To illustrate the modern controversy it is comparatively easy to refute the suggestion that for example ‘her coming will be a good thing’ expresses excitement at a friend's arrival. It might instead be a parent's unemotional assessment of the prospect of her daughter's friend visiting.6 But Chrysippus’ analysis is both less easy and more easy to attack: less easy because it is so much more subtle involving not only a second judgement but also reflection on the difference between goodness and preferred indifference (Chapter 12 below) and between judgement and appearance. At the same time it is more easy to attack because it is more rigorous in form committing itself rightly or wrongly to a single formula rather than offering illustrations.
PART I: Emotions as Judgements versus Irrational Forces
6: Posidonius on the Irrational Forces in Emotion Galen's Report
The Stoic Chrysippus regards emotions we have seen as consisting in evaluative judgements. There is the judgement that something good or bad is at hand and the judgement that it is appropriate to react. I want to suggest that Posidonius writing in the same school 150 years later offers a brilliant critique. He raises often in sharper form the kind of questions that have been put in the modern philosophical debate. For now too the thesis has been defended that emotions are evaluative judgements1 and now too objections have been raised that judgements are not necessary or sufficient2 that something else is required in addition perhaps desire3 or that emotions are not reducible to judgement desire or anything else.4 But the ancient debate was much sharper and more thorough than most of the modern one for the simple reason that Chrysippus specified exactly which two judgements constituted emotion in his view. This made it possible for his opponents to offer precise and detailed counter-examples and for his defenders to elaborate his analysis to meet the problems. It has been a major complaint about modern versions of the judgemental analysis that they have not specified the precise judgements they had in mind.5 Some offer only examples some not even that. As a result objections to them have had to be correspondingly vague.
Although I do not believe that the modern debate has been equally sharp there may have been a comparable debate in the fourteenth century. Simo Knuuttila has told me of debates in Adam Wodeham which parallel those between Zeno and Chrysippus on the role of judgements and those between Posidonius and Chrysippus on the role of assent complete with counter-examples. I look forward to his book describing these.
I believe it is necessary to assess the impact of Posidonius’ objections to Chrysippus’ analysis if we are to assess—as I shall try to (starting in Chapter 10)—the efficacy of Chrysippus’ therapy.7
Chrysippus’ fellow Stoic Posidonius (135–51 BC) writing 150 years later offered a very different viewpoint. He thought Chrysippan intellectualism had taken a wrong turn in forgetting Plato's teaching that besides reason and its judgements there were irrational forces at work in the soul. Plato had spoken of two irrational parts of the soul. Posidonius normally calls them capacities (dunameis)8 and says that one the thumoeides is concerned with anger and domination the other the epithitmētikon with lower appetites. Stoicism would never understand the emotions if it forgot Plato's point and the emotions were central to everything else in ethics.
Plato in an image that influenced Freud9 had compared reason and the two irrational forces with a charioteer and two horses10 and had used the psychodynamic language of tugging (helkein anthelkein).11 Posidonius repeats the analogy including the reference to the psychodynamic tug (holkē).12
Before turning in the next two chapters to Posidonius’ particular objections to Chrysippus I should sketch his general viewpoint a subject which will recur in Chapter 17 when I consider how it may differ from that of Galen who reports him.
