Skip to main content Accessibility help
Hostname: page-component-55597f9d44-t4qhp Total loading time: 0.357 Render date: 2022-08-19T10:54:16.804Z Has data issue: true Feature Flags: { "shouldUseShareProductTool": true, "shouldUseHypothesis": true, "isUnsiloEnabled": true, "useRatesEcommerce": false, "useNewApi": true } hasContentIssue true


Published online by Cambridge University Press:  04 March 2020

Benjamin Harriman*
The University of Edinburgh


In his Meditations, Marcus Aurelius repeatedly presents a disjunction between two conceptions of the natural world. Either the universe is ruled by providence or there are atoms. At 4.3, we find perhaps its most succinct statement: ἀνανεωσάμενος τὸ διεζευγμένον τό⋅ ἤτοι πρόνοια ἢ ἄτομοι (recall the disjunction: either providence or atoms). The formulation of the disjunction differs; at 7.32, being composed of atoms is contrasted with a stronger sort of unity (ἕνωσις) that may survive death. In 10.6 and 11.18 Marcus simply offers φύσις (nature, construed in the Stoic manner as providentialist and causally efficacious) in opposition. On the surface, the contrast between the theory of atomism and the acceptance of providence seems to not warrant the term ‘disjunction’; it seems possible to accept both atomism and a causally determined providential universe. Yet, it is agreed on all sides, in the recent literature, that the relevant contrast for Marcus is not between the atomist and the non-atomist views of the constitution of the natural world as such but between two entailments that follow from the atomist Epicurean and the non-atomist Stoic advocacy of these positions. The contrast is between the providential ordering of the Stoic universe and the chaotic chance-ridden Epicurean model.

Research Article
Copyright © The Classical Association 2020

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)


I would like to thank David Sedley, George Boys-Stones, Simon Trépanier, the anonymous referee and the editors of this journal for their helpful criticism and advice. The Leverhulme Trust generously supported this research.


1 4.27, 6.10, 6.24, 6.44, 7.32, 7.50, 9.28, 9.39, 10.6, 11.18. A similar disjunction is implied at 6.44 and 8.25.

2 The disjunctions do not form a homogeneous group but are of several different types. This is explored further below.

3 Cooper, J., ‘Moral theory and improvement: Marcus Aurelius’, in id., Knowledge, Nature, and the Good (Princeton, 2004), 335–68Google Scholar.

4 Rist, J., ‘Are you a Stoic?’, in Meyer, B.F. and Sanders, E.P. (edd.), Jewish and Christian Self-Definition, vol. 3 (London, 1982), 2359Google Scholar.

5 Brunt, P.A., ‘Marcus Aurelius in his Meditations’, JRS 64 (1974), 120Google Scholar, at 3 n. 15, citing 1.17.8 and 8.1. The criticism of syllogistic reasoning is familiar from Seneca; cf. Ep. 45, 48, 49, 82 and 83.

6 See 7.67, where Marcus appears to give up hope of becoming a φυσικός or a διαλεκτικός. We shall soon return to this passage.

7 2.2; cf. 12.3, where τὸ ἡγεμονικόν is replaced with νοῦς. See Gill, C., ‘Marcus Aurelius’ Meditations: how Stoic and how Platonic?’, in Bonazzi, M. and Helmig, C. (edd.), Platonic Stoicism, Stoic Platonism (Leuven, 2007), 189207Google Scholar. Gill provides an account of Marcus’ apparent flirtation with Platonist psychology that emphasizes his Stoic credentials.

8 3.16; cf. 6.32. This circumscribed identification of pneuma with the soul and not as the divine power enabling the tensional hexeis and individuating qualities of items in the world is itself somewhat at odds with orthodox Stoicism. See Sedley, D., ‘Marcus Aurelius on physics’, in van Ackeren, M. (ed.), A Companion to Marcus Aurelius (Malden, 2007), 396407Google Scholar, at 397.

