¹ Such a view would be found most famously in W. Jaeger, Aristotle: Fundamentals of the History of His Development (Oxford, 1948), 39–53, followed by A.H. Chroust, ‘Eudemus or On the Soul: a lost dialogue of Aristotle on the immortality of the soul’, Mnemosyne 19 (1966a), 17–30; A.H. Chroust, ‘The psychology in Aristotle’s lost dialogue Eudemus or On the Soul’, AClass 9 (1966b), 49–62; W.D. Ross, ‘The development of Aristotle’s thought’, in I. Düring and G.E.L. Owen (edd.), Aristotle and Plato in the Mid-Fourth Century (Uppsala, 1960), 1–17; F. Nuyens, L’évolution de la psychologie d’Aristote (Louvain, 1948). A developmentalist view, albeit substantially less dramatic than Jaeger’s rendering, can be found even in early attempts at histories of philosophy and in works of the (late) Renaissance: E. Berti, La filosofia del primo Aristotele (Padova, 1962), 9–33.

² D.A. Rees, ‘Theories of the soul in the early Aristotle’, in I. Düring and G.E.L. Owen (edd.), Plato and Aristotle in the Mid-Fourth Century (Uppsala, 1960), 191–200; H. Cherniss, Aristotle’s Criticism of Plato and the Academy (Oxford, 1944); I. Düring, ‘Aristotle and Plato in the mid-fourth century’, Eranos 54 (1956), 109–20.

³ V. Rose, Aristoteles pseudographus (Leipzig, 1863); V. Rose, Aristotelis qui ferebantur librorum fragmenta. Opera Aristotelis, vol. 5 (Leipzig, 1870, repr. 1967); V. Rose, Aristotelis qui ferebantur librorum fragmenta (Leipzig, 1886); R. Walzer, Aristotelis dialogorum fragmenta (Florence, 1934); W.D. Ross, Aristotelis fragmenta selecta (Oxford, 1955); O. Gigon, ‘Librorum deperditorum fragmenta’, in Aristotelis opera, vol. 3 (Berlin, 1987). Less commented on is E. Heitz, ‘Aristotelis fragmenta’, in Aristotelis opera omnia graece et latine, vol. 4 (Paris, 1868).

⁴ I refer here to Ross’s collection for convenience; there is little significant difference among the collections of fragments, except that Rose originally included also the passages from Pseudo-Plutarch’s De musica (1138C–1140B = fr. 47 Rose), which do not so clearly fit in with the topics covered in the Eudemus. Walzer considers the De musica passages to belong rather to De philosophia, and Ross follows Walzer (fr. 25 Walzer/Ross).

⁵ Moreover, while the late antique commentators make it seem like Aristotle and Plato overlapped in their endeavours to argue against the view of the soul as a harmony, one can only reasonably consider that there were two arguments against the view in the Eudemus: the one found also in De an. 408a1–5, and another which rests upon the ‘un-Platonic’ division between substance and quality. For a presentation of the harmony between the supposedly ‘later’ Aristotle and the arguments against the soul as a harmony in the Eudemus, see M. Vogiatzi, ‘Aristotle on the soul as harmony’, Elenchos 41 (2020), 245–68.

⁶ Elias in Cat. 114.25–115.12 = Eudemus fr. 39 Rose = 3 Walzer = 3 Ross = 61 Gigon.

⁷ in Cat. 114.25–7 κατασκϵυάζων δὲ τὴν ἀθανασίαν τῆς ψυχῆς κἀν τοῖς ἀκροαματικοῖς δι’ ἀναγκαστικῶν λόγων κατασκϵυάζϵι, ἐν δὲ τοῖς διαλογικοῖς διὰ πιθανῶν ϵἰκότων. Translations of Elias are mine.

⁸ in Cat. 115.11–12. ὁ δὲ Ἀριστοτέλης ἐν τοῖς διαλογικοῖς μάλιστα δοκϵῖ κηρύττϵιν τὴν ἀθανασίαν τῆς ψυχῆς.

⁹ H. Tarrant, Proclus. Commentary on Plato’s Timaeus, Volume 6, Book 5: Proclus on the Gods of Generation and the Creation of Humans (Cambridge, 2017).