Only when the cause of emotions is understood can we understand what is to be desired (orekta) or avoided (pheukta) or what exercises (askēsis) should be given to children or why the impulse in emotion depends on imagery.13 Indeed the whole process of education depends on our understanding human psychology.14 Moreover the emotions occupy a central position in the explanation. No other ancient philosopher makes emotions so central. In Posidonius’ words:
I think that the examination of things good and evil the examination of goals (telē) and the examination of virtues depend on a correct examination of emotions.15
Posidonius’ own explanation of emotions depends on his understanding of Plato's division of the soul into three. He agrees the capacities are three in number16 and he says that understanding them binds all the doctrines of ethics as if by a single cord.17 We shall see that Posidonius is said to give Platonic names to the three capacities of soul: rational irascible and appetitive (logistikon thumoeides epmumētikon). And he calls the last two powers the emotional element (to pathētikon) of the soul. He describes Plato as divine admires him and ranks him first on this subject.18 In the first book of his On the Emotions he wrote something like an epitome of Plato's discussion of how to train the rational and irrational parts of the soul in children.19 Posidonius thinks that very different training is needed for the irrational and emotional elements in the soul:
For it is necessary both that this [the rational element] should acquire understanding of the truth and that the emotional movements (hat kata pathos kinēseis) should be blunted through habituation to good practices if one is going to display a person with a better character.20
The emotional movements which it is here necessary to train are described in the preceding lines as being involuntary although we can be trained to control them. They depend on physiognomy and on the mixture of chemicals that make up our bodies.21
The implications for education are further spelt out in what immediately follows in line with the epitome of Plato that Posidonius wrote. One must have foresight as Plato said about the seed from which the child springs and then about the regimen of the pregnant mother as regards her food drink exercise rest sleep waking appetite and anger.22 This will help to make the emotional element of the soul amenable to the rational. The rational should be able by age 14 to take control following the imagery used by Plato in the Phaedrus like a charioteer controlling the two horses of appetite and anger:
This [rational element] is small at first and weak but finishes up large and strong around the fourteenth year by when it is right for it like a charioteer to take control (kratein) and rule (arkhein) over the pair of horses naturally conjoined with it appetite (epithumia) and anger (thumos).23
From this Posidonius concludes again that two different types of training are needed:
The education and virtue of this [rational] element is understanding of the nature of things as that of the charioteer is understanding of the instructions (theōrēmata) for driving chariots. For understanding (epistēmē) does not get generated in the non-rational capacities of the soul any more than in the horses. For these the proper virtue accrues from a kind of non-rational habituation (ethismos alogos) for charioteers from rational instruction (logikē didaskalia).24
The non-rational habituation will include the musical modes which have an effect quite independently of any words on children with too much or too little high spirit:
We shall prescribe for some people a regimen in rhythms scales and exercises of one sort for others another sort as Plato taught us. We shall rear the dull heavy and spiritless (athumoi) in sturdy (orthioi) rhythms scales which stir the soul strongly and similar exercises but those who are too spirited (thumikōteroi) and dart around too madly in the opposite.25
An official British document of 1905 agrees Handbook of Suggestions for the Consideration of Teachers. It considers that music has an effect independently of words on the emotional tendencies of children:
By wisely chosen songs the natural play of the healthy emotions of childhood can find an expression at once ample and controlled… but it is not necessary that infants should understand all the words they sing as the chief appeal is not to the intellect the training of which is the purpose of almost every other subject in the curriculum but through the spirit of the song to the unconscious mind of the child.
There is also we earlier learnt from Posidonius a kind of habituation presumably for the rational element of dwelling in advance (proendēmein) on the unwelcome things that might happen so that as and when they do we shall not find them unfamiliar. The result of this can be that the emotional movements occur not at all or only a little. The value of familiarity is that it prevents our beliefs being ‘fresh’—something whose relevance to emotion Chrysippus left unexplained.26
The distinction of capacities enables Posidonius to draw conclusions not only for training but for virtues and for the end (telos) to be pursued in life.27 As regards virtues we must distinguish unlike Chrysippus between the virtue of the rational element which is understanding (epistēmē) and the virtues of the irrational capacities.
