The only direct source of evidence which we possess for the historical murder-laws of Attica—for the murder-laws of other Greek States we have no direct evidence at all—is a fragmentary inscription of the year 409-8 B.C., containing a few lines written in the old Attic alphabet, which, though ‘restored’ in a manner sufficiently satisfactory to render it trustworthy and intelligible, gives us nevertheless the most rudimentary information about the Attic murder-code. The real value of this inscription has been indicated by Lipsius.[1] The fragments of laws which are found on the inscription are so closely identical with the corresponding portions of the Draconian laws as they are cited by Demosthenes that they must, he says, be regarded as furnishing a convincing proof of the validity of the remaining laws which Demosthenes has cited. Now, these remaining laws are written in the Ionic alphabet, which was used by Athenian writers in the fourth century and in the latter half of the fifth century B.C., and it so happens that the date of the change in the alphabet used in Attic inscriptions, namely, the year 403-2 B.C., was also the date of what Glotz[2] describes as ‘la grande révision législative qui signala l’archontat d’Euclide.’ If, then, any changes occurred in Attic murder-law, in the period which elapsed between Dracon and Eucleides, it was probably in the year 403-2 B.C. that such changes were finally incorporated in the written code. We shall see that there is no reference to the Areopagus in our fragment, but it may of course have been mentioned in the missing portion of the inscription which referred to wilful murder. Pollux and Plutarch[3] state that the Areopagus was created by Solon. We shall see later[4] what elements of truth this dictum may contain. The only change which we believe to have been made in the period from Dracon to Eucleides was the isolation of the Areopagus from the general list of the Ephetae courts.[4] This change we attribute to Solon, and with this exception we accept the murder-laws which are quoted by Demosthenes as the original code of Dracon. We have already[5] argued against the theory of Glotz that the clause μηδ’ ἀποινᾶν was a Solonian innovation. The alteration which we attribute to Solon was not properly speaking a change in the murder-code, but merely a change in the distribution of pleas in the judicial system. Hence we accept the ancient tradition[6] that ‘Solon changed all the laws of Dracon except those relating to homicide.’ The Solonian legislation was less severe and more humane than Dracon’s code. If Solon did not alter the murder-laws, it was probably because they were, so to speak, so non-Draconian, because they did not bear the stamp of Dracon’s own peculiar genius. They were, we have said,[7] an eclectic codification of the unwritten laws of the Ephetae and the Exegetae. Religiously consecrated by their joint tribal and Apolline inspiration, they stood above the gales of Athenian political ferment. It was only in the personnel of the judicial system that a loophole was left open for political intrigue. In this respect alone was alteration easy and obvious: and in this respect alone do we suppose that alterations took place.

The original inscription of 409-8 B.C. consists of forty-eight lines, of which six are undecipherable, and nine others badly mutilated. We will give here just four of the best lines, from which the condition of the remainder may be inferred.

The most important portion of the inscription, as restored by Köhler, is given by Dareste,[8] and transliterated into the Ionic alphabet reads as follows:

This inscription has been restored, mainly, from quotations in the speeches of Demosthenes. But before attempting to translate it, we must point out that even in its restored form the inscription is archaic and obscure, and the meaning is not always certain. The first half of the inscription seems to refer to involuntary or accidental homicide. But the end of the second line, as it stands, cannot possibly be taken to refer to accidental homicide, because the verb βουλεύειν usually means ‘to plot’ or ‘to resolve,’ and therefore implies an element of deliberation. Wilful murder is not expressly mentioned, save in so far as the slaying of a homicide exile abroad is decreed to be equivalent to murder. There is also a reference to justifiable homicide in self-defence. But most of the fragment consists of an enumeration of the persons who are by law entitled to share in the acceptance of gifts of ‘appeasement’ from an involuntary slayer: and of the judges by whom the various kinds of homicide must be decided. We do not believe that any judicial distinction is intended in the use of the two verbs δικάζειν and διαγιγνώσκειν. Both words mean, we think, ‘to adjudicate.’ There is no question of preliminary investigation as distinct from final decision. In regard to the second line, the restoration ὡς βουλεύσαντα can only mean ‘on the ground of having plotted (to kill).’ Did the restorer mean by this clause ‘attempting murder’ (when death did not ensue) or ‘contriving murder’ (when death did ensue)? The noun βούλευσις can have both these meanings, but the verb βουλεύειν cannot, we think, denote attempts to kill. If the restorer meant ‘contriving murder,’ such an interpretation is open to the following objections: (1) ‘contriving murder’ ranked with wilful murder in Attic law, and was tried by the Areopagus, not by the Ephetae[10]: (2) it is rightly maintained[11] that the presence of καί at the beginning of the inscription indicates that a portion is missing, and it is natural to assume that this missing portion contained the law relating to the graver kinds of homicide, including not only wilful murder, but also contriving murder. In order to obviate such objections, Philippi abandons the verb βουλεύειν and proposes to read βουλεύσεως τὸν ἀεὶ βασιλεύσαντα, ‘the King-archon for the time being shall judge concerning attempted murder.’ But this suggestion is open to the following objections: (1) we are compelled to render διαγνῶναι (l. 13) ‘to adjudicate finally,’ and we do not think that it bears this meaning in the inscription; (2) τὸν ἀεὶ βασιλεύσαντα is a very questionable Greek rendering for ‘he who is King-archon for the time being’: (3) while it is true that attempted murder was tried, in Aristotle’s time,[12] in the Palladium, it has no real affinity with manslaughter. It is impossible to suppose that the ‘appeasement’ mentioned in the inscription could have ever been applied in cases of attempted murder. It would be absurd to compensate relatives who had lost nothing, and to ignore the person on whose life the attempt was made. Demosthenes definitely cites[13] this law of ‘appeasement’ as referring to manslaughter. Hence, as we believe that the second line of our inscription refers to manslaughter and as λευσαντα (sic) is found in the unrestored part of the inscription, we propose to restore μὴ βου]λεύσαντα instead of ὡς βου]λευσαντα and understanding κτεῖναι with αἰτιᾶται, we translate ‘if anyone accuses a person of slaying without deliberate resolve.’

We will now suggest a translation of this passage, reading μή instead of ὡς in the second line.

‘And if a man slays a man not with intent (to kill), let him be put on trial (φεύγειν), and let the “Kings” judge of the causes of death, or, if anyone accuses a person of slaying without deliberation (μὴ βουλεύσαντα), let the Ephetae adjudicate. And the “appeasement,” if there is a father or (and) brother(s) or (and) sons (of the slain), let all (accept) or let one objector hold the field: if there be none of these, let (the “appeasement” extend) to cousinship and cousins, provided all consent to be “appeased” having sworn the (customary) oath: if there be none of these (i.e. cousins) and if the man slays involuntarily, and the Fifty-one, the Ephetae, decide that he slew involuntarily, let ten phrateres permit his return from exile,[13A] if they (all) agree, and let the Fifty-one select these (ten) according to birth (or rank or merit—ἄριστίνδην), and let (all) previous slayers be bound by this law: ... and if any person slays a manslayer or causes (i.e. plots) his death while the manslayer abstains from the boundary markets and from Amphictyonic games and festivals, let him be liable to the same penalty as if he had slain an Athenian (citizen): and let the Ephetae judge the case: ... it is lawful to kill manslayers or to arrest them, in our territory (ἡμεδαπῇ) but it is not lawful to torture them or to amerce them: the fine payable shall be twice the amercement: ... if any person slays on the spur of the moment in self-defence a man who tries by violence unjustly to rob and plunder him, let his act of bloodshed go unpunished.’

Philippi[14] finds the reference to ‘the kings’ in this inscription rather difficult to explain. He thinks that the allusion can only be to archons, but he feels also that ‘it seems inadmissible to assume collegiate functioning after the archonship became annual.’ He therefore views with sympathy the extraordinary suggestion of Köhler, that τὸς βασιλέας means τοὺς ἀεὶ βασιλεύοντας, that ‘the kings’ are ‘those who from time to time held the office of king-archon.’ The solution of this, as of other difficulties in the inscription, is, we believe, to be found in a correct analysis of the word ἄκων, which means ‘involuntarily’ or without intent.

Let us suppose that a man A caused the death of another man B. Obviously this event could occur either (1) in an accidental manner, without the least possible foresight or culpable neglect, as for instance in a wrestling-match or in a javelin-throwing competition: or (2) in circumstances which implied a certain amount of culpable neglect, or ἀφυλαξία, because the slayer did not take the usual or the necessary precautions—as, for instance, if a drug was administered, in illness, to B, and A did not see to it that the drug was of the proper kind: or (3) in a manner which involved a certain amount of intent or deliberation, though not necessarily ‘malice aforethought,’ on the part of the slayer, as, for instance, if A struck B in a drunken bout, or in a sudden fit of anger, jealousy or revenge. Plato,[15] in the Laws, makes the clearest possible distinction between these cases, and so does Antiphon[16] in his Tetralogies. But the Greek words ἄκων and ἀκούσιος were applied indiscriminately to all three cases!

The Greeks of historical times actually put on trial inanimate objects which had slain a man. Why? Was it because these objects were regarded as polluted and it was necessary to discover the extent of the pollution? We do not think so, for such objects were either polluted or they were not. There could have been no question of degrees of pollution. The purpose of such a trial was rather, we think, to inquire whether the objects were guilty or not. But why was this question of such importance? Clearly because there was a human, as distinct from a divine, interest in such trials. We suggest that these trials were instituted primarily in order to establish the innocence of an accused man. In Greek law, unlike modern law, it was necessary for a man to prove his innocence. He could only do this, very often, by proving that somebody else, or something else, was guilty. We do not agree with Müller[17] and Philippi[18] in regarding these trials (δικαὶ ἀψύχων) as sham trials. Presided over by five ‘kings,’ as Aristotle[19] assures us that they were, they cannot have been so altogether meaningless and absurd. They were, we think, almost as important as a modern Coroner’s inquest. Now, who, we may ask, were the five ‘kings’ who sat at the Prytaneum ‘murder’ court in the time of Aristotle? They were, simply, the King-Archon, and the four Phylobasileis, or Tribe-Kings, who still survived as the religious and judicial representatives of the old Ionian tribes of Attica. These kings are therefore the aristocratic descendants of the Elders who ‘sat on smooth stones in a sacred circle,’ in the Pelasgian Age.[20] The Prytaneum, as Glotz[21] points out, was the oldest court at Athens. Coulanges[22] connects this court with the worship of the ancestral-hearth; it was, he thinks, the divine ‘hearth-stone’ of the nation, the source of its vitality, the symbol of its immortality. Yet this court Müller and Philippi regard as a mock or sham-court, in which a number of respectable but unintelligent nobles persisted in upholding the obsolete traditions of a ridiculous past!

We believe that ‘the kings’ of the Prytaneum Court are identical with ‘the kings’ of our Draconian inscription. The first two lines of the fragment refer, in our view, to accidental slaying, in which there was no degree of guilt attaching to the human agent, but in which it was necessary to prove that the guilt was attached to an animal or an inanimate object. We think it quite probable that such cases were tried at the Prytaneum.[23] We may go so far as to say that such cases were the raison d’être of the survival and the historical importance of such a court.[24] The legislator, in our inscription, says: ‘If a man slays another without intent, let him be put on trial, let “the kings” judge of the causes of death’ (δικάζειν αἰτιῶν φόνου).

We shall discuss[25] later the function of the Ephetae judges who are mentioned in this inscription as collaborating with the kings in the judicial investigation of homicide-guilt, and we shall suggest an explanation of the fact that they were invariably fifty-one in number.[26] In regard to the adverb ἄριστίνδην, which means, in general, ‘according to excellence,’ we agree with Philippi[27] that in the context it refers to birth rather than to social rank. The selection of the phrateres would probably have been made from ‘brethren’ who were not kinsmen of the slayer, but merely related by ties of ‘affinity’ or of local contiguity with him. Plato[28] suggests that in certain cases of homicide the judges of guilt (and probably therefore of atonement) should not be akin to the criminal. The fact that the father and the brothers (we assume that the singular form ἀδελφός includes all the brothers) and the sons of the slain could, if unanimous, have accepted ‘appeasement’ and have legalised the manslayer’s return from exile, shows how far from, and yet how near to, the wergeld customs of Pelasgian days were the historical murder laws of Greece. Yet here we have not wergeld proper, but only a survival, a reflection, of its ancient vigour. Nothing could show more clearly than this law does the validity of our theory[29] which finds in a ‘compromise’ between different forces the origin of the historical homicide-code of Greece.

Glotz[30] holds that the objection of a single relative to ‘appeasement’ could neutralise the will of the other kinsmen because, if he were obdurate, he could prevent the unanimity which was required by law for such return. But we shall argue, later,[31] that while the relatives had considerable legal powers if they were unanimous, they were probably subject to superior control if they disagreed. It is difficult to suppose that one bitter enemy amongst the relatives of the slain could, in practice, have imposed a penalty of perpetual exile for manslaughter.

We have discussed[32] the theory of Glotz that ‘private settlement’ was legal, even for wilful murder. How can Glotz reconcile such a theory with this Draconian law which provided for[33] a trial and a verdict even in cases of accidental slaying? According to our interpretation of the restored inscription, the relatives of the slain may not always agree, but the kings and the Ephetae must adjudicate in each case.

Glotz suggests,[34] further, that Dracon first introduced the distinction between murder and manslaughter. Is this the view which is suggested prima facie by the restored inscription? To us it seems quite obvious that the inscription assumes, as a familiar fact, an already existing distinction, not merely between murder and manslaughter, but also between manslaughter and accidental slaying. If the distinction appeared as a legal innovation in the Draconian legislation, surely such a distinction would have received some emphasis, since it would have been necessary to enlighten an uncivilised public opinion; surely the definitions of the various kinds of homicide would have been more clearly marked and the penalties more clearly indicated.

Since the Draconian inscription has been restored from quotations in Demosthenic speeches, we shall turn to those speeches for a more complete account of Attic homicide law. But the Demosthenic references must be supplemented from other sources—especially from Plato’s penal code.

Homicide Laws in Plato and Demosthenes

If we accept the opinion of Coulanges[35] that the synoekism of Attica did not abolish the local prytanies and magistracies, it will be readily conceded that the Athenian city courts, that is, the Attic State courts, did not necessarily adjudicate in all cases of homicide. Owing to the civic and religious aspect of wilful murder and kin-slaying—crimes which involved the penalty of death or the confiscation of property—we may feel certain that the State courts had exclusive jurisdiction in such cases.[36] But we cannot be sure that the same principle applied to manslaughter and minor degrees of guilt, except when such deeds occurred between parties who had only one civic bond between them, namely, the political union of the State. Most frequently, we admit, the parties involved would be of such a kind. The rise of political democracy and of a new nobility of wealth led to the accumulation, in the cities of Attica, of a vast multitude of persons who did not belong to any of the old tribes or religious corporations.[37] The common worship of the clan, the phratry and the tribe did not receive their allegiance. Hence, probably, the courts of such organisations would not, even if they could, adjudicate in their case. But there survived in Attica, all through the historical era, families who still belonged to these more primitive groups. They were the old nobility, the country gentry, scattered over rural Attica,[38] who continued to obey and, where possible, to exercise the old jurisdictions of the clan, the phratry, and the tribe. We have shown that local tribal courts still functioned, with State-sanction, in historical Attica.[39] We have quoted a passage[40] from Plato which suggests that some such local courts had power to condemn to death a person who maliciously wounded one of his parents.

We agree with Müller[41] and Coulanges,[42] in opposition to Glotz[43] and Philippi,[44] in the view that Plato’s Laws are based, in the main, upon the Attic legal codes. There are certain points in which Platonic law seems independent of Attic law. Are these variations to be attributed to the fancy of an idealist or are they rather a supplement, an incorporation of local and tribal laws which the State codes did not mention but always presupposed? So far as homicide at least is concerned, we prefer the second alternative: and we shall give at a later stage the reasons for our preference.

In describing the trial of inanimate objects and of animals which were guilty of human bloodshed, Plato says[45]: ‘If a beast of burden or any other animal shall kill any person (except in a public contest) let the relatives (of the deceased) prosecute the cause of death: and let the wardens of rural areas (ἀγρόνομοι) upon whom ... the relatives shall impose this task, decide upon the matter: and let them destroy the animal (if) condemned and cast it beyond the boundaries (of the State). If any inanimate object deprives a person of life (except lightning or such god-sent bolt ...) either by the person falling upon it or by its falling upon the person, let the nearest of kin appoint the nearest neighbour to act as judge, and (thus) free from pollution himself and his whole kindred, and cast the condemned object beyond the boundaries.’ There is no mention of the Prytaneum Court or of the Tribe-Kings. We can explain the omission by supposing that Plato is referring to local courts and local cases of bloodshed, in which the relatives had not to go outside their immediate neighbourhood to obtain jurisdiction. The ancient phratry was an assembly of local clans: neighbourhood was the essential factor in the bond which the phratry religion represented. The ‘nearest neighbour’ in this quotation would have been a member of the phratry, if not of the clan, to which the slain person belonged. The duty of prosecution which is here referred to was no sham duty[46]; it was a serious religious obligation. Failure to prosecute would have ‘polluted’ the relatives of the slain.

So far there is no question of any human guilt. But such a question might have easily arisen. In the Hebrew murder-code,[47] if an ox gored a man to death, it was necessary to inquire whether the ox had been ‘let out’ by the owner, and whether the ox was previously ‘known to be dangerous.’ If so, the owner could have been put to death, unless he ransomed his life. Let us suppose, furthermore, that the object had not ‘fallen,’ but was such that it must have been ‘thrown.’ Two cases might now arise: (1) the ‘thrower’ might confess that he threw the object, say, a stone or a piece of wood, but at the same time deny that he threw it with the intention of hitting, much less, of killing, any person: or (2) the ‘thrower,’ guilty of intent to kill, might escape undetected, perhaps concealed by a wall or a boulder or a shrubbery, from which he had hurled the fatal missile. Thus, the trial of inanimate objects, and also, but to a less extent, the trial of animals, might have had a close connexion on the one hand with the question of accidental homicide, committed by a human agent, and on the other with the question of ‘murder by persons unknown.’ Upon the precise circumstances of each case would have depended the question whether local magistrates and tribunals would have possessed jurisdiction in the matter, or whether it would have had to be referred to the central State authority at Athens. But to this same authority would naturally also have fallen the decision as to the guilt of animals or objects which had caused the loss of human life within the city of Athens and its environs: and hence we can understand why the central Prytaneum court had to adjudicate not only upon guilty animals and inanimate objects, but also, and with much more serious possibilities, upon murder by persons unknown.

In the case of objects which could only have proved fatal if they were thrown by a human agent, a verdict of acquittal, in regard to such objects, would have logically involved a verdict of murder by persons unknown; for, if we suppose that the object was accidentally thrown, it is probable that the thrower would have come forward and established the blood-guilt of the object concurrently with his own innocence. Demosthenes[48] says in regard to the Prytaneum court: ‘If a stone or piece of wood or iron or anything of the kind falls upon and strikes a man and we are ignorant who it was that threw it, but know and have in our possession the instrument of death, proceedings are taken against such instruments here.’ Plato asserts that the objects mentioned were prosecuted by the relatives of the slain: but may we not also assume that a man who had thrown one of these objects without malicious intent, and who was accused of murder or manslaughter, would have lodged an accusation against the ‘object’ at the preliminary inquiry[49] before the King-Archon, that is, at the Prytaneum? If the Prytaneum found the object guilty, would not the verdict have prohibited any further proceedings? If, on the other hand, the object was clearly hurled by a human agent with malicious intent, and if the agent was unknown, proceedings, of a most formal kind, were taken against the unknown slayer.

Similar proceedings would of course be taken if there was no ‘object’ involved, as, for instance, in case of death by strangling. Such proceedings are thus described by Plato[50]: ‘If anyone,’ he says, ‘is found dead and the murderer is not known, and is not discovered by careful search-parties, let there be proclamation against the murderer as in other cases, and let the heir-at-law (i.e. the nearest relative of the deceased) proclaim in the market-place that the murderer, whoever he is, must not, since he is guilty of bloodshed, set foot in any sacred place in his native State or in that of his victim, or if he does, and he is discovered and identified, he shall be put to death and cast unburied beyond the boundaries.’ We have already pointed out[51] that the object and purpose of trials for homicide in Greece was not so much the establishment of guilt, as it is in modern States, but rather the establishment of innocence. Now, our last quotation from Plato suggests that a man who came to be suspected of homicide some time after the crime was committed, and who was never formally prosecuted and convicted, could, nevertheless, be put to death! But we shall see[52] that one refuge still remained to the ‘unfortunate wretch.’ He could have pleaded innocence, in the presence of the avengers, and this plea compelled ipso facto a recourse to trial: he could of course be arrested on the spot and imprisoned, but he could challenge a verdict at a court of summary jurisdiction, the prison court, known as ‘the Eleven,’[53] and if he proved his innocence to the satisfaction of more than four-fifths of his judges, his accuser paid a fine of one thousand drachmae! Thus, he could not be slain on the spot by the avengers if he pleaded innocence: but unless he proved that he was innocent he was ultimately put to death.

Aristotle may be taken to suggest that there was no essential connexion between the trial of inanimate objects and the verdict of murder against a person unknown. He says[54]: ‘if the name of the homicide is unknown, the indictment is prosecuted in general terms against the unknown author’; but in the next line he adds: ‘The King-Archon and the Tribe-Kings have competence in indictments against lifeless objects and the brute creation.’ The juxtaposition of such references is sufficiently significant. Pollux[55] is more definite: ‘The Prytaneum court,’ he says, ‘adjudicates concerning slayers if they are unknown, and also concerning lifeless objects that have fallen and caused death.’

Though none of these authorities say anything to support, neither do they say anything which refutes, our opinion that the Prytaneum court could also try cases of ‘accidental’ slaying in which a person accused of manslaughter pleaded an entire absence of neglect[56] or passion or intent. Our view is however rendered probable by the fact that preliminary inquiries in homicide cases were made in this place which, in addition to being a court, was also the official residence of the King-Archon and of the Prytaneis[57]; but the most cogent argument in favour of our hypothesis is to be found in the first two lines of the Draconian inscription if our interpretation of these lines is correct. We fail to see how the Draconian reference to ‘kings’ as ‘the judges’ in cases of homicide committed ‘without intent’[58] can be otherwise satisfactorily explained.

If it be objected that pleas of ‘accidental’ homicide were regularly tried at the Palladium court,[59] we may reply that the proper function of this court was subsidiary or supplemental to that of the Prytaneum. In the Palladium the accused in his plea denied, indeed, any guilt, but he would have found it difficult to prove his innocence unless he could transfer the guilt to another person. In the Prytaneum, as we conceive it, he had often an opportunity of laying the blame upon an inanimate instrument of death. Such a plea of accidental slaying involved no question of human guilt, as the accusation was centred upon an inanimate ‘object.’ Again, whenever the plea of the accused differed from the charge of the accuser, it was the duty of the King-Archon to decide on the probabilities of the case, before he relegated the trial to its appropriate court.[60] If, then, a person accused of murder or manslaughter could advance a plea of accidental homicide by accusing an inanimate object, the Prytaneum court adjoining the official residence of the King and the Prytaneis would have been at the immediate disposal of the defendant. No long period of time, such as ordinarily had to elapse between formal accusations and homicide trials, preceded the trials at the Prytaneum; and we may infer from Plato’s account that the verdict of ‘death by persons unknown’ was normally brought in by the Prytaneum court before any formal proclamation of the unknown murderer was made by the relatives of the slain.

Involuntary Homicide

In regard to pleas and charges of manslaughter, we hope to show that there is a very substantial agreement amongst the ancient authorities. Once more[61] we must call attention to the possibility of local as distinct from central jurisdiction. Demosthenes[62] quotes a law of Dracon relating to the ‘appeasement’ of the relatives of the slain, which is practically identical with the law which we have quoted from the restored inscription. ‘Proclamation to (or against) the slayer shall be made in the market-place (by all relatives of deceased) within the degrees of cousinship and by cousins; in the prosecution there shall act jointly with these, the sons of cousins, the sons-in-law (γαμβρούς) and the fathers-in-law (πενθερούς), the cousins-in-law, the sons of such cousins and the phrateres. If “appeasement” is prescribed (δέῃ), if there is a father or (and) brother(s) or (and) sons, let all (these) be appeased or let one objector hold the field: if there are none of these, and (the accused) slays involuntarily, and the Fifty-one, the Ephetae, decide that he slew involuntarily, let ten phrateres decide about appeasement, if (all) consent. These let the Fifty-one choose according to birth (or merit).’ We give below[63] the Greek version of the latter portion of the law, so that it may be the more easily compared with the corresponding portion of the Draconian inscription. In this inscription, there are two lines which