¹⁰ in Tim. 3.323.31–324.4 ὃ δὴ καὶ Ἀριστοτέλης ζηλώσας ἐν τῇ Πϵρὶ ψυχῆς πραγματϵίᾳ ϕυσικῶς αὐτὴν μϵταχϵιριζόμϵνος οὔτϵ πϵρὶ καθόδων ψυχῆς οὔτϵ πϵρὶ λήξϵων ἐμνημόνϵυσϵν, ἀλλ’ ἐν τοῖς Διαλόγοις χωρὶς ἐπραγματϵύσατο πϵρὶ αὐτῶν καὶ τὸν προηγούμϵνον κατϵβάλϵτο λόγον. Translation: Tarrant (n. 9).

¹¹ Procl. in Remp. 2.349.13–16 λέγϵι δὲ καὶ ὁ δαιμόνιος Ἀριστοτέλης αἰτίαν δι’ ἣν ἐκϵῖθϵν μὲν ἰοῦσα ἡ ψυχὴ δϵῦρο ἐπιλανθάνϵται τῶν ἐκϵῖ θϵαμάτων, ἐντϵῦθϵν δὲ ἐξιοῦσα μέμνηται ἐκϵῖ τῶν ἐνταῦθα παθημάτων; my translation.

¹² Düring (n. 2); Chroust (n. 1 [1966b]), at 51–2; Rees (n. 2).

¹³ in de An. 106.29–107.5 = Eudemus fr. 38 Rose = 2 Walzer = 2 Ross = 58 Gigon.

¹⁴ De an. 106.29–107.5 καὶ οἱ λόγοι δὲ οὓς ἠρώτησϵ (Plato in the Phaedo) πϵρὶ ψυχῆς ἀθανασίας ϵἰς τὸν νοῦν ἀνάγονται σχϵδόν τι οἱ πλϵῖστοι καὶ ἐμβριθέστατοι, ὅ τϵ ἐκ τῆς αὐτοκινησίας (ἐδϵίχθη γὰρ ὡς αὐτοκίνητος μόνος ὁ νοῦς, ϵἰ τὴν κίνησιν ἀντὶ τῆς ἐνϵργϵίας νοοίημϵν), καὶ ὁ τὰς μαθήσϵις ἀναμνήσϵις ϵἶναι λαμβάνων καὶ ὁ τὴν πρὸς τὸν θϵὸν ὁμοιότητα· καὶ τῶν ἄλλων δὲ τοὺς ἀξιοπιστοτέρους δοκοῦντας οὐ χαλϵπῶς ἄν τις τῷ νῷ προσβιβάσϵιϵν, ὥσπϵρ γϵ καὶ τῶν ὑπ’ αὐτοῦ Ἀριστοτέλους ἐξϵιργασμένων ἐν τῷ Eὐδήμῳ. ἐξ ὧν δῆλον ὅτι καὶ Πλάτων τὸν νοῦν ἀθάνατον μόνον ὑπολαμβάνϵι … Translation: R.B. Todd, Themistius: On Aristotle on the Soul (London, 1996).

¹⁵ See n. 1 for references.

¹⁶ Ps.-Simpl. in De an. 221.20–33 = Eudemus fr. 46 Rose = 8 Walzer = 8 Ross = 64 Gigon.

¹⁷ B. Effe, Studien zur Kosmologie und Theologie der Aristotelischen Schrift Über die Philosophie (Munich, 1970), 150–3.

¹⁸ Iambl. Protr. 7.41.27–42.4 Pistelli = Protrepticus fr. 6 Ross = 73 Gigon; my translation.

¹⁹ Jaeger (n. 1), 49.

²⁰ H. Flashar, ‘Platon und Aristoteles im Protreptikos des Jamblichos’, AGPh 47 (1965), 53–79. D.S. Hutchinson and M.R. Johnson, ‘Authenticating Aristotle’s Protrepticus’, OSAPh 29 (2005), 193–294, at 251 (Book 7) and 258 (Book 8) admit that it is difficult to determine the origin of the ideas contained in the seventh and eight books of Iamblichus’ Protrepticus, although they still wish to reconstruct Aristotle’s Protrepticus on the basis of many of these passages. A.P. Bos, ‘Aristotle’s Eudemus and Protrepticus: are they really two different works?’, Dionysius 8 (1984), 19–51 questions whether or not the Protrepticus and the Eudemus were two separate dialogues.

²¹ D. Runia, ‘Festugière revisited: Aristotle in the Greek Patres’, VChr 43 (1989), 1–34, at 21.

