Hostname: page-component-745bb68f8f-b6zl4 Total loading time: 0 Render date: 2025-02-11T15:41:14.923Z Has data issue: false hasContentIssue false
  • English
  • Français

THEMISTIUS AGAINST PORPHYRY (?) ON ‘WHY WE DO NOT REMEMBER’

Published online by Cambridge University Press:  16 April 2021

Robert Roreitner
Robert Roreitner*
Affiliation:
Charles University in Prague
*
robert.roreitner@ff.cuni.cz
Rights & Permissions [Opens in a new window]

Abstract

This article sheds new light on Themistius’ argument in what is philosophically the most original (and historically the most influential) section of his extant work, namely On Aristotle's On the Soul 100.16–109.3: here, Themistius offers a systematic interpretation of Aristotle's ‘agent’ intellect and its ‘potential’ and ‘passive’ counterparts. A solution to two textual difficulties at 101.36–102.2 is proposed, supported by the Arabic translation. This allows us to see that Themistius engages at length with a Platonizing reading of the enigmatic final lines of De anima III.5, where Aristotle explains ‘why we do not remember’ (without specifying when and what). This Platonizing reading (probably inspired by Aristotle's early dialogue Eudemus) can be safely identified with the one developed in a fragmentary text extant only in Arabic under the title Porphyry's treatise On the soul. While Themistius rejects this reading, he turns out to be heavily influenced by the author's interpretation of the ‘agent’, ‘potential’ and ‘passive’ intellect. These findings offer us a new glimpse into Themistius’ philosophical programme: he is searching for an alternative to both the austere (and, by Themistius’ lights, distorted) Aristotelianism of Alexander of Aphrodisias and the all too Platonizing reading of Aristotle adopted by thinkers such as Porphyry.

Keywords

ThemistiusAristotelianismPlatonismDe animaintellectimmortality
Type
Research Article
Information
The Classical Quarterly , Volume 71 , Issue 1 , May 2021 , pp. 379 - 385
Check for updates
Copyright
Copyright © The Author(s), 2021. Published by Cambridge University Press on behalf of The Classical Association

On Aristotle's On the Soul (In De anima) 100.16–109.3 is unique in the extant philosophical work of Themistius. The author of what often read like literal paraphrases of Aristotle's words goes out of his way here in dealing with the final sentence of Aristotle's De an. III.5 (430a23–5). This section is longer than Themistius’ preceding discussion of the rest of De an. III.4–5 and longer than his subsequent discussion of De an. III.6–8 as a whole.Footnote 1 What attracts so much of Themistius’ attention is Aristotle's claim that ‘we do not remember because this [sc. the agent νοῦς] is impassive, while the passive νοῦς is perishable’ (οὐ μνημονεύομεν δέ, ὅτι τοῦτο μὲν ἀπαθές, ὁ δὲ παθητικὸς νοῦς φθαρτός). This claim is notoriously ambiguous. It is not clear whether Aristotle is asking about (i) the reason why it can happen during our lives that we are unable to remember what we once knew, or (ii) the reason why we cannot remember things from the time before we were born,Footnote 2 or (iii) the reason why we will not be able to remember things from this life after death. Themistius insists that (iii) is the correct understanding, that, by implication, in Aristotle's view ‘we’ will exist after our death, and thus that his claim according to which the passive νοῦς is perishable cannot concern our intellect proper (which Aristotle thinks is immortal) but only some lower capacity of the soul.

The reasons why Themistius’ discussion relating to De an. III.5 430a23–5 is so long and elaborate seem to be that (a) he is well aware that his interpretation is highly controversial, and (b) he sees a great danger in this passage for his overall understanding of Aristotle: it turns out to be the only passage in De anima which seems, at least prima facie, to claim explicitly that our νοῦς is perishable, so that no personal immortality is possible—just as Alexander of Aphrodisias argued before Themistius.Footnote 3 Themistius is determined to do everything he can in order to disqualify Alexander's interpretation of these lines. From this perspective we can well understand why Themistius attacks Alexander's famous reading of De an. III.5 (identifying the immortal agent νοῦς with the first god whose immortality has nothing to do with us) at 102.36–103.19, and spells out in detail his sophisticated alternative at 103.20–105.12.