9 See 2.2, 4.39, 11.19, in particular for Marcus’ identification. For the influence of Posidonius, see Rist (n. 4), 31. Posidonius' integration of Platonist elements within his philosophy is controversial. Galen (see, for example, Hipp. et Plat. Books 4–5) insists that Posidonius accepted Plato's tripartition of the soul. This characterization has been subject to recent, widespread scepticism; see, in particular, Tieleman, T., Chrysippus' On Affections (Leiden, 2003), 198287CrossRefGoogle Scholar. See Ju, A.E., ‘Stoic and Posidonian thought on the immortality of the soul’, CQ 59 (2009), 112–24CrossRefGoogle Scholar for an account of Posidonius’ view of the soul's post-mortem persistence.

10 See Hadot, P. (transl. Chase, M.), The Inner Citadel (Cambridge, MA, 1998), 165–8Google Scholar. Hadot finds at 6.13 one of his strongest pieces of evidence for the theory that the Meditations is a personal text intended as a series of therapeutic exercises. Here Marcus seems to be advocating a method of eliminating excess pride and false beliefs in non-virtuous pursuits by examining the physical realities of things. This applies, mutatis mutandis, to those things that are unpleasant (6.36). The point, as Hadot suggests, is that the process is intended to help overcome prejudices about the world.

11 Marcus’ disgust with his own physicality is evident in a well-known passage from 8.24: ὁποῖόν σοι φαίνεται τὸ λούεσθαι: ἔλαιον, ἱδρώς, ῥύπος, ὕδωρ γλοιῶδες, πάντα σικχαντά: τοιοῦτον πᾶν μέρος τοῦ βίου καὶ πᾶν ὑποκείμενον.

12 Annas, J., ‘Marcus Aurelius: ethics and its background’, Rhizai 2 (2004), 103–19Google Scholar, at 107–14.

13 See also Asmis, E., ‘The Stoicism of Marcus Aurelius’, ANRW 2.36.3 (Berlin, 1989), 2228–52Google Scholar, at 2252. Asmis, though, adopts an eclectic view of Marcus at odds with Annas's presentation.

14 See, for example, Gill, C., Marcus Aurelius: Meditations, Books 1–6 (Oxford, 2013), lxixlxxivGoogle Scholar. Gill provides a very helpful overview, though no firm conclusion other than advocating against putting too much stress on these passages for the overall interpretation of the Meditations. See also Sedley (n. 8) and Bénatouïl, T., ‘Theôria and scholê in Epictetus and Marcus Aurelius: Platonic, Stoic, or Socratic?’, in Long, A.G. (ed.), Plato and the Stoics (Cambridge, 2013), 147–73CrossRefGoogle Scholar. Although I agree with Annas on Marcus’ general attitude towards Stoic physics, I offer quite a different reading of his approach to Epicureanism and a different solution to the most difficult group of disjunctions.

15 Rist (n. 4), 33 notes this connection. Hadot (n. 10), 151–3 recognizes the complexity of Marcus’ use of disjunctions by charting the further ‘sub-disjunctions’ generated in 9.28.2. My conclusions about the significance of this complex structure very much differ in so far as I do not accept Hadot's identification of the parts of philosophy with the three Epictetan disciplines (τόποι), and thus do not think that Marcus’ disjunctive framework serves to reinforce the ‘discipline of desire’ which Hadot identifies with physics.

16 10.7, 9.28 and 4.21 respectively. On this point, see Gill (n. 14), lxviii and Sedley (n. 8), 405–6.

17 One might think here of Anaïs Nin's widely quoted observation that ‘we write to taste life twice, in the moment, and in retrospect’. The point is that we need to take into account how Marcus would have made use of his work after its composition.

18 Brunt (n. 5).

19 Cooper (n. 3), 366–8. For Fronto and his historical context, see Champlin, E., Fronto and Antonine Rome (Cambridge, MA, 1980)CrossRefGoogle Scholar.

20 That we have some forty instances of Marcus telling himself ‘to always remember’ is frequently cited. See Brunt (n. 5), 3 n. 18 and Rist (n. 4), 24. In this respect, I find much to agree with in Ackeren, M. van, Die Philosophie Marc Aurels (Quellen und Studien zur Philosophie 103.2), 2 vols. (Berlin, 2011), 428–43CrossRefGoogle Scholar. van Ackeren's view of the Meditations as a self-directed dialogue in which the form in which Marcus engages helps to structure his self-improvement project aligns closely with what I will argue. The crucial difference, however, is that I see no reason to explain away Marcus’ interaction with non-Stoic sources. The contemplative framework I suggest provides a justification for Marcus’ apparent receptivity.

