οὐσίαν ἔχει, κἂν ἀΐδιος ᾖ. καὶ οὐ διὰ τοῦτο οὐχὶ πρόεισιν ἀπὸ νοῦ, διότι ἀΐδιος· οὐδὲ γὰρ διὰ τοῦτο οὐκ ἐπέστραπται, διότι ἀεὶ τέτακται· ἀλλὰ καὶ πρόεισιν ἀεὶ καὶ ἀΐδιος κατ’ οὐσίαν, καὶ ἐπέστραπται ἀεὶ καὶ ἄλυτος κατὰ τὴν τάξιν.

From this it is apparent that as the intellect is an object of desire to all things, so all things proceed from the intellect, and the whole world, though eternal, has its essence therefrom. The eternity of the world affords no ground for denying that it proceeds from the intellect; just as it keeps its own station forever, yet is nonetheless reverted upon the intellect. It proceeds eternally, and is eternal in its being; it is eternally reverted, and is steadfast/indissoluble in its own station.

(38.3–38.8)

Proclus describes here intellect as a final and efficient cause of all beings. That does not mean that intellect is the cause of all things tout court but only of those which are. This, of course, excludes the One and the henads which transcend even being¹⁵⁵ and do not desire it.¹⁵⁶ Although he refers to the causation of the cosmos, it is unclear whether νοῦς should be understood here in a generic sense as the hypostasis νοῦς or more specifically as the demiurgic νοῦς. Both are certainly involved in the cosmos’ causation, the latter, however, more directly. Moreover, he makes clear here that procession and reversion are not distinct processes which occur at a specific time or in time at all. Rather, they describe an atemporal causal relationship between the intellect and the cosmos. This secures the eternity of the world (38.5) against possible objections such as ‘if the world is caused by the intellect, it is not eternal’ or ‘if the world is eternal, it is not caused by the intellect’. Both of these were objections common at the time: the former stemming from certain Platonists and Christians, among whom also Philoponus, the latter from Alexander.¹⁵⁷ Clearly, for Proclus the world can be both eternal and caused, insofar as its causation – that is, its procession from intellect – does not refer to a temporal process.

In summary, Proclus incorporates in ET the explanation of the intellect’s causation in his general theory of causality, without having recourse to other philosophical authorities in his explanations. This has the advantage of offering a solution to a specific problem by using universal laws which are purportedly the result of strict deductions in ET. Such arguments can be more persuasive since they are not based on the exegesis of a specific text – as elsewhere in Proclus (or other Neoplatonists, for that matter). Additionally, this discussion in ET underlines why Proclus and likeminded Platonists considered it absolutely crucial to attribute both types of causality to intellect. A failure to do so amounts indeed to a grave misconception of metaphysics as the existence of reality depends to a significant degree on the intellect’s causality. Denying one type of causality means disturbing either the procession from the One or the reversion to this principle. That is, if the intellect is not efficient, the procession of reality stops at the level of intellect, as the last entity to proceed from the One. However, if the intellect is not a final cause, there is no reversion of lower beings to the One.

4.5.3 Platonic Exegesis in the Commentary on the Timaeus

Besides these systematic considerations, Proclus attributes efficient and final causality to the demiurge on exegetical grounds.¹⁵⁸ In this he resembles other Neoplatonists, including Ammonius and Simplicius.¹⁵⁹ For Proclus the evidence for the demiurge’s efficient causality is easy to produce, as Plato often refers in the Timaeus in ‘efficient’ terms to his activity: e. g., ἀπεργάζηται (28a8); δεδημιούργηται (29a7); συνέστησεν (29e1); δρᾶν (30a7); συνετεκταίνετο (30b5). The productive activity is already indicated by the name δημιουργός but also by its other terms such as πατήρ and ποιητής (28c3) as well as συνιστάς (29e1). Yet, in the Timaeus Proclus also finds corroboration for the demiurge’s final causality. The latter might seem surprising, as the very presence of final causes in the Timaeus has been questioned in recent scholarship.¹⁶⁰ Thus, when interpreting Plato’s demiurge we have the reverse situation to Aristotle’s intellect: the demiurge’s efficient causality is obvious, while evidence for his final causality is more difficult to produce. Due to the demiurge’s centrality in cosmology and metaphysics Proclus dedicates long discussions to this metaphysical principle in his commentary on the Timaeus.¹⁶¹ Since these have been the object of extensive studies by Opsomer, to which I refer the reader,¹⁶² I will only focus here on the demiurge’s causality and not, for example, on his identity or place in the metaphysical hierarchy.