All this is easily understood by an attentive reader. What is less clear is that at 100.16–109.3 Themistius is entering into discussion with another important figure who is no less crucial for the structure of his overall argument than Alexander. This has not been properly appreciated and understood largely owing to an unresolved textual issue at 101.36–102.2.

FIXING THE TEXT AT 101.36–102.2

After setting out his arguments for why we should understand the question of De an. III.5 430a23–5 as being about post-mortem memory at 100.37–101.36 (the main support being a connection established between III.5 430a23–5 and I.4 408b18–30), Themistius turns to a view of other interpreters (101.36–7):

ὥστε παρηνέχθησαν ἅπαντες τῶν δοκούντων ἐγκαλεῖν τῷ φιλοσόφῳ, ὅσοι καὶ ἀπορεῖν αὐτὸν καὶ διαλύειν ᾠήθησαν.

Thus all those who seem to criticize the philosopher were led astray in believing that he was both raising and solving difficulties.

One can easily understand Richard Heinze's desperate comment about this sentence: non intellego.Footnote 4 Heinze thought that a satisfying solution could only be reached by complete rewriting.Footnote 5 After Heinze, it was Paul Moraux who suggested a more elegant solution consisting in substituting the first καί at 101.37 with κακῶς.Footnote 6 The censured interpreters would thus be qualified as those who

κακῶς ἀπορεῖν αὐτὸν καὶ διαλύειν ᾠήθησαν.

believed that he [sc. Aristotle] was raising and solving difficulties in a wrong way.

This solution was followed by Robert B. Todd in both of his translations of Themistius.Footnote 7

But even if the problem with 101.36–7 can be solved in this way, another difficulty awaits us in the sentence that immediately follows, opening a new paragraph and a new page in Heinze's edition (we are about to see that this apparently insignificant circumstance is likely to have played some role in the story). Here Themistius seems to be turning to a new question (102.1–2, transl. R.B. Todd):

διὰ τί ποτε οὖν οὐ μεμνήμεθα ὧν ὁ ποιητικὸς νοῦς ἐνεργεῖ καθ' ἑαυτὸν καὶ πρὶν εἰς τὴν σύστασιν συντελέσαι τὴν ἡμετέραν;

Why then do we not remember the objects of the productive intellect's activity on its own, i.e. before it contributed to our constitution?

This question expresses exactly one of the three understandings of Aristotle's question at III.5 430a23–5 distinguished above, namely (ii)—an understanding different from the one Themistius has just defended, namely (iii). Why is he now raising this question? Neither Moraux nor Todd has a good answer. And this worry becomes still more urgent as we read on. At 102.2–17 and 102.18–24 Themistius rehearses his arguments (based on the allegedly parallel passage from De an. I.4) for interpreting the question of De an. III.5 430a23–5 as concerning post-mortem memory and not our capacity of remembering, during this life, the objects thought by the agent νοῦς before we were born, that is, as concerning (iii) and not (ii). And at 102.17–18 he points out that this latter question cannot be Aristotle's question because it does not make good sense at all. So, how should we understand the text of 102.1–2, where Themistius seems to be raising this question for himself?

The impression that there is something wrong with these lines is further strengthened by what immediately follows (102.2–3, transl. R.B. Todd):

φθειρομένου γάρ φησι τοῦ κοινοῦ οὐχ οἷός τέ ἐστιν ὁ ποιητικὸς οὔτε διανοεῖσθαι οὔτε μνημονεύειν.

For he says that when the common [intellect] perishes the productive [intellect] can neither think discursively nor remember.

What is the role of γάρ here? Is Themistius explaining his question about our memory of the objects thought by the agent νοῦς before we were born (or providing an answer to it) by insisting again that Aristotle's question at the end of De an. III.5 should be read in a different way, as concerning post-mortem memory?

I preface my formulation of the solution to these difficulties by highlighting the importance of the extant Arabic translation of Themistius’ In De anima for our understanding of the text. This translation (available in Lyons's excellent edition)Footnote 8 was produced in the ninth (or early tenth) century and thus constitutes an indirect witness of a much older manuscript tradition than the most ancient extant Greek manuscript used by Heinze (Parisiensis Coislinianus 386), which dates from the eleventh century. Dozens of potentially interesting variants and implications for the Greek text were helpfully underlined and discussed by Gerald M. Browne.Footnote 9 The present case shows that his discussion was not exhaustive.

