¹ Bernays, J., ‘Über die fälschlich dem Aristoteles beigelegte Schrift περὶ κόσμου’, in Gesammelte Abhandlungen (Berlin, 1885), 2.278–81Google Scholar, argues that the addressee is Tiberius Alexander, Philo's nephew and governor of Egypt in the latter half of the first century a.d. This is accepted by Pohlenz, M., Die Stoa (Göttingen 1948–9), 1.361 and 2.177Google Scholar.

² Exceptions are Gohlke, P., Aristoteles an König Alexander über die Welt (Paderborn, 1968 ³)Google Scholar; Reale, G., Aristotele: Trattato sul cosmo per Alessandro (Naples, 1974)Google Scholar; Bos, A.P., Aristoteles: Over de kosmos (Meppel, 1989)Google Scholar; Bos, A.P., ‘Considerazioni sul De mundo e analisi critica delle tesi di Paul Moraux’, Rivista di filosofia neoscolastica 82 (1990), 587–606Google Scholar. Cf. Reale, G. and Bos, A.P., Il trattato sul cosmo per Alessandro attribuito ad Aristotele (Milan, 1995)Google Scholar.

³ For an overview of the various conjectures concerning the date and authorship of the treatise, see Furley's introduction in Forster, E.S. and Furley, D.J. (edd.), Aristotle: On Sophistical Refutations; On Coming-to-be and Passing-away; On the Cosmos (Cambridge, MA, 1955), 340–1Google Scholar; Schönberger, O., Aristotle: Über die Welt (Stuttgart, 1991); 46–53Google Scholar; Reale and Bos (n. 2), 25–57: Thom, J.C., ‘The power of God in Pseudo-Aristotle's De mundo: an alternative approach’, in Helmig, C. and Markschies, C. (edd.), The World Soul and Cosmic Space: New Readings on the Relation of Ancient Cosmology and Psychology (Berlin, forthcoming)Google Scholar.

⁴ We follow Lorimer's Greek text in the Budé edition (1933) and Furley's English translation in the Loeb edition (1955). On occasion, Furley's translations are slightly modified.

⁵ See the parenthetical remark in ch. 4, 394b10, and also ch. 6, 398b20–3.

⁶ ἡ τῶν ὅλων τάξις τε καὶ διακόσμησις, ὑπὸ θεοῦ τε καὶ διὰ θεὸν ϕυλαττομένη. Some MSS read διὰ θεῶν ϕυλαττομένη.

⁷ This view can be found in e.g. Xen. Mem. 4.3.13–14; Pl. Resp. 506d–e and 508a–509a; in Arist. Metaph. Λ.10.1075a11–25; Theophr. Metaph. 4b11–18.

⁸ It is not clear whether this refers to each of the seven city walls of Ecbatana or each city wall in the empire. The latter seems to fit the intended conclusion better.

⁹ E.g. Pohlenz (n. 1), 1.361; Strohm, H. ‘Studien zur Schrift von der Welt’, MH 9 (1952), 137–75Google Scholar, at 160; Moraux, P., Der Aristotelismus im I. und II. Jh. n. Chr., Band 2: Der Aristotelismus bei den Griechen von Andronikos bis Alexander von Aphrodisias (Berlin, 1984), 37–8Google Scholar; Duhot, J.-J., ‘Aristotélisme et stoïcisme dans le ΠΕΡΙ ΚΟΣΜΟΥ pseudo-Aristotélicien’, RPhA 8 (1990), 191–228Google Scholar, esp. 203–4; and Thom (n. 3).

¹⁰ For an informative discussion of the way in which these devices are used in analogies, see Barryman, S., ‘Ancient automata and mechanical explanation’, Phronesis 48 (2003), 344–69CrossRefGoogle Scholar; Henry, D., ‘Embryological models in ancient philosophy’, Phronesis 50 (2005), 1–42CrossRefGoogle Scholar. See also Gregoric, P. and Kuhar, M., ‘Aristotle's physiology of animal motion: on neura and muscles’, Apeiron 47 (2014), 94–115CrossRefGoogle Scholar.