²² For the presence of the lost Aristotle in Tatian, see L. Alfonsi, ‘Echi del giovane Aristotele in Taziano’, Revue d’études augustiniennes et patristiques 2 (1956), 251–6. A similar hesitation to say that the soul is immortal for Aristotle is also found in Pseudo-Justin Martyr, Cohortatio ad gentiles 7.E.2. For the reception of the lost works of Aristotle in this text, see L. Alfonsi, ‘Traces de jeune Aristote dans la Cohortatio ad Gentiles faussement attribuée à Justin’, VChr 2 (1948), 65–85.

²³ M. Edwards, Aristotle and Early Christian Thought (London, 2019), 38–54 likewise singles out Origen and Clement, as well as Tertullian and Basilides, as being among the figures of the Patristic period who were somewhat in favour of incorporating the texts of Aristotle.

²⁴ Clem. Al. Strom. 5.13.88.1–2 ἐντϵῦθϵν οἱ ἀμϕὶ τὸν Πυθαγόραν θϵίᾳ μοίρᾳ τὸν νοῦν ϵἰς ἀνθρώπους ἥκϵιν ϕασί, καθάπϵρ Πλάτων καὶ Ἀριστοτέλης ὁμολογοῦσιν; W. Wilson, Clement of Alexandria, Stromata: Ante-Nicene Fathers, vol. 2 (Buffalo, NY, 1885).

²⁵ Origen, C. Cels. 3.80.14–16 ἀγωνισάσθω οὖν μηκέτι κρύπτων τὴν ἑαυτοῦ αἵρϵσιν ἀλλ’ ὁμολογῶν ἐπικούρϵιος ϵἶναι πρὸς τὰ παρ’ Ἕλλησι καὶ βαρβάροις οὐκ ϵὐκαταϕρονήτως λϵγόμϵνα πϵρὶ τῆς ἀθανασίας τῆς ψυχῆς ἢ τῆς ἐπιδιαμονῆς αὐτῆς ἢ τῆς τοῦ νοῦ ἀθανασίας …; F. Crombie, Origen: Contra Celsum. Ante-Nicene Fathers, vol. 4 (Buffalo, NY, 1885).

²⁶ C. Cels. 2.12. Nevertheless, he does not refer to Aristotle especially negatively throughout the rest of the text, but rather is quite even-handed, e.g. Christ’s advice to flee the city that persecutes you (Matthew 10:23) is exemplified by Aristotle’s second departure from Athens (C. Cels. 1.65).

²⁷ For the twenty-one fragments (in Rose’s edition) derived from Clement, see E.A. Clark, Clement’s Use of Aristotle: The Aristotelian Contribution to Clement of Alexandria’s Refutation of Gnosticism (Lewiston, NY, 1977), 10–13. Bernays analyses parallels between Clem. Al. Strom. 2 and Eth. Nic. 1111a16–18, although he does not conclude that this is sufficient evidence that Clement had a copy of the Nicomachean Ethics in its entirety. J. Bernays, ‘Zu Aristoteles und Clemens’, Symbola philologorum Bonnensium in honorem Friderici Ritschelii collecta (Leipzig, 1864), 301–12. For more on Clement’s use of the lost writings of Aristotle, see L. Alfonsi, ‘Motivi tradizionali del giovane Aristotele in Clemente Alessandrino e in Atenagora’, VChr 7 (1953), 129–42; G. Lazzati, L’Aristotele perduto e gli scrittori cristiani (Milan, 1938), 9–34; A.J. Festugière, ‘Aristote dans la littérature grecque chrétienne jusqu’à Théodoret’, in L’idéal religieux des Grecs et l’Évangile (Paris, 1932), 221–63. For Clement’s appropriation of Aristotle’s Categories in the Stromata, see M. Havrda, ‘Categories in Stromata VIII’, Elenchos 33 (2012), 197–225. None of this implies that Clement had full copies of Aristotle’s writings, and it seems likely that he used manuals: E. de Faye, Clément d’Alexandrie: étude sur les rapports du Christianisme et de la philosophie grecque au II ^e siècle (Paris, 1906), 333–6. de Faye relies heavily on H. Diels, Doxographi Graeci (Berlin, 1879) and seems to allude to the Vetusta Placita as a common source for Clement, Philo and Cicero. This would be disproven by J. Mansfeld and D. Runia, Aetiana: The Method and Intellectual Content of a Doxographer. Vol. 1: The Sources (Leiden, 1996), 327–32 on the Placita of Aetius. The chance that Clement, Cicero and Philo share a common source would be even slighter given the (controversial) view of A. Lebedev, ‘Did the doxographer Aetius ever exist?’, in Philosophie et culture. Actes du XVIIe Congrès mondial de philosophie (Paris, 1988), 813–17, which denies even the existence of Aetius as the single author of the Placita. This is all to say, it seems rather unlikely that Cicero and the school of Alexandria would have had access to the same manuals.