Proclus emphasises the need for a single efficient cause of the universe at In Tim. 2.79.4–87.6 [1.258.12–264.3], when he discusses Tim. 28a4–5 (πᾶν δὲ αὖ τὸ γιγνόμενον ὑπ᾽ αἰτίου τινὸς ἐξ ἀνάγκης γίγνεσθαι: παντὶ γὰρ ἀδύνατον χωρὶς αἰτίου γένεσιν σχεῖν).¹⁶³ Proclus regards this as one of five ‘fundamental principle[s]’ (In Tim. 2.85.15 [1.262.29]: ἀξίωμα) in Plato’s cosmology, showing that the whole realm of becoming derives its existence from a cause.¹⁶⁴ For Proclus this cause is efficient, as also evidenced by the preposition ὑπό, and, more specifically, demiurgic, as this is the term Plato uses for efficient causes ‘in relation to becoming’ (2.82.12–13 [1.260.25]). Thus, while all efficient causes bring something about, only demiurgic causes produce that which comes to be insofar as it comes to be. As Opsomer (2017: 144) explains: ‘Hence the Good is the cause of being (cf. Resp. VI 509b6–10), also for the material world, but not its demiurgic cause. For it does not produce the world qua becoming. The same is true for the highest intelligible and the intelligible-intellective deities. They play a causal role, but not a demiurgic one’. Among the different demiurgic causes the universal demiurge, which Plato introduces as the ‘maker and father of the universe’ (Tim. 23c4–5), is the highest cause. The demiurgic cause is responsible not just for producing (In Tim. 2.81.13 [1.260.3]: ποιοῦν) the cosmos but also for maintaining and preserving it (2.81.1–2 [1.259.22]: τὸ σῴζεσθαι καὶ τὸ συνέχεσθαι).¹⁶⁵ Moreover, while the paradigmatic cause brings about the immanent form of the beings, the demiurge is the cause of order (2.96.12–14 [1.270.24–6]: τάξεως γὰρ ὁ δημιουργὸς αἴτιος, εἴδους δὲ ἁπλῶς τὸ παράδειγμα αἴτιον τοῖς μετέχουσιν) by implementing these forms correctly in the universe and preventing a disordered participation in the forms (2.95.20–96.14 [1.270.8–26]).¹⁶⁶

What about the demiurge’s final causality? For Proclus the ultimate final cause of all reality, including the cosmos, is the transcendent One/Good (e.g., In Tim. 1.4.1–2 [1.3.6], 2.102.14–16 [1.274.28–30], 2.220.14–15 [1.356.13–15]) for whose sake the cosmos has been produced by the demiurge (2.221.8–11 [1.356.31–357.2]).¹⁶⁷ Proclus finds an allusion to this principle at least twice in Tim., since he takes the ‘reason why’ (29d7: δι’ ἥντινα αἰτίαν) the creator made the universe and the ‘most important principle/reason’ (29e4: ἀρχὴν κυριωτάτην) for the cosmos’ generation as alluding to the Good.¹⁶⁸

In this context Proclus’ interpretation of Tim. 29e1–2 (‘he was good, and one who is good can never become jealous of anything’) at In Tim. 2.225.3–229.5 [1.359.20–362.16] is crucial, since he discusses there the relationship between the One/Good and demiurge. In regarding this passage as evidence for the demiurge’s final causality, Proclus is in good company: Syrianus (In Met. 82.9–11), Ammonius (ap. Simpl. In Phys. 1360.31–3), Simplicius (In Phys. 464.3–6) and Asclepius (In Met. 21.20–1) also mention it. In his interpretation, Proclus states that the ‘final cause is this: goodness, both absolute goodness and demiurgic goodness’ (2.226.9–10 [1.360.16–17]) and he continues ‘one goodness is absolute (ἁπλῶς) and the other is that in the demiurgic intellect, and the former is the source of all goods, intelligible and intellective, hypercosmic and encosmic, and the latter, being a particular good, is the cause and source of some things, but has been allotted to a lower order than others’ (2.226.15–19 [1.360.22–6]).¹⁶⁹ Thus, for Proclus the demiurge is a final cause insofar as it participates in the One. Through their goodness both the One and the demiurge are final causes; however, the latter clearly only insofar as it derives its goodness from the One.¹⁷⁰

The problem is that being good does not necessarily imply being a final cause. Only if something causes qua good simpliciter (ἁπλῶς) and not accidentally (κατὰ συμβεβηκός) can it be considered a final cause – at least according to Aristotle in Met. 1.7.988b6–16.¹⁷¹ Proclus seems to accept this condition as well, as reversion to a final cause requires first of all a desire for the cause. While the latter aspect is not found explicitly in Tim., Proclus simply seems to assume it, since, as we have seen before, he regards the intellect as an object of desire to all beings.¹⁷² This characterisation satisfies the condition of being a final cause, since the intellect’s goodness is the reason for its (almost) universal desirability – just as in the case of the absolute Good.¹⁷³ However, the exegetical background for assuming this type of causality remains weak and rather unpersuasive.

It is possible to assume that Proclus’ main influence in this respect was Aristotle, since he himself sometimes emphasises the Aristotelian heritage: ‘therefore the intellect is both an object of love (ἐραστόν) and of desire (ὀρεκτόν), as Aristotle says’ (In Alc. 317.22–318.1; tr. mine).¹⁷⁴ Still, we have to be cautious here, as already the Middle Platonist Alcinous claimed that the intellect moves the cosmos as an object of desire (Didask. 10.164.24–31). There, it is clearly an Aristotelian borrowing. Yet, by the time of Proclus this view was already so dominant that the genuinely Aristotelian import was probably no longer visible to the Neoplatonists. Instead, Proclus saw Aristotle simply as rehashing Platonic ideas and even partly doing this incorrectly. It generally seems that when Proclus cites Aristotle explicitly, he intends to reveal Aristotle’s Platonic heritage and not to introduce a foreign doctrine into his Platonist system. Nevertheless, Proclus shows a remarkable independence in dealing with Aristotle’s arguments, as I have shown.

4.6 Conclusion

My reconstruction of one of the central debates on Aristotle’s metaphysics in late antiquity revealed Proclus’ and Ammonius’ views on Aristotle’s intellect and its relation to Plato. I have shown that the two Neoplatonists offer contrasting views of Aristotle’s metaphysics as well as divergent strategies of approaching it. For Proclus, Aristotle’s arguments force him to accept the efficient causality of the prime mover; yet Aristotle himself fails to acknowledge this. In contrast, Ammonius believes that Aristotle was actually committed to these arguments as well as their result. Crucially, Proclus and Ammonius share the infinite power argument but disagree on whether Aristotle himself drew the conclusion that the prime mover is the efficient cause of the cosmos’ being. As I have emphasised, Proclus’ interpretation is part of his more general conception of Aristotle’s metaphysics, which he regards as flawed primarily due to Aristotle’s elimination of the Platonic One and the ensuing misalignment of metaphysical principles. Whereas for Proclus Plato is an indispensable corrective to Aristotle, Ammonius (and, more clearly, Simplicius) does not share this view but rather regards Aristotle’s metaphysics as essentially in agreement with Plato. Most significantly, this serves as further evidence that Proclus – unlike Ammonius and his pupils – is not committed to the harmony-doctrine. This makes Proclus’ approach, as I have argued, more sensible and, indeed, closer to our modern understanding of Aristotle, since Aristotle’s metaphysical system differs significantly from the Neoplatonist view of Plato’s metaphysics. Methodologically, there is also a divergence between Proclus and Ammonius: the former is more argumentative, while the latter focuses more extensively on the actual text and its exegesis. In part these differences can be accounted for by the context – Platonic in Proclus and Aristotelian in Ammonius. But they also demonstrate different exegetical strategies.

Additionally, this division has an important historical dimension, as it presents us a dynamic intellectual environment with a variety of individual approaches. To show this I emphasised how Ammonius responded in his μονοβίβλιον partly to Proclus’ interpretation and then went on to influence his pupils, Simplicius, Asclepius et al. and, ultimately, certain medieval philosophers. In producing a monograph on this issue, Ammonius played a crucial role in the interpretation of Aristotle’s prime mover. Based on the scant evidence, Proclus’ reading seems to be heavily inspired by Syrianus – just as his overall critical approach to Aristotle. These philosophers interact with each other’s interpretations and demonstrate a heightened awareness for subtle differences in their readings.

While Proclus was certainly interested in Aristotle’s views on the causality of the intellect, he also goes at length to set out his own reasoning behind adopting the final and efficient causality of the intellect. As I have shown, Proclus’ arguments are philosophical as well as exegetical, whereby both aspects are interrelated and sometimes indistinguishable. In ET, he tries to remain faithful to the treatise’s axiomatic character, which is presented as unaffected by authoritative views, by deducing the intellect’s type of causality from general metaphysical presuppositions such as the triadic structure of reality as μονή – προόδος – ἐπιστροφή. Given his premises, Proclus’ solution appears quite economical and compelling. In his commentary on the Timaeus he provides textual reasons for his position: while he presents convincing evidence for regarding the demiurge as an efficient cause, the reasons for its final causality are less persuasive. In this way, his theoretical reflection on the intellect’s causality is more successful and offers a stronger argumentative foundation than his exegesis of Plato.