Tragedy and history are related - both high mimetic genres, indebted to epic, much concerned with leaders and nobles, politics and wars, nations or individuals in lengthy conflicts, often describing sequences of events that span more than one generation of human experience (besides Aeschylus, note in tragedy the handling of past time in Euripides’ Phoenissae and Orestes). But their relationship is not clear-cut. If we may extend the family-tree metaphor, they may seem to have only one parent in common, epic, married by turns to lyric poetry on the one hand, geography or genealogical catalogue on the other. There are obvious problems in identifying the links between history and tragedy more precisely. Sometimes we may prefer merely to speak of affinities between the genres, rather than direct influence. It is naturally difficult to define a relationship between genres which are themselves varied and evolving. As with any comparison, we must guard against the temptation to simplify one pole of the opposition: thus the ‘‘typical’’ tragic plot cannot be reduced to some simplistic formula such as ‘‘pride goes before a fall,’’ or ‘‘great man brought low.’’ Chronology and sociopolitical circumstances matter: tragedy was far more important, and more formative of the imaginative world of the citizens, in fifth-century Athens than in first-century BCE Rome, but even in the time of Cicero the theater might provide role-models or apt citations (Cic. Sest. 115-126; Att. 2.19.3,15.11.3, etc.; see further Griffin 1985: ch. 10). It is in the Athenian context that we may be tempted to see the closest affinity between the genres. Athenian politics and values are prominent in both forms. Herodotus is said to have known Sophocles and was evidently read by him; Thucydides and Euripides are not linked by any such biographical data, but both were Athenian citizens, and they share many preoccupations. By contrast in later periods, when tragedy meant primarily the classic Athenian dramas of earlier times while history was generally composed elsewhere and on non-Athenian themes, the gestures of historiography towards tragedy almost inevitably become more allusive, a kind of homage paid in passing, not an indication of an integral or fundamental resemblance between the genres.
It would be convenient if we had some guidance from ancient theoretical or critical writing, but notoriously there is no full-scale treatise on the writing of history in antiquity before the pamphlet by the satirist Lucian. Many passing references exist, not all of them well understood: a few will be mentioned in due course (an example of a passage much cited but misused is Duris, FGrHist 76 F 1, on the absence of mimesis in earlier writers such as Ephorus; see the reinterpretation by Gray 1987). In particular the polemics of Polybius against his rivals and predecessors have sometimes been given much more weight than they should be: his caustic references to the ‘‘tragical’’ writing of Phylarchus and others cannot support the reconstruction of an actual school of‘‘tragic history’’ (rightly Walbank 1960, cf. 1962; cf. Meister 1975).
Aristotle famously juxtaposes ‘‘poetry’’ (which in this context means primarily tragedy, as the highest poetic form) and ‘‘history’’ (Poet. 9.1451b4-7): ‘‘the difference lies in this, that the historian writes of what has happened, and the poet of what might happen. For this reason poetry is more philosophic and more serious than history; for poetry speaks more of the general, history of the particular.’’ The passage has been much discussed. It evidently implies that the plots and narratives of poetry exclude the clutter of detail and incidentals which complicate those of history; but we may note that this is described as a tendency (‘‘speaks more of...’’), for of course poetry cannot do without all particulars, and the historian must select, sift, and shape his subject matter in order to provide any kind of analysis and generalization. Insofar as he does so he is acting as a poet might; insofar as he does so with a view to creating dramatic and emotionally intense effects, he may be said to resemble the tragedian in some degree. This tendency was detected by ancient readers in several of the ancient historians, above all in the work of Thucydides (e. g., D. Hal. Thuc. 15).