21 I follow the translation of Farquharson, A.S.L., The Meditations of Marcus Aurelius (Oxford, 1944)Google Scholar, with modifications throughout. Here I keep particularly close to avoid tendentiousness; however, I take him to significantly undertranslate διαλεκτικός and φυσικός.

22 Fronto reminds Marcus of this point; see his De Eloquentia 2.11–14, in the edition of van den Hout, M.P.J., M. Cornelius Fronto: Epistulae (Leipzig, 1988), 140–2Google Scholar. The difficulties of his duties as emperor are a frequent theme. Bénatouïl (n. 14), 155 helpfully cites 2.1, 6.13, 9.36 and 10.31.

23 Annas (n. 12), 116 n. 34.

24 Pl. Ap. 18b–c, 19c and 23d.

25 Farquharson (n. 21), 487. For Marcus’ response to Aristo's philosophy, see Fronto, Ad M. Caes. et invicem 4.13, in van den Hout, M.P.J., M. Cornelius Fronto: Epistulae (Leipzig, 1988), 67–8Google Scholar. That it is the dissident Stoic that Marcus has in mind is agreed by most but not all. See Hadot (n. 10), 12–14 for a defence and bibliography. For Aristo's claim that physics is beyond human, see SVF 1.352.

26 Marcus often uses the language of nakedness and of stripping things down to their essence. See 3.11, 6.13, 11.2 in particular. Gill (n. 14), xli–xlii makes this point.

27 See, in particular, 2.17, 4.43 and 9.28. For a list of further references, see Long, A.A., ‘Heraclitus and Stoicism’, in id., Stoic Studies (Cambridge, 1996), 3557Google Scholar, at 56–7. Long provides the classic account of the role of Heraclitus in the Stoic school. I have argued elsewhere that Heraclitus proved a decisive influence in helping to inform Cleanthes’ method of developing the foundation-stone of Stoic epistemology, the kataleptic impression. See also Asmis (n. 13), 2246–9.

28 Stoic interest in Pythagoras and Pythagoreanism as an important precursor to Plato has received an interesting recent treatment by Ju, A.E., ‘Posidonius as historian of philosophy’, in Schofield, M. (ed.), Aristotle, Plato and Pythagoreanism in the First Century b.c. (Cambridge, 2013), 95117Google Scholar. She concludes that Posidonius sought to examine the relationship between Platonic and Pythagorean thinking and that this allows him to recast his Platonic inheritance by mathematizing it and, in doing so, he ‘endorsed Pythagoreanism as an august precedent’.

29 See, for example, Brown, E., ‘Contemplative withdrawal in the Hellenistic age’, Philosophical Studies 137 (2008), 7989CrossRefGoogle Scholar.

30 Bénatouïl (n. 14), 157.

31 See Bénatouïl, T. and Bonazzi, M., ‘θεωρία and βίος θεωρητικός from the Presocratics to the end of antiquity: an overview’, in Bénatouïl, T. and Bonazzi, M. (edd.), Theoria, Praxis, and the Contemplative Life after Plato and Aristotle (Leiden, 2012), 114CrossRefGoogle Scholar.

32 See, for example, 9.30.

33 Bénatouïl (n. 14), 156–7 helpfully points to Sellars, J., The Art of Living (Ashgate, 2003), 150–65Google Scholar.

34 εὐφροσύνη ἀνθρώπου ποιεῖν τὰ ἴδια ἀνθρώπου, ἴδιον δὲ ἀνθρώπου εὔνοια πρὸς τὸ ὁμόφυλον, ὑπερόρασις τῶν αἰσθητικῶν κινήσεων, διάκρισις τῶν πιθανῶν φαντασιῶν, ἐπιθεώρησις τῆς τῶν ὅλων φύσεως καὶ τῶν κατ’ αὐτὴν γινομένων.