¹¹ So the puppets in question should not be confused with marionettes, in the case of which different body parts are moved by different strings.

¹² In Cic. Fat. 41–3 (= SVF 2.974, L.–S. 62C) and Gell. NA 7.2.6–13 (= SVF 2.1000, L.–S. 62D).

¹³ Lorimer, W.L., Some Notes on the Text of Pseudo-Aristotle “De mundo” (Oxford, 1925), 63–5Google Scholar; cf. Maguire, J.P., ‘The sources of Pseudo-Aristotle De mundo’, YClS 6 (1939), 109–67Google Scholar, at 151–2, and Duhot (n. 9), 207–11.

¹⁴ If this interpretation is correct, it seems at first sight that the author used the older eight-note celestial harmony scheme (seven planets plus the sphere of the fixed stars) that is attributed to the Pythagoreans in Aristotle's Cael. 2.9.290b12–291a9 and in Eratosthenes’ Hermes. This scheme is to be distinguished from the later versions of the doctrine using only seven tones (see Burkert, W., Lore and Science in Ancient Pythagoreanism [Boston, MA, 1972], at 351–2)Google Scholar. On the other hand, our text does not speak about the harmony of the spheres per se, but rather of the celestial bodies. Such a view would create obvious difficulties in the case of the fixed stars – we would simply have too many notes. Perhaps the stars emit a single note collectively.

¹⁵ For an overview and analysis of the evidence, see Wilson, P., The Athenian Institution of the Khoregia: The Chorus, the City and the Stage (Cambridge, 2000), esp. 134Google Scholar.

¹⁶ See also Pl. Leg. 665a, 653d–654a: Apollo, the Muses, and Dionysus as συγχορευτάς τε καὶ χορηγούς.

¹⁷ See e.g. Xenophon of Ephesus, An Ephesian Tale, 1.2.2, on Anthia, the fourteen-year-old beauty who led the chorus at the local festival in Ephesus and who was revered by locals as Artemis herself (see Wilson [n. 15], 349 n. 23).

¹⁸ Retaining the emendation of Wendland and Wilamowitz, followed by Lorimer and Furley.

¹⁹ See Lloyd, G.E.R., Polarity and Analogy (Cambridge, 1966), 252–3Google Scholar.

²⁰ ‘As our soul, being air, holds us together, so pneuma and air surround the whole world’ (οἶον ἡ ψυχή ἡ ἡμετέρα ἀὴρ οὖσα συγκρατεῖ ἡμᾶς, καὶ ὅλον τὸν κόσμον πνεῦμα καὶ ἀὴρ περιέχει, DK 13 B2 = Aëtius 1.3.4).

²¹ This localization of Phidias’ portrait is in contrast with other ancient descriptions of Athena's shield. Some of the numerous surviving accounts of the ornaments and scenes on the shield contain precious details about the depiction of the fight between the Greeks and the Amazons (e.g. Plin. HN 36.18 and Paus. 1.17.2.). Some others, most notably Plutarch (Vit. Per. 31), also mention that among the Greeks fighting the Amazons, an ageing bold figure is the depiction of Phidias himself, whereas a man raising a spear is supposed to portray Pericles. Yet – and here comes the interesting part – it seems fairly certain from these accounts that the middle of the shield was occupied by a Gorgon head, whereas the two fighters purportedly representing Phidias and Pericles were at the upper or more probably the lower extremity of the shield. Indeed, this is what we can see on the so-called ‘Strangford shield’, customarily supposed to be a copy of Athena's shield. Incidentally, Mansfeld, J., ‘Two attributions’, CQ 41 (1991), 541–4CrossRefGoogle Scholar, points out that the apparently mistaken localization of Phidias’ portrait is a further argument against attributing the De mundo to Aristotle. It is highly unlikely, he argues, that Aristotle, having spent so many years in Athens, could be so mistaken about the position of the portrait, forgetting also the Gorgon head. For a response, see Reale and Bos (n. 2), 170–1.

²² As Lloyd (n. 19), 58–9, observed: ‘The connection between the heavenly regions and divinity is a constant feature of Aristotle's theology. He often refers to religious beliefs shared, he says, by Greeks and Barbarians alike, according to which the heavenly bodies are gods and the heaven itself (the “uppermost region”) is divine.’