The Arabic translation of 101.36–102.2 runs as follows:

fa-yarā an yakūna qad ḥāda ʿammā kāna yarā-hū l-faylasūfu ǧamīʿu man yaẓunnu bi-hī anna-hū yaʿtibu ʿalayhī fī ẓanni-him anna šakka-hū wa-ḥalla-hū šakkun fī l-sababi llaḏī la-hū ṣirnā lā naḏkuru mā yafʿalu-hū l-ʿaqlu l-faʿʿāl ʿalā infirādi-hī qablu an yaṣīra ilā taqwīmi ḏāti-nā.

We can leave aside the convoluted and not quite transparent beginning of this Arabic sentence. What the part in bold shows clearly is that the translator understood Themistius’ characterization of the critics of Aristotle in question in the following way: they believed that Aristotle was raising and solving difficulties concerning the cause of why we do not remember the objects thought by the agent νοῦς before we were born. To understand the text in this way, the translator had to take the clause διὰ τί ποτε οὖν οὐ μεμνήμεθα at 102.1 not as starting a new sentence (of a new paragraph) but as being subordinated to ἀπορεῖν καὶ διαλύειν from the preceding clause at 101.37. And there is one more thing we learn from the Arabic translation: the translator clearly read a text in which there was no οὖν after ποτε at 102.1.

So, the text of 101.36–102.2 he was translating must have run like this:

ὥστε παρηνέχθησαν ἅπαντες τῶν δοκούντων ἐγκαλεῖν τῷ φιλοσόφῳ, ὅσοι καὶ ἀπορεῖν αὐτὸν καὶ διαλύειν ᾠήθησαν διὰ τί ποτε οὐ μεμνήμεθα ὧν ὁ ποιητικὸς νοῦς ἐνεργεῖ καθ᾽ ἑαυτὸν καὶ πρὶν εἰς τὴν σύστασιν συντελέσαι τὴν ἡμετέραν.

Thus, all those who seem to criticize the philosopher were led astray in believing that he was raising and solving the difficulty as to why we do not remember the objects of the productive intellect's activity on its own, that is, before it contributed to our constitution.

On the basis of this evidence, I venture to formulate the following hypothesis about how the manuscript tradition arrived at the version printed by Heinze. A simple dittography of οὐ at 102.1 was enough to start the process whose next logical step was a correction of the first οὐ into οὖν which implied the separation of 102.1–2 as a self-standing question. The historical accident consisting in the fact that the two alleged sentences were separated by a page break in Heinze's edition might help to explain why the error has escaped modern philology for so long.

Actually, when I said that it has escaped modern philology, I was not being completely fair to Omer Ballériaux, who suggested already in his 1941 dissertationFootnote 10 that 101.36–102.2 should be read as a single sentence. He repeated the point in two later articles,Footnote 11 complaining that Todd's translation did not follow his suggestion. We can understand, however, why Ballériaux's point was not taken on board, although he was in fact closer to the truth than were his contemporaries. He never even mentions that there is a problem with οὖν at 102.1. But as long as this οὖν remains in the text it is difficult to see how 101.36–102.2 could really be read as a single sentence. Only after bracketing (with the support of the Arabic translation) the redundant particle as a corruption do we arrive at a smooth text. Two textual problems are solved at a stroke.

THEMISTIUS BETWEEN PORPHYRY (?) AND ALEXANDER

This having been done, let us return to the philosophical issue about Themistius’ opponent at 101.36–102.24. It is someone who read De an. III.5 430a23–5 as asking and explaining why we do not remember during this life the objects thought by the agent νοῦς before we were born. This does not seem to be Alexander, not least because we would not expect Alexander to criticize Aristotle in this way.Footnote 12 Wilhelm Kutsch has suggested that the reading under consideration may derive from a treatise whose fragment is extant in an Arabic translation under the title Porphyry's treatise On the soul (Maqālat li-Furfūrīūs fī n-nafsi).Footnote 13 This identification is very likely since (a) the interpretation of De an. III.5 430a23–5 targeted by Themistius at In De an. 101.36–102.24 is indeed proposed by the Arabic text as a central point to which it returns again and again, and (b) there is no other extant ancient text or even any mention of a text which would defend this reading.