Similarly with the end (telos) of life: Zeno or Cleanthes is said to have defined this as living in conformity with nature.28 But the nature with which we are to conform is the more divine element in the soul not the more beast-like.29 Posidonius’ own definition of the end we learn elsewhere covers both elements of the soul. It is Mixing in contemplation of the truth and order of all things… in no way distracted by the irrational part of the soul’.30 Other definitions of the end are attacked.31
Posidonius’ distinction of capacities also enables him to clarify what is to be desired or avoided (orekta pheukta) namely wisdom not the pleasure and domination which are desired by the animal-like element of the soul.32
Posidonius’ repeated complaint against Chrysippus is that he cannot explain things. He cannot explain impulses being excessive (pleonazousa hormē).33 He cannot explain variations in emotional reaction as between different people.34 For example he cannot explain why the wise person is not emotional at possessing the greatest good virtue or why the person who is merely progressing in that direction is not emotional at lacking that good.35 He cannot explain why emotions fade with time although judgements remain intact.36 He cannot explain the unruliness of tears which come unbidden or fail to come when called.37 He cannot explain how music can control emotions.38 He cannot explain why children go to the bad because he thinks that they have an affiliation (oikeiousthai) with virtue so that moral corruption must be due to outside factors. But he overlooks the affiliations of the emotional powers of the soul39 and their emotional movements.40
The attack on Galen as a source for Posidonius
The main evidence for Posidonius’ view comes from a much later author Galen. In Chapter 17 I shall consider how Galen differed from Posidonius and shall say more about both of them. But now I need to consider Galen for a different reason that his credibility as a reporter of Posidonius has been attacked. One author says:
Too many scholars have been victims of this intrepid polemicist of his sectarianism and persuasive ardour. It is he who transforms Posidonius into a Platonic philosopher and presents as ‘contradictions’ in Chrysippus things which his successor saw only as aporiai or puzzles to be got out of.41
Another author speaks of Galen's interpretation of Chrysippus as a highly partisan and misleading picture of the Stoic debate.42 Again Galen has been accused of ‘significant distortion’ and ‘unenlightening polemic’ so that he is ‘obviously suspect’ and only his direct quotations from Chrysippus can be used.43
Certainly some caution is needed. Galen is a polemicist intent on attacking Chrysippus in favour of Plato. But he is an intelligent and skilful polemicist who fastens on the details of Chrysippus’ strategy and wording.44 And he is not an indiscriminate Platonizer: he doubts for example a central tenet of Plato's the immortality of the soul.45 If we go too far in dismissing our main source of evidence for Posidonius on the emotions we shall be left with only our own ideas about what Posidonius as a Stoic must have said. Indeed I think this has happened when modern conceptions of Stoic orthodoxy have been used to decide what Posidonius must have thought.46
Criticism has also been extended to Posidonius. ‘Posidonius’ doctrines’ it has been said ‘do far less than Chrysippus’ to promote understanding of what passion is and its relation to reason.’47
The picture I should like to put in the place of this is a quite different one. Chrysippus went as far as anyone could in making plausible the idea that emotions consist in evaluative judgements. Correcting his predecessor Zeno who made emotions result from but not consist in judgements he was able to go further in one respect than his modern counterparts. He was able drawing on Zeno to specify precisely which two judgements are involved in all emotions so that the theory becomes highly testable. Emotions all involve a judgement of good or bad and the judgement that it is appropriate to react. This theory has the permanent value of showing us two points at which we may seek to attack unwanted emotions. In the end the theory will not work partly (not only) because there are cases at least at the margins which will not fit. It is to the further credit of Chrysippus and Zeno that they foresaw some of these cases brought them out and suggested some answers. Posidonius was to put forward an importantly different view of the emotions and at the same time to attack Chrysippus’ theory at some of its most vulnerable points. Some of his attacks would raise questions for modern versions of the idea that emotions are judgements.
Other ancient authors too come out of this analysis well. In Seneca we can for the first time see clearly how Chrysippus’ view could be defended. Whether these clarifications are Seneca's own contribution or whether he is making clear what had already been said we should in either case be grateful to him. Finally Galen himself can be taken as a polemicist who often has some basis for complaint in the text he is attacking.
First objection to Galen: gratuitous fault-finding
I shall address four of the objections to Galen's reporting. The first is that his complaints against Chrysippus are gratuitous and that this is particularly so when he accuses Chrysippus of contradicting himself on emotions occurring without judgement or reason. On one view there is no incompatibility to be complained of.48 I have argued in Chapter 3 that on the contrary there is a real incompatibility between Chrysippus’ account of emotion as mistaken judgement and Zeno's account of it which Chrysippus seems to present so favourably as akratic disobedience to recognized truth. If Chrysippus is to be defended it needs to be shown that Chrysippus was not endorsing Zeno. But the quotations from Chrysippus do not make this clear and I have argued in Chapter 3 that Posidonius thinks he was.49 Seneca's later attempt at reconciling Zeno with Chrysippus being written in Latin would not have been available to Galen.
Second objection: who is orthodox Posidonius or Chrysippus?