Before examining the tragic elements in history further, I consider more briefly the other side of the comparison, since tragedy did on occasion deal with ‘‘what has happened.’’ Historical dramas are attested, chiefly from the very early period and from the post-classical age, but only the Persians of Aeschylus survives (472 bce; on its relation to Herodotus, see, e. g., Pelling 1997a: 2, 3-8: most probably the historian knew the play, but it does not seem to have shaped his own treatment significantly). It is striking that this drama, while in one sense deeply political, engages with the recent events only at arm’s length: the scene is set in the Persian court, the exaltation of the Greek victory is handled indirectly, through the eyes of the enemy, and no individual Greek is named. Aeschylus’ predecessor Phrynichus dramatized a defeat, not a victory, in his Sack of Miletus, and was penalized by the Athenians ‘‘for reminding them of their own misfortunes’’ (Hdt. 6.21): tragedy required greater distancing to have its proper effect. Other evidence suggests that later Greek historical dramas tended to be set in a distant time or place or both (Hall 1996: 7-10; Bowie 1997: 40-45). The famous Gyges-drama (probably of Hellenistic date) is a case in point (TrGF Adesp. 664). There is a significant contrast here with Roman dramas on contemporary themes, which seem sometimes to have been put on with openly encomiastic purpose; on the other hand, even at Rome these seem to have been relatively rare events (Flower 1995).
Most tragedy dealt with mythical characters in the heroic age (Easterling 1997): distancing is achieved, even if the characters were regarded as in some sense historical
(on the ambiguities here see Veyne 1983). Historical influence is not thereby excluded, and is very clearly present in some plays, especially Aeschylus’ Eumenides (e. g., Dodds 1960b; Sommerstein 1996: ch. 12). The relation of the various dramas to their times is much disputed: perhaps the only approaches we can firmly rule out are the search for explicit reference to the present and for political allegories (Zuntz 1955 is the classic refutation). Certainly the tragedies are rich in anachronisms and aetiologies: they often prefigure or predict institutions, cultic practices, and dilemmas of the poet’s own time. Athenian ideology is reflected in the plays, though not uncritically. Hostile references to Sparta, discussions of the merits of forms of government, suspicious comments on clever rhetoric or corrupt politicians, doubts about the value of divination, are among the motifs with obvious contemporary relevance. The orators of the fourth century could cite speeches from the dramas as inspiration for patriotic citizens (Lycurg. 1.100, quoting Eur. Erechtheus F 360; cf. Wilson 1996). In some cases, we may feel, their citations simplify or deny the tragic complexities of the original (Dem. 19.246-248, citing Soph. Ant. 175-190). Yet we need not doubt that many of the audience recognized some at least of the episodes on stage as more violent and passionate versions of events experienced or imaginable in their own time: whether tragedy encouraged them to face private and public disaster with greater resolution or merely taught them to indulge their emotions is a question already raised by Plato, and the debate remains unresolved today (on the civic aspect of the dramas see further, e. g., Meier 1993; Cartledge 1997a; Griffin 1998, with the responses by Seaford 2000; Goldhill 2000; Rhodes 2003).
Tragedy was only one of a number of genres which tackled themes that we would call historical in verse. In eschewing too close an engagement with the contemporary world it followed the precedent of Homeric epic, but other early poetry, both hexameter and elegiac, had been less scrupulous (West 2003: esp. 25-31, 220-285; Bowie 1986, supplemented by Bowie 2001). Poetic treatments of genealogy, civic origins, travel, and geography preceded prose. Historiography proper, however defined, only emerges when such themes are transferred to narrative prose. The questions surrounding the early emergence of historical writing are addressed elsewhere in this collection (see also esp. Fowler 1996; Marincola 1999). Here we need only note that the debt of early history writing to tragedy cannot be viewed in isolation: it is part of the genre’s debt to poetry in general, especially epic. Indeed, the paradox is that while Herodotus in some ways seems most closely akin to tragedy in outlook and narrative patterns (e. g., his concern with the ironic fulfillment of oracles), there is every reason to suppose that his world view was fully formed before he ever saw a tragedy performed in the Athenian theater. (Cf. Lloyd-Jones 1971: 66 and 142; contra, Chiasson 2003, who sees the Lydian logos as a kind of contestation with Attic tragedy.) Herodotus influenced Sophocles (West 1999), but it is less clear that the reverse was true. We may be dealing more with a fondness for certain story-patterns and types of plot that transcend genres and that have their roots in epic or (perhaps) folk-tale.