35 See Bénatouïl (n. 14), 159–60 and Rutherford, R.B., The Meditations of Marcus Aurelius: A Study (Oxford, 1989), 157–8Google Scholar.

36 Cf. Plut. Cic. 3.

37 See, for example, DK B31, DK B76 and DK B90 on elemental transformation.

38 περισκοπεῖν ἄστρων δρόμους ὥσπερ συμπεριθέοντα καὶ τὰς τῶν στοιχείων εἰς ἄλληλα μεταβολὰς συνεχῶς ἐννοεῖν: ἀποκαθαίρουσι γὰρ αἱ τούτων φαντασίαι τὸν ῥύπον τοῦ χαμαὶ βίου. An intriguing parallel to Marcus’ discussion of the importance of elemental change to therapeutic contemplation is found in Musonius Rufus, fr. 42 Hense. Musonius shows a similar interest in Heraclitus as well (fr. 18a Hense). This should not be very surprising; Fronto claims that his teacher, Athenodotus, studied with Musonius, suggesting a direct paedagogical line between Musonius and Marcus (Ad Eloquentia 1.4, in van den Hout, M.P.J., M. Cornelius Fronto: Epistulae [Leipzig, 1988], 135)Google Scholar.

39 Heraclitus DK B36; cf. DK B76 and DK B77.

40 Heraclitus DK B71 and DK B72.

41 Heraclitus DK B73.

42 Heraclitus DK B74; cf. DK B70 and DK B79.

43 λόγῳ τῷ τὰ ὅλα διοικοῦντι is, no doubt, a Stoic gloss and not a part of a verbatim quotation from Heraclitus; see Kahn, C.H., The Art and Thought of Heraclitus (Cambridge, 1979), 104Google Scholar. The notion of logos as a governing principle harmonizing opposites is perhaps more akin to its Stoic conception. Yet, we are primarily concerned with what Marcus took to be Heraclitean, and not with what we can safely ascribe to the earlier figure from our vantage point. With this in mind, it seems reasonable to assume that Marcus’ gloss represents his understanding of what Heraclitus said and thought. Long (n. 27), 51 observes that the Stoics took their notion of the logos to overlap significantly, if not fully, with that of Heraclitus.

44  δὲ βίος πόλεμος; cf. DK B48, DK B53 and DK B80.

45 Cf. Plutarch's quotation of the fragment (De E apud Delphous 393B = DK B91). Plutarch paraphrases this fragment at Mor. 559C.

46 SVF 1.141.

47 That the Stoics would find a natural connection between questions of psychology and respiration in Heraclitus is suggested by Aristotle's remark at De an. 405a25–7: ‘Heraclitus too says that the arche is soul, since it is vapour from which everything is composed. And it is very rarefied and in ceaseless flux.’

48 SVF 1.140.

49 On Hippocrates’ Epidemics 6.270.26–8 (SVF 2.782, LS 53E).

50 See Hierocles, Elements of Ethics 1.15–27. Plut. Stoic. Repugn. 1052F–1053A attributes such a view to Chrysippus.

51 ὁ δὲ ψυχὴν λογικὴν καθολικὴν καὶ πολιτικὴν τιμῶν οὐδὲν ἔτι τῶν ἄλλων ἐπιστρέφεται.

52 Gill (n. 7), 200–7.

53 ἡ δὲ τῆς ἰδίας διανοίας αἰδὼς καὶ τιμὴ σεαυτῷ τε ἀρεστόν σε ποιήσει καὶ τοῖς κοινωνοῖς εὐάρμοστον καὶ τοῖς θεοῖς σύμφωνον.

54 Gill (n. 7), 205–6.

55 Gill mentions the Alcibiades (128e–130c), the Republic (611d–612a) and the Phaedo (68a–69d, 78d–84b). He considers this feature in Aristotle in his Personality in Greek Epic, Tragedy, and Philosophy (Oxford, 1996), 356–83.

56 Cf. 10.11 and 3.11. While the latter does not specifically mention theōria, the process described there (methodically examining each thing one encounters and its relationship to the cosmos as a whole) makes it clear that the process imagined of elevating one's mind is largely the same.