A second objection against Galen's reporting turns on the concept of orthodoxy. It has been argued that Posidonius is too orthodox a Stoic to abandon Chrysippus’ view that emotions are judgements of reason. He cannot have reverted to the Platonic view that there are two non-rational parts of the soul called appetitive and irascible or collectively the emotional part which play a role. Galen therefore must be disbelieved.50 This objection must also be addressed.
The concept of orthodoxy has a forward-looking and a backward-looking face. We tend to think of Chrysippus as representing Stoic orthodoxy because we look forward and see how influential he was. But if we look backwards as Posidonius would have done to see who is most in line with preceding views it is not necessarily Chrysippus who looks orthodox.
There is a further point. The first three Stoic heads Zeno Cleanthes and Chrysippus were all open to reinterpretation. Moreover Chrysippus was thought to have wavered in his views. All this made it easier for Posidonius to think of himself as faithful to the first two. Many of the relevant points have been made by Ian Kidd.51
Posidonius thought of both Zeno52 and Cleanthes as being on the side of the Platonists. The point about Cleanthes is well known. For Posidonius argues that Cleanthes divided the soul in Plato's way by quoting a passage in which Cleanthes represents Reason and Anger (logismos thumos) as distinct entities talking to each other.53 Galen doubts if Zeno is equally Platonist and admits his position is not obvious54 which is incidentally another sign that Galen is not himself an indiscriminate Platonizer. But the oral character55 of Zeno's teaching on the subject may have left room for doubt. Certainly Posidonius claims that Chrysippus was out of line with both Zeno and Cleanthes in not referring to an emotional part of the soul to explain the abatement of emotion.56
Galen implies that Chrysippus is unorthodox when he describes Posidonius as praising the ancient (palaion) view and rejecting Chrysippus’ idea that the soul has only a rational not an appetitive and an irascible element.57 In line with this Galen describes Chrysippus as using the arguments of an alien school (allotria hairesis) and not the hypothesis that belonged to his own (oikeia hupothesis).
This illustrates how easy it is for an ancient philosopher to represent his own views as the orthodox ones whatever he wants to say. It was standard practice to present oneself as orthodox where modern philosophers might prefer to claim credit for novelty. As for Chrysippus there would be no necessity to express agreement with him since he was not viewed by the Stoics themselves as a founder figure. (David Sedley has argued that it is only with founder figures that disagreement has to be disguised.58)
There is more. We find Galen and Posidonius arguing that Chrysippus after all agrees with them despite himself—another typical strategy in ancient philosophy. Galen's claim is that Chrysippus contradicted himself between different books. In his work On the Soul as opposed to the main treatise under discussion On the Emotions Chrysippus is said to assign love to the appetitive capacity (epithumētikē) and anger to the irascible (thumoeides).59 In fact this is merely Galen's interpretation. All Chrysippus actually does is to leave the distinction among the soul capacities unchallenged while he argues that at any rate they cannot be assigned to different parts of the body.60
Posidonius also uses Chrysippus as a witness (martus) in favour of his own views. Chrysippus is quoted in order to argue that he is committed to agreeing that emotions fade because of satiety. Moreover with satiety reason makes its way in and finds space as it were. This appears to treat reason as a distinct entity which is displaced by others at the height of emotional turmoil.61 Of course we are not told the context of Chrysippus’ remarks. Here too he could have avoided contradiction if he made the emotions that subside in the face of reason to be themselves acts of reason. But it is implied that he did not.
Another Stoic who may have diverged from Chrysippus is Posidonius’ teacher Panaetius. There is evidence quite widely accepted that Panaetius treated impulse as a capacity distinct from and opposed to reason:
For the capacity (vis) and nature of minds is twofold. One part is placed in impulse (appetitus) which in Greek is hormē and which snatches a person hither and thither. The other part is placed in reason which teaches and explains what is to be done and avoided. So it comes about that reason is in charge and impulse submits.62
Panaetius is also reported as saying that
temperance holds in control the disturbed movements of the mind which the Greeks call pathē and makes the impulses (appetitiones) which they call hormai obedient to reason.63
Someone may object that there is nothing un-Chrysippan here. For the impulse which in Panaetius’ account is contrasted with and subjected to reason may itself be viewed in Chrysippus’ manner as an act of reason. After all it is a Stoic orthodoxy that reason and impulse are distinct capacities.64 They must be distinct since they develop at different ages.65 Since this in no way prevents impulse in adult humans from being an act of reason in Chrysippus why it may be asked should not impulse be an act of reason in Panaetius too? Still this is not the impression created by Cicero's silence when he presents Panaetius’ impulse as not merely distinct from reason but opposed to it without any hint that it is an act of reason.