This cautionary note leads on to some further reservations. ‘‘Tragic history’’ has become a cliche, but what is it meant to convey? What kind of influence or cross-fertilization is envisaged, and how far is it formative in any specific text or part of a text? In what follows I suggest a framework for further discussion. While my general argument is skeptical, I do not wish to suggest that the terminology is never appropriate, and my comments above about the complexity and range of the material should be borne in mind. The comparison itself, even if our verdict on influences may be negative, can serve to bring out some significant features of the historiography of antiquity.
1. Is history adopting the form of tragedy? This would be hard to maintain on any literal level - tragedy is verse, and, in its lyrics especially, highly elaborate, selfconsciously poetic verse; history is prose. The two genres are also strongly contrasted in scale or amplitude. Tragedies are typically compressed (this is one way in which they achieve their intensity): the longest is much less than 2,000 lines. No ancient drama approaches the length of Hamlet. History by contrast is expansive, even diffuse, and prone to digressions. Polybius needed forty books, Dionysius twenty, Livy 142, Nicolaus of Damascus 144, Ammianus thirty-one. It goes without saying that massive works such as these must embrace many sequences of action: if we are to speak of tragic form, it will be a matter of particular episodes rather than an overarching structure. Perhaps shorter historical monographs, particularly if focused on a single individual, might be more amenable to analysis utilizing the terminology of tragedy - a point adumbrated in the well-known comments in Cicero’s letter to Lucceius, in which he emphasizes the dramatic potential of a briefer work concentrating on the fortunes of one distinguished participant in the historical process (Fam. 5.12.2; Rudd 1992). This also suggests that biography may be more amenable to ‘‘tragic’’ coloring than historiography proper.
More specific formal similarities can be found, but they are not all that frequent. We can make little of the fact that both genres make extensive use of speeches: on the one hand this follows the precedent of epic, on the other it shows the influence of the rhetorical schools (Walbank 1965). Nevertheless, the agonistic use of speeches which we find in both genres suggests an important affinity: audiences and readers are encouraged to contemplate both sides. Insofar as the audience enjoys a superior position, discerning the weaknesses, distortions, or self-deception in a speech, there is scope for irony of a dramatic, sometimes a tragic variety. (Cf., e. g., Macleod 1983: 154-156, comparing the exchange between Hecabe and Odysseus in Euripides’ Hecabe with the Plataean debate in Thucydides.) More important still, both genres have the potential to be non-partisan (though this is achieved to widely varying degrees); both make us see the complexity of perspectives and issues, forcing us to judge and to think.
Other formal analogies might not be easy to identify conclusively. One would have to consider more limited approximations. One approach would be to consider the presence of poetic (or specifically tragic?) vocabulary in historical texts: here again, however, the most conspicuous debt is probably to epic (for sample studies see Avery 1979; Chiasson 1982 on Herodotus; Gries 1949a on Livy; Miller 1961-1962 on Tacitus and Vergil; various skeptical pronouncements in Goodyear 1972: 31, 325; 1981: 108-109; 1992: 234-235). There are many problems here, of course: not only the difficulties of identifying vocabulary as markedly poetic (and still more genre-specific), but the complexity of judging the effect of such language in a new context. Emotional effect in prose is often achieved by means other than diction, e. g., through imaginative use of metaphor, which is not confined to prose or verse.
Another, perhaps more reliable approach might be the deployment of certain typical scenes or effective formal devices with a tragic resonance. The best example is perhaps the tragic messenger. A famous case is Thuc. 3.113, in which a herald arrives to ask for the return of the Ambraciot dead. Here the messenger is caught unawares, since the situation is even worse than he had thought: we catch a glimpse of the tragic technique most powerfully represented in the Oedipus of Sophocles, whereby reversal coincides with recognition (Stahl 1966: 133-135 = 2003: 133-136). Closer still is the scene in Ctesias (F 24 Lenfant), in which a messenger releases the bad news of Cyrus’ death gradually to the horrified Persian queen.