In fact, there is much more to be said about the connection of Themistius’ discussion in In De an. 100.16–109.3 and this Arabic fragment than scholars have so far noted. It is not just that this is the only known text developing the interpretation of De an. III.5 430a23–5 targeted by Themistius at 101.36–102.24. There are several striking points of agreement between it and Themistius’ discussion throughout 100.16–109.3.

  1. (a) Porphyry (?) distinguishes (268.10–11 Kutsch)Footnote 14 between the imperishable ‘material’ (hayūlānī) intellect and a perishable ‘passive’ (munfaʿil) intellect (cf. In De an. 105.13–108.34), which he identifies with wahm (φαντασία); without it—he interprets Aristotle to be saying at 430a25—the material intellect cannot think anything, at least while inhabiting the body (cf. In De an. 101.27–30, 102.23–4).Footnote 15

  2. (b) Porphyry (?) insists (268.12–13 Kutsch) that each of ‘us’ (naḥnu) is an agent νοῦς claimed to be separable and immortal by Aristotle at 430a22–3, while also being a ‘psychic’ or a ‘material’ νοῦς more or less unified with the agent νοῦς (cf. In De an. 100.16–37).

  3. (c) Porphyry (?) takes (268.6, 268.9 Kutsch) the νοῦς, analysable into a material aspect and an agent aspect, to be the form of the soul (cf. In De an. 100.28–37).

  4. (d) Porphyry (?) uses (268.15 Kutsch) the (Plotinian) metaphorFootnote 16 of the air and the light contained in it to describe the composed character of our νοῦς, while insisting that the comparison is imperfect (cf. In De an. 103.20–105.12).

  5. (e) Porphyry (?) talks (268.18, 269.5 Kutsch) about our νοῦς ‘mixing’ with the material world or with ‘our traces’ (cf. Them. 108.29–31).

What emerges from this comparison is a picture of Themistius being heavily inspired throughout 100.16–109.3 by Porphyry(?)'s exegesis of De an. III.5, while rejecting a central tenet of his interpretation.Footnote 17

Themistius’ approach to Porphyry (?) here is typical. It is typical of Themistius to take the Platonist reception of Aristotle very seriously, to be heavily inspired by Neoplatonist readings of Aristotle, and to accept the challenges raised by the Platonists against Aristotle—while at the same time attempting to defend an Aristotelian perspective against Platonism. In this way Themistius is developing an alternative Aristotelian approach to various issues: he is often attempting to beat a middle path between Platonism and the Aristotelianism of Alexander of Aphrodisias.Footnote 18

To get a more concrete idea about what is at stake at In De an. 100.16–109.3 and what it can mean here to offer an alternative Aristotelianism, contrasting with both the Neoplatonist approach to Aristotle and Alexander's interpretation, I add one final consideration. Porphyry (?) is very likely to have read the question of De an. III.5 430a23–5 against the backdrop of Aristotle's early dialogue on the soul entitled Eudemus (now lost). Proclus, In Remp. II 349.13–26 seems to imply that in this dialogue Aristotle was raising the question of why the soul, when incarnated, forgets what it saw before but does not forget what it saw during this life when it departs the body.Footnote 19 Once this early text is taken as the key to De an. III.5 430a23–5, Porphyry(?)'s reading, that is, (ii), indeed, suggests itself as the most natural one.

Now one can imagine how someone like Alexander would react to this Platonizing reading of Aristotle. No matter what Aristotle wrote in his youth under the direct influence of his teacher, the only thing which counts is his mature view formulated in De anima; and in the context of De anima itself Aristotle cannot meaningfully raise the question of why we do not remember what we saw before being born for the simple reason that, according to this treatise, there was nothing like us before we were born.

Themistius seems to share this kind of misgivings and he is ready to defend a reading of De anima which shows Aristotle as decisively departing from his earlier views presented in the Eudemus.Footnote 20 Like Alexander, Themistius is interested in Aristotle's mature position which seems to be genuinely novel with respect to Plato. But Themistius is not ready to go as far as Alexander here. His Aristotle is not as Platonized as Porphyry(?)'s Aristotle, but he still shares much more common ground with Plato and later Platonists than Alexander's Aristotle does.