This is not to say that Panaetius’ distinction of capacities already anticipates the much more Platonic one of Posidonius. It is a twofold distinction and makes no use as Posidonius does of the terminology of Plato's threefold division. But it reveals once again that considerations of orthodoxy did not require agreement with Chrysippus.
Third objection: do emotions involve or are they identical with judgements?
The third objection to Galen's reporting exploits the fact that Posidonius does after all allow a major role to judgements in emotion. This is perfectly true but the important question is whether Posidonius accepts Chrysippus’ view that the emotions are simply identical with judgements. I doubt if the texts in which Posidonius gives a role to judgements go as far as this. But first we must look at Galen's report and at the evidence which has been thought to conflict with it.
Posidonius completely departed from both opinions. He does not think the emotions (pathē) are judgements (kriseis) nor supervenient (epigignesthai) on judgements but in everything he follows the ancient account and thinks they are produced by the spirited and appetitive capacities.66
So far this might suggest that emotions never involve judgements. But according to Plutarch or pseudo-Plutarch if we can believe him67 Posidonius regarded appetite fear and anger as based on judging and apprehending (en krisesi kai hupolēpsesin):
Posidonius says that some afflictions (pathē) are (i) mental and others (ii) physical and some (iii) though physical not mental are found in (peri) the mind while others (iv) though mental not physical are in (peri) the body (i) Mental without qualification he says are those based on (literally those in) judging and apprehending (ta en krisesi kai hupolēpsesin) for example appetites fears anger. (ii) Physical without qualification fevers chills fattening thinning. (iii) Physical but in the mind lethargy melancholy bites (dēgmoi) appearances expansions (diakhuseis). (iv) Contrariwise mental but in the body trembling pallor and changes of appearance in fear or distress.
Further connections between emotion and judgement are suggested by Posidonius’ reference to thinking (putare) in the definition of anger which Lactantius ascribes to him and which had probably been recorded by Seneca:
an appetite for punishing the person by whom you think (putes) you are unjustly harmed.68
The conflict in these sources is best explained I believe if Posidonius denied judgement in some cases of emotion but saw judgement as being present in the standard cases. And this fits exactly with the detailed criticisms by Posidonius which Galen records some of which cite examples where judgement is missing but others of which presuppose the presence of judgement. If emotions are standardly characterized by judgements this will be why Posidonius chooses this way of picking them out for contrast with bodily afflictions like fever. As for Posidonius’ definition of anger as involving thoughts about harm (putes) this word will cover the fact that anger normally involves judgement though sometimes only an appearance or feeling ‘as if’. As regards Galen's description of Posidonius it is misleading only to the extent that he is putting in a rather unqualified way the actual state of affairs which is that for Posidonius the emotions are not identical with judgements as Chrysippus supposes. This is not to deny that the emotions still typically involve judgements.
I think that scholarship has been tempted to ask too black-and-white a question: does Posidonius regard all emotions or no emotions as involving judgement? If I am right the truth is that he sees them as standardly but not always involving judgements and as not identical with judgements.
Fourth objection: would not an emotional element in the soul prevent eradication of emotion and viewing indifferents as indifferent?
It may be thought that Galen must be wrong to ascribe to Posidonius Plato's two emotional powers of the soul. For Posidonius would then have to abandon the Stoic belief to be discussed in Chapters 13 and 14 that emotion can be eradicated (apatheia) and (it might seem) the belief to be discussed in Chapter 12 that everything except character and rationality can be viewed as indifferent.