A final formal aspect is of fundamental importance. Tragedy is drama, set before us without a mediating voice. History is narrated, and the authorial voice, though varying greatly in manner and authority in different works, is a constant of the genre. Even when the historian was himself a participant in the events, he writes after a significant interval of time, and otherwise distances himself from them (often through the ‘‘objectivizing’’ device of third-person narration: Marincola 1997: 183ff.). It is true that tragedy, especially in the choral odes and (on occasion) in messenger speeches, draws back from the action and contemplates the events from a longer perspective, but the general tendency of tragedy is to involve the audience in a far more immediate way, intensely and protractedly (cf. 3 below). The narrative process of historiography does not exclude such effects at crucial moments, but it militates against them; and many of the historians clearly regard this distance from the subject as a crucial part of their self-presentation. Thucydides mentions his exile, but does not launch upon self-justification. In summary, tragedy engages the emotions of the audience by direct enactment; history sometimes does this, but the episodes in which this happens are framed by the stabilizing narrative voice of the historian, who guides the reader and suggests evaluations and explanations much more frequently and explicitly than is possible in drama. The contrast is a crucial one, but it must be added that it is not total. Some historians do this, some of the time: but authoritative summing-up, telling the reader what to think, is a device less frequently used than we might expect; and it is striking that these ‘‘final’’ statements often seem to be qualified or contradicted elsewhere (the problems involved in Thuc. 2.65 or Tac. Ann. 6.51 come to mind). It is too simple to say that tragedy shows, history tells: history too has a strong tendency to prefer dramatic enactment to summary analysis. There is an affinity here which again we may attribute to the epic tradition. (On various aspects of the didacticism of historiography, see Rutherford 1994.)
2. Does history deal with the characteristic subject matter of tragedy? This seems much more promising. Not only does history sometimes discuss events of the mythical age, it also deals with historical events of comparable scale and intensity - wars and empires, family feuds, expeditions which end in disaster or the deaths of leaders. On closer examination, however, doubts arise. Even in the earliest authors we can discern a determined tendency to scale down the exaggerations and impossibilities of myth (e. g., Hecataeus on Cerberus, FGrHist 1 F 27: cf. generally Wardman 1960; with different emphases: Wiseman 1979: ch. 9; Marincola 1997: 117-127). Herodotus notoriously took an unorthodox line on Helen’s presence or absence at Troy (2.120, drawing on but rationalizing Stesichorus): Priam and the others would never have been so insane as to fight a war for Helen if they had been in a position to surrender her. (Contrast Euripides’ Helen, also drawing on Stesichorus but making the story more fantastical.) This detached and critical attitude to the legends is carried still further by Thucydides, notably in the archaeologia (1.1-19). It is not Agamemnon the sacrificer of Iphigenia or the victim of Clytemnestra that interests him, but the overlord who could assemble a massive force because of the greater power at his command (1.9). Thucydides’ exclusion of the ‘‘story-telling element’’ from history was influential (1.22.4), though not absolute even in his own case (2.29, 102, with Hornblower, CT ad locc. ). Even though later historians reinstated many mythical or effectively mythical tales, they did so with reservations: mythology was suited to provide pleasant digressions, no more, said the hard-headed Polybius (38.6). Ephorus passed over the muthologiai and started his work with the return of the Heraclidae (Diod. 4.1.3). Livy admitted that the material on the founding of Rome was closer to poetic fancy than to uncontaminated chronicle (praef. 6), but he still sought in the early books to distinguish the more reliable versions (e. g., 1.16). Plutarch’s qualifications and adjustments in his lives of Theseus and Romulus similarly show a marked awareness of the problems of dealing with this very dubious material (esp. Thes. 1; Pelling 1999b). Reservations are especially evident when the historians are concerned with tales which traditionally involved divine intervention, an area where they wish to be seen as exercising epistemological caution. More relevantly for our theme, when they actually engage with the matter of classical tragedy the treatment is often notably un-tragic: thus we may contrast the flatness of Thuc. 2.29 with what we can surmise of Sophocles’ Tereus (other cases include Diod. 4.62 [a flat summary of the story of Hippolytus], 64-65 [Oedipus and family], D. Hal. AR 1.46-47 [the taking of Troy]). The opening pages of Herodotus (1.1-4) illustrate both these tendencies: avoidance of explicit reference to divine actions, and a certain brisk, summary, even occasionally flippant handling of older and more questionable material.
We may find it more profitable to contemplate the passages in which the prose writers deal with catastrophic events of historical times and rise to the manner of tragedy. There are undeniably places in which the agonies of an individual or the downfall of an army or city are described in powerful and moving language: the finale of the Sicilian expedition leaps to mind (esp. Thuc. 7.71 and 75). As Dionysius comments, ‘‘every man’s heart is carried away by such language’’ (Thuc. 27; cf. Brunt 1993: 189, 205-206). Many other passages could be cited (e. g., Hdt. 1.45; Livy 30.12-15; Tac. Hist. 3.70-72). It is relevant that the ancient critics laid heavy emphasis on the quality of enargeia, vividness (e. g. [Long.] Subl. 15.1; 26.2; Quint. 6.2.29; Walker 1993). A notable passage on this theme is Plutarch’s comment (deglor. Ath. 347A): ‘‘Thucydides is always striving for this vividness in his writing, since it is his desire to make the reader a spectator, as it were, and to produce vividly in the minds of those who peruse his narrative the emotions of amazement and consternation which were experienced by those who beheld them’’ (he goes on to cite Thuc. 4.10-12, 7.71). Thucydides himself had given the lead with the reference to the anxious army on shore, an audience watching the sea-battle in process (7.71 again; cf., e. g., Sall. Jug. 60; Tac. Hist. 1.32.1; 3.83), and many modern studies have taken further the notion of the gaze, the onlooker, the spectator within the text (e. g., Dewald 1999 on Herodotus and Thucydides; Davidson 1991 on Polybius; Feldherr 1999 on Livy). No doubt later authors often did think in terms of dramatic performances, but a caveat is in order, for Homer already set this pattern unforgettably in Book 22 of the Iliad, in which the Trojans (and indeed the Greeks) watch the pursuit of Hector by Achilles and observe the humiliation of Troy’s champion, while on a higher plane a different audience, more detached but still passionate, also witness the events: ‘‘all the gods were looking on’’ (22.166; Griffin 1978). It is evident that enargeia alone cannot guarantee a diagnosis of ‘‘tragic influence.’’ There is a danger that ‘‘tragic’’ may be diluted to mean little more than ‘‘dramatic.’’ (A similar conflation of ideas underlies the suggestive paper ofWiseman 1994: 1-22, who sees Roman drama as strongly influential on Roman historiography: to my mind his argument rests too heavily on passages which compare episodes to theatrical scenes. For further discussion Flower 1995: esp. 172-174, and Zwierlein 2003 may be consulted.)
Marincola in a valuable discussion (2001: 69-73) has suggested that the affinities between history and tragedy should be seen in terms of plot-patterns, especially those involving advance preparation, repetition with difference, ironic and unexpected reversal (peripeteia), and so forth. The difficulty is that again this opens the door to wider comparisons. Plato, Aristotle, and others observed that Homer was the pathfinder of tragedy, not only in the intensity of suffering he portrays and the highly dramatic manner in which he does so, but in the construction of plots which combine reversal and recognition (Rutherford 1982; Herington 1985: 213-215). The fall of a city, that quintessentially tragic theme, while not enacted in the Iliad, is constantly foreshadowed. (On the theme see Paul 1982. See esp. Livy 1.29; Pol. 2.56 on the excesses of Phylarchus; also Pol. 39.5, with Hornblower 1981: 104-105.) It is in fact difficult to isolate a theme common to tragedy and history which is not to some degree present in Homeric epic (cf. Macleod 1983: 157-158).
3. Is history seeking to arouse the same emotions as tragedy? (Cf. esp. Marincola 2003 on ‘‘the emotions of history,’’ with much earlier bibliography.) Much that falls under this heading has already been anticipated. Recent study has made much more explicit that ancient historians wrote rhetorically, in order to intrigue, astound, excite, distress, and persuade their audiences (Wiseman 1979, etc.; how far this involves falsification or exaggeration of the facts they include is a separate inquiry). A part of that rhetorical process is the emphasis on the greatness of their subject, and often this involves emphasis on the magnitude of suffering which that subject entails: Thucydides’ statement of intent of course combines both (1.23; cf. Tac. Hist. 1.2-3), and Herodotus had already made reference to the misfortunes which resulted from many of the events he narrated (e. g., 5.97.3; 6.98; 8.20). Naturally the historians seek to make their readers feel the impact of these moments of disaster and reversal, and to this extent there is a recurring need to arouse pity and fear (but not only these: horror, wonder, and admiration are among the other reactions sought). Moments of desperation are marked out (Pol. 3.118), and anxious deliberation made explicit.
In highlighting the emotions of the participants, the historians focus not only on the leaders but also on the larger community involved: the interest in mass morale of an army or a population is a typical feature of historiography. We note also a willingness in many of the writers to concentrate on extremes (Grant 1974; Macleod 1983: 140-141) and excesses, as in vicious forms of revenge (Hdt. 1.119, 212-214; 8.105-106), terrible deaths (Hdt. 4.205; Jos. BJ 5.512-519), crimes involving family members (e. g., Tac. Hist. 3.25, 51), or horrific mutilation (Hdt. 6.75). We may well feel that mythological tales have influenced the historical imagination in some of these cases (the parallel between Harpagus and Thyestes has long been noticed); and certainly the willingness to make the audience’s flesh creep may owe something to tragedy. This seems even more likely in Rome, where our limited evidence suggests that drama was both more grandiloquent and more prone to the macabre (Holford-Strevens 1999; see now Ferri 2004 on the pseudo-Senecan drama Octavia).
It seems important, however, that even in set-piece scenes of traumatic experience, classical historians do not go the full lengths of tragedy. Extended lamentation, protracted focus on the suffering of named individuals, prolonged dramatization of the reactions to loss and disaster, are not typical of the historical genre - yet they are inseparable from tragedy. The description of the event may be powerfully described (the killing spree at Corcyra, Thuc. 3.79-81), but the aftermath is seldom dwelt on. Thucydides gives us both his description of the plague’s symptoms and his sociological analysis of resultant moral breakdown, but he does not mention, still less narrate, the particular details of Pericles’ bereavement (contrast the more personal account in Plut. Per. 36-37). Moreover, such set-pieces of narrated disaster may be among the highlights of historical reportage, but they are hardly typical of it. Many pages in all of the historians go by without evoking our emotions to anything like this degree. I do not dispute that some passages of ‘‘quieter’’ emotional quality may subtly prepare for or contrast with moments of higher intensity still to come, or that ‘‘dead-pan’’ sentences may pack a powerful moral and emotional punch (e. g., the implicit condemnation of the Athenians’ inertia in Thuc. 3.68.5). But the overall effect of a historical work is, I would insist, different in quite fundamental ways from that of a tragic drama.
This point need not, of course, be presented only in negative terms. Tragedy helps bring the characteristic detachment of history into relief. Again the historian’s narrating voice is an important factor. In tragedy, all those on stage, including the chorus, are affected by the events (the gods, normally appearing above the stage proper, are perhaps exceptional, though Artemis does feel grief at Hippolytus’ death). In history the narrator evaluates and makes a considered judgment: the baffled aporia of Xenophon at the end of the Hellenica (7.5.27) is quite exceptional. If emotion is aroused by the narrative, it is because the historian thinks this appropriate to the magnitude of his subject; but it is rare for him to be overwhelmed by it. Passages such as those in which Tacitus declares himself or his readers wearied by continuous horrors are introduced boldly and unconventionally: it would be naive to take them at face value (see Ann. 4.32, 6.7, 16.16; Hutchinson 1995: 61).
4. Do history and tragedy share some kind of intellectual attitude to their subject matter - if not a world view, at any rate some form of bleak or disillusioned pessimism, recognizing the costs of great achievement or the limited scope for human success? This view has been adumbrated by distinguished scholars (e. g., de Romilly 1977). This is perhaps the most difficult category of all, and the one in which most careful differentiation between authors would be necessary: clearly it will often be related to questions concerning the implied presence of some superhuman power influencing historical events, whether this is described in terms of‘‘god,’’ ‘‘the gods,’’ ‘‘fate,’’ or ‘‘necessity.’’ To impose on one genre an interpretive framework derived from the other (or from an abstract construction loosely based on the other) is a hazardous procedure: if we doubted this, the object lesson of Cornford’s attempt to read Thucydides in terms which he judged appropriate to Aeschylean tragedy should be sufficient warning (Cornford 1907: part 2, esp. ch. 13; for criticism see, e. g., Fisher 1992: 390-391). Even when the language of divinity is used, such expressions should not be read too literally in any of the historians: in some cases the gods seem to stand for ‘‘historical inevitability,’’ in others perhaps they merely add a note of dignified mystery. ‘‘Never surely did more terrible calamities of the Roman people, or evidence more conclusive, prove that the gods take no thought for our happiness, but only for our punishment,’’ writes Tacitus (Hist. 1.3: nec enim umquam atrocioribus populi Romani cladibus magisve iustis indiciis adprobatum est non esse curae deis securitatem nostram, esse ultionem). Elsewhere he ascribes the ascendancy of Sejanus to ‘‘the wrath of the gods against the Roman state’’ ( Ann. 4.1: deum ira in rem Romanam), but such pronouncements are hardly typical of the author or the genre. (‘‘A striking and ominous phrase, but no confession of a creed,’’ opined Syme 1958a: 521. For the complexities of tone in Tacitus see the discussion by Hutchinson 1995: 241-250.) Some passages do indeed suggest something of what we might call a tragic view of life: at the very outset of his work Herodotus comments that he is well aware that human fortune never remains in the same place: great cities have become small, and small cities great (1.5; cf. very differently Amm. Marc. 14.11.25-26). The same ‘‘moral,’’ if that is the right word, seems implicit in the passage late in Polybius’ work (39.5), in which Scipio watches Carthage burn and, quoting Homer (again, not tragedy!), fears for the future fate of Rome. (In general, however, the strongly patriotic strain in Roman historiography, while still allowing pessimism, is at least discouraging to a truly tragic vision.) The question warrants further investigation: here it is enough to observe that history and tragedy both partake of a broader pessimistic outlook which seems altogether characteristic of Greek thought (Burckhardt 1998: 85-124, remains an outstanding survey).
5. In view of the points made above, we can perhaps find more solid ground in a more limited inquiry. Are there cases where history deliberately evokes and explicitly refers to tragedy, even using examples from the classic plays to provide a parallel or precedent for the historian’s own subject? Undoubtedly this is so. Walbank in a well-known paper examined Polybius’ treatment of the last years of Philip V, a highly colored account elaborated further by Livy, and concluded that the ‘‘tragic version’’ was the historian’s own construction (Pol. 23.10-11, supplemented by Livy 40.3-16; Walbank 1938; note 23.11.1ff. for references to tragic exempla). Another famous instance, stHl more explicit, is Livy’s introduction to the tale of Tarquinius the elder: ‘‘for the royal house of Rome too brought forth a paradigm of tragic criminality’’ (1.46.2: tulit enim et Romana regia sceleris tragici exemplum). There follows a grim narrative of fraternal conflict, tyrannical rule, and monstrous female brutality (note especially the reference to Furies in 1.48.7). It is in general striking that Roman authors seem to associate tragedy with monstrous crimes, especially within the family (Hor. Ars 89-91, 185-186; Pers. 1.17f.; Juv. 6.628-637; Hutchinson 1995: 63 n. 47). This has some relevance to the historiography of Tacitus, and especially to the culmination of the Annals in the Neronian books (matricide, fratricide, poisoning, incest, wife-killing all figure in these dark books).
