SOURCE: Carawan, Edwin. “The Tetralogies and Athenian Homicide Trials.” American Journal of Philology 114, no. 2 (1993): 235-70.

[In the following essay, Carawan presents an analysis comparing the methods of argument used in the Tetralogies attributed to Antiphon with court arguments he is actually known to have made.]

The Tetralogies attributed to Antiphon are documents of singular importance for the evolution of rhetoric at Athens: they give an artful demonstration of opposing arguments in three hypothetical cases of homicide; their methods are not unlike those of the showpieces of Gorgias or the paired speeches in Thucydides' History.¹ Basic questions of authorship and date remain unresolved, but it now appears to be the majority view that the Tetralogies can be at least identified as a product of Antiphon's era, if not his own work. The author's aims, however, remain uncertain: his work is often treated as a practical demonstration of model arguments for aspiring advocates to imitate (it is thus considered typical of early technai); but it may also be read as abstract speculation on prominent issues of the fifth-century Enlightenment. By either interpretation the artificial scenarios and theoretical aims account for the discrepancies of language and legal background that separate the Tetralogies from the other Antiphontian speeches which represent the actual arguments in homicide trials.² In response to this emerging consensus, it seems warranted to undertake a more thorough comparison of the common topics and techniques found in the Tetralogies and the relevant court speeches. Commentators have generally been content to remark certain similarities of theme, to acknowledge that the Tetralogies offer a condensed version and a different perspective on topics also found in the court speeches, and not to analyze the parallels and discrepancies systematically.³ Such is the chief purpose of this study. If we are to draw any valid conclusions about the author's aims and his place in early rhetoric, we should arrive at a better understanding of how his work compares with the one body of material that offers the clearest parallels in subject matter and method: the speeches for homicide trials. To this end we shall first consider the “questions at issue” in the Tetralogies alongside the court cases (with their peculiar mechanism for defining the dispute); second, the ethical argument that is premised upon miasma doctrine; and third, the problematic legal argument regarding “the law that prohibits (even) justifiable killing.”

I. INVENTION IN THE TETRALOGIES

A preoccupation with defining the question at issue is evident in each of the three Tetralogies; the second (in a case of accidental slaying in javelin practice) poses this problem most directly. Some matters, decided by the laws and decrees, are beyond dispute: “what is agreed upon as to the facts of the case, the laws and the voters have judged … ; if there is an issue in dispute (amphisbētēsimon), this is assigned to you [as judges] to decide”(2.2.1). The clear implication of this opening statement is (1) that the defense will have no basis to dispute the facts which are agreed upon (τὰ ὁμολογούμενα τῶν πραγμάτων) and (2) that the legal basis of the charge, “which the laws and the voters have judged,” is not an “issue in dispute … for [the judges] to decide.” The plaintiff has agreed that the act was unintentional, but he claims that defendant is nonetheless liable. “The facts of the case” are not disputed; neither the letter nor the intent of the law is at issue. The author thus suggests to his audience a basic division of issues as questions of fact, law, or definition.

Although the systematic division of issues—what would later be the province of stasis theory and what this author and Aristotle call “issue in dispute” (ἀμφιsβητηsιμόν, ἀμφιsβήτηsις)—appears to be largely a product of fourth-century rhetorical analysis (Aristotle Rhet. 1374a, 1417b), the author of the Tetralogies seems to assume that the rudiments of this classification are familiar to his audience.⁴ He may well have chosen dikē phonou as his format because it readily lends itself to experimentation on “issues in dispute.” In other procedures it remains doubtful how much of the case was decided in anakrisis; but in dikai phonou the pretrial hearings (over the course of three months) would ordinarily have given every opportunity to decide such basic questions as the facts of the case and the relevant law, and thus to define the “issue in dispute” before trial. In the court speeches this aspect of the proceedings is generally reflected in an apodeixis, or exposition of the opposing claims, at the outset of the proof.

In the opening speech of Tetralogy 2 the author seems to declare, in essence, that the procedural mechanism for defining the question at issue and the conventional apodeixis of courtroom argument are utterly ineffectual and irrelevant in this case. In the plaintiff's second speech, moreover, it is clear that the author intends to disregard or defy the ordinary rules and assumptions that governed the special courts for homicide. Once his opponent has spoken, the plaintiff reacts: “he has shown to what extremes dire necessity can drive anyone … I would not have supposed that he would (even) answer the charge” (2.3.2). The prosecutor contends that he has lost half of his allotted time by dispensing with the first speech when he thought that the trial was a mere formality.

This tactic corresponds to a familiar device in the court speeches (“surely my opponent will not have the audacity to dispute,” vel sim.).⁵ But in the court speeches this tactic draws upon what has been discovered in the pretrial hearings. Thus in Antiphon 6, where the apodeixis (6.15-17) seems clearly intended to make nonsense of the charges—as though the plaintiffs have no case—the defendant builds upon the preliminaries: he knows precisely the wording of the charge and how the key terms will be interpreted, and he is confident that the plaintiffs cannot dispute the testimony of his witnesses. In Tetralogy 2, however, there is the clear implication that the prosecution do not know how the defense will argue their case or even the wording of their plea. If this were an actual case for trial, the issues would have been clearly defined in the three preliminary hearings. The statements on both sides would have been attested under the solemnest of oaths (Dem. 23.67-68). Antiphon 1.2 refers to a written indictment; Antiphon 6.14-17 shows that the sworn statements were read out and affirmed at the beginning of the trial and that each side knew the content of the other's affidavit.

The author of the Tetralogies seems purposely to distort or defy this aspect of the proceedings. He treats the case as though the two parties have gone to trial without defining the real issue; the plaintiff pretends to have no knowledge of how the defense will plead—indeed he had supposed that his opponent would abandon the case in desperation and send his son into exile in the midst of the trial. In alluding to the provision that allows voluntary exile before the final speech the author is again treating the conventions in a provocative way. The provision for escape before the final speech is properly exploited in Tetralogies 1 and 3, where intentional killing is alleged and the death penalty looms. But this provision is irrelevant in cases of unintentional homicide such as this, where the defendant is guaranteed a safe path of exile after the final verdict. There was nothing to be gained by forgoing the trial.⁶ Unless we are to suppose that the author was wholly ignorant of this aspect of his chosen format, it is reasonable to conclude that he has intentionally seized upon a case in which the ordinary mechanism for defining the question at issue would prove inadequate.

The concise, elliptical approach to the scenario also suggests that the author has taken up a familiar paradox, one in which the “issue in dispute” might well seem elusive. The sole narrative, which must suffice for the four speeches, is the one brief sentence: “My son was struck by the javelin thrown by this youth in the gymnasium and immediately died.” The author evidently assumes that his audience will recognize the case—much as Gorgias could assume that his audience knew the stories of Helen and Palamedes. In this case Plutarch reports that such an incident was indeed publicized (Per. 36): we are told that Pericles and Protagoras debated whether the weapon itself or the officials were culpable.

If we assume that this case was well known to the audience of the Tetralogies, the initial formulation regarding the “issue in dispute” (“what is agreed upon as to the facts of the case, the law and the voters have judged; … if there is an issue in dispute it is [for the judges] to decide”) suggests that the author intended to discount the familiar solutions. Later in the tetralogy (3.7) the prosecution expressly discounts the view that Pericles supposedly debated, that the officials were culpable; the author sees no need to explore it. It was probably also a common assumption that the defendant in such a case might well attempt to plead that the killing was “justifiable” by the law of the Delphinium court. By the Delphinium law, however, the defendant may admit to the killing but claim legal justification only if he can invoke a specific provision of statute; among such instances were death in athletic competition (ἐν ἄθλοις)—not, as our author has construed the case, in training.

In the treatment of “issues in dispute” with which he begins this case, the author seems rather pointedly to reject a plea of legal justification. What “the law and the voters have judged” (issues which are not for the judges to decide) evidently refers to statutory rules, which, upon strict construction, would render a plea of legal justification invalid.

In all three tetralogies we find a similar allusiveness; the succinct, abbreviated narrative seems to presuppose a certain audience awareness of the issues in the famous or typical case. Tetralogy 2 is again the clearest instance, and what we learn of the controversy surrounding this case tells us something about the author's inspiration.

If we accept the historicity of the anecdote and assume that indeed Pericles and Protagoras debated this very theme, it is possible that the incident occurred in the 440s (or very soon after)—probably the era of Antiphon's earliest involvement as legal counsel. But if the debate between Pericles and Protagoras was first publicized or fabricated by Stesimbrotus, the terminus post quem must be rather later. Plutarch's account clearly suggests that Stesimbrotus was the ultimate source for the tale of Pericles' quarrel with his son Xanthippus; and it is in this context that we are told how Xanthippus exposed the debate with Protagoras.⁷ Plutarch himself is skeptical of Stesimbrotus' account, and if, as seems likely, the tale is a rank fabrication by Stesimbrotus, the story was current no earlier than the late 420s, probably later.⁸ On this assumption the Tetralogies would belong to the same period as Antiphon 5 and 6, with which they are so much at odds stylistically. They may yet be a work of the same author, utilizing the literary dialect and departing widely from his court practice; or it may perhaps be the work of a contemporary or “student” of Antiphon but, if so, not necessarily written before his death.⁹ And a third possibility is not out of the question, that the tale told by Stesimbrotus was elaborated by a later author, and it was this later publication that prompted the author of the Tetralogies to write an adaptation. In any case the tetralogy was clearly inspired by the cause célèbre.

In subject matter the other two tetralogies are similarly artificial controversiae: both may be linked to common topics in the rhetorical handbooks, and both seem to anticipate audience interest in certain sensational themes. Tetralogy 1—in which the defendant must argue that he is not likely to have killed the victim precisely because he is the most likely suspect—parallels a famous topic posed by Tisias; it may also recall a famous incident or a recent precedent.¹⁰ There are obvious parallels between the issue in Tetralogy 3—in which the defendant must plead that he was not the initiator of a brawl with an older man—and various common topics of the later rhetorical literature. The topic owes something to the fact that such brawls were undoubtedly commonplace in all periods; the scenario in Tetralogy 3 is also remarkably similar to a case reported by Demosthenes.¹¹

In his selection of material the author also reveals an interest in cases that involve fundamentally legal or juristic problems, areas where the rules and principles that shape the courtroom arguments prove inadequate or suggest problematic consequences. Thus in Tetralogy 2 the author has focused upon an incident where the accepted notions of causation and legal justification involve obvious contradictions or ambiguities which seem to defy the ordinary solutions and the very assumptions underlying dikai phonou. The hypothetical “who-done-it” in Tetralogy 1 has similar legalistic implications.