The first response to make is that we need to look at the evidence whether or not Posidonius did share these beliefs. He is against our living our whole life in accordance with emotion zēn kata pathos.69 Moreover Galen tells us if emotion is defined in Chrysippus’ way as an irrational movement of the soul contrary to nature Posidonius agrees that it will not be found in the souls of refined people (asteioi).70 But it is very significant that the asteios is said to be free of emotion only if emotion is taken as something contrary to nature. For Posidonius rejects that definition since as pointed out above he regards our nature as including an emotional element.71 Neither point then endorses Chrysippus’ view that (apart from such exceptions as the eupatheiai) we should have no emotion at all and the second point actually calls it in question. Posidonius does mention Chrysippus’ view on apatheia but again does not endorse it in a dialectical passage directed against Chrysippus’ comparison of mental and bodily health. There is no bodily analogue Posidonius complains of freedom from emotion (apatheia).72 But he does not say whether he shares Chrysippus’ belief in such apatheia. John Cooper offers a further telling point which is more decisive still. Posidonius describes some people as lacking in anger dull and sluggish (athumoi ambleis nōthroi). They will need rhythms and scales to stir the soul up.73 This implies that Posidonius actually does not want them to be emotionless. There is more. Zeno will have wanted people to be free of pathos because he defines pathos as excessive. But it was argued in Chapter 3 that his concept of pathos did not cover all or most of emotion but only those Medea-like emotions which involve disobedience to reason. Posidonius would then have had a precedent in thinking that apatheia or freedom from pathos involves only freedom from emotions which are excessive.
Qualifications to the ideal of freedom from emotion may also be present in Posidonius’ teacher Panaetius. Admittedly he far more explicitly than Posidonius endorses some kind of freedom from emotion. For to judge from Cicero's report Panaetius speaks of being free (liber vacandum) from all perturbation of the mind (perturbatio being Cicero's translation of pathos).74 Yet he also talks of making the impulses (hormai) obedient to reason and of treating what other people think glorious as small (parva).75 Hence there is much to be said for the suggestion that for Panaetius freedom from pathos consists in no more than moderating emotion by reason. And it turns out this is the view of Panaetius which Aulus Gellius actually takes.76
So far the evidence on apatheia actually supports Galen's attribution to Posidonius of an emotional element in the soul. But what about the evidence on indifferents?
This evidence is mixed. Two reports in Diogenes Laertius and one in Epiphanius treat Posidonius and Panaetius as dissenting from Chrysippus on indifferents. But the claim that Posidonius put wealth and health in the class of things good not indifferent is contradicted in Seneca's much fuller report and probably by Cicero's story about Posidonius’ reaction to an attack of pain in the joints which is in any case relevant. Discussing the general subject of indifference Posidonius met attacks of pain by exclaiming ‘It is no good pain although you are troublesome (molestus) I shall never concede that you are an evil (malum).’77 When Cicero reports that Panaetius in a letter on enduring pain refrained from denying that pain is an evil we cannot accept Cicero's inference that Panaetius accepted pain as evil. For Stoics often tailor their advice to the person addressed. With grief Chrysippus himself recommends attacking the idea that it is appropriate to indulge rather than saying that everything is indifferent.78 And there is a more extreme example of this technique in Seneca's On Consolation to Marcia which not content with passing over in silence the indifference of her bereavement actually emphasizes its magnitude by comparing the virtues of other sons who died prematurely.79 But this is done for purposes of persuasion and in fact it provides a very effective preparation for other techniques.
The further report that both Posidonius and Panaetius rejected the claim of virtue to be sufficient for happiness and desiderated health strength and provisions (khorēgia) as necessary could be true although it gets no support from Cicero.80
But what if Posidonius did accept to a greater or lesser extent the view that most things are indifferent? This would in no way impugn Galen's ascription to him of an emotional part of the soul. Even the sage will need to use the spirited and appetitive powers of his soul and not only when he engages in eupatheia. Simply in his correct selection of indifferents he must avoid the listlessness which we have seen Posidonius condemns. Of course in engaging in emotion Posidonius’ sage is not erroneously taking indifferent things to be good or bad. For Posidonius simply does not accept that Chrysippan account of what an emotion is. Nor is there any reason why he should since as was argued in Chapter 3 Zeno did not accept that definition of emotion either.
From the book: