I

The Stoics commonly maintained that only virtue (ἀρετή) and what participates in it is a genuine good (ἀγαθόν), and that, therefore, virtue is the necessary and sufficient condition for happiness (εὐδαιμονία). According to Diogenes Laertius, Posidonius took a different stance. First, Diogenes tells us that Posidonius also counted wealth and health among the goods (7.103 [F 171]):Footnote ³

Later Diogenes states that Posidonius as well as his teacher Panaetius maintained that virtue alone is not enough to achieve happiness (7.127–8 [F 173]):

Kidd rightly remarks that these two passages are complementary: ‘If virtue is self-sufficient for happiness it must be the only “good”; if there is need of other factors such as wealth and health, then virtue cannot be self-sufficient’.Footnote ⁴ However, Kidd dismisses both passages as spurious because they seem to be contradicted by Seneca, who in one of his letters reports an argument by Posidonius that shows why wealth and health should not be counted as goods.Footnote ⁵ According to this argument, health and wealth might also have negative effects, whereas a true good should only benefit and be completely free of harm. Therefore, health and wealth can only be ‘advantages’ (commoda), that is, a special sort of ‘preferred indifferents’ (προηγμένα), but no genuine goods.Footnote ⁶ Judging from Seneca's account, Posidonius in this context followed the orthodox Stoic doctrine, while the passages in Diogenes suggest that Posidonius abandoned the traditional Stoic view. In the following, I argue that we should not take the apparent contradiction between these two testimonies as a reason to dismiss Diogenes’ report.

To begin with, let us consider what could have induced Diogenes to make the above-mentioned claims about Posidonius. If his account should indeed be incorrect, the reason for that may be, as Kidd suggests, some sort of confusion or misunderstanding over a Stoic technical term, over a philosophical argument or over some genuine Posidonian innovation.Footnote ⁷ That being so, even if one or both of Diogenes’ statements are strictly speaking false, they might still reflect in some form a doctrinal innovation of Posidonius. We do not need to assume that Posidonius explicitly spoke of wealth and health as ‘goods’. It suffices to suppose that he attributed to wealth and health some specific sort of value that was not in line with the orthodox Stoic view. This could have then induced Diogenes to report that he regarded them as goods.Footnote ⁸ Similarly, we should leave room for the possibility that Diogenes’ statement about Posidonius’ and Panaetius’ heterodox opinion on virtue is indicative of some innovation that took place in the Middle Stoic period—even if the statement might not be true in its literal sense.

To explore this possibility, it will be necessary to first discuss the broader philosophical context of the passages in question. The most important source for Posidonius’ ethics and psychology is Galen's De placitis Hippocratis et Platonis. As several passages in De placitis testify, Posidonius considered a correct understanding of the soul's powers and of the emotions as the key to all ethical doctrines.Footnote ⁹ The studies by Fillion-Lahille, Cooper and Gill have shown that Posidonius’ theory of emotions is in fact not, as Galen insinuates, a complete return to Platonic psychology, but is still based on the traditional Stoic tenet that emotions (πάθη) are effectively judgements—or the results of judgements—of the soul's rational faculty.Footnote ¹⁰ However, while Chrysippus held that the adult human soul is fully rational, Posidonius maintained that the soul also comprises certain non-rational faculties. According to Galen, he conceived of them not as different forms (εἴδη) or parts (μέρη) of the soul, but only as different faculties of a single substance (δυνάμεις μιᾶς οὐσίας).Footnote ¹¹ As Cooper has shown, Posidonius must have held that the energy involved in an emotional impulse comes not solely from the rational faculty but also from the affective movements (παθητικαὶ κινήσεις) that originate from the non-rational faculties. Although Posidonius does not claim emotions to be identical with these affective movements, he maintains that the affective movements contribute to the energy of the emotional impulse and may influence our judgements which then result in an emotion.Footnote ¹² Thus Posidonius occupies a middle ground between the Chrysippean and the Platonic position. He apparently thought that with his model, several phaenomena can be explained much better than with Chrysippus’ model, for example the excessive nature of the emotional impulse or the fact that an emotion subsides over time even if the underlying evaluative judgement remains the same.Footnote ¹³