Plutarch in his biographies is particularly fond of the trope of associating his narrative with the tragic stage. In a highly dramatic scene he describes Demosthenes taking poison: the orator’s last words are an admonition to the Macedonian Archias to ‘‘waste no time in playing the role of Creon from the tragedy, and cast this body out without burial’’ (Dem. 29.6, anticipated by 29.2). Sustained play on imagery of the comic and tragic theater guides the reader’s response to the shifting tone of the pair Demetrius and Antony (de Lacy 1952; Pelling 1988: 21-22). Above all there is the Crassus, in which the hapless triumvir, vanquished at Carrhae, is decapitated and his head used as a gruesome prop in the production of the Bacchae at the Parthian court. The grotesque mingling of reality and drama in this scene is worthy of Euripides himself (Crass. 33, concluding ‘‘and in such a final exit did Crassus’ generalship find its conclusion, in the very manner of a tragedy’’). (On this life see further Braund 1993.)
Again, however, we must note that historians are not unique in using tragedy in this way. Horace in a powerful ode celebrates the imminent publication of Pollio’s history of the civil wars in language which suggests that the distinguished author’s gifts as a tragic poet may be transferable to the medium of history ( Carm. 2.1.9-12; Nisbet and Hubbard 1978: 9; Woodman 2003: 196-213). It is possible that Pollio had made some similar generic claim in his preface or early in his work. But the tragic note is also struck explicitly by orators (Andoc. 1.129; Antiph. 1.17; Dem. 21.149), epic poets (Verg. Aen. 4.469-473), even satirists (Juv. 6.634ff., 643, 655f.; 15.27-32). Historians were not the only writers to don the tragic buskin.
Walbank in a famous paper (1960 = 1985: 241) effectively refuted the proposition that a school of tragic historiography inspired by Peripatetic theories existed in Hellenistic times, and argued that insofar as tragic effects could be discerned in history, they were of a kind that went back to the early masters of the genre. He recommended that the term ‘‘tragic history’’ should be dropped, and others have echoed his doubts (Hornblower 1994b: 44). Experience suggests that once such phrases have become embedded in the discourse of academic debate they are hard to exorcise (witness ‘‘the heroic code,’’ ‘‘Euripidean melodrama,’’ ‘‘hubris and nemesis"). I only suggest that this one be used rather more self-consciously than it has sometimes been in the past. Affinities may be more suggestive than influences; specific scenes may be more profitably compared than whole works; the assumption that only bad historians ‘‘go tragic” needs to be firmly dismissed; above all, the different rhetorical and epistemological preoccupations of the two genres must always be borne in mind. Tragic history is not a self-standing genre or a phase in a genre's development: it is more like a particular color in an artist's palette, used in specific places for a particular effect.
FURTHER READING
Cornford 1907 is an interesting but perverse reading of Thucydides in ‘‘tragic” terms; more careful and valuable are the treatments by Stahl 1966; Macleod 1983: 140ff.; Hornblower 1994a: 111-120. Studies of Herodotus in these terms have been less successful, but see Stahl 1975; Riecks 1975; Chiasson 2003 (with much earlier bibliography); Griffin 2006. For later writers Walbank 1960 remains fundamental; cf. Walbank 1972: ch. 2. Brunt 1979 has many pertinent remarks and useful references. Woodman 1993 is a brilliant essay on theatricality in Tacitus on Nero’s court.
A wide-ranging paper on the importance of Homer for ancient historiography is Strasburger 1972; cf. his longer study of 1966 (Strasburger 1975) (with the reservations of Murray 1968).
I owe very valuable comments on a draft of this chapter to Professors Gregory Hutchinson and Christopher Pelling.
World History