Now there has been a general tendency to emphasize the importance of argument from probability in the Tetralogies and to date them accordingly to the height of the sophistic era, when such tactics were new and intriguing. The author's method, however, is not such as to suggest that probabilities per se were among his chief interests; his use of eikos is in fact rather limited and narrowly focused. He uses probabilities only in the first and third tetralogies; and in the third, only in regard to the commonplace “The young are more prone to reckless violence.” Tetralogy 1, to be sure, involves a tour de force in ethical reasoning on probable motive; but this exercise in probabilities is not by any means the whole of the argument. Whether it is sufficient to justify the usual assumption, that the author's chief aim was to give a demonstration in probabilities, seems to me questionable.

The use of probabilities in Tetralogy 1 is directed to the question of fact, whether the defendant is indeed the unknown agent who ambushed the victim. This line of argument is clearly a speculative exercise; but it is not without legal or juristic significance. From all available evidence, both the extant speeches for court cases and the record of statute law, the question whether the defendant or some other unknown assailant was in fact the culprit was rarely (if ever) the issue in dikai phonou. The homicide procedure that Dracon devised was intended primarily to facilitate private settlement, providing for exile and aidesis or for self-help; trial before the ephetai was invoked only where this primitive form of “judgment” was disputed by one party or the other.¹² The function of the Draconian trial was not to investigate or make a determination upon the facts, but to give judgment regarding liability, either for the plaintiffs or for the defendant. From these origins, the argumentation in homicide trials tends to focus not upon questions of fact or circumstance per se, but upon the formal instruments of proof—oath, witness, testimony under torture, and other documents. These instruments originally served as the decisive means of resolving the case and later became the focus of preliminary hearings where the issues of fact and law were decided, the plaintiffs established a prima facie case, and the defendant made his plea.

More than sixty years ago Solmsen suggested that Antiphon stands at the transition from the earlier mode of reasoning from formal instruments of proof to the era when eikos and circumstantial evidence became prevalent; he observed that “rational argumentation” in Antiphon's court speeches tends to center upon the atechnoi pisteis (oath, witness, etc.), which thus seem to be still the decisive “barycenters” of the argument—“Gravitationszentren.”¹³ This model of Antiphon's argumentation has been recently challenged, and some useful clarifications have been made; the notion that Antiphon was somehow conceptually constrained by the older mode of argument is ultimately misleading, especially in regard to Antiphon 5.¹⁴ But in the arrangement of Antiphon 1 and 6, the two court speeches for dikai phonou, Solmsen's observation is strikingly apt: the arguments are clearly structured around the validation of oaths, testimony, and challenges to torture. In Antiphon 1 the speaker's principal argument is directed against the defense oath—“How can he swear that he ‘knows full well’ the defendant is innocent,” when he was neither witness to the events himself nor willing to torture the slaves who have knowledge of the events? (There are secondary arguments regarding challenge to torture.) Again in Antiphon 6 the speaker addresses himself to the wording of the diōmosia and the value of testimony; rather than argue directly on the factual issues and the crucial concept of indirect causation (bouleusis), he fixes upon the self-refuting claims and actions of the prosecution. Thus in the extant court speeches in dikai phonou Solmsen's finding largely holds true, that arguments from probability and circumstantial evidence tend to center upon the “nonartificial” or evidentiary proofs. This is true of Lysias 1, however, as well as Antiphon 1 and 6; whereas in Lysias 13, just as in Antiphon 5, in homicide cases prosecuted by “warrant and arrest” before ordinary dicastic juries, rather different methods prevail. The reason for this shift may not be as Solmsen supposed, that Antiphon's mode of argument was in transition, but that the traditional instruments of proof retained a special function in dikai phonou, and therefore circumstantial evidence and other artful tactics tended to focus upon oath and testimony, rather than directly upon questions of fact or motive.

The author of Tetralogy 1 is in fact remarkably attentive to this aspect of the proceedings. He seems to assume that the plaintiff must rely on some, at least minimal, evidentiary proof (in this case the dying testimony of a slave). Bear in mind that if he had not wished to be at all constrained by court practice, he might have ventured upon a purely circumstantial case (dispensing with the dying slave). Instead he adapts the conventional arguments on testimony under torture: where the prosecutor insists upon the validity of the slave's testimony, the defendant discredits it by invidious comparison to freemen's testimony, which is subject to penalties for perjury.¹⁵ But the author has skewed his scenario in such a way as to subvert the ordinary rules of evidence. Ordinarily the plaintiff's oath would rely upon firsthand knowledge of the events or eyewitness testimony. In this case the sole basis for the plaintiff's oath is the reported testimony of the dying slave. Hearsay of living witnesses was not allowed, and the testimony of slaves was ordinarily admissible only if properly obtained under torture. Our author has, in effect, acknowledged but evaded both restrictions; his case is technically admissible, but he could not have contrived more tenuous grounds for the charge.

So far as we can judge, circumstantial evidence alone was not felt to be sufficient for bringing a case to trial (though once at trial the “probabilities” were often decisive). Where there were no qualified witnesses or other evidentiary proof upon which to base their charges, the accepted course of action for the victim's family would be to make their proclamation at the Prytaneion against the unknown killer.¹⁶ That proclamation of the unknown killer was sometimes the only recourse is shown by Dem. 47.69-70: the exegetes told the plaintiff, “since you yourself were not present, but only your wife and children, and there are no other witnesses, make proclamation against no one by name but against ‘the perpetrators and killers.’” For trial before the ordinary homicide courts, prosecutors would have to swear, either on their own conviction or on the basis of other firsthand witnesses, that so-and-so was “the killer” (directly implicated in cause of death). Tetralogy 1 goes as far as the procedure will allow to dispense with these traditional rules of evidence.

The issue in Tetralogy 1 has, indeed, its parallels among the court speeches. In both Antiphon 1 and 5 the charges amount to premeditated murder. Antiphon 1, Against the Stepmother, is the only extant speech prosecuted on the same charge in the same jurisdiction as envisioned in Tetralogy 1, namely intentional homicide before the Areopagus.¹⁷ The question of fact, however, whether the defendant was indeed implicated, does not appear to be the principal issue. It is evident from the argumentation that the question of intent was crucial to the defense, and here the prosecution has only dubious evidence.¹⁸ The plaintiff is prosecuting in obedience to his father's dying charge, and he claims that his father had actually caught the woman attempting to give him drugs, though she claimed to have given them as love potions (1.9). The slaves' testimony under torture would supposedly have confirmed this incriminating fact.¹⁹ Aristotle alludes to a similar case in Magna Moralia 1188b29-38—indeed it is very possibly the same case (perhaps as notorious as the scenario in Tetralogy 2). Here intent is clearly the issue, and we learn that the woman was acquitted on grounds that she acted from love rather than malice. In any event, in Antiphon 1 the question of fact, whether the stepmother was implicated, does not appear to be the issue.

Other than Tetralogy 1 we have no direct evidence that the question of fact, whether the defendant was or was not implicated, was ever at issue in dikai phonou. Antiphon 5 (Herodes) is indeed a case of “who-done-it” similar in some respects to that of the Tetralogy 1 but prosecuted by the procedure for “warrant” and summary arrest (endeixis/apagōgē). Here the prosecution rely upon the testimony of a slave under torture (evidently handled improperly and discredited by the slave's dying recantation). The defendant suggests that he should be tried by dikē phonou, presumably before the Areopagus;²⁰ but the treatment of the legal issue is highly argumentative and certainly not reliable evidence that such cases were regularly tried at the Areopagus. Indeed, it is a reasonable inference from the speaker's argument (and one his judges were clearly intended to make) that the plaintiff's case would be untenable before the high court.

The very problem of how to convict a murderer who plots to evade detection is, to be sure, a recognizable topos in the extant court speeches, but it serves a different function: by the very nature of the proceedings it was almost impossible to convict a murderer without witnesses; that fact became axiomatic. We find this topic developed in Antiphon 6 (Choreutes): “in cases where murder is planned and executed without witnesses, judgment must be rendered solely on the basis of such claims as the prosecution and the defendant put forward … rather upon conjecture (εἰaάzοντας) than sure knowledge of the facts” (6.18). In this case there are many witnesses and no dispute on the facts; therefore the judges should put little credence in the artful conjectures of the prosecutors. The chief function of the topic is to emphasize by contrast the certainty of the defendant's proofs. When we turn from this usage to the proem of Tetralogy 1, it looks as though the author has chosen his scenario precisely to demonstrate how the paradigmatic “insoluble case” may in fact be solved. He has reversed the ordinary assumptions: assume there are no reliable witnesses, the bare minimum of formal proof for admitting the charges. He extends the concept to its logical though unconventional conclusion—to try the case solely upon probable motive and prejudicial appeals. In this way the topic in the court speeches may have served as a model for Tetralogy 1.²¹ The reversal of this argument such as we find in that tetralogy, on the other hand, would not have been much use in the homicide courts; and it has, in fact, left no mark upon the extant court speeches.

In Tetralogy 1 we are told that crimes done on the spur of the moment are not difficult to prove; but where the perpetrators anticipate the danger and take every precaution to avoid suspicion, they are difficult even to discover, let alone convict (1.1.1-2). Upon this principle the author urges the judges to rely upon the probabilities; he proceeds to incriminate the defendant by showing that no chance assailant (for theft or other common motive) was likely to have done the killing. The device is not unlike what we find in Gorgias' Palamedes or, for that matter, in the extant fragment of Antiphon's defense: consider all possible motives and opportunities, and exclude the alternatives. In this technique Tetralogy 1 perhaps imitates the famous showpieces.²² It also draws upon conventional arguments in the court speeches, such as we find in Lysias 1.43-46: the defendant eliminates the usual motives and then concludes, “If I intended to murder him, would I have involved (others as) witnesses, when it was possible for no one to know?”

In a case where the evidence is not so clearly stacked in his favor we might expect a resourceful logographer to reverse the argument, much as the author of the Tetralogies has done; but consider the parallel in Antiphon 5 (Herodes). In this case we have, indeed, other plausible suspects (fellow travelers on the fateful voyage); we might expect the prosecution to exclude them from suspicion, one by one, and the defendant in turn to raise a reasonable doubt by suggesting that some other assailant did the killing (as the defendant does in Tetralogy 1). But instead he simply cites examples to support the conventional topic, that murder without witnesses is impossible to prove outright (64-69). He reminds his judges of the murder of Ephialtes, for which no one was ever charged (though some were suspected for motive).²³ There was also the case of a slave boy who murdered his master and would have gone undetected had he not lost his nerve. Rather than answer circumstantial evidence with probable suspects (as in Tetralogy 1), Antiphon tends to focus his argument upon the formal instruments of proof; his probabilities are addressed to the validity of the evidence, not the ultimate questions of motive and opportunity.²⁴

Compare also the version of this argument in Antiphon 1.28, against the oath sworn by the defense: “How can anyone know what he was not present to witness? Those who plot murder … do not make their plans and preparations in the presence of witnesses, but with as much secrecy as they can manage, so that no one may know.” The victim was taken unawares but survived long enough to call upon his son, the plaintiff, for retribution; otherwise no one would have suspected. In this passage once again the topic (“murder without witnesses, unproveable”) is introduced not in regard to the facts of the crime but to the particular evidentiary or assertory proofs that constitute the basis of the case: it provides an argument against the defense oath and in favor of the plaintiff's. In all three court speeches, in fact, the topic is directed against artful conjecture, in favor of the speaker's oath and other formal instruments of proof. In Tetralogy 1, on the other hand, the topic serves as a justification for basing the case on circumstantial evidence, disregarding the customary validity of oaths and other documents.

In general the Tetralogies differ widely from the court speeches in their handling of evidentiary proof. The only treatment of witnesses is found in Tetralogy 3, where they are mentioned sporadically and the author seems to assume that witnesses can be found to substantiate any assertion. This feature is not surprising if we assume that the Tetralogies are a purely theoretical essay; but it may also indicate a cynical view of the traditional rules of evidence. The treatment of testimony under torture is particularly suggestive.

In all three of Antiphon's court speeches the speakers claim to have challenged their opponents to decide certain crucial questions by torture; the opponents, of course, refused.²⁵ In Antiphon 5 (Herodes) the slaves would presumably have established the defendant's alibi; in Antiphon 6 (Choreutes) they would have established that the defendant had no involvement in the events leading to the victim's death. Again, in Antiphon 1 the plaintiff challenged the defense to submit their slaves to torture: the slaves would presumably have established that the step-mother once attempted to drug the victim. These challenges were lodged in preliminaries, and had they been accepted the procedure would have been administered before the trial, with the wording of the interrogation submitted in writing beforehand and the responses witnessed by both sides.

The author of Tetralogy 1 is obviously familiar with the rhetorical uses of such evidence but he is certainly not bound by the procedural rules. There is a good deal of argument devoted to the dying servant's testimony (as crucial or questionable without torture). But in the closing speech for the defense we meet with a direct challenge by the defendant to submit the question of his alibi to his slaves under torture. At this point in the proceedings there was no longer any practical opportunity for this procedure; the author of the court speeches never even suggests such a ploy. The author of the Tetralogies evidently ventured upon a surprise tactic, which would have been doubtful of success in court, in order to achieve a sensational effect with another audience.²⁶

In all, the author of the Tetralogies seems reasonably attentive to the framework of Athenian law. He knows, for instance, that the attending physician cannot be charged in the death of his patient; he is also aware that the famous case in Tetralogy 2 was not technically within the jurisdiction of the Delphinium court; he is evidently familiar with the language of Dracon's law regarding the instigator (βούλευsας) and the initiator of a violent quarrel (ἄϱξας χειϱῶν ἀδιaῶν). None of the details can be controverted by other evidence.²⁷ But in his handling of “issues in dispute,” evidentiary proof, and probabilities the author of the Tetralogies contradicts the very principle of homicide proceedings. His subject matter suggests an interest in peculiarly legal or juristic problems, and his method seems to reflect a certain skepticism regarding the traditional homicide procedure: How can the case be limited to a foregone decision on the “issue in dispute”? How can a man be condemned on so flimsy a pretext as hearsay of a dead slave's accusation?

It is reasonably clear, in any event, that the author was chiefly concerned with creating provocative effects rather than illustrating practical technique. By the theoretical model, it is usually assumed that the author's aims are purely abstract or technical: the Tetralogies have little or no connection with actual court proceedings, though some social critique may be implied.²⁸ But it is also reasonable to suppose that an experimental adaptation of this kind—following a specific trial format and responding to certain modes of argument peculiar to that procedure—is likely to follow or parallel developments in court practice. The kind of critique implied in the Tetralogies is more likely in an era when the failings of the traditional procedure were all too apparent and trumped-up charges on perjured testimony became a notorious tool of litigation.²⁹ As we shall see in the following sections, the author's treatment of other conventional topics suggests a similarly critical perspective on the arguments that prevailed in the courts of the ephetai.

II. THE BURDEN OF BLOODGUILT

Regarding the consequences of defilement that a killer brings upon the community the Tetralogies adopt a prejudicial tactic that is very much at odds with the court speeches: the speakers repeatedly call upon the judges to realize the threat to themselves and to the polis, and to cleanse the city of this evil.³⁰ In the court speeches the threat of defilement and avenging spirits is raised solely in regard to the principals themselves: they risk pollution and the wrath of the victim's spirit if they fail to honor him or to abide by their oath; but there is no overt consideration of a threat to the judges or their responsibility to safeguard the community. The threat of defilement is invoked only to prove credibility of the speaker's assertions against those of his opponents. Thus the defendant in Antiphon 5 (11, 82) argues that his safe passage at sea, where a tainted killer would be likely to meet with misfortune, is proof of his innocence. In Antiphon 6.40 the speaker recalls that the prosecutor met with him publicly soon after the death, as proof that his accuser did not then believe him guilty. Again, in both 5 and 6, we find the commonplace that even a man who kills his own slave, with no one to avenge the death, will purify himself as religion requires, without any legal sanction to compel him (6.4 = 5.87). The killer cannot evade the consequences of his guilt by deceiving the judges, and therefore defendant's plea of innocence has the greater claim to credibility.

Such is the characteristic use of miasma doctrine as a source of arguments in the court speeches: the fear of defilement and avenging spirits is an assumed premise in proof of the speaker's credibility or his opponent's deceit. The Tetralogies, however, repeatedly invoke the fear of defilement to different effect: it is a matter for the judges to weigh directly as a threat to themselves and to the community they represent. Thus in Tetralogy 1 the prosecutors begin with a version of the standard argument—“the defilement falls entirely upon us” if the wrong man is convicted and the real killer goes free (1.3)—but they quickly proceed to the more ominous implications. In the conclusion to the opening speech we find the threat to the greater community now elaborated as the peculiar responsibility of the judges: “It is inexpedient for you to allow this man, tainted and unholy as he is, to enter the shrines of the gods and defile their sanctity; … for it is from just such desecrations as this that public disasters befall. … It is rather your personal obligation to effect vengeance, regarding his acts of impiety as your own disaster, thus to absolve the community of defilement” (1.10-11). In response (2.11) the defendant claims that it will not be he who pollutes the shrines of the gods, but the prosecutors, if they falsely convict him: they will be the cause of calamity for persuading the judges to offend the gods.

It is perhaps reasonable to assume that such beliefs were implicit in Antiphon's court speeches, but none of them expresses this ominous injunction to the jury. The closest parallel comes in the epilogue to Lysias' speech Against Eratosthenes (12.99-100), but even this instance points up the difference between miasma arguments in Antiphon's court speeches and the more provocative version in the Tetralogies. Lysias does indeed refer to the defilement of temples and public places by the Thirty, and it is therefore all the more striking that he seems to avoid laying the responsibility directly upon the judges. He contends that his own ardor in prosecuting the case has left no stone unturned in defense of the temples that these men defiled. In keeping with the standard miasma argument, he defends his credibility against the obvious charge of vindictiveness. He is also acting, he contends, in defense of the dead: “I believe they are listening and will know how you vote, and they regard those who acquit these men as casting a verdict of death upon them, but those who exact punishment take vengeance in their behalf.” Threatening as it seems, there is still a noteworthy contrast between this epilogue and the miasma argument characteristic of the Tetralogies: Lysias does not warn that the danger of defilement, as a threat to the community, must weigh upon the judges; he may suggest to the superstitious among his audience that they will be haunted by spirits of the dead, but he does not say so. He adheres to conventional reasoning on the problem of bloodguilt—the judges are not implicated. It is perhaps only in a speech such as this that Lysias could even insinuate such a threat.

This speech for publication follows, perhaps faithfully, the practice of the courts; but we should bear in mind the turn of events and the type of case which it represents. It could not be prosecuted by homicide procedures, which were severely restricted by the oath against vendetta, μὴ μνηsιaαaεiν, by the rule against charges of instigation (bouleusis) in dikai phonou, and by the rule disallowing “warrant and arrest” for crimes done in the recent civil conflict.³¹ Eratosthenes' complicity in the killing of Polemarchus could only be addressed in the special accountings to which members of the oligarchic regime must submit. As a result, any allegations of homicide against such defendants became more a criminal matter—relevant to the interests of society—than simply a matter of private wrongs and personal liability (as dikai phonou were traditionally prosecuted). Even in that setting Lysias evidently did not yet feel he could claim outright that “these men bring damnation upon the community, and it is your duty to see that the polis is absolved.” Instead he is largely faithful to the conventions of legal argument on these religious matters. It may have been the suggestion, however, of a superstitious threat to the jury in this famous speech, or others like it, which prompted the author of Tetralogies to elaborate the theme as he does. Consider the peculiar turn that this argument has taken in the tetralogy: “If the defendant is wrongly acquitted, the avenging spirit of the dead (prostropaios) will not trouble us but will haunt you. Realizing this threat, avenge the dead … cleanse the city. For thus … you absolve yourselves of defilement on his account” (1.3.10-11).

Again in the first speech of Tetralogy 2 the author appears at first to adopt the traditional miasma argument but then goes far beyond the practice of the court speeches. The plaintiff first observes that the vengeful spirit of the victim is only a threat to the willful murderer (and in this distinction he is faithful to traditional belief), but he then turns this principle into a surprising emotional appeal: the unintentional killer “has roused no rage of vengeance in the dead but in the living”; the judges should be as much concerned for the distraught family as for the deceased (2.1.2).³² Following this appeal, the author warns of the threat of defilement to the city as a special concern of the judges: “Do not permit the whole community to be defiled by his guilt.” And he returns to this theme in the epilogue of his second speech: “The mark of defilement … weighs upon you … : if you convict and bar him from the prohibited areas, you (yourselves) shall be free of any claims; but if you acquit him you become responsible. … Do not share in his defilement.”³³

This theme is finally formulated as a mythos in Tetralogy 3: a man's life is a gift of the god, and the spirit of the dead is roused to vengeance by the theft of what god has given; thus he bequeaths the wrath of avenging spirits as god's retribution, and his curse pursues those who wrongly bear witness for his killers and those who wrongly judge against him (3.1.1-4). The speaker invokes this specter as a guarantee of his credibility but also as a threat to the judges.³⁴ The defendant takes up the argument (3.2.8), threatening the judges with avenging furies if they should convict him. In the epilogue to the final speech (3.4.10-11), the defense urges the judges to fear for themselves defilement and avenging furies.

By contrast, the standard injunction to the jury in Antiphon 6.3-6 emphasizes the judges' piety toward the gods and the laws, not the consequences of defilement.³⁵ Again in the parallel passage Antiphon 1.3-4, there is no reference whatever to miasma or avenging spirits as a concern of the judges; and in this speech, if anywhere, such an appeal would seem fitting, since the plaintiff calls upon the judges to act as his kinsmen (anankaioi), as his legal relations have forsaken him. Nonetheless, the implicit doom of defilement and wrath of the dead is conjured only as a guarantee of the speaker's conviction.

The more ominous turn of the argument, particularly the generalized threat of miasma in cases of unintentional homicide (Tetralogy 2) and the myth of haunting as the spirit's vendetta for the theft of what god has given (Tetralogy 3), has a suggestive parallel, however, in Plato's adaptation of the popular mythology in Laws 9.865d-e: “They say that the victim of a violent killing, … soon after death, overwhelmed with fear and dread at the violence done him, rages against his killer; and seeing his killer frequent his familiar haunts, the spirit is frightened and disturbed and himself endeavors to wreak havoc with every power he has, with guilty conscience as his ally, upon the perpetrator, both the man himself and all his affairs.” This “ancient mythos” is introduced to justify the rules requiring exile and purification even in cases of “unintentional” killing. It must be given due reverence (Plato suggests) not because it conveys a literal truth but because it embodies and enforces a useful moral. We find, I think, a similar perspective on traditional belief in the Tetralogies. The author's archaizing piety is a formal posture; his evocative language—aηλίς, πϱοsτϱόπαιος, aτλ.—is borrowed from the drama, not from the courts. His pronouncements on threat of haunting and defilement are certainly no reliable indication that the Tetralogies belong to the era of Sophocles' Antigone, as they are sometimes taken to be.

The evidence on this issue is, of course, no proof against authenticity. It is at least reasonably clear, however, that the emphasis upon defilement as a threat to the polis is uncharacteristic of the court speeches; and it is a strong indication that the Tetralogies would have been of little value as a “practice and demonstration text” in the age of Antiphon. On balance, the evidence on evolution of procedure and argumentation should weigh against the views of Zuntz, Decleva Caizzi, and others, who see this feature of the Tetralogies as inspired by an authentic, presophistic religiosity. Considering this feature in isolation, we might simply conclude that it is yet another instance where the Tetralogies are purely theoretical in their aims and irrelevant to court practice. But the very notion that the trial is fundamentally concerned with this threat to the polis, and that the judges' verdict should be directed against this urgent danger, implies an immediate understanding of homicide as a public wrong, not merely a private offense to be settled by a decision regarding liability as in the traditional procedure. If, as all the evidence indicates, Dracon's law was strictly addressed to the issue of liability—religious concerns being parallel and independent of statute—then on this parameter we have a simple linear development from Dracon's trial procedure to the extant court speeches: dikai phonou remained quintessentially a form of settlement regarding private liability. In the Tetralogies, however, we find an emphatic declaration that homicide must be addressed as a public wrong.

III. THE LAW PROHIBITING JUSTIFIABLE KILLING

In Tetralogies 2 and 3 the author invokes “the law prohibiting just as well as unjust killing” as the basis for a puzzling sequence of argument. This prohibition is treated as a rule of statute law, yet it seems inconsistent with what we know of the wording of Athenian laws: in phrasing it lacks the conditional form characteristic of statute (e.g., “if a man kill … let him be exiled”), and in substance it seems to be flatly contradicted by considerable evidence that “justifiable homicide” was indeed sanctioned in a number of special circumstances. There is also an unlikely ambiguity in the meaning of “just” or “justifiable” killing itself: it seems to be synonymous with accidental or unintentional homicide in Tetralogy 2, but equivalent to killing in self-defense in Tetralogy 3.

To account for these inconsistencies it is often supposed that “the law prohibiting just as well as unjust killing” is invoked as a principle of popular morality, in essence a rule of unwritten law rather than statute.³⁶ “Just as well as unjust” is a polar expression; the prohibition amounts to a moral absolute—all killing, right or wrong, is forbidden.³⁷ But this explanation does not account for the fact that the prohibition in question is treated as at least the substance of statute law: in both speeches it is the law upon which the plaintiffs base their case (2.2.9, 3.2.5). It has been suggested that the author intends to show how unwritten law might actually overturn statue law in the courts.³⁸ But there are no clear parallels for the legal validity of unwritten moral rules. We shall be better able to comprehend the author's purpose if we first determine more clearly what is meant by “justifiable” killing in each context. The more straightforward instance is found in Tetralogy 3.

In Tetralogy 3 the meaning of phonos dikaios seems at first equivalent to “killing in self-defense.” The defendant claims that the victim struck the first blow, and he was justified in using equal force; indeed, he would have been justified in responding with greater force against the aggressor. He is evidently alluding to one of the provisions of Dracon's Law, of which enough of the wording remains for us to see that there was a specific rule regarding justifiable response against the initiator of a violent altercation, ὁ ἄϱξας χειϱῶν ἀδίaων.³⁹ But the defendant cannot simply rely upon the protection implied in the law: he must prove he is not the “killer,” the agent causally responsible.

The defendant introduces “the law forbidding just as well as unjust killing” (3.2.3) as a form of dilemma: he is surely not guilty of unjust killing, since he responded justifiably to provocation; were he the killer, the killing would be justifiable (though culpable); but he is not, he insists, the causal agent—that responsibility lies elsewhere. In short: if he were guilty, it must be for either just or unjust killing; his action was justifiable, not unjust; his justifiable response was not in fact the cause of death; therefore he is guilty of neither just nor unjust killing. The “law” prohibiting justifiable killing is a purely argumentative device, and we should be wary of reading a moral absolute into it.

From this argument it is reasonably clear, in any event, that “justifiable homicide” involves an act that the author himself regards as retributive or retaliatory violence: “Had the man died in immediate consequence of the beating, he would have died justifiably at my hands”—for those who start a fight deserve to suffer more harm than they do (3.2.3). There is no pretense that the defendant was in fact endangered by the victim or acted truly in self-defense; rather he claims to have acted justifiably in meeting injury with retribution.⁴⁰

In the second defense speech (3.4.8) those who speak for the defendant return to this argument, but they now link the rule against justifiable killing with the concept of “culpable error” and “mischance” (ἁμάρτημα, ἀτύχημα). The law forbidding just as well as unjust killing has been answered, they say, inasmuch as the victim died from ill effects of medical treatment and not from the blows themselves. Such mischance is the aggressor's doing rather than the responsibility of one who responds to another's aggression. The defendant acted entirely unwillingly and suffered misfortune brought on by another's error, while the victim acted entirely voluntarily and brought on his own misfortune.

The argument in Tetralogy 3 is consistent with that in Tetralogy 2 to this extent, that both arguments for the defense ultimately attempt to reduce the issue to one of culpable error for which the victim himself is to blame. The homicide is “justifiable” in this sense, that it was brought on by the victim's culpable error. The so-called self-defense thus involves some of the same factors as the unintentional killing: it is the victim himself who set the sequence of events in motion; the “justifiable” killing is his doing, not the defendant's. “Justifiable homicide” is not, however, equivalent to unintentional killing, neither in Tetralogy 3 nor, as we shall see, in Tetralogy 2.

The original meaning of “justifiable homicide” as retributive killing is clear in a number of fifth- and fourth-century references. In the fifth century vindictive killing would appear to be the usual, proper meaning of phonos dikaios; a certain moral repugnance is evident, but there is nothing parallel to the notion that it was prohibited by law, written or unwritten.⁴¹ Indeed, the issue before the judges in Eumenides is whether the killing was justified or not (611-13).

The clearest parallel to the defendant's position in Tetralogy 3 is to be found in Lysias 1.37. The speaker argues that he would have been justified in luring the adulterer to his doom, in retaliation for earlier injustice: “I would have thought myself justified (δίaαιος) in apprehending by whatever means the man who had corrupted my wife.” Such “justifiable,” i.e., retributive, killing must be disavowed, however, and defendant must base his defense upon an argument from causation, that he did not entrap the victim and was not the initiator of the causal sequence. The argumentation in Lysias 1 will need more detailed treatment elsewhere; but there can be little serious dispute that the real issue is the question of entrapment and that the defendant cannot simply rely upon the statutory provision against adulterers caught in the act. He must show that he did not plot to bring about the fatal outcome, that it was caused by other factors (including the victim's actions). There is, in any event, a singular parallel to Tetralogy 3 in the turn of argument, whereby the defendant insists that retributive killing would have been justified but that he did not kill even justifiably.

Now it is reasonable to suppose that the special jurisdiction at the Delphinium court—where such cases as Lysias 1 would be tried—prompted legal controversy on such issues, perhaps as early as the era of Antiphon but surely by the early fourth century. The Delphinium was commonly regarded as the court of “justifiable” homicide; its jurisdiction included retributive killings, as in cases of sexual violation or violent theft or abduction, along with some cases of unintentional homicide, such as death in athletic competition or by mistaken identity in warfare. From the latter applications of the Delphinium law, however, we should not assume that the concept phonos dikaios was in any sense equivalent to unintentional killing. Though popular usage refers to the Delphinium as a court of phonos dikaios, it is clear from Ath. Pol. 57.3 that the jurisdiction was properly restricted to those cases where the defendant admitted to the killing but claimed legal justification by a specific statutory provision.⁴²

Controversy over the ambiguities latent in the Delphinium jurisdiction is reflected in the often puzzling arguments of Demosthenes in defense of the traditional courts in Against Aristocrates; there it is evident that phonos dikaios retained a strong sense of retributive killing. In the section covering the Draconian laws Demosthenes deals with several statutes affecting the Delphinium jurisdiction. Nowhere does he suggest that “justifiable homicide” was a defined legal category, but it is clear that the speaker and his audience would readily interpret cases within this jurisdiction, even unintentional killing in athletic contest, as “justifiable homicide.” In such instance justification lies in the victim's “culpable error.” Thus in explicating the Draconian law (53-55) Demosthenes argues that one who kills (accidentally) in athletic competition is free of guilt, since the victim proved unequal to the contest: “[the lawgiver] considered the victim himself responsible for his own suffering, if he was too weak to endure the struggle for victory.” This interpretation of the lawgiver's intent appears to be Demosthenes' own inference, but it is likely to reflect popular understanding of the law, and it is entirely consistent with the topic in the Tetralogies.⁴³ “Justifiable killing” includes those deaths that come as a consequence of the victim's culpable error.

Demosthenes devotes considerable attention to those killings that are “lawful and justifiable.” Thus he reports the Draconian law allowing self-help against theft or abduction (60), where the defender may kill with impunity; that the killing is “justifiable” is implicit in the condition that the offender be caught in the act of “violent and unjust” theft or abduction. Aristocrates' illegal measure, by contrast, condemns anyone who might kill Charidemus “even if justifiably, even if according to law.” Again, in regard to the Delphinium court itself (74-75), though there was apparently no preamble or other text defining “justifiable homicide” per se, the essence of the jurisdiction seems to be retributive killing. From the example of Orestes' trial for vengeful killing, Demosthenes assumes, the ancient lawgivers once deduced that there are indeed some “justifiable” killings, “for the gods would not have given unjust verdicts.” Aristocrates' bill has not exempted such cases but outlaws anyone who kills Charidemus “even if justifiably, even if according to law.”⁴⁴

The problematic concept of “justifiable homicide” appears to be a substantive issue in the case against Aristocrates. “Of all acts and speech,” Demosthenes urges, “one of two epithets apply, ‘just’ or ‘unjust’”; no one act or word could be both “just” and “unjust,” “for how could the same things be both just and not [just]?” He concludes, “Every act is deemed to have one quality or the other and if it seems ‘unjust,’ it is judged despicable, but if ‘just,’ worthy or noble.” Of course Aristocrates' bill makes no such distinction. It is perhaps conceivable that Demosthenes is simply elaborating upon an all too obvious commonplace, a distinction without a difference in the law. But it is more reasonable to assume that the argument has relevance to the legal issues of the day, and that it might be argued, in support of Aristocrates' bill, that even so-called justifiable killing was forbidden unless prescribed by law. Thus Demosthenes, in his defense of the traditional homicide courts, attempts to formulate the principle of “justifiable” killing, and in doing so he seems to be countering an argument based on the Delphinium jurisdiction: only where the law explicitly provides for a plea of “lawful homicide” is such a defense admissible; in all other cases even “justifiable homicide” is prohibited.

In the same period we find further evidence of an ongoing debate on legal justification and a suggestive parallel to the case in Tetralogy 3: the case reported in the speech Against Meidias, 21.72-75, against Euaion, convicted in the killing of Boiotos. Euaion struck back in anger against a drunken assailant, who was evidently the weaker of the two (just as in Tetralogy 3). He was convicted by one vote. The issue, Demosthenes supposes, was not whether the defendant was justified in striking back, but whether his action was the proximate cause of death.⁴⁵ Those who voted for acquittal (he observes) evidently judged that even lethal force, though “excessive retribution,” was granted by law in the case of one who suffered violent insult: the action of his assailant was regarded as the direct cause of his justifiable response. Similarly in Tetralogy 3 the defendant asserts that it is justifiable to respond with equal and greater force against the aggressor.⁴⁶

Demosthenes treats the case against Euaion as a famous precedent of an earlier era (though perhaps within living memory). It is not unlikely that this case or one like it served as a model for Tetralogy 3—just as the case that Pericles and Protagoras debated served as a model for Tetralogy 2. From the parallels alone it is reasonable to assume that “the law prohibiting just as well as unjust killing” is a formulation of the principle that simple retributive killing is in itself wrongful and illegal; only those instances which statutes prescribe are protected by the Delphinium jurisdiction. The legalistic controversy on this issue and the connected problem of culpable error may well go back to the era of Antiphon, but there our sources fail us. For a better understanding of the argument in Tetralogy 2 we must turn, again, to the fourth century.

The issue of “culpable error” appears as a conventional topic in the fourth century. Aristotle, in the fifth book of Nicomachean Ethics as in Rhetoric, gives a concise but revealing distinction of the key ethical terms, “wrong” or “injury” (ἀδίaημα), “error” (ἀμίaτημα), and “mishap” or mischance (ἁτάχημα). The report in Ethics 5.8 is especially revealing in regard to the reasoning of Tetralogy 2:

Of three types of harm (or damage, βλαβή)… those done in ignorance are “errors” (ἁμαϱτήματα), as when one fails to affect the (right) person, or effect the (right) outcome, by the means or for the reason intended; for [the agent] either thought he would not hit (the mark), or not with the weapon (that struck), or not the (right) person. … Now when the error is contrary to reasonable expectation it is a mischance; but when it is not (unforeseeable) yet without malice, it is an “error,” for the agent “errs” [or “is at fault”] when the origin of causation is in him, but he suffers mischance when the cause is external.

The phrase “at fault” is perhaps the best English approximation to the sense of liability, contained here in ἁμαϱτάνειν/ἁμάϱτημα. “Error” is causal: it is thus related to “injury.” We find a similar distinction in Rhetoric and yet a third version of the commonplace in Anaximenes' Ars Rhetorica.⁴⁷ There is also a similar conception of injury from error, as liable for compensation or retribution, in Plato's Laws (as we shall see). In all of these treatments, “error” is an unintended wrong, yet it is akin to “injury” and ordinarily involves liability.⁴⁸ Bear in mind that these are not sophisticated philosophical constructs but the philosopher's and the rhetorician's rendering of an ethical commonplace.

Such would seem to be the line of reasoning in Tetralogy 2.2.7-9, where it is usually assumed that “justifiable” killing is equivalent to unintentional. The defense first establishes that the boy who threw the fatal lance did not himself err but acted precisely as intended: the javelin was thrown on the right trajectory at the proper moment and would have struck the target. By Aristotle's model, he neither erred in regard to the act itself, the instrument, or the aim, nor in regard to the person affected. Rather the error lies in the action of the victim, who moved into the path of the javelin at the fatal moment. It is he then who erred, causing an unintended outcome, from which he himself suffered just retribution: “and for the error he has been avenged upon himself and has exacted punishment. … As the error redounds to him, the act belongs not to us but to the one guilty of error, and since suffering comes upon the agent himself, we are acquitted of blame. He has been justly avenged upon the agent of his wrong in the moment of his error” (2.2.8). It is at this point (2.2.9) that the defendant invokes “the law prohibiting just as well as unjust killing” as the basis for his acquittal: “for this boy is acquitted of unintentional killing by the error of the victim himself; and since he is not even charged with intentional killing he is acquitted on both counts.”⁴⁹

The same kind of dilemma is utilized here as in Tetralogy 3: wrongful death is either just or unjust; the defendant is not even charged with an unjust act of malice; the justifiable killing is the victim's own doing; the defendant is therefore not guilty of the killing on either count. This passage should not be construed as a straightforward equation of “justifiable” with unintentional killing; one stands for the other only as a term in the dilemma. What renders the killing “justifiable” is that the victim's unintended but no less culpable error has been justly punished by his own hand. That this is the thrust of the argument is proven by the reply in the second prosecution speech: “I, too, assert that the law rightly demands killers be punished: for it is just that the unintentional killer met with unintended harm, and the one who has perished … would be wronged if he were not avenged” (2.3.7-9). The plaintiff accepts the notion that retribution for error is “justifiable,” but he insists that it is the defendant and not the victim who committed the error.

Thus in both these tetralogies “the law prohibiting just as well as unjust killing” serves as the basis for a sophistic dilemma, and in both demonstrations phonos dikaios means essentially retributive killing, extended to retribution for culpable error. This device, I suggest, was inspired by controversy regarding the Delphinium jurisdiction. Properly interpreted, the Delphinium rule amounts to a restriction upon those cases where the defendant can plead legal justification. A plea of lawful homicide is admissible only where the killer can invoke a specific provision of statute (allowing self-help or absolving the athlete or combatant); otherwise even “justifiable” killing is prohibited. In Lysias 1 the killer of the adulterer must therefore prove that he did not plot the adulterer's murder even justifiably. The author of the Tetralogies has adapted this rule (in the terms in which it was popularly understood) as a device for transferring blame in cases where culpable error may be assigned to the victim: the law prohibits just and unjust killing; all killings are either just or unjust (as Demosthenes insists, 23.75); if the defendant's act is neither just nor unjust killing, he is not the killer. In each case the killing is clearly not unjust (as no malicious intent is alleged); the justifiable killing is the victim's doing, since he has brought retribution upon himself for his culpable error. By this device for transferring blame (what later theory would call metastasis) the author has, in effect, reversed the conventional bearing of the rule that even justifiable killing is prohibited.

Thus in the reports of Aristotle and Demosthenes we discover the premises for the otherwise puzzling argument in Tetralogy 2 that treats an unintentional killing as retribution for culpable error—which, though “justifiable,” is nonetheless prohibited.⁵⁰ There is also a suggestive parallel in Aeschines 2.87-88, though it is difficult to construe what precisely Aeschines means by “justifiable” killing. He contrasts Demosthenes' unscrupulous charges with the final oath in homicide proceedings where the victor must forswear a “just” but wrongful verdict. Whatever we are to make of Aeschines' argument, this passage at least confirms that it was meaningful in the fourth century to speak of law and procedural rules—not unwritten law—as prohibiting both just and unjust killing.⁵¹

There is further evidence of popular controversy on the concept of “justifiable killing” in the treatment of homicide in Plato's Laws, closely contemporary with Demosthenes' Against Aristocrates. The laws regarding bloodshed are easily the most detailed and systematic set of statutes in the whole of Plato's code, and perhaps more than any other part of the Laws they seem to be a calculated response to current fallacies of popular reasoning regarding ancient legal principles and institutions. The various preambles and statutes that have to do with what Athenians called “justifiable homicide” (for trial before the Delphinium court) constitute the outer frame of the homicide code, and in the general introduction we could ask for no clearer statement that this theoretical construct is intended to address popular misconceptions of “justifiable” retribution (cf. 859d-60c).

Plato's “Athenian” attempts to reconcile the popular distinction between willful and unintended wrongs with the principle that none do wrong willingly, that wrongdoing originates in a disorder of the soul (860d-e); he concludes that the two categories are indeed to be recognized but are based on a different, more meaningful distinction (861c-d). The lawgiver must recognize two separate criteria of judgment and their respective aims: the law is to redress both “injustice” (ἀδιaία) and “damage” or harm (βλαβή); confusion of the two concerns has led to popular misconceptions of justice and justifiable acts. The character of an act as “justifiable” cannot, as commonly supposed, be determined purely according to the agent's conscious intent. Justice and injustice are matters of moral order; damage is a wholly separate concern (862b). Plato makes quite clear that certain acts proceeding from a righteous disposition and “just” (in the proper sense) are nonetheless liable to claims by an injured party: “The lawgiver must look to both concerns, injustice and harm, and by the laws make good what has been damaged, restore what was lost and set right what was overturned; killing or wounding he must endeavor to heal, by reconciling the suffering (family of the victim) to those responsible, by compensation for their injuries” (862b-c).

Following these principles of moral legislation, Plato amends the traditional divisions of jurisdiction; he dispenses with the category of “justifiable homicide” altogether. He begins with what the Athenians would have regarded as lawful or “justifiable” homicide, to be tried at the Delphinium court, but which he calls “violent though unintentional” (βίαια aαὶ ἀaούsια). They include (1) accidental death in athletic competition and (2) accidental death in war or military training, including practice with the javelin and other weapons.⁵² These cases are remedied by ritual purification. A similar remedy applies (3) in cases where a patient dies under a doctor's care. These cases all have obvious parallels in the Tetralogies.

Self-help killing with impunity (ἀνατί) is granted if an intentional killer returns from exile (871c). And in the final section of the homicide code, there are legal protections for self-help killing against (1) theft by night, (2) “mugging,” (3-4) sexual violation of blood kin or wife, and (5) in defense of kin against lethal threat. These cases, and those described as “violent but involuntary,” would all come under the jurisdiction of the Delphinium court at Athens if the killer pleaded legal justification. They are nowhere described as “justifiable” but rather as cases where the killer is “free of taint” (aαθαϱός) or “the wrongdoer be slain with impunity” (νηποινεί), or as “those cases where the law allows killing” (876c). In this regard Plato reflects the wording of Athenian law, but he is also following out the principles of moral legislation that he proclaimed in his preamble in answer to popular notions on such problematic concepts as “justifiable homicide.”

Demosthenes in the speech Against Aristocrates takes up similar issues. There are also suggestive parallels in the fragments of fourth-century thinkers that Aristotle reports as indicative of common topics. The ancient problem of Orestes' guilt came to represent the fundamental paradox of “justifiable” killing.⁵³ The traditional procedure, which had evolved from customary rules of self-help as a mechanism of private settlement, was now regarded with a deeper comprehension of bloodshed as a public wrong. Popular controversy on such issues is perhaps indicated in Plato's Euthyphro; it is plainly a preoccupation of Lysias' speeches in the early restoration era, when he attacks Eratosthenes, his brother's killer, as a threat to society, and when the killer of the adulterer Eratosthenes must disavow even “justifiable” retribution.⁵⁴

CONCLUSIONS

The positive findings of this study may be briefly summarized; the further implications may be more difficult to determine. In his treatment of the major topics of dikai phonou, where the Tetralogies are most clearly parallel to the court speeches, the author has methodically undertaken to confound or reverse the rules and assumptions that governed the court proceedings. This pattern may conceivably be unintentional or incidental to some larger theoretical aim; but it is reasonable to suppose that the author follows the format of dikai phonou and deals with notorious cases and problematic concepts peculiar to that jurisdiction precisely because these legal issues are in themselves of some interest to him.

Whatever the author's aims, we can conclude with some confidence that his method involves a systematic reversal of topics and techniques found in the court speeches. Thus in Tetralogy 2 we find a virtual parody of the standard apodeixis that ordinarily set forth the sworn statements and the “issue in dispute” that preliminary hearings had defined. The extraordinary exercise in probabilities in Tetralogy 1 responds to the standard topic that a murder without witnesses cannot be proved beyond conjecture. The treatment of the religious issue, the threat of defilement and avenging spirits, may also be read as a reversal of the traditional court argument. In the court speeches the threat of miasma and avenging furies is mentioned only in regard to the antidikoi; the peril of bloodguilt is invoked as proof of the speaker's credibility (he would not knowingly defy religious sanctions). In the Tetralogies the religious consequences are invoked as a matter for the judges to weigh directly, as a threat to themselves and to the community, in a legal setting where the capacity for either side to fabricate evidentiary proof is taken for granted. Finally, in the treatment of “law prohibiting just as well as unjust killing,” the author of the Tetralogies has extended a concept proper to the Delphinium court into an area where it was never intended to apply. It is properly a rule restricting retributive killing; the author of the Tetralogies has adapted this rule as a device for transferring blame to the victim: “the killing is either just or unjust; the defendant's action is clearly not unjust (i.e., malicious); the killing is the justifiable consequence of victim's culpable error, and it is therefore his own act and not the act of defendant.” In demonstrating this device the author is responding not to the moral conflict of written and unwritten law but to a legalistic issue that seems inherent in the Delphinium jurisdiction and the popular conception of phonos dikaios as retributive killing.

Such in brief are the positive findings of this study. It is beyond the scope of this investigation to argue the question of authorship or period-authenticity in detail; but the further implications of this finding—that the method of the Tetralogies is in large part a reversal of the court arguments—should at least convey a caveat against the prevailing and often unquestioning assumption that the Tetralogies are Antiphon's work or a product of his era. The legal or juristic focus of these exercises, together with the sum total of connections and discrepancies with other sources, would suggest a certain ideological context such as we more clearly discern in the age of rhetoric and legal reasoning after Antiphon, sometime in the half-century from Lysias 1 to Plato's Laws.

The author's preoccupation with “issues in dispute” (Tetralogy 2) and his ingenuity in devising the hypothetical “who-done-it” (Tetralogy 1) suggest an era in which the traditional procedure was subject to some skepticism. The ancient procedure that Dracon devised and that Antiphon treated with such reverence (6.2 = 5.14) was designed to facilitate private settlement; it is singularly unsuited to investigating and deciding such preliminary questions as whether the defendant is or is not implicated. By the very nature of the procedure, the issue was largely a matter of assigning liability in cases where the involvement of the defendant was admitted but his intent or culpability was in doubt. In my view, we are likely to find an exercise such as Tetralogy 1, which assumes that the evidentiary proof is virtually meaningless and the case should be decided solely on circumstantial evidence, only when the cynical abuses of the traditional procedure were notorious.

From the same perspective the treatment of miasma and avenging furies as a threat to the greater community seems out of place in the era of Antiphon's court speeches. It is an important topic in all three Tetralogies but nowhere clearly articulated in the extant speeches for dikai phonou. It is reasonable to suppose that these religious concerns were present to the minds of judges and antidikoi throughout the long development of dikai phonou. The religious sanctions and consequences so powerfully conveyed in drama appear to parallel the legal proceedings. But they are not issues for judges in the courts to weigh. Neither in Dracon's Law nor in Antiphon's court speeches do we have any clear sense of a “criminal” aspect to dikai phonou, that homicide should be remedied as a public wrong or danger, such as we find, indeed, in Lysias' treatment of the killing of Polemarchus—and in the Tetralogies.

The Tetralogies belong to an era in which rhetoric and the law that shaped it were undergoing fundamental change. They appeal to an audience who regard the laws with a new ideological consciousness, such as may well have been inspired by the restoration of democracy, recodification, and nomothesia. In Tetralogy 2 the plaintiff seems to acknowledge the new distinction of nomoi and psēphismata, and he addresses the people as lawmakers, “sovereign in the state.” We find similar formulations in Andocides' On the Mysteries, where we also learn that traditional rules for prosecution and self-help against homicides were restricted by the very settlement that restored democracy. In the new age of legalism the traditional procedures for homicide disputes, dikai phonou before the ancient courts of the ephetai, were increasingly regarded as antiquated and undemocratic. Demosthenes anticipates just such a reaction in his labored defense of the traditional courts in Against Aristocrates. Of all homicide cases known to us after Antiphon, barely half were still prosecuted by dikai phonou; and in those cases where the traditional procedure was followed, it appears to have become chiefly a tool of litigation. Dikai phonou were often superseded by the summary procedures for “warrant and arrest”; and those cases that were still prosecuted by dikē phonou were often decided without trial either by private settlement or by suits for false witness. Since homicide charges traditionally rested upon the oaths of plaintiffs and witnesses, it was sometimes possible to entangle an opponent in awkward and lengthy proceedings on largely groundless accusations. A special procedure for trial of perjury charges arising in dikai phonou was evidently instituted to remedy such abuses.⁵⁵ The Tetralogies in fact—alone among the extant homicide speeches—allude to this procedure against perjury in dikai phonou; and their very method seems to reflect an utter lack of faith in the traditional rules of evidence and liability. They are in some sense a “polemic” against the peculiar logic of dikai phonou.⁵⁶ They are, in all likelihood, the rough product of an era when the workings of law were subject to a new scrutiny and the tools of face-to-face debate were first fashioned into literary technique.⁵⁷

Notes

1. For general reference see Blass, Attische Beredsamkeit 149-74; Lipsius, Attische Recht 602, 615; Gernet, Antiphon 6-16; Dover, “Chronology” 56-59; MacDowell, Homicide Law 80; Kennedy, Art of Persuasion 130-31; Decleva Caizzi, Tetralogiae; Gagarin, “Prohibition of Homicide” and “Proofs in Antiphon”; Sealey, “Tetralogies.” See also Cole's valuable study, Origins of Rhetoric 73-82, treating the Tetralogies as an early “proto-rhetorical practice and demonstration test.”

2. In favor of period authenticity and discounting discrepancies see Dover, “Chronology” 58; Decleva Caizzi, Tetralogiae 21-44; Gagarin, “Prohibition of Homicide” 303; Cole, Origins of Rhetoric 73-78. An intriguing solution to both linguistic and legal discrepancies has now been offered by Innes, “Sophistopolis,” arguing that the rhetorical setting is not Athens but a broader sophistic construct; for the view that the law of the Tetralogies is fictitious or universal cf. Thiel, Erste Tetralogie 13-15. Against authenticity: on legal and ideological grounds, see most recently Sealey, “Tetralogies.” The dispute goes back to Dittenberger, “Tetralogien,” of 1896-97 and 1905 (responding to Lipsius, “Tetralogien,” 1904). Against Antiphon's authorship on linguistic grounds see Gernet, Antiphon 15-16 (answered by Dover, “Chronology” 58-59); cf. von der Mühll, “Unechtheit” 1-5. The related problem of whether Antiphon the Sophist and the speechwriter are one and the same is beyond the scope of this paper, but see note 28 below.

3. Thus Zuntz, “Earliest Prose Style,” treats Tetralogy 2 as an earlier model than the court speeches; again, in “Tetralogies” 102, on the treatment of “judicial murder” in Ant. 5.91 and Tetr. 3.2.8, he reasons: “The [earlier] parallel in the third Tetralogy is wholly dominated by the idea of the avenging spirit … [which] fails to reappear in extant speeches. The abandonment of this traditional motif is characteristic evidence of a changed ‘Zeitgeist.’”

4. The manuscript hypotheses identify the issues as sτοχαsμός, question of fact (Tetralogy 1), or μετάsταsις/ἀντὲγaλημα, transferred blame (Tetralogies 2 and 3). The hypotheses of course do not go back to the author of the Tetralogies himself, but cf. Cole, Origins of Rhetoric 97-98.

5. Cf. Ant. 1 (Stepmother) 5-7, 28-30.

6. For the guarantee of safe passage to those convicted for unintentional killing see Dem. 23.72. The author's confusion about the penalty has been often observed—defense refers to the prospect of a verdict against him as though the boy's life were at stake (διαφθορὰ, 3.2.10); cf. Sealey, “Tetralogies” 77.

7. Stesimbrotus, FGrHist 107 F 11; Plut. Per. 36.4-6. Xanthippus reportedly told the tale to discredit his father as an effete intellectual. The whole anecdote (36.2-5) probably derives from Stesimbrotus; he is cited by name at 36.6, and a connection between the name-citation and the previous anecdote is indicated by the turn of phrase, πιὸς δὲ τούτοις aαὶ τὴν περὶ τῆς γυναυaὸς διαβολὴν ὑπὸ τοῦ Kανθίππου φηsὶν ὁ Στηsίμβροτος … διαsπαρῆναι.

8. Plutarch seems anxious to quash the slanders of Stesimbrotus (Per. 13-16). See my “Thucydides and Stesimbrotus.”

9. For the ancient tradition that Antiphon of Rhamnous was a teacher of rhetoric cf. Menex. 236a; Anon. Vit. Thuc. 1-2; Radermacher, Artium Scriptores B X.2-3, 5-9.

10. Plato (Phaedrus 273b) attributes the technique to Tisias; cf. Aristotle Rhet. 2.1402a17-22. Gagarin, “Proofs in Antiphon” 30, aptly calls this method “reverse εἰaός.” The hypothesis to Tetralogy 1 cites Lysias Against Mikines as a parallel. The author may also have had in mind the unsolved murder of Ephialtes, cited in Ant. 5.69; see below at note 23.

11. With the ethical proof in Tetralogy 3, regarding youthful akolasia, cf. Aristotle Rhet. 1389a. On the case reported in Dem. 21.72-75 see below at note 45.

12. For this model of homicide procedure cf. Thür, “Zum διaάzειν bei Homer” and “Urteil aus Mantineia.” See also my “Ephetai” and “Phreatto.”

13. Antiphonstudien: on this transitional role see esp. 33-46; Solmsen doubted the authenticity of the Tetralogies and excluded them from his study (3 n. 1).

14. Gagarin, “Proofs in Antiphon”; cf. Goebel, Early Greek Rhetorical Theory, esp. 49-55. The view that traditional modes of proof continued to prevail in dikai phonou will be argued in detail elsewhere.

15. Cf. Tetr. 1.4.7, 10. Witnesses in dikai phonou took the same oath as the parties themselves (cf. Ant. 5.12). The earliest known perjury suit in dikai phonou is reported in Isoc. 18.52 (after 403), possibly given greater emphasis as a novel procedure. Ant. 5.95 (not long after 420) suggests there was as yet no remedy against perjury in such cases.

16. For judgment against the unknown killer cf. Ath. Pol. 57.4; Dem. 23.76. The obstacle against prosecuting without evidentiary proof should not be construed as a statutory rule, barring the basileus by law from letting such a case go forward: in Dem. 23.70 the exegetes' response makes it clear that this was in the nature of a customary rule, whose violation would only tend to discredit the plaintiff.

17. See Heitsch, Antiphon 21-32, following Wilamowitz (“Erste Rede”) in the assumption that the concubine who administered the potion had implicated the stepmother. See also Gagarin, “Bouleusis”: intentional murder by “planning” (bouleusis) would be tried at the Areopagus.

18. A plea of innocence of intent is clearly anticipated in §§ 22-23: the defense will admit complicity in the killing but will plead that the killer acted rashly, “without design.” For the manuscripts' ἀβούλως τὲ aαὶ ἀθέως Gernet unnecessarily emended to ἐπιβούλως, aτλ.

19. It is likely that the concubine (who administered the drug) had implicated the stepmother under torture; cf. Heitsch, Antiphon 25 with n. 57.

20. On the legal issues in this case, see Heitsch, Antiphon 44-56; Gagarin, Murder of Herodes 17-29.

21. Von der Mühll, “Unechtheit” 4-5, offered a similar hypothesis.

22. Tetralogy 1 is in some respects closer to the Ulysses attributed to Alcidamas than to Gorgias' Palamedes; see below note 57. The extant fragment of Antiphon's Defense presumably derives from a post-eventum publication. Von der Mühll, “Unechtheit” 3, even speculated that certain passages in the Tetralogies may directly imitate Antiphon's Defense. For comparison of this technique in the Defense and Ant. 5 see Solmsen. Antiphonstudien 59-62.

23. Plutarch reports (Per. 10) that some suspicion attached even to Pericles.

24. Thus in Ant. 5.43 the defendant attempts to discredit the slave's testimony under torture by protesting “Would I have enlisted an accomplice to be a witness to the crime?”

25. On challenge to torture see Thür, Basanos, esp. 11, 132-33, on Ant. 1.

26. Cf. Lämmli, Prozessverfahren 106; Thiel, Erste Tetralogie 14-15, taking this divergence from court practice as an indication against authenticity; Thür, Basanos 223-24, suggesting that Aeschines 2.126-27 imitates the device in Tetralogy 1. It remains uncertain whether the challenge to settle the case by this “ordeal” was legally binding. The author of the Tetralogies is perhaps misled by Ant. 6.23-24, where the challenge is decisive under exceptional circumstances, in a major political trial before the heliaia.

27. A suggestive detail is the date of the crime, during the Dipoleia, which gives the defendant his alibi (Tetr. 1.4.8). The detail was probably chosen for literary interest (cf. Clouds 983) and as a legal curiosity: the ritual involved the slaying of a bull and search for the unknown killer, evidently parallel to the proclamation of impersonal and unknown killers at the Prytaneion; cf. Parke, Festivals of the Athenians 162.

28. Decleva Caizzi, Tetralogiae 50, 69, suggests that certain arguments are in fact a conservative polemic against sophistic probabilities; cf. Gagarin, “Prohibition of Homicide” 303, regarding this line of argument as “a moral comment on the Athenian legal system.” Those who believe that Antiphon the politician and the author of On Truth are one and the same will perhaps see a connection between the sort of critique suggested in Tetralogy 1 and the treatment of oath-taking in the sophist's treatise; see most recently Gagarin, “Identity of Antiphon.”

29. The reference to perjury proceedings in dikai phonou (1.4.7, 10) is the only such reference in the extant homicide speeches. See above at note 15; on perjury in dikai phonou cf. Carawan, “Ephetai” 3-6.

30. Cf. Sealey, “Tetralogies” 74, against Adkins, Merit and Responsibility 100-106. For the religious sanctions as a parallel mechanism see Parker, Miasma 104-43; and, against the usual view that the Prytaneion and Phreatto courts were instituted to safeguard the community against pollution, see my “Phreatto” 61-67.

31. Ath. Pol. 39.5; cf. Andoc. 1.91, 94.

32. The sentence “He has roused no rage … in the dead but in the living” was bracketed by Gernet; but, for the older belief that the avenging spirit haunts the “intentional” killer, cf. Parker, Miasma 105-30.

33. Tetr. 2.3.11-12. The “mark of defilement,” kelis, is common in tragedy but is not found in Antiphon's court speeches or in Lysias. Cf. Parker, Miasma 107: “murder-pollution as a stain on the hands … is manifestly a symbol of something beyond itself, since the stain is invisible; the Tetralogies unabashedly substitute the thing symbolized for the symbol.”

34. Tetr. 3.1.4-5, ἅπαsαν τὴν πόλιν aαθαϱὰν τοῦ μιάsματος aαταsτήsετε. Cf. 3.3.7, ἀντὶ τοῦ ἀποθανόντος ἐπιsaήπτομεν ὑμην. τω τούτου φόνῳ τὸ μήνιμα τῶν ἀλιτηϱίων ἀaεsαμένους πasαν τὴν πόλιν aαθαϱὰν τοῦ μιάsματος aαταsτῆsαι.

35. Ant. 6.3: “Trials for homicide are of great importance for you, the judges, rightly to decide, most of all for the sake of piety to the gods, and also for your own sake”; cf. §6, where a wrongful verdict is “a wrong and a sin against god and law.” Ant. 1.31 makes a similar appeal to piety toward the gods and the laws.

36. Cf. Gagarin, “Prohibition of Homicide,” esp. 300-303; he concedes that there is no clear parallel in the most likely sources contemporary with Antiphon.

37. The Pythagorean commandment, for which Aristotle cites Empedocles, μὴ aτείνειν τὸ ἔμψυχον (Rhet. 1372b6 = DK 31 fr. 135), is a very dubious parallel, since it amounts to a prohibition against killing any animate creature. For the polar expression “right or wrong” cf. Aristophanes Ach. 373; Andoc. 1.1-2 and 135-36.

38. Decleva Caizzi, Tetralogiae 32-44.

39. With Tetr. 3.2.1, cf. IG I³ 104.33-36, and see Stroud, Drakon's Law 56.

40. Tetr. 3.2.3, Iἰ μὲν γὰρ ὑπὸ τῶν πληγῶν ὁ ἀνὴρ παραχρῆμα ἀπρθανεν. ὑπ' ἐμοῦ μὲν διaαίως δ' ἂν ἐτεθνήaει—οὐ γὰρ ταὐτὰ ἀλλὰ μείξονα aαὶ πλείονα οἱ ἄρξαντεο δίaαιοι ἀντιπάsχειν εἰsί. Cf. Tetr. 1.2.10: “If I am the probable killer, I am justified (δίaαιός εἰμι); for clearly I would have acted in defense against injury (ἀδιaούμενος ἠμυνάμην).”

41. Cf. Aeschylus Agamemnon 1604, Choephori 988; Sophocles Electra 33-37; Euripides Hecuba 263, Electra 1094-96, 1189. Repugnance against this form of “just killing” is expressed in Eur. El. 1050-51, where Clytemnestra challenges, “Say that … your father did not die justly (ἐνδίaως); and the chorus answers, δίaαι' ἔλεξας· ἡ δίaη δ' αὶsχρῶς ἔχει. Cf. the fragment of Theodectes' Orestes in Ar. Rhet. 2.24.3, and see note 53 below.

42. Ath. Pol. 57.3, ἐὰν δ' ἀποaτεĩναι μέν τις ὁμολογη, φη δὲ aατὰ τοὺς νόμους, aτλ.

43. The laws protecting those implicated in accidental death on the playing field or the battlefield were probably dictated by public policy and not originally conformable to the principle of dikaios phonos in the provisions regarding retributive killings.

44. The same phrase is used as earlier: 60-61, aἂν διaαίως, aἂν ὡς οἱ νόμοι διδόαsιν.

45. Dem. 21.75, aαταγνόντας μὴ ὅτι ἠμύνατο … ἀλλ' ὅτι τοũτον τὸν τϱόπον ὥsτε aαὶ ἀποaτεĩναι, τοὺς δ' ἀπογνόντας aαὶ ταύτην τὴν ὑπεϱβολὴν τως τιμωϱίας τω γε τὸ sωμ' ὑβϱιsμρνέ δεδωaέναι.

46. Tetr. 3.2.2, οὐ ταὐτὰ ἀλλὰ μείzονα aαὶ πλείονα δίaαιοι οη ἄϱχοντες ἀντιπάsχειν.

47. Aristotle Rhet. 1374b: “‘errors’ and ‘injuries’ are not to be judged on the same scale, nor are ‘mischances’ (commensurate with) ‘errors.’ For mischances are contrary to reasonable expectation and not of malicious intent; while ‘errors’ are those outcomes not unforeseeable, though yet without malice.” Cf. Anaximenes Ars Rhet. 4.7-9: “you regard a wrong done with intent as ‘injustice’ and say that such wrongs deserve the most serious retribution; but to do harm through ignorance must be called ‘error’; and to fail to accomplish one's intent … because of others or by chance you regard as ‘mischance.’”

48. On ἁμάϱτημα as both “error” and “fault” or blame (errore e colpa) cf. Decleva Caizzi, Tetralogiae 52-58. She emphasizes the connection between ἁμάϱτημα and ἀsέ]βημα. 'Ατυχία/ἀτυχῆsαι is also extremely rare in fifth-century sources—nowhere found in the court speeches of Antiphon—but occurs fourteen times in the Tetralogies. Decleva Caizzi sees this as yet another instance of Antiphon's formative usage, “con valore pregnante” (63-69).

49. Gagarin, “Prohibition of Homicide” 295-97, takes this passage as evidence that “just and unjust” homicide is equivalent to “unintentional and intentional.” He discounts the connection of thought that I have suggested here. It is clear in 2.2.9, in any event, that the assertion that “the law [prohibiting just and unjust killing] acquits us” comes as a conclusion or further implication from the preceding argument: “for [the defendant] is acquitted … by the error of the victim himself.”

50. Cf. Gorgias Helen 122-23, εἰ δ' ἐsτὶν ἀνθϱώπινον νόsημα aαὶ ψυχῆς ἀγνόημα, οὐχ ὡς ἁμάϱτημα μεμπτέον ἀλλ' ὡς ἀτύχημα νομίsτεον. However, ἁμάϱτημα may have the sense of “wrongdoing,” regardless of intent; cf. Antigone 914, 926-27; Ajax 1094-96. In Ant. 6 (Choreutes), which involves the defendant's liability for an arguably culpable error, the defendant never addresses the problem in these terms.

51. See esp. Gagarin, “Prohibition of Homicide” 304-6, arguing that here “unintentional” equals “justifiable” homicide, as in Tetralogy 2 by his interpretation.

52. There is a problem with the text here (865a-b), all manuscripts reading “officers” (ἀϱχόντων) where editors assume “javelin” (ἀaοντίων). Some reference to lethal weapons is in order, since the protection for lawful homicide is valid for training both with and without armor.

53. With Dem. 23.74, cf. Theodectes' Orestes (F5): “‘It is just for the woman who slays her husband’ to die, and just for the son to avenge the father; but not for the son to take vengeance upon the mother” (Rhet. 2.24.3, 1401a); also in Alcmaeon (Rhet. 2.23.3), Theodectes posed the same problem, that retribution may be justified with regard to the offender, but those who presume to exact retribution may be nonetheless denied the right. Similarly Alcidamas in Messeniacus recast Empedocles' μὴ aτείνειν τὸ ἔμψυχον as a principle contradicting conventional justifications: Rhet. 1.13.2, 1373b.

54. Ath. Pol. 39.5; cf. Andoc. 1.91, 94. The council was to grant no “warrant” or arrest (endeixis, apagōgē) for wrongs under the Thirty. By the traditional procedure for endeixis against homicides (Dem. 23.51) plaintiffs had been protected from prosecution if the accused were killed in the arrest; cf. Hansen, Apagōgē 16.

55. Carawan, “Ephetai,” esp. 3-6, 13-16.

56. Decleva Caizzi, however, interprets the argumentation as a conservative “polemic” against argument from probability (Tetralogiae 50, 69).

57. Cf. Cole, Origins of Rhetoric 113-38, esp. 120-23, on “writing out loud” and the adaptation of literary effects commensurate with those of oral performance. Among such effects, I suggest, was a somewhat more realistic representation of court proceedings, such as we find in Lysias 12 and in the treatments of Socrates' defense. As we turn from Gorgias' Palamedes to Alcidamas' Ulysses (Radermacher B VII.44; B XXII.16), in the latter demonstration one is struck by the use of witnesses, the mention of a tenfold penalty, and the intriguing description of lost documentary evidence—the arrow with hidden, winged words. This aspect of the Tetralogies will require further treatment elsewhere.

Bibliography

Adkins, A. W. H. Merit and Responsibility: A Study in Greek Values. Oxford: Clarendon Press, 1960.

Blass, Friedrich. Die attische Beredsamkeit. 2d ed. Leipzig: Teubner, 1887-98.

Carawan, E. M. “Ephetai and Athenian Courts for Homicide.” CP [Classical Philology] 86 (1991) 1-16.

———. “Thucydides and Stesimbrotus.” Historia 38 (1989) 144-61.

———. “Trial of Exiled Homicides and the Court at Phreatto.” RIDA [Revue Internationale de Droits de l'Antiquite] 37 (1990) 47-67.

Cole, Thomas. The Origins of Rhetoric in Ancient Greece. Baltimore: The Johns Hopkins University Press, 1991.

Decleva Caizzi, Fernanda, ed. Antiphontis Tetralogiae. Testi e documenti per lo stúdio dell'antichità 28. Milan: Varese, 1969.

Dittenberger, W. “Antiphons Tetralogien und das attische Criminalrecht.” Hermes 31 (1896) 271-77, 32 (1897) 1-41.

———. “Zu Antiphons Tetralogien.” Hermes 40 (1905) 450-70.

Dover, K. J. “The Chronology of Antiphon's Speeches.” CQ [Classical Quarterly] 44 (1950) 44-60.

Gagarin, Michael. “The Ancient Tradition on the Identity of Antiphon.” GRBS [Greek, Roman, and Byzantine Studies] 31 (1991) 27-44.

———. “Bouleusis in Athenian Homicide Law.” Symposion 1988 (1991) 81-99.

———. The Murder of Herodes. Studien zur klassischen Philologie, 45. Frankfurt: Lang. 1989.

———. “The Nature of Proofs in Antiphon.” CP 85 (1990) 22-32.

———. “The Prohibition of Just and Unjust Homicide in Antiphon's Tetralogies.” GRBS 19 (1978) 291-306.

Gernet, Louis. Antiphon, Discours. Paris: Les Belles Lettres, 1923.

Goebel, George. Early Greek Rhetorical Theory and Practice: Proof and Arrangement in the Speeches of Antiphon and Euripides. Dissertation, University of Wisconsin, Madison, 1983.

Hansen, M. H. Apogōgē, Endeixis and Ephēgēsis against Kakourgoi, Atimoi and Pheugontes: A Study in the Athenian Administration of Justice in the Fourth Century B. C. Odense: Odense University Press, 1976.

Heitsch, Ernst. Antiphon aus Rhamnous. AAWM 1984, no. 3. Wiesbaden: Steiner, 1984.

Innes, D. C. “Gorgias, Antiphon and Sophistopolis.” Argumentation 5 (1991) 221-31.

Kennedy, George. The Art of Persuasion in Ancient Greece. Princeton: Princeton University Press, 1963.

Lämmli, Fritz. Das attische Prozessverfahren in seiner Wirkung auf die Gerichtsrede. Paderborn: Schöningh, 1938.

Lipsius, J. H. Das attische Recht und Rechtsverfahren. Leipzig: Reisland, 1905-15.

———. “Über Antiphons Tetralogien.” Berichte der Königlichen Sächsischen Gesellschaft der Wissenschaften (Leipzig) 56 (1904) 191-204.

MacDowell, D. H. Athenian Homicide Law. Manchester: The University Press, 1963.

Parke, H. W. Festivals of the Athenians. Ithaca: Cornell University Press, 1977.

Parker, Robert. Miasma: Pollution and Purification in Early Greek Religion. Oxford: Clarendon Press, 1983.

Radermacher, L. Artium Scriptores: Reste der voraristotelischen Rhetorik. Sitzungsberichte der Österreichische Akademie der Wissenschaften, philosophisch-historische Klasse, 227, no. 3. Vienna, 1951.

Sealey, Raphael. “The Tetralogies Ascribed to Antiphon.” TAPA [Transactions of the American Philological Society] 114 (1984) 71-85.

Solmsen, Friedrich. Antiphonstudien. Neue philologische Untersuchungen, 8. Berlin: Weidmann, 1931.

Stroud, Ronald. Drakon's Law on Homicide. Berkeley and Los Angeles: University of California Press, 1968.

Thiel, J. H., ed. Antiphons erste Tetralogie. Groningen: Wolters, 1932.

Thür, Gerhard. Beweisführung vor den Schwurgerichtshöfen Athens: die Proklesis zur Basanos. Sitzungsberichte der Österreichische Akademie der Wissenschaften, philosophisch-historische Klasse, 317, no. 1. Vienna, 1977.

———. “Zum διaάzειν bei Homer.” ZSS [Zeitschrift der Savigny-Stiftung] 87 (1970) 426-44.

———. “Zum διaάzειν im Urteil aus Mantineia.” Symposion 1985 (1989) 55-69. von der Mühll, Peter. “Zur Unechtheit der antiphontischen Tetralogien.” MH [Museum Helveticum] 5 (1948) 1-5.

Wilamowitz-Moellendorff, Ulrich von. “Die erste Rede des Antiphon.” Hermes 22 (1887) 194-210.

Zuntz, Günther. “Earliest Attic Prose Style.” C&M [Classica et Mediaevalia] 2 (1939) 121-44.

———. “Once again the Antiphontean Tetralogies.” MH [Mediaevalia et Humanistica] 6 (1949) 100-104.