This is of relevance to us because Posidonius’ theory of virtues is closely connected to his account of the soul and of its functions. The following passage in Galen illustrates this point, starting with a direct quotation from Posidonius (PHP 5.5.33–8 [F 31, part]):

According to Galen, Posidonius thought that reason becomes fully operational at the age of fourteen (2), which tallies with the orthodox Stoic account.Footnote ¹⁴ Until then, one should seek to prepare the children's non-rational faculties (here collectively called παθητικόν) for the future rule of the rational faculty (λογιστικόν) (1). In Posidonius’ simile, which is clearly inspired by Plato's Phaedrus, the non-rational faculties are likened to the horses of a chariot (3). They must not become either too strong or too week and need to be made obedient to reason (4). The training/education (παιδεία) and virtue (ἀρετή) of the rational faculty, which is represented by the charioteer, differs from that of the non-rational faculties in that it consists in knowledge (ἐπιστήμη) (5). The non-rational faculties, just like the horses of the chariot, instead of knowledge have their own proper virtue (οἰκεία ἀρετή) which can only be achieved through non-rational habituation (ἐξ ἐθισμοῦ τινος ἀλόγου), whereas the charioteer, symbolizing the soul's rational faculty, can be educated through rational instruction (ἐκ διδασκαλίας λογικῆς) (6).

That Galen here paraphrases a Posidonian text is, as Kidd observes, strongly suggested by the usage of indirect speech right after the direct quotation in (1).Footnote ¹⁵ In conjunction with some further remarks in De placitis, the passage can thus be taken to attest Posidonius’ distinction between rational and non-rational faculties of the soul.Footnote ¹⁶ We may want to be cautious about assuming that the total number of those faculties was necessarily three and that Posidonius applied Platonic names to them.Footnote ¹⁷ Obviously, Galen has an interest in making it seem as if Posidonius fell back onto a (pseudo-)Platonic model of the soul. Yet even if we grant that Galen, where he uses indirect speech, may have employed his own rather than Posidonius’ terminology, I do not see sufficient grounds for doubting the general authenticity of his report.

If we follow Kidd in accepting the passage as authentic, how does this further our understanding of the two passages in Diogenes Laertius? From Galen's account we learn that Posidonius not only distinguished between rational and non-rational faculties but also between two different types of virtue: virtue of the rational faculty, which consists in knowledge and can be acquired through rational instruction, and virtue of the non-rational faculties, which can only be acquired through non-rational habituation. Unfortunately, Galen does not explain this division of virtue in much detail, and he provides no concrete examples of either of the two types. Since the virtue of the non-rational faculties does not consist in knowledge, we may speculate that it involves some non-propositional content that cannot be fully expressed in language and therefore cannot be acquired through rational instruction. But since we have no explicit evidence for that, this remains a matter of conjecture. As we shall see below, more can be said about the process of non-rational habituation and the role of the virtue of the non-rational faculties. Nevertheless, due to the scarcity of our sources, many details of Posidonius’ theory have to remain unclear.

However, it is evident that this division of virtue marks a difference to the traditional Stoic doctrine. Earlier Stoics like Zeno or Chrysippus considered virtue as essentially rational and posited that it always consists in knowledge or prudence (ἐπιστήμη/φρόνησις).Footnote ¹⁸ In addition, Cicero and Seneca later regarded the view that virtue is perfected reason as standard Stoic doctrine.Footnote ¹⁹ Therefore, if Posidonius should have thought that also virtue that does not consist in knowledge, namely virtue of the non-rational faculties, is required for happiness, this would indeed mean that he regarded virtue in the traditional Stoic sense as insufficient for happiness.

Against this background, we should be careful to outright dismiss the passage in which Diogenes reports that Posidonius did not consider virtue as self-sufficient for happiness (7.127–8). It is possible that the Posidonian division of virtue found its way into Diogenes’ doxographic account in a somewhat convoluted form. Thus, the actual meaning of Diogenes’ statement might be that Posidonius—in contrast to earlier Stoics—did not consider virtue of the rational faculty as self-sufficient for happiness. This would fit in well with the fact that, as the passage in Galen attests, Posidonius demanded that all faculties of the soul, including the non-rational ones which were not acknowledged by earlier Stoics, need to be trained and be led to virtue.

Diogenes attributes the heterodox opinion on virtue not only to Posidonius but also to Panaetius. This makes it less likely that his account is merely based on a misunderstanding and does not reflect some actual doctrinal innovation. There are further remarks by Diogenes which in this context need to be taken into consideration. While Zeno and Chrysippus considered virtue to be the perfection of human rationality and therefore identified it with knowledge or prudence, Panaetius is reported by Diogenes to have distinguished between theoretical (θεωρητική) and practical (πρακτική) virtue (7.92).Footnote ²⁰ Diogenes also tells us that Hecato, another Middle Stoic and disciple of Panaetius, divided virtue into intellectual (θεωρηματική) and non-intellectual (ἀθεώρητος) virtue (7.90).Footnote ²¹ It remains unclear whether these classifications are in any way comparable or equivalent to the Posidonian division of virtue reported by Galen. It may well be, as is often supposed, that Panaetius’ distinction between theoretical and practical virtue was intended to be applied to the four cardinal virtues in the sense in which Cicero presents it in Off. 1.15–7.Footnote ²² Nevertheless, the evidence cited at least suggests that in Middle Stoicism, a more differentiated concept of virtue was developed which, in the case of Posidonius, led to the inclusion of virtues belonging to the non-rational faculties. This need not have been a deliberate break with the school's founders, but was likely intended as an expansion or improvement of the orthodox doctrine. Still, the philosophical consequences of this change for Posidonius’ ethics will require further elucidation.

II

I have suggested that the basis for Diogenes’ report in 7.127–8 is that Posidonius did not consider virtue of the rational faculty as self-sufficient for happiness. This does not yet answer the question of why Diogenes also tells us that Posidonius regarded health and wealth as necessary for happiness and allegedly even classified them as goods. Before I turn to this question, I want to address a possible objection. In contrast to the interpretation outlined above, Tieleman has argued that Posidonius’ distinction between rational and non-rational faculties is not an innovation of the Middle Stoic period, but in fact goes back to Chrysippus and the Early Stoa. The underlying dichotomy, Tieleman claims, consists in the difference between the soul's mental and physical aspect. On Tieleman's interpretation, Posidonius’ introduction of non-rational faculties and his corresponding division of virtue only explicates further what was already present in the early Stoic model.

For the Stoics, the soul is of course a body, namely pneuma in a particular tension.Footnote ²³ Tieleman refers to PHP 4.6.2–3 to claim that according to Chrysippus, some wrong actions are due to false judgements while others are due to the soul's lack of tension.Footnote ²⁴ From there, it might seem like a small step to distinguishing between one type of virtue which ensures the soul's physical tension, and another type of virtue which consists in knowledge and belongs to the mental or intentional aspect of the soul. Thus Tieleman thinks that the distinction between theoretical and practical, and intellectual and non-intellectual virtue attested for Panaetius and Hecato, as well as Posidonius’ division of virtue and his distinction between a λογιστικόν and a παθητικόν of the soul, are essentially just reformulations or small expansions of the orthodox Stoic doctrine.Footnote ²⁵

This interpretation supposes that Chrysippus distinguished between a physical and a mental (or intentional) aspect of the soul, and that this distinction forms the basis for certain developments in the Middle Stoic period. We should be careful, however, not to blur all the differences between Posidonius and Chrysippus. In Posidonius’ model, the movements of the non-rational faculties exert an affective pull (παθητικὴ ὁλκή) that may influence the rational faculty in its judgements.Footnote ²⁶ As a result, Posidonius sees the non-rational faculties as potential sources of evil—due to their ability to cause our rational faculty to err. Allegedly, he even censured Chrysippus in this context, because he thought the latter was unable to provide a convincing explanation of how evil or vice (κακία) arises (PHP 5.5.9–11 [F 169, part]).

The potential of the non-rational faculties to cause evil also shines through in Posidonius’ definition of the end in life (τέλος). Traditionally, the Stoics defined the end as ‘living in accordance with nature’, although different generations of Stoics produced different variations of this formula.Footnote ²⁷ Posidonius, however, defined the τέλος slightly differently to other Stoics. Clement of Alexandria reports the following definition by Posidonius (Strom. 2.21.129.4 [F 186, part]):

Apart from the small inconsistency concerning the expression ‘part’ (μέρος) of the soul where ‘faculty’ or ‘power’ would be more appropriate,Footnote ²⁸ this formula illustrates very well that according to Posidonius the soul's non-rational faculties have their own causal efficacy. The two parts of the Posidonian formula correspond (1) to the λογιστικόν and (2) to the παθητικόν (or ἄλογον) of the soul.Footnote ²⁹ We may assume that to achieve both the goals presented in the Posidonian definition of the end, one must acquire both types of virtue: in order to contemplate and promote the truth and order of all things, one needs the virtues associated with reason. However, one will only be able to act reasonably and live a harmonious life if one also possesses the virtues belonging to the non-rational faculties, otherwise those faculties may influence reason in a negative way and lead it astray.

It should thus be obvious that for Posidonius, the παθητικόν or ἄλογον has its own causal efficacy and needs to be controlled through certain virtues. In contrast to that, Chrysippus’ doctrine is commonly understood in a completely different way. The proper tension of the soul, which is equated with the soul's physical health (ὑγίεια) or strength (ἰσχύς), is not construed by Chrysippus as a separate faculty, able to exert its own influence over reason. Rather, the soul's strength or proper tension can be identified with one's adherence to a harmonious and consistent set of beliefs. Conversely, the soul's weakness or lack of tension can be identified with one's adherence to a conflicting and inconsistent set of beliefs. To say that someone is liable to falter in their judgements due to the weak τόνος of their soul would be to say that they are liable to do so due to the inconsistency of the belief system they entertain. In Chrysippus’ model, the soul's lack of tension can only be seen as responsible for errors and emotions in so far as it is itself the physical manifestation of certain mental states and processes, not in so far as it constitutes a separate non-rational faculty.Footnote ³⁰

This is not to say that Posidonius rejected the doctrine of the soul's physical tension. Since there is no evidence to the contrary, he must have granted, as all Stoics did, that the soul is a body and that our impressions, judgements and virtues can in principle all be explained in physical terms. This is a direct consequence of Stoic corporealism, according to which the soul, just as all other things that exist, has to be a body. However, the differences between the Chrysippean and the Posidonian model remain. Posidonius’ non-rational faculties have the potential to misguide reason and to cause evil; they cannot be compared to the physical aspect or tension of the soul.

In addition, it seems questionable whether we should even attribute to Chrysippus a clear distinction between the soul's mental and physical aspect. In fact, the Stoics in their terminology did not clearly distinguish between the two and often used ambiguous terms to describe the functions of the soul.Footnote ³¹ This further corroborates the impression that for the Stoics, the mental and the physical aspect coincide, which makes it futile to compare them to Posidonius’ λογιστικόν and παθητικόν. With this in mind, we should not try too hard to harmonize the Chrysippean and the Posidonian position. We have to acknowledge that in attributing causal efficacy to separate non-rational faculties, Posidonius did not follow the orthodox Stoic doctrine.

III

Once we acknowledge the decisive role the παθητικόν plays in Posidonius’ ethics, the rest of Diogenes Laertius’ remarks in 7.103 and 7.127–8 should become easier to understand. As I have pointed out in Section I, Posidonius believed that the non-rational faculties cannot be trained or educated in a rational way. Rather, these faculties need to be moulded through some sort of non-rational habituation (PHP 5.5.33–8). According to a passage in Galen, Posidonius recommended to ‘dwell in advance’ (προενδημεῖν) upon things that might happen in the future so as to get used to them and to not get emotionally overwhelmed if they happen (PHP 4.7.7–9 [F 165, part]). It seems plausible that this ‘dwelling in advance’ is supposed to play a role in moulding the non-rational faculties and developing the virtues that are necessary to control them. But there is more to be found in Galen's report. Apparently, Posidonius also argued that certain musical modes and rhythms are appropriate for moulding the children's παθητικόν.Footnote ³² And more generally, Posidonius thought that one's whole environment and bodily condition has an influence on the affective movements of the soul; he maintained that certain bodily features correspond to certain traits of character. For instance, someone who is broad-breasted (εὐρύστερνα) and warmer (θερμότερα) is also more given to anger (θυμικώτερα). In addition, a person's character differs depending on the area they live in (κατὰ τὰς χώρας); and in general, the affective movements of the soul always follow the condition of the body (τῶν παθητικῶν κινήσεων τῆς ψυχῆς ἑπομένων ἀεὶ τῇ διαθέσει τοῦ σώματος).Footnote ³³

If Posidonius advocated such a doctrine, it is easy to see why he may have been committed to attributing special value to external things like health and wealth. Evidently, health is a bodily condition and as such, according to the above argument, influences the soul's affective movements. We may surmise that it can have a positive effect on the non-rational faculties, making them easier to control. In the case of wealth, the influence may be less direct than in the case of health. However, if a person is wealthy, they should in principle be able to adjust their environment so that it has a positive effect on their physical well-being and consequently on their affective movements. In this sense, also wealth can be regarded as potentially influencing the non-rational faculties of the soul. This might be the reason why according to Diogenes, Posidonius valued health and wealth more highly than other Stoics did. But should we go so far as to say that Posidonius considered health and wealth as necessary for happiness and as genuine goods?

Kidd firmly rejects this possibility.Footnote ³⁴ He refers to PHP 5.6.17–19 (F 161), where Galen talks about what Posidonius took to be akin or proper (οἰκεῖος) to the rational and the non-rational faculties respectively. According to that passage, Posidonius thought that emotions arise when people mistake things that are proper to the non-rational faculties (οἰκεῖα ταῖς ἀλόγοις δυνάμεσι) for things that are proper without qualification (ἁπλῶς οἰκεῖα). In fact, only things belonging to our rational and divine aspect (τοῦ λογικοῦ τε καὶ θείου), such as wisdom (σοφία), are proper without qualification, whereas things like pleasure or power over one's neighbour, which belong to the animal-like aspect of the soul (τοῦ ζῳώδους τῆς ψυχῆς), are only proper to the non-rational faculties. Kidd argues that for Posidonius, things like health and wealth must fall into the second category. They may be regarded as οἰκεῖα to the non-rational faculties but not as ἁπλῶς οἰκεῖα—and only things that are ἁπλῶς οἰκεῖα can be genuine goods.Footnote ³⁵

The same argument can be applied to Posidonius’ division of virtue and his definition of the end. Given Posidonius’ distinction between οἰκεῖα and ἁπλῶς οἰκεῖα, also the virtues of the non-rational faculties cannot be regarded as absolute goods. They seem to be responsible only for making the soul's affective movements controllable by reason and in that sense are subordinate to the virtues of the rational faculty.Footnote ³⁶ If we follow this line of reasoning, it becomes clear that for Posidonius, absolute goods can only be affiliated with one's rationality.

However, even if Posidonius did not regard things like health and wealth as goods in an absolute sense, it should be obvious that the habituation that is required to make the non-rational faculties obedient to reason must in some way be affected by or even dependent on one's environment and bodily condition. Regardless of how exactly Posidonius envisioned the process of non-rational habituation, it would be inconsistent for him to maintain that environment and bodily condition exert an influence over the non-rational faculties, while at the same time maintaining that we can mould those faculties independently of our environment and bodily condition. If a person is healthy and wealthy, this must influence the affective movements of their non-rational faculties. As a possible result, those movements may become less frequent or less intense and hence easier to control. Thus, even if health and wealth are not ἁπλῶς οἰκεῖα and therefore cannot be unconditionally beneficial, they may still be seen as conducive to happiness in so far as they have the potential to make it easier for the λογιστικόν to operate undisturbed by the affective movements of the παθητικόν. I hence conclude that based on Posidonius’ tenets, health and wealth must at least have some instrumental value for achieving happiness.

Such an interpretation is not only well compatible with our sources, it also has the advantage that we do not have to completely dismiss Diogenes’ report. Diogenes tells us that Posidonius regarded health and wealth as necessary for happiness and as goods. I suggest not taking this report too literally, while still assuming that there has to be some basis for Diogenes’ claims. And indeed, if Posidonius regarded health and wealth as potentially conducive to happiness and therefore had to assign some instrumental value to them, this could have induced Diogenes to report in a simplified way that Posidonius regarded health and wealth as goods. In other words, if we grant that Diogenes was somewhat imprecise in his report, his remarks can be reconciled with the rest of our evidence and need not be completely rejected.

If this reconstruction of Posidonius’ doctrine is correct, things like health and wealth that potentially enhance our control over the non-rational faculties gain a status that cannot be compared to the usual status of Stoic indifferents (ἀδιάφορα).Footnote ³⁷ In Stoic texts, health and wealth are usually given as prime examples of promoted indifferents (προηγμένα). Thus, the Stoics would claim that, if circumstances allow, one should choose health and wealth over sickness and poverty. Yet they would also maintain that we achieve happiness through virtue alone, regardless of whether we actually attain specific indifferents such as health and wealth. This orthodox Stoic concept of indifferents does not seem compatible with Posidonius’ approach. Posidonius in this context seems to be closer to the Peripatetics than to the Early Stoics. We know from Cicero's account in De finibus of the lively debate between Stoics and Peripatetics on the use and value of external goods. While the Stoics argued that virtue alone is sufficient for a happy life and that it is impossible to further increase one's happiness, the Peripatetics claimed that the possession of external and bodily goods does make life happier (Cic. Fin. 3.43). We do not know to what extent Posidonius participated in that debate and exactly which position he took. Even if he, in contrast to orthodox Stoics, thought that health and wealth can help us to achieve happiness, he may still have defended the traditional Stoic doctrine that happiness does not admit of degrees, and that a person, once happy, cannot become happier by amassing further goods. I consider this a likely option, although there can be no certainty in these matters. In any case, if Posidonius was willing to attribute instrumental value to things like health and wealth, he was in a sense closer to the Peripatetic than to the orthodox Stoic position.

There remains the question of whether Posidonius also considered health and wealth as necessary for happiness. In principle, if health and wealth are of instrumental value for achieving happiness, this does not yet entail that they are necessary for it. But Diogenes suggests exactly that by stating that according to Posidonius and Panaetius, there is ‘need’ (χρεία) of health, wealth and strength (7.127–8). Since I do not see in Diogenes’ claim a direct contradiction to other sources, I believe that we should accept his testimony. The orthodox Stoic definition of goods, evils and indifferents does not seem fully compatible with Posidonius’ idiosyncratic approach. Posidonius may have regarded health and wealth as necessary for happiness even if he did not classify them as genuine or absolute goods. The passage in Seneca can be reconciled with such an interpretation—despite the fact that Kidd adduces it as the main reason why we should doubt the correctness of Diogenes’ report.Footnote ³⁸

Let us take a closer look. As Seneca tells us (Ep. 87.31–40 [F 170]), Posidonius argued that wealth (diuitiae) can potentially be a cause of evil (causa malorum)—not in the sense of a necessary, self-sufficient cause (causa efficiens) but in the sense of an antecedent cause (causa praecedens).Footnote ³⁹ Although wealth in itself does not harm, it may still incite us to evil by provoking false beliefs in us. Health (bona ualetudo) can be classified in a similar way. According to Seneca, Posidonius reasoned that both health and wealth can also have negative effects and therefore must not be confused with genuine goods. In contrast to things like health and wealth, goods are necessarily free of all harm and cannot influence us in a negative way.

Prima facie, Seneca's account might seem to contradict Diogenes’ report, in which we read that Posidonius counted health and wealth among the goods. However, as pointed out above, we should always consider the option of interpreting Diogenes’ statements in a less literal sense. The passage from Seneca highlights that Posidonius did not regard things like health and wealth as absolute goods. But Seneca also reports that according to Posidonius, health and wealth are ‘advantages’ (commoda) because their positive potential outweighs their negative potential.Footnote ⁴⁰ Thus Posidonius may still have thought that health and wealth—notwithstanding their negative potential—can in principle be conducive to happiness. There is no inherent contradiction in assuming that something has instrumental value and can help us to achieve happiness while holding that the same thing may also have negative effects. If we interpret the evidence in that way, the passages from Diogenes can be reconciled with Seneca's account and need not be completely rejected.

Also the other reasons Kidd gives for his dismissal of Diogenes’ report do not seem compelling to me.Footnote ⁴¹ The fact that according to Cicero, Posidonius maintained that virtue was the only good (Tusc. 2.61 [T 38]) does not preclude the possibility that Posidonius attributed instrumental value to things like health and wealth while not regarding them as goods in an absolute sense. And finally, Kidd's argument ex silentio seems the least convincing. After all, we have Diogenes’ report itself, and since it is possible to interpret Diogenes’ statements in such a way that they are compatible with the rest of our evidence, this solution is preferable to the outright dismissal of his report.