Punishment in Ancient Athens
Danielle S. Allen, University of Chicago
Part I) Introduction
Ask any modern citizen to name a punishment meted out by their state and odds are the first thing she will say is "imprisonment." Ask the same citizen then to say why modern states use imprisonment as their preferred penalty, and he'll say, "We need to keep the bad guys off the street! Prisons are for deterrence." Or, the odds are just as good that the answer will be, "Prisons are places where criminals can take classes, get religion, hold a job and so be reformed for their re-entry to society." Or, again, still the odds are the same, this citizen may rather say, "Criminals should pay and life in prison should be as hard as possible." Modern citizens are notoriously fixated on imprisonment as the central penalty and just as notoriously unable to reach consensus on whether retribution, deterrence, or reform should be the central principle of punishment.
As philosopher John Rawls put it in 1955:
The subject of punishment has always been a troubling moral question. The trouble about it has not been that people disagree as to whether or not punishment is justifiable. . only a few have rejected punishment entirely . . . The difficulty is with the justification of punishment: various arguments for it have been given by moral philosophers but so far none of them has won any sort of general acceptance; no justification is without those who detest it [emphasis added] ("Two Concepts of Rules" John Rawls, 37).
When in 1968 legal theorist H. L. A. Hart wrote about theoretical efforts to justify punishment, he too displayed unease:
Many are now troubled by the suspicion that the view that there is just one supreme value or objective (e.g. Deterrence, Retribution, Reform) in terms of which all questions about the justification of punishment are to be answered is somehow wrong. . . no clear account of what the different values or objectives are, or how they fit together in the justification of punishment can be extracted [first emphasis added]. (Punishment and Responsibility: Essays in the Philosophy of Law. Oxford: Clarendon Press, p. 16).
Rawls' remark from the 50's and Hart's from the 60's still apply: modern penal theory consistently evinces an unease, a certain troubled-ness, about why we punish. The Athenian democrats, in contrast, to the citizens of modern democracies, knew why they punished. And so we will begin where they did, with this question, "Why did the Athenians punish?" as the first of four questions to be explored in this lecture.
Let me outline the four questions. The first two are both "Why" questions. First, we will ask why the Athenians punished in the sense of, "How come?": "What caused them to punish and when did they think it necessary to punish?" Second, we will ask another "why" question, not "what initiated punishment?" but rather "to what ends did the Athenians punish?" This is "why" in the sense of "What were their objectives in punishing?" Third, we will ask how the Athenians punished. By what processes and procedures did they move from identifying a wrongdoer to assigning him (and upon rare occasion her) a penalty? And fourth, we will ask what penalties they finally imposed on the wrongdoer.
A society's approach to punishment reveals its soul: how it understands cause and responsibility; what its utopian hopes are; and how it has decided to approach conflict. These four questions (why, in the sense of "what cause?"; why, in the sense of "what purpose?", how, and what) should open up Athenian punishment in ways that convey a living society that continually had to make choices about how to construct authority. This story about Athens should, by way of contrast, provoke our own thoughts about how punishment serves the construction of authority in modern states.
Part II) Why (What Cause)?
The Athenians had no doubts about why they punished: it was simply because someone was angry at a wrong and wanted that anger dealt with. Specifically, the anger of the victim necessitated punishment, and the Athenians made this idea central to their penal practice. Although the city's penal laws allowed any citizen to prosecute on behalf of someone who had been the victim of a crime, or on behalf of the city in general, in 96% of the cases for which we still have copies of the courtroom speeches, the prosecutor was in fact either himself the victim of the wrong done or else he was personally involved in some dispute with the wrong-doer. In court, one after another prosecutor would launch his case by invoking and explaining his personal animosity toward the defendant. This is what Aeschines (see Orator Biographies) is doing here in 330BCE in the speech from his prosecution of Timarchus for speaking in the assembly despite having worked as a prostitute:
When I saw that Timarchus was, though disqualified by law, speaking in your assembly, and when I myself was personally being slanderously accused [by him and his allies], I decided it would be most shameful not to help the whole city and the laws and you and myself. It would seem, O Athenians, that the usual saying about public trials is not false: i.e. the saying that private enmities do indeed correct many public matters ( Aesch. 1.1-2).
After the initial story of personal outrage, one after another prosecutor would move beyond that to argue that his jury should also adopt an anger equivalent to his own. Here is Demosthenes (see Orator Biographies) doing exactly this in the 360s:
It's not right that Meidias' behavior should arouse my indignation alone and slip by, overlooked by the rest of you. Not at all. Really, it's necessary for everyone to be equally angry! (21.123).
With rare exceptions, cases of punishment in Athens were directed at resolving a problem that had arisen between two people and that were identified when someone said he was angry. Anger was so central to the Athenian experience of wrong-doing and punishment that courtroom litigants could describe laws as having been established for the purpose of establishing what levels of anger were appropriate for various acts of wrong-doing (e.g. Dem. 21.43; Aes. 1.176). Thus Demosthenes writes:
Observe that the laws treat the wrong-doer who acts intentionally and with hubris as deserving greater anger and punishment; this is reasonable because while the injured party everywhere deserves support, the law does not ordain that the anger against the wrongdoer should always be the same (Dem. 21.42, 43).
Anger was thus assumed to be not only the source of particular punishments but also at the root of law itself. The Athenians accordingly felt relatively little uncertainty or unease about why (that is, in response to what causes) they punished: they acted in response to anger.
Part III) Why (To What End)?
The Athenians, then, punished in answer to someone's anger, but to what end did they do so? If a modern citizen were to hear that someone, a parent or teacher, or a state, had punished out of anger, he would expect the motives of the punisher to be essentially vindictive. Anger, we think, leads directly to a desire for payback of the eye-for-an-eye variety. In contrast, the Athenians developed a far more nuanced view of what it meant to take anger as the starting point of punishment. Anger might be the origin of punishment, but they also conceded that it was a disease.
In tragedy characters regularly invoke anger as the reason to punish but they also reiterate the idea that wrongdoing and its punishment involved the community in some sort of communal sickness. This is especially evident in the tellings and re-tellings of the myth of the House of Atreus, the story of how King Agamemnon won the Trojan War and returned to his hometown of Argos only to be killed by his wife Clytemnestra who was is turn killed years later by their son Orestes. He then is driven out of the city by the Furies. All of the versions of this story use the metaphor of disease to describe the effect of wrong-doing on the diverse members of a community who participate in an event of wrong-doing and its punishment.
Euripides, for instance, describes the victim, that is, the murdered Agamemnon, as a festering wound within the household (Euripides' Electra, 318). In another play, he makes the wrong-doer, Orestes, diseased and calls him a disease in the land (Euripides' Orestes, 395, 831). Aeschylus, in contrast, treats would-be punishers, namely, the Furies, as bearers of illness to the land; he says that their disease drips from their eyes (Aeschylus Eumenides, 480). In the mythical tradition of the House of Atreus all the parties to wrong-doings and the responses to it—victim, wrong-doer, punisher, and the community or "land"—somehow share in a "disease"; and this surely symbolizes the idea that no party to the experience of wrong-doing is exempt from the trouble it introduces to the community. But in exactly what sense is each of these parties diseased?
When Aeschylus describes the Furies' disease, the sickness of their anger, as dripping from their eyes, he employs the common Athenian habit of drawing connections among vision, anger, and the spread of the disease of social disruption. Those who looked upon a murderer were polluted by the sight; and a murderer's glance was said to spread poison just like the look of a snake. In Greek conceptions of vision, sight involved the physical transfer of particles and properties from one person to another. Aristotle provides a graphic example of the idea that vision was a physical transfer of properties from seer to seen when he writes that whenever a woman who was menstruating looks into a mirror, the glass ends up covered with blood (Deinsomnis 2.495b.25-3). Vision was a two-way exchange between seer and seen and so an exchange of glances provided a figure for intersubjectivity in general. Wrong-doers and their acts of wrong-doing were poisonous and were like poisonous snakes, because they introduced anger to the community: glares, glances, and poisonous looks or, simply, negative forms of intersubjective exchange among citizens. They were "plagues" to the community as a whole precisely because sight of them made people angry. Whereas the victim and would-be punisher were diseased because they felt anger, the wrongdoer transmitted disease because, in angering people, he upset the harmony of social relations. Anger justified punishment since, as a disease, it demanded a cure.
In Euripides' play Orestes, one of the characters gives his city the following advice on how to cure the city in respect to Orestes' pollution:
If the wife who shares his bed kills a man and the son of this one kills the mother in turn, and afterwards the one born of this one does away with murder by means of murder, where will a limit of these evils be reached? The ancient fathers handled these matters nobly: whoever was stained with blood, they did not allow to come near to the sight of their eyes, nor to encounter them — but rather required such a person to make matters holy by exile and not to exchange blood for blood (Eur. Ores. 508).
Here the speaker recommends exile as a way to deal with wrongdoing and to avoid cycles of angry vendetta. Exile is useful precisely because it removes the wrong-doer from the sight of those who are angry. As we shall see, the Athenians often used extremely violent methods of punishment in their attempts to cure the community and to restore its peacefulness, but the notion that punishment cures the community does not necessarily require a turn to violence. Tragedy itself reflects an awareness that the problem of anger can be addressed with words, and with attempts to restore friendship, as well as with exile. How then did the Athenians try to restore peace when real dramas of wrongdoing unsettled the city? How did the Athenians identify wrongdoers, negotiate the question of their desert, and then sentence them?
Part IV) How?
When it came time to punish, the Athenians acted out of anger and to cure anger, but this does not mean that they acted in anger. Rather, they interposed an extensive institutional system between the moment when an angry victim pointed to a wrong-doer and the infliction of punishment. The purpose of this system was to allow the citizens to convert a moment of private anger into a public decision crafted with a view to curing the community through a restoration of peace. But exactly how did the Athenians convert the anger of their citizen-prosecutors into public decisions about punishment?
First, it is important to grasp the basic institutional structure of the city. Athenian institutions were not neatly divided into legislative, executive, or judicial bodies as are the institutions of modern democracies. They were distinguished rather according to how citizens took part in them. Citizen men gathered en masse in the Assembly and on popular juries that never had less than 200 members and might have as many as 6000. Or they participated as individuals holding one of the 700 or so magistracies in the city. The high drama of Athenian punishment occurred when private citizens prosecuted one another and found themselves arguing their cases (and their views of law, justice, and democracy) before juries of their peers (how often does any of us ever speak before 500 of our fellow citizens, let alone 1000 or 6000?). But magistrates probably carried out the humdrum work of punishment in Athens.
Let me, then, go over their administrative duties before I return to the courtroom trials. Included among the 700 magistrates were eleven men in charge of the prison who were called, simply, "the Eleven" (see Glossary); officials like the treasurers of Athena who supervised the city's public funds; the agoranomoi who were responsible for maintaining order in the marketplace; and also the 500 members of the Boule or Council, which was the central executive body in the city and the agenda setting body for the Assembly. The members of the Council assigned among themselves jobs such as those of the thirty "cataloguers of the people" who disciplined public slaves and supervised various religious matters and also themselves supervised a group of enslaved Scythian archers who served as a minimalist police force. These magistrates (with the help of their subordinates, slave and free) could fine, arbitrate, and prosecute or preside at a trial. The Council could act as a court and impose fines up to 500 drachmae, but most officials were limited to 10 or 50 drachmae fines. Probably the same limits applied in arbitrations.
These administrative powers were, in fact, quite important. Most male citizens would have served as magistrates at some point because most magistracies were filled by random lottery, carried a one-year term, and could not be held more than once. Moreover, the cost of an adult male's food for a year was, on estimate, 36 drachmae, and the daily wage for an unskilled laborer at the end of the fourth century was 1.5 dr. so the power to fine to the tune of 10 and 50 drachmas was consequential. In contrast, even a relatively minor court case could carry a penalty of up to 1000 drachmae. Most social disruptions — petty and even not so petty crime — must have been handled not in the courts but by the ordinary citizen magistrates, and the courtrooms must have been the province of the city's elites, those who could afford the hefty fines.
Here it is worth noting that women, just like the non elite male citizens, could participate in the penal system both in trivial dealings with magistrates and as witnesses at arbitrations and as active participants in private arbitrations. They were not, however, allowed in jury trials except as defendants or as material proof, and if a woman were to wind up as a defendant in a court case, a male citizen would have to speak on her behalf. The cases that actually came to trial would have been the flamboyant conflicts among high-powered men, and so the most charged as far as their potential to affect their city was concerned. They would have been the disputes in which it was most important that the citizenry, acting as a collective body of men, confirm its authority over the establishment of norms that would govern and shape life in the city. Let us turn, then, to those judicial dramas.
How did a trial work? Here are the basic parameters: Any citizen could initiate a trial (there were no public prosecutors in Athens) simply by registering it with the magistrate under whose jurisdiction it fell; the magistrate would preside over a trial to be judged by a jury of 200+ randomly selected men who would listen first to prosecutor and then defendant and then, without any deliberation, vote by secret ballot; majority vote carried the day; a tie went to the defendant. When a citizen brought a case to court, he had a dizzying array of procedures from which to choose, but we need only pay attention to two absolutely pivotal procedural distinctions: first, these was the distinction between public suits — whether brought as a graphe, phasis, endeixis, apagoge, eisangelia, or probole — and private suits or dikai. Second, there was the distinction between two methods of sentencing convicted wrongdoers. Let's turn first to the difference between private and public suits.
Private suits or dikai (see Glossary) were heard before juries that had 200 or 400 citizens; penalties were smaller than in public cases and the bulk of the penalty was paid to the prosecutor (like damages). Nor did prosecutors in private cases risk any penalty for losing their prosecution. In contrast, public cases, and graphai (see Glossary) in particular, were heard before juries of at least 500. The penalty was much larger than in private cases. Worse still, prosecutors in many types of public case had to pay a penalty if they failed to secure at least 20% of the jurors' votes. The stakes of prosecution were thus much higher in a public than in a private case. But the most important distinction between private and public cases hinges on the method of sentencing that predominated in each case.
The Athenians had two methods for sentencing wrongdoers at trial. Either the law under which the defendant was charged prescribed the penalty or, after a conviction, both prosecutor and defendant had to propose a penalty and the jury, again without discussion, had to vote between the two options. The median voter theorem (in rational choice theory) has it that each litigant would aim to capture the imagination of the median voter and so would moderate his own extreme position for the sake of carrying the vote. The prosecutor would need to suggest a penalty that his audience would not find too extreme; the defendant, one that they would not find too lenient. We see this procedure, which was called timesis, in operation in Socrates' Apology where Socrates makes a joke of it by proposing that he be punished with free dinners for life. If Socrates had been willing to play by the rules and had made a serious proposal, what had started out as a contest between polar opposites, should have become rather an exercise in finding a middle of the road solution. After having had a chance to vent their anger and enmity, the litigants would be forced, by the procedure of timesis, to greater moderation. The anger that had inspired punishment would at last be channeled into a more restrained outcome oriented toward securing public peace and satisfying all parties, prosecutor, defendant, and community.
Significantly, the procedure of timesis bore a close resemblance to the practice of arbitration where each party had to offer a resolution proposal. Both dikai and graphai could be sentenced by either method, but the procedure of timesis was the more regular method of sentencing in dikai. For graphai it was more common to have sentences defined by law. Private cases thus bore a closer resemblance to the practice of arbitration than did public cases. Here vocabulary becomes important. The word dike was used for arbitration as well as for private cases, but its most basic and common meaning was simply, "justice." The methods employed for resolvin disputes in arbitrations and private cases thus exemplified key Athenian ideas about justice: parties would participate in the venting of emotion and then watch it reined in again, as all parties collaborated (if forced to by institutions rather than voluntarily) in achieving social cohesion by finding moderate outcomes. Thus, in private cases Athenian jurors and litigants cultivated their skills at justice understood as procedures for releasing, and then restraining, powerful negative emotions.
If private cases were associated with such general ideas about justice, what symbolic force did public cases have? A graphe was always, from its inception, a case that was heard before a court, and never before an individual magistrate. These public cases were therefore those having to do with social disruptions that, the Athenians believed, the community needed to confront as a community. Moreover, the word "graphe" means "written thing," and Athens was a society in which writing played a significant role in memorializing. Thus a conviction under this procedure was followed by the erection of official memorials of punishment, inscriptions in bronze or stone set up in public places for all to see. Public cases, then, were those, a raft of evidence suggests, that the Athenians thought needed especially to be written into social memory. Presenting such cases before large juries was a way of ensuring that the community would remember important conclusions reached about social norms.
The Athenians thus developed methods for responding to anger that either moderated it or else converted it into a public memory. But what were the penalties that they used to moderate and memorialize their anger?
Part V) What?
Here we can no longer avoid turning to the gory details. I will begin with a simple list of the penalties imposed in Athens, say a few words about the most interesting penalties, and then draw some brief conclusions about their symbolic weight.
On fellow citizens, the Athenians imposed fines, imprisonment, a set time of public humiliation in the stocks, limited loss of political rights, total disfranchisement, exile from the city (which could be amplified with the confiscation of property and/or the razing of the convict's house), and death (which could be amplified with the confiscation of property and/or the razing of the convict's house and/or a refusal of burial). Women could not, of course, be subject to a loss of political rights but they could lose their rights to participate in religious spaces and events. On the resident foreigners in their midst, the Athenians imposed all of the above penalties, with the exception of disfranchisement. As for slaves they fined masters and executed slaves, and also imposed whippings and beatings. (And they also seem to have imprisoned slaves in "mill houses" on a regular basis!).
In respect to their punishment of citizens, the Athenians have often been thought lenient. Socrates' execution by hemlock has seemed to some like a humane precursor of lethal injection. In fact, the standard means of execution was not poison but a form of bloodless crucifixion in which the convict was (probably) fastened to a board with iron collars around wrists, ankles, and neck, and the collar around the neck was tightened to strangle the wrongdoer. Socrates too would probably have suffered such a crucifixion had he not had wealthy friends. From the end of the 5th century, the Athenians seem to have been willing to let wrongdoers convicted to death use hemlock to commit suicide in advance of their execution provided they could afford to pay for the dose. It was expensive — 12 dr. a dose at the end of the 4th century — no doubt because it grew only in cold, shady, and distant spots like Susa in Asia Minor and Crete. Yet even if the bloodless crucifixion was not lenient, it did have an element of moderation. Generals on the battlefield had the authority to execute citizens and this they did with a swift blow of the sword. The purpose of the unusual crucifixion an its elaborate effort to avoid blood seems to have been to distinguish judicial punishments, and penalties in the peaceful city, from the violence of the battlefield. On some level, the bloodless crucifixion protected the body of the citizen from abuse even in death.
In contrast, the Athenians were indeed lenient in their willingness to let convicts on death row escape prison and flee into exile. Even convicted murderers, who were being held in prison while they awaited execution, were expected to make a jail break and flee the land (Plato Crit. 44b-c). And the expectation that wrong-doers would simply take themselves into exile was such that a defendant in a murder trial was given the chance, after his first speech in the trial, to leave the country if he wanted (Dem. 23.69-70). Exile wasn't the easiest burden to bear, for an exile might become "a beggar in a strange land, an old man without a city" (Ant. 2.2.0). But exiles could also re-establish themselves in another city and even, in some cases, gain citizenship in their new homes. The Athenian preference for exile over execution is the best evidence of their desire to use punishment to cure all parties to the wrongdoing. In departing the community, the wrongdoer freed the victim and the prosecutor of the anger, and put an end to the social disruption plaguing the city but he also himself gained the chance to start a new life in a context where he would not be the focus of anger and social conflict. Peace in the community was restored and the wrongdoer was also restored to life.
The single greatest difference between ancient and modern penalties is, then, the prominence of exile in the former context and of imprisonment in the latter. The Athenians did use imprisonment as a penalty but this developed out of the custom of imprisoning wrongdoers who were unable to pay their fines. Impoverished Athenians who could not pay their fines ended up imprisoned for indefinite periods of time, and over time the city seems to have developed means whereby such citizens could propose set time limits for their imprisonment, to replace their fines. But imprisonment was never one of the penalties mandated by law. Indeed, the modern rise of imprisonment as the basic sentence mandated by law coincides neatly with the near total disappearance of exile in the 19th century. Exile was still in use in colonial America and, of course, modern Australia has its roots in serving as a penal colony for Britain. But the modern prison now serves the function formerly served by exile: it allows the community to forget, almost entirely, about particular wrongdoers and so restores a sense of order to the community. Yet in the transition from a world that relied heavily on exile to one that relies most of all on incarceration, something valuable has been lost. Although prisons may help communities forget about particular wrongdoers who disappear into them, they have not achieved the second function of exile, which was to restore the wrongdoers also, allowing them to enter a new community in contexts in which they would have a new chance at life.
For all this emphasis on the Athenian use of penalties like exile that allowed them to forget all about the wrongdoer, it is important to remember that in other cases the Athenians preferred to memorialize punishments for eternity. As we have seen, the procedure of the graphe, and the inscription that would follow it, were especially used for such memorialization. The Athenians thus developed techniques for punishment that drew on the capacities of the community's memory and others that drew on its ability to forget. In general, it placed heavy emphasis on memorializing punishments in those contexts where the wrongdoing had an especially political significance (treason, temple-robbery, impiety, etc.). In contrast, when the wrongdoing primarily concerned particular individuals and their personal conflicts, the city was willing to let it go.
Part VI) Conclusion
Moments of anger gave the Athenians reason to punish. Acting out of anger, they wished to cure themselves, the victim and the wrongdoer of the trouble and unease provoked by the wrongdoing and its emotional aftereffects. In punishing, they distinguished between those wrongs that were trivial enough for a single magistrate to restore the peace and those that required the work of the community. In respect to those more significant disruptions of the peace, they again made a crucial distinction, this time between private suits in which participants were to be given the chance to cultivate their skills at justice, at venting and then restraining their emotion, and public suits in which the community was given a chance to reflect on its norms publicly and to issue a decision that would be dramatically recorded in public memory. Then, when it came to the final moment in punishment, the moment of the actual administration of the penalty, the Athenians flexibly turned sometimes to penalties that would help them remember the wrongdoer, the wrong, and the community's decision about it and at other times to willed forgetfulness. The Athenians thus employed an idea of punishment that focused primarily on a consistent recognition of the need to restore communal peace in face of a disruption. Anger led not to retribution but to restoration.
Finally, the Athenian focus on anger reveals two things about punishment. First, punishment arises simply from the desires of the punishers, which is to say, from the desires of the community. Questions of how to punish therefore involve us in asking who we want to be and what our relations to our desires are as those are expressed by anger. The Athenians dramatized this idea once a year at a festival called Thargelia where they administered a "cure" for themselves as a whole city. The festival involved an especially violent ritual. It was said that the Athenians had once killed a Cretan man named Androgeos and had afterwards repented of their own act of wrong-doing. Every year thereafter, to deal with the problem of the city's guilt and implication in the murder, they drove two undesirable members of the community out of the city in rituals resembling stonings. Such a scapegoat was called a pharmakos. It is related to the word pharmakon which means both medicine and poison and from which we get "pharmacy" and "pharmaceutical." The ambiguity of the word pharmakon reveals two things, the first being the paradoxical nature of punishment as viewed from the Athenian perspective. Punishment forces a community into facing the idea that acts of violence are expected to cure a community that otherwise disavows acts of violence.
The second feature of punishment revealed by the ritual requires that we know a bit more about it. The festival of Thargelia marked the end of the year, and the day after saw a festival marking the beginning of the new. The Athenians moved through the year knowing that they would conclude the year with a moc stoning, in which they would mimic a communal act of passion and admit to the communal desire to inflict harm. This they would do in order to cure themselves. The festival dramatized the certainty that the community's rules against violence would eventually break-down; it dramatized that everyone was mutually implicated in the break-down; and that everybody was mutually implicated in a system for restoring order which inevitably treated certain citizens as means to the ends of other citizens. The festival was an admission that the origin of punishment in anger implicates all citizens in a set of disordered relationships, which must be restored but which can be restored only by a process that imposes itself on different citizens in different ways and to different degrees. We can and should deplore the violent means that the Athenians used to make such an admission, but surely the admission itself is an important one: dealing with wrong-doing and punishment requires that we think about the community's desires and the problem of anger within the community and about how best to respond to those desires and that anger.
Sophocles gave an accurate description of Athenian politics when, in the "Ode to Man," of the Antigone, he had a chorus praise humankind for having taught itself political skills that include not only judgment and voice and wind-swift thought but also "constitutional anger," an anger that both regulated the city and was regulated by it (354-55). Punishment may originate in anger, but one need not satisfy it in order to resolve it and restore peace.
Allen, D. The World of Prometheus: the politics of punishing in democratic Athens. Princeton University Press.
Hart, H. L. A., Punishment and Responsibility. Oxford University Press.
Hunter, V. Policing Athens. Princeton University Press.
Keuls, E. The Water Carriers in Hades: A Study of Catharsis through Toil in Classical Antiquity.
Lloyd-Jones, H. The Justice of Zeus. Berkeley.
Mackenie, M.M. Plato on Punishment. Cambridge.