Chapter 4 - Autonomy

Summary

Chapter 4 discusses autonomy (or its lack) as a key element in characters’ fictional ontology. It divides the topic into three themes - freedom, revenge, and suicide - each of which plays a vital role in Seneca’s dual concept of personal and fictional independence. The first section addresses Stoic isolationism as an assertion of individual sovereignty and demonstrates that this trait is shared by Seneca’s sapiens and tragic characters alike. The second section examines revenge as, simultaneously, an amplification and limitation of characters’ autonomy. Finally, the chapter considers Senecan suicide as an expression of free agency in the midst of misfortune, charting its importance across both the prose and the dramatic works.

Forming an identity implies the ability to do so. In more profound terms: self-definition, of the kind discussed throughout this monograph, is predicated upon individuals’ relative autonomy and agency, their capacity to effect changes in their own lives and in the world around them, to act in accordance with their own wishes, to avoid undue social, bodily, or moral constraints imposed from outside. The obverse also holds, because compromised autonomy is typically accompanied by a disruption or diminution of personal identity, with the oppressed individual becoming, at worst, an object, a vessel, an instrument.

Autonomy constitutes a, if not the, major distinction between the categories of ‘character’ and ‘person’, because fictional beings claim no real capacity for self-determination, no matter how full or dominant their personalities may seem. Characters are trapped within texts, at the mercy of authors and readers alike, and have no contingent futures on which to exercise their powers of choice – they have no powers of choice. Instrumentality, object-status, is their inescapable fate, a point brought home all the more forcefully in Senecan tragedy where dramatis personae openly acknowledge their positions within pre-existing and concurrent literary traditions, their identities predicated not just on authorial invention, but also on the demands of genre and of prior poetic models. Against such a background, their notorious tendency for self-assertion may seem a pitiful mirage. However often Medea affirms herself as Medea, or Hercules as Hercules, they remain unable to alter themselves or their circumstances. The desire of so many Senecan protagonists to push forward and dare the undareable, do the impossible, acts as a foil to the agency they do not ultimately possess. They may rant as much as they wish: it won’t change anything.

That is, of course, a reductive view of the tragedies, and deliberately so, because it signals the impression conveyed by Seneca’s characters of violently insistent selfhood, pitted against all opposition, bursting through constraints and achieved whatever the cost to the world’s moral and social fabric. A combined love of force and control, along with ambitions of absolute sovereignty are some of Seneca’s characters’ most defining traits. Even their habit of self-citation, so patently metapoetic, can be interpreted as bullish solipsism and imperious self-confidence, a sense of importance so great that it evaluates itself in the third person, as though from the perspective of an awestruck spectator.¹ One question, therefore, is how to balance these two views of selfhood in the tragedies: do characters cede autonomy to their literary pedigrees, or do they pursue it in spite of – or by means of – the traditions lying in their wake? A second, and related issue is how to measure these tragic characters’ pursuit of autonomy and agency against the ideas set out in Seneca’s philosophical writings, because that will help elucidate the degree of importance and seriousness accorded to them in the tragedies. This does not mean entering upon questions about free will,² since those are, at base, irrelevant to fictional beings. Instead, this chapter focuses on the topics of personal and political sovereignty and freedom, and self-sufficiency as indices of an individual’s capacity for uninhibited action. Following an initial discussion of autarky in the prose and dramatic works, I examine how Senecan concepts of autonomy play out in adjacent themes of revenge and suicide. Bids for power and control, in Seneca, almost invariably lead to destruction, whether of others or of the self. The concept of sovereignty displayed in both the tragedies and the philosophy is essentially a zero-sum game in which freedom is achieved by rendering others subordinate and powerless. Self-definition and self-assertion come at a heavy price.

While autarkic selfhood is by no means a new topic in studies of Senecan drama, most of this scholarship comes from the field of English Literature rather than Classics,³ and there remains substantial scope for fresh investigation, especially on the theme of revenge, which has received surprisingly little treatment given its outsized role in Seneca’s tragedies, and more broadly still, in Seneca’s thought. The present chapter not only expands on existing ideas of Senecan autarkeia, but also brings a new dimension in the form of characters’ autonomy, that is, how Seneca’s interest in actual, human self-determination intersects with his characters’ awareness of their purely fictional ontology. Their pushing against the boundaries of their texts coincides with their impulse to pursue ever greater criminal acts as expressions of their unassailable agency. And their pursuit of sovereign control – over their own bodies and reputations, over events and other people – mirrors in unsettling ways the sapiens’ lonely reign in his self-conferred kingdom of virtue. This Senecan ideal of autonomy is distinctly problematic, not because of the difficulty involved in attaining it, but because it lays waste to its surroundings. Self-determination happens amid the ruins.

4.1 Freedom

Sages and Other Tyrants

To begin this investigation, we need to ask whether Seneca’s characters really can claim autonomous, autarkic selfhood, or whether it is a delusion born of their spiralling criminality. From a strictly Stoic point of view, it has to be the latter, because true freedom for Stoics lies in exercising ratio to make correct judgements, following the dictates of natura, and expunging the passions, all of which Seneca’s dramatis personae fail to do on a spectacular scale. Fired by anger, jealousy, fear, and greed, they assent to undertake monstrous acts and attribute immense value to the kinds of external concerns – power, wealth, reputation – that Stoics classify as ‘indifferent’.⁴ Seneca’s characters are in this regard slaves both to fortune and to their own unchecked desires, which can make their pursuit of autonomous selfhood seem like a wry joke on the part of their author, a negative exemplum of how misguided priorities cloud the mind.

This is correct up to a point, but Seneca’s celebration of autonomy also displays deeply negative elements, which, like his treatment of constantia, leave it open to misappropriation and misuse. Gordon Braden and Cedric Littlewood both detect in Seneca’s sapiens a marked tendency for solipsistic superiority, Braden comparing the wise man to the madman, and Littlewood comparing him to the tyrant.⁵ Each presents a valid argument, and the purpose of this section is to tease out their ideas in more detail, as a prelude to understanding the tragedies’ quasi-Stoic representation of autonomy.

Certainly, Seneca’s vision of autarky is extreme. He glorifies the sapiens as a figure of supreme independence and self-containment, someone who stands apart from regular human society – in a moral and, more often than not, in a physical sense – and someone whose indifference to the loss of possessions, family, or even body parts makes him untouchable. The sapiens aspires through virtue to equal the gods, and his moral outlook is such that he can never really be harmed, victimised, oppressed, or enslaved. This prospect of spiritual indomitability must have held deep appeal for disenfranchised members of society (Epictetus, we may recall, was a slave),⁶ and especially for the Roman elite, who felt disenfranchised by the principate.⁷ Yet the very fact of this appeal also points to a combative element in Stoic ethics, its encouragement of individuals to flout oppressive conventions and claim a kingdom for themselves, however immaterial that kingdom may be. Braden describes this aggressive trait as the residue of Stoicism’s Cynic focus, ‘a commitment to the self’s superiority to all public ambitions and intimidations’.⁸ Littlewood similarly identifies in the Roman Stoics an appetite for moral conflict and isolationism that puts them on a par with autocratic rulers.⁹ The resemblance does not go unremarked in the ancient world either, albeit meant in more positive terms. Diogenes Laertius reports the Stoic belief that, ‘not only are the wise free, they are also kings; kingship being unaccountable rule, which none but the wise can maintain’ (οὐ μόνον δ᾿ ἐλευθέρους εἶναι τοὺς σοφούς, ἀλλὰ καὶ βασιλέας, τῆς βασιλείας οὔσης ἀρχῆς ἀνυπευθύνου, ἥτις περὶ μόνους ἂν τοὺς σοφοὺς συσταίη, 7.122 trans. Hicks). Cicero expresses a parallel idea in the de Officiis, in the context of describing how some wise men retreat from public affairs: ‘they had the same aim as kings, to lack nothing, to obey nobody, to enjoy liberty, that is, essentially, to live as one pleases’ (his [philosophis] idem propositum fuit quod regibus, ut ne qua re egerent, ne cui parerent, libertate uterentur, cuius proprium est sic vivere ut velis, 1.69–70).¹⁰ Although this comment about kings may be no more than a gloss on the quality of philosophical freedom and the expectations it entails,¹¹ nonetheless Cicero constructs a troubling parallel between the sage’s self-sufficiency and that of an autocrat.¹² Theoretically, there is no danger in the Stoic living as he or she wishes, because what he or she wishes will always be virtuous, but the comparison with monarchs emphasises free agency and absolute self-determination over any moral concerns. By citing both sapientes and kings as the epitome of supreme freedom, Cicero highlights (presumably unintentionally) the possible negative consequences of Stoics transcending standard morality and not feeling the loss of social bonds. Such autonomy has the potential to become, or to seem like, hostile, arbitrary wilfulness.

Seneca, too, enjoys comparing the sapiens to an absolute ruler. ‘If you wish to subject everything to your authority, submit yourself to reason; you will rule many if reason rules you’ (si vis omnia tibi subicere, te subice rationi; multos reges, si ratio te rexerit, Ep. 37.4). Submission is the path to domination: even though this implies a degree of humility, Seneca’s notion of self-control rapidly extends to control over others (multos reges, not multa).¹³ The sapiens’ inward, self-reflexive concern for his own morality is unsettlingly comparable to a monarch’s institutionally sanctioned egotism. If, at base, an absolute ruler has subjects chiefly as material on which to exercise his power, the same applies to Seneca’s view of life’s moral and physical adversities.

Permutations of this sapiens–monarch binary are found throughout Seneca’s prose. Epistle 114.23 declares the soul a king (rex noster est animus), while de Beneficiis 7.6.2 avers that the wise man possesses everything ‘in the manner of a king’ (regio more) meaning that he has power over everything.¹⁴ In Epistle 108.13, Seneca reports that one of his teachers, Attalus, ‘called himself a king’ (ipse regem se esse dicebat) and that Seneca thought him ‘more than a king, because he was entitled to pass judgement on kings’ (sed plus quam regnare mihi videbatur, cui liceret censuram agere regnantium, trans. Gummere). Once again, terrestrial concerns usurp Stoic claims of metaphorical rulership. Attalus’ paradox that the wise man is a true king (because, presumably, he exercises strict dominion over himself and minimises his earthly needs cf. Ep. 108.14–15), becomes in Seneca’s mind the power of regulating public morals in the manner of a censor (censuram, Ep. 108.13). The authority acquired through Stoic renunciation aspires to civic supremacy; jurisdiction over the self is thought to justify jurisdiction over others.

It is telling that another permutation of this motif occurs in the Thyestes, where the second chorus lauds the simple life as ‘a kingdom [that] each man grants to himself’ (hoc regnum sibi quisque dat, Thy. 390). Like the paradox expressed by Attalus, the line may be taken as a comment on the benefits of Thyestes’ exilic poverty, to the effect that his quiet sylvan existence has conferred upon him a kingship more meaningful than his prior rule in Argos.¹⁵ Yet, no sooner do we recognise this quasi-Stoic attitude than Thyestes himself reveals an ingrained desire for the Argive throne, praising its wealth in terms that all but confess his greed.¹⁶ Does this contradict the chorus’ vision? Is the choral ode a foil for Thyestes’ later conduct? I would argue that the two inclinations – to rustic simplicity and to tyranny – are not as mutually exclusive as they seem. They are, rather, points on a spectrum of autarkic aspirations, since in each case Thyestes chases the freedom to live as he pleases. Senecan autonomy lends itself well to absolutist claims.

The same logic emerges from the anecdotal encounters between tyrants and their victims that pepper Seneca’s prose.¹⁷ In the context of praising self-sufficiency in de Tranquillitate, Seneca reports that a tyrant threatened the philosopher Theodorus with death and Theodorus replied, ‘you have … the right to please yourself, you have within your power only half a pint of my blood’ (habes … cur tibi placeas, hemina sanguinis in tua potestate est, Tranq. 14.3 trans. Gummere). Essentially, Theodorus limits the tyrant’s power by declaring himself unaffected by physical pain; the prospect of death is recast as a petty half-pint of blood. The philosopher’s autarky enables him to triumph over the tyrant, to hold sway over the tyrant (even if only in an abstract sense), and this, at base, is what Seneca’s anecdote celebrates: the sage slipping through the ruler’s grasp, proving that terrestrial absolutism is not so absolute after all. Seneca explores a similar idea in the very next story, of Julius Canus, whom Caligula had sentenced to death. Not only does Canus react calmly to the announcement of his impending execution, but, when the guards arrive at his house, they find him playing latrunculi, a battle-game somewhat like chess (Tranq. 14.4–8). The metaphor is clear, but Seneca spells it out anyway: ‘do you think Canus was playing a game? He was making mockery!’ (lusisse tu Canum … putas? Inlusit! Tranq. 14.8). The encounter between philosopher and emperor is figured as a competition in hegemony, a zero-sum game in which one achieves control by wresting it from one’s opponent. Canus’ composure belittles the emperor’s power, and even goes as far as allowing him to ‘win’ at the game of absolutism. Both the sage and the tyrant aspire to moral autonomy, but the sage does it better.

The most detailed example, and the final one I wish to consider in this section, is that of the quasi-Cynic, quasi-Stoic philosopher Stilbo, whose story Seneca tells in Epistle 9.¹⁸ Stilbo’s home city of Megara has been destroyed by Demetrius Poliorcetes, but the sage accepts his loss with equanimity:

Hic enim capta patria, amissis liberis, amissa uxore, cum ex incendio publico solus et tamen beatus exiret, interroganti Demetrio, cui cognomen ab exitio urbium Poliorcetes fuit, numquid perdidisset, ‘omnia’ inquit ‘bona mea mecum sunt’. Ecce vir fortis ac strenuus! ipsam hostis sui victoriam vicit. ‘Nihil’ inquit ‘perdidi’: dubitare illum coegit an vicisset. ‘Omnia mea mecum sunt’; iustitia, virtus, prudentia, hoc ipsum, nihil bonum putare quod eripi possit.

This man, his homeland captured, his children and wife lost, when he emerged from the general conflagration alone yet happy, and Demetrius, whose last name, Poliorcetes, referred to his destruction of cities, asked whether he had lost anything, this man said: ‘all my goods are with me’. What a brave and tough fellow! He vanquished his enemy’s victory. ‘I have lost nothing’, he said: he forced Demetrius to wonder whether he had actually conquered. ‘All my goods are with me’; justice, virtue, wisdom, in other words, he considered nothing that could be taken from him to be a good.

(Ep. 9.18–19)

Placed in the position of a victim, standing amid the rubble of conquest, loved ones gone, and brought face-to-face with his sardonic enemy, Stilbo vaunts his freedom. The narrative indulges in a cheeky bit of misdirection, designed to upset our assumptions about conqueror and conquered. Has Stilbo lost anything? Omnia … bona mea … mecum sunt. Demetrius, we may assume, could reasonably have expected to hear the first three of those words, but the last two come as a surprise. Stilbo’s solution to victimhood is to move the goalposts; in refusing to attribute true value to any of life’s externals, he empties Demetrius’ conquest of meaning. He also destabilises the conqueror’s claim to autonomous action, since Demetrius’ role as conqueror depends on Stilbo’s acceptance of victimhood, whereas Stilbo depends on nothing. The philosopher’s self-contentment at once reflects and exceeds his oppressor’s power, enabling him to triumph over Demetrius (ipsam hostis sui victoriam vicit) and to undermine his opponent’s sense of superiority (dubitare illum coegit an vicisset). Seneca imagines Stilbo as the equivalent of a bellicose tyrant, who, brave and tough (fortis ac strenuus), asserts his unassailable agency and self-determination in the context of violent desolation.¹⁹ Both figures in the anecdote are defined by their separation from social bonds, the one because he destroys them, the other because he functions despite their loss. Both claim the power not to be affected by the wreckage around them, both retain their self-possession (and in Stilbo’s case, this may be more than self-possession of the psychological/emotional kind, given the very real prospect of enslavement after one’s city has been captured).²⁰ In the confrontation between sage and tyrant, the sage emerges victorious, not just because he evades the ruler’s grasp, but because he beats him at his own game. This need to ‘win’, which comes across so strongly in Seneca’s accounts of Stilbo and Canus, seems to contradict the sage’s professed disregard for ‘indifferents’; why would Canus or Stilbo, and above all, why would Seneca care who emerges victorious?

It is this competitive dominance that gives the lie to Seneca’s vision of virtuous self-government. While one could argue that Stilbo represents the laudable moral equivalent of Demetrius’ sovereign independence, that very equivalence leaves Stilbo tainted by association. His interior hegemony of soul and spirit is less a foil to Demetrius’ external rulership than a version of it. Granted Stilbo is unlikely to raze a city – he is not about to become a second Poliorcetes – but Seneca’s military language indicates an aggressiveness embedded in Stilbo’s autarkic ideals, which are, of course, Seneca’s autarkic ideals. Essentially, Senecan autarky celebrates power and control so much,²¹ and celebrates it so persistently on the model of worldly autocracy, that it risks valuing absolutist tendencies over morally informed independence. Both Stilbo and Demetrius aspire in their various ways to dictate the very shape and meaning of the world around them; the similarity may present an enticing prospect for disempowered individuals, but it is also a moral problem for Seneca’s definition of freedom.