²⁸ Clark (n. 27), 145.

²⁹ Only the passage in Iamblichus is included in the collected fragments, always as having belonged to the Protrepticus: fr. 60 Rose = 10b Walzer = 10b Ross = 73 Gigon. See n. 20 above for literature that has questioned the thesis that Iamblichus only relied on Aristotle’s Protrepticus, as opposed to actively synthesizing different texts. J. Brunschwig, ‘Aristote et les pirates tyrrhéniens (A propos des fragments 60 Rose du Protreptique)’, RPhilos 153 (1963), 171–90 considers this narrative in particular to belong to the Eudemus.

³⁰ V. Limone, ‘Origen’s explicit references to Aristotle and the Peripateticians’, VChr 72 (2018), 390–404, at 403 n. 60 also notes this passage.

³¹ P. Tzamalikos, Origen: Philosophy of History & Eschatology (Leiden, 2007), 244–7.

³² For more on Origen’s access to Aristotle, in addition to Runia (n. 21), see also Limone (n. 30), 390–404; G. Bardy, ‘Origène et l’aristotélisme’, in Mélanges Gustave Glotz, vol. 1 (Paris, 1932), 75–83; P. Tzamalikos (n. 31), 4–5, 84, 244–7 and passim. For more specific examples see the indexes of P. Koetschau (ed.), Origenes Werke, II. Buch V–VIII Gegen Celsus. Die Schrift vom Gebet (Leipzig, 1899), which also indicate that Origen often references Aristotle without citing him explicitly. These studies, however, are inconclusive; I will return to this matter at the end of this article.

³³ Pace Edwards (n. 23), 38–9, who considers that ‘the writings of Tertullian give little evidence of a close reading of Aristotle’. From this assessment he excludes Tertullian’s De anima, which in his view shows great precision in reading and interpreting Aristotle. Tertullian’s greatest detail in his attention to Aristotle is indeed to be found there. Edwards also well notes that Tertullian subtly distinguishes the views of Aristotle and the Peripatetic Strato against Dicaearchus, in so far as Strato and Aristotle both posit the existence of a hegemonic faculty of the soul (De an. 14.1 = Strato, fr. 56 Desclos–Fortenbaugh), although Aristotle locates that in the heart and Strato in the head (cf. De res. carn. 15.3–5). For discussion of this passage, see: M.-L. Desclos and W.W. Fortenbaugh, Strato of Lampsacus: Text, Translation, and Discussion (London, 2010), 308 and 407–8. The term which Tertullian uses in this passage as well as in 15.1, hegemonic (ἡγϵμονικόν), is significant in so far as Iamblichus uses the same term in the above-mentioned Protr. 7.41.27–42.4. John Philoponus likewise attributes this view to Aristotle at in De an. 195.10–11 (καὶ αὐτὸς ὁ Ἀριστοτέλης ϕαίνϵταί που δοξάζων ἐν τῇ καρδίᾳ τὸ ἡγϵμονικὸν ἱδρῦσθαι). To this list, we may add Philo (De somn. 1.30–1), who presents a number of different theories as to what ὁ ἡγϵμὼν νοῦς is—one of which is ἐνδϵλέχϵια (a term associated with Aristotle’s dialogues, as opposed to the ἐντϵλέχϵια of De anima). Furthermore, we find Origen twice using the terminology of ἡγϵμονικόν to describe the hegemonic faculty of the mind as being in the heart (Hom. I in Ps. 36.3–6), and how Christ illumines the hegemonic faculty of the mind (Comm. in Joann. 1:25.160–1). It is hard to discern a common source for all these rather disparate figures. While Philo or Tertullian may lift this term from the Stoics, e.g. from Posidonius (Diog. Laert. 7.138–9), there is little evidence that Iamblichus or Origen would be relying on Posidonius here. Moreover, the idea that they reference is distinct both from the Stoic idea of hegemonic nous (which is rather a cosmic principle) and the materialist view of hegemonic nous being located physically in the head (i.e. Strato’s view: fr. 121 Wehrli ap. Pollux 2.226).