57 Cf. 9.29: χειμάρρους ἡ τῶν ὅλων οὐσία: πάντα φέρει … μὴ τὴν Πλάτωνος πολιτείαν ἔλπιζε, ἀλλὰ ἀρκοῦ, εἰ τὸ βραχύτατον πρόεισι, καὶ τούτου αὐτοῦ τὴν ἔκβασιν ὡς μικρόν τί ἐστι διανοοῦ.

58 I take it that this is complementary with Gill's understanding of the Meditations as combining, at a fundamental level, the twin themes of ethical self-improvement and the transient flux of human nature, particularly in relation to death: Gill (n. 14), xxxiv–lii.

59 Cf. 2.11, 6.10, 6.24, 7.32 and 8.25.

60 Annas (n. 12), 108–10.

61 ταὐτὰ οὖν ἐκείνῳ, ἐν νόσῳ, ἐὰν νοσῇς καὶ ἐν ἄλλῃ τινὶ περιστάσει: τὸ γὰρ μὴ ἀφίστασθαι φιλοσοφίας ἐν οἷς δήποτε τοῖς προσπίπτουσι μηδὲ ἰδιώτῃ καὶ ἀφυσιολόγῳ συμφλυαρεῖν, πάσης αἱρέσεως κοινόν. πρὸς μόνῳ τῷ νῦν πρασσομένῳ εἶναι καὶ τῷ ὀργάνῳ, δι᾽ οὗ πράσσεις.

62 While it is true that atoms are not explicitly mentioned here, there is every reason to believe that the atomist world-view is what Marcus has in mind. A parallel use of σκεδάννυσθαι is found within the atoms/providence disjunction at 6.4; σκεδασμός occurs within such a disjunction at 6.10, 7.32 and 10.7.

63 Cf. 4.21.

64 εὐφροσύνη ἀνθρώπου ποιεῖν τὰ ἴδια ἀνθρώπου, ἴδιον δὲ ἀνθρώπου εὔνοια πρὸς τὸ ὁμόφυλον, ὑπερόρασις τῶν αἰσθητικῶν κινήσεων, διάκρισις τῶν πιθανῶν φαντασιῶν, ἐπιθεώρησις τῆς τῶν ὅλων φύσεως καὶ τῶν κατ᾽ αὐτὴν γινομένων.

65 Sedley (n. 8), 399 helpfully notes the first chapter of Alcinous’ Didaskalikos (152.10–11), where the objects of knowledge are explicitly said to be intelligible and fundamentally stable. This Platonist position, then, is directly at odds with Marcus’ view of the transience of ousiai.

66 See, for example, 4.3, 6.10, 6.24.

67 Rist (n. 4), 36–9 takes the pessimism which this entails to be the primary instance of Heraclitus’ influence on Marcus’ world-view.

68 Cf. 4.10.

69 See Gill (n. 14), lxxii.

70 Farquharson (n. 21), 823.

71 Sedley (n. 8), 405 n. 2.

72 See, for example, 6.10.

73 This is reminiscent of the familiar Stoic argument that the whole is never inferior to its parts (Sext. Emp. Math. 9.85).

74 Farquharson (n. 21), 824–31 details the issues and scholarly controversies of the late nineteenth and early twentieth centuries related to this passage. Most we may safely set aside.

75 Elemental transformations, in addition, are caught up in τὸν τοῦ ὅλου λόγον. What are they in addition to? I suggest it is the dissolved atoms themselves, as will become clearer below; however, it is somewhat opaque how those that take the view that Epicurean atomism is in mind here could construe this in their favour, unless we were to take this phrase in a remarkably weak manner.

76 So Farquharson (n. 21), 825, citing Lucr. 3.1076.

77 Farquharson (n. 21), 735–6 discusses the main textual issues. δὲ μόνα is Casaubon's correction of δαίμονα, ἐτεῇ is Usener's suggestion for the manuscripts’ ἔτι εἰ and ἔτι ἢ.