Nor is this problem solved by accepting Seneca’s sporadic distinction between kings as beneficent rulers and tyrants as cruel autocrats.²² According to this line of reasoning, if the soul is a king, and by extension the wise man is too, then however monarchical its rule, it is nonetheless founded upon virtue. Perceptive readers will have noted that the excerpt I cited above, from Epistle 114, was an opportunistically truncated version of the full passage, which reads: animus noster modo rex est, modo tyrannus. Rex, cum honesta intuetur, salutem commissi sibi corporis curat, et nihil imperat turpe, nihil sordidum. Ubi vero impotens, cupidus, delicatus est, transit in nomen detestabile ac dirum et fit tyrannus (‘our soul is at one time a king, at another a tyrant. A king, when it has regard for honourable things, cares for the health of the body in its charge, and gives no disgraceful, no base commands. But when it is uncontrolled, greedy, self-indulgent, then it changes into that detestable and dire term and becomes a tyrant’ Ep. 114.24). This seems at first blush to resolve the ambiguities of Senecan autonomy, by granting true authority and freedom only to those pursuing Stoic self-control. Philosophical self-government is virtuous; the passions, in contrast, usurp power like tyrants. But even here there is a snag that threatens to unravel Seneca’s logic: the verb imperat, which situates the virtuous soul in the realm of supreme military and political command. Senecan autarky is a form of imperium, all the more absolutist for being self-granted: imperare sibi maximum imperium est (‘command of the self is the greatest empire’, Ep. 113.31).²³ Although it can and has been argued that Seneca reinvents imperium as an internalised, intangible alternative to the principate’s expanding power,²⁴ still the term belongs to the discourse of autocracy and as such, it establishes a competitive relationship between sapiens and monarch, not separation. In the words of Gordon Braden: ‘imperium remains the common value, the desideratum for both sage and emperor.’²⁵ Time and again, Seneca’s concept of moral autonomy cannot break free from terrestrial, political paradigms. As a result, his definition of true, Stoic autarky begins to look a lot like its opposite, the false, immoral freedom claimed by tyrants.

Overlap between these two categories is therefore the main issue, and one that has important consequences for our understanding of Senecan tragedy. The wise man’s freedom from oppression, deprivation, the torments of appetite, is imagined on the model of freedom to oppress, deprive, torment. Because Seneca depicts the sapiens’ autonomy in terms of imperialist conquest and forceful domination it can easily be mistaken for the earthly independence exercised by those in power. The reverse applies as well: political autarky comes to resemble Stoic self-sufficiency. For Seneca’s dramatis personae, this means that even the most ruthless, unscrupulous pursuit of independence can take on Stoic colouring and lend itself to analysis in (at least partially) Stoic terms. If the would-be sapiens aspires to absolute control over his circumstances and indeed, over his opponents, then the same can be said of the tyrant. If the sapiens revels in his supreme isolation, celebrates unfettered individual agency, and regards his moral life as a self-conferred kingdom, then how different are the attitudes and aspirations of Seneca’s Atreus, or Medea, or Hercules? Like the sapiens, the characters of Senecan tragedy refuse to be dominated by others, they grasp at omnipotence, they exercise fierce (if misguided) self-control to achieve their desired ends. In this regard, it is not only valid to discuss their criminality in terms of autonomous selfhood; it is necessary.

All by Yourself

Before turning attention to the tragedies, however, I wish to consider one other, crucial aspect of Senecan autarky: solitariness.²⁶ For Seneca, solitude provides the right environment for self-determination and self-assertion, and these activities, in turn, breed yet more solitude. Social entanglements are often presented as damaging to virtue; Seneca urges withdrawal. He treats family relationships and friendships with a similar degree of detachment: specific individuals are replaceable, and the true sage will avoid any interdependence likely to expose him to their loss. The sapiens’ insulation from all contingency means that he stands alone, exercising supreme subjectivity and sovereignty, godlike not just in his virtue, but in his invulnerability and capacity for self-directed action. Once again, these characteristics form an important background for Seneca’s dramatis personae.

Retreat from public life is such a varied and pervasive theme in Seneca’s writings²⁷ that the following discussion does not aspire to comprehensive coverage. Rather, I examine the specific issue of how solitude affects and protects the autonomy of the Senecan self. At the core is Seneca’s concept of self-sufficiency, which is fundamentally introspective, and thus presents a substantial deviation from earlier, Ciceronian traditions of personal autonomy coupled with political involvement.²⁸ For Seneca, public life represents perilous enslavement to other individuals, to the pursuit of wealth, power and influence, to the many endless and (from a Stoic perspective) pointless demands of the workaday world. There may seem nothing remarkable in this – similar opinions can, for instance, be found in Lucretius and in Roman satire – but Seneca’s portrayal is distinctive for its emphasis on subjectivity and personal sovereignty. A brief glance at the de Brevitate Vitae, for instance, shows men devoted to business and politics becoming objects of passional forces (both grammatically and figuratively): alium … tenet avaritia, … alium mercandi praeceps cupiditas … ducit; quosdam torquet cupido militiae (‘greed ensnares one, reckless desire for trade propels another; passion for war torments others’ Brev. Vit. 2.1). Public life requires one’s subordination to other, more powerful people (Brev. Vit. 2.2) while throngs of clients curtail one’s freedom on a spiritual as well as physical level (Brev. Vit. 2.4). Men embroiled in such situations are ‘never able to return to their true selves’ (numquam illis recurrere ad se licet, Brev. Vit. 2.3). In the helter-skelter of public life, Seneca concludes, ‘no-one belongs to himself’ (suus nemo est, Brev. Vit. 2.4). By contrast, Seneca elsewhere describes Stoic autarky as an act of self-ownership: te dignum putas aliquando fias tuus (‘you think yourself worthy of at last becoming your own master’ Ep. 20.1); ubicumque sum, ibi meus sum (‘wherever I am, I am my own master’ Ep. 62.1). Epistle 75 concludes with a particularly forceful version: absoluta libertas [est] … in se ipsum habere maximam potestatem. inaestimabile bonum est suum fieri (‘absolute freedom [is] … holding supreme power over oneself. Being master of oneself is a priceless good.’ Ep. 75.18).

The upshot is that the Senecan self rarely, if ever prospers in the public sphere.²⁹ Withdrawal from social and political duties is carried out less with the aim of forming one’s own alternative social group (like an Epicurean circle of friends) than for the sake of cultivating the self’s inner sanctum, a lonely and self-absorbing task. Implicit throughout Seneca’s descriptions is the idea that public life interferes with one’s capacity for self-government, hence his use of reflexive language to explain the philosopher’s autarky. While Seneca’s propensity for reflexive phrasing has often been remarked as a novel development in self-awareness and self-care,³⁰ I believe its chief purpose is more grammatical and political than ethical or aesthetic. First, it celebrates the philosopher’s unfettered subjectivity by making him the subject and object of his own action; he is under his own dominion, not another’s, and with the responsibility of such self-monitoring comes the freedom of being in charge.³¹ Second, many of Seneca’s reflexive phrases originate from the juridical language of ownership: vindica te tibi (‘claim yourself for yourself’, Ep. 1.1); suum esse (‘to be master of oneself’, see Ep. 20.1; 62.1; 75.18, above); se habere (‘to possess oneself’ Ep. 42.10; Brev. Vit. 5.7).³² By implication, the philosopher’s judgement is superior to the institutional control exercised by courts, and it duly supplants them. These phrases affirm the philosopher’s supreme will and power by insisting on his inviolable claim to be his own person. When the Senecan philosopher withdraws into his introspective domain, he forsakes worldly institutions only to set up superior versions within his own soul. Like the competition between philosopher and absolute ruler, explored in the preceding section, Senecan ideals of seclusion invest the sage with unbridled personal sovereignty.

Thus, philosophical autonomy exceeds terrestrial power in the act of its retreat as well as in moments of confrontation. It is, for instance, this idea that motivates Seneca’s portrait of Augustus in section 4 of the de Brevitate Vitae, where the princeps is described as longing to retire from government and enjoy contemplative otium: hoc labores suos … oblectabat solacio, aliquando se victurum sibi (‘he would make his work pleasant via this consolation, that one day he was going to live for himself’ Brev. Vit. 4.3). Reflexive language brings us back to the self-contained realm of the Senecan sapiens, whose access to autonomy and solitude outstrips even that of the world’s most formidable ruler. Augustus’ lonely post at the peak of Rome’s hierarchy is nothing compared to the philosopher’s self-appointed aloofness.

One last example should suffice to clinch my point. Epistle 14 contains one of Seneca’s many exhortations to shun public life, in this case chiefly with the aim of evading vim potentioris (‘the violence of the stronger’ Ep. 14.4). The worst of all our terrors, Seneca maintains, originates ex aliena potentia (‘from other people’s power’ Ep. 14.4), which he proceeds to depict in terms of mob violence, torture, and public execution (Ep. 14.4–6). Besides illustrating the intemperate nature of worldly might, these examples are significant for turning human individuals into depersonalised objects, literally dividing them into limbs and fluids. The philosopher, however, escapes such violation through a combination of physical retreat and spiritual inviolability: ‘let us withdraw into ourselves in every way’ (undique nos reducamus, Ep. 14.10 trans. Gummere). Once again, reflexive language positions the philosopher as arbiter of his own personal circumstances, while the evocation of the immaterial inner realm, in contrast to the tangible facts of bodily penetration, affirms the philosopher’s ultimate unassailability; no-one can reach this private, internal region, not even the fiercest tyrant. Withdrawal from public life is accompanied by withdrawal behind the barriers of one’s spirit. This is the best, indeed the only, method of asserting one’s subjectivity, whereas full engagement with the social world will only lead to one’s enslavement and oppression, whether physical or spiritual or, most likely, both.

Similar assertions of subjective control inform Seneca’s views on friendship, albeit in more moderate fashion. While it would be wrong to think that Seneca denies the value of having friends – on several occasions, he actually affirms their importance for a full and joyous human life (e.g. Ep. 9; 19.10; 48.2; Ep. 109) – nonetheless his Stoic beliefs involve a certain amount of indifference. One must not grieve for a friend’s death or absence, and such bereavement will not in any way affect the sage’s happiness, nor will it curtail his ability to function as a self-sufficient individual and paragon of virtue.³³ One friendship can be substituted for another (Ep 9.5–6) and even though friends are not exactly interchangeable – since memory ensures their distinctness³⁴ – the value of their company and converse will be the same as that with any other sage.³⁵ As is apparent in Seneca’s Stilbo anecdote, the specifics of personal attachment must be approached with equanimity, no matter how inherently worthwhile such attachments are as a facet of lived experience.³⁶ Significantly, Seneca argues against the Epicurean view that friendships are formed for the purpose of help and comfort:

Sapiens etiam si contentus est se, tamen habere amicum vult, si nihil aliud, ut exerceat amicitiam, ne tam magna virtus iaceat, non ad hoc quod dicebat Epicurus … ‘ut habeat qui sibi aegro adsideat, succurrat in vincula coniecto vel inopi’, sed ut habeat aliquem cui ipse aegro adsideat, quem ipse circumventum hostili custodia liberet.

The wise man, even though he is self-sufficient, nonetheless wants to have a friend, if for no other reason than to practise friendship, so that his great virtue does not lie idle, not for the reason Epicurus states … ‘so that he should have someone to sit by him when he is sick, to come to his aid when he has been cast into chains or has become poor’, but so that he should have someone by whose sickbed he may sit, and whom he himself may free from the surrounds of hostile imprisonment.

(Ep. 9.8–9)

There is a strong altruistic element to this: friends should not be self-serving nor should amicitia be purely transactional because at some point the transaction will fail its recipient. But in elaborating this principle, Seneca also emphasises the sapiens’ agency and control: he is the one looking after the friend, the one freeing the friend, just as, on a more abstract plane, he frees himself. Again, the wise man occupies a superior position, a position of active subjectivity as opposed to the status of passive object. Although the sentiment is well meant, it is not hard to see how such assumption of control reinforces notions of the philosopher’s supreme sovereignty.³⁷ Not only does he evade the tyrant’s grip and the potential degradation that accompanies much social activity, but he also slips through the knot of interpersonal interdependence, for relying on another individual exposes one to contingency, which a true Stoic will, of course, transcend.

The self comes before the friend in Seneca’s thought, and although this is not a selfish principle per se, it should give us pause, nonetheless. At the close of Epistle 6, Seneca quotes with approval Hecato’s summary of moral progress, ‘I have begun to be a friend to myself’ (amicus esse mihi coepi, Ep. 6.7). This is progress indeed, Seneca remarks, because such a man ‘will never be lonely’ (numquam erit solus, Ep. 6.7) and will be ‘a friend to everyone’ (hunc amicum omnibus esse, Ep. 6.7). The idea is that only proper self-government allows one to be a proper friend and good global citizen; one must secure one’s own moral basis first, before benefitting others. But the reflexive language of self-friendship, coupled with the assurance of self-sufficient solitude, suggests that the chief beneficiary is the philosopher himself, who maintains subjective control and secluded autonomy even in contexts of social exchange. Whether making friends or losing them, the sapiens appears a lonely figure, self-directed and self-determined.

Along with friendship and seclusion, there is another crucial component of Senecan autarky that requires consideration, a component with substantial ramifications for the tragedies as well: divinity. In Stoic thought, the wise man is equal to a god: an entity of perfect reason, in tune with natura, above Fortune, needing nothing beyond itself. Although intended to elevate and celebrate human aptitude for virtus, this concept shares with Seneca’s other thoughts on autarky the capacity to be twisted in less scrupulous directions, as its affirmation of supreme agency and solitude can be co-opted all too easily into the service of megalomania. If the philosopher aspires to godhead, so do the selfish and the power-hungry, the imperialists and the madmen. Granted their motives and means of achievement are the antithesis of Stoic virtus, but their desire for control and invulnerability – the rewards of divinity – resemble the Stoic’s in arresting, sometimes disturbing ways.

As befits his interest in absolute self-government, Seneca emphasises the concurrence between sapiens and deus. ‘This is what philosophy promised me: to make me god’s equal’ (hoc enim est quod mihi philosophia promittit ut parem deo faciant, Ep. 48.11). In Epistle 31.9, Seneca assures Lucilius that he, too, ‘will rise equal to god’ (par deo surges) if he takes nature as his guide. Whoever attains flawless reason deos aequat (‘is on par with the gods’ Ep. 92.29). In fact, the sapiens can even be said to outstrip divinity inasmuch as he achieves rationality via his own efforts, rather than merely embodying it, as the Stoic god does. Divinity, by its very nature, cannot partake of evil, but the wise man emerges superior even to this level of perfection because he can recognise moral evil and overcome it.³⁸ ‘There is a way in which the wise man surpasses god: god fears nothing because of nature’s favour; the wise man because of his own’ (est aliquid, quo sapiens antecedat deum: ille naturae beneficio non timet, suo sapiens, Ep. 53.11). Another permutation of the idea occurs in the de Providentia, where Seneca advises his imaginary interlocutor to bear misfortune bravely: ‘in this respect you surpass god; he is beyond suffering from evils, you are above it’ (hoc est quo deum antecedatis: ille extra patientiam malorum est, vos supra patientiam, Prov. 6.6). In each case, it is the sage’s capacity to act on his own behalf that places him above the static, unchanging essence of the deity. Active achievement of perfection is presented as more impressive than perfection itself. On the basis of such claims, it is not surprising that later interpreters of Seneca, from Augustine to modern scholars, sometimes accuse him of hubris,³⁹ and although mistaken, this reaction is evidence of the megalomaniacal potential embedded in Seneca’s theology. The issue is of course more complex than mere disdain for divine power; it is about the philosopher’s self-conferred independence, its challenge to godhead being a virtually incidental consequence, albeit one that is liable to misuse.

Agency and autarky are the key themes in Seneca’s portrait of the quasi-divine sapiens. Not only does the sage attain the freedom enjoyed by god, but he attains it actively and self-reflexively, suo beneficio (Ep. 53.11, above), a result of the jurisdiction he exercises over himself. Seneca quotes approvingly Sextius’ view that ‘Jupiter has no more power than the good man’ (Iovem plus non posse quam bonum virum, Ep. 73.12), where posse evokes raw potential for action as well as the authority that accompanies and guarantees such potential.⁴⁰ The core meaning of Sextius’ claim is that the sapiens and Jupiter are equally capable of bestowing benefits and forgoing external possessions, which makes them equally complete in happiness. No sooner is the comparison made, however, than Seneca avers the sage’s superiority in the matter of possessions, because while ‘Jupiter cannot use them, the wise man does not want to’ (quod Iuppiter uti illis non potest, sapiens non vult, Ep. 73.14); the sapiens’ act of willing ranks him above Jupiter’s abstention by default. This sense of superiority even originates from the sapiens himself (hoc se magis suspicit; ‘in this regard he esteems himself more’, Ep. 73.14), which makes his eclipse of divine power entirely self-directed.

In a related vein, the sage resembles a god in his invulnerability.⁴¹ He is impervious to injury, physical or psychological, and remains unaffected by loss. This means, Seneca affirms in de Constantia Sapientis, that he is ‘a next-door neighbour to the gods, and resides closest to them, like god in everything except mortality’ (vicinus proximusque dis consistit, excepta mortalitate similis deo, Const. 8.2). Once more, the main issue here is autonomy, of which divinity represents the apex. A divine being is, in Patrick Gray’s terms, impassable, that is, not susceptible to being acted upon.⁴² The definition is particularly useful for thinking about de Constantia section 8, where gods and wise men are portrayed as exempt from object status. Their inability to receive injury – because perfect ratio does not allow for the existence of such a category – epitomises their broader freedom from submission, oppression, others’ control. God and sage, like sage and tyrant, represent a duo of sublime subjectivity and self-determination. Better yet: the sage gains the upper hand on both of these counterparts, because unlike the tyrant he is not subject to contingency and unlike the god, he actively generates his own conditions of self-government.

The Stoic sage’s proximity to the divine also reinforces his solitariness, most obviously because the Stoic god is not part of a pantheon, but also because the sapiens’ singular virtue enables him to transcend the rabble and its earthly preoccupations. Epistle 9 likens the philosopher’s self-sufficiency during times of hardship to Jupiter’s calm acceptance of ekpurosis: sage and god both retreat into themselves, yield themselves to quiet contemplation (Ep. 9.16–17). The comparison stresses the sapiens’ untouchability, the self-containment that insulates him from worldly shocks, and makes him an essentially lonely figure even when there are other people in his life. Just as the Stoic divinity does not depend on anyone, and has no need of anything, so the Stoic philosopher aspires to a sublime level of freedom, the price of which is isolation.

Tragic Freedom

Following this lengthy but important (de)tour through Senecan autarky, I return now to the tragedies, specifically, to discussion of how Seneca’s dramatis personae envisage and pursue freedom. The aim of the preceding two sections, besides providing expositional material, was to argue that Seneca creates accidental equations between autarky acquired through virtus and ratio, and its opposite, the irrational, immoral autarky of the tyrant or egoist. The existence of such parallels allows for – one might even say, encourages – analysis of autarky in the tragedies. It prompts us to take Seneca’s characters more seriously, and not just dismiss their desire for autonomy as a parody or inversion of Stoicism, since Seneca’s Stoicism encompasses that inversion already, in its very definition.

A key example in this regard is Seneca’s Hippolytus, who resembles a proficiens in his dual aspiration to self-mastery and independence. When the Nurse encounters him in a forest glade, halfway through Act 2, she urges him to exchange his lonely chastity for the joys of youthful love (Phaed. 435–82) and he replies with an encomium on the wholesome pleasures of life in the woods (Phaed. 483–564). Against the Nurse’s vision of human intercourse, Hippolytus sets the freedom afforded by seclusion and simple needs. Granted these themes are declamatory and poetic commonplaces, they are also, in this context, reflections on Senecan autarky. The Nurse sets the tone via her cheeky appropriation of Stoic discourse, concluding her praise of sexual pursuits by commanding Hippolytus to ‘follow nature, life’s guide’ (vitae sequere naturam ducem, Phaed. 481).⁴³ Stoic concepts are also unmistakably present elsewhere in her speech, even though they have received no scholarly attention. Thus, for instance, she speaks of ‘the proper duties god has allotted’ to the different stages of human life, namely that ‘joy befits youth and a grim brow old age’ (propria descripsit deus / officia … / laetitia iuvenem, frons decet tristis senem, Phaed. 451–3), where the collocation of propria, officia, and decet cannot help but recall Stoic notions of decorum/τὸ πρέπον (‘appropriateness’) and καθῆκον (‘fitting behaviour / proper function’). She enunciates an even more explicitly Senecan form of Stoicism when she tells Hippolytus, ‘I am anxious with worry about you, because, hostile, you discipline yourself with harsh punishments’ (anxiam me cura sollicitat tui, / quod te ipse poenis gravibus infestus domas, Phaed. 438–9). Though a negative attribute from the Nurse’s perspective, Hippolytus’ self-control (te ipse … domas) epitomises the reflexive subjectivity of the Senecan sage: his withdrawal from human commerce frees him from being the object of another’s power and leaves him to shape his life as he sees fit. Further, Seneca’s choice of the verb domare does double duty in evoking, on the one hand, Hippolytus’ desire for authority and power, and on the other, his eventual, fateful similarity to a wild beast.⁴⁴ His aggressive self-control will end in violence; withdrawal, in this case, spells destruction.

Hippolytus’ reply picks up on this question of autarky and pursues it with a vengeance. Life in the forest is, he maintains, ‘free from hope and care’ (spei metusque liber, Phaed. 492); the forest-dweller ‘serves no kingdom’ (non ille regno servit, Phaed. 490), nor does he chase in vain after wealth and honour (Phaed. 491); he knows nothing of crime (Phaed. 494–5) nor, more importantly, of lies (Phaed. 496). In all respects, he is his own master, his autonomy being simultaneously a freedom from oppression and a freedom to act as he wishes. Echoes of Seneca’s philosophical otium abound, even if both portrayals owe their genesis to standard poetic topoi.⁴⁵ Hippolytus, like the Stoic sage, defines his moral outlook in opposition to the popular values of society, and removes himself from that society the better to pursue his life. Also like the Stoic sage, he imagines his isolation as a supreme form of power: the forest-dweller is ‘lord over empty fields’ (rure vacuo potitur, Phaed. 501), an image not far removed from Stilbo’s triumphant stance in a devastated landscape.

As my reference to Stilbo suggests, however, Hippolytus’ autarky is far from being unproblematic or morally pure. His disavowal of love transforms itself all too rapidly into an exercise in hate. While Hippolytus wishes to preserve his freedom by avoiding being ‘conquered’ by a woman (victus, Phaed. 573),⁴⁶ he also regards his mother’s death as granting him ‘the license to hate all womankind’ (odisse … feminas omnes licet, Phaed. 578–9). Several commentators note that this sentiment reworks a fragment from one of Euripides’ lost Melanippe tragedies (498 Kannicht; 498 Collard and Cropp): πλὴν τῆς τεκούσης θῆλυ πᾶν μισῶ γένος (‘I hate the whole race of women apart from my mother’).⁴⁷ There is, however, an essential difference in Seneca’s version: the verb licet, which shifts focus from misogyny per se to the fact of Hippolytus’ freedom to indulge in it. Crudely put, Antiope’s death removes from Hippolytus yet one more constraining social bond, which affords him the licence to behave as he pleases. Rather than being the object of someone’s love (cf. the passive form, victus, Phaed. 573, above), Hippolytus uses his solo status to assert active control of the situation. This self-focused isolation is, moreover, a particular characteristic of the Senecan Hippolytus, who shuns all society, right down to the family unit, in contrast to his Euripidean counterpart, who refuses merely to worship Aphrodite.

Of course, the autonomy avowed by Seneca’s Hippolytus is not fully Stoic. It would be too simple a syllogism to say that isolation augments Hippolytus’ independence and that it does the same for the Senecan sapiens, therefore Hippolytus exemplifies a sapiens. This is not defensible, nor is it what I am arguing. Instead, the point is that Hippolytus’ angry, isolated, solipsistic view of autarky does not undermine Seneca’s Stoic principles so much as extend them. Granted Hippolytus misconstrues freedom as licence and, despite all his protestations of independence, actually enslaves himself to anger (Phaed. 566–8), nonetheless he embodies an extreme version of Senecan principles in his withdrawal from social activity, in his preoccupation with personal freedom, and in his condemnation of mob morality. Even his particular emotional weakness, rage, is the one Seneca confesses most likely to befall a sapiens: ‘the wise man will not stop being angry, once he begins’ (numquam irasci desinet sapiens, si semel coeperit, Ira 2.9.1); ‘if you expect the wise man to be as angry as the shamefulness of criminality demands, he must not just grow mad, but go insane’ (si tantum irasci vis sapientem, quantum scelerum indignitas exigit, non irascendum illi sed insaniendum est, Ira 2.9.4).⁴⁸ Although unlike Hippolytus, the sapiens does not fall into this trap, temptation in both instances comes in the form of moral outrage. Acute awareness of vice coupled with the desire to protect oneself prompts irate withdrawal. The comparison should give us pause.⁴⁹ Hippolytus is no sage, certainly, but neither can his autarkic aspirations be dismissed as mere delusion, or as foils to the authentic Stoic views expressed in Seneca’s prose. If Hippolytus’ Stoic ideals are warped – and they are, undeniably – that is partly because Seneca’s ideals are too.

The flipside of Hippolytus – a potential sage overlaid by angry, selfish tendencies – is Atreus, an angry tyrant with Stoic inclinations. Since I have dealt already with some of Atreus’ Stoic traits in Chapter 1, I shall restrict myself to a brief summary here. I have discussed how self-knowledge and firmness of purpose lend Atreus’ actions a quasi-Stoic tint; likewise, he echoes in distorted form philosophical concepts of the summum bonum (Thy. 205–6) and is presented as ‘untroubled’ by the chaos around him (securus: Thy. 720; 759).⁵⁰ This Stoic framework prompts – even if it does not outright confirm – the attribution of Senecan ideals to some of Atreus’ other activity as well. For instance, Atreus eschews commonly accepted moral principles as a barrier to his autonomy: in response to the satelles, who urges honourable conduct as the only source of genuine popular support, Atreus retorts that a ruler’s true power lies in being able to disregard the populace and trample on its values (Thy. 207–18). While the satelles advocates Senecan principles in this scene (to the extent that many critics, including the Octavia’s unknown author, have cast Seneca in the satelles’ role),⁵¹ Atreus’ disavowal of them can also be read in Senecan terms. Like Atreus, the Senecan sapiens pits himself against prevailing, popular morality, which he regards as a threat to his autarky. Sage and tyrant coincide in their desire not to come under another’s control: qua iuvat reges eant (‘kings should go where they please’, Thy. 218) – the sentiment needs but little modification to fit the sapiens as well.

Atreus also recalls the sapiens in his self-deification,⁵² a topic I have examined in Chapter 1, and now reprise in the light of my preceding comments about Senecan theology. Significant in Atreus’ case is both his equivalence to the divine (aequalis astris gradior, ‘I stride equal to the stars’, Thy. 885 resembles language used in Ep. 92.29: deos aequat, ‘is equal to the gods’), and the fact that his status is self-conferred. By the conclusion of his revenge, Atreus’ power and independence exceed those of the mythological pantheon, checked only in his inability to drag the gods back from their flight (Thy. 893–5). Certainly, one can see in his divine pretensions the megalomania of a figure like Caligula,⁵³ or more simply, the tradition of imperial deification (which was not always strictly posthumous). But the divine aspirations of the Senecan sapiens also belong within this nexus, since the wise man, too, is portrayed as outranking the gods in his capacity for autonomous, autarkic action, a position achieved via his own relentless effort. Here, as on so many other occasions in Seneca’s work, tyrant and sapiens share essential aims and qualities, albeit ones that originate in vastly different value systems. This is what prevents Atreus from being pure parody; his questionable traits belong to the sapiens, too, just in a different guise. Thus, the juxtaposition that many scholars detect ultimately fails to hold. John Stevens, for instance, suggests that Atreus ‘does not wish to join the heavenly community by perfecting his virtue, but to supplant the gods by perfecting his vice’.⁵⁴ True, up to a point, but the common goal of perfection indicates a degree of complementarity, and the idea of supplanting or somehow exceeding the divine is already there in Seneca’s portrait of the sapiens. The wise man’s relationship to the gods is just as competitive as Atreus’.

The most powerful version of this quasi-Stoic autarky is not Atreus, though; it is Medea. Her exchange with the Nurse, in particular, is laced with sentiments that would fit just as easily in the mouth of a Senecan sapiens:⁵⁵

Med: fortuna fortes metuit, ignavos premit.

Nut: tunc est probanda, si locum virtus habet.

Med: numquam potest non esse virtuti locus.

Nut: spes nulla rebus monstrat afflictis viam.

Med: qui nil potest sperare, desperet nihil.

Nut: abiere Colchi, coniugis nulla est fides

nihilque superest opibus e tantis tibi.

Med: Medea superest: hic mare et terras vides

ferrumque et ignes et deos et fulmina.

Med: Fortune fears the brave, but crushes cowards.

Nur: Courage must be put to the test if there is occasion for it

Med: There will never not be an occasion for courage.

Nur: No hope shows the way out of your afflictions.

Med: One who has no hope despairs of nothing.

Nur: The Colchians have gone, your spouse is unfaithful,

nothing remains of your once great wealth.

Med: Medea remains; here you see sea and earth,

and steel and fire and gods and lightning bolts

(Med. 159–67)

Medea is safe because nothing else can be taken from her.⁵⁶ She is self-reliant in the face of Fortune’s onslaught and responds to the deprivations of victimhood – the loss of husband, home, and resources – by affirming her self-possession and freedom to act on her own behalf. Her praise of virtus encompasses both the masculine, heroic ideal of ‘courage’ and the Stoic ideal of moral ‘virtue’, which, Medea affirms, remains not just a but the only constant in any situation.⁵⁷ Line 163 – qui nil potest sperare desperet nihil – finds partial echo in Epistle 5.7, where Seneca quotes Hecato: desines timere si sperare desieris (‘you will cease from fear if you cease from hope’).⁵⁸ More broadly, Medea resembles a sapiens in her disavowal of ‘externals’ and in her ability to function fully, autonomously, without them: fortuna opes auferre, non animum potest (‘fortune can take away my wealth, but not my spirit’, Med. 176). Later, when Jason confronts her, she will likewise claim, ‘Fortune, in every form, has always stood below me’ (fortuna semper omnis infra me stetit, Med. 520), and, ‘my mind is able and accustomed to despise royal wealth’ (contemnere animus regias … opes / potest soletque, Med. 540–1). Both statements could stand alone as Senecan, Stoic assertions of autarkeia, making Medea, on the face of it, equal to Stilbo: iustitia, virtus, prudentia, hoc ipsum, nihil bonum putare quod eripi possit (‘justice, virtue, wisdom, in other words, he considered nothing that could be taken from him to be a good’ Ep. 9.19).

The preceding qualifications – ‘could stand alone’; ‘on the face of it’ – are crucial, though, because Medea’s actions within the tragedy actually demonstrate an excessive, destructive concern for externals, especially for her reputation and for her hold over Jason.⁵⁹ Contrary to her disavowals of loss, she is affected by the drastic change in her circumstances, which she plans to rectify to her satisfaction, even if only through the emptiness of revenge. In addition, Medea’s forceful self-mastery is offset by her describing herself as the object of passional forces (e.g. Med. 937–44), which, by strict Stoic standards, makes a mockery of her desire for independence.

And yet, as in the case of Atreus and of Hippolytus, the equation is not so simple, because what Medea hopes to regain most of all is the capacity to control her fate, and this pursuit of self-determination coincides on many levels with the philosopher’s, whose inclination for terrestrial mastery I have outlined already, above. Medea celebrates solitariness because it confers invulnerability and the release from being subjected to another’s power. In the midst of disaster, surrounded by threats, Medea finds strength in the thought that she has herself to fall back on. Paul Veyne’s characterisation of Stoic self-reliance could just as easily be applied to Medea 166–7 (if it is not already a citation of it): ‘when all seems lost, the only thing that really counts and acts, the I, remains’.⁶⁰ Medea superest: hic mare et terras vides / ferrumque et ignes et deos et fulmina (Med. 166–7). The real import of these two remarkable and enduring lines⁶¹ lies not in their evocation of Medea’s magical powers,⁶² nor in their self-conscious citation of her mythological-literary pedigree,⁶³ but in their affirmation of steadfast, autarkic selfhood. Medea, and no one else, will dictate what ‘Medea’ means. Just as it is in her power as a witch to summon or even to embody the natural phenomena of sea, earth, fire, and lightning, so it is within her power as an individual to shape herself and ensure her own security. The lines’ rhetorical punch comes from their celebration of agency, agency in its most naked form: the pure power of sea or flame.⁶⁴ Like Seneca’s sapiens, Medea claims to be untouchable, indomitable, at precisely the moment when she is most in danger of being dominated.

Lurking behind all of Medea’s quasi-Stoic assertions is the promise of revenge, which she deems the chief means of re-establishing her autonomy. Her defiance of fortune is not just an acknowledgement of inviolable inner strength but also a guarantee that her crimes will outdo anything fortune has wrought against her, overturn it, control it. The same goes for her self-affirmation in response to loss and victimhood: besides highlighting Medea’s fierce autarkeia, lines 166–7 foreshadow the gathering storm of her vengeance, which, she avers later in the play, will uproot and flatten everything with its violent onrush (Med. 411–14). Paradoxically, her Stoic professions of detachment actually contribute to her earthly triumph.

Nor is this quasi-Stoic dynamic of revenge exclusive to the warped world of the tragedies; we can also see it, faintly, in Seneca’s stories of sage versus tyrant, where the former’s moral victory resembles a kind of retribution for what he has been made to suffer. I note above the competitive relationship Seneca envisages for these two figures, and the sage’s need to ‘win’ at the game of possession and control. Such competition approximates retaliation in the relationship of equivalence it creates between sapiens and ruler: the former responds to the latter’s aggression on equal and opposite terms, triumphing over his adversary because he engages in a superior version of the harm he has experienced.⁶⁵ Demetrius robs Stilbo of his city and family, Stilbo robs Demetrius of his victory; Caligula sentences Canus to death, Canus trumps Caligula’s power by counting death as nothing.⁶⁶ Just as the avenger typically assumes and exceeds his opponent’s characteristics, so these ‘victories’ are described as mimicking the rulers’ military and political sway. The impression is mild, but unmistakable: the sapiens, like Medea, uses his self-sufficiency as a form of revenge.

4.2 Revenge

Medea: Vengeance, Identity, Autarky

It is not surprising to see Seneca’s vision of Stoic autarky gravitate towards revenge, because aside from the particulars of the sapiens’ competitive stance, vengeance itself is an exercise in autonomy. Francis Bacon called it ‘a kind of wild justice’.⁶⁷ Revenge mimics the law while occupying a space beyond its reach; it is the flipside of judicial procedure. αὐτο/νομία: independence, and more literally, taking the law into one’s own hands. Linguistic derivation points to an underlying union of ideas, namely that personal autonomy is deeply implicated in the pursuit of revenge and that vengeance amounts to the search for individual – moral, political, social – freedom. As Eric Dodson-Robinson rightly recognises, revenge is a declaration of agency in response to personal disaster.⁶⁸ Being made into a victim, being made to suffer loss, dishonour, or physical damage deprives one of sovereign jurisdiction over one’s own life and body. The injury endured by the victim – whether corporeal, psychological, social, or any combination thereof – represents his or her helpless submission to external forces and a consequent distortion of authentic selfhood. Vengeance is a means of reasserting control over one’s life, reclaiming one’s capacity to act, and reconfiguring one’s identity in response to its (often) violent disfigurement at another’s hands. The avenger seeks to transform him- or herself from passive object into active, aggressive subject; such reciprocal retaliation could just as easily be called ‘a kind of wild self-fulfilment’. And crucially, a bid for self-determination. There is much in the impulses of vengeance that reflects the desires of the Senecan sage, even if he employs vastly different means to realise them.

This section moves away from Stoic preliminaries, however, to consider how Medea’s vengeance shapes her autonomy and identity as a quasi-human within the fictional world of her play. Successful retaliation grants Medea the freedom to define herself, and her sexual and social status, as she wishes. It enables her to reassert control over her body, her future, even over the record of her past. And, like so many other instances of Senecan autarky, it leaves her adrift in a solitude of her own making.

Every element of Medea’s revenge is geared towards recalibrating her sense of self in the wake of Jason’s betrayal. In both the Euripidean and the Senecan version, Medea’s vengeance works through exact reciprocity to deprive Jason in the same way his actions would have deprived her. He abandons their marriage; she prevents his remarriage. More specifically, her murder of Creon and Glauce/Creusa sabotages Jason’s prospective kinship ties and his place within Corinth’s socio-political order, just as his remarriage meant severing ties with Medea and jeopardising her socio-political status to the point of consigning her to exile. Having left her homeland, abandoned her father and murdered her brother all for Jason’s sake, Medea takes as recompense Jason’s future father-in-law and adopted home.⁶⁹ Since Jason had intended to keep the children from their marriage, leaving Medea bereft, she pre-empts him and bereaves him of them permanently. She ensures that Jason, too, will have to endure wandering in exile, tainted by criminal associations, shunned by other kings and communities. By the time her revenge is complete, the only connection remaining to Jason is the only one he tried to break: his union with Medea.⁷⁰

Such acts of reciprocal, pre-emptive desolation are Medea’s ways of recovering her status and identity and reaffirming her capacity for self-determination. As Gianni Guastella has demonstrated in a perceptive article on the revenge dynamics of Seneca’s version, Jason’s plans imperil the social roles Medea has built for herself. Not only would his remarriage invalidate her position as spouse and – to a lesser extent – as mother, it would also render meaningless all of Medea’s prior, often criminal actions in the service of Jason’s safety; if she loses him, her past loses its purpose.⁷¹ Thus, when Seneca’s Medea asks her husband coniugem agnoscis tuam? (‘do you recognise your wife?’ Med. 1021), she is – besides the other interpretations discussed in Chapter 1 – emphasising her faculty of self-definition and ensuring Jason acknowledges it as such. He has tried to change her status, to remove her role as wife; she has wrested back that power. Similar logic drives her infanticide, since Medea asserts possession of her children by disposing of them as she wishes (there is, perhaps, a hint of this in Med. 935: pereant, mei sunt; ‘let them perish, they are mine’)⁷² and at a deeper, unspoken level, by fixing them as hers for all eternity: they will never grow up, never leave, never change. Empty victories, of course, but that is the price of revenge, which activates the victim’s agency and confirms his or her identity at the expense of the social bonds that constitute that identity in the first place.⁷³ Retaliation reinstates Medea qua Medea, but, paradoxically, without the relationships that made her so.

Identity formation through vengeance is a pervasive theme in the play, and, as the preceding example of Medea 1021 indicates, it is especially noticeable in the heroine’s habit of self-reflexive speech. Repeated utterance of her own name and role represents for Medea a self-exhortation to retributive action, a totemic guarantee of what she is capable of and who that capability marks her out as being. Often overlooked in favour of metapoetics and dramatic self-awareness, this aspect of Medea’s illeism is equally as crucial for our understanding of Seneca’s composition. When, for instance, Medea goads herself to ‘embark on all that Medea can do, and all she cannot do’ (incipe / quidquid potest Medea, quidquid non potest, Med. 566–7), a meta-literary interpretation would highlight the character’s acknowledgement of her own abilities and storyline, as well as Seneca’s ambition for his Medea to surpass all her previous incarnations. On this reading, the lines’ self-reflexivity would be a combined declaration of poetic aspirations, belatedness, and anxiety of influence. It is all of these things. Yet it is also the heroine’s promise to attain self-definition via successful pursuit of revenge: what Medea can and will do, after all, is harm Jason, an act that reinstates her sense of self by allowing her to control her fate. The two readings tend to pull in opposite directions: either Medea is caught in a cycle of pre-scripted activity or, as a human analogue, she uses revenge to achieve self-government and fully realised subjectivity. Both interpretations are valid; each identifies one of the lines’ fundamental features. But we should be wary of stressing metapoetics to the detriment of Medea’s implied humanness, because, besides acknowledging her literary pre-destination, Medea 566–7 is also a fierce celebration of individual agency in which Medea qua person promises to overturn all checks and limitations, to exceed all constraints, and to achieve something beyond the expected, beyond even the pedestrian realm of the possible. Aspiring to do what she cannot (currently) do is Medea’s way of attaining greater autonomy and freedom, and of realising her selfhood via the absolute independence to act as she wills. In a similar vein, the famous Medea— fiam (‘Medea— I shall become her’ 171) is not just a promise to fulfil a pre-existing dramatic role, but a guarantee of Medea’s ability to fashion her own identity as she, and she alone, wishes, no matter what anyone else tries to do to her (we might want to stress the first-person: ‘Medea— I shall become her’). Such affirmations of sovereignty reveal the self-creation inherent in the heroine’s project of revenge: she and no other will decide what ‘Medea’ represents and who Medea is.

In addition to recalibrating her future, moreover, revenge also confers control over the production of her past. It dictates how she will be remembered – not as the victim, but as the perpetrator, the active party in the event (e.g. Med. 52–3; 423–5). Likewise, it facilitates the recuperation and reformulation of what she has lost, as in her counterfactual claim to have regained father, brother, homeland, and virginity (Med. 982–4).⁷⁴ While none of these things has (or can!) be reinstated in actuality, they encapsulate the autonomous self-fashioning and self-legitimisation afforded by Medea’s vengeance. By re-establishing her dominance, revenge enables her to believe in and to impose whatever version of the past best suits her. History belongs to the winner.

The autarkic impulse of Medea’s vengeance extends further still, from the immaterial realm of her reputation to her flesh-and-blood presence as a maternal body. Of all the uniquely Senecan elements in this play, many of which bear directly on the interlinked issue of vengeance and identity, her vow to scour her womb for any remaining embryos ranks as one of the most memorable: ‘if any love pledge still lies hidden in the mother, I shall search my womb with a sword and drag it out’ (in matre siquod pignus etiamnunc latet / scrutabor ense viscera et ferro extraham, Med. 1012–13).⁷⁵ In one vicious image, Medea sums up the agency conferred by revenge. If children symbolise a diminution of her autonomy – through her dependence on a spouse, lack of control over her own body, and pledged bond (pignus) to another person – then the prospect of abortion represents its reinstatement. As a woman, Medea achieves independence from socio-political constraints by first achieving independence from corporeal ones. Her willingness to engage in self-harm also verges on Stoic contempt for bodily pain, as though Medea has to subjugate herself to herself, and refuse the lure of externals in order to become fully autarkic.⁷⁶ Moreover, like Senecan concepts of individual sovereignty, Medea’s self-government hinges on increased removal from human society: the image of abortion shows her cutting ties to Jason at a most visceral level. Once again, her project of revenge acquires a quasi-Stoic dimension, as her desire to obliterate damaged personal relationships amounts to a fiercely defended form of self-mastery.

Such defiance of limitations, Medea’s assertions of agency, her self-fashioning and desire to dictate her future – all of this activity draws attention to her status as an implied human figure. Even though vengeance is built into her story, still she pursues it on the assumption that her circumstances are contingent, otherwise retribution would be pointless. Revenge’s transformative power,⁷⁷ its promise to elevate Medea from victim to aggressor, passive to active participant, implies her possession of quasi-human agency and potential for individual change. Even her penchant for self-naming represents an assurance of future development and maturity, of control over both self and world, in addition (and contrast) to its well-recognised emphasis on the repetitive, fictional quality of Medea’s being. One could, of course, ask whether Seneca’s audience needs to believe in Medea’s implied humanness first, before accepting these traits as evidence of it, but the question is unnecessarily chicken-and-egg (and in any case, similar acceptance of Medea’s fictional status must precede awareness of her metapoetic qualities). What matters is that Medea’s quasi-human characteristics should not be overlooked, not least because it is these features, rather than ironic metapoetics, that endow her rhetoric with such force, and lend her behaviour an urgent, troubling moral dimension. Medea qua textual construct may play with the contours of her narrative, but Medea qua person explores the limits of human constraint and capacity for action.

In contrast to Medea’s hard-won autonomy, the Jason of Seneca’s version appears perpetually subjugated and hemmed in, which accentuates his wife’s power all the more; brief discussion of this binary rounds out my present analysis of Medea’s revenge. Seneca’s Jason is a notably weaker and more minor figure than Euripides’, partly because he has fewer lines but also because Seneca depicts him as the constant victim of other people’s dominance.⁷⁸ When he arrives on stage, he confesses not to have broken faith with Medea of his own free will, but under compulsion from Creon, who has forced him into a marriage alliance in exchange for protecting him and his sons from Acastus’ vengeance (Med. 434–9). Having been duly separated from her husband and children, Medea may be handed over to Acastus for punishment, in retaliation for her prior killing of Pelias. No mere decoration, inclusion of this backstory is designed to minimise Jason’s independence, something the chorus, too, acknowledges when it asks the gods to ‘spare a man who acted under orders’ (parcite iusso, 669). Unlike Medea, Jason seems forever unable to assume control of his situation. He complains of being bound by fata (Med. 431) whereas Medea vaunts her superiority to fortuna (Med. 520). Word-choice is significant, too, since fatum implies a pre-ordained sequence of events while fortuna designates something fickler and more changeable.⁷⁹ In contrast to Medea’s confident dismissal of externals, Jason is at their mercy; he seems unable to rely solely upon himself and he constantly denies responsibility, as though he were not the source of his own actions. Far from making him an honourable or innocent figure,⁸⁰ these traits cast Jason as Medea’s feeble foil, an individual whose misguided concept of security has greatly diminished his agency and independence. Whereas Jason relies on Creon (e.g. Med. 538–9, where he promises Creon’s money rather than using his own; cf. Eur. Med. 610–3), Medea relies on herself. True, she strives for power over her oppressors and seeks to master fortune rather than, in Stoic guise, to conform to its demands, but her independence outstrips Jason’s because she regards herself, not others, as the only real source of safety, of fairness, even of meaning. Jason may crumble, but Medea superest.

Revenge and Fictional Autonomy

I mentioned near the close of the preceding subsection that Medea’s revenge highlights her quasi-human features chiefly by accentuating her capacity for autonomous action. The effect is hardly exclusive to Medea, or to Seneca. Rather, it is a consequence of revenge narratives more generally, which orbit around questions of self-assertion and self-determination, and which propel fictional characters into independent, largely self-motivated action. Revenge in literature distils issues of choice and intention and imbues them with particular urgency. While many other fictional scenarios also achieve this variety of effects, I have chosen to focus my present discussion on revenge because it is a distinctive element of Senecan tragedy and of Western theatre more broadly.⁸¹ Further, it has the advantage over other fictional scenarios of telescoping all these scattered facets of autonomy into one, climactic event.

The revenge plot tends to focus on control, which is perhaps the most obvious means of its emphasising characters’ humanness. A major difference between fictional and actual beings is the level of mastery they exercise over their own existence, for, though both groups are inhibited by circumstance, by the demands of others, by convention, and in some belief systems, by ineluctable fate that plays itself out like a narrative, still human freedom exceeds that of characters’ in its capacity for choice and change. One cannot assume control when one’s context is not contingent, and to the extent that characters are imprisoned within their scenarios, they are powerless to govern their own affairs. On a metaliterary plane, the avenger’s explosive anger articulates frustration at such restrictions: it crashes through the status quo, pushes beyond the expected and the possible, and rearranges its milieu radically, violently, on its own terms. From Aeschylus’ Clytemnestra to Shakespeare’s Hamlet and beyond, the avenger assumes a quasi-authorial, quasi-directorial role, constructing his or her own scenarios, plots (in both senses of the term!), tricks, contraptions and performances.⁸² The avenger aspires to dictate how subsequent action will unfold in his or her fictional world, an act of control that imagines, simultaneously, the possibility of contingency – futures can be altered; circumstances and people can change – and its lack – all events must come under the command of a single, directorial will. Choice also plays a role here, because on the one hand, avengers are compelled to act by a host of forces beyond their immediate control,⁸³ but on the other, must decide whether to accept the call to retribution and how to execute it. It is from Hamlet’s indecision over whether revenge is the right course of action that his character acquires much of its complexity and depth, its illusion of intimate individualism. In a very different fashion, but still on the topic of choice, it is Atreus’ deliberation over the method of revenge that delineates so clearly his moral and behavioural traits. The revenge plot’s attention to decision-making and to eventualities accentuates the avenger’s status as a human analogue while at the same time acknowledging – one might even say sympathising with – the limitations of fictional existence.

Accompanying the idea of contingent futures, moreover, is the idea of contingent selves. As I remarked in the preceding analysis of Medea’s revenge, acts of retaliation typically entail self-(re)creation or development as part of the avenger’s escape from victimhood. In Seneca, revenge is as much about self-discovery as it is about righting perceived wrongs. Though Senecan avengers do not undergo any radical shifts of personality, they can still be said to enlarge their capabilities and increase the sheer force of their presence over the course of the play. Medea, for instance, declares in Act 5 that her character ‘has grown through evils’ (crevit ingenium malis, Med. 910), where malis most likely signifies both the crimes she has committed against others and the prior suffering she has endured at their hands.⁸⁴ Pursuit of vengeance has increased her psychological and moral stature even if her identity has proceeded along the same continuum throughout. And in terms of social status – for this ‘outward’ form of selfhood is one of the avenger’s prime concerns – Medea transforms herself from marginalised fugitive into a powerful manipulator of other people’s fates, and, in less positive terms, moves from being wife and mother in actuality to being them in name only. In effecting a transition from victim to agent, passive to active, the fictional avenger embodies a distinctly human capacity for change, and particularly, for self-directed, self-motivated change, which forms the basis of so much autonomous action in the non-fictional universe.⁸⁵

Another notable consequence of the revenge plot is its filtering of events through the avenger’s perspective, a focalisation that happens just as much in dramatic as in narrative literature,⁸⁶ and is especially pronounced in Senecan tragedy, where avengers dominate the dialogue and overrule all opposition with their superior wit. Of course, first-person viewpoints are far from unique in ancient literature, but revenge plots are distinctive for their sustained reliance on a single character’s perception of events in a genre where focalisation is more usually dispersed across multiple speakers. The tragic avenger is our confidant and commentator: we know what Atreus and Medea are plotting, and this knowledge, besides generating ample dramatic irony, grants us privileged access to their intentional and emotional states.⁸⁷ The illusion of their humanness grows in proportion to this access, as their revenge becomes an expression of agency and individual will. Emphasis on the avenger’s perceptual activity adds an intimate, private dimension to the character, as though he or she were endowed with fully functioning consciousness and such hidden realms of thought as necessarily accompany a first-personal perspective. Nor does the character have to be particularly ‘round’ or ‘deep’ for this rule to apply. Although Seneca’s avengers occupy the opposite end of the spectrum from, say, Hamlet’s anguished complexity, still their aggressive focalisation of the tragedies’ events creates an impression of internality, of decision-making and moral sensibility, no matter how rhetorically expressed. This internality is the wellspring of autonomy, the self behind the action, the doer behind the deed.

On the topic of doers and deeds, however, one could demur that the avenger’s role as an agent of causality is no different from, say, Greimas’ theory of the actant, a narrative element that propels action and may be instantiated by animate and inanimate objects alike, and even by abstractions.⁸⁸ While my preceding discussion takes for granted a link between agency and human or quasi-human autonomy, Greimas divorces the two categories, prioritising the former in such a way as to dismiss the latter, alongside refusing to accord any special status to fictive agency in its human as opposed to non-human forms. The avenger, on this reading, becomes a sophisticated species of plot device, an initiation (rather than the more personal ‘initiator’) of subsequent fictive events and of no more significance to the narrative syntax than any other catalyst for action. To take an example from Seneca’s Phaedra, the revenge unleashed against Hippolytus is activated as much by the sword (Phaed. 898–900) as by Theseus himself. In Greimas’ view, the two would claim equivalency as spurs to the ensuing sequence of events.

The theory falters, though, in its failure to acknowledge how central a concept of human agency is to our understanding and appreciation of fiction, so central in fact that the agency fictional works accord to objects, animals, plants, and abstract phenomena – to name just a few – tends to be framed in human terms, modelled on a broadly accepted (if culturally conditioned) understanding of human capacities.⁸⁹ Many of these fictional actants are endowed with intentional and emotional states or treated as though they possess them; many exhibit enduring traits, both physical and psychological. The sword in Seneca’s Phaedra ‘speaks’ Hippolytus’ name to Theseus (hic dicet ensis, Phaed. 896), despite its inanimate existence. Thus, in contrast to Greimas’ equating fictional beings with objects and impersonal forces, the association is more likely to work in the opposite direction, as the impersonal is typically imbued with person-like qualities. Of course, the sword is not capable of self-directed action, and in this regard, its narrative agency does not lead to or derive from any impression of autonomy. But just because some agents of causality are non-autonomous does not mean that all are, and characters, as human analogues, have a special claim to being measured against human models of action. To see Theseus’ vengeance purely as a narrative prop is to erase his responsibility for what happens next, and therefore to erase the audience’s sympathetic involvement with his character (is he likeable, or not? Is he justified?) and besides, to erase the very thrust of tragedy – of an unfair fate engineered by accident – that emerges from the ensuing events.⁹⁰ To interpret the play in these terms, which is by and large a critical norm in Senecan studies, is to ascribe, tacitly, a degree of autonomy to Theseus, whereby he exercises his fictional independence to make a crucial – and damning – choice. What is true of Theseus, moreover, is true of most if not all avengers in drama, not to mention of most characters in literature more broadly. Their structural agency as elements that propel the plot is complemented to the point of being overshadowed by their thematic agency as quasi-human figures whose actions resonate across their fictional landscape. Though we must guard against overstating fictional autonomy, we must also guard against eradicating it.

Turning back to the Senecan avenger, we can see this balance in play, because besides accentuating characters’ implied humanness, acts of vengeance on the Senecan stage also call attention to their status as fictional constructs. I remarked above that avengers tend to assume a directorial or authorial role within their dramas, a circumstance that summons the shadow of metatheatricality and self-conscious performance. As always, Seneca’s Medea furnishes excellent examples. Her skill in magic, for one, encapsulates simultaneously her power to effect change through vengeance and that power’s circumscribed, fictional nature. Witchcraft enables her to dictate the course of events and to orchestrate Jason’s downfall, activities that assimilate her to Seneca qua author. The latter half of the tragedy is a performance directed by Medea herself, in which she also plays the starring role.⁹¹

This authorial function is particularly evident in the magical power Seneca ascribes to Medea’s voice.⁹² In Act 4’s spell-casting scene, the Nurse reports how Medea ‘summons plagues’ (pestes vocat, Med. 681); how a ‘scaly crowd’ of snakes ‘is drawn forth by her magic chanting’ (tracta magicis cantibus / squamifera … turba, Med. 684–5); how a serpent ‘is stunned at hearing her song’ (carmine audito stupet, Med. 689). Medea, too, recognises and revels in her voice’s magnetic quality: ‘may Python come’, she pronounces, ‘at my songs’ command’ (adsit ad cantus meos / … Python, Med. 699–700); ‘I have summoned rain from dry clouds’ (evocavi nubibus siccis aquas, Med. 754); ‘the summertime earth has shivered in response to my chanting’ (aestiva tellus horruit cantu meo, Med. 760); the forest ‘has lost its shade at my voice’s command’ (amisit umbras vocis imperio meae, Med. 767); ‘the Hyades are shaken by my song’ (Hyades … nostris cantibus motae, Med. 769). The metapoetic sense of these references is not hard to find: Medea’s poetry (carmen; cantus) conjures the world into being and arranges it according to her liking. As the terminology suggests, this is solemn poetry in an elevated genre: both carmen and cantare can be used in reference to tragedy. In addition to casting her as a dramaturg, moreover, Medea’s vocal abilities associate her with the actor, whose task involves ‘positing the existence of fictional space and fictional objects’ through the sheer power of speech acts.⁹³ Things happen, things exist, because Medea says so.

If Medea’s vocal power symbolises her mastery – over words, over the environment, over the play’s events – it also indicates her ultimate lack of such control, by signalling that she, too, is the fictional product of a carmen, the object of somebody else’s imagination, subject to somebody else’s will. By fashioning her as an author/director figure, Seneca reminds his audience of Medea’s fictive status as a character in his play. Her occasional similarity to an actor likewise contributes to this effect, because it celebrates her ability to manipulate spectators both internal and external to the drama (cf. her calling Jason spectator at Med. 993), while also acknowledging her subordination to a script. Thus, her magic simultaneously guarantees her autonomy and divests her of it, creating the illusion of her omnipotence only to stress that it is just that: an illusion.

While Act 4 provides the most plentiful crop of examples, Medea’s quasi-authorial role is cited at other points in the play, too. When Creon denounces her as a malorum machinatrix facinorum (‘a contriveress of evil deeds’ Med. 266) and when the Nurse, in quaking admiration, calls her a scelerum artifex (‘an artist of crime’ Med. 734), each underscores her creative abilities as a practitioner of wickedness. artifex in particular is a word that refers not only to authors, but also to actors and stagehands (e.g. artifices scaenae: Sen. Ben. 7.20.3; Suet. Jul. 84.4; Gell. 3.3.14), which situates Medea’s metapoetic power in a solidly theatrical context. Jason, similarly, calls her a sceleris auctor (979), recognising her authorship of crime via an appropriately generative metaphor, since it is through her increase of children that Medea achieves her scelus. Not only that, but her crime itself grows from, builds upon and extends the scale of her earlier forays into wickedness. Like Seneca’s portrayal of her magic, each of these appellations articulates Medea’s power as an agent of vengeance while at the same time admitting the limitations imposed by her fictional existence: she can contrive anything to suit her angry purpose … but only at the behest of her own auctor, Seneca. The idea resurfaces, climactically, when Medea prepares herself for infanticide by ordering her grief to ‘seek out material’ for its revenge (quaere materiam, dolor, Med. 914). A common metapoetic marker, materia here signifies, all at once, the means of Medea’s vengeance – the very stuff of her children’s bodies and of her maternal role in producing them – the content of Seneca’s version, and Medea’s own status as materia for Seneca’s tragedy.⁹⁴ In other words, it expresses both her authorial aspiration to shape events and her subjugation to another’s authority. Her implied humanness as an individual, purposive agent within the drama’s universe is shackled to her purely fictional ontology.

Shackling and limitation are core elements of the avenger’s experience and this is another reason why revenge plots convey so precisely the problem of characters’ autonomy. At base, vengeance is a response to powerlessness. The avenger chafes against constraint and consequently explodes into anger against his or her perceived oppression. This quality of weakness and subjugation is fundamental, though critics often lose sight of it: we may, for instance, be inclined to regard Atreus as all-powerful from the play’s outset, but he makes it clear that he feels vulnerable and victimised as a result of Thyestes’ adultery. His revenge represents re-instatement, a resumption of confidence and autocratic sway (a theme explored below, in the next subsection). As an expression of fictional agency, therefore, Atreus’ vengeance exalts his dominance while never once losing sight of his containment within a given literary form. His desire to surpass all kinds of limitation, crystallised in his repeated use of modus (Thy. 255; 279; 1052) combines the tyrant’s with the avenger’s inherent inclination to overreach, while at the same time acknowledging that his fictional power is bred of constraint.

The avenger’s freedom is never complete, either, because the act of vengeance itself is always (over)determined by forces beyond the avenging agent’s control. Betrayal happens, murder is committed, and the victim-cum-avenger makes a move in response. Re-venge is inherently re-active, a secondary event conditioned by other, arguably (or seemingly) freer agents and imposed upon the avenger not just by dint of circumstance but, often, by other individuals – or ghosts – seeking personal recompense via the avenger’s hands.⁹⁵ Viewed from this angle, vengeance becomes a duty, and the avenger more of an instrument than an agent.⁹⁶ Although Seneca’s chief avengers, Atreus and Medea, act on their own behalf and set out to gain satisfaction only for themselves, still their activity is predicated on a host of preceding events and prevailing influences that ensnare them within a particular storyline. Atreus is spurred into action by his brother’s transgressions and by the spectral inspiration of Tantalus, who is in turn goaded by the Fury.⁹⁷ His presumption of individual control seems paltry against this backdrop, but that is the central dynamic of the revenge plot, in which fictional autonomy is at once granted and withheld, stimulated and suppressed by the self-same forces.

Lastly, it is helpful to think about the avenger’s autonomy in a specifically theatrical context, because if the act of retaliation distils issues of individual agency, so too does stage performance, with a comparable degree of clarity and urgency.⁹⁸ Both centre upon the need for action, upon action as a determinant of identity, upon the performer’s power to effect change in his or her surroundings and to manipulate an audience. As I note above, the actor resembles the avenger in experiencing a compromised autonomy. On the one hand, he or she enjoys the freedom of doing things on stage (δράω – drama) and of being an active subject in contrast to the audience’s physical passivity as recipients of the performance.⁹⁹ In ancient Rome, where theatrical performers were typically disenfranchised and occupied the lowest rungs of society, the contrast must have been starker still, as theatre gave otherwise powerless individuals the opportunity to appear as self-directed, self-determined beings.¹⁰⁰ Although the Roman actor’s slave status classifies him, socially, as an object (e.g. Varro Rust. 1.17.1), he may become on stage a thinking, acting, intentional subject. On the other hand, though, this subjectivity is conferred by prevailing social and dramatic conventions, before which the actor must (literally!) bow and which imprisons him/her as the object of the audience’s gaze and approval. The performer’s freedom is short-lived, dependent not only on the play’s duration, but also on the authority of playwright, director, and spectators. Like the tragic avenger, the actor’s explosive agency reveals a relentless chafing against the very restrictions that produce it. In this respect, the tragic avenger could even be considered an Ur-figure for the actor, making it unsurprising that revenge plots underpin so much of Western drama: their obsession with agency and autonomy holds a mirror up to theatre itself.

Atreus: Vengeance, Identity, Agency

Seneca’s Atreus is an avenger par excellence, in his affirmation of indomitable individual agency, and in his converse role as a product of family entanglements and an already overdetermined genre. He epitomises, simultaneously, the avenger’s license to do as he pleases and his subjection to powers and processes beyond his control. He embodies what Curtis Perry defines as ‘a core dialectic [of Senecan tragedy] … in which hyper-assertive selves are set against ironizing structures of predetermination’.¹⁰¹ This section and the following one examine, respectively, the autonomy Atreus pursues through vengeance and the restrictions he inherits as part of his literary and genealogical background.

Like Medea, Atreus begins his play desperate to reclaim the identity and status he feels have been stolen from him. Cuckolded by his brother, his paternity uncertain and his kingship vulnerable, Atreus hungers to reconfirm his social, sexual, and political dominance. His sense of manhood, in particular, is a crucial though under-explored¹⁰² aspect of this capacity for action, for it is by reaffirming his sexual vigour that Atreus prevails over Thyestes and, concomitantly, asserts himself as an actively self-determined individual. Thyestes is about adultery as much as it is about tyranny. Atreus’ opening words of self-excoriation, ignave, iners, enervis (‘useless, feckless, impotent’ Thy. 176), besides criticising his present inactivity as unworthy of a tyrant, also evoke the sexual passivity to which he feels Thyestes’ adultery has consigned him: he is unmanned, impotent.¹⁰³ Emasculation carries with it the entire baggage of objectification, marginalisation, and oppression that the avenger, too, experiences and fights against. For Atreus, this victimhood is the equivalent of occupying a woman’s role. When, for instance, he cites the myth of Tereus, Procne and Philomela, he aligns his suffering with that of the story’s female characters (Thy. 275–6),¹⁰⁴ as though Thyestes’ adultery and prior usurpation of the throne qualified as acts of rape, or as though he were experiencing infidelity and its ensuing family confusion from the wife’s perspective, like Procne. Either way, Atreus associates his damaged virility with the Greco-Roman woman’s social and sexual subordination to men, portraying his lack of control over his marriage and kingdom as equivalent to a lack of personal autonomy. Similar hints of effeminisation emerge from his desire ‘to be filled with greater monstrosity’ (impleri … / maiore monstro, Thy. 253–4), a phrase that conjures images of pregnancy through its use of impleo (cf. Ov. Met. 6.111),¹⁰⁵ its anticipation of Thyestes’ eventual fate, and its allusion to poetic inspiration, which is often depicted as a procreative act.¹⁰⁶ The implication is simultaneously positive and negative: Atreus qua avenger will swell with the productive power of his retaliation, but this is a situation made necessary by his fear of having been relegated to a sexually submissive role. Pregnancy is the perfect symbol of Atreus’, and indeed of any would-be avenger’s, compromised autonomy, as it destabilises everything from identity to corporeal integrity, and carries with it the stigma of female passivity, of being an object or vessel for somebody else’s use. Atreus’ masculinity will increase only in proportion to the agency, subjectivity and self-assertiveness proffered by revenge.

Over the course of his tragedy, Atreus carves out for himself a renewed role as a paterfamilias, with all its implied masculine dominance. He exchanges his initial state of enervation (enervis, Thy. 176) for harshness/hardness (durus, Thy. 763) as he hews his nephews’ bodies in preparation for cooking.¹⁰⁷ He also stops being a vessel and assumes instead the dominant part of an impregnator: ‘I shall fill up the father with the death of his sons’ (implebo patrem / funere suorum, Thy. 890–1), he promises the audience at the opening of Act 5. It is a promise he repeats in Thyestes’ presence: ‘now I shall fill up the father completely with his own throng’ (totumque turba iam sua implebo patrem, Thy. 979). This transferral of pregnancy from Atreus to Thyestes encapsulates the success of the former’s revenge. To reinstate his virility, Atreus compels his brother to undergo a transformation equal and opposite to his own: Thyestes begins as durus (Thy. 299) and ends up effeminised, his bulging gut an ugly parody of a full womb (Thy. 999–1004; 1041–4).¹⁰⁸ Grotesquely, Atreus proves his manhood by burdening Thyestes with children, an act that confirms his agency at the same time as it curtails that of its victim, for the female role thrust upon Thyestes renders him socially, politically, and sexually subservient to his brother, in addition to its depriving him of bodily autonomy.

Interlinked themes of sex, revenge, and personal agency also cluster around the play’s imagery of fullness and gratification, especially around the term sat/satis, which Atreus employs throughout. In Act 2, he complains that the ‘fire burning [his] breast is not big enough’ (non satis magno meum / ardet furore pectus, Thy. 252–3) and declares of his proposed attack on Thyestes, ‘I shall leave no outrage undone and none is enough’ (nullum relinquam facinus et nullum est satis, Thy. 256). The motif returns in Act 5, when Thyestes’ glut of wine and flesh (iam satis mensis datum est / satisque Baccho; ‘enough has now been given over to feasting, enough to wine’ Thy. 899–900) leads Atreus first to celebrate and then to doubt the fulfilment of his revenge: ‘it is good, it is ample, now it is enough even for me. But why should it be enough?’ (bene est, abunde est, iam sat est etiam mihi. / sed cur satis sit?, Thy. 889–90). Although he cannot compel the gods to witness his atrocity, ‘it is enough that the father view it’ (quod sat est, videat pater, Thy. 895). In the brothers’ final clash, Atreus gloats that Thyestes will shortly ‘have [his] fill’ of his children (satiaberis, Thy. 980), while Thyestes admits his innocent enjoyment of the meal: ‘I have my fill of feasting, and no less of wine’ (satias dapis me nec minus Bacchi tenet, Thy. 973).

As an allusion to the play’s theme of transgressive consumption, the satis-motif is well recognised.¹⁰⁹ It has also been studied as a technique of rhetorical/generic amplification with equal degrees of insight.¹¹⁰ But its other associations have so far gone unremarked. Its evocation of psychological fulfilment, for instance, relates directly to Atreus’ pursuit of vengeance, his desire to receive recompense, to achieve ‘payback’ (satisfacere), and his nagging feeling that no penalty, no matter how severe, will ever erase this sense of injury. Although the term satisfacere does not feature in the Thyestes, the avenger’s repeated use of satis suggests its underlying presence.¹¹¹ Similarly, the thematic centrality of food is something this play shares with broader Western cultural definitions of vengeance, many of which employ metaphors of alimentary overindulgence: the avenger experiences his lack as hunger, strives for fulfilment, and, frequently, suffers from dissatisfaction at the end.¹¹² What is specific to Atreus and to the play’s plotline is generic to vengeance itself, something the Thyestes’ satis-motif suggests by alluding not just to food, but to food as punishment and desire as physical appetite. Atreus sets out to sate his anger, to satisfy his soul by filling Thyestes’ body, to give his brother, too, a bitter taste of victimhood. And his own fullness remains uncertain even at the play’s end, a circumstance he shares with many tragic avengers. Revenge resists closure, and its perpetrators rarely feel replete.

In this last regard, the Thyestes’ satis-motif is also a barometer of Atreus’ autonomy, for it suggests his continued enslavement to desire even when revenge has been brought to completion. Successful retaliation may ensure Atreus’ sexual domination, but does it, can it, ever guarantee his freedom from the impulse of vengeance itself, which tends towards addiction and imprisons its protagonists in an endless loop of wanting more. Like Tantalus’ hunger, revenge resists gratification; its innate excessiveness always admitting the possibility of going further still and committing a yet more perfect crime.¹¹³ Although Seneca’s Atreus appears content with his final achievement (Thy. 1096–9), there remains a lingering sense that he could have engineered an even greater atrocity: he could have forced the gods to watch (Thy. 893–5), forced Thyestes to commit cannibalism knowingly (Thy. 1053–6; 1065–8), could, perhaps, have restored his marriage in actual rather than rhetorical fact (Thy. 1098–9). Every shortfall indicates a curtailment of Atreus’ agency, as though all of his sweeping anger could not overcome the barricade of its own ineradicable presence. This is a crucial caveat to my arguments about the avenger’s agency: retaliation impedes autonomy just as it confers it; no matter how powerful the avenger becomes, he or she must eventually reckon with anger itself. At the same time, though, this reckoning underscores the avenger’s dominance. The sheer magnitude of Atreus’ aspirations, their proximity to the impossible, emphasises the extent to which his agency already reaches: he may not be able to compel the gods’ return, but he has caused their flight. Essentially, the play’s theme of satisfaction illustrates the enormity of Atreus’ power by highlighting its occasional limitation. It also confirms his autonomy by establishing a contrast between active aggressor and passive victim, between the one who demands satisfaction and the one who will pay for it, for although Atreus himself is consumed by rage, he is the one in charge of Thyestes’ more literal consumption: the passive satiaberis (Thy. 980) shows all too clearly where the balance of power lies.

In its evocation of food and fullness, satis also contributes to Thyestes’ themes of sexual anxiety and the gendered active/passive binary. Interestingly, Petronius (75.11 and 87.1)¹¹⁴ uses satisfacere in reference to sexual gratification, which raises the tantalising possibility of vengeance tout court embodying a sexual act. Of course, Atreus’ revenge does have a sexual dimension, but that is due to the specifics of plot, it seems, rather than to the individual quality of revenge itself; there is too little evidence to be certain on this point. A more fruitful set of parallels, however, appears in Plautus’ Amphitruo, where satis is used to characterise Alcmena’s insatiable sexual appetite: she complains in her canticum of pleasure’s insufficiency (Am. 633), and Jupiter, in their first scene together, asks why she cannot be satisfied with his love of her (Am. 509).¹¹⁵ Jupiter’s own indulgence is called satietas at Am. 472, and when the offended Alcmena seeks an apology from Amphitruo, her request – satis faciat mi ille (‘he must make amends to me’, Am. 889) – acquires undeniable sexual connotations in light of her preceding characterisation. While I do not propose that Plautus influenced Seneca directly,¹¹⁶ this comedy’s use of satis is a helpful measure of what occurs in the Thyestes: besides configuring Atreus’ retaliation as hunger, satis also configures it as lust, a burning desire that Atreus struggles to gratify. Further, Atreus’ persistent yearning seems to place him in a sexually subordinate role, as implied by the parallel of Alcmena, and by Greco-Roman cultural norms, which tend to associate sexual insatiability with feminine lack of control. This fact of Atreus’ lust, like the fact of his anger, represents a potential check to his free agency.

Unlike anger, though, the check seems temporary, because, as I explore above, Atreus manages to trap Thyestes into ‘pregnancy’. The protagonist’s lack of satietas is cancelled out by his antagonist’s surfeit: satis finds its echo in Thyestes’ being ‘stuffed’ (satur est, Thy. 913). Comparison with Plautus’ Alcmena is instructive here, too, because the same term is used to describe her bulging pregnant belly, in the context of a joke about food (Am. 665–8). Just as Thyestes’ overeating resembles gestation, so Alcmena’s gestation resembles overeating. The parallel highlights Thyestes’ increasing feminisation across the course of the drama, and the concomitant reinstatement of Atreus’ dominant masculinity. Rage may not admit of full satisfaction, but lust has a generative telos.

Although a more generic form of identity than Medea’s, Atreus’ manhood is nonetheless pivotal to the sense of self he seeks to recalibrate in revenge. It is an integral part of his social position as a father, as the head of a household, as an autocrat, that he appear sexually powerful.¹¹⁷ He says as much in his opening collage of the expectations inherent in a tyrant’s role (Thy. 176–8): the list shows clearly what Atreus wants to be, and further, that vengeance is his chief means of achieving this identity. His political and sexual statuses largely coincide in a relationship of mutual reinforcement, so that his securing of power within his immediate domestic sphere (domus) confirms the authority and prestige of his lineage (domus) as a ruling family.¹¹⁸

At the epicentre of these themes is the question of paternity, which exemplifies Atreus’ combined ambition for domestic and social control, and over which Thyestes’ adultery has cast a long shadow. Atreus begins the play in doubt over his sons’ parentage: he worries about their ‘dubious bloodline’ (dubius sanguis, Thy. 240) and hopes for ‘proof of uncertain paternity’ (prolis incertae fides, Thy. 327) by involving Agamemnon and Menelaus in his plot. First, he aims to determine his sons’ loyalty by disclosing his full intent, reasoning that if they defend their uncle, they must in fact be his offspring (Thy. 328–30). No sooner has he devised this primitive DNA test, however, than he discards it for fear that his children will betray him even unwittingly, through the apprehension writ across their faces (Thy. 330–1). Instead, he treats Thyestes’ cannibalism as confirmation of bilateral legitimacy, quipping gleefully that Tantalus, Plisthenes and the un-named third boy are ‘definitely’ Thyestes’ sons (certos, Thy. 1102) and concluding the same for his own, in a passage whose ‘mixture of logic and sheer delusion’¹¹⁹ still defies scholarly subdual:

     nunc meas laudo manus,

nunc parta vera est palma; perdideram scelus,

nisi sic doleres. liberos nasci mihi

nunc credo, castis nunc fidem reddi toris

     now I praise my handiwork,

now the true palm of victory is won; my crime would have been wasted

had you not grieved. Now I believe the children are mine,

now trust and chastity have been restored to my marriage

(Thy. 1096–9)

Schiesaro attempts to untangle this claim by proposing that a) Thyestes’ grief proves his parentage, for he would not, presumably, lament what was not his, and b) disproves his parentage of Agamemnon and Menelaus, for he would, presumably, grieve less if he knew he still had sons remaining; ergo the children belong to Atreus.¹²⁰ It is a feasible proposition, especially given Atreus’ later assertion about Thyestes’ wanting to commit an equivalent crime but refraining because he thought the children were his (hoc unum obstitit: / tuos putasti; ‘one thing stopped you: you thought they were yours’ Thy. 1109–10).

The logic remains tortuous, though, and its rationalisation merely deflates Atreus’ powerful rhetoric.¹²¹ These claims do not arise from careful calculation but from the vicious glee of payback: Atreus feels assured of his paternity because he is victorious tout court, because victory guarantees his dominance and re-establishes his manhood (he must be the father: he is powerful and Thyestes weak), because successful vengeance grants him the power of self-determination and self-creation (he can be whatever and however he wishes), and because his political supremacy imbues his words with an almost performative quality (what he says goes). His sexual and domestic ascendancy is confirmed more through symbolism than through coherent reasoning, such as when he turns Thyestes into a quasi-female vessel. Likewise, he establishes Thyestes’ paternity via equally symbolic means, as the father’s ingestion of his sons represents an indissoluble merging of genetic and corporeal substance.¹²² From a rational perspective, it may appear that Atreus has slim grounds for insisting on the legitimacy of his sons and the faithfulness of his marriage, but the point at issue here is that Atreus dictates reality, not the other way around. The tyrant’s agency and autonomy are so vast that logic cannot restrain them (nor can time, for that matter; Atreus claims to have reversed its effects). Thus, through vengeance, Atreus assumes the authority to shape the world around him, and to shape it to his advantage.

Such autonomy breeds isolation. Thyestes 1096–9 shows Atreus asserting power over his family members – the power to reconstitute and redefine them according to his will – at precisely the moment he steps free from their messy interpersonal nexus. Although he declares his marriage true and chaste, he has not reconciled with his wife, who, for all her thematic significance, does not even feature as a character in the play. His children are similarly instrumental, important only to the extent that they reflect on him, as living symbols of his virility and dominance. He has, of course, cast aside any remaining ties to his brother. As is the case for Seneca’s Medea, and indeed for much of Seneca’s approach to matters of freedom, destructive solitude is the ultimate guarantee of Atreus’ individual autonomy. In the words of Gordon Braden, for Seneca’s tragic characters, ‘the devastation of emotional ties is an unanswerable gain of power and control’.¹²³ In releasing Atreus from the danger of subjection to another person, vengeance cuts him loose from the sustaining bonds of human society. Just as the tyrant finds freedom in his radical separation from those he rules, and just as the avenger finds freedom in slicing through the social ties that ensnare him, so Atreus exalts in having reached the lonely pinnacle of self-sufficiency and individual sovereignty, states that he (and, arguably, Seneca too) imagines existing beyond the reach of societal and legal norms. Freedom, for Atreus, is just another word for tyranny, for revenge, for murder.

Political supremacy is another core aspect of Atreus’ autonomy, one I have so far remarked upon only in passing. Accompanying his sexual and domestic potency, it, too, is strengthened by vengeance because even though Atreus begins the tragedy already enthroned in Argos and already capable of violent coercion, nonetheless he views his rule as unstable, under threat from Thyestes’ nefarious scheming. Atreus insists throughout the play that his brother is his mirror image, devising the same crimes (Thy. 193–5; 201–4; 314–16; 917–18; 1105–9) and coveting the throne with the same intensity (197–9).¹²⁴ The imputations verge on paranoia, especially given that Thyestes appears in person weak and gullible, the opposite of Atreus’ conniving cleverness. Still, his brother’s previous usurpation makes Atreus wary of future attacks. Thyestes’ prior theft of the talismanic ram, along with Atreus’ wife, Aerope (Thy. 222–35), is an act of public and private sabotage that disrupts the Atrean domus on a political as well as personal level. Atreus has experienced exile at his brother’s hands (Thy. 237) and even though he has since regained control in Argos, the mutual nature of their conflict prevents him from feeling secure.

Consequently, while in comparison to Seneca’s Medea, Atreus does not pursue vengeance from a position of total social or political marginalisation, even so he rejoices that his completed act of retaliation guarantees his absolute rule. ‘Oh I am the highest of heavenly beings, king of kings’ he crows upon witnessing Thyestes’ meal (o me caelitum excelsissimum / regum atque regem, Thy. 911–12), in celebration of a power so total that it brings other rulers and gods under its sway. Later in the same Act, he announces to Thyestes, ‘This is the day that confirms my kingship and establishes the solid trust of definite peace’ (hic est, sceptra qui firmet mea / solidamque pacis alliget certae fidem, Thy. 971–2). Situated in the false context of reconciliation, the statement extends the promise of political harmony achieved through the brothers’ newfound unity, with alligo hinting further at ties of kinship and affection.¹²⁵ Atreus’ real meaning, however, is that he has restored his own supremacy by neutralising Thyestes’ political threat (namely, by removing his heirs and souring his appetite for power). Lurking underneath the lines’ suggestion of plurality and co-operation is the tyrant’s obsessive drive for solo control. The ambiguity of Atreus’ rhetoric likewise illustrates his power over language and hence, over the very shape of the world around him. Whereas Thyestes cannot seem to extricate himself from lexical and rhetorical conventions, Atreus is their undisputed master – another gauge of his autonomy.

Finally, Thyestes 971–2 is also notable for its conflation of political with sexual dominance, because the terms certus and fides convey distant echoes of Atreus’ cuckoldry and subsequent doubts about paternity (cf. Thy 327 prolis incertae fides and Thy. 1099 fidem … toris, above), while the sceptrum functions in Senecan tragedy as a symbol of patriarchal (political, sexual) power.¹²⁶ Consolidating his hold on the Argive throne is what enables Atreus to feel assured of his masculinity and position within the family: one form of autonomy ensures the other.

In sum, Atreus demonstrates his agency and autonomy by using revenge to wrest back control of his identity and to alter his situation for – as he sees it – the better. Like Medea, he pursues vengeance as a means of self-transformation and self-creation, riding its swelling tide from a position of relative passivity to one of absolute sovereign dominance: from quasi-feminine to fully masculine, from political insecurity to perfected tyranny. Implicit in the roles he assumes is the power to dictate the shape of the world around him, a power he exercises on everything from the heavens to the form of Thyestes’ body. More fundamentally still, Atreus’ retaliation ensures his agency because vengeance itself is ‘a thrust toward action’,¹²⁷ a transition from endurance to perpetration, as Seneca signals so clearly in Atreus’ opening monologue. In all of these respects, revenge emphasises Atreus’ quasi-human features, namely his capacity for independent action, his assertive subjectivity, his (illusion of) contingent selfhood. But this very position of control also draws attention to its own limitations, which marks Atreus’, and all literary portrayals of vengeance as a straining of human capability against the restrictions of fictional form. The ensuing section takes up this question of textual identity to explore how Atreus’ genealogical and literary background propel him into action.

A History of Violence

Atreus’ revenge reveals its explicitly fictional dimension in a multitude of ways: it prompts the protagonist’s resemblance to a playwright/director; it gestures self-reflexively to the very genre of tragedy and to vengeance as one of its principal conventions; it flags Atreus’, and Seneca’s, debt to prior literary models; it acknowledges, by enforcing, characters’ circumscribed autonomy. The first of these items has been treated already in considerable detail by Alessandro Schiesaro, Cedric Littlewood, and most recently, Anthony Boyle.¹²⁸ In brief: Atreus’ manipulation of his brother corresponds to a director’s handling of rehearsal and performance processes. Not only does Atreus devise a role for Thyestes, in the manner of a playwright, but he also orchestrates a reconciliation, commands his brother to change garments/costume, frames the feast as an inset performance and gazes upon it as a satisfied spectator (libet videre; ‘it is pleasant to watch’ Thy. 903) before entering to take part in the denouement. The nefas Atreus plans, perfects, and commits is the very substance of the play; in fashioning his revenge, he all but writes the Thyestes itself, in addition to embedding his own drama within its bounds.¹²⁹

Themes of literary and generic self-consciousness have likewise received an ample share of scholarly scrutiny, but I revisit them here because they encapsulate perfectly the conflicting dynamics of Atreus’ revenge, and because they have rarely, if ever, been measured against the contours of vengeance itself, its very nature as a human act and a fictional trope. I note in my preceding analysis the tendency for Atreus’ vengeance to generate dissatisfaction and yearning; another crucial formulation of these emotions, which accompanies the satis-motif, is the tragedy’s frequent recourse to magnus and maior, alongside more diffuse expressions of magnitude. Atreus characterises his desire for revenge as a persistent longing to exceed limits and achieve something greater than what has gone before. I have already had occasion to quote Thyestes 253–4 – impleri iuvat / maiore monstro; ‘I long to be filled with greater monstrosity’ – in the context of discussing the play’s gender roles; the lines also announce the excessiveness germane to all acts of retaliation and, at a more specific level, the relationship of Atreus’ own revenge to its prior fictional and genealogical instantiations. maior is the key word in this regard, and it is multivalent. First, as a metapoetic marker, it signals the ‘lofty’ genre of tragedy and Atreus’ self-conscious operation within it.¹³⁰ In wanting to commit greater outrages, Atreus expresses a metaliterary desire both to act in accordance with his given genre and to outdo all prior dramatisations of revenge. Second, and relatedly, maior at Thyestes 254 signals the specific intertext of Accius’ Atreus, who declares of his brother’s heralded attack:

iterum Thyestes Atreum adtrectatum advenit;

iterum iam adgreditur me et quietum suscitat.

maior mihi moles, maius miscendumst malum

Once again Thyestes comes to assault me;

now, again, he attacks me and rouses me from my rest.

Greater trouble for me, a greater crime to concoct

(Atr. 198–201 Ribbeck TRF²)

Excessiveness and repetition are present already in the Accian version,¹³¹ as the threat of Thyestes’ renewed hostility reiterates the brothers’ earlier confrontation – when Thyestes wrested the throne from Atreus (cf. Sen. Thy. 222–37) – as well as reiterating this well-known story’s representation in a poetic text. Reiteration is built into the myth’s plotline and reinforced, metapoetically, by its multiple treatments on the tragic stage. Concomitantly, Accius’ Atreus aspires to surpass his prior mythological conflict with Thyestes and prior fictional instantiations of his trademark aggression. And Seneca’s Atreus aspires to surpass even this already overdetermined claim to greatness.

The third important feature of Seneca’s maius-motif is its evocation of literal as well as literary genealogy. Paul Hammond remarks that maior at Thyestes 254 calls to mind the ancestral quality of Atreus and Thyestes’ hatred, the maiores from whom the present conflict originates and to whose models Atreus looks for inspiration (Thy. 242–3).¹³² Tantalus in particular provides not just a template for Atreus’ behaviour but also direct influence in the form of his ghostly presence, which sets the play’s events in motion. This ancestral background functions alongside the lines’ references to poetic tradition, in highlighting Atreus’ enmeshment within larger structures of causation and agency. For all of his personal sovereignty, he cannot avoid being an instrument of forces greater than himself, whether those forces comprise the myth’s pre-established patterns and its prior poetic handling, Seneca’s authorial control (likewise conditional upon literary precedent), or the Pelopid family’s curse and its seemingly genetic predisposition for alimentary wickedness. As the meeting point of all these vectors, Atreus’ revenge begins to seem unavoidable and pre-determined, not the wild cry of independence that I (and on many occasions, Atreus himself) have characterised it as being. It becomes instead the embedded textual act of a patently fictional entity.

Such intimations of fictionality are not confined to one or two lines, either, but traverse the entire play, as Seneca’s language from the very first scene indulges in a rapid spill of comparatives and images of excess. ‘Let Thracian impiety happen in greater number’, (Thracium fiat nefas / maiore numero, Thy. 56–7) roars the Fury in simultaneous allusion to the play’s Ovidian intertext of Procne and Philomela (Ov. Met. 6.424–674), and to this myth’s (re)transposition into the genre of tragedy (with numerus indicating not just the number of children Atreus will sacrifice, but also the elevated metre of tragic drama).¹³³ In echo of this hellish prologue, Atreus himself promises to perpetrate ‘something that does not cleave to the limits of ordinary pain’ (nil quod doloris capiat assueti modum, Thy. 255), describes his ‘mind swelling with something greater, larger than normal, beyond the boundaries of human custom’ (nescioquid animus maius et solito amplius / supraque fines moris humani tumet, Thy. 267–8) and confesses that he must perform ‘something greater’ (maius … aliquid, Thy. 274) than Procne did against Tereus. Upon hearing all of this, Atreus’ minister shudders that the intended crime ‘is greater than anger’ (maius hoc ira est malum, Thy. 259). These examples pair the tyrant’s and the avenger’s shared impulse for over-reaching with expressions of metaliterary competitiveness. Atreus qua fictional construct acknowledges his confined place within a poetic tradition and admits the pressure he feels to distinguish this particular version of his story from other, similar versions, whether those versions involve him or other figures (e.g. Procne) engaged in acts parallel to his own.

As always, though, Seneca’s portrayal of autonomy is more complex than mere metapoetics. While it is undeniable that Seneca employs the maius-motif to signal his rivalry with earlier poets and to enrich his own text by echoing them, nonetheless Atreus’ desire to surpass, simultaneously, his predecessors and the confines of possibility is also emblematic of the avenger’s bid for total control over his opponent and his circumstances. maius for Atreus represents the ‘more-ness’ of vengeance, and its attendant magnification of the perpetrator’s agency. Essentially, the motif articulates a definition of vengeance itself, not just Atreus’ particular pursuit of it, for retaliation is, by nature, an excessive act. Payback is not a matter of pure equation or recompense, but overpayment, as the victim attempts to extract compensation for things that can never be recovered or repaired, whether they include deceased friends / family members or, at a more basic level, the very fact of one’s own past suffering.¹³⁴ Atreus freely admits that his vengeance must go beyond anything Thyestes has done to him: ‘you do not avenge crimes unless you outdo them’ (scelera non ulcisceris, / nisi vincis, Thy. 195–6), he remarks in Act 2, half to the minister, half to himself. The idea reappears in Act 5, when Thyestes complains of discomfort in his belly:

Thy:       genitor en natos premo

premorque natis; sceleris est aliquis modus

Atr: sceleri modus debetur ubi facias scelus,

non ubi reponas.

Thy:      Look, I, the father, crush my sons

and am crushed by them; the crime has some measure

Atr: A crime needs measure when you commit it,

not when you repay it.

(Thy. 1050–3)

This terse exchange encapsulates perfectly the dynamic of revenge. Thyestes’ description of his physical state – equal parts himself and his sons – presumes the equilibrium of payback and at the same time, indicates its excessiveness, its blurring and overturning of boundaries. His comment in 1051, sceleris est aliquis modus, can be read either as a gauche reference to the symmetry of his bodily suffering,¹³⁵ or – as it is sometimes translated – a vain plea about crime’s limits.¹³⁶ Actually, both meanings are present, and their combination points to revenge’s uneasy union of balance and immoderation.

It is a union Atreus comprehends to the core, as he rebuffs Thyestes’ complaint with, effectively, an explanation that despite its ‘re’ prefix (re-venge; re-ponas), retaliation is no mere ‘equal and opposite reaction’. Though talio assumes an eye-for-an-eye exchange, the reality is more like two, or even a whole face, for one. In an innovative twist on the theme of curtailment and boundary violation, Atreus proposes that the original perpetrator, not the avenger, is the one most limited in his criminal activity, presumably because the originary offence is not compelled to push beyond a prior model, whereas vengeance is, by definition, responsive and competitive in its drive to replicate while outdoing the very event to which it owes its existence. Revenge – Atreus’, anyone’s – outstrips established parameters in a manner that is simply not incumbent upon the initial crime.

It is no accident if this competitive repetition sounds a lot like intertextuality. To put it another way: the poetic imitation ingrained in Atreus’ act is matched by, and finds direct reflection in, the avenger’s need to repeat and outperform his predecessor’s moves. Revenge’s combination of recurrence and innovation, confinement and transcendence, resembles the challenge faced by poets – and poems … and characters – situated within an established tradition which they must acknowledge and defy, as these respective needs arise. The same union of opposites underpins the fictional avenger’s agency: imitation of the past threatens his/her capacity for self-directed action at the same time as stimulating the desire to break free. A injures B, and B, by replying, cedes autonomy to the discourse that A has formulated and within which all subsequent action must occur. But, by expanding on A’s initial deed, B asserts the ability to advance beyond mere copying and into a realm of self-conferred sovereignty. Paradoxically, it is this very dependence upon past paradigms that secures a kind of freedom for the avenger, at least in the sense that he/she can see where potential limitations lie and can work to overturn them.

In Atreus’ case, the pre-established parameters that guide his vengeance include not only Thyestes’ initial crime, but also the intergenerational wickedness of the Tantalid dynasty, the existence of comparable acts of vengeance in other mythological narratives, the existence of other tragedies dealing with the same topic, and the pre-determined nature of revenge as an event intrinsic to the tragic genre. Atreus knows from the outset that he is in a revenge tragedy, and consequently, that he must fulfil the demands of this particular artform (cf. Thy. 176–80).¹³⁷ All of this may appear to quash any possibility of his having free agency or claiming independent action, but the reality is actually more complex than this, because these forces promote Atreus’ conduct even as they restrict it. Essentially, Atreus’ recourse to prior examples, his reflexive awareness of genre and the self-consciously literary texture of his thought need not, per se, negate impressions of his quasi-human autonomy. His fictional and implied human identities are not opposites, in this instance, but two sides of the same coin. Atreus wants to perform greater crimes not just out of rivalry with earlier literature, but also because the extent of his transgression verifies the extent of his power. Flouting modus is central to his expression of agency, but in order for this to occur, a modus must first be established. Audiences can appreciate Atreus’ push towards domination only once they have recognised the very dynamics of that domination in Atreus’ own sense of victimhood.

This intersection I propose between the mechanics of revenge and those of literary appropriation has important consequences for our understanding of fictional character in general, not just of Seneca’s Atreus. Fundamentally, it demonstrates that fictional beings may seem quasi-human when they are at their most metapoetic, and vice versa. A character’s status as a textual construct not only does not preclude its equally significant status as an implied human personality but may actively give rise to it. This is not a question of ‘either/or’ but ‘both/and’. We may take vengeance, broadly speaking, as an event that emphasises characters’ person-like aspects, chiefly because it is a personal act: the avenger takes his or her injury to heart – this is a deed meant and received in deeply personal terms – and targets specific individuals in return. Vengeance is meaningful violence and meaningful suffering; it is not directed against anonymous, faceless groups or institutions, nor is it performed by them. Although it may be perpetrated by groups, vengeance in literature is typically a solo act that distils in a single figure pressing human problems of justice, self-determination, and moral choice (to name but a few). Its emphasis on action and on change invites writers and audiences alike to frame the avenger in human terms despite his or her purely fictional ontology.

The consciously literary texture of much fictional vengeance – its awareness of poetic and/or dramatic tradition; its inclination for performance and role-play; its similarity to the very act of literary composition – may seem to override its ‘personal’ quality, but the dichotomy is false, as is the broader dichotomy from which it derives, namely that of character-as-text versus character-as-person. Granted metapoetics, metatheatrics, and intertextuality, by signalling characters’ fictional fabric, prevent us from assuming too much of their motives and psychology, but these very conventions also work to frame characters’ motives and psychology in the first place, to deepen their significance by placing them within a tradition. The same applies to audiences’ emotional investment in and response to given fictional scenarios: while metapoetic techniques may seem, on the one hand, to curtail these by minimising readers’/viewers’ sense of immersion in the events portrayed, they can, on the other hand, activate such immersion – at base, simply by imposing shape and form on what might otherwise appear meaningless, random activity. If the avenger resembles a poet/playwright in plotting retaliation, and an actor in executing it then, yes, self-awareness of fictionality may generate a certain amount of audience detachment. But it can invite involvement, too, because, as I have outlined above, the avenger’s scheming focalises events from his/her perspective, encourages audiences to sympathise with his/her grievances, affirms the character’s capacity for independent action, and contributes to illusions of contingency. It fosters the avenger’s self-assertion as much as it potentially denies it. These two facets work in tandem; characters’ textual ontology cannot really be divorced from their implied humanness, and vice versa.

A brief coda about Atreus, in light of these remarks: although vengeance against Thyestes springs from a nexus of factors, not all of which are under Atreus’ direct control, it is nonetheless individually, personally meant. However much Tantalus and the Fury propel him into action, however much Procne and Philomela (and Ovid) provide him with a template, however much his actions cleave to conventions of tragic vengeance, the brutality he visits upon his brother amounts to so much more than the outcome of an impersonal or supra-personal system that uses Atreus as its instrument. Literary and genealogical inheritance may influence the form of Atreus’ revenge, but they do not occlude his (relatively) autonomous performance of it. The subjectivity he (re)claims through revenge is just as much (if not more) a matter of being active and in command, of bringing matters under his control, as it is of being subjected to external forces, people, and sources of motivation.¹³⁸ Likewise, his hatred of Thyestes is not just a mythological or literary datum, nor the inevitable result of family feuds; it is these, granted, but it is also an immediate, passionate, implied human response to the circumstances in which Atreus finds himself and to the individual he credits with creating them. To forget this latter dimension is to bleach Atreus’ project – and any avenger’s, for that matter – of its most fundamental colours, and to see only half of the story. Served hot or cold, vengeance is the discriminate means of satisfying individual grievance. As the saying goes, it’s personal.

4.3 Suicide

Self-Enforced Endings

From vengeance I move to suicide, as a complementary and likewise deeply Senecan expression of individual agency. Though they may seem like odd bedfellows, the two acts actually have a lot in common. Like the avenger, the Senecan suicide takes action from the margins, and achieves in self-inflicted death a vital demonstration of freedom and self-government. Both acts also usurp the law, a similarity particularly apparent in the case of Roman political suicide, which often took the place of state-sponsored punishment.¹³⁹ Both represent the dissolution of oppression and victimhood, and the aggressive reinstatement of subjectivity in response to intolerable circumstances. Both strive to achieve dominion, over others or over the self. Whereas the avenger pushes outward to destroy those around him/her, the suicide turns inward to the task of making the self inviolate from all future attacks, material or immaterial. These parallels, combined with the scope of suicide’s portrayal in Seneca’s work, make the motif particularly apt for my present study of characters’ fabricated versus quasi-human identities. It is also an appropriately closural gesture for this chapter and for the book overall, a self-chosen, self-directed ending that lends additional significance to many of my preceding arguments and promises freedom for reader and writer alike.

Let us begin this final episode by glancing back, briefly, to the Thyestes. When the victim understands the true contents of his meal, his first thought is to free his progeny from entombment in his belly:

volvuntur intus viscera et clusum nefas

sine exitu luctatur et quaerit fugam:

da, frater, ensem (sanguinis multum mei

habet ille): ferro liberis detur via.

negatur ensis? pectora inliso sonent

contusa planctu—sustine, infelix, manum,

parcamus umbris.

My guts churn inside me and an enclosed evil

struggles with no way out; it seeks escape:

give me your sword, brother (it has much of my blood

already): let the blade grant passage to my children.

You refuse? Let this bruised breast resound

with sorrowful blows – stay your hand, wretch,

spare the dead.

(Thy. 1041–7)

The grotesque gesture amounts to a suicidal wish: Thyestes will spill his guts, in the process releasing what remains of his children and, just as vitally, releasing himself from the prospect of future suffering and oppression. If his current state represents a diminution of his autonomy and confusion of his individual agency, the prospect of suicide represents their dramatic reinstatement: Thyestes taking charge of his own fate, wresting control from Atreus. Though the lines themselves exhibit an undeniable debt to Ovid’s Tereus, who similarly responds to revelations of paternal cannibalism by contemplating disembowelment (Met. 6.663–4),¹⁴⁰ the statement in Thyestes’ mouth is wholly Senecan. It is, as Brad Inwood observes of Seneca’s overall notion of suicide, ‘a mark of agency even amidst misfortune’.¹⁴¹ The difficulty for Thyestes is that the contents of his stomach render him incapable of self-determination, at least symbolically. For Thyestes is simultaneously himself and alien to himself; he and his children have blended together; he is not physically discrete and cannot act purely in his own interests. In a perceptive if harrowing comment on the nature of blood ties, Seneca shows how biological proximity limits Thyestes’ capacity for autonomous action. If he cannot hurt himself without also ‘hurting’ his offspring, then Atreus has attained a complete victory over him, having foreclosed even the consolatory possibility of agency proffered by suicide.

Bolstering this interpretation of Thyestes 1041–7 is the famous ‘ode to suicide’ at de Ira 3.15.4, which follows directly upon the tale of Harpagus’ unwitting cannibal banquet.¹⁴² Astyages, furious at what he perceives as Harpagus’ disloyalty, cooks and serves up the latter’s sons. Upon the meal’s completion, Astyages orders the children’s heads to be brought in and presented to their father, whereupon he asks Harpagus what he thinks of his reception. Harpagus suppresses his anger and responds with flattery, a reaction that elicits first guarded approval from Seneca, then an outburst in praise of suicide’s liberating power. The endless possibilities of death by one’s own hand are presented as ‘escapes from servitude’ (effugia servitutis, Ira 3.15.4) and, by implication, as opportunities, however terminal, for autonomous action. Harpagus may be oppressed by a dominus (Ira 3.15.4) – and it is significant to Seneca’s point that Harpagus lives under an autocratic regime – but suicide would enable him to regain control over himself, his life, his circumstances. Parallels to the Thyestes are obvious. It could even be said that Seneca treats the theme more intricately in the tragedy than in the dialogue, because he acknowledges the complex, thwarted agency of one who has just consumed the products of his own flesh.

Preoccupation with agency and autonomy underpins much of Seneca’s thinking about death,¹⁴³ an association that contributes, in turn, to his vision of the sapiens’ competitive autarky. Of course, suicide in Seneca is a well-trodden topic¹⁴⁴ that I do not intend to reprise here. Instead, I present a swift overview of its relationship to self-determination and self-government in Seneca’s work, as a prelude to considering its bearing on characters’ autonomy in the tragedies.

The paradigmatic case for Seneca, and arguably, for all aristocratic suicides in imperial Rome, is Cato the Younger, whose act is celebrated, variously, as a consummate gesture of political and spiritual freedom, of moral virtus and self-control, of bravery and aristocratic dignity, of Stoic contempt for ‘indifferents’, and of good old-fashioned republican defiance of an aspiring dictator.¹⁴⁵ While Seneca’s recurring, multifaceted treatment of Cato touches upon all of these elements to some degree, the most relevant to my present discussion is this combined concept of political and personal libertas. Basically, in opting to die rather than endure the prospect of Caesar’s clementia, Cato trumps the ruler’s claim to absolute sovereignty.¹⁴⁶ Like the Senecan sage who outmanoeuvres the tyrant, Cato slips through Caesar’s grasp, thereby demonstrating its limitations and the superiority of his self-conferred autarky in opposition to Caesar’s self-conferred autocracy. In de Providentia 2.10–11, Seneca’s Cato declares that his sword will grant him the (individual) freedom it could not grant his (collective) fatherland, and muses that it is just as disgraceful for him to seek death at another’s hands as it is for him to seek life. Both assertions pivot around issues of individual sovereignty, as Cato counters the now inevitable fact of his political subjugation with the ultimate and absolute subjective agency of self-imposed death. This is a particular kind of freedom, situated at the intersection of public and private spheres, the freedom to be as and what one chooses, even if that ‘being’ means, finally, not being. Catherine Belsey remarks of suicide in Renaissance drama that it ‘re-establishes the sovereign subject’ and fulfils the subject’s desire for autonomy, ‘to be not just free, but also the origin and guarantee of its own identity, the source of being, meaning, and action … In the absolute act of suicide, the subject itself is momentarily absolute.’¹⁴⁷ The same judgement may profitably be applied to Cato in the de Providentia, who presents his self-inflicted death as the essence of what it means to be ‘Cato’,¹⁴⁸ and who frames his suicide as an explicit challenge to Caesar’s absolute power. By depicting Cato’s death as the alternative to political liberation (Prov. 2.10), Seneca hints at equivalence between terrestrial control and spiritual mastery: the sage emulates and outstrips the ruler’s role, albeit only in the singular realm of the self.

Seneca further emphasises Cato’s agency by calling him ‘a most fierce avenger of himself’ (acerrimus sui vindex, Prov. 2.11), a striking phrase that twins the projects of suicide and retaliation, imputing to each a commensurate degree of freedom. Just as the avenger reinstates his or her autonomy via retributive violence, so the suicide discovers the purest form of agency in visiting violence upon the self; self-destruction is revenge, in another sense, for the position one has been placed in and for what one has been made to suffer.

In this regard, the political aspect of Cato’s suicide is relevant only to the extent that it illustrates his autonomy. What matters is not Cato versus Caesar or republic versus prospective dictatorship, but, more fundamentally, one man assuming the power to dictate another’s fate. While Seneca is undeniably alert to the event’s historical background,¹⁴⁹ and to the specific political implications of aristocratic suicide (which he himself will commit), the concept that unites this anecdote with his other portrayals of suicide is more basic: it is about the capacity to choose and control the form of one’s death, thereby escaping oppression and victimhood.

Examples of this idea abound in Seneca’s prose works, especially in the Epistles, where Seneca, writing in old age, is even more than usually concerned with death.¹⁵⁰ Epistle 26.10, for instance, pairs death with emancipation: qui mori didicit servire dedidicit; supra omnem potentiam est, certe extra omnem. Quid ad illum carcer et custodia et claustra? Liberum ostium habet. (‘one who has learned how to die has unlearned servitude; he is above all external power, certainly beyond it. What do prison and guards and bars matter to him? He has a way out.’) That final image of the liberum ostium, so difficult to capture in translation, combines concepts of physical escape – from prison, from one’s own body – with social enfranchisement, the condition of no longer being subjected to another’s potestas. Self-inflicted death is the equivalent of terrestrial autonomy. Seneca reprises the idea at Epistle 51.9: Libertas proposita est; ad hoc praemium laboratur. Quae sit libertas quaeris? Nulli rei servire, nulli necessitati, nullis casibus, fortunam in aequum deducere … ego illam feram, cum in manu mors sit? (‘Freedom is placed before me; I work towards that reward. And what is freedom, you ask? It means not being enslaved to any circumstance, to any need, to any chance; it means engaging with Fortune on equal terms … shall I endure her, when death is within my reach?’). Here the Stoic concept of ‘indifferents’ is used to illustrate a freedom at once social and spiritual, mundane and transcendent, for the individual who manages to reject the false lure of earthly concerns (res, necessitas, casus) and to view death with detachment, as neither intrinsically good nor evil, resembles a fully enfranchised member of society, not beholden to anybody or anything else.¹⁵¹ Death as physical escape is matched with the sage’s less tangible freedom from life’s ups and downs: both enable the individual to evade others’ grasp, and to claim sovereign dominion over his/her own existence.

Fascination with personal autonomy is also the reason why so many of Seneca’s anecdotes about suicide focus on slaves or prisoners of war. Epistle 70 tells the story of a German bestiarius who escaped performing in the arena by choking himself with a toilet brush (Ep. 70.20–1), and of another gladiator who committed suicide on the way to his morning munus by drooping his head in feigned sleep until it was caught and broken in the wagon’s spokes (Ep. 70.23). Seneca accords these examples significance for proving that anyone, no matter how lowly, can defy death (Ep. 70.19), but his choice also springs from a deeper, unexpressed premise, namely that gladiators are prime examples of social disenfranchisement and objectification. The gladiator resembles a slave (and often is one) in the lack of control he exercises over his own body and person. His subjectivity and agency are compromised to the point of being almost non-existent. His very job allows him to be wounded or killed with impunity and even if he is the one doing the killing, he does so in obedience to his trainer, editor, and the rules of the game. Outside the arena, he is infamis: open again to being beaten with impunity, debarred from public office, and unable to bring lawsuits or to represent others in court.¹⁵² He is the object of others’ gaze, the source of their enjoyment, and identified almost solely in terms of his body. Little wonder, then, that Seneca selects gladiatorial suicide as the epitome of agency and subjectivity, for here is a class of person possessing almost no opportunity for self-determined action beyond the deed of suicide. But suicide is enough to guarantee his ultimate, irreversible self-government. In the deed’s self-reflexive microcosm, the gladiator discovers the freedom and subjectivity denied him by Roman society as a whole.

Much the same set of assumptions applies to Epistle 77’s anecdote about the young Spartan prisoner of war, who dashed his brains out against a wall rather than perform the menial task of fetching a chamber pot (Ep. 77. 14–15). Once again, the autonomy and agency afforded by suicide are set in contrast with the slave’s bodily and social subjugation, and more deeply still, with the inevitable corporeal needs and appetites that ensnare every human being. Suicide represents corporeal domination to the same extent that urinating (signalled by the chamber pot) represents our corporeal reliance on fluids. Self-inflicted death frees the self at every level. The obverse is Epistle 70’s tale of Telesphorus of Rhodes, who clings to life despite having been imprisoned in a cage by the tyrant Lysimachus and treated like a wild animal (Ep. 70.6–7). Telesphorus’ reluctance to end his own life is equated to his bestial existence as an oppressed, tormented object. His unwillingness to assert agency through suicide is not only weakness (Ep. 70.6), but also makes it seem as though he deserves his fate. What he suffers, after all, is merely a magnified, reified version of the curtailment experienced by anyone who places undue value on physical existence and lacks the self-control required for attaining subjectivity in death.

It may be objected, at this point, that Seneca’s view of individual agency is not as clear and straightforward as I have presented it, mainly because he is also inclined to praise those who forego suicide and instead submit themselves to torture or execution.¹⁵³ He often pairs Socrates with Cato, for instance, although the former was, stricto sensu, executed (despite administering the poison himself), while the latter took his own life.¹⁵⁴ Another example, cited for other purposes earlier in this chapter, is Julius Canus, whom Seneca applauds for his calm acceptance of and submission to a death sentence from Caligula (Tranq. 12.4–10). Epistle 70.8 sees Seneca approve of people who refuse to commit suicide when faced with torture, on the grounds that enduring torment likewise demonstrates one’s virtus. While it is true that such forms of death and suffering do not entail the absolute agency epitomised by suicide’s self-reflexivity, Seneca still treats them as instances of autonomy, chiefly by accentuating the element of choice. As James Ker has shown, Epistle 70 unites its scattered examples of libertas achieved through death by emphasising ‘our inalienable freedom to choose (or to resist) death whenever we wish, and to choose between whatever methods are available’.¹⁵⁵ Anyone who remains steadfast under torture evades the oppressor’s grasp just as much as those who take their own lives. The core issue is outmanoeuvring one’s aggressor, typically a tyrant or autocrat, and avoiding victimhood.

Further, Seneca’s focus on contexts of political oppression highlights his interest in death as a form of autonomy and agency superior even to a ruler’s sovereignty. Like the sapiens’ autarky, with which it is often paired, the self-chosen end to life symbolises for Seneca a power equivalent to or surpassing that of any monarchical figure. While Nero, for instance, is the ‘arbiter of life and death’ (vitae necisque … arbiter, Clem. 1.1.2) for his subjects, the suicide exercises his own arbitrium (e.g. Ep. 70.19) in deciding when and how to end his life.¹⁵⁶ The parallel is far from coincidental, and it envisages the suicide’s hegemony as equal, momentarily, to the princeps’. In fact, it lessens the princeps’ authority by usurping his role and asserting the individual’s fundamental immunity to subjugation. Suicide is another version of Seneca’s celebrated ‘empire over the self’.

Self-Harm in the Tragedies

Self-inflicted death and harm also claim prominence as assertions of autonomy in the tragedies. We have seen already how Thyestes responds to his predicament with an (ultimately ineffectual) gesture of suicide; Hercules reacts in a similar manner upon learning the true extent of his misfortune. Initially, before realising that he is himself responsible for his family’s death, he contemplates revenge as the way to resolve his suffering and restore his honour (Her. 1186–91). Once his guilt is made clear, however, he shifts the focus of this act from outside to inside; retaliation becomes self-destruction. Like vengeance, suicide presents Hercules with the opportunity to remedy his losses, both the tangible loss of wife and children, and of abstract qualities such as fama (Her. 1260). His promise to ‘find a way to death’ (mortis inveniam viam, 1245) signals the agency implicit in this deed via evocation of his earlier, heroic activity. For instance, Amphitryon declares at 276–7 that Hercules will not stay trapped in the underworld forever: ‘either he will find a way [out] or he will make one’ (inveniet viam / aut faciet).¹⁵⁷ When the phrase recurs in the context of disaster, it signals how suicide can reprise Hercules’ indomitable faculty for doing; the strength and power that enabled the hero to overcome hell’s boundaries will enable him now to return, permanently.

Hercules’ desire for agency is also apparent in his lurid catalogues of self-harm: he will cremate himself on a huge pyre (1216–17), burn himself along with Thrace’s groves and Cithaeron’s ridges (1285–7); he will drag down onto himself the whole of Thebes (1287–90), and if that is insufficient, the entire firmament (1293–4). The same colossal heroism as made his labours possible will now be employed for the greatest, most self-defining task of all. That Hercules frames prospective suicide in terms of his heroic past (e.g. Her. 1279–82) underscores the agency he perceives it as conferring. If he acted as Juno’s instrument in murdering his wife and children, now he aspires to full subjective control of his circumstances. When Amphitryon tries to absolve Hercules of responsibility by pleading, ‘Juno launched this arrow, with your hands’ (hoc Iuno telum manibus immisit tuis, 1297), Hercules refuses consolation in favour of the self-determination offered by suicide: ‘now I shall use it’ (hoc nunc ego utar, 1298). He will turn his weapon against himself, ensuring that his instrumentality is matched or exceeded by his capacity for independent action.

Hercules’ suicidal inclinations are eventually trumped by an even more solipsistic ambition to preserve his reputation, but other Senecan characters, finding themselves in equally impossible circumstances, actually follow through with the deed.¹⁵⁸ In the Troades, when Astyanax is due to be hurled from Troy’s battlements in sacrifice, he takes the initiative and leaps instead (sponte desiluit sua, Tro. 1101). It is a brief moment, and Seneca does not dwell on matters of agency the way he does with Hercules, but he still manages to suggest it in passing, via a simile comparing the Trojan boy to an immature wild beast, fierce though incapable yet of any real harm (Tro. 1093–6). The image revisits in positive terms a simile uttered by the distraught Andromache at the close of Act 3: Astyanax is a tiny calf torn from its mother by a savage lion (Tro. 794–8), where the young animal’s helplessness and need for parental protection confirm its lack of autonomy. Species is significant, too, for as a iuvencus (795), Astyanax embodies a common sacrificial animal, while his later status as a ferocious wild creature removes him from this category. From domestic to untamed, helpless victim to proud aggressor, Astyanax achieves in suicide an autonomy and agency he could not find in life.

The boy’s independence is likewise stressed by incessant comparison to Hector, a topic I have explored at length in Chapter 2, where I focus mainly on its obscuring Astyanax’s individuality. While this is undeniably the case for most of the tragedy, there are moments when Astyanax’s resemblance to his father signifies not just subordination but also a growing capacity for self-determination, which culminates in the sovereign act of suicide. His jumping from the same tower from which he once watched his father’s feats in war suggests both the pitiful curtailment of his heroism and its drive to achieve something comparable to Hector’s intimidating deeds. If Hector’s fighting symbolises the hero’s raw ability to perform independent actions and effect changes in the world around him,¹⁵⁹ then Astyanax’s self-conferred end grants a similar degree of agency, albeit in terminal form. It is an achievement matching his father’s, and the appropriateness of the setting suggests that Hector is also, in some oblique way, Astyanax’s inspiration; suicide claims the same value as martial valour.

Oedipus, too, discovers autonomy in self-harm, although he stops short of complete obliteration. Throughout the play he has decried fate’s hold over him, his prophesied parricide and incest, the praedicta (Oed. 915) that represent his inescapable domestic imbroglio and its prior tellings (prae/dico) in earlier literature. As Curtis Perry remarks: ‘Seneca’s Oedipus is about what happens to Oedipus’ massively assertive and tyrannical self as he becomes increasingly entangled with various forms of unwelcome contingency or limitation: fate, the family, literary belatedness, the mother as origin and terminus.’¹⁶⁰ If Seneca’s Oedipus seems more boastful and autocratic than Sophocles’, that is largely a function of his thwarted desire for independence and control. Recollection of his encounter with the Sphinx (Oed. 92–102), for instance, is programmatic both in its portrait of Oedipus’ self-assured intelligence and more broadly, as an example of his dominance. Oedipus faced an external threat and triumphed over it, and he is moved to recall this when Jocasta accuses him of cowardice, that is, when he feels his power is under siege (Oed. 81–6). Further, his recollection can be read as an avowal of literary independence as well, by which I mean that he overcomes the Sphinx qua poet,¹⁶¹ and thus assumes, implicitly, a certain freedom from the skeins of carmina that threaten to entrap him. Of course, this freedom is utterly illusory, as is his autocratic sense of control, but his persistent expression of them throws his actual lack of autonomy into sharp relief. I note in Chapter 3 that Seneca’s Oedipus is more a passive object of other people’s knowledge than an active possessor of critical insight; the same dynamic applies to all of his actions in the play, which swing between aggressive claims to sovereign power and wretched acknowledgements of constraint. The heavier the constraint, the fiercer Oedipus’ claims become, until, in the wake of Jocasta’s suicide, he challenges the knowledge of Apollo himself: ‘O lying Phoebus! I have surpassed my wicked fates.’ (O Phoebe mendax, fata superavi impia, Oed. 1046).¹⁶² There is a feeling of exultation in Oedipus’ despair, as though he has overcome his life’s impediments and achieved more than was strictly necessary. This is Oedipus in his ‘domineering and world-dominating’ mode,¹⁶³ a man who wants to smash through limitations and outplay even the gods.

In its reference to Apollo, moreover, Oedipus’ outburst evokes the shadow of poetic composition and literary precedent, as though Oedipus was also claiming to have outplayed his own prior fictional instantiations. Though speculative, the possibility is worth considering, and on this reading, Oedipus 1046 becomes a statement about poetic innovation. Does Oedipus imply that his dual role as matricide and parricide (Oed. 1044–5) push him beyond the typical requirements of his dramatis persona? Or is Seneca indicating, obliquely, his novel scripting of Jocasta’s death, on stage via a sword thrust to her womb?¹⁶⁴ In either case, Oedipus (and Seneca) suggest they have surpassed the decreed (see fata) contours of this particular story.

Such hopes and intimations of autonomy find their fullest expression in Oedipus’ self-blinding. Although his initial rush into the palace sees him considering only external sources of punishment – someone to stab him (927–8); animals to maul him (929–32); Agave to, presumably, behead him (933) – Oedipus quickly turns his attention to self-inflicted forms of atonement. He insists in Stoic vein that ‘death alone frees the innocent from Fortune’ (mors innocentem sola Fortunae eripit, 934), clearly in reference to himself, despite preceding and ensuing admissions of scelus.¹⁶⁵ Having been dogged by fortune throughout his life and over the entire course of his dramatised existence, Seneca’s Oedipus seeks in suicide his final – only – means of attaining sovereign control. That he changes tactics and foregoes suicide in favour of blinding may seem, initially, to diminish that autonomy, but Seneca continues to emphasise throughout the passage the idea that self-harm constitutes a vital form of self-determination. When, for instance, Oedipus characterises his actions as poenae (937; 976) and supplicia (944; 947) he effectively assumes the role of judge in deciding on the nature of his case, its specific features and permutations (936–51), and meting out punishment accordingly. This puts Oedipus in a dominant, authoritative position. Significantly, this is the moment when he asserts control over what happens to his own body, whereas for most of his life that body has been subjected to other people’s whims and plans, especially in the case of his pierced ankles. The ensuing account of his plucking out his eyes reads like autonomy in technicolour, as the messenger’s narrative draws attention not only Oedipus’ agency – he directs his hands, he gouges and rips – but also the agency of his body: the eyes throb (963); they ‘hurry to meet their wound’ (vulneri occurrunt suo, 964); the hands cling (967); the nails tear (968); the torn head vomits blood (979).

Upon completing his self-inflicted brutality, Oedipus is described as victor (974), as though the sovereign power invested in his actions amounted to military conquest. When he returns to stage at the beginning of the subsequent Act, his overtly theatrical reference to his visage – ‘this face befits Oedipus’ (vultus Oedipodam hic decet, 1003) – can also be construed as a celebration of the self-determination facilitated in acts of self-harm: Oedipus has created this face with his own hands; this is an act of self-construction, not formation on somebody else’s terms or through somebody else’s agency.

His authority even acquires a metatheatrical/metapoetic dimension, in the sense that his face doubles as a mask (see discussion in Chapter 3), and that the words eruentis (961), scrutatur (965) and evolvit (967) assimilate his self-blinding to an act of literary interpretation, a search for textual meaning. I propose in Chapter 3 that Oedipus’ self-mutilation likens his body to a text and casts him as an object to be deciphered; there, I stress the generally passive role that this consigns him to. While in no way negating the force of this prior argument, here I draw attention to Oedipus’ active treatment of his own body for the space of the messenger speech. Throughout the rest of the tragedy, other characters claim the authority to interpret Oedipus’ corporeal signa, and Seneca invites the audience to do likewise; here, and only here, that authority falls to Oedipus himself. He is the one in charge of unrolling, analysing, and coming to know his features, a capacity that aligns him – momentarily – with the play’s audience/readers. Even if Oedipus does not attain the status of a poet or dramaturg, still his literary autonomy is conveyed through his ability to interpret and thus assert control over the material placed before him. As much as Oedipus’ self-mutilation configures him as a literary work, it also implies his brief independence from being, merely, the playwright’s instrument or the audience’s object.

There is, then, a strong sense in which the tragedies’ portrayal of suicide and self-harm accentuates characters’ quasi-human aspects by presenting them as autonomous beings capable of deciding their own fates. Like the mechanism of revenge, a character’s choice and pursuit of death is based on the assumption of a contingent future and on one’s individual ability to intervene and alter the course of one’s life. Death implies characters’ escape not only from others’ undue influence, but from the confines of the text itself, a release from authorial control. Yet, as Oedipus’ example demonstrates, it can also function as acknowledgement of the limitations imposed by a purely literary existence, by reminding audiences of a character’s ultimate instrumentality and of the unavoidable fact that deaths, too, are scripted. However much Seneca’s Oedipus interprets his blinding as a declaration of freedom, the event is, nonetheless, an anticipated, much replicated part of his story, so that the autonomy he exercises in this instance, which he seeks so desperately and wins with so much pain, is in the end something subordinate to Seneca and to literary tradition. If Oedipus’ face befits him, it is chiefly because that’s what his story demands.

As a coda to this section, I consider one last Senecan suicide: Jocasta’s, which is noteworthy in both form and presentation. Like Oedipus, Jocasta regards self-inflicted death as punishment for her (unintentional) misdeeds (poenas, Oed. 1035, 1040), and although Seneca makes no specific mention of her death as a form of escape or self-determination, still hints of agency and autonomy are present in her choice of weapon, for using a sword associates Jocasta with masculine activity, both martial and sexual, though the latter connotations prevail here. That the sword belongs to Oedipus only reinforces its sexual symbolism,¹⁶⁶ as does Jocasta’s word choice when she declares rapiatur ensis (‘the sword must be seized’ 1034) because rapio is commonly used to describe acts of sexual violence and domination. Her decision to pierce her womb, moreover, evokes sexual intercourse recast in the self-reflexive form demanded by suicide.¹⁶⁷ Just as suicide splits the human agent into active subject and passive object, so does Jocasta’s death encapsulate both active and passive, masculine and feminine sexual roles.

This attack on her womb also implies agency through its defiance not just of pain and mortality – a standard goal in Senecan suicide – but also of the subjugated female body that must cede independence to children. I remarked earlier in this chapter that Medea’s grim desire to scrape embryos from her belly (Med. 1012–13) amounts to a reassertion of control over her corpus and concomitantly, over the social structures that have trapped her within this maternal function. Similar intimations are present in Jocasta’s act, inasmuch as it frees her from the particularly tangled nexus of her own family and prevents any further possibility of incestuous birth. Its symbolism also reflects, once again, the self-reflexive quality of the suicidal deed, for Jocasta’s womb is a source of both origins and ends, birth and death, imprisonment and freedom.

It is just possible, too, that Jocasta claims poetic/fictional autonomy in the sense that her particular mode of death seems to be the origin of a tradition rather than a reprisal of established and expected motifs. First, the unknown author of the Octavia appropriates it and applies it to Agrippina, who demands that the soldier arriving to kill her strike her womb, for its guilt in bearing Nero (386–72). Tacitus follows suit in attributing to his Agrippina much the same sentiment and actions (Ann. 14.8.4), possibly via the Octavia’s influence.¹⁶⁸ Dio’s account of Agrippina’s murder (61.13.5) likewise echoes this tradition, albeit in more distant and muted form. The only viable parallel to predate Seneca’s Jocasta occurs in the Elder Seneca’s Controversiae 2.5.7, where the tyrant in a fictional legal scenario threatens to beat a woman’s belly, to prevent her from giving birth to a potential tyrannicide.¹⁶⁹ The resemblance is loose, at best, and even if Seneca took some inspiration from it when composing the final Act of his Oedipus, still it is the distinct quality of his image, not the declamatory one, that proceeds to spawn a tradition. Of course, this argument must remain speculative in the absence of further evidence, but it seems feasible to grant some novelty to Seneca’s Jocasta, and in so doing, to liberate her, if only marginally, from the pre-determination that most Senecan dramatis personae take for granted as an essential part of their characters. Though she cannot escape her purely fictional ontology, there is in her death the slimmest suggestion of literary as well as personal agency, of the ability to assert one’s independence from established motifs and to exert one’s power in influencing others.

Conclusion

I have pursued throughout this book the idea that Seneca’s dramatis personae articulate simultaneously the constructed, textual, and implied human facets of their existence, but autonomy is one instance in which these two facets exhibit an occasional dynamic of competition or tension, as characters’ fierce pursuit of independence collides with the inescapable fact of their curtailed fictional being. In revenge and suicide, two of the tragedies’ most prevalent themes, characters’ assertion of limitless agency is brought up short by the very boundaries of the play’s themselves, their enactment, their scripted nature, their engagement with literary traditions, their status as products of Seneca’s authorial, authoritative imagination. Metapoetics aside, moreover, revenge and suicide are two actions that likewise affirm an individual’s autonomy only to question its ultimate fulfilment, since both represent empty victories for self-determination: the latter by cancelling the agent’s existence, the former in its equally self-destructive drive to attain dominion by overriding the social connections via which that dominion is constituted. The same patterns repeat across Seneca’s prose, as the sapiens’ celebrated sovereignty often seems to confer little more than self-satisfied isolation. This is self-determination, certainly, but at the expense of so much that gives the self meaning.

On the other hand, Seneca’s tragedies do furnish instances of characters’ autonomy coinciding with their fictional makeup, as is the case for Medea’s magic, Atreus’ conscious magnification of his deeds, and Oedipus’ active blinding of his passive, textual body. Although typically recognised for their literary connotations, these examples underscore the significance of human autonomy as a theme in Senecan drama, for it is as expressions of independence and control that they acquire much of their power. What makes Medea’s and Atreus’ revenge so formidable, and so memorable, is less their obviously fictional texture than their engagement with notions of sovereignty, both personal and political, that resonate as loudly in the world outside the tragedies as within them. Hence the need to view these motifs in conjunction with Seneca’s prose works, not only for the purpose of better detecting their presence in the dramas, but also, more significantly, for comprehending that the tensions and instabilities exposed by Senecan tragedy are already present, lurking, in Seneca’s philosophical precepts. Like the exchange between fictional and quasi-human elements of character, the dialogue between Senecan philosophy and tragedy is precisely that, a dialogue; it goes both ways.