One of these shared features, which Themistius is eager to emphasize against Alexander, is that of personal immortality.Footnote 21 It is exactly this part of Themistius’ conciliatory moderate Aristotelianism that exerted, for better or for worse, the strongest influence on the subsequent Aristotelian tradition in both the Arabic and the Latin worlds.Footnote 22

Footnotes

Many thanks to the participants of the workshop ‘Theophrastus on νοῦς’ at HU Berlin in June 2019, where the main ideas of this paper were presented and discussed for the first time. I would like to thank, above all, Stephen Menn, who organized this workshop and served, so to speak, as the midwife of this paper. Thanks also to Nicholas Aubin, who helped me with understanding the intricate structure of the Arabic translation of Themistius at In De an. 101.36–102.2, as well as to the anonymous referee of CQ for valuable comments. This work has been supported by Charles University Research Centre programme No. 204053.

References

1 There are other passages where Themistius is offering an original and insightful discussion of Aristotle's thought (see below, n. 18), but none comes even close to the intricacy and audacity of In De an. 100.16–109.3.

2 This can be further fleshed out either in a Platonist or in an anti-Platonist way, that is, either as asking why we first do not remember and need to recollect or as asking why recollection is impossible. The former will play an important role below. For the latter, see Menn, S., ‘From De Anima III 4 to De Anima III 5’, in Guyomarc'h, G., Louguet, C. and Murgier, C. (edd.), Aristote et l’âme humaine: Lectures de ‘De anima’ III offertes à Michel Crubellier (Leuven, 2019), 95–155, at 140–1Google Scholar.

3 Alexander of Aphrodisias, De an. 90.14–16; cf. 90.23–91.5.

4 Heinze, R., Themistii In libros Aristotelis de anima paraphrasis (Berlin, 1899)Google Scholar, ad loc.

5 He seems to have the following text in mind: ὥστε παρηνέχθησαν ἅπαντες τῶν δοκούντων τῷ φιλοσόφῳ, ὅσοι μὴ περὶ τοῦ ποιητικοῦ νοῦ ἀπορεῖν αὐτὸν καὶ διαλύειν ᾠήθησαν.

6 Moraux, P., ‘Le De Anima dans la tradition grecque: quelques aspects de l'interpretation du traité, de Théophraste à Thémistius’, in Lloyd, G.E.R. and Owen, G.E.L. (edd.), Aristotle on Mind and the Senses: The Proceedings of the 7th Symposium Aristotelicum (Cambridge, 1978), 281–324, at 324 n. 137Google Scholar.

7 Schroeder, F.M. and Todd, R.B., Two Greek Aristotelian Commentators on the Intellect (Toronto, 1990)Google Scholar; and Themistius, On Aristotle's On the Soul, transl. R.B. Todd (Ithaca, NY, 1996).

8 M.C. Lyons, An Arabic Translation of Themistius, Commentary on Aristoteles, De Anima (Oriental Studies 2) (Thetford, 1973). See already Lyons, M.C., ‘An Arabic translation of the commentary of Themistius’, Bulletin of the School of Oriental and African Studies 17 (1955), 426–35CrossRefGoogle Scholar.

9 Browne, G.M., ‘Ad Themistium Arabum’, ICS 11 (1986), 223–45Google Scholar.

10 Ballériaux, O., Thémistius. Son interpretation de la noétique aristotélicienne (Liege, 1941)Google Scholar, 183 n. 1.

11 Ballériaux, O., ‘Thémistius et l'exégèse de la noétique aristotélicienne’, Revue de Philosophie ancienne 7 (1989), 228–9Google Scholar; Ballériaux, O., ‘Thémistius et le néoplatonisme. Le ΝΟΥΣ ΠΑΘΗΤΙΚΟΣ et l'imortalité de l’âme’, Revue de Philosophie ancienne 12 (1994), 171–200, at 178–9 n. 23Google Scholar.

12 From the extant writings of Alexander, it is not possible to say for sure what his understanding of Aristotle's question at De an. III.5 430a23–5 was. It may well be that he thought the question is about forgetting knowledge that we once had and so making errors in a field in which we were once experts, i.e. (i). This seems to have been Theophrastus’ understanding: see Themistius, In De an. 108.18–28, that is, fr. XII (B), according to E. Barbotin, La théorie aristotélicienne de l'intellect d'après Théophraste (Louvain, 1954); or fr. 320A (part), according to W. Fortenbaugh, P. Huby, R. Sharples and D. Gutas (edd.), Theophrastus of Eresus. Sources for his Life, Writings, Thought and Influence, vol. 2 (Leiden, 1992).