²³ All the MSS have ἐν πόλει δὲ νόμος, which is emended into νομοθέτης by Lorimer, followed by Furley. Lorimer (n. 13), 114–19, convincingly defends his emendation. Briefly, the emendation is justified because the author will continue by drawing a contrast between this group of analogies and the analogy of the law of a city. This would obviously not work had the law of a city already been included among this set of analogies. The point of contrast, i.e. that the command in the first group is wearisome, would just as obviously be inapplicable to the law. And the corruption is easy to explain in view of the prominence of the law analogy at the end of the chapter.

²⁴ Helmsman: Pl. Plt. 272e, Criti. 109c; Arist. De an. 2.1.413a8–9. Charioteer: Pl. Phdr. 246e–247a. For more references, see Lorimer (n. 13), 115–17.

²⁵ Translation from Hardie, R.P. and Gaye, R.K., rev. Barnes, J., The Complete Works of Aristotle, vol. 1 (Princeton, NJ, 1984), 446Google Scholar.

²⁶ Considering the dialectical position of the De mundo, it is worth noting that, in later Peripatetic tradition, claims about the trouble-free existence of the cosmic god are often connected with a direct criticism of the immanence of the Stoic god – just as we find these elements side by side in the De mundo. See e.g. Alexander, De mixtione, 226.24–9, and the careful analysis in Bénatouïl, T., ‘How industrious can Zeus be? The extent and objects of divine activity in Stoicism’, in Salles, R. (ed.), God and Cosmos in Stoicism (Oxford, 2009), 23–45CrossRefGoogle Scholar.

²⁷ Although final causation certainly is not prominent in this treatise, we would not agree with Duhot (n. 9), 215 and 224, that the author of this treatise is ignorant of final causation, or, for that matter, that ‘the noetic nature of God is equally absent from the treatise’ (224). van Nuffelen, P., Rethinking the Gods: Philosophical Readings of Religion in the Post-Hellenistic Period (Cambridge, 2011), 137CrossRefGoogle Scholar, also thinks that in the De mundo ‘god seems to have become reduced to a giant efficient cause rather than a final cause’.

²⁸ See Lloyd (n. 19), 335.

²⁹ Shelley, C., Multiple Analogies in Science and Philosophy (Amsterdam, 2003)CrossRefGoogle Scholar. It is important to note, however, that Shelley seems to restrict the term ‘multiple analogy’ to this type of multiple analogy. In our taxonomy, this is only one subclass of multiple analogies, and hence we give it a different name.

³⁰ Spiro, R.J., Feltovich, P.J., Coulson, R.L., and Anderson, D.K., ‘Multiple analogies for complex concepts: antidotes for analogy-induced misconception in advanced knowledge acquisition’, in Vosniadou, S. and Ortony, A. (edd.), Similarity and Analogical Reasoning (Cambridge, MA, 1989), 498–531CrossRefGoogle Scholar, seem to restrict the term ‘multiple analogy’ to this type. Once again, in our taxonomy this is only a subclass of multiple analogies.

³¹ See ibid.

³² Ibid., 528.

³³ Moraux (n. 9), 40.

³⁴ Ibid., our emphasis.

³⁵ This crucial aspect of the author's procedure has been overlooked by Duhot (n. 9) in his otherwise insightful discussion of the text. Failing to see that the analogies progressively emend one another, he thinks that they disconnectedly introduce discrepant or incompatible points, which leads him to believe that the argument is muddled and the treatise ‘of low philosophical value’ (233). Similarly, van Nuffelen (n. 27), 136, thinks that the author just ‘heaps comparison on comparison’. Van Nuffelen's failure to observe the connection between the Great King analogy and the law analogy leads him to claim that ‘Whereas Maximus [of Tyre in Or. 11] incorporated the notion of immobile law into his image of the Great King, On the World likens god separately to the Persian King and to law’ (136), and to make a negative appraisal of the author's use of the image of the Great King (138). Nevertheless, Chapter 6 of van Nuffelen's book is very informative about that image in Post-Hellenistic literature.