³⁴ J. H. Waszink, ‘Traces of Aristotle’s lost dialogues in Tertullian’, VChr 1 (1947), 137–49. Not all of Waszink’s conclusions regarding the questions of mediation between Tertullian and Aristotle’s original texts are persuasive, however. For example, he argues that Tertullian could not have had a full copy of the Eudemus because he knows of the narrative of Midas and Silenus, also found in Aristotle’s Eudemus, via Theopompus (Eudemus fr. 44 Rose = 6 Walzer = 6 Ross = 65 Gigon = Ps.-Plut. Cons. ad Apoll. 115B1–E9; cf. Cic. Tusc. 1.114, Arist. Pol. 1257b14–17). This is far from obvious. Indeed, while Tertullian does claim to know of Silenus auctore Theopompo, this does not mean that he only knows of this single narrative of Silenus. Rather, there is good reason to think that Tertullian had a number of accounts available to him, or at least two: we have no reason to think that Theopompus ever mentioned Midas’ ears in so far as Aelian does not include this part of the narrative. Yet in De an. 2.3, Tertullian specifically references the ears of Midas (Silenum Phrygem, cui a pastoribus perducto ingentes aures suas Midas tradidit). This cannot simply be an allegorical ‘lend ears’, as in ‘listen to’, because the case of ‘ears’ is wrong (aures should be auribus).

³⁵ Tertullian also probably receives Aristotle’s embryology via Soranus: R. Polito, ‘I quattro libri sull’anima di Sorano e lo scritto De anima di Tertulliano’, RSF 49 (1994), 423–68. For the similarity between Tertullian’s De carn. Christ. 5 and Arist. Rhet. 23.22 see J. Moffatt, ‘Aristotle and Tertullian’, JThS 17 (1916), 170–1.

³⁶ Tert. De an. 12.3–6 Waszink = Gigon 971 (Fragmente ohne Buchangabe). Gigon alone includes this fragment in his collection, but does not assign it to any specific work. That this passage is peculiar is noted also by E. Barbotin, ‘Deux témoignages patristiques sur le dualism aristotélicien de l’âme et de l’intellect’, in Autour d’Aristote – Recueil d’études de philosophie ancienne et médiévale offert à Monseigneur A. Mansion (Leuven, 1955), 375–85. However, he does not connect it to a lost work of Aristotle but attributes the inaccuracy of Tertullian’s summary of Aristotle’s views simply to his own Christian interpretation of Aristotle. This is difficult to believe, as will be shown below.

³⁷ My translation. Here, we should also add Theodoret, who communicates the same position as Tertullian but in a shortened form (i.e. that Aristotle divides soul and intellect and maintains intellect alone as immortal), as Barbotin (n. 36) highlights, although Theodoret does not have access to the Eudemus but relies on intermediary sources (Graec. affect. cur. 5.28.1–29.1).

³⁸ Translation: W.S. Hett, Aristotle: On the Soul. Parva naturalia. On Breath (Cambridge, MA, 1957), slightly modified.

³⁹ Eth. Nic. 1177b30–31 ϵἰ δὴ θϵῖον ὁ νοῦς πρὸς τὸν ἄνθρωπον, καὶ ὁ κατὰ τοῦτον βίος θϵῖος πρὸς τὸν ἀνθρώπινον βίον; translation from H. Rackham, Aristotle: Nicomachean Ethics (Cambridge, MA, 1926).

⁴⁰ Runia (n. 21); Bardy (n. 32), 75–83; Limone (n. 30), 390–404. Limone (pages 403–4) concludes from his study that what he has collected as explicit references to Aristotle in Origen ‘neither persuade[s] about his in-depth knowledge of Aristotle’s writings nor exclude[s] his first-hand access to them’.

⁴¹ There are three examples of this: goal, homonymy and verb: Limone (n. 30), 395–6.

⁴² This cannot be taken as entirely certain. Iamblichus must have had access to some dialogues of Aristotle, or some extensive reports. J. Pépin, Théologie cosmique et théologie chrétienne (Paris, 1964) controversially considers that Ambrose of Milan had access to the lost dialogue De philosophia.