78 See, for example, Diogenes of Oenoanda (fr. vi Williams): Δημόκριτος τὰς ἀτόμους μόνας κατ᾽ ἀλήθειαν εἰπὼν ὑπάρχειν ἐν τοῖς οὖσι, τὰ δὲ λοιπὰ νομιστεὶ ἅμαντα. Cf. Sext. Emp. Math. 7.135.

79 In this way, I take it that Marcus’ judgement is along the same lines as what Aristotle holds at Ph. 8, 252a32–b5, on Democritus’ failure to establish a strong enough determining ἀρχή over and above pointing to what was prior to any given event. On this reading, Democritus accepted causal determinism in some form but failed to extend this analysis beyond something akin to a theory of regularity; see Barnes, J., The Presocratic Philosophers (London, 1982), 430–2Google Scholar.

80 See 4.27 for another example of a limited sort of atomism being contrasted with providence. There the atomist universe is a ‘medley’ (κυκεών) but an ordered one (κόσμος). Contrast 6.10.

81 See Philo, Aeternitate Mundi 2.497 M (EK F 99b). See also SVF 3.27 (Diogenes Bablonius) and SVF 3.5 (Zeno Tarsensis).

82 εἴτε ἀιδίοις ἀμοιβαῖς ἀνανεουμένου, cf. Heraclitus, DK B90: πυρός τε ἀνταμοιβὴ τὰ πάντα καὶ πῦρ ἁπάντων ὅκωσπερ χρυσοῦ χρήματα καὶ χρημάτων χρυσός. The Heraclitean formulation of the opposed positions on conflagration is prefigured by τροπή at the start of (2); cf. Heraclitus, DK B31a. The Heraclitean flavour of this passage is often noted, but little interpreted; see Kirk, G.S., Heraclitus: The Cosmic Fragments (Cambridge, 1962), 337Google Scholar.

83 This analysis owes a great deal to the exposition of Sedley (n. 8), 400–1. Note too that both the body and the soul are depicted as undergoing continuous flux. Farquharson (n. 21), 396 takes the ‘peculiarly qualified individual’ to be a concession to the Epicurean view, but this seems to ignore Plut. On Common Conceptions 1083C–D. For more on the Stoic response to the ‘Growing Argument’, see Sedley, D., ‘The Stoic criterion of identity’, Phronesis 27 (1982), 255–75CrossRefGoogle Scholar and Bowin, J., ‘Chrysippus' puzzle about identity’, OSAP 24 (2003), 239–51Google Scholar.

84 Some of the worry commentators have about this passage (see Gill [n. 14], lxxiii, who references Cooper [n. 3], 346–51) is that Marcus appears to raise for examination different models of the natural world which support his ethical conclusions without ever making it clear why the Stoic position must be adopted. Rather Marcus proceeds as if he has indeed demonstrated the Stoic position. 8.25–6 suggests an answer to Gill, but the larger problem with such a worry is that it illicitly expects an explicit argument. If we agree that Marcus is writing for himself and his own self-improvement, we should not think he spelled out every argument he assented to. On the account I have given, this is where the appeals to memory become relevant. Marcus is piecing together the tools he needs to satisfy his own enquiry into Stoic philosophy; he is not attempting to persuade a sceptical audience.

85 Annas (n. 12), 112.

86 See Diog. Laert. 7.39–41.

87 See τί οὖν ταράσσῃ; at 9.39.

88 There is something of G.E. Moore's ‘here is one hand’ in this. See his ‘Proof of an external world’, Proceedings of the British Academy 25 (1939), 273–300. As Moore appeals to the certainty that one is looking at her hand, Marcus falls back on the certainty of her access to her hēgemonikon. In both cases, the examples are given as more certain than one's belief in the arguments of an opponent. In Moore's case, the sceptic's; in Marcus’, the atomist's.

Save article to Kindle

To save this article to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle.

Note you can select to save to either the or variations. ‘’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

Available formats

Save article to Dropbox

To save this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Dropbox account. Find out more about saving content to Dropbox.

Available formats

Save article to Google Drive

To save this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Google Drive account. Find out more about saving content to Google Drive.

Available formats

Reply to: Submit a response

Please enter your response.

Your details

Please enter a valid email address.

Conflicting interests

Do you have any conflicting interests? *