13 Kutsch, W., ‘Ein Arabisches Bruchstück aus Porphyrios (?), ΠΕΡΙ ΨΥΧΗΣ, und die Frage des Verfassers der “Theologie des Aristoteles”’, Mélanges de l'Université Saint-Joseph 31 (1954), 265–85CrossRefGoogle Scholar.

14 I refer to pages and lines in the article quoted in n. 13 (lines are not marked by Kutsch himself).

15 If, as seems to be the case, Themistius is exploiting this Porphyrean (?) thought in his In De anima, he introduces at least one important innovation: he does not identify φαντασία with the passive intellect itself but rather with its proximate matter (see 100.30; for a potential indirect Porphyrean [?] inspiration behind this move, see [c] below). The identification of the passive intellect with φαντασία is common in Proclus, who is here probably drawing on Porphyry: see e.g. Proclus’ Commentary on the Republic [In Remp.] II 52.4–8 and 107.14–108.16, or his Commentary on the Timaeus I 244.11–22 and III 158.5–11.

16 Plotinus, Enn. IV.3.4.18–21.

17 What should we make of the plural ἅπαντες … ὅσοι at In De an. 101.36–7 (a question raised by CQ's referee)? Such a plural is not uncommon in Themistius when criticizing some view, even though this view can quite safely be traced back to a concrete individual (this individual may have had followers and the plural may also serve the function of making Themistius’ attack less personal). At In De an. III.5 102.36–103.19, for instance, Themistius introduces his attack on the reading of De an. III.5 that seems to go back to Alexander of Aphrodisias by referring to οἱ τὸν πρῶτον θεὸν λέγεσθαι παρ᾽ αὐτοῦ τὸν ποιητικὸν νοῦν ὑπολαμβάνοντες (102.36–7). In his polemic with ‘a critic of Aristotle’, i.e. probably Porphyry (see n. 18 below), at In De an. I.3 16.19–18.37 Themistius also occasionally switches from the singular to the plural; see especially 18.30–7 (cf. 17.12–16). For a similar switch, see also In De an. I.4 25.33–6; the doctrine of ensoulment here can probably be traced back to Porphyry as well, as suggested by the fragment of Contra Boethum in Euseb. Praep. evang. XV.11.2–3 (cf. Moraux [n. 6], 322 n. 114).

18 Another instance of the same strategy is Themistius’ account of knowledge acquisition (and especially concept formation) in On Aristotle's Posterior Analytics I.1 and II.19 (see R. Roreitner, ‘Themistius on concept formation’, Archiv für Geschichte der Philosophie 103 [forthcoming]). One more example would be Themistius’ polemic concerning the nature of perception at In De an. I.3 16.19–18.37 (reflected at In De an. I.4 27.8–29.21) with a ‘critic of Aristotle’, who is probably to be identified with Porphyry, too (so already a marginal note in Laurentianus 87,25; for more on this, see Moraux [n. 6], 320 nn. 102 and 103). It can be argued that Themistius is looking for an alternative to both Alexander's understanding of the perceptive soul as strictly impassive and a Neoplatonist account of perception as a kind of self-motion of the soul.

19 To explain this apparent asymmetry, Aristotle is reported to have used an analogy with health and illness: if incarnation is like an illness for the soul, we can perhaps better understand its forgetfulness during the embodied life and its alleged ability to remember this life after death, i.e. recovery.

20 That Themistius approached the Eudemus with a certain reservation is also suggested by In De an. III.5 107.3–4 (see n. 21 below).

21 The way in which Themistius approaches Plato's proofs of immortality is also typical of his moderate Aristotelianism. He maintains, in a conciliatory tone, that Plato's proofs work but not for the soul as intended: they only apply to the νοῦς (In De an. III.5 106.29–107.3). The same manoeuvre is undertaken with respect to the proofs presented in Aristotle's Eudemus (107.3–4).

22 I explain elsewhere why Themistius’ interpretation of Aristotle is most probably mistaken and why, none the less, he has a point in resisting Alexander's approach: ‘The νοῦς-body relationship in Aristotle's De anima’, in Fink, J.L. and Gregorić, P. (edd.), Encounters with Aristotle's Philosophy of Mind (Abingdon – New York, 2021), 247–78Google Scholar.