PART ONE
THE CANON AT RISK

Chapter One
Paradigm, Precept, and Message

In the last lines of the preface to his translation of Thyestes (1560), Jasper Heywood reports his feelings after waking up from a dream meeting with Seneca:

This said, I felt the Fury's force enflame me more and more,
And ten times more now chaf'd I was than ever yet before,
My hair stood up, I waxed wood, my sinews all did shake,
And as the Fury had me vex'd my teeth began to ache.
And thus enflam'd with force of her, I said it should be done,
And down I sat with pen in hand, and thus my verse begun.
(J. Daalder, ed., 1982, p. 21)

The fury is Megaera, who with some justice could be called Seneca's tragic muse.

Some sixty years ago Otto Regenbogen, in a remarkable essay published by the Warburg Institute, declared that Seneca was the first to write what is today understood by the term "tragedy."^[1] The Greeks of the fifth century B.C. wrote tragoidiai that continue to serve as models of significance and power. But their plays do not invariably exhibit the peculiar combination of elements that since the earliest Renaissance, and in the wake of Seneca, has embodied the tragic vision: an unhappy and mournfully moving end supervening upon an abrupt fall; the centrality of the hero and his failure; the prominence of nefas, iniquity;^[2] grandiloquence, ghosts, and magic; an appeal to learning; a measure of didacticism; and all the qualities summed up under the triad atrocitas, maiestas, and gravitas: vehemence, grandeur, and high seriousness.^[3] This is not to say that the Greek repertory does not also, within its varied compass, exhibit these qualities. But Aeschylus's Eumenides, Sophocles' Philoctetes, and a majority of Euripides' extant plays are

[1] Regenbogen 1927–28, repr. 1961. Subsequent references to this seminal essay will he by the 1961 pagination.

[2] Opelt 1972.

[3] Regenbogen 1961, p. 451.

living proof that a Greek tragoidos could perform his task without many of the ingredients that the later European tradition considered essential in a tragedy. A play like Sophocles' Oedipus Rex may at first blush be thought to answer to the Renaissance demands, but it does not. The grandeur is not sufficiently selfconscious or spectacular, the vehemence sufficiently sensational or internalized, to authenticate the play as a paradigm of what the Renaissance critics, from Julius Caesar Scaliger to Johann Christoph Gottsched, required in high tragedy.^[4]

Friedrich Leo thought that by the first century of our era serious drama had shifted from the exploration of ethos, character, to the treatment of pathos, passion.^[5] The formulation is only partly apt. Hellenistic criticism used the two terms to designate a range of emotion from gentleness to fury. But even if we understand the terms in their pre-Hellenistic sense, the precise function of "character" in Greek tragedy has been questioned,^[6] and it is doubtful that any tragedy can do its work without passion playing its part. Yet of the enormous difference between the two kinds of drama, the classical Greek and the Senecan, there can be no doubt. Of all the plays in the Greek corpus, the one that comes closest to the Senecan type in its emphasis on the lability of characters and the frivolity of the gods, and in its admixture of the macabre, is Euripides' Rhesus, hardly a cherished jewel in the classical crown, and a play whose marginal standing has driven scholarship to extreme positions about its date and its authenticity.^[7]

The distance that separates Seneca from his Greek predecessors has been obscured by attempts to discover what unites the Roman versions with the Greek treatments of the myths on which they are based. Apart from Octavia, which is neither mythological nor Senecan, all the extant tragedies of Seneca have their analogues in the Greek repertory (though we have little more than the titles of some of the plays involved). Comparisons, though initially intriguing, tend to create the impression that Seneca meant to emulate or compete with his Greek

[4] For a resumé of Regenbogen's essay, see Coffee 1957. A representative list of the principal general discussions of Senecan drama in the modern period would have to include, at a minimum, the following: Klein 1865; Fr. Leo 1878; Herrmann 1924; Regenbogen 1961; Friedrich 1933; Pratt 1939; Zwierlein 1966; Herington 1968; Seidensticker 1970; Dingel 1974; Heldmann 1974; Pratt 1983; Braden 1985. For a brief discussion, with extensive bibliography, of the past century's work on Senecan drama, see Seidensticker and Armstrong 1985.

[5] Leo 1878, p. 148.

[6] Jones 1962.

[7] Ritchie 1964.

forerunners. That, as an educated Roman with a documented interest in Greek letters, he knew the work of the ancient tragedians is certain. It is less certain whether as a dramatic poet he was more stimulated by the Greeks or by republican dramatists, such as Ennius and Accius, of whose writing we possess only unenlightening portions or fragments. The consensus is that Seneca probably owes most to his immediate predecessors, and especially to the dramatist Varius, who also wrote a Thyestes .^[8] Seneca's debt to the republican dramatists, and also his desire to rival the ancient Greeks, may currently be underestimated. Quotations from and references to tragedy are so rare in his prose work that no comparative inferences can be drawn with assurance.^[9] For our needs it will be best to leave aside questions of literary debt and comparison and to concentrate on what makes Senecan drama the peculiar phenomenon it is.

To see the issue in the proper light, we must devote some attention to Seneca's total literary output. In his prose writings Seneca considers himself a Stoic. He is, to be sure, an eclectic philosopher. Like bees, he says (Ep. 84.5), we must gather our readings from various sources and use our intelligence to make them over into one authentic essence. There are those who are reluctant to regard him as a systematic philosopher at all. In many of the Dialogi and Epistles, he cites Epicurus with great veneration. But he explains (Ep. 8.8) that the opinions cited are in the public domain, and that he could equally well have gone to the poets. Occasionally he questions or even pokes fun at the Stoics and their tenets (Ep. 83; NQ 4.6.1, 7.22.1). His opinion of Chrysippus or Zeno is not always positive, but it is clear from his criticisms (e.g., Ben. 3.8) that he read them in the original. What matters is that the Stoics are nostri ; they are the community of speculative thinkers in which he is confessedly at home (NQ 7.22.1; CS init.; Ot. 3.1). The chief desiderata listed in the preface of book 3 of the Naturales quaestiones —control of vices, disregard of fortune, cheerful endurance of pain, authority over one's own life, purity, concentration on essentials—have an unmistakably Stoic look about them. Seneca's pronounced admiration for Cato, the saint of the imperial Stoicizers, confirms the impression.

[8] Tarrant 1978; cf. also the summary in Tarrant 1985, pp. 16–19. But cf. Dingel 1985, pp. 1053–54. For the prehistory of Thyestes in particular, see Tarrant 1985, pp. 40–43.

[9] Dingel 1974, pp. 48ff.

Both in his letters and in his treatises, Seneca demonstrates again and again that his knowledge of the history of thought is extensive, that he is familiar with the rules of logic and argumentation, and that he has been relatively successful at carving out his own decently consistent position from among the conflicting views available to him. He is not an original thinker in the strictest sense of that idea. But like Lucretius and Cicero, the only Roman writers with whom he deserves to be compared, he has made the thinking of his sources and his teachers his own. He convinces us with the earnestness of his inquiry and the firmness of his choices, and he advances his arguments with an engagement of self, and with a sense of drama, that are often exhilarating.

Between Chrysippus, the third head and dominant intellect of the Stoa in the third century B.C.E. , and Seneca, in the first century of our era, a number of forceful and relatively independent Stoic thinkers enriched the ideological storehouse of the school. "Among the Stoics, from the founders till today, opinions differ," was the view of Numenius, a scholar who lived a century after Seneca (Hans von Arnim, Stoicorum veterum fragmenta [henceforth SVF ] 2.20). Stoicism was the only branch of Hellenistic philosophy that did not sanctify or try to congeal the founder's positions. Still, by comparison with the achievement of Chrysippus, in the area of thought with which we are going to be concerned, the innovations and adjustments introduced by his successors relate to details rather than substance. On points of ethics and of psychology, Seneca can be shown to have adopted or modified the teachings of Posidonius and of Panaetius where those differ from the pronouncements of the Old Stoa. But Chrysippus remains the principal inspirer of the aspects of Stoicism crucial to our understanding of Senecan drama. Seneca introduces the name Chrysippus again and again. And though his own flexible and humane genius does not, at first, appear to have much in common with the tough intellectualism ascribed to Chrysippus, I shall have occasion to argue that the difference between the two, and especially the so-called intellectualism of Chrysippus, must not be magnified.^[10]

As with Marlowe, or Chapman, critics have responded to the power of Senecan drama with a sense that it harbors a striking measure of

[10] Hall (1977) demonstrates that Seneca is not given to exactitude in quoting from his sources. This is an important caveat in any attempt to recover the precise wording of the writing upon which Seneca relies. But we are more interested in the broader affinities between Stoic science and Senecan thought.

authorial self-revelation. If we had the theoretical writings on the stage with which the tradition credits Sophocles, or if Aeschylus and Euripides had left us such inquiries, it is unlikely that those documents would make it easy for us to connect the preoccupations, much less the speech, of their plays with the private passions of the writers. In Seneca's plays, on the other hand, every character is felt to offer a bit of the author himself as he stands revealed, or wishes to have himself thought of, in his prose writings.^[11] And that exposure of personality stands in the closest relation to the characteristic ways of analyzing conduct that Zeno and Chrysippus had made available.

At one level, the Stoic nature of the concerns is obvious. Amphitryon in Hercules Furens tries to console his son by calling his crime a mistake:

Quis nomen usquam sceleris errori addidit?
Who is so bold to call an error "crime"?
(HF 1237)

Hercules replies that, generally speaking, there is no difference:

Saepe error ingens sceleris obtinuit locum.
A major error usually ranks as crime.
(HF 1238)

This is orthodox, mainstream Stoicism, though qualified by temporal modifiers—usquam, saepe —that might not have satisfied the more doctrinaire members of the school. With one stroke the old Aristotelian conundrum of whether hamartia is an error or a vice becomes irrelevant: under the exacting rules of Stoic ethics, the two are indistinguishable. Another line in the same play, the last line of the first chorus (201), provides a near-emblematic instance of the Stoic reinterpretation of an old Greek moral: alte virtus animosa cadit . Various translations are possible. We should probably put considerable emphasis on the force of the adjective: "Heroic manhood plummets deep." It is difficult to imagine a more pregnant formulation of the need for the traditional hero, with his great soul and his exalted virtue, but also with his aggressiveness and his passion to lord it over others, to fall. These chiselled phrases are easy to fit into the Stoic canons of moral and social conduct. That they bespeak a larger and more intimate engagement will, I hope, become clear in the course of this essay.

[11] Shelton 1979.

Nonetheless, critics have always been readier to acknowledge the Stoicism of Seneca's essays and letters than to recognize the same principles at work in the dramatic corpus. Superficially, to compare Seneca's prose with his drama might seem like comparing the writings of the Reverend Norman Vincent Peale with the Grand Guignol of Titus Andronicus and similar dramatic documents of terror. Indeed, soon after Seneca's lifetime some readers decided that the two bodies of work were so different in their ideological orientation that they could not be the work of the same man.^[12] One scholar who argued this position was the fifth-century churchman and public verse orator Sidonius Apollinaris, who wrote (9.229), in one of his many praises of unworthy emperors, that there were two or even three Senecas, one of whom was the philosopher, and another the dramatist. Similar statements are found among the Renaissance humanists, including Erasmus. Justus Lipsius, one of the key figures in late Renaissance neo-Stoicism, thinks that only one or two of the plays are by the philosopher, and that the rest are by various other Senecas. Lessing, also, in the youthful work entitled Von den lateinischen Trauerspielen welche unter dem Namen des Seneca bekannt sind, engages in similar speculations, as did Diderot before him.^[13]

The cutting in two of Lucius Annaeus Seneca is the most radical step taken by those who are embarrassed and irritated by their own failure to find anything essentially Stoic or philosophical in Senecan drama. One of the latest critics to come to this conclusion—though he does not go all the way but keeps the man Seneca unsplit—is F. H. Sandbach, the author of an authoritative handbook on Stoicism. Sandbach has a brief section on Senecan drama, in which he refuses to see any appreciable trace of Stoicism, except for Hercules Oetaeus, which, he thinks, exhibits a Stoic saint in action. The antinomies Sandbach establishes will continue to occupy our attention. The plays, he says,

are concerned with the effects of the passions and the blows of Fortune. For the Stoic, Fortune was to be identified with Fate and Providence, for the dramatist it is a blind and hostile power; for the Stoic the passions are sequels of faulty judgment, for the dramatist they are independent

[12] For the ancient and medieval periods, the separatist hypothesis is best traced in Trillitzsch 1971 and 1978. With variations, it continued to be widely argued well into the ninteenth century. See also Häuptli 1983, pp. 77–105, for testimonia, from Suetonius to Stravinsky.

[13] Erasmus 1934, pp. 37–38. Lipsius 1611, intr.: only Medea is by the philosopher; the other plays are by various other Senecas. Lessing 1890, pp. 167–242. Diderot 1821, bk. 1, chaps. 125ff.

forces that fight with reason and pervert it for their own ends. Seneca's characters are not so much human beings as simplified exponents of anger, jealousy, cruelty . . . and the no less dangerous love, passions which brush aside the arguments of those who speak for reason and morality.^[14]

The burden of this contrastive analysis is, it seems, the following: if the drama were truly Stoic in complexion and intent, it would feature believable human beings in action, and reason would win out; that is to say, it could not be tragic, and would have considerable difficulty being drama. Similar objections have been raised by others. Joachim Dingel, the author of a searching analysis of this question, finds that the plays do not conform to Seneca's thinking as it emerges from the philosopher's remarks about aesthetics and education: "Der Dichter Seneca ignoriert die Kritik des Philosophen Seneca."^[15] Senecan drama realizes all the horrors the philosophical writings repudiate: scenes of hell, the topos of the double night, Jupiter's adulteries, and much more. His conclusion is that the plays show us the real Seneca, while the philosophical writings give us a mask, an official pose, feeding copiously and facilely upon the ready-made materials of the philosophers.^[16] His hesitant suggestion that the dramas document a "negative Stoicism," dwelling on what we do not understand about the gods and Fate, seems to leave the door open for the conclusion that they involve some kind of Stoicism after all. Yet that insight is not developed.^[17]

Today, Sandbach and Dingel represent, each in his own way, a minority position. The majority of readers and interpreters (not to mention producers), though equally unwilling to look at the plays as Stoic documents, are not troubled by the scruples that worry them.^[18] But the difficulties they have spelled out remain, and one is bound to admire the courage with which the modern questioners have turned against the received opinion, particularly where the received opinion is based on a sense, found even among critics who ought to know better, that Stoicism cannot really be taken seriously as philosophy (for example, see R. A. Brower 1971, p. 143). As I have indicated, I have consider-

[14] Sandbach 1975, pp. 160–61. In this judgment Sandbach agrees with a number of scholars, including students of later tragedy who look upon Hercules as a prime model for Corneille and others: see Wanke 1964, pp. 176ff., and Sokel 1964, p. 23f.

[15] Dingel 1974, p. 65.

[16] For a similar position, see also Henry and Walker 1963.

[17] For a brilliant review of Dingel's position, see Timpanaro 1981, pp. 117–18 n. 6.

[18] Among the critics who maintain and emphasize the Stoic identity of Senecan drama, see especially Regenbogen 1961; Egermann 1940; Knoche 1941; Marti 1945; Lefèvre 1969; Cacciaglia 1974.

able sympathy with the view that we must be on our guard against a criticism that insists that a literary text conform to a specific philosophical thesis. But the case of Seneca is a special one. The writer whose name is attached to the dramas in the manuscript tradition is a self-confessed Stoic. What is more, the point is to show, not that Seneca was trying to be a Stoic in his dramas, but that the dramas make better sense, or come across more powerfully, if understood as emanating from a Stoically trained perception.

But again, among those who are willing to accept the philosopher as the author of the majority of the plays, and who recognize a moral function in the drama, there is little agreement about the nature of that function. Some believe that Seneca tried to fashion the myths into moral fables—and stumbled. Kurt von Fritz, a subtle critic of the history of dramatic theory and a leading connoisseur of the fragmentary tradition of Stoicism, finds that Seneca's Oedipus represents an attempt to construct a moral-exemplary tale, and that it fails.^[19] According to another view, everything in the action of a Senecan play tends to confute the claims of Stoics, but the choral sequences show a way out of the impasse.^[20] They are the lessons that bring home to us the way of redemption; they may be irrelevant to the immediate aims of the particular plot they embellish, but they furnish the therapy whose need the drama demonstrates. On the face of it, as dramatic criticism, this is a desperate proposal. But it is one that would have been understood by many of Seneca's later imitators, who cared less about drama than about salvation, and it is not so very different from an interpretation of Senecan drama in terms of crime and punishment.

Those who wish to dissociate the philosopher from the dramatist cite the denunciations of poetry in the prose writings.^[21] There at one

[19] Von Fritz 1962, pp. 21–26. This curiously circumscribed judgment of von Fritz, which in effect says that Senecan drama is melodrama, is surprising in the light of his mature understanding elsewhere of what Stoicism, in all its varied aspects, may contribute to the experience of serious drama. But the notion of crime and punishment is at the root of most analyses of Senecan drama as Stoic manifesto.

[20] The choruses most often put under this obligation are Aga 589ff., Oed 882ff., 950ff., Phae 959ff., 1123ff., Thy 336ff., Tro 371ff. Cf. Marx 1932.

[21] For Seneca's discussion of poetry and of the liberal arts in general, see Mazzoli 1970; Dingel 1974; Stückelberger 1965. Mazzoli provides an intelligent and accurate introduction, but goes too far in constructing a Senecan poetics out of the disjecta membra of Seneca's remarks. Dingel recognizes that Seneca has no consistent philosophy of poetry. In his view, the tragedies give us the authentic voice of Seneca, while the discussions of poetry in the prose writings are subject to the limitation of those prose writings, which constitute a mascheratura, an attempt to rationalize and render innocuous the evil in the world. Cf. also the pertinent discussions in Berger 1980, and Hadot 1984.

point or another Seneca berates scenes in hell, legends of the gods misbehaving themselves, the concept of Chance working at random, and other topoi favored by the dramatists and prominent in Senecan tragedy. One might object that Zeus's rape of Io hardly exhausts the full range of significance of Aeschylus's Suppliants, or Apollo's constraint of Cassandra that of his Agamemnon, or even Hercules' raking of Hell that of Seneca's Hercules Furens . But that is too easy an answer. Seneca himself says in one of his Consolations (Pol. 11.5): "There is no book among the writings of the poets that does not furnish you with a great many paradigms of the varied nature of men and of the uncertainty of events and of the many causes that make tears flow." The causes are crucial (as we shall see again later in this essay); the philosopher can extract his evidence from literature, and will find ready-made grounds to fit into his argument. The inherence of causes in the literary complex gives to that body of writing the potential for philosophical development and a philosophical base of its own. In fact, a thorough study of Seneca's pronouncements on the uses and risks of reading shows a variety of often conflicting responses, some of which will be taken up directly. The philosopher cannot think about poetry without a passionate interest in its effects.

It is true that in his treatises and letters, Seneca's thinking about poetry and the liberal arts mostly cleaves to ethical concerns. Seneca rejects disinterested philological or historical inquiry (where did Ulysses go in his travels?) in favor of the sort of questions moral philosophy asks, so we will not stray ourselves (Ep. 88.7). We are all Ulysses, tempted by vicious stimuli to roam and blunder. The writer must teach us how to navigate, especially when we are temporarily shipwrecked, and to love everything that is honestum, respectable and civilized. Reading and listening are legitimate and profitable only if out of the words there come to be works: ut quae fuerint verba sint opera (Ep. 108.35). The paradigms must come to rest in the blood of the listeners. Philosophy is to be tested against the substance of life as recorded in the poems, a life that demands action. The Stoic debate concerning the virtues of quiet wisdom and its realization in the active life can still be heard in Antonio's rejection of Alberto's counsel of patience:

Patience is slave to fools, a chain that's fixed
Only to posts, and senseless log-like dolts.
(John Marston,
Antonio's Revenge 1.5.36–37)

In all endeavors to assess the nature of the Stoic share in the plays, the sights are trained on moral doctrines, the values that are most effectively communicated via exempla, the cautionary figures and actions of prominent men and women, and via precepts and sententiae, the often moving, but always safe, generalities offered by the characters and particularly by the chorus.^[22] Is this what Seneca has in mind when, in one of his Epistles (20.1), he asks his correspondent to "plunge the philosophy into the bedrock of his heart"—ut philosophiam in praecordia ima demittas —and to test his progress, not through speech or writing, but by the experience of his soul? Stoic choruses, a modern critic says, are "pitched to inflame rather than exorcise,"^[23] a remark that mirrors Scaliger's (and, before him, Quintilian's) insistence on the emotions, the affectus, of the audience as the target of the dramatist's persuasive power.

Let me anticipate and say that if precepts and paradigms were all that Stoicism had to contribute to the dramatic construct, the conclusion would have to be, not only that Senecan drama is only intermittently Stoic, but that the whole question is skewed, and that the drama as drama, particularly in its tragic aspect, is incompatible with genuine Stoic belief.^[24] Diogenes the Cynic, a spiritual ancestor of some of the more radical Stoic moralists, wrote an Oedipus in which he sought to show that it was silly of Oedipus to be exercised over the marriage with his mother, on the grounds, presumably, that events beyond our control should not be permitted to disturb us. Diogenes' genial (or brutal?) confection, whether it was a drama or not, is unfortunately lost. But the Stoic potential for serious drama cannot be shrugged off quite so easily. If the Stoic moral is merely a conclusion e contrario, an ethical or religious message suspended, like a Brechtian placard, in the interstices of demonstrations of human misery, we might as well admit that Senecan drama is best defined in the narrow terms heralded by Nietzsche's slogan, at the head of his Streifzüge eines Unzeitgemässen: "Seneca or the Toreador of Virtue."^[25]

There is some irony in the spectacle of Nietzsche falling a victim to what must be regarded as a Christian narrowing of the Stoic legacy.

[22] For Aristotle on exempla and sententiae, see Rhet. 2.20.1393a28ff. and 2.21.1394a19ff.

[23] Braden 1970, pp. 38–41.

[24] Hence the speculation of Marti (1947) that Senecan tragedy may be derived from earlier tragedies composed by Stoic philosophers has not found acceptance.

[25] Friedrich Nietzsche 1966, p. 991.

The neo-Stoicism of the Renaissance and of the Baroque, of Erasmus and Justus Lipsius, of Monchrestien and J. du Vair and Andreas Gryphius, has usually been studied as a momentous exercise in merging the postulates of Christian morality with those of Stoic ethics, under the guidance of a post-Reformation consciousness of sin.^[26] Christianity and Stoicism meet on the plane of stubborn resignation. Papinianus, the hero of Gryphius's play of the same name, is an avowed Stoic as well as a messenger of unmistakably Christian obligations. Justus Lipsius's De constantia, arguing that all men are guilty in the sight of God (2.16), reinforces the point that the Christianization of Stoicism, or the Stoicizing of the patristic tradition, is most keenly observed where the subject is sin and sinners. The easy acceptance, through the centuries, of a spurious exchange of letters between Seneca and Saint Paul endorses the tradition.^[27]

But Lipsius himself expounds a body of thought that goes far beyond the tight strictures of a moral code. Wilhelm Dilthey^[28] and others have demonstrated that much of the European philosophical and literary tradition, from the late Middle Ages to the nineteenth century and beyond, may be interpreted as the working out of Stoic impulses, and that the Stoic seeds survive in the most varied sectors of human thought. What is more, even if we restricted ourselves to tracing the echoes of Stoic ethics and psychology, it would be a mistake to look only for direct and positive signals. The success of a play, and the identification of the elements that have gone into its creation, are gauged by more encompassing questions than the plotting of an intermediate maxim or a terminal moral openly carried into the field. Here is an excerpt from Lipsius's De constantia (2.13), in the translation of Sir John Stradling. The speaker is Langius:

Tell me, in beholding a tragedy, will it stomacke thee to see Atreus or Thiestes in the firste or second acte walking in state and maiestye uppon the scene? To see them raigne, threate and commaund? I thinke not, knowing their prosperitie to be of small continuance; and when thou shalte see them shamefullie come to confusion in the last Acte. Nowe then in this Tragedy of the World, why art thou not so favourable towards God, as to a poore Poet? This wicked man prospereth. That Tyrant liveth. Let be awhiles. Remember it is but the first Acte, and consider aforehande in thy mind, that sobs and sorrowes will ensue uppon their

[26] Zanta 1914; Simon 1955; Dilthey 1960; Abel 1978.

[27] Sevenster 1961.

[28] Dilthey 1960.

sollace. This Scene will anon swimme in bloud, then these purple and golden garments shal be rowled therein. For that Poet of ours is singular cunning in his art, and will not lightly transgress the lawes of this Tragedie. In musicke, doo we not allowe sometimes disagreeing soundes, knowing that they will all close in consent? But the parties iniured doo not alwaies see the punishment. What marvels is that? The tragedy commonly is tedious, and they are not able to sit so long in the theater: yet others do see it, and are worthily stricken with feare when they perceive that some are reprived before the severe throne of justice, but not pardoned: and that the day of execution is prolonged, not wholly taken away.

It is difficult to know what to make of this notion of God as a tragic poet, though there are some Plotinian texts canvassing its implications.^[29] It is the converse of the contemporary Platonist elevation of the poet to the rank of divine maker. But the idea that tragedy can have a moral effect only if the viewer can supplement its meaning with his own historical and eschatological imagination is one that runs counter to the demand that Stoic tragedy carry its message upon the sleeve. Marcus Aurelius said (11.6) that tragedy was invented to teach you to regard the untoward events in life with the same unruffled feeling or even pleasure as their imitations on the stage. But Marcus comes to this insight from a direction opposite to that of Lipsius: dramatic performance has its place in a wonderfully contemptuous catalogue of meaningless and frivolous activities (7.3), in the midst of which we are asked to maintain our good humor; a philosopher stands at an enormous distance from those who strut tragically across the stage of life (9.29). Lipsius, on the other hand, is fascinated by tragedy. But the reader senses that Lipsius is impressed with tragedy for reasons that have little to do with an overt ethical didacticism. Whether in his graphic description of what happens on the stage, as on the larger boards of the world, he has in mind Seneca or one of his sixteenth-century imitators is uncertain. What he finds impressive about this kind of drama is difficult to accommodate to a simple calculation of sin and punishment duly apportioned.

The time has come to look more closely at the role of paradigm and precept in Senecan drama, and to show why they are, by themselves,

[29] Plotinus 3.2.16–17 discourses upon the world as a space created for dancers and actors.

incapable of explaining the power of the plays, or of vindicating their standing as Stoic documents. Seneca says (Ep. 95.65) that according to Posidonius, paradigms, or exempla, are part of a larger machinery of moving the listener, called paraenetic in Greek, which also includes precepts, consolation, persuasion, and exhortation. Zeno stated (SVF 1.84) that a paradigm is "the recollection of a past action by way of approximation to what is now being sought," a complicated formula designed to cover a wide usage of the term, particularly within the medium of ethics. In the psychology of drama, the effect intended is not so much recollection but the vivid presentation of an act or an agent on the stage. The paradigmatic figure or action may itself be linked with anterior (mythical or philosophical) paradigms. Terpnos, the celebrated actor of Nero's time, "sang" lamentations of Priam, imprecations of Oedipus, the madness of Orestes, the death of Icarus, and much else,^[30] and Nero himself delighted audiences with similar evocations of past figures. But the great tragic character does not follow an exemplum so much as create one.^[31] If the dramas are, as the moralists believe, suasoriae, moral briefs of a kind, what is the function of the paradigms in them? In what follows, I shall at first present my argument as if it were a foregone conclusion that Senecan drama means to be educational, and that the focus of this educational mission is in the area of moral and political behavior.

To begin with, a paradigm can work either positively or negatively.^[32] In the former case, it is intended to produce a fruitful message, a hortatory or suasive signal. The noble hero or the wise king relying upon his healthy understanding and acting sensibly, or suffering without breaking, contributes a model calculated to improve the audience, or at least to sway them from a course of folly. Jupiter as a governor is a paradigm for rulers (NQ 2.43): let them follow his reasonableness and his clemency.

To turne to our domesticke hystories, what English blood seeing the person of any bold English man presented and doth not hugge his fame, and hunnye at his valor. . . . so bewitching a thing is lively and well spirited action, that it hath power to new mold the harts of the spectators and fashion them to the shape of any noble and notable attempt.

[30] Note the sources given by Cattin 1963a, pp. 101ff.

[31] For the whole subject, see Egermann 1940, and Hadot 1969.

[32] On the importance of exempla of vice as well as virtue for acquiring moral awareness, see Ep. 120.3–11.

Thomas Heywood's words (in An Apology for Actors )^[33] reflect a long tradition of the dramatic use of paradigms, a tradition of which Seneca, with such figures as Polyxena and Astyanax in Troades, was one of the principal architects. When Oedipus, in his bitterness and despair, marvels at the miracle of a woman as great-hearted as his daughter Antigone

Unde in nefanda specimen egregium domo?
How does a vicious house come by this singular model?
(Phoe 80)

his speech records the impact a public figure, from history or legend, should have upon a miscreant society. This is what Albertino Mussato, the re-founder of Senecanism in the early Renaissance, in his first Epistle hoped for among the projected effects of the tragic models: to produce constancy and assurance in the face of flux and adversity. And this is a tradition that remained powerful into the eighteenth century, at which point the "ironic" cast of enlightenment historiography killed the easy reliance upon the authority of positive paradigms.^[34]

With Mussato, we have already lighted upon the other variety of exemplum, the cautionary paradigm. At the conclusion of Seneca's De constantia sapientis the emperor Caligula is brought in, by way of a supplementary playlet, to cement the positive lesson with a demonstration of the inadmissibility of its converse. Many of the old Greek heroes and heroines, with their lives of passion and torment and the disastrous consequences of their choices, were eminently suitable as warning examples. Those who feel that it is the business of the theater to teach, and that includes the vast majority of critics over the past two millennia, advertise a policy of keeping the categories, the suasive and the deterrent, clearly distinct.^[35]

In Senecan drama, the cautionary prevails. Atreus in Thyestes and Clytaemestra in Agamemnon teach us, if teaching is the word, how not to live. The tragedy appeals, not to our crudely imitative instincts—

[33] Heywood 1941, p.7.

[34] Koselleck 1967.

[35] The distinction may be effaced by the critical compendium "fear." Some Stoic sources cited by DeLacy (1948, pp. 249, 270) suggest that the poet uses fear to influence the listener. In the case of cautionary paradigms, the fear would be a negative force; in the case of hortatory exempla, fear equals awe and admiration. This is the background against which Corneille developed his concept of admiration as the principal intended effect of high tragedy.

what Freudians call the repetition compulsion—but to a discriminating moral sense that shrinks from the severity of the vices acted out and belabored. Antigone and Polyxena move us less through the manner of their action than through the admiring reports about them. Some profess to see Hercules in Hercules Oetaeus in the same light, as an elaborate instance of the imitabile exemplum, but the difficulties with that proposition are insurmountable. He, like all the central figures of Senecan drama who interest us, illustrates the cautionary, if not the repellent, mode. Generally speaking, positive paradigms are, if cast in major roles, dramatically unpersuasive. In George Chapman's The Revenge of Bussy D'Ambois, Bussy's brother Clermont is conceived as a Stoic sage, as is Cato in the same author's Caesar and Pompey . They are embroiled in the Machiavellian troubles of their times, and go under while maintaining a principled stance. Clermont takes his own life rather than yield to the impulse for revenge. The result is a dramatic letdown and embarrassment.

Most critics would agree that where the suasive paradigm carries the weight of the action, the play's intensity suffers. In several chapters of the third book of his Manductio ad philosophiam Stoicam (1604), Justus Lipsius demonstrates, by implication, the uselessness of a Stoic saint on stage.^[36] And this is often interpreted to mean that such plays go wrong because they preach a Stoic message—that is, because their hortatory mission is the chief office their commitment to Stoicism is designed to convey. "The failure of Chapman's tragical drama ever to achieve fully the stature of great tragedy is in large measure due to the fact that Stoicism negates the premises from which such tragedy develops. Mere chance did not determine that Bussy D'Ambois, Chapman's outstanding success in tragedy both on the stage and in the acclaim of literary history, proves to be the only play on which the ethics of Stoicism exerted so slight an influence," observes J. W. Wieler.^[37]

With his "the ethics of Stoicism," the critic is half right. But it is a reasonable assumption that what he means by this phrase is the whole

[36] Chap. 5, 3rd paradox; chap. 6, 4th paradox; chap. 7, 5th paradox; chap. 10, 8th paradox; etc. For a useful survey of Lipsius's Stoicism, see Saunders 1955. In spite of Lipsius's earnest attempt to reconcile Stoicism and Christian faith—in Physiologia 1.18ff. he argues that the many gods permitted in Stoicism are really no different from Christian angels, and that even their corporeality ought not to be pressed—his understanding of the major tenets of Stoic science was extensive.

[37] Wicler 1949, p. 163. Most modern studies of Chapman refer to Stoicism or Seneca only in discussing style.

range of what Stoicism could offer to a dramatist; and here I hope to suggest that Bussy D'Ambois is a gloriously Stoic drama through and through.^[38] Leaving aside this particular play, it is wrong to jump to the conclusion that the cautionary paradigm, with all its packed freight of murder and lust, is less indebted to Stoic impulses than the admired image of nobility. Even a casual reading of the canonic Stoic texts reveals a constant preoccupation with the seamy and the sinister, an almost luxuriant dwelling on the vices the Stoic hopes to avoid. Quintilian (10.1.129) calls Seneca an egregius insectator vitiorum, a rare compliment that, though meant to characterize his philosophical prose, is equally applicable to the drama. Evil, Chrysippus said (SVF 2.1175), is an instrument whereby God educates men.

For obvious psychological reasons, the dividing line between the suasive and the cautionary is, in practice, not as neat as the classification tends to promise. From the point of view of his father, Amphitryon, the infant Hercules, remembered in a wistful speech, is an accomplished Stoic, facing the fiery snakes with a calm countenance (HF 215ff.; cf. also 1200–1201). Ideally, as a vanquisher of monsters, the Hercules of myth should have an easy enough time making himself over into a Stoic saint, and indeed, many Stoics, like the Cynics before them, adopted him as their patron hero. But in the dramas—both Hercules Furens and Hercules Oetaeus —the auspicious features are blotted out by the recognition that the monsters the hero is fated to overcome are largely in his own breast, and that they vanquish him more devastatingly than his legendary self ever defeated them. The dusdaimon of Euripides, the hero brought low by divine forces beyond his control, turns into a nocens, an authoritative destroyer of what is good. Lower down on the scale of significance, Agamemnon, in the second act of Troades (250ff.), counsels Stoic moderation. But as his opponent, Pyrrhus, gathers the character evidence against him, and casts in his teeth his fickleness, his fear, his weakness for women, and his greed, Agamemnon's credibility as a Stoic sage, never very compelling, is demolished, and the suasive and the cautionary are once again brought into collision, or better, into fateful convergence.

Taking another look at Chapman, even the noble Clermont is an equivocal symbol of Stoic virtue. The challenges of blood, of personal loyalty, and of the cosmic catastrophe invoked by the ghost of his

[38] Haydn (1950) has some discriminating comments on neo-Stoic elements in Renaissance drama. But his distinction between doctrinal and bastard Stoicism seems to me unpersuasive.

brother Bussy speak to him so urgently that his Stoic orations, largely Epictetan rather than Senecan, appear destined not so much to voice his nobility as to help him find his balance and hold his crumbling world together. It must be said that without a large amount of adulteration, so that the saintly nobility comes to be marbled with streaks of fretfulness or distemper, if not downright madness, the virtuous hero leaves us cold. The Renaissance martyr play, such as Gryphius's Katharina von Georgien (1647), in which the heroine is put through a series of murderous and revolting tests, demonstrates that our interest in the saint is pleasingly fanned by his palpable torments.^[39] But pure suffering linked with shining virtue is not a promising foundation for tragic empathy. The conflation of the saintly with the questionable is imperative, far beyond Zeno's delicate limits quoted in Seneca's Ira: "In the mind of the sage also, even after the wound is healed, a scar remains, and so he will experience certain suspicions and dark hints of passions, though of the passions themselves he will be free."^[40] The hortatory and the deterrent must enter into a potent fusion. As it has been put recently with regard to the titular hero of Seneca's Oedipus, hardly a paragon of Stoic virtue, but not for that another Atreus: if Seneca was interested in providing a lesson, he did so by hinting at the possibility of salvation even for Oedipus, and not by insisting that Oedipus, in falling short of that salvation, automatically assumed the character of a villain.^[41]

But does this glancing away from the mutual exclusiveness of the negative and the positive not put the whole idea of an educative Stoicism in doubt? Doesn't Seneca's dramaturgy prove that the cautionary paradigm reigns supreme, and that the lesson conveyed by it is easily discernible? As Atreus, near the end of Thyestes, rings down the curtain on his pleasure in the crime, only to turn around and prick the bubble of that very pleasure:

bene est, abunde est, iam sat est etiam mihi.
sed cur satis sit? pergam . . .

It is enough, more than enough, and I
Am satisfied. Or am I? No, I will go on . . .
(Thy 889–90)

[39] For an excellent appreciation of Gryphius and the Dutch and German Senecans who preceded him, see Regenbogen 1961, pp. 426ff.

[40] 1.16.7: In sapientis quoque animo, etiam cum vulnus sanatum est, cicatrix manet. Sentiet itaque suspicion quasdam et umbras adfectuum, ipsis quidem carebit.

[41] Heldmann 1974, p. 194.

the warning about ravenousness, the major theme of the play, is obvious enough. Atreus resembles Hannibal, who, when he saw a ditch full of human blood, exclaimed over the beauty of the sight; or Volesus, who went into similar raptures as he walked among the corpses of three hundred victims of execution (Ira 2.5.4–5). A staging of such ferocity, such amoral indulgence, portending the loss of all civilized control, cannot but turn the audience in a salutory direction. The severity of the vices makes us shrink and think again. The later branches of the Senecan tradition, both the gory Italian and the slightly less sanguinary French, draw their satisfactions and their social legitimacy at least in part from their adherence to the Stoic view transmitted by Horace: tragedy, in featuring an exemplary deterrence, educates. If in Marlowe the moral lesson is undermined by the hero not coming a cropper as the moral would demand it, but crashing in unexpected circumstances, the reason lies with the greater complexity of the dramatic issues. But the audience is likely to disregard the aesthetic and psychological dislocations and take the warning straight.

Yet the example of Marlowe raises a principal question. If the suasive or the subtly mimetic is dovetailed with the cautionary, can we be certain of the appropriate response? Compare Corneille's Auguste:

Ces exemples récents suffiroient pour m'instruire,
Si par l'exemple seul on se devoit conduire:
L'un m'invite à le suivre, et l'autre me fait peur;
Mais l'exemple souvent n'est qu'un miroir trompeur;
Et l'ordre du destin qui gêne nos pensées
N'est pas toujours écrit dans les choses passées.
(Cinna 2.1.31–36)

Seneca tells a fascinating anecdote about the performance of a play by Euripides, Bellerophon, now lost, in the course of which much was said on behalf of the possession of wealth, to the apparent detriment of the standing of innocence, fair health, and good reputation (Ep. 115.14ff.). To what extent the arguments pro and con were conveyed by the lives of exemplary characters or were developed through more or less detached maxims (for this topic, see below), is now impossible to say.^[42] At a certain point in the performance, Seneca tells us, the spectators rose in disgust and chided Euripides for what they conceived to

[42] For an attempt to reconstruct the contents of the play, see Webster 1967, pp. 109–111.

be his campaign on behalf of filthy lucre. It took all the playwright's authority to get the audience to resume their seats and to look at the remainder of the play, at the end of which they would realize, because of the fall of the hero, that a delight in wealth for wealth's sake has dire consequences. It is conceivable that Seneca did not know the play, and had the anecdote from somebody else; the fragments that have come down to us do not make it likely that it was Bellerophon himself who was the devotee of affluence. But the tale, like Lipsius's remarks about the divine playwright, scores a significant point. A complex dramatic structure, featuring a seemingly positive or, worse, innocuous paradigm subsequently shorted into a negative moral, can be counted on to spark the right reaction only if allowed to do its work slowly, without interruption, and perhaps with a surprise twist toward the end. There is a lesson here about timing, a lesson Seneca's plays can hardly be said to certify. If, in a play that offers models both for imitation and for eschewal, the relation and the timing of the elements are not expertly handled, the consequences for the drama can be disastrous. What if Euripides had not risen to reassure the fans, and the misleading inchoate signals of the exempla had so taken root in their tempers that the final reversal had no chance to rally them?

Though Seneca says that "nothing is as useful as first to look upon deformity, and then upon the danger" (Ira 2.35.3), he is himself alive to the risk inherent in conflicting signals and their combination, whether arranged in sequence or not. Stoicism is keenly aware of the precariousness of a morality that is preached or implied e contrario . "Evils sometimes offer the aspect of the good. . . . vices border upon virtues; the incorrigible and the sinful have the likeness of the righteous" (Ep. 120.8).^[43] Seneca warns that we must watch closely to make the necessary distinctions, not only between virtue and vice, but also between the agent and his quality or, to put it more pertinently, between the choices made and the issues made to triumph. Two similar lines of conduct can be morally at odds. The ingredients of action are so finely calibrated in their nexus that, in a forceful stage presentation as in life, it is usually hard to separate what is imitable from what is to be avoided. And perhaps, from the point of view of the intel-

[43] Mala interdum speciem honesti obtulere. . . . sunt enim, ut scis, virtutibus vitia confinia et perditis quoque ac turpibus recti similitudo est. Dingel (1974) argues unsuccessfully against Knoche 1941, Mazzoli 1970, and others that this text proves that we cannot extract the notion of a cautionary paradigm from Seneca.

ligent playgoer who prefers moral opacities, radical separation is counterproductive.

The poets, conjuring up Jupiter's amorous exploits, could be thought to be excusing human errors; lustful characters certainly avail themselves of the precedent. It is said that hellfire and brimstone preachers are depressingly familiar with the experience of building up the challenge of the demon rum, only to have the parishioners, overwhelmed by the attractions of the build-up, storm out to the nearest bar. The danger is greater in the literary artifact, and especially in drama, because the supposedly cautionary tale is invested with formal and psychological allurements that get in the way of the moral. The theatricality, the sparkling rhetoric, and the proud vitality of the Senecan villain stand ready to transform the cautionary, if not into a positive model, into a new compound whose educative dimension is inscrutable. Lessing's early essay on Senecan drama makes much of this difficulty, an interesting by-product of Lessing's enlightenment attempt to strip away the didactic incrustations upon Aristotelian theory.^[44]

In philosophical Stoicism, in the treatises and letters aimed at unencumbered reflection and persuasion, the purity of virtue and the interdependence of virtues is taken for granted, and virtue is allied with reason. "Virtue is according to nature; the vices are contrary and inimical [sc. to nature]" (Ep. 50.8), a remark that does not stop Seneca from pointing out the rampant currency of vices. Virtue is coupled with fortitude, justice, prudence, temperance, all of them issuing from, and ultimately identifiable with, the rational principle in man, which Seneca calls ratio . A man or woman truly possessing one virtue is, by associative implication, endowed with all of them. Like stones in a vault, they support one another. On rare occasions, at brief moments of a delusory reflectiveness, drama subscribes to the same postulate. In Hercules Furens (737–47) Theseus distinguishes between sinners and saints, and terminates his account with an address to kings squarely based on the premise that the Stoic king is an achievable reality. Theseus's philosophical reductionism is short-lived; the premise is undone by the tragedy that passes him by.

The philosophical texts also know another, quite different, story. Taking his cue from Plato's Timaeus, Chrysippus declares (SVF 2.1170) that in nature's production line, the most valuable commodities are

[44] Lessing 1890; cf. also Barner 1973. Braden 1985 is largely concerned with the issue sketched above.

often also, because of their delicacy, most vulnerable to corruption. In the words of Aulus Gellius, who reports this:

morbi quoque et aegritudines partae sunt, dum salus paritur.
sicut hercle . . . dum virtus hominibus per consilium naturae
gignitur, vitia ibidem per adfinitatem contrariam nata sunt.

Disease and illnesses are born as health is being produced.
Indeed, . . . while men are endowed with virtue by the design of
nature, vices are generated alongside by an adverse affinity.
(Noct. Att. 7.1)

Another explanation of evil, also associated with Chrysippus, and akin to the one just cited, says that it exists to show off the good. In the world we know, virtue and vice are functions of each other, with the latter encompassing and harrying the former. Now and then, on rare occasions, the confinement of the good works to its advantage. The encirclement by the forces of evil gives it a strength that allows it to shine forth, in brief flashes, and establish a precarious authority.

Virtue is
Only the irremediable logic of all the anguish
Your cunning could invent or heart devise.
(Robert Penn Warren,
Brother to Dragons, p. 30)^[45]

Or, as Samuel Butler has it more wittily:

The devil, in fact, when he dresses himself up in angel's clothes, can only be detected by experts of exceptional skill, and so often does he adopt this disguise that it is hardly safe to be seen talking to an angel at all.
(The Way of All Flesh, chapter 19)

Whatever the various formulations, experience teaches us that an individual life is a paradigm case of the mutual implication of good and evil. The tragic entailment is a position to which Seneca returns again and again in his philosophical writings, and to which I shall return be-

[45] When Robert Penn Warren's Brother to Dragons was first published, Leslie Fiedler greeted it with a remarkable review entitled "Seneca in the Meathouse," repr. in No! In Thunder (1960). Some of Fiedler's remarks are worth quoting here: "It is a Senecan tragedy that Warren has composed, a play that cannot be played, a poem that must be imagined as acted in the high, ranting style, complete with ghosts and prophecies and dire forebodings, shakings of the earth and raw skulls, suicides and obscene murders and crimes too horrible to define. . . . We have come somehow to believe that bombast and melodrama are hopelessly corrupted, that these shrill, honorable means are in themselves despicable. But we need them to complete ourselves, to do justice to the absurdity of terror, its failure to be well-behaved" (pp. 129–31).

low. Drama, with its temporal limitation and its thematic compression, but also with its age-old call for a pity and a fear responding to human ambivalence, does not run counter to the philosophical expectations. It supplies pregnant vignettes in evidence of their truth. A Stoic king on the stage, if there is such a commodity, must have his faults along with his excellences, orthodox Stoic assumptions about the achievability of pure virtue notwithstanding.

Again, philosophical Stoicism cannot sidestep the inevitable tension between the demand for excellence, with its Homeric implications of strenuous living and the exacting needs of the performative self; and the challenge of social awareness, of humanitas, to philanthropon, which Seneca discusses in De clementia . The debate is continued by the Christian Stoics, who, from Montaigne (Essays 1.1) on, worry over the dubious status of compassion, regret, and other humane stirrings in the heart of the man who wants to be at peace with himself and with the world.^[46] The canonical intellectualism of Chrysippus, already tempered by some of his successors, is shunted aside to leave room, in Seneca's more tolerant taxonomy of morals, for the saving grace of selected emotions. I cannot, Seneca writes, produce a wise man who, like a rock, has no feelings whatever (Ep. 71.27–28). Man consists of two parts, one irrational, the other rational. Seneca's difficulty with this problem shows up in a degree of inconsistency. On other occasions (e.g., Ep. 116; also Ep. 85), he goes along with Posidonius, and perhaps Chrysippus himself: let us stifle the irrational part as much as is in our power.^[47] But the programmatic emphasis on clementia, mildness, in the tract devoted to that political and social disposition, marks his reluctance to "suffer the soul's energy to decline and freeze" (Ep. 16.6).^[48]

And yet, in the prose works, Seneca's view that a good man, even under Stoic auspices, is not devoid of all feeling does not, strangely, extend to misericordia, the compassion a human being feels for the sufferings of another. On this score the plays show a much greater openness for the feelings that bind men together. Note the great choral

[46] The history of the discussion is traced by Dilthey 1960, p. 37 and passim.

[47] For Chrysippus, see SVF 3.431, and Long and Sedley 1987, pp. 419ff.

[48] Ne patiaris inpetum animi tui delabi et refrigescere. For Seneca's view of clementia, and its difference from venia of which he disapproves, see Griffin 1976, pp. 154ff. Cf. also Cicero's contrasting of the Platonic-Aristotelian gratia (tolerance) with the unforgiving hardness prized in the Stoic saint (Pro Murena 61ff.).

odes, or essays, on the sharing of grief, in Agamemnon (esp. 664–69) and Troades:

Dulce maerenti populus dolentum,
dulce lamentis resonare gentes;
lenius luctus lacrimaeque mordent,
turba quas fletu similis frequentat.

It is sweet for one in grief to know
That he but feels a common woe:
And lighter falls the stroke of care
Which all with equal sorrow bear.
(Tro 1009–12)

In effect, the signals built into Senecan drama take us back to an Aristotelian compassion that the early Stoics had rejected as a basis for healthy human intercourse. Seneca also defies Stoic radicalism by coming out for a graduated scale of punishments (NQ 2.44). Meanwhile, more simply, Seneca accepts the special contribution of Stoicism to the old tension: a blackening of the Homeric ideal. On the one hand, Stoicism commends heroism, particularly the Odyssean heroism of patient fortitude. On the other hand, Stoicism turns its back on the heroism of bloody exploits celebrated in the Iliad and the bulk of Greek mythology. As I have mentioned, both Cynics and Stoics on occasion refashioned the figure of Hercules into a model of patience and self-sufficiency. But the Hercules of serious drama usually embodies violent aggression and self-centered pride. In his De tranquillitate animi (2.12), Seneca deplores the character of Achilles: a maladjusted and changeable delinquent. In Hercules Furens, Hercules is the overreacher, the remover of natural restraints (279–93, 955ff.), the destroyer of barriers that guarantee civilized life.^[49] To Amphitryon, who tries to make apologies for his son, the tyrant Lycus, a curious choice for the expositor of the Stoic arraignment of heroes, cites as examples of Hercules' "relaxations":

Hoc Euryti fatetur eversi domus
pecorumque ritu virginum oppressi greges;
hoc nulla Iuno, nullus Eurystheus iubet:
ipsius haec sunt opera.

[49] See especially the good remarks of Owen 1968, pp. 302–8. Also Zintzen in Lefèvre 1972. Zwierlein 1984 now offers an interesting, but in the end unpersuasive, argument in support of the thesis that Hercules, in HF and elsewhere, is an admirable hero, unhubristic, a Stoic saint.

Look at the house of Eurytus uprooted, and flocks
Of innocent girls subdued like helpless sheep.
No Juno gave these orders, or Eurystheus:
These are his  works alone.
(HF 477–80)

Stoic drama continues to throw the spotlight on those whose greatness spells their doom. Caesar, in Caesar and Pompey, and Bussy have persuaded themselves that, in their craving for greatness, they can flout the laws that apply to ordinary men and women. In Seneca's own plays the laws count for less. The polarities that articulate his dramas are dictated by the guiding principles of an individual life, not by the needs of the commonwealth. The human failures are enacted within the area of private contentions. Hippolytus, clearly designed to charm us with his purity and his thoughtfulness, is not a sage, or even a proficiens, a man attempting to train himself in the good. The corollary of his purity is a loathing of women sealed with a fourfold argument from impossibility:

ignibus iunges aquas
et amica ratibus ante promittet vada
incerta Syrtis, ante ab extremo sinu
Hesperia Tethys lucidum attollet diem
et ora dammis blanda praebebunt lupi,
quam victus animum feminae mitem geram.

And sooner shall you scramble fire and water;
Sooner shall dangerous quicksands offer safe
Anchorage to ships; and sooner yet
Shall Tethys from her utmost western bounds
Bring forth the shining day, and savage wolves
Smile fondly on the timid does, than I,
Subdued, will melt in kindness before women.
(Phae 568–73)^[50]

The ineradicability of his hatred of women is endorsed by a reach into unthinkable disruptions of nature. In the same play the nurse, the stock figure designed to frame heroic excess with popular shrewdness or timidity, delivers a Stoic sermon, shot through with aspersions cast on popular mythology, soft living, and indulgence in one's passion (195ff.). But before long the nurse surrenders principle and offers her support to her mistress, out of fear that she might kill herself. Thus

[50] For Seneca's partiality to the argument from impossibility, see chap. 7, pp. 194–203.

decency tempers her resolution. Fellow feeling issues in corruption and sin just as surely as in that ultimate reductive specimen of philosophical tragedy, Brecht's The Measures Taken . In a Stoic drama, the hero and his party owe the darkening of their characters to a variety of sources: the turbulence endemic in the classical models, the corrosive spell of the temptations and the dilemmas a tragedy cannot do without, and last, but not least, the debilitation with which radical Stoicism saddles both political necessities and humane forbearance. In Oedipus it is the environment that closes in on the hero and lays him low; in Troades it is Andromache's fear for her child and Ulixes' obedience to reasons of state that tilt the complexion of heroism toward the black.

At its worst the old Sophoclean hero, legatee to Homeric arete, becomes, under the pressure of the Stoic insistence on the life of reason, a hero-villain, a Satan as much as an Adam. By a potent anticipation of Dante's insight into the near-identity of punishment and crime, the Senecan hero-villain creates for himself a life of greed or lust or fear that is both his dereliction and his penalty. In Seneca's words: sceleris in scelere supplicium est; the punishment is in the crime (Ep. 97.14). The miasma draws the gods into its fold. Juno proposes to inflict mighty suffering upon her stepson Hercules (HF 110ff.); in the process she ravages herself. In using Thyestes as an instrument for his own chastisement (Thy 259), Atreus unknowingly prepares his own person for the same experience. The outward flow of evil is irreversible. "Senecan characters . . . do not commit evil out of calculation for specific gain but because they feel they ought to";^[51] or, which is saying the same thing, because the Stoic experiment in rationalizing and taming the soul makes excessive demands on them. Because victims as well as victors share in the passions—Megara's hatred of the oppressor (HF 380ff.) is a case in point—everybody who counts for something in the plays is, on the Stoic scale of values, a delinquent, if not a villain. The refined verse and the glitter and the subtlety of the psychological analysis refract the sense of radical evil and suffer us to continue to talk of heroes. The imperatives to which the Senecan characters respond are difficult to accommodate within a clearly weighted scheme of moral differentiae. But the larger than life impression of tragic achievement remains.^[52] And it is because of this that the didactic force,

[51] Braden 1970, p. 17. See also Pack 1940 on the terms for guilt, sin, and error in what Pack calls Senecan "melodrama."

[52] This is ably argued by Braden 1985.

the discriminability between models, and the very possibility of translating them into temporal action, must remain in doubt. The conceit of the ghost of Tantalus, that he would rather be in hell than on the Atreidae's earth (Thy 68ff.), and his advice to the dwellers in the underworld to love their tortures—amate poenas! —are the logical consequence of a radical Stoicism echoing the old Socratic conceit that an unjust man being punished is happier, or at least better off, than one who is not. The ghost poses as the warner (90ff.), but the ingrained fallibility that informs his own past life cancels the warner's voice. Stoic ethics stipulates an enormous distance between the ideal Stoic king and the king of past history and myth.^[53] Stoic drama obliterates the distance. A leader, whether in legend or in contemporary experience, is nothing without his vitality, his passion to lead. To appeal to him to strangle his passion, or to expect of him that his smallish foibles will not grow into massive ills, is to defy the deepest political instincts and the stage traditions that feed on them.

"The road through precepts is long; short and productive the one through paradigms."^[54] For the purposes of his philosophical prose, Seneca considers vivid paradigms to be more readily appropriable and economical than the ubiquitous obbligato of commonplaces and aphorisms and precepts, what a modern writer, citing Bertolt Brecht, calls "reach-me-down sloganeering" or plumpes Denken .^[55]Ep. 95, indeed, in tandem with Ep. 94 Seneca's most extended consideration of the usefulness of precepts, finds that for the acquisition of wisdom the "preceptive part" of philosophy is powerless unless the listener's soul is already battened down by doctrinal certainty. The large bulk of the letter, by way of showing what it means not to be properly secured, enlarges on the grossness of modern culture, particularly in the matter of eating and drinking. The language is concrete, the tableaux are vulgar, and one is reminded of scenes in Senecan drama where, presumably, precepts would fall on the same infertile ground.

The conventional distinction is between precepts, intended to stimu-

[53] For a portrait, roughly contemporary with Seneca, of the Stoic king, see Musonius chap. 8 in Lutz 1947, pp. 60ff.

[54] Ep. 6.5: longum iter est per praecepta, breve et efficax per exempla. Herington (1966, p. 443) speaks of "This truly Roman principle."

[55] Eagleton 1981, pp. 62–63.

late thinking about action or the mental equivalent of action, and maxims, pithy formulations of recognized (or paradoxical) truths, from which the precepts derive their value. In his prose Seneca often talks about the relation between sententiae, the adages of tribal wisdom, or decreta, the dogmata of a philosophical creed, on the one hand, and precepts, exhortations, dictates, and injunctions, on the other.^[56]Decreta, Seneca says (Ep. 95.12), are designed to fortify us and to guard our security and tranquillity. It is obvious that, by themselves, they have no business in tragedy; their usefulness depends on how they are blended with other maneuvers coming out of the arsenal of wisdom literature. In philosophical discourse, the plumpe Gedanken, comprising both maxims and precepts, stand out from their environment. Though Seneca habitually uses both maxims and precepts to initiate and endorse his arguments, they are, by virtue of their condensed and self-sufficient form, detachable. The first full-size English commentary on Senecan drama, Sir William Cornwallis's Discourses upon Seneca the Tragedian (1601), is a set of reflections upon eleven Senecan sententiae . In his case they are drawn from the plays, but for all that Cornwallis does with them they might equally well have come from the Dialogi or the letters. The same is true of Fulke Greville's Treatises of Monarchy, which were originally intended as choral songs for tragedies, units built up of maxims and precepts, and then proved too long and were brought out separately.^[57]

In the plays, maxims and precepts are as common in dialogue as they are in the choral essays, often the dramatic equivalents of Seneca's philosophical writings. The interesting difference from the prose writings is that in the plays the categorical separation of the premises—that is, the maxims—and the conclusions—that is, the precepts—is effaced. One is reminded of Paul De Man's showing that in Nietzsche's theatricalized philosophy the distinction between constative and performative is suspended.^[58] In addition, the dividing line between paradigms and sententiae turns problematic. In Senecan drama, as in all tragedy bearing on the great issues of life, it is difficult to distinguish between a proposition that carries its weight by virtue of the person

[56] Bellincioni 1978, pp. 87ff. Note, however, that Seneca's flexible system allows the relations between sententiae, praecepta, and decreta a certain fluidity: Ep. 94.27–18, 95.9–10, and passim.

[57] Charlton 1946, pp. clxxxiif.

[58] De Man 1979, chap. 6.

who offers it, and a truth or a briefing or warning that has no such backing. In the last analysis, because of the play of characters acting upon our imagination, drama contains few statements that are not geared to choices associated with the dramatic agents. What is more, as a dramatist Seneca endows the rhetorical commonplaces with an excitement and a stylistic elegance that sets them far apart from the sober premises of the prose essays.^[59]

We all have a gift for virtue (Ep. 108.8). Even a vicious person, Seneca warns, is capable of rising to the stirring sounds of noble sentiments in the theater. A scoundrel may have enough vestigial goodness in him to respond in his heart when the appropriate virtue is eulogized. But equally so, and more profoundly, Seneca's reliance on encapsulated precepts points in quite another direction: the miscreant applauds the call to rectitude because its respectability permits him vicariously to live a life from which his own diseased counsels should by right exclude him. The converse would also be true: a moderately good soul may be expected to thrill to the fiendishly immoral maxims delivered by an Atreus or an Aegisthus, vouchsafing a proxy admission to a realm of forbidden feeling. Thus every pronouncement, every clever aphorism or urgent piece of advice, is potentially counter-productive. The more impressive the formulation, or the more astutely positioned within the psychological curve of the scene or the drama, the more corrupting it could be. The question discussed at length in Epistles 94 and 95, whether, in emergency situations that leave no room for both, the palm is to be given to decreta, doctrine, without precepts or to precepts without doctrine,^[60] is, as we have noted, moot in drama, where the two merge on the level of the characters' intentions, and where the missionary effect of the pregnant formulations is at best uncertain. In Renaissance drama, which operates with a different understanding of the human individual, an excessive appropriation of maxims may have the effect of reducing that individuality; in the course of Marlowe's The Jew of Malta "the effect of their [sc. the maxims'] recurrent use by Barabas is to render him more and more typical, to de-individualize him."^[61] At the same time, an audience can be roused to a great height of enthusiasm by a proper concatenation of

[59] For Seneca's creativeness on this score, see Owen 1968, p. 313. For the use of commonplaces in a particular play, Phaedra, see Croiselle 1964.

[60] For this, see Macl. Currie in Dudley 1972, pp. 45ff.

[61] Greenblatt 1980, p. 107.

the fragments of tribal wisdom as in Gaunt's great speech in act 1, scene 2 of Richard II, where the maxims carry a patently implied prescriptive message for his son, Bolingbroke.

The urgency with which Senecan drama harps on the pitfalls and perversions of wealth does not sit well with modern readers, and it is doubtful that Roman listeners or readers would have allowed it to move them closer to vows of poverty. In the choral utterances and the dialogue of the older tragedians, such hectoring was incidental, and its import was often structural as much as ideological. In Aeschylus, or Pindar, gnomai served as devices of segmentation. In Seneca they are pervasive; they assume the prominence given to them by the Cynics, the systematic foes of a cultivated social order. Counsels concerning kingship, war, love, fortune, the golden mean and the middle way, death, and sundry other Stoic themes resist dramatic plausibility, whether reflectively delivered by the chorus or shot back and forth in anger between contestants. But their implausibility does not deprive them of a certain grandeur, because they are the themes to which a responsible member of the commonwealth must always return. Seneca's own estimate of the value of wealth, for example, crops up repeatedly in his writings. Characteristically, it varies; at one point (VB 22.4, 23.3) he finds that wealth is not needed for happiness, but that it is capable of making a contribution to it; at another (Ep. 84.11) he stresses its inherent threat.^[62]

Amphitryon and Megara, while waiting for Hercules to return from the underworld and rescue them from the scheming of the tyrant Lycus, engage in a dialogue in which they test the strained relations between wish and belief, between fear and assumption, between faith and fright, between fortune and goodness, between danger and safety, between luck and being caught (HF 309ff.). The series of propositions and repartees is elegant and witty; they have also been regarded as dramatically dead. It has been argued that the same nice distinctions might have been put in the mouth of just about anybody in the play; that, because the truths are self-evident, they defy a linkage with character. But that is underrating the complex affinity between maxim and paradigm, and the special ironies that stamp the Senecan character. Note what the tyrant Lycus, characterized by his opponents as saevus, savage (329), says a little later (402ff.) about war. He promulgates a

[62] Cf. Motto 1970, pp. 225–27.

little essay about fighting and its effect that stamps him, temporarily, as a man of sound understanding. He offers less rant, more reflexion, more sense, and even more sententiae than his adversaries. He concludes: war is wrath; let us put aside hatred. We might have looked to the innocent sufferer, Megara, for these innocent reputable sentiments. It is a sign of Seneca's mature exploitation of the problematic standing of paradigm and precept that, temporarily and ironically, he engages Lycus in their service. Megara, meanwhile, is busy with the full unfolding of her hatred.

It might be argued that the tyrant's lecture is a piece of sophistry. Euripidean antecedents show the way; Jason, Helen, Menelaus pronounce fragments of wisdom gainsaid by their personal ambitions, but convenient in their passing difficulties. In Euripidean drama, however, the deceitful nature of the sermonizing is apparent and an enjoyable part of the intrigue pattern. In Senecan drama, where the hero is stigmatized and the martyr has a share in the world's sinfulness, where the chorus is radically removed from the arena of the action, the dramatist has greater license to attach the shorthand thoughts to the agents of the moment. The looseness is compounded by yet another factor. Seneca is fond of bundling maxims together in staves of five or more, with the not unexpected result that the immediate import gets clouded in the aggregate, with occasionally interesting aesthetic and dramatic results.^[63]

Let us say that where sentiments are introduced in apparent compliance with the speakers' temporary needs, the total picture is, didactically speaking, impenetrable. In the first act of Phaedra (186ff.) the heroine extols the power of Love, who defeats even the gods. The nurse counters with an explanation that Love is nothing but a projection of purely human drives. The chorus, introducing the second act, reverts to the perspective of Phaedra and acclaims the tangible lordship of the god Amor. In this case we are dealing not so much with brief maxims as with fully orchestrated points of view. But, like the shorter formulations, these extended arguments lack lasting resonance, and at the end we are left to wonder, not only what to believe, but whom to credit. In each case the coupling of speaker with what is

[63] For examples, see Canter 1925, pp. 87–89. Canter's collection of materials, though assembled under the auspices of an obsolete understanding of what is meant by "rhetoric" and relatively innocent of analytic development, still has considerable usefulness.

said is, for the moment, authentic and dramatically persuasive. But, fortunately for the tragic substance of the play, the prescriptive momentum lapses into illegibility.

We conclude that sententiae and their combinations, no less than paradigms, lend themselves to a kind of mannerism, a pointed display of literary effects, with little overall relevance to the Stoic burden of the dramatic experience.^[64] Not surprisingly, those who doubt the Stoic complexion of Seneca's plays have pounced on precisely this kind of evidence, even though, as they admit, there are many choral essays that fit smoothly enough into a handbook of Stoic teachings: on the advantages of the simple life, on life as a preparation for death, and so forth.

Near the end of Oedipus, the messenger's speech detailing Oedipus's blinding is framed by two brief choral passages (882–914, 980–97):

Fata si liceat mihi
fingere arbitrio meo
temperem Zephyro levi
vela . . .

If it were mine to shape my fate
To my own pleasure, I would trim my sails
To the gentle breeze . . .
(882–85)

Fatis agimur: cedite fatis;
non sollicitae possunt curae
mutare rati stamina fusi.

By fate we are driven; then yield to fate.
No anxious care can change the thread
Spun by the unswerving spindle.
(980–82)

The refusal of drama to limit itself to an adequately defined lesson is unmistakable. Assuming that for once the chorus are deeply concerned about their king, Oedipus is made to stand as a paradigm for both positions: he failed to take the middle way, and Fate got the better of him. He serves both as a cautionary and as an exemplary model for the thoughts his sufferings stimulate in the hearts of his people. Or take Troades . In the second chorus we are told, in elaborate detail, that the

[64] For mannerism in Seneca, see Shelton 1974, chap. 5, and 1979, and, earlier, Burck 1971. The standard discussion of mannerism is by Hauser 1964 . Cf. also Segal's "Senecan Baroque" (1984). For the difficulties of the term "mannerism" as applied to literature, see Barner 1970, pp. 33–46, and Weitz 1973, pp. 15ff.

underworld is a myth no right-minded person will accept: the stories about hell are "empty noise and hollow words, and a nightmarish tale" (405–6). This is an age-old intellectualist stand, issuing in the precept to keep one's mind healthy and fearless by not accepting the eschatological fables about Hades and Orcus and punishment and rewards. The stand is voiced again and again in the treatises and letters. But then Seneca turns around, and his drama achieves some of its most telling effects by the loving care with which it dwells on the horrors of the netherworld. Once again, the charge is shorted, and reflection or incitement within the drama are stripped of their decisiveness. Seneca himself, in several notable passages of his prose writings, makes no bones about the slipperiness of sententiae and chains of sententiae:

Petis a me, ut . . . scribam tibi, an haec pars philosophiae, quam Graeci paraeneticen vocant, nos praeceptivam dicimus, satis est ad consummandam sapientiam. scio re in bonam partem accepturum, si negavero.

You ask me to . . . write to you, whether the field of philosophy called paraenetice by the Greeks and praeceptiva by us is sufficient for the achievement of wisdom. I know you will take it in the proper spirit if I say it is not.
(Ep. 95.1; cf. also 108.8–9)

Like the paradigms, precepts forfeit a too easy legibility by remaining true, in their aggregate, to the fullness and the caprices of life, and to a philosopher's perception of that fullness. One might ask where this leaves, if not pity and fear, then the moral effect upon the audience, of which the chorus is the built-in representative. The answer is: in the position of the soul, which is most attuned to man's misery and least equipped to help him in his need.

"Some things, we say, recoil from custom, but then, by another route, they return to custom."^[65] No persuasion, in the full sense recommended by the Socrates of the Phaedrus, is to be expected. The rhetoric is, technically speaking, imperfect, because it is not aimed at a defined target. Drama arouses a different kind of persuasion, which has nothing to do with the rhetoric of the school books; and that persuasion is not primarily dependent on precepts or paradigms or lectures for its success. To the extent that they come in, that they fit into the mosaic of the drama's rhetorical energies, they serve such momen-

[65] Ben. 2.35.2: A consuetudine quaedam quae dicimus abhorrent, deinde alia via ad consuetudinem redeunt.

tary purposes as irony, frustration, enrichment, iconographic amplification, and, intermittently, psychological relief. Does this mean that the Stoic cast of the drama is thereby compromised? I doubt that it is, for in their dispersion, in their failure to engage the purposes of an integral agenda, they are not completely unlike the loosely organized topoi of the prose writings, not only of Seneca, but of other Roman Stoics also.

The third chorus of Thyestes, one of the most remarkable constructions in Roman literature (546–622), starts with a praise of fraternal amity, and then proceeds to develop the opposition between war and peace, with a full orchestration of the contrast through scenes from the land, from the larger world, and from mythology. Insensibly, however, logical contrast, designed to throw into relief the loveliness of peace, metamorphoses into the movement between contraries and into the Heraclitean interdependence of opposites. The essay terminates with a voicing of the Herodotean moral of mutability:

Nulla sors longa est: dolor ac voluptas
invicem cedunt; brevior voluptas.

No lot endures. Grief and Joy, each in turn,
Depart; Joy leaves the sooner.
(Thy 596–97)

Like some of the choral odes of Aeschylus, the chorus opens on a note of joy and closes on a note of despair. Unlike Aeschylus, Seneca manages the trajectory smoothly, without the jagged turns that define the Greek dramatist's vision of life. The line of thought is something like this: (1) war has ceased: Atreus loves Thyestes; (2) the cessation is an instance of mutability; (3) (not expressed, but understood by the audience as a likely consequence) mutability will cause Atreus and Thyestes to fall out once again. Individual segments of the Senecan passage are notorious sententiae: nothing is constant, pleasure is short-lived, Fortune and her wheel are in control. Some of the aphorisms cancel each other. The whole poem has a range and a power that far exceed any narrow discursive orientation. But the elements absorbed into it can be traced also in the treatises, and individual letters to Lucilius exhibit a similar smooth restlessness.

To sum up: Stoic drama incorporates components that we associate with orthodox Stoicism, including the suasive and the cautionary use of exempla and the rhetorical use of maxims, aphorisms, and injunc-

tions, without necessarily in each case pledging itself to canons championed by Stoic philosophers. The reason for this is to be found not only in the special conditions of drama, but also in an insight of which Seneca, along with others, is richly aware in his own philosophical writings: that both paradigm and maxim can lead lives of their own, defying the ostensible matrix from which they spring, and that this freedom is not entirely alien to the function of paradigm and maxim in Stoic thought.

If this were all, we might perhaps be allowed to say that Senecan drama, in these instances, draws upon the materials of Stoicism. But we would hardly be entitled to conclude that Senecan drama, as an instrument of moral appeal, is Stoic drama. As long, that is, as we measure the Stoic component in Senecan drama only by glancing at the moral and educative contribution of the instruments of persuasion, the Stoic thesis must remain in doubt. For the neo-Stoics of the Renaissance, as for most of the Roman Stoics (but not for Seneca), Stoicism was a matter of ethics and of psychology, with some attention paid also to the relation between man and the gods. But the great shift from Greek tragedy to Senecan drama of which I spoke earlier is hard to account for solely in terms of how the Stoics defined moral man and his conduct and his religion. For a better understanding, we shall have to explore other aspects of the teachings of the Porch.

Chapter Two
Truth, Speech, Posture

If compared with other poetic genres or literary universals, all drama, and especially tragedy, carries a Stoic stamp. Perhaps this is merely another way of saying that the Stoic saint, the disciplined, articulate leader presenting himself as a role model to the admiration of the flock and acting out his life and death in full awareness of the impression made upon the gallery, is cast in the dramatic mold, as is the Stoic sinner, the villain swollen with his passion. Epic, as pure narrative, untouched by the dramatic incubus that Aristotle piles upon its back, delineates the play of public forces, the lines of energy stretching between competing warriors. Epic realizes multiple collision, conquest and defeat, and the tidy adjustments and reconciliations demanded by the sanguine realism of the heroic order. It is transmitted by a narrative voice pretending to simulate a historical record. Lyric is, in antiquity, either a further development, an interiorization, of epic descriptiveness, or it examines the feelings, the perceptions and the biases, of a single speaker whose solitary voice reaches us from a distance. That voice is often addressed to a companion or a body of companions. In the bond between singer and addressee, the tangles of the lyric sensibility are appraised or unravelled. Neither in the epic nor in the lyric, if these modes can be thought to exist in a pure state, are we brought face to face with a struggle, a precarious moral conflict, a daring championing of unusual values, acted out with an eye to the impression registered directly upon members of the tribe.

To be sure, in Horace's Roman odes, Vergil's Aeneid, and other lyrics and epics derived from these imperial models, we encounter a fair measure of the selfconsciousness and the clash of responsibilities that we associate with serious drama. In their abstracted forms, however, as isolated by genre specialists, neither the epic nor the lyric builds on issues of incremental questioning or thrusts the contestants into a position of role-playing. That is left to the drama.

Once we look at the preoccupations that characterize mature Stoi-

cism, especially in its Roman form, in Musonius,^[1] Seneca, Epictetus, and Marcus Aurelius, they are recognizably dramatic: the soul in conflict; the ego as exemplar; the agent as performer within the sight of an audience and answerable to it; death as a source of moral edification; the great imperative of either/or, which, on the level of intellectual preference, grants a man the choice between being a sage or being a fool. Above all, this is true of the test-tube situation, the great pointed dilemma, spilling furiously across the bounds of natural law and artistic decorum: Medea having to choose between humiliation and murder, Pentheus hovering anxiously between pride and prurience, Antigone discoursing about husbands and brothers. They find their closest analogue in the so-called Stoic paradoxes, the outrageous fictions under the pressure of which all moral niceties are thrown into doubt. "Cato believes that it is the same crime whether somebody strangles his father or a chicken."^[2] The ramrod dementia that empowers this paradox assisted at the birth of many European tragedies, from Sophocles' Ajax to Artaud's The Cenci . It is difficult to fathom how William Archer, quoted with approval by Christopher Ricks, could have said of Bussy D'Ambois : "Dramatists who could produce effects with such total disregard of nature, probability, and common sense, worked in a soft medium."^[3] Cleanthes is credited with saying: paradoxa, ou men paraloga, which we might paraphrase: paradoxes go against the grain of doxa = normal expectation, but not of logos = larger truth.^[4] Offenses against nature, probability, and common sense form the humus in which the European tradition of high tragedy took root.

The origin of no human action,
No matter how sweet the action and dear, is ever
Pure like the flower. For if sweetness is there, then
bitterness too,
In that hell-broth of paradox and internecine
Complex of motive and murderous intensity
We call the soul . . .
(Robert Penn Warren,
Brother to Dragons, p. 56)

[1] For Musonius, see Lutz 1947 and Geytenbeck 1963.

[2] For the Stoic paradoxes, see Molager 1971. The above aperçu, illustrating the third of Cicero's Paradoxa Stoicorum, may be a version of a remark by Zeno. See also Lefèvre 1970, pp. 69–74, for the function of paradox in Seneca, as compared with its uses in Ovid.

[3] Archer, The Old Drama and the New (Boston, 1923), p. 46, cited by Ricks 1971, p. 349.

[4] SVF 1.619.

Stoicism, alone among the Hellenistic and Greco-Roman philosophies, offers a body of thought and a language that are temperamentally adequate to the demands of serious drama. All tragedy strives for a constellation of issues, many of which are prefigured in the Stoic texts. Our task is to determine whether Senecan drama and its like-minded descendents share in properties that are in excess of that modicum of Stoic affinity. The point of this essay is precisely this, to confirm that, over and above the kinship with the Stoic manner that most drama exhibits, the Senecan revolution made for a large-scale intensification of the obvious analogy between Stoicism and drama. A full answer to this question will become apparent only in the latter part of our inquiry. But some preliminary observations are possible even now.

Can we get help from what Seneca says about tragedy in his prose writings? Regrettably, like other Stoics, and unlike Academics and Peripatetics, Seneca has relatively little systematic commentary on this head. With the exception of a few interesting offsiders, notably Ariston of Chios (third century B.C.E.) and Diogenes of Babylon (second century B.C.E.), the Stoa seems not to have interested itself in poetics, in spite of frequent quotations from the poets, and in spite of Cleanthes' own considerable achievement as the author of the Hymn to Zeus .^[5] Seneca's feelings about poetry and drama emerge only marginally and are not always consistent.^[6] As a critic of literary genres, Seneca is ill at ease and uncertain. On the one hand, he reasons that verse can lighten suffering, and that some poets have, as it were inadvertently, expressed philosophic truths (Helv. 17.3; Ep. 8.8). On the other hand, in certain contexts he despises drama for its immoral tales and for its meretricious appeal to the senses (VB 26.6–8; Ben. 1.4.5). Most of the pronouncements in this vein are glancing and summary. Seneca refuses to face head-on the role of poetry in the life of the mind. Even the famous Epistle 88, on the educational significance of the liberal arts, shies away from the obvious difficulties posed by the achievements of drama. But here and there an incidental remark promises more, and I shall revert to some of them as I continue with the topic.

In Epistle 88 the theater is discussed under the heading of what the

[5] For Stoic poetics, see DeLacy 1948.

[6] See Knoche 1941; Mazzoli 1970 (esp. pp. 122–48, on Seneca's view of dramatic poetry); Dingel, review of Mazzoli 1970, Gnomon 46 (1974): 213. Also Baäumer 1982, pp. 130–36, who is too ready to believe that Seneca concedes to theater a positive moral effect.

Platonizing Stoic Posidonius categorized as the "playful art," ars ludicra, or, to give it its Greek name, techne paizousa or paidiodes .^[7] By this Posidonius means the kind of aesthetic exercise that offers pleasure chiefly to the eyes and ears and must, it appears, be engaged in at one's peril. A caution is in order here, applying to much that is said by the philosophers about the Roman experience of drama. Customarily, when Seneca talks about the theater, the kind he has in mind is the mime or pantomime and related art forms, spectacular representations common in his time that did in fact offer little beyond the satisfaction of crude sensory needs.^[8] The recurrent emphasis on stage machinery and production tricks shows what is involved.

But it will not do to seek an alibi in a laudable contempt for a vulgar entertainment. Seneca's rejection of the theater goes beyond this; it comports with both his training as a philosopher and with his own private sensibilities. In Epistle 7.2 he reveals his larger revulsion: "The bigger the crowd, the greater the danger. There is nothing quite so damaging to a good character as sitting down in a theater. Immediately, because of the pleasure experienced, vices slip in."^[9] Just by joining in with the company of theatergoers, no matter what the fare represented, the life of reason, which is the only one that counts, is disrupted in two ways. The contact with the undisciplined multitude befuddles the mind and undermines opportunities for quiet contemplation, such as is recommended by Horace in an amusing and enlightening anecdote about an old soldier in an empty theater (Ep. 2.2. 128–40). More important, the exposure to the delights or, as Plato puts it, to the flatteries or humoring qualities of the presentation is equally disabling.

At best, Epistle 108 informs us, poetry, and that includes drama, has a propaedeutic value, or it helps to publicize teachings that depend on the sweetening of verse to reach larger numbers of people. But the passage in which Seneca says this employs an image, borrowed from Cleanthes, that seems to take us far beyond the argument from public relations or elementary education.

[7] For a full discussion of Ep. 88, see Stückelberger 1965. Though the letter contains much of interest, it is on the whole a fatuous and unprofitable attack on the humanities, contrasting in its details and throughout with Seneca's many positive statements elsewhere; cf. Stückelberger, pp. 76ff.

[8] Chap. 2 of Barish 1981 canvasses the prejudice fired by this conception of the theater.

[9] Quo maior est populus cui miscemur, hoc periculi plus est. nihil vero tam damnosum bonis moribus quam in aliquo spectaculo desidere: tunc enim per voluptatem facilius vitia subrepunt.

As our breath produces a brighter sound when a trumpet drives it through the narrows of an extended channel and spreads it at last through a wider opening, so our perceptions are rendered brighter by the narrow constraints of verse. The same material, when expressed in prose, meets with little interest and is less effective. But where verse accedes, and well-defined meters sharpen and expose the sense, that perception acquires the thrust of a hurled javelin.^[10]

The image from athletic contests is traditional, prepared for by Pindar and Cicero and others, but in the mouth of Seneca it takes on a special significance, as we shall see directly. Its association with the simile from the acoustics of brass music shows Seneca for once fully alive to the power that poetry, and especially public poetry, may under the proper circumstances possess. If he continues to have reservations about literature, and indeed about all the so-called liberal arts except philosophy, it is because only philosophy, he thinks, succeeds reliably in bringing about salutary action. In this conviction he is at one with the dominant philosophical schools, from Xenophanes and Plato down. The context in which he airs his conviction favors the primacy of philosophy. Still, the passage from Epistle 108 shows what Seneca is capable of thinking when he lets his hair down.

Poetry can be examined by a variety of interested people, by the philologus (historian-scholar-critic), by the grammaticus (linguist-grammarian), and by the philosopher (Ep. 108.3 off.). The same meadow, Seneca allows, offers grass to the ox, a coney to the dog, and a lizard to the stork. But only the philosopher, not the historian or grammarian, can translate the poetic impulses into moral energy. Conversely, and this takes us back to Cleanthes and the image of the trumpet, certain philosophical truths are of such consequence that they cannot be adequately and widely communicated without meter, rhythm, and song. Seneca would go along with his critic Quintilian, who says that what is needed in effective communication is a style (compositio ) that reaches not only the ear but the feelings (affectus ), to create the desired motus animorum, the intellectual and moral animation that both philosopher and poet, not to mention the orator, hope

[10] Ep. 108.10: Nam ut dicebat Cleanthes, "quemadmodum spiritus noster clariorem sonum reddit cum illum tuba per longi canalis angustias tractum patentiore novissime exitu effudit, sic sensus nostros clariores carminis arta necessitas efficit." eadem neglegentius audiuntur minusque percutiunt quamdiu soluta oratione dicuntur: ubi accessere numeri et egregium sensum adstrinxere certi pedes, eadem illa sententia velut lacerto excussa torquetur. Cf. DeLacy 1948, p. 271.

to elicit. The style needs to be equipped, as a hand that is armed is the more capable.

Quintilian's phrase, in which old ideas about psychagogia, going back to Gorgias and Plato, have been refined by a Stoic concern with the passions, says nothing overtly about instruction. The accent is on the generation, deep within the emotive being, of impulses that translate into right action. To repeat what Seneca says, ista ediscamus ut quae fuerint verba sint opera: our reading must translate diction into action (Ep. 108.35). Because the truly profound insights, those that shape a moral stance, are difficult to convey, a departure from quotidian speech may be more productive. At the same time, Seneca does not hold that the speech should stray far from the intended meaning; he does not subscribe to the allegorical method, the interpretive technique that became almost synonymous with the Stoic understanding of older poetic texts. Allegorization as an exegetical tool relies on the Aristotelian assumption that poetry is superior to other forms of discourse in that it is more philosophical. It also assumes that the poetic utterance can be translated into good discursive prose without damage to its complexity or truth, or rather to the advantage of its concealed meaning. It is as well that Seneca shows little interest in the opportunities or the systematic challenges of allegorization. His opposition (Ep. 88.5) relieves us of the need to look for hidden solutions to the difficulties posed by Senecan drama. It is a splendid vindication of Seneca's ability as a writer of transparent verse and untransparent tragedy that in his pronouncements on literature he stands clear of the farfetched explanations by ambitious critics who try to make sense of Homer and Hesiod and Plato by pulling them down with them into a theological underbrush.

In sum, there are occasions when Seneca objects to poetry and to tragedy as he knows them; but in a number of texts he leaves room for a larger appreciation. As he says in a remarkable passage on classical writings, there is no book among them that will not bring home to you a wealth of examples of how men differ and of the uncertainty of their lives and of tears prompted by innumerable causes.^[11] The philosophical essayist who follows his mentors, both Epicurean and Stoic, in an

[11] Pol. 11.5: nullus erit in illis scriptis liber qui non plurima varietatis humanae incertorumque casuum et lacrimarum ex alia atque alia causa fluentium exempla tibi suggerat. It should be noted that these appreciative comments come in a letter of consolation and are colored by the contemplation of death.

easy condemnation of the meretricious arts is also alive to the emotional impact of great verse. Contrast a later courtier who had difficulties with his prince and who had read his Seneca very carefully. At the beginning of his Consolatio, Boethius has Philosophy chase away the scenicae meretriculae, the "theater molls," on the grounds that they have no remedy for his grief, but on the contrary will aggravate it with their sweet poisons.^[12]

In his role as a moral guide and imperial voice, Seneca naturally preferred to express himself in the flexible, but unequivocal, cadences of the philosophical tract. But no great leap of the imagination is required to see him turning to verse and to fully dramatized argument where he wished, not to teach accessible lessons, but to submit the more controversial, less settled movements of his thinking to the test of a confrontation with the realities of a familiar world. In the process, one imagines, he was also intrigued by the possibility of creating something new, an art that might show him to have written finis to a moribund tradition, and to have put in its place a worthy and enriching successor.

A detailed appreciation of the speech patterns of Senecan drama must wait upon a later discussion. But it is worth pointing out here that his dramatic language shows the same tendency toward rapid dialectic, the same strenuous exploitation and exhaustion of an issue by means of antitheses and progressive qualification, that is also found in his essays.^[13] Some of our best critics have said that the preference for antitheses that marks the Senecan argument is an index of the discord in his being. But the antithetical style marks, not the idiosyncrasies of an unsettled individual, but the climactic distillation of the insights of a whole movement, reinforcing the natural tendencies of serious drama. Seneca's use of antithesis is singular for the simplicity of its expression, the pared-down moves of active minds locked in combat. The self-questioning of Clytaemestra in Agamemnon (108ff.), the dialogues be-

[12] For Seneca's influence on Boethius, see Lerer 1985.

[13] For recent summary discussions of the stylistic properties of Senecan dramatic writing, see Fantham 1982, pp. 24–34, and Tarrant 1976, pp. 25–27. For the "dramatic" style of the essays, see Traina 1974. It remains true, as Tarrant (1978, p. 257 n. 179) comments: "There is no comprehensive study of Seneca's tragic language." On stylistic agreements between Seneca, Lucan, and Corneille, see Wanke 1964, pp. 117ff.

tween Atreus and his minister (Thyestes 245ff.) and between Medea and the nurse (Medea 155ff.) proceed with a spare and concentrated ferocity, without the artificial flourishes and the levelling symmetries of the Gorgianic or Euphuistic style. Here is the gist of what Phaedra's nurse says to her mistress in a vain attempt to dissuade her from her impulse (Phaedra 145ff.): Theseus does not see the crime—but Minos does; we can hide the crime—Sol and Jupiter will uncover it; the gods look away—your conscience cannot.

In the plays this taut dialectic is supported by a consummate skill in shuttle-speech (stichomythia ), the highly stylized arrangement of one-or two-line utterances and counterutterances.^[14] The sequences are always brief; the longest piece of shuttle-speech in Senecan drama is 13 lines long; Euripides' Ion has one of 115 lines. As against the practice of the Greek tragedians, who use shuttle-speech for a number of purposes, Seneca mostly reserves it for attack and counterattack, with key phrases or terms signalling a tortured and oppressive line of development. The combat is not one of close engagement, but of the refusal to consider what the opponent has to offer. Later I shall have more to say about the reasons that prompt the participants in these altercations to face away from each other. Here it suffices to stress the remarkable economy, indeed, the density, of Senecan stichomythia .

Seneca tells us himself, in the 59th Epistle, that he prefers a simple, compressed type of speech, emboldened by striking, but not elaborate, images. His principal advice is to keep the clauses brief. Helpful communication between decent people does not demand an elegant or contrived idiom; still, polish is permitted to the degree that it does not obscure the message (Ep. 75.1–6). The model is that of medical advice; we do not expect the physician to wax eloquent, but if he does there is no objection as long as his briefing is effective. In all these respects the essays and the dramas obey the same rules, although Seneca had his critics—the emperor Claudius, Quintilian, the orator Fronto, and others—who felt that the writing in his essays lacked punch.^[15] In the Renaissance the anti-Ciceronians, notably Erasmus, readily acknowledged Seneca as a model for their plainer, more pointed, less broadly academic speech, misleadingly called "the Senecan amble."

[14] Seidensticker 1970. My sketch of the chief properties of Senecan stichomythia is heavily indebted to Seidensticker's exhaustive and enlightening treatment. The book is useful also for an understanding of the practices of the three Greek tragedians.

[15] The best compilation of ancient judgments of Seneca's work is found in Trillitzsch 1971. For a continuation into the Renaissance, see Trillitzsch 1978 and Schmidt 1968.

The same aggressive simplicity is the norm also in the images and tropes that enliven the Senecan style, again both in his prose and in the plays. This may seem surprising. The common notion of Senecan rant, with its presumption of prolixity and bombast, is firmly entrenched. For anyone who is familiar with the rhetoric of the Renaissance dramatists, it is easy enough to suppose that Seneca's tropes must be swollen, unwieldy, and indigestible. Not so. Both dialectic and imagery, in their relative economy and transparency, conform to the preferences of the Stoic-Cynic diatribe as attested, for instance, in Cicero's Stoic Paradoxes . The imagery employed by the Stoic teachers is not rich and allusive, but designedly natural, even obvious.^[16] Much of it is taken from the texture of everyday experience, from the customs and the furnishings of the ordinary household. It has been conjectured that Vergil's most homely image, the bucket of water reflecting the light of the sun (Aeneid 8.22–25), derives, via Apolionius Rhodius, whose Argonautica Vergil is imitating, from a Stoic source.^[17] In Stoic treatises the images are employed to illustrate or illuminate topics in the areas of ethics, of physics, and of anthropology. In Senecan drama also, the images, though indebted to the practice of the epic more than to that of the philosophers, serve as models of demonstration or illustration. They are external to the surface meaning to be clarified, but never so distant as to produce difficulty or friction.^[18] Even where the imagery is grandly developed, as in the lengthy sequence of the behavior of Cerberus in Hercules Furens (783–827), the pictorial scheme does not pull the understanding about, and has little of what Bottom calls "Ercles' vein."

What is true of the images is true also of Seneca's diction and syntax as a whole. Why then do so many critics feel that one of the contributions of Seneca's dramatic art to the Renaissance stage was to encumber the language? As one writer says about Bussy D' Ambois (the same might have been said about Tamburlaine or Antonio's Revenge, not to mention the Tudor Senecans): "All the notorious characteristics of the Senecan aura are present here: the epic similes, classical mythology, eruptions of Latinized vocabulary, syntactical tangles which sometimes defy unravelling."^[19] The Latinate diction may be in imitation of

[16] Rolke 1975. For the dramatic appropriateness of Seneca's similes and comparisons, see Primmer 1976; also Liebermann 1974 chap. 2.

[17] Fränkel 1968, pp. 376–80.

[18] The issue of imagery will be taken up again in chap. 7.

[19] Brooke 1964, p. xxxii.

Seneca's speech, but cannot very well be thought a fault of Seneca's; syntactic tangles are pervasive in Chapman, but much rarer in Seneca than they are in, say, Sophocles; epic similes and mythological references do not necessarily create stylistic corruption.

What Atlas or Olympus lifts his head
So far past covert, that with air enough
My words may be informed, and from his height
I may be seen and heard through all the world?
A tale so worthy, and so fraught with wonder
Sticks in my jaws and labors with event . . .
(Bussy D'Ambois 2.1.25ff.)

It is the bathos of the thought expressed, rather than the speech itself, that grates upon our nerves.

The more serious objection to Seneca's, and the Senecan, style rests on weighty authority. Bacon's words are typical of the Renaissance fault finders:

Little better is that kind of stile . . . which neer about the same time succeeded. . . . The labour is here altogether, That words may be aculeate, sentences concise, and the whole contexture of the speech and discourse, rather rounding into it selfe, than spread and dilated : so that it comes to passe by this Artifice, that every passage seemes more witty and waighty than indeed it is. . . . And this kind of expression hath found such acceptance with meaner capacities, as to be a dignity and ornament to Learning; neverthelesse, by the more exact judgments, it hath bin deservedly despised, and may be set down as a distemper of Learning, seeing it is nothing else but a hunting after words, and fine placing of them.^[20]

Oratio potius versa quam fusa, discourse modelled rather than extended:^[21] the distinction is elegant and suggestive; what usefulness it possesses concerns the prose style (Including that of Justus Lipsius) that is the immediate object of attack, rather than the drama, in which Bacon took less interest.

It will be noted that the complaint is not about extravagance or rant but about conciseness and wit and the sharpness of internal tensions. T. S. Eliot confirms this for the drama: "Though Seneca is long-winded, he is not diffuse; he is capable of great concision; there is even a monotony of forcefulness; but many of his short phrases have for us as much

[20] Francis Bacon 1640, p. 29.

[21] Here Bacon would differ with Dr. Kettel, a seventeenth-century president of Trinity College, who is quoted as having said that "Seneca writes as a Boare does pisse, scilicet by jirkes" (Tarrant 1985, p. 22 n. 97).

oratorical impressiveness as they had for the Elizabethans."^[22] What gives the language its special character, and in this it differs greatly from the speech of the prose essays, is the barely controlled hysteria of clashing passions and drives unlimited in their hunger. The brittle concatenation of pointed phrases, rhetorical questions, maxims and puns and shuttle-speech serves the purpose of defining a range of appetites and resentments. The severity of the control generates its own heat, which materializes as linguistic selfconsciousness. The long chains of references and examples, catalogues geographical and mythological, the inability to come to a stop once a topic has been broached and exhausted, may appear to err on the side of indulgence. In a subsequent chapter I hope to show that Senecan copiousness is in fact not a matter of mismanagement but part of his larger loyalty to the Stoic matrix. At any rate, Eliot's "short phrases" and Bacon's "aculeate words" are reminders that succinctness and concision are marks of the Senecan manner, no less than the tendency to string the units out in ringing, repetitive proliferation.

The world's a Theater, the earth a Stage
Which God, and nature doth with Actors fill,
Kings haue their entrance in due equipage,
And some there parts play well and others ill.
. . . . . . . . . . . . . . . .
All men haue parts, and each man acts his owne.^[23]

Stoicism prompts theatrical tropes.^[24] If the Roman Stoics, including Seneca and Lucan, focus on playacting in the presence of others, they do so especially at the point where the play draws to its conclusion, the plethora of possible masks has been discarded for the one authentic role, and the snuffing out of life has to be cast in a heroic mold. The hero's eagerness to put his suffering or his passions on display matches the Stoic's penchant for exhibitionism and truculence. Homeric heroism is direct, spontaneous, and untheatrical. It functions best in the heat of the battle when all energies are aimed at the sole purpose of erasing another life. It requires no audience, except for the subsequent

[22] T. S. Eliot 1932, p. 59.

[23] Thomas Heywood 1941: "The Author to his Booke." For the "theatrical" element in Marston's Antonio's Revenge, see Altman 1978, p. 293 n. 53.

[24] See also Epictetus 1.2.16, 1.29.42, 3.22.26.

glorification of the heroic deed. Stoic heroism is a planned, a highly contrived and intellectualized activity. It achieves its full meaning only if it draws attention to itself as the central spectacle in a crowded arena. Self-dramatizing, seeing oneself as an actor with an audience, entails the admission that life has meaning only as a performance, as an aesthetic experience. Otto Regenbogen^[25] refers to Dio of Prusa (13.36) for a vision of Rome that could stand as a motto for Seneca's vision of his world: imperial Rome seems to Dio like a huge pyre Achilles has constructed for the dead Patroclus from many spars and beams, from the bodies of victims and precious garments, sprinkled with fat and oil. Now he towers there with offerings and prayers and invokes the winds to fan the flames and devour the pyre. This grandstanding Achilles, with an eye upon the multitude witnessing the ceremony, is a far cry from the Homeric hero alone with his dead friend.

When the nurse ventures the opinion that a hatred openly confessed undermines the opportunity for action, Medea responds with disdain:

Levis est dolor, qui capere consilium potest
et clepere sese: magna non latitant mala.
Slight is the grief that yields to counsel
And drops from sight; great troubles do not hide.
(Med 155–56)

The theatrical metaphor does not play an important role in Seneca's Dialogi, but comes in prominently in the Epistles (for example, 74.7, 76.31, 108.6–8, 115.14ff.). Seneca compares the students who look for philosophical instruction to spectators in the stalls. The world is an amphitheater in which an audience of immortals watches Cato struggle with Fortune.^[26] In the plays Seneca comes close, on a number of occasions, to making it appear as if attending an exhibition or a performance defined the agents appropriately and constituted their principal fulfillment as human beings. In Troades 1068–1103 the messenger contributes a remarkable sketch of the "theater" within which Astyanax performs his great leap. The messenger's speech is introduced by someone, Andromache or Hecuba (it does not matter; any experienced sufferer of the woes of this world will do), addressing him as follows:

[25] Regenbogen 1961, p. 461.

[26] Prov. chap. 2. From among the many works dealing with the topos of theatrum mundi let me cite Barner 1970, pp. 86ff.; Curtius 1953, pp. 138ff.; Cope 1973; Geertz 1981. For the Cynic view, see esp. Helm 1906, pp. 44ff.

Expone seriem caedis, et duplex nefas
persequere: gaudet magnas aerumnas dolor
tractare totas.

Detail the butchery, and itemize
The brace of horrors. My grief delights in savoring
The gruesome narrative entire.
(Tro 1065–67)

The speaker is prepared for the rehearsal of what is central in Senecan tragedy: the sequence and interconnection of misdeeds, in their mobilized particularity, designed to appeal to listeners who expect the worst and need to have their fears corroborated. Tractare has its own quality: handle, caress, but also manage philosophically, consolatorily. The speaker calls for a demonstration rather than a report. The demonstration comes with its own proper setting. The natural scenery of ruined tower, rocks, and hills, with hundreds of men in various positions, clinging to outcroppings and trees to watch the executions from a distance, suggests a cavea as it might have been sketched by Piranesi. The soldier-spectators weep and regret what they have done to Astyanax, and then turn round to do likewise to Polyxena (1118–20). Her death (1188ff.) renders the stage analogy, suggested in the act of Astyanax, fully explicit. The messenger openly compares the scene to the theater (1123–25). When, in his report, she meets her end, the killing is in slow motion: the sword entering deep into the flesh, the blood rushing from the huge wound, and the "angry" fall are all calculated to bring tears to the eyes of both Trojans and Greeks watching the maiden's martyrdom. The playbill is set and the scenario cannot be changed. This is an instance, in embryo, of the play-within-play convention that was to become so powerful in the later European theatrical tradition.^[27] The fully realized play within the play, often engineered to pull the wool over the eyes of the characters, who are cast in the role of the entertained, maximizes the attractions of manufacturing lives, measuring them against the workings of a broader scheme of things, and displaying them for public approval or disapproval. A later period draws for this purpose on the convention of courtly masques and other entertainments.

In Seneca's own plays there is no scope for such diversions. The sce-

[27] Nelson 1958. Cf. Braden 1985, p. 211: "Hieronimo's play-within-a-play has become a powerful explication of the theatrical structure latent in the Senecan tragic climax."

nic action is too compact and the dramatic energy too turbulent to permit the interposition of a distancing interlude. Still, contrast the theatricality of the deaths of Polyxena and Astyanax with Talthybius's account of the sacrifice of Polyxena in Euripides' Hecuba (521ff.). In the latter the same crowd is said to be in evidence, though there is no mention of Trojans. But the action is handled as a ritual occasion. Nothing is said about the feelings or the postures of the audience; attention is paid only to their silence, their amens, and their moves to honor the dead.

Stoic writings are full of this elementary conviction that men are either actors on a stage or witnesses in the orchestra. Zeno's student Ariston of Chios compares the wise man to a good actor who knows how to play Thersites as well as Agamemnon (SVF 1.351). The conceit was originally associated with the Cynics, with Diogenes and his followers, whose paradoxical doctrines were marked by a histrionic emphasis that made of every act of "philosophical" instruction a piece of grandstanding. Teles, an early itinerant preacher of that movement who lived in the third century B.C.E. , is said to have compared Fortune to a playwright, and to have remarked that you have to know whether you have been appointed a principal or not, and that there is no point in trying to switch.^[28] In the very last paragraph of his memoir, Marcus Aurelius revives a simile used on a number of occasions in his book:

What is so terrible about being expelled from the city, not by a tyrant or an unjust judge, but by nature who put you there in the first place? It is like a magistrate hiring and then dismissing a player from the stage. "But I have not played all five acts, only three." Agreed; but in life three acts complete the drama.
(12.36)

Marcus Aurelius and Epictetus tend to slip into the theatrical trope particularly when they highlight the alienation of the responsible man from his fellows and the loneliness that makes him see himself imprisoned at the center of a watchful universe. The central Stoic dogma that the self is under obligation of attempting to assimilate itself to nature produces a degree of reflection and argumentation that, paradoxically, problematizes the dimensions of the self. Seneca sees the danger: one of the causes of nervousness, he says, is the habit of striking a pose. Let us not dwell on the thought that others are always appraising

[28] Hense 1889, p. 3.

us (TA 17.1–2). Marcus's vision is the more dramatic because he writes about himself, with an urgency that puts his anxieties close to the experience of tragedy. So much so, that in a wonderful passage on the triviality of kleos, reputation (4.3) he can argue the moral: concentrate upon your self (here understood as the limits of the moral man) and regard the world (kosmos ) as transience and life as a figment of the imagination.

The needs of the theatrum mundi affect the plays in various ways. We have to distinguish between self-dramatization at climactic moments; role-playing throughout an action; selfconsciousness based on an inner weakness; dissembling for purposes of intrigue; mediating between the play and the audience for the sake of laying open the fictiveness of the action; and much else that comes under the heading of the theatricality of the Senecan play. The neo-Stoics of the Renaissance perpetuate the image of the theatrum mundi . Juan-Luis Vives, for instance, in his Fabula de homine (1518) exhibits man putting on a variety of masks, of which the mask of homo sapiens is only one; his real being, briefly revealed in a surprise flash of human optimism, is identical with that of Jupiter. More pessimistically, Justus Lipsius, in his essay on dissimulation (De constantia 1.8), comments that the whole world is a stage play. Here the moral is that when people grieve for themselves, they often pretend that they are grieving for a public ill. This raises a difficult issue: where is the boundary between performance and pretending? Stoicism does not recognize a revolt against Rollenzwang such as is brooded over by Strindberg.^[29] Seneca acknowledges that it may be difficult to play one's role satisfactorily (Ep. 80.7). He also sees that it is not easy to distinguish between a healthy and effective role-playing and a tortured selfconsciousness, and recommends an unadorned simplicity, nihil obtendens moribus suis (TA 17.2), avoiding the masking of one's true character. At such moments the Cynic and the Stoic teachings find themselves in minor conflict, resolved along Epicurean lines, with an accent on the delights of simplicity, a common theme in the choral essays of the plays, but significant also as a temporary, and ultimately abortive, truth in the mouths of some of the characters, including Thyestes (Thy 446ff.) and Hippolytus (Phae 483ff.). Such moments of utopian wishfulness are experienced as desperate attempts to get away from the grimmer message

[29] Karnick 1980, p. 161.

that the fiction of peace and simplicity may in fact involve the most contorted playacting within the world of drama.

In Seneca, life as play irrupts in the form of self-dramatization, the refurbishing of the self as a paradigm. It is a function of Seneca's own brooding tendency to see himself as the lonely source of moral energy. Several letters, taken in combination, bring out the implications. In Epistle 10 Seneca calls for self-reliance; we do not, he says, need the active companionship of others. But then, in Epistle 11, he suggests that it is good to be able to think of somebody else, some prized fellow being, watching your every act. Similarly, in Epistle 19, he argues that it is important to examine with whom one eats and drinks, more so than what one consumes. Again and again the moralist seems to be recommending self-reliance and autonomy, but it is clear also that the old shame-consciousness of the culture inevitably calls for the approving presence of others. Without their express sanction, the achievement of the solitary agent would forfeit its value. To quote from the final chorus of Troades again:

gaudet in multos sua fata mitti
seque non solum placuisse poenae.

He is glad to see that many share his fate,
That he is not the only plaything of his doom.
(Tro 1014–15)

Or, more disturbingly, witness the speech of Medea, pointing to Jason as her spectator:

derat hoc unum mihi,
spectator iste. nil adhuc facti reor:
quidquid sine isto fecimus sceleris perit.

One thing, I feel, was lacking:
That Jason saw it not. Count it as not yet done:
Without his presence my crime is nullified.
(Med 992–94)

Self-dramatization, the character's intense recognition that he is creating his role for the delectation and horror of others, and that their presence adds substance to his standing, is also a kind of existential exercise. Medea wishes to become Medea, and Hercules Hercules, to conform both to their own expectations and to those of their enemies and friends.^[30] Even the inner conflicts, the protracted demonstrations

[30] Thomann 1961, 1: 43.

of instability and lack of center, prominent in Agamemnon, but evident in all the dramas, are in the end folded into the dynamics of dramatic careers poised toward self-revelation. "Zerrissenheit " joins with obsession to shape a furious spectacle. The character starts ab ovo and hesitatingly or impetuously grows into what his public mission (and, in the case of Senecan drama, the bettering of his sources) demands. The process becomes particularly effective on the Senecan stage when it informs the climax of a heroic life drawing toward its close. Near the end of Hercules Furens, after a remarkable speech by Amphitryon and a moving, thoroughly unhistrionic dialogue between father and son, Hercules has one last oration, the most hyperbolically theatrical statement of the play. Its extravagance is sparked by its terminal position. He sees himself as if he were another:

iamdudum mihi
monstrum impium saevumque et immite ac ferum
oberrat: agedum, dextra, conare aggredi
ingens opus . . .

Now long enough has there been hovering
Before my eyes that monstrous shape of sin,
So impious, savage, merciless and wild.
Then come, my hand, attempt this mighty task . . .

and calls upon his father to give him the arms with which to kill himself:

arma nisi dantur mihi
altum omne Pindi Thracis excidam nemus
Bacchique lucos et Cithaeronis iuga
mecum cremabo . . .

Give me my arms,
Or else I shall from Thracian Pindus strip
The woods, the groves of Bacchus, and shall burn
Cithaeron's ridgy heights along with me . . .
(HF 1278–94)

The imaginative violence of the speech is a consequence of Heracles' despairing awareness that he has reached the end of his road. His frustration generates the images of an actor carving out ever greater roles for himself. Othello, in his final speech, plays to the gallery, or rather to an audience of his own selection. The terms of his address show him to be an actor on the boards, preparing for an exit whose rhetorical and psychological satisfactions will compensate for the ordeals of his tragedy. Contrast Marston's Pandulpho, pointing to his murdered son's breast:

Man will break out, despite philosophy.
Why, all this while I ha' but played a part,
Like to some boy that acts a tragedy,
Speaks burly words, and raves out passion:
But when he thinks upon his infant weakness,
He droops his eye.
(Antonio's Revenge 4.5.46–51)

Here the theatrical trope is tied to revulsion and disavowal. Pandulpho thinks self-dramatizing a weakness and prepares to escape the convention. But the convention is deeply entrenched in the Senecan manner, which Marston continues to use to good effect, in spite of the potential for comedy that self-inflation and cocking a jaundiced eye at one's own posturing carry with them.

Charles Rosen, in a commentary on Walter Benjamin's analysis of German baroque drama, speculates that self-dramatizing draws upon the admission that life has no meaning, or rather that it has meaning only as performance, as an aesthetic creation in the midst of a senseless void.^[31] This comes fairly close to Eliot's view of the matter in "Shakespeare and the Stoicism of Seneca." In the Stoic play the hero clutches his flaw or his crime, and his exquisitely contrived end, as if they were things of beauty, and the resonance of the final speech converts the weakness into triumph and the ugliness into splendor. And because the exemplary Stoic life is a constant preparation for death, the terminal coloring may be found at every step. The ghost of Thyestes appearing at the beginning of Agamemnon (31) speaks lines that in their hyperbole are a fine instance of this consummative cast on the part of a villain who has reached the end of the line:

non pavidus hausi dicta, sed cepi nefas.

No translation could do justice to the Latin, if only because, as often, it dispenses with the possessive adjectives, and thus absorbs a whole world into the first-person action. Approximate overtranslations might be: "I did not timidly sip the ordinances of Fate through a straw; I

[31] Review of Walter Benjamin 1977, New York Review of Books, 27 October 1977, pp. 31–40. Benjamin's epoch-making book on the German Trauerspiel does not pay much attention to Seneca, but many of its insights are pertinent to our inquiry. But cf. the running critique of Benjamin in Rumpf 1980. Rumpf questions many of the details of Benjamin's historical reconstruction, without seeming to realize that historical reconstruction plays roughly the same role in Benjamin's book as in Nietzsche's The Birth of Tragedy .

drained it all, and it was crime." Or again: "Unafraid, without hesitating, I ingested the ordinances of Fate, and made them mine: a crime." The line may be an evocation of Aeneid 6.624,

ausi omnes immane nefas ausoque potiti,
All dared a heinous crime and won their gamble,

the summary statement about the great sinners in Hell. But Aegisthus's first person singular, the built-in negative foil augmenting the effect, and the simple, but telling, imagery from heavy drinking, point up the Stoic sharpening of the concern with the self.

Similarly Megara, in Hercules Furens (380ff.), regrets that she has to share her loathing of the tyrant with the people of Thebes. "Masochism" is too tepid a term to designate the fervor with which these Stoic heroes and heroines attest publicly their proud bondage to suffering and ugliness. In the same play Hercules describes what he is going to do for penance (1279ff.): he threatens to pile the whole city of Thebes upon his body and bury himself beneath it and be smothered by the collapsing structures. The prediction is exaggerated to the point of forsaking any claim to dramatic or even psychological probability. That is part of its meaning; it is designed to ring out as a heroic vow, a piece of exorbitant propaganda surrendering all moral or biographical significance to the sheer impact of physicality and horror.

In Caesar and Pompey the dying Cato, for once a virtuous and imitable champion, pulls out his entrails for all to see. It is an act of atrocious defiance, or a perverse display of narcissicism, but even more an authentication of his commitment to his cause as he sees it. It is also a demonstration of the Stoic hero's obsession with the repellent, and of the aspiration to convert ugliness into the spectacular. This is in response to the Stoic insistence that the virtue of the sage shines most brilliantly when the obstacles, in the shape of misery and sickness, are most oppressive. Self-dramatizing is given a further edge by the hero's loathing for his surroundings and, by association, for himself. On the Senecan stage, the hero and the villain are, through the agency of this loathing, if for no other reason, one. The villain's macabre delight in his villainies generates the same fearful concentration on the repulsive. Evil, everything that the schooling of the acolyte is meant to fend off, is made concrete in the guise of ugliness, and ugliness, under the pressure of self-dramatization, acquires a lurid glamor.

As the messenger details the ghastliness of what happened to As-

tyanax's body at his fall (Tro 1110ff.), Andromache rejoices in the knowledge that in this also, in the laceration of his body, her son is like his father. Even the most protracted scene of physical violence in the Senecan corpus, Hercules' murder of his wife and children (HF 994ff.), cannot compete with Andromache's joy for its canonization of horror and its enchantment with disgust. Ovid had shown the way,^[32] especially with the hideous details of the combat of Centaurs and Lapiths (Metamorphoses 12.210–535), and Hesiod is a distant ancestor, particularly the "Hesiod" of the Shield of Heracles . But within the Stoic orbit, under the aegis of the conversion of significance into the sensory impact of theatricality, meaning reveals itself as shape, and shape, honed and stroked and feverishly embraced, turns grotesque.

The cult of ugliness is the price this kind of drama has to pay, and pays gladly, for its fixation upon the hero's selfconsciousness. Plutarch is not the most unprejudiced judge of Stoicism, but there is virtue in his comment: "Just as the beetles are said to eschew the fine scent and to seek out the stink, so the Stoic love keeps company with what is most ugly and misshapen, and turns away from beauty."^[33] Why Plutarch associates Stoicism in general with a penchant for ugliness we shall have to explore later. Meanwhile, the disembowellings, the cutting out of tongues and hearts that garnish the intrigues of the revenge plays of the Renaissance and of the martyr dramas of the Baroque, not to speak of Artaudian scenarios on the modern stage, are a further harvest of this obsession with the repellent.

Julius Klein, whose account of Senecan drama, published in 1865, is in many ways still the fullest and one of the most original, though not the most sympathetic, introduced into the critique of that drama the concept of athleticism.^[34] He felt that the common experience of

[32] Fuhrmann 1968, pp. 41ff. Fuhrmann emphasizes that Ovid's interest in "das Grausige" is confined to the Metamorphoses, and that even there it crops up only intermittently, mostly lightheartedly, and without ideological implications or overtones. The importance of Ovid to his successors is, on this score, a matter of technique and aesthetics.

[33] Stoic. absurd. poet . 3.1058a.

[34] Klein 1865–76, 2: 356f., 405. Cf. also Artaud 1958 chap. 12: "An Affective Athleticism," which is, however, addressed to the actor's task rather than to the spirit of the drama. But within Artaud's feverish scheme the actor occupies much of the space Seneca reserves to the character.

gladiatorial games was at the very heart of Seneca's version of the heroic stance. The gladiatorial combat became for many contemporary writers a fitting image of extreme exertion and bloody accomplishment. The analogy is ubiquitous; Seneca rewrites myth so that it conforms to the grandiose or vicious practices of the Neronic period. Choral passages and speeches give him an occasion to enlarge on the building of the Corinth Canal, the drying of the Lernaean swamp, the imperial achievements in song and sports, and in amours. Also in poisoning, the murder of relatives, the dethronement of rivals, and all the other practices that haunt the records of the age. With all this there is a greater concentration upon concrete and differentiated agony than there is in Greek tragedy.^[35]

Seneca's prose also testifies to the popularity of the pregnant sense of "exercise." Misfortune and danger exist, Seneca says (Prov. 4.7) to elicit a brave response, to try the mettle, to summon the quality of the soldier: Hos itaque deus quos probat, quos amat, indurat recognoscit exercet . Even without the provocation of Fate or God, the moralist wrestles with himself. In the words of Marcus Aurelius (7.61): life resembles wrestling more than dancing, because we have to stand ready and without slipping against all assaults, foreseen or not. Stoic language is full of the play of muscles and weaponry, to a degree never developed by the Greeks, who had their own games to draw on had they wished. In the prose works the final conquest, always hinted at but never consummated, often presupposes a twisting and a battering of the soul, which has its analogue in Christian saints' lives. The gladiatorial spirit bears its most exotic fruit in the anecdote told, on whatever authority, by Thomas Heywood, about Julius Caesar playing the principal part in Hercules Furens (he means Hercules Oetaeus but forgets that the death of Lichas is reported, lines 808–22, rather than acted out) and getting so carried away that he killed the slave (a condemned criminal?) acting the part of Lichas, swinging him about his head, "terque quaterque (as the Poet says)."^[36]

In Greek tragedy the agent establishes his commitment and broadcasts his desires and his aversions. Only rarely does he grant us the glimpse of an inner conflict. Where this happens, as in some plays of Euripides, the clash of loyalties or ambitions is shaped as a dialogue,

[35] Thomann 1961–69, 1: introduction.

[36] Heywood 1941, bk. 2, p. 14. By "the Poet" Heywood means Vergil, e.g., Aeneid 1.94.

with much of the decorum of a philosophic exchange or a political debate. This is true even of the most excruciating case of decision-making, Medea's harrowed monologue about the fate of her children (Eur. Med. 1021ff.). The inner conflict is played out with little recourse to the language of warrior or athlete. The Senecan hero's struggle is one of flesh and blood. The self-absorption of the militant who poses as the paradigm observed obstructs the balanced disclosure of natural impulses. To borrow words that William Hazlitt used of another writer:

His strength and his efforts are convulsive throes—his works are a banquet of horrors. They are full (to disease) of imagination—but it is forced, violent, and shocking. This is to be expected, we apprehend, in attempts of this kind in a country, like America, where there is, generally speaking, no natural imagination . The mind must be excited by overstraining, by pulleys and levers.^[37]

May we think of the Greeks as natural Englishmen, and of the Romans as prevenient Yankees? What Hazlitt says about Charles Brockden Brown is not intended by way of a compliment. But we might remind ourselves that what works in a certain kind of drama is less pleasing in a romance. And as Nietzsche said: "Our most sacred convictions . . . are judgments of our muscles."^[38]

Stoic sensationalism and Stoic athleticism are the informing principles by which to distinguish the Stoic hero's death from the death of Socrates, whom many Stoics regarded as a Stoic before his time. Both Plato and Seneca can be quoted to the effect that all life is a form of dying. But Socrates died to live; the death of the ironist, in Plato's Phaedo, is sketched in lightly, almost shrugged off, because it is only a means, a passport to a greater being, or a convenient closure to a full life. The Stoic hero makes of his death a production.^[39] He insists on controlling, prolonging, hastening, enjoying, protesting his death, not

[37] Quoted by Ricks, New York Review of Books, 22 July 1971, p. 12.

[38] Friedrich Nietzsche 1968, p. 173.

[39] The bibliography on the attitude toward death and suicide in Stoic thought, and in Seneca, is enormous, stimulated in part by the fact of Seneca's own suicide (Tacitus, Ann . 15.62–64). All critics concerned with the two subjects have devoted some pages to the topic. Concerning Seneca, special mention might be made of Fantham 1982, pp. 78–92; Griffin 1976, chap. 11; Argenio 1969; Henry 1985, pp. 116ff.; and, either more marginally or more adventurously, Campbell 1930, pp. 134ff. (on "To be or not to be"); Barthes 1972; Daraki 1982; and Tynan 1950, chap. 6, "The Invincible Must," which has much of interest on Seneca, less the dramas than the epistles, and concludes with the judgment that Marston is a faulty Senecan and Chapman a superb one.

for what it promises, or for what it shuts off, but for the expenditure of manifest energy it makes possible. True, of Seneca's heroes only Phaedra, Deianira, and Jocasta fall by their own hands.^[40] But the spirit of the dying is always there, with even Jason and Aegisthus, in all else cowards confessed, showing a readiness for death equal to that of the hardiest. Antigone tries to argue Oedipus out of his death (Phoe 182ff.), pointing out that he has nothing more to fear, and that he would merely be demonstrating cowardice if he were to decide on suicide now. Oedipus, resorting to the same cold, yet impassioned, logic, defines the reasons for his resolve to die. Deianira, the instrument of Hercules' death, announces (HO 842ff.) that she is going to kill herself. But like Oedipus in his play, she decides that mere death would not be condign punishment, and considers various forms of death, including everybody else ganging up on her to murder her, and Juno killing her out of annoyance at Deianira having succeeded where she has not. The argument continues at great length (the longueurs of this play have persuaded most to declare it spurious), and at one point she asks Hyllus to do the killing.

The choruses too, clouded mirrors of the principal messages in the treatises and letters, dwell on the various aspects of death that a Stoic thinker needs to consider. Death as terror, death as liberation, death as one phase in a general cosmic collapse, death as a challenge and death as transubstantiation: these are some of the themes taken up in the choral essays. Some of the choral pronouncements are moving specimens of grand poetry. This is particularly true of the third chorus of Hercules Furens (830ff.), with its haunting picture of vast crowds moving silently through Hades, and its closing lines on the ineluctability of death:

Qualis est vobis animus, remota
luce cum maestus sibi quisque sensit
obrutum tota caput esse terra?
stat chaos densum tenebraeque turpes
et color noctis malus ac silentis
otium mundi vacuaeque nubes. . . . . . . . . . . .
. . . tibi, mors, paramur.
So our spirits mourn, when each feels crushed
In darkness underneath the weight

[40] On the issue of suicide in Senecan drama, see Tarrant 1976, pp. 286–87, and Fantham 1982, chap. 5.

Of this great earth. There chaos reigns,
Repulsive gloom, the hateful dark
Of night, the torpor of a silent world,
And barren clouds.
. . . . . . . . . . . . . .
. . . life is but practicing for Death.
(HF 858–72)

The essay, addressed to Hercules and taking off from his visit to the underworld, fills the dramatic space within which Hercules is off killing Lycus. Other choruses, notably the third chorus of Agamemnon (589ff.), proclaim the gentle benefits of a willing death over one imposed, or over a continued, but useless, sojourn in the guesthouse of life. But the choruses can do no more than furnish a discursive obbligato to the solo profession of the hero. The Herculean spirit refuses to acknowledge a single natural cause of death; all nature conspires to make death, and the death in life, an autonomous reality, without the need of a specific trigger mechanism. Bussy's dying speech may stand for many:

My sun is turned to blood, 'gainst whose red beams
Pindus and Ossa, hid in endless snow,
Laid on my heart and liver, from their veins
Melt like two hungry torrents, eating rocks
Into the ocean of all human life,
And make it bitter, only with my blood.
(Bussy D'Ambois 5.3.182–87)

"Bussy's proudly affirmed Stoicism . . . erupts into an overwhelming sense of disaster," observes Nicholas Brooke.^[41] But the sense of disaster is also a truculent perception of the magnificence of a world in dissolution, and Bussy's indispensable part in it. His gladiator's body absorbs the ravage, as his grammar reflects the dissolution.

"Tragedy is . . . a dress rehearsal for death; it is life breathing, moving, and talking in a winding sheet. . . . it is death wearing a splendid, gaudy mask." But then Kenneth Tynan continues: "There would have been splendid English melodrama had Seneca written no plays; but our tragedy would be miserably depleted had he written no Epistles ."^[42] As should be clear by now, I cannot go along with this separation of the dramatic from the non-dramatic in Seneca's oeuvre. And Tynan's de-

[41] Brooke 1964, p. xxxii.

[42] Tynan 1950, pp. 185–86.

preciation of the importance and influence of Seneca's plays is an instance of his bracing fractiousness. But the compliment paid to the Epistles is helpful. On the level of explicit argument, the Senecan obsession with the challenges and the splendor of death is more easily instanced, and the instances are more easily quoted, in the philosophical works. Epistle 101, for instance, is entirely devoted to the proposition that the only alternative to a life that is not fully and joyously lived is a speedy death, and quotes Maecenas with disapproval because he advised postponement. Epistle 82, on the Stoic evaluation of death, is one of the best: both cheerful and solemn, it celebrates the Stoic virtue of meeting adversity rationally, and that means with full allowance for death as an exercise of reason. One is reminded of Goethe, during the time of Werther, sleeping with a loaded pistol under his pillow. Perhaps the vision of the last chapter of the second book of the Naturales quaestiones is more expressive:

eo itaque fortior adversus coeli minas surge et, cum undique mundus exarsit, cogita nihil habere re tanta morte perdendum. quod si tibi parari credis illam coeli confusionem . . . si in tuum exitium tanta ignium vis excutitur, at tu solatii loco numera tanti esse mortem tuam.

Confront the threats of the heavens, and when the universe is in flames around you, consider that in such a mighty ruin you have nothing to lose. But if you can bring yourself to believe that that convulsion of heaven . . . is aimed at you . . . then you may surely regard it as some consolation that your death is costing so dear.
(NQ 2.59.11–12, tr. J. Clarke
[1910], modified)

This grandly aesthetic conception, which is especially common in this treatise, is, of course, only one of the many, often contradictory, ways in which Seneca talks about death, in both media. The cataclysmic dimension of this vista will be the subject of a later discussion.

Seneca's view of the possibility of survival after death varies from the notion that death is final to the expectation that in the beyond the wise are reunited with other wise men.^[43] The latter vista is, understandably, rare in the dramas. In any event, the dramatic formulations, not to mention the enactments of deaths, are often more powerful. The bravura mood of the many reflections upon death offered by the choruses and heroes aligns them with the other instruments of self-dramatization we have inspected. Tynan contends that "the topmost

[43] Currie 1972, pp. 34ff.

utterances of tragedy hold an essential residue of belief, an inflexibility of mental stance,"^[44] and chooses to recognize them in Seneca's prose writings rather than in his dramas. By the same line of reasoning he is driven also to declare Macbeth nontragic, because of the sea change in the king's resolution at the moment of death. But this narrowing of the conditions for tragedy seems to me to fly in the face of what the European tradition of high tragedy has encouraged. That tradition calls for a consistency of mood, and of power, rather than of belief, though these are difficult to separate, especially in a Stoic environment. If, for the Stoic believer, life is a sustained and declaratory testing unto death, the histrionic stance is bound to undermine the momentum of ideological consistency or philosophical dedication. Marcus Aurelius expresses himself in a way that is unrealizable on the stage:

You have subsisted as a part. You will pass from the scene in what begot (you). Better, you will be taken up again, as change dictates, into its seminal spirit.
(4.14)

Or again, more lucidly:

Do not despise death, but take pleasure in it, for it, too, is willed by nature. . . . As you now wait for the emergence of your baby from the womb of your wife, so you should accept the hour when your tiny soul will drop from this shroud.
(9.3)

Marcus's privileged language and composed faith derive from the particular application of a highly specialized body of doctrine, presented by one who knew how to integrate the discordant elements of that doctrine in a near-saintly sobriety. The stage, even in its choral extensions, calls for a different direction. In this area, the area of life and death and the energies summoned to shape the arc of a heroic career, the analogies between the drama and the Stoic faith are to be sought in the moods and the tangible feel of what it meant to be a Stoic in a hostile world, rather than in points of doctrine. For doctrinal coincidences we shall have to look elsewhere.

[44] Tynan 1950, p. 183.

Chapter Three
Causes, Necessity, Gods

In the search for the Stoic identity of Senecan drama, I have looked at the function of Stoic topoi; at the use of paradigms and precepts to support or, often, undermine the authority of the topoi; at the reflection of the Stoic life-and-death struggle and its histrionic exuberance in choral disquisitions and in climactic scenes of heroism and villainy; and at other, less obvious links between the ancient dramatic tradition and Stoic insights or formulations. There is much here that is impressive or suggestive, and that gives some inkling of why and in what sense Roman imperial tragedy is different from its Greek ancestors. Certainly these are some of the elements that the Renaissance Senecans and the authors of German baroque Trauerspiel looked to in the fashioning of their own dramatic poems. Yet our findings leave something to be desired.

All serious drama, including Greek tragedy, exhibits many of the features emphasized, albeit in a less consistent and concentrated manner. Aeschylus is sententious. The Heracles of Sophocles' Trachinians perches perilously on the line dividing heroism from villainy. The life-and-death struggle is of the very substance of Sophoclean drama, though its artistic execution lacks Seneca's unusual talent for generating hysteria and near-madness. Those who say that Senecan drama cannot rightfully be termed Stoic tend also to minimize the systematic difference between Greek tragedy and its Roman successor. I would agree that it is difficult to pinpoint precisely where the differences lie between the ethical assumptions of, say, Euripides' Hercules Furens and those of Seneca's play of the same name. A close study of the literary structures and of the contingent formal characteristics would demonstrate the ideological and temperamental divergences that do indeed exist. A number of excellent analyses along these lines are available, not only for the Hercules plays but also for others that invite comparison.^[1]

[1] See, e.g., the pertinent chapters in Lefèvre 1972, and particularly the work of Friedrich.

As I now continue to search for the reasons that set Senecan drama apart from other theatrical traditions, and as I begin to focus more narrowly than before on a possible Stoic legacy, I turn to a topic that will, I suggest, bring us closer to our quarry, though the route will be slow and circuitous. I start with a quote from a recent study of genre theory. "To interpret a work as dramatic is to interpret the relation [sc. between mind and world in the work] in terms of causality," observes William E. Rogers.^[2] It is only because it is felt that one thing should follow unmistakably and disturbingly from another that the sequence of events, or the sequence of sufferings, is thought to be tragic. Aristotle made much of cause and effect and their psychological vagaries in drama.^[3] The plot he regards as the best owes its supremacy to the cogent force of its causal structure, to the compelling ligatures that, under the aegis of the probable and the necessary, make of every turn in the action a consequence or a result. Tragedy may well claim to be the principal literary and scenic demonstration of the causal compulsion, of the fiction that everything has a cause in conformity with a larger dispensation, even if that dispensation is intuited rather than clearly discerned. The commitment to causality remains paramount, by implication, even in those examples of modern dramaturgy in which special effects are achieved by the programmatic subversion or attrition of the principle.

Historians of science have argued that the earlier Greeks were beginning to gain an insight into the causal nexus predicated on a notion of natural law, but that the Stoics were the first to make a concerted and penetrating analysis of the cause-and-effect relation and its varieties; that they were the first to state the connection between causal law and induction, and to take the revolutionary step of distinguishing between an understanding of causes and an understanding of functions.^[4] The details of these advances are part of the history of logic and of the history of science, especially mechanics. In one of his Epistles (65; cf.

[2] Rogers 1983, p. 49.

[3] See Else 1957 on Poetics 1452a18ff.

[4] The principal evidence for the Stoic theory of causes is translated and discussed in Long and Sedley 1987, pp. 333ff. For a clear exposition, see further von Fritz 1972, p. 105, and Sambursky 1959, pp. 49ff. Also Christensen 1962, pp. 52ff. The fullest ancient formulation of the Stoic law of causation is found in Alexander of Aphrodisias's De fato, chap. 22; cf. Sorabji 1980, p. 64. For Chrysippus on causes, see SVF 2.336ff. and 945ff. Forschner 1981, p. 85, reminds us that our sources on this head are often defective or contradictory. But there is enough of a scholarly consensus to satisfy the needs of our discussion.

also 19.6), Seneca himself offers a discussion of the Stoic view of causes, dwelling on the delight to be derived from a study of their complexities, of their imperfect anticipation in the thought of Plato and Aristotle, and, in the end, of their coincidence under the one axiom of causality. He maintains that the active cause has a privileged status, with all other "causes" being necessary conditions rather than proper causes.^[5] His essay, not one of his better or more reliable commentaries on the thinking of his predecessors, terminates in praise of God, who, for him, evidently embodies the unitariness of the causal principle.

Underlying all thinking about this subject in Stoic theory was the hypothesis of the pneuma, the corporeal continuum first postulated by Aristotle and the physicians, but worked up into a universal explanatory canon by the Stoics (see chapter 4 below). The pneuma, the all-pervading stuff of divine coherence, came to be the material coefficient of the causal chain. Continuity or, with some allowance for our experience of discrete events, contiguity was recognized as the essence of causation; causes are bodies in motion, making contact and affecting other bodies. Of the Aristotelian causes, the efficient comes to absorb the rest.^[6] The material cause turns arche, originating principle, and virtually disappears from explanations of empirical experience. But that does not mean that the Stoics subscribed to the axioms of traditional mechanics. Their concept of body does not primarily turn upon dimensions or solidity. Rather, it is talked about in terms of acting and being acted upon; it is action that authenticates body.^[7] In Seneca's own words, quod facit, corpus est (Ep. 106.4). Note that God, the corporeal source of all action, has no fixed form.

But this is to anticipate. In the earlier Stoic examinations of cause, prior to Posidonius, who reopened the question and enriched the analysis, the aim is above all to elucidate human action and accountability, and less to explain natural phenomena. But this is a matter of emphasis and illustration rather than a categorical distinction. It goes against the grain of Stoic science to restrict causes, and action and being acted upon, to one realm or another. Issues of human responsibility called for criteria little different from those pertinent to relations within the physical world. Seneca's own imaginative procedure in the

[5] Frede 1980, pp. 227–28. Frede's contribution is superior to all previous discussions of the topic.

[6] Weil 1964, p. 564, of whose formulations I make liberal use.

[7] Cf. Long and Sedley 1987, pp. 272–74.

Naturales quaestiones shows everywhere how the natural cosmos is made to serve as a trope for movements within the psychological spectrum, and vice versa, and that the two form a sustained band of action and reaction.

Chrysippus's inquiry into the relation between fate and (free) will benefited particularly from the distinction, prominent in the Stoic sources, between an antecedent cause, whose effect persists after the cause has ceased its work, and an operating cause, which subsists along with its effects.^[8] Precisely what is involved in this distinction, and how it is made to bear on the vital issues of answerability and attribution, is not entirely clear. But it appears as if the former, the antecedent cause, might be illustrated by the pulling of the trigger that propels a bullet, while the latter, the operating cause, can be seen in the force or velocity that speeds the bullet on its course. In the terms accepted by Stoic thinkers, such a force, and in fact both factors, would have to be regarded as corporeal.

In the contexts of volition and of the play of emotions, the operating causes are carried by the horme, the drive or appetite, and it is they that form the nerve structure of tragedy. The Stoics do not recognize a causeless change; the continuum hypothesis guarantees that everything that happens, and everything that is done and thought and enunciated, is both effect and cause of another effect. In our daily lives the causes that make us act or refuse to act are not always apparent, and are often desperately hidden. It is the business of the dramatist, as it is that of the social philosopher, to employ the probe of his idiom, not so much to lay them bare, as to construct a likely model of their interaction. Our appreciation of the meaning of what characters say or do in a play is shaped by our sense of what makes them speak or act, and how the cause-and-effect filiation is distributed over the arc (or, depending upon the type of play, the vortex) of the aggregate action. (We are here talking about the sequentiality of an experience, or of a text, and not about the coherence of a character and its traits.) The pinpointing of a cause, either immediate or remote, often contributes to the shattering impact of the tragic exposition, especially where, as Aristotle recommends, the cause uncovered is found to be different from the one that had been tracked or suspected.

[8] Sambursky 1959, pp. 60ff., cites the ancient texts.

studious contemplation sucks the juice
From wizards' cheeks, who, making curious search
For nature's secrets, the first innating cause
Laughs them to scorn as man doth busy apes
When they will zany men.
(Antonio's Revenge 4.1.45–49)

That kind of surprise will be less important in Senecan drama than it is in some of the intrigue plays of Sophocles or Euripides. But, as we shall see, the dependence upon a scheme of constraining causality is, if anything, even stronger.

With Aristotle one has the impression that logic and psychology, questions of necessity and probability, are not forced into neat separation, and that some of his pronouncements in the Poetics owe their power and their suggestiveness to the unacknowledged friction between the two realms of consistency. In Stoic thought, the filiation of various kinds of causes becomes ever more complex. The assumption of corporeality, that all causes are bodies in motion, is taken so seriously that it is largely a lost hope to want to isolate the rules of logic and the patterns of psychology from the material behavior of physical masses. Causality, in the words of Johnny Christensen, turns "unitary, universal, and absolute, the basis of a methodological monism."^[9] The violent objection to Epicurus's atomic swerve is an index, among others, of the steadfastness with which the Stoics rejected the possibility of chance events.

Neither Plato's deflection of natural causes via the medium of the errant cause (Timaeus 48a) nor Aristotle's virtual subsumption of other causes under the umbrella of the teleological lent themselves particularly well to the moral and biographical paradoxes of serious drama. The Stoic formulations are more congenial. Everything that is enacted has a cause; the cause may not be apprehended by us;^[10] each cause is also, though we are unlikely to know precisely in what way, a result of a prior causation. This synthesis of dogmatic belief with admission of ignorance, without the derogatory implication of ignorance stipulated in Plato's metaphysics, is more or less what the needs of tragedy require. As one student of the problem puts it in his remarks on the inclusion of the possible in Stoic determinism: "Instead of

[9] Christensen 1961, p. 21.

[10] SVF 2.937. Cf. Edelstein 1966, p. 28.

seeing causation as a one-dimensional chain of actual occurrences they [sc. the Stoics] saw it as a many-dimensional network of potential occurrences . . . out of which, and in accordance with the rules of disjunction, only one course will be actualized."^[11] Here we find both the suspense of the moment prior to the actualization of the effect, insistence upon the need for a causal factor, and an understanding of the complexity of the system within which the operating cause is made to work, all of which call to mind the preoccupations and tensions of great drama. Chrysippus appears to have talked of a "swarm of causes," smenos aition (SVF 2.945). There is, in fact, a remarkable and ironic similarity between this kind of thinking and that of Nietzsche, who, in his Gay Science and elsewhere, protested against the very concept of causality.^[12] To cite a recent authority: "Nietzsche believes that every event in the world is inextricably connected with every other. . . . He thinks that the history of the whole world, or, in more modest terms, the history of each person, is totally involved in every moment."^[13] Starting from the same premise of a pervasive network of interrelation, Nietzsche and the Stoics arrived at opposite conclusions. Nietzsche abandoned the notion of cause as inadequate to the purposes of a life vigorously lived, while the Stoics ratified the pervasiveness as a scientific dogma.

The elevation of the causal principle to absolute rule is known as determinism. The enemies of the Stoics objected to the Stoic hypostases of Fate (heimarmene ) and Necessity (ananke ), and alleged them to signal an impossibly rigid conception of the workings of the world and of man's position in it.^[14] In the Stoic texts, we find subtle distinctions between Fate (what will be), Necessity (what must be),^[15] Providence (what God foresees and assigns), and Fortune (what happens, seemingly at random). For Chrysippus, these are merely different ways

[11] Sambursky 1959, p. 77.

[12] Friedrich Nietzsche 1966, pp. 119f.: "Ursache und Wirkung: eine solche Zweiheit gibt es wahrscheinlich nie—in Wahrheit steht ein Kontinuum vor uns, von dem wir ein paar Stücke isolieren."

[13] Nehamas 1985, p. 149.

[14] The most important ancient sources are, besides the Senecan texts: Cicero De fato 39 and De natura deorum 2.73ff.; Manilius 4.1–121; Plutarch De fato ; Alexander of Aphrodisias De fato . Among the recent modern discussions, I would single out Rist 1969, chap. 7, Sorabji 1980a (esp. pp. 70–88) and 1980b, Sharples 1981, Long 1971, chap. 8, and Long and Sedley 1985, pp. 340ff. For theories of determinism in the first and second centuries A.D. , see Theiler 1946.

[15] Sorabji proposes to distinguish further between Necessity, as it is generally understood, and exceptionless regularity.

of talking about one and the same fundamental state of affairs, and he may have been more latitudinarian than his critics allowed. Alternatively, he may not have been able to solve the difficulties posed by a strict determinism.^[16] Once again the difference between cosmology and psychology, between whether the causal system is viewed as covering all there is or whether the focus is on a human life and its coordinates, remains secondary.

By and large Seneca, in his prose writings, agrees, though on occasion he is intrigued by other ways of ordering the evidence, such as subordinating one of the hypostases to another, or breaking free of the constraints of the system and claiming that the soul is stronger than Fate (Ep. 98.2.). Cleanthes, for one, could not admit that evil as well as good was brought about by divine cause, and drew a line between events providential and fated on the one hand and events fated but not sanctioned by Providence on the other (SVF 2.933). Neo-Stoicism is given to similar scruples. Lipsius objects to the rule of immutable law—his revealing name for it is "Violent Destiny"—which he contrasts unfavorably with God's Providence and the destiny innate in the created world.^[17]

"Violent Destiny," perhaps inspired by Seneca's discussion of the immutability of physical laws (NQ 2.36), is a fitting label for the spirit of causal compulsion alive in Seneca's dramatic world.

quae nexa suis currunt causis.
(Oed 990)

Opponents of the idea that Seneca writes Stoic drama have argued that his tragedy eschews the benign Providence linked with heimarmene in the more mercurial passages of Stoic treatises, and that it repudiates the self-determination of man and the perfection of the gods or God, both of them axioms prominent in the Stoic texts, which we know from ancient attacks upon them. And Kurt von Fritz,^[18] as we have seen, though a supporter of the view that Senecan drama is Stoic, confesses his embarrassment by declining to call it tragedy. In his opinion, heimarmene and tragedy are irreconcilable.

The beauty of tragedy is precisely that the precise affiliation be-

[16] Von Fritz 1972, p. 107.

[17] Lipsius 1939, bk. 1, chap. 18 (p. 39). For the influence of Stoic determinism, from Petrarch to Zwingli's De providentia, see esp. Dilthey 1960, pp. 154–61.

[18] Von Fritz 1962, chap. 1.

tween the various partial perceptions of the principle of causality is not clear.

There is a deep nick in Time's restless wheel
For each man's good, when which nick comes, it strikes;
As rhetoric yet works not persuasion,
But only is a mean to make it work,
So no man riseth by his real merit,
But when it cries "clink" in his raiser's spirit.
(Bussy D'Ambois 1.1.134–39)

Chapman's implicit analogy between the forces that determine a man's life and the powers that shape a work of literature is instructive. Chapman follows this up with a deeply pessimistic sequel:

Many will say, that cannot rise at all,
Man's first hour's rise is first step to his fall.
I'll venture that; men that fall low must die,
As well as men cast headlong from the sky.
(ibid., 140–44)

There are here memories of the kind of thinking incorporated in The Mirror for Magistrates, according to which every casus has its causa, everything that befalls a man is likely to lead straight to his fall.

But seeing causes are the chiefest thinges
That should be noted of the story wryters,
That men may learne what endes al causes bringe . . .^[19]

The conviction that Necessity is operative also in the various clicks of Fortune may be found scattered through Seneca's writings, both prose and dramatic. Note the exchange between Clytaemestra and the nurse

Clyt.:  ubi animus errat, optimum est casum sequi.
Nurse:  Caeca est temeritas quae petit casum ducem.

Where the soul is adrift, it is best to follow fortune.
Blind rashness chooses Fortune as its leader.
(Aga 144–45)

Casus is, perhaps, a term from dice. In this case, with Clytaemestra deciding to let the chips fall where they may, casus is the chance direc-

[19] Campbell 1936, p. 12; the speaker is the ghost of Lord John Tiptoft, Earl of Worcester, in The Mirror for Magistrates (1559) fols. 64 v. and 65 r., lines 29–31 of Tragedy 15.

tion her feelings and actions will take, thus an inner dynamic rather than external cause. The nurse, however, interprets casus as inescapable Fortune.

Omnes cum fortuna copulati sumus ; all of us are fastened to Fortune (TA 10.3), is neither a desperate cry nor a shout of triumph but a gentle, open-minded admission that we are all in different ways wards of the same custodia, stewardship. Fortune can be many things, all the way from seemingly random chance to the strict pattern of metastasis into the opposite: in se ipsa fortuna ruit (BV 4.1), to the consolatory (and, one presumes, ultimately delusory) notion that the soul is stronger than any kind of fortune (Ep. 98.2). Seneca's thinking about Fortune and Fate and their relationship is unusually rich and varied, and much has been written about it.^[20] A fascinating index of Seneca's interest in the issue is his verse translation of four lines by Cleanthes (SVF 1.527) in Ep. 107.10, terminating in the notorious conclusion

Ducunt volentem fata, nolentem trahunt,^[21]

Fate gently guides you, if you consent; if not,
It drags you.

a formulation that recalls many similar conceits in the dramas. Man's proper relation to whatever Fortune has to give is illustrated by means of an image taken from Attalus (Ep. 72.8): like dogs, gulping down the scraps thrown them by their masters and immediately clamoring for the next scrap, many of us are not satisfied with the gifts of Fortune; only the wise man is content, or, as Seneca puts it, plenus: he accepts the gifts joyously, and is at rest with himself. Here we see that Fortune is easily accommodated within the ambit of the Stoic wise man. As Chrysippus said (SVF 2.973), what men call chance is really an aitia adelos, an illegible cause. Atreus's

quis influentis dona fortunae abnuit?

Who'd spurn the gifts of an accommodating Fortune?
(Thy 536)

is the villain mirror image of the saint's contentment.

The refusal of the drama to align itself unequivocally with a primer of tenets deflates any attempt to define the Stoicism of the tragedy in

[20] Pfligersdorffer 1961; Patch 1922.

[21] See the brilliant discussion by Dahlmann 1977, important also for an understanding of Seneca's style. See further, Setaioli 1979.

dogmatic terms. Seneca himself and many of his contemporaries were quite aware of the nice, if varying, differentiae, argued by a succession of Stoic writers, between Fate and Fortune, between Destiny and Chance, between the will of the gods and the constraints of nature. But the heavy pull of paradox and the manipulation of human aggression for a poetry of fullness, rather than clarification, blunt the distinctions, a process our own less sensitive ears help to complete. Let us return to Seneca's observation that "we are all fastened to Fortune" (TA 10.3). The whole passage is worth quoting because of its imagery of the puppet action that goes back to Plato's Laws (645a), but more particularly because of the almost playful imbalance of its details and the suggestion that the total picture counts for more than a specific affirmation.

omnes cum fortuna copulati sumus: aliorum aurea catena est, laxa, aliorum arta et sordida, sed quid refert? eadem custodia universos circumdedit alligatique sunt etiam qui alligaverunt, nisi forte tu leviorem in sinistra catenam putas. . . . quibusdam aliena supra caput imperia sunt, quibusdam sua . . .; omnis vita servitium est.

We are all fastened to fortune. In the case of some the chain is of gold and extended, with others it is short and mean. But what does it matter? The same guardianship has enveloped all; those who have strung the cable are themselves attached, unless perhaps you think that the chain on the left is lighter. . . . Some have their heads bowed down by the power of others, some by their own. . . . All life is dependence.

In this text Fortune is either a subspecies of Fate or identical with it. Chance is a general mischance. Similarly in the plays Fortune is not only the force that directs men and their affairs, but a collective term. for those affairs (Aga 88–89).^[22] The chorus of Phaedra (959ff.) appears to differentiate radically between nature and the supreme deity, controlling the regularities of the larger world from their distant seat; and blind Fortune, championing the cause of the worse among men. But the sequel of the action, with its dovetailing of the natural and the monstrous, and its intrusion of the Olympian into the terrestrial, shows up the untenability of the distinction. In fact that distinction is more absolute than comparable demarcations in the treatises and letters. It is a conventional index of the chorus's shortsightedness rather than a reliable commentary on the action.

Still, as might be expected, in the plays the qualifications of the radi-

[22] For the relation between fortune and guilt in Agamemnon, see the perceptive comments of Lefèvre (1966, 1973). My own views on guilt will be taken up below.

cal concept of Fate are richer than they are in the treatises. A moment such as the one in Hercules Furens where the chorus calls on Hercules to break the bonds of destiny—fatum rumpe manu (566)—to go against the law of nature, is not within the scope of the theoretical writings. Hercules' "rupture of fate,"—that is, his crossing of the boundary between life and death and back—in the end does not do the chorus's bidding, for it embroils him even more tightly in the meshes of another, more constricting fate. This is the kind of modulation some would call irony, which it is the special privilege of tragedy to arrange. A slightly different version of the same dislocation is found at Troades 360ff., the speech in which Calchas announces that the fata demand blood, even more blood than has been spilt by the heroes. This is Seneca's way of motivating the murders that are mythologically given and dramatically necessary. Thus the text and its requirements, the world of discourse, calls for its own species of determinism. Naturales quaestiones 2.34–38 raises the question: if fatum is immutable, what is the point of praying and supplication? and answers it by the seemingly specious solution that the results of prayer are also part of fatum (cf. also Pol. 4.1). The answer acquires a much stronger appropriateness in a dramatic setting whose outcome is largely fixed in the tradition.

Hercules, in Hercules Furens, is hardly a Stoic saint, but the world in which he is a player, and the challenges to which he responds, are most easily understood against a background of Stoic discussions of Necessity and Fate. The same is true of the many references to Fortune in the plays. If the fourth chorus of Phaedra (1123–53) pictures Fortune striking those most highly placed and sparing the lowly,

Raros patitur fulminis ictus
umida vallis

The verdant valley rarely attracts
The thunder's stroke
(1132–33)

the topos is not only a hoary property of the tragic idiom. It is also a ready reflection of the common Stoic idea that our perceptions are not capable of isolating each element in the swarm of causes, and that random chance is an unavoidable corollary of the inadequacy of knowing. Even the concept of an incremental Fortune, of a random shower of boons and benefits (Thy 536), should be appreciated as a signal of human blindness and not as an authorial rejection of Stoic determinism.

Fortuna vires ipsa consumpsit suas.
Fortune herself has swallowed up her powers.
(Aga 698)

Cassandra's resignation (Aga 698) converts the disabling force of fortune into a process destructive of itself. Often Fortune, as random chance, turns out to be a projection of the villainous hero insisting on his fiendishness and deluding himself into regarding it as a token of human freedom.^[23] Stoic pessimism, combining with its creed of causality a willing admission that we cannot hope to discern the various strands of the causal tissue and that we are reduced to manufacturing our own crude triangulations, was well suited to merge with Pauline Christianity as the Renaissance rediscovered it.^[24]

But historians of philosophy have also made clear, in a series of subtle and compassionate reconstructions of the fragmentary evidence, that "Stoic determinism does not exclude a coherent theory of voluntary human action," and further, "that the Stoic concept of moral responsibility . . . represents an advance on Aristotle in raising sharply the problems of heredity and environment."^[25] The latter is important. In Stoic thinking, ethics, theology, cosmology, biology, and psychology are closely intertwined because of the basic premise that the pneuma, the stuff of life of which all vital entities are manifestations, is corporeal. Hence an examination of moral action cannot be conducted without a full accounting of the various biological and environmental factors that enter into it. I shall return to that subject. For present purposes, however, the first part of the statement quoted above is crucial. It is only by hindsight, via the structured retrospective glance enjoyed by audiences in the theater, but rarely by anyone else, that a complex of actions and events and responsibilities can appear to be perfectly determined. Heimarmene as a working assumption makes due allowance for the contributions of the human will, which must be counted as particles in the causal swarm:

[23] Pfligersdorffer 1961.

[24] Schings 1966 and Stalder 1976 have much to say about the influence of Stoicism, and especially of the vagaries of Fortune, in German baroque drama.

[25] Long 1971, pp. 173f. See also Forschner 1981, p. 111, and van Straaten 1977. Sorabji (1980b, p. 282) distinguishes between hard and soft determinism, a distinction by which he means to express his doubts about the success of Chrysippus's project, thus putting himself in league with the ancient critics of the Stoics, especially Alexander of Aphrodisias, who, according to Long, did not give the Stoics a fair shake. See now also Frede 1982.

if the thing
Accomplished would seem to accomplish only its own
Inevitability, and the thing that exists
Would seem to fulfill only its own being,
And to be but the Q.E.D. of a fatal sorites,
Yet the accomplished was once the unaccomplished,
And the existing was once the non-existing,
And that transition was the agony of will
And anguish of option . . .
(Robert Penn Warren,
Brother to Dragons, p. 111)

The most celebrated Stoic text advocating modifications of strict external determinism and preparing the way for the recognition of an inner cause, and of human volition, is, characteristically, not about a human act but about a natural motion. If we want to savor the full implications of the passage, we must go to the parallels Seneca draws in the Naturales quaestiones between physical events (usually meteorological phenomena) and human experiences (usually passions). As reported by Cicero and Aulus Gellius (SVF 2.974 and 1000), Chrysippus said that if a cylinder rolls down an incline after it is pushed, the cause of this is not only the push but also the shape and the rollability of the cylinder. Thus antecedent and operating causes are jointly credited. What follows in Gellius's Latin text may be a somewhat garbled account of what Chrysippus actually wrote:

sic ordo et ratio et necessitas fati, genera ipsa et principia causarum, movet, impetus vero consiliorum mentiumque nostrarum actionesque ipsas voluntas cuiusque propria et animorum ingenia moderantur.

Thus order and reason and the necessity of fate, the very essences and principles of causation, does [sic! ]^[26] the moving, but the force of our intentions and minds and our very actions are governed by each person's will and by his native talent.

It is curious that the human will, participating in the complex of causes, but also, at least rhetorically, standing aside from it, should be illustrated by the image of the cylinder, with its native capacity for rotary motion. The Stoic axiom of universal corporeality presses for the use of examples from the world of physical objects. In any case, with this concession to the power of causes at home in the heart of our moral being, the Stoics, including Chrysippus, who is generally chided

[26] If movet is sound, the singular may reflect Gellius's understanding that the various causal factors cited are synonyms of a unitary principle.

as the most chillingly deterministic of them, come closer to Aristotle's argument (Nic. Eth. 1110b9ff.) against any position that implies that all action is involuntary because it is prompted by external causes. In its essence the Stoic concession is not a concession at all, merely a further assertion of the amplitude of a causal nexus whose strands may show up in the least expected quarters.

Without an acknowledgment of the role of human initiative, an important element in the moral and social sector studied by the philosopher would be missing. It has been claimed that Seneca, perhaps because he was a Roman, and for that reason not tied to traditional Greek ways of defining human motivation, was the first to import voluntas, an understanding of the contribution of the will, into the Stoic debate.^[27] Albrecht Dihle has modified this claim in important respects, though he too believes that it was the Romans, and especially the Roman lawyers, who fully opened up this dimension in our discussion of ethical and psychological realities.^[28] Though the Greeks did not quite have a word to correspond to the Roman voluntas (which is not exactly the same as the English "will" or the German Wille ), it would be difficult to deny that the Greek writers had a fair understanding of volition, in spite of the emphasis on logos and dianoia, reason and intelligence, as the prompters of human purpose and decision. In the protreptic of Seneca's philosophical writings, however, velle and voluntas play an unusually large role: Quid tibi opus est, ut bonus sis? velle : What do you require to be a good man? To have the will (Ep. 80.4). The emphasis is such that it is in Seneca's interest to soft-pedal the Chrysippean talk of heimarmene . True, the Stoics had from the very beginning wrestled with the problem of how the moral requirement of free choice is to be integrated within their larger vision of the causal network. Chrysippus had written on this (SVF 2.974ff.), as had Cleanthes and others.^[29] But where earlier Stoic authors had attached the greatest value to the human instinct for self-preservation, Seneca came to stress the will to live, and the will to make life worth living. Under the pressure of his ethical concerns, the old distinction between the

[27] Pohlenz 1948–49, 1: 319.

[28] Dihle 1982, pp. 134ff. But see the critique by Adkins 1985.

[29] Cf. what Furley (1967, p. 220) says about Aristotle: "the criterion of the voluntary act is not that it is 'spontaneous' or 'freely chosen' or that 'he could have chosen otherwise,' but that the source of the act cannot be traced back to something outside the agent." That is to say, voluntary action is not uncaused action.

wise and the fool is converted into one between two wills, the wills for good and for malevolence. In this respect Seneca may be said to have played into the hands of the ancient critics of Chrysippus and of the strict constructionists of Stoic determinism, and to have once again anticipated the neo-Stoics.^[30]

But leaving aside this particular development, if a tragedy can be said to achieve its effect by cultivating the obliquity of the relation between freedom and necessity, between voluntary action and external constraint, between the moral and the universal, then traditional Stoicism is beautifully equipped to provide the medium, and Senecan drama, with its strained, but programmatic, shuffling between contraries, exploits the tensions to the hilt. These tensions derive from the recognition that specific predicates, at home in the realm of the finite and the particular, can trigger surprising contradictions when fitted into a more comprehensive order. When Atreus says, scelus iuvat ordinare (Thy 715–16), we may freely, if circumstantially, translate it: there is pleasure and usefulness in an aesthetically pleasing, meticulous structuring of the crime. The two planes of ethics and aesthetics, or of law and art, are made to collide. Again and again in Senecan drama, through the dovetailing of irreconcilables, judgments and feelings are made to clash, and are compacted into a poetry of oppression. For in the drama the discordant strains are not serialized and disengaged, as they often are in philosophical prose, but are forged into an overwhelming tissue of syntheses. Arguments for this or that position, for or against wealth, or freedom, or death, will still put in their isolated appearance, but their isolation is not a matter of argumentative clarity or conclusiveness.

The example of Bellerophon cited above (p. 20) warns that a play must be taken complete, in one gulp, and not as a series of separate, or even sequentially unfolding, propositions. In the drama the Stoic alternatives are stripped of their explanatory or referential power and fused together to set up a field of energy within which the agents plot their moves (or rather, as we shall see later, within which the agents are conditioned to move). The language shifts back and forth between Fortune and Destiny, between random chance and the guiding hand of the

[30] Justus Lipsius, e.g., in his more mature works, the Manductio and the Physiologia (note esp. Phys. 1.12, "Fatum et Libertas"), sees no real difficulty in reconciling Stoic fatum with the Christian tenet of the free will of man, as ordained by God; see Saunders 1955, p. 54.

gods. The audience is not invited nor are the characters perceived to subscribe narrowly to one or the other of these categories or formulations, in spite of the patent commitments attached to them in Seneca's philosophical corpus. The effect of a speech on the subject of Fortune, of which there are many in Senecan drama (and which remains a favorite gambit in later drama, such as that of the Viceroy of Portugal in Thomas Kyd's The Spanish Tragedy 1.3), is not to extend and harden our imagination along ideological lines, but to enrich our sense of the fullness and contrariness of the living scene. Thus the drama reasserts its claim to be an imitation of life.

Philosophically we distinguish between opposed stances and accord to one of them our (provisional) approval. The Senecan tragic vision, readied by the weighing of systematic alternatives, grants simultaneous asylum to Jealous Fortune and Blind Error, to Necessity and Will, to the angry power of the gods and the dispassionate functioning of causation. All are featured at a level of intensity that builds up their status as constraints without explaining anything. Fortune, random chance, is in these texts often little more than a sham perpetrated by a speaker who has his own interests at heart, and thus a perverted emblem of human freedom. This would explain Thyestes' remarkable line cited earlier:

Non pavidus hausi dicta, sed cepi nefas.
(Aga 31)

Fortune, he appears to be saying, was the originator of the crime, but he, Thyestes, drained or highjacked Fortune's dictates, and thus commandeered the unspeakable (nefas ) offense.

The drawing of unambiguously logical lines between conflicting assumptions is of merely local usefulness. Against the paradoxes and the fullness of the whole, such topical gains are doomed to insignificance. The Senecan scene is a nexus of seemingly dominant operational causes belying faith in an orderly system, voiced in a language that resolves into darkness. The diction is Stoic; it is the signature of Stoic thinking, stripped of an evident or lasting adherence. The individual terms and many of the ideas are taken from the Stoic canon; the burden of the argument—"argument" here taken in its technical sense, appropriate to drama—derides the more settled findings of the philosophical model.

The victim
Becomes the essential accomplice, provocateur—
No, more, is the principal—and the real victim

Is he whose hand was elected to give the stroke,
But is innocent.
(Robert Penn Warren,
Brother to Dragons, p. 139)

The Stoic paradoxes win out over the Stoic sermonizing.

I have argued that much that is valuable in Stoic thought comes to be carried over into the speech and thought of Senecan drama, though with a difference. Once such item is the Stoic interest in divination. Because of the doctrine of the interconnectedness of all fibers of the causal nexus, and of our blindness in the face of it, induction is readily sacrificed in favor of divination, mantike, educated guessing at probable effects, even far in the future, on the basis of signs.^[31] Cicero's De divinatione, especially book 1, is the principal document permitting us to glimpse Stoic thinking on the subject. Posidonius is said to have written a work in five books on mantike . Seneca's Naturales quaestiones 3.32.ff., on lightning bolts and bird omens, gives us his ideas about divination. We differ from the Etruscans, he says, in that we believe that only certain things that happen regularly, ratione, allow us to make the predictions of the future that come under the head of divination. But as he goes through his evidence, the gulf between himself and the Etruscans narrows. The discrimination between regular, or natural, events and chance events is clearly vitiated by his admission that we just do not know enough about the incidence of lightning and about birds' calls to make truly verifiable inferences from them. With greater knowledge, more links in the causal chains might be discoverable.

Of those associated with the Stoic succession, only Panaetius (second century B.C.E. ) disagreed with the acceptance of divination as a scientific technique faute de mieux . The conservatives assumed that the world is full of signs guaranteed by the benevolence of the gods, a notion repeated by Plotinus (3.3.[48]6.19), and that errors were caused by signs badly read, and not by the unreliability of the signs. Scenes of divination, such as the inspection of entrails for the purpose of reading the will of the gods or the prospects of the future, are turned in Seneca's hands into powerful vehicles of the macabre. Oedipus features

[31] For Chrysippus on divination, see SVF 939ff., 1187ff. Cf. Sambursky 1959, pp. 65ff., and now Long and Sedley 1987, pp. 263–66.

two sequences of divination. In Thyestes (755ff.) the inspection of the entrails of the slaughtered boys—tremunt . . . spirant . . . calentes —is, in its horrible way, a celebration of the vital spirit that infuses all creation, even where it is temporarily stunted. Why Seneca should be drawn to these scenes of black augury will become clear later.

Divination and hieromancy testify to the continuing importance of the gods in a philosophy otherwise committed to the search for natural causes, which deals with the gods under the rubric of physical science. The reading of signs, however mysterious, has been known to proceed without a belief in divine agents. The inspection of unusual signals for the purpose of extrapolating a distant cause or an equally distant effect has its own perfectly secular status. But because the signs deciphered are often taken from the realm of meteorology or planetary motion, areas traditionally associated with divine acts, the gods are absorbed into the inquiry. The Epicureans pushed the gods back to the frontiers of the empyrean, at most bringing them back into the picture as a trope for the sacredness of the bond of friendship. The Stoics preserved the divine metaphor, and indeed accepted both the traditional deities and the long-established unitary God of philosophy, as part of their acknowledgment of the interlocking forces of the universe. The Homeric model of political and social relations within the divine pantheon lent itself elegantly to the task of fixing the attractions and repulsions in the physical world, especially with audiences who needed to fall back on the familiar language of religion and myth.

In perusing the Stoic philosophical texts, one cannot escape the impression that the recognition of the gods is determined by a variety of contexts and designs. On the one hand, as we have seen, the gods are cyphers, publicistic necessities for the sake of interesting the masses, shorthand for more complex or abstract intuitions. The gods are introduced in human form to teach the uneducated that they, the gods, exist.^[32] At a higher level of sophistication, the word "god" was taken in various senses. God as ultimate principle is ontologically different from God as fire, for the latter has form and is capable of quantitative change. Both, in line with the demands of Stoic corporealism, are bodies, or body.^[33] Again, the sources suggest that the gods perpetrate no evil, but are responsible for some evil.^[34] Note the contemptu-

[32] DeLacey 1948, pp. 269–70.

[33] Weil 1964, p. 567 n. 26.

[34] Gould 1970, pp. 156–58. It should be said that Gould has not convinced all his peers of the validity of his thesis.

ous tone with which Cicero's Velleius, an Epicurean (De nat. deor. 1.15.39ff.), accuses Chrysippus of assembling a motley flock of divinities, including reason, nature, fate, necessity, fire, flux, the world itself, and many more. On the other hand, in Marcus Aurelius's notations, the god who counts is a personal divinity from whom he expects nothing but good.

The roots of this conceptual pluralism are found in the older Greek tradition from the Iliad on. Its elaboration in Senecan drama squares with the protean formulations in the Senecan treatises. In Naturales quaestiones 2.45 Seneca emulates the catholicity of Chrysippus by identifying Jupiter with the ruler of the universe; with the world soul, animus et spiritus mundi; with fate; with causality, causa causarum; with providence; with nature; and with the world. Comparable extensions of the semantic field of divinity are found elsewhere in his oeuvre, as in De beneficiis 4.7–8 where fortune is also brought in, and fatum is equated with Liber (= Bacchus), Hercules, and Mercury. There is, we are told in Epistle 41.1, no need to raise your arms in prayer: prope est a te deus, tecum est, intus est; your god is near, with, within you. But lest we put too spiritual an interpretation upon this piece of aculeate Senecanism, we also learn (NQ 6.3.1) that when the sky or the earth gets shaken, this is not owing to any wrath of the gods; there are natural causes; sky and earth are unbalanced by certain faults, just like our own bodies; when they seem to be doing harm, they are really receiving harm. The implication is that the gods can neither do nor receive harm, nor experience wrath. No wonder that Lipsius came to the conclusion (Const. 1.18) that in his remarks about God as a maker of destiny, Seneca did not really mean what he said.

In the drama the gods are endowed with a more formidable, though equally multiform, presence; the publicistic need has become generic necessity. In the epic the gods form an apparatus; they are a mirror in which men can recognize themselves, or they function as the levers of momentous change. But in contrast with the heroes who are the prime objects of the bard's attention, the gods lack a meaningful life of their own. In the lyric, the gods figure as ceremonial authorities, or as the tyrants of this or that passion. Only in the drama do the gods occupy a position and assume a variety of functions that render them indispensable. They represent the order or disorder, and hence the mystery, of the contingencies and the impediments without which the tragic action would be frivolous motion in an empty setting. They can be fellow

sufferers, or tyrants, or objects of inquiry. But above all they are the old dramatic mechanism for evoking the interconnectedness of causes in a resistant cosmos.

Occasionally, in the Senecan plays, we find rather pallid conversations about what constitutes divinity. Hercules Furens 448ff. features a debate between Amphitryon and Lycus. Ostensibly it is about what makes a hero, but mixed in, as a minor premise, is the question of the difference between men and gods. It is Lycus, the Machiavellian realist, who argues the complete separation of the two, while Amphitryon wants to efface the boundary line between them, citing gods who were temporarily men, and men who became gods. Unlike Amphitryon in Euripides' Heracles, Seneca's Amphitryon does not rail against the gods, perhaps because of his investment in the cause of his son, whom the tradition makes the arch-straddler of the line between the two species. Like Lycus, Amphitryon is capable of employing syllogistic reasoning in support of his conviction (445–46). But the conviction is authentic, and his response to the first hints of Hercules' return from the underworld,

audimur! est est sonitus Herculei gradus

Our prayer is heard! It is his step, the step
Of Hercules!
(HF 523)

has all the earmarks of the announcement of a divine epiphany, which indeed had been prayed for by Megara (279ff.). Hercules himself thinks he is, or wants to be, a god (958ff.).

But Hercules and the feelings expressed about him are unusual. Once in a while someone, like the nurse in Phaedra (195ff.), will try to humor or console a sufferer by suggesting that a god is a fiction invented to make sense of an inexplicable occurrence or passion. But as a rule the human agents view the gods as literature and philosophy have taught them to view them, at a distance from themselves and active or inactive in the various guises that human distress precipitates. Indeed, John Herington has argued that Senecan tragedy, like Aeschylus's, is religious theater. Seneca, he says, leaves no ultimate questions open. "Our earliest ancient tragedies, the first plays of Aeschylus, were composed when the Western world was just emerging into an era of free inquiry. . . . Our latest tragedies, those of Seneca, seem . . . to mark the beginning of the reverse process, the transition from free inquiry to an era of religion." But the two playwrights share between them a

"desperately urgent sense of the absolute reality of sin and virtue."^[35] I cannot subscribe to all the terms of this comparison. I do not think that either Aeschylean tragedy or Senecan drama "shows an unquestioning faith in the ultimate workings of the world." They would, in fact, not be tragedies if this were so. But there is some truth to the proposition that Senecan drama is religious drama, in the sense that the presence of the gods is not only taken for granted but exploited for powerful ends.

For "gods" we may also write "demons."^[36] With the exception of the fragmentary Phoenissae, all of Seneca's plays, including the spurious Octavia, have within them demons or, as we would call them, ghosts. In three of them, Troades, Medea, and Oedipus, messengers report the appearance of ghosts; in three, Agamemnon, Thyestes, and Octavia, a ghost or ghosts appear on the scene; and in one, Hercules Oetaeus, the voice of the hero at the end is that of a hero turned demon. It is true that many of these ghosts are revenant humans. But their survival in another world has given them the intermediate, quasidivine status to which the Greeks, from Hesiod on, have accorded the label "demon." Both Chrysippus and Posidonius are known to have written books on demons. As encyclopedic students of human experience they could not disregard the ubiquitous worship of the lesser divinities. The difference between gods and demons is largely one between remoteness and propinquity, between spirituality and incarnation. Demons were thought to be more concretely or devastatingly embroiled than gods in the affairs of men. They were appealed to in the matter of curses and defixiones . For the purposes of drama it is helpful that demons appear, and that their appearance is automatically disruptive and violent.

In Seneca, the distance between gods and demons is diminished. Drama can do little with the Stoic God, whom Cleanthes and Chrysippus and Posidonius identified with the life-giving pneuma . Where the gods are introduced in their own persons, as is Juno at the beginning of Hercules Furens, their malevolence shows them to be indistinguishable from demons. Gods and demons alike are conceived as visual manifestations of the engines of action, of the intractable causes, the unlucky coincidence of errant motives that circumscribe the road the hero has to travel.

[35] Herington 1966, pp. 460–61.

[36] Braginton 1933.

More properly, their horrific presence hints at a fundamental instability to which the world of Senecan drama is heir. Juno runs in her own person through all the changes of temper and of will that are subsequently exhibited in the violent turns of the hero's career. The ghost of Tantalus, in the prologue to Thyestes, embodies several contrary conceptions at war with one another. On the one hand, he is needed in, the palace to initiate and preside over the crime (62ff.); on the other, he must not tarry (105–6) because otherwise the whole world will be desiccated with the thirst of which he is the mythological guarantor; and finally (95) he expresses a desire to stay and prevent the crime, a desire he is incapable of translating into action because he will not enter the palace.^[37] The incongruities underscore both the enormity of what is afoot and the illegibility of the causal skein. And so the supposed engines of action are stripped of their power as intelligible initiators and turn into allegories of uncertainty and terror.

On the Tudor and the Jacobean stages, where Greek and Roman gods are admitted only in very special circumstances, the ghost survives to preserve the sense of divinatio, of the authority of the undecipherable.^[38] The entrance of the demon carries with it a glimpse of the future, of a more inclusive reality. In Seneca's Troades, the ghost of Achilles is merely mentioned; in Thomas Heywood's translation, the ghost appears and has a soliloquy of thirteen stanzas between acts 2 and 3, "in a tone which Peele could hardly outdo."^[39] But the irruptive vehemence of his entry also points up the weakness of those exposed to the apparition. In Ben Jonson's Catiline the two prologues of Envy and of Sylla's Ghost sound a lasting echo of the Stoic belief that the hazards of enforcing circumstance are most strikingly caught in a formulation that conserves, though in a censored shape, the old divinities of the genre. The ghost is a contribution of science rather than religion, the echo of a speculation about the world gone sour. In the absence of an ideological commitment, let alone a confessional identity, it is difficult to see how either Greek tragedy or Senecan drama can be called religious in the sense in which a medieval mystery play or baroque martyr spectacle or a work by Claudel is religious.

In Seneca the gods give themselves either not at all, as in the first

[37] Anliker 1960.

[38] Dahinten 1958. See also Whitmore 1915 and West 1939. Cf. Garton 1972, p. 199: "A drama substituting the merely supernatural for the numinous incorporated in its structure an increasing number of ghost scenes."

[39] T. S. Eliot 1932, p. 84.

chorus of Thyestes, or they parade their malignity for all to see, in the guise of demons and ghosts. Divination furnishes an imponderable link between Stoic science and Senecan dramaturgy. The ghosts document the Stoic awareness that the structure of the world is not completely amenable to methodical inspection, or rather that even the most complete inspection will have to allow for inaccessibilities and surprises. What the Senecan drama makes of this insight goes far beyond the careful adumbrations of a scientific skepticism, and comes down hard on the noxiousness of the inscrutable. Confidence in the face of complexity has turned into black magic. But the language continues to remind us of the vigorous exploration of the causal network that entangles man in a larger cosmic web. As Oedipus sees the ghost of his father, and wonders whether Antigone sees him too,

sequor, sequor, iam parce—sanguineum gerens
insigne regni Laius rapti furit;
en ecce, inanes manibus infestis petit
foditque vultus. nata, genitorem vides?
ego video.

I'll follow, yes, I must; and you, let go.
Look where Laius comes in rage, grasping
The bloodstained scepter of his ravished realm;
Look, with malevolent hands he seeks and tears
My empty sockets. Daughter, do you see
My father? I do.
(Phoe 40–44)

we recognize the source of the ghost of Hamlet's father. Euripidean ghosts, including that of Polydorus in Hecuba, are calm, narratively informative; they are little more than messengers of history, retailers of the antecedents or consequences of the text. Greek gods demonstrate a heavenly assurance. Aphrodite, in Hippolytus, and Dionysus, in Bacchae, can be cool and imperious because in the vision of the playwright they represent forces that, though by no means entirely legible, are thought to be dominant and unrefracted. In Senecan drama, gods and demons lack this assurance. They, like the men and women they can neither assist nor, of their own volition, destroy, are the furious, but impotent, prisoners of an inscrutable universe.

Finally, by way of a postscript to the chapter, a few words about freedom and its implications in Stoic thought and in Seneca, a topic to

which I alluded briefly above. Within the constraints of the causal and motivational machinery, the dialectic between heroism and freedom is unusually problematic. The Cynics, who elevated freedom into a policy of unconditioned liberation, had had no difficulty with the concept because they made it ride roughshod over all other obligations, including those of a civilized life and of the duties of leadership. By identifying heroism, as in the figure of their cult hero Heracles, with the license to act as one wishes, they dispensed with whatever worries attached to the relation between freedom and the challenge of social action. Cleanthes wrote a book entitled On Freedom, and other Stoics have things to say about the idea of an action that is unhampered by manifest constraint. Still, considering the importance we give to the notion of the freedom of choice in the history of philosophy and in the annals of political thought, it is remarkable, or rather understandable in the light of Stoic determinism, how rarely the Greek terms eleutheria and exousia have that meaning in the extant Stoic texts. The Stoic emphasis, where it is found, is on being able to act without being dependent on factors beyond one's control. Freedom means "doing one's own," that is, refraining from any act that would enmesh the agent in the foment of his environment.^[40] As we shall see later, Stoic physics stamps this aspiration as an irreclaimable hope.

Zeno and his successors, moreover, following Plato's identification of the ruler with him who rules over himself, equated the wise man and the "king," the responsible leader with his obligations (SVF 3.332, 617, 691), and thus perpetuated the political accent Greek ethics had borne from the start. This is an emphasis congenial to tragedy, where an affirmation of complete freedom, freedom of choice unchecked by obligations, can only debilitate the tragic passion. Further, in Greek and Roman Stoicism, the wise man is not, even outside of the political arena, completely free to act, but is bound to the logos, the code imposed by his insight into levels of controllability. The language of Origen belongs to a later era. Origen says that the wise man is free inasmuch as he has received the power of willing from the divine Creator.^[41] No such liberation by the grace of God is available to the Stoic. His notion of freedom is coupled with that of the observance of the law itself, and of the limitations thus imposed on the human agent. The Stoa expressly rejects the view that freedom, what is eph' hemin, allows

[40] For what the early Stoics had to say about the psychology of action, see Inwood 1985.

[41] Origen De principiis 3.1.19–20.

for the possibility that we could do the opposite. The good can only act well, the bad ill.

All Hellenistic philosophy, of course, shows a gloomy interest in the possibility of one supreme self-determined enterprise, the radical act of the free choice of death. But once again, this belief in the autonomy of suicide is hedged in by a whole battery of limiting conditions. Where suicide is not prompted by fear or desperation, it is dictated by a rational decision derived from the sanctions of the logos . In any case, such a decision can be made only by those who have steeled themselves by the constant practice of abnegation expected of the Stoic saint. Likewise in tragedy, a decision that is made freely, that might, in fact, have gone the other way, is useless or trivial. The dramatic concern with constraint and compulsion leaves freedom far behind. We must remember also that the Stoic wise man, or saint, or king, is a figment of the utopian imagination. Our world knows only prokoptontes, proficientes, men and women who are attempting to get closer and closer to that ultimately inaccessible state of wisdom.^[42] Even if Stoic kings were, by definition, free, which, in the strict sense of the word, they cannot be, the learners of this world, including those venerated by disciples, are not.

The topic of the Stoic king, of his rights and opportunities and obligations, is a constant one in Seneca, both in his philosophical writings and in the drama. Medea assumes that kings have power; they can be useful, and they can protect and save the weak:

hoc reges habent
magnificum et ingens, nulla quod rapiat dies:
prodesse miseris, supplices fido lare
protegere.

This is what kings can do,
A mighty privilege that time cannot extort:
To succor the afflicted, to provide a firm
Refuge for those in need.
(Med 222–25)

She might equally well have said that kings have the obligation, rather than the opportunity or freedom, to assist the helpless. In Epistle 73 Seneca condemns some radical Stoics for their opposition to monar-

[42] In VB 17.1ff. Seneca rehearses the accusations his enemies have levelled against him. In effect they say that he is a man given to luxurious living and to emotional indulgence rather than a true Stoic. His answer: he is a proficiens rather than a sage and wealth gives him a chance to benefit others. Here we have a portrait of the good, but imperfect, man such as is adumbrated in some of the choral essays.

chy, on the grounds that after all the institution guarantees communal well-being. Public service is, in his eyes, a Stoic demand; he acknowledges this even in the treatise in which he reviews the joys of retirement, De tranquillitate animi . His own forced retirement from politics is certain to have embarrassed him. Curiously, in the second chorus of Thyestes (336ff.), the portrait of the Stoic king is conceived in the terms of the Epicurean lathe biosas, of the call for a life away from the center of the storm. This coincides with the Sophoclean and Euripidean choral sentiment, diametrically opposed to the heroic temper, that it is better not to be exposed to the stresses of politics and leadership.

The Senecan plays and treatises abound in debates concerning the requirements of kingship, and concerning what makes a strong ruler as opposed to a Stoic king. This is particularly true of Thyestes, the play in which the temptations of kingship are graphically tested. Thyestes holds out against the crown, but is softened up by his son Tantalus (Thy 404–90) and yields before the siren call of Atreus's deceitful surrender (534–45).^[43] It is obvious that the term "king" can, under these circumstances, be invested with the most spectacular ironies. The Renaissance stage perpetuates the discussion and the ironies. Andrugio, the attractive victim in Antonio's Revenge, has a fine Stoic speech on kingship in act 4 of the preceding play, Antonio and Mellida . In the later play, act 2, scene 2, Piero and Pandulfo engage in a shuttle-speech on tyranny that is literally translated from Thyestes . A passage in The Spanish Tragedy (3.1.1–11), on the tribulations of being a king, is modelled on Seneca's Agamemnon (57–73). The new political constellations of the Renaissance helped to restore the relevance of those old debates. In Seneca's own plays, however, the Stoic message of what is expected of the perfect king is even more equivocal than it is in the treatises. The debates are conducted, not with the evident aim of helping one or the other view to acceptance, but to define the fluid social matrix within which human purpose and human welfare are buffeted.

One prominent thought stands out from the confusion of conflicting findings, and achieves prominence within the Senecan dramatic corpus: the thought that perfect wisdom imposes the recognition of ties and the willingness to live up to them, and that it is the tyrant's delusion and the source of his crimes that he considers himself free, without obligation:

[43] The process is well described by Seidensticker 1970, pp. 104ff.

Who to himself is law, no law doth need,
Offends no king, and is a king indeed.
(Bussy D'Ambois 2.1.203–4)

The couplet purports to stake out the privileges of true kingship; Bussy's own subsequent dereliction and failure demonstrate that the freedom of the king must turn into the license of the tyrant. And the tyrant necessarily develops a suicidal hunger for greater and greater license.

Wise men can be truly royal only if they cease aspiring to temporal power. Thyestes is made to think he can carry his newly discovered attunement to the simplicities of his sylvan retreat with him into the royal palace (542–43); he crashes in the attempt. But a sage who separates himself from the bonds of the community and refuses to respond to its demands has no business on the tragic stage. The tradition of the polis in Athenian tragedy paradoxically requires the presence of the king at the heart of the drama. Without the active role of princes and kings and temporal leaders and the ruling heads of households, tragic issues fall away. Aeschylus's Persae could not have been conceived as a tragedy if the playwright had put Athenian democracy at the center of the stage.

Stoicism gratefully adopts the old political orientation, along with the ancient myths to which the medium was attached. But Stoic drama also draws attention to the precariousness of the old tragic perception that a king is political man generalized and raised to his highest potential. It further collapses the established dramatic bond between royalty and passion, or rather it exaggerates the bond and makes the king over into a living incarnation of wrath, thus creating a paradigm to face the Stoic "king" from the opposite end of the spectrum. As Medea confesses:

Difficile quam sit animum ab ira flectere
iam concitatum quamque regale hoc putet
sceptris superbas quisquis admovit manus,
qua coepit ire, regia didici mea.

How hard it is to turn the soul from anger
Once it is roused, and how the seizure of
The sovereign scepter spurs persistence in
The sovereign path, I know: the crown has taught me.
(Med 203–6)

The more violent the king, the kinglier he is: this is a favorite conceit of the stage tyrants of Senecan drama. Thus heroism is questioned, and

reduced to the hard contours of parody. The Stoic philosopher-king merges with the tragic hero-villain. Some say that Greek tragedy, especially that of Euripides, is a poetic rebuttal of the pretensions to progress and perfectibility associated with the movement of enlightenment, the Sophists, in the second half of the fifth century B.C.E. Senecan tragedy is not a protest but a continuance. Its kings and queens exploit the warnings and the precepts that Seneca's own philosophical discussions of kingship have made available. But in their own tragic careers they confirm the insufficiency of those precepts and the doom dimly perceived in the warnings.

What should not be overlooked, in the end, is the usefulness of the royal cypher in a dramatic tradition that stresses the imponderability of causes. The royal ideal hints at a maximum of control, at an efficacy of will and determination that human society rarely concedes. The tyrant is the champion of the special presumption that claims freedom from the causal network. As the plays, with their repeated discussions of what it means to be a king, show unmistakably, the concept of the king is just as delusory as the assurance of the tyrant is vicious. Both of them belie the necessary truth that all of us are hemmed in by a swarm of causes that shape our very being and mold our actions. Many of the causes are internal to ourselves. But neither the king nor the tyrant is capable of exploiting this internality for lasting purposes of his own. Within the Senecan world, a clear sight or control of the causes of action is denied to all. Only in a very limited way can Polyxena or Antigone, much less Hercules, be said to be in command of their lives. The reason is not just that the complexity of the network defeats all hope of mastery, but that the causal system may itself be inescapably flawed and diseased, and intrinsically corrupt. The image of the Stoic king, carefully maneuvering between his own needs and the intuited forces of a lawful universe, is a mirage. Senecan drama appears, now and then, to hold out the comfort that the reins of our existence are within our grasp. But because it is tragedy, and clinical tragedy at that, the comfort is extended only in order to be thoroughly demolished. Mastery might be possible in a universe that obeys fixed laws, in a world that is healthy and theoretically analyzable. The universe of Senecan drama is diseased; its causes are those a physician rather than a governor is best qualified to attempt to track. It will be the task of the following chapters to establish that this vision is acceptable within a Stoic understanding of how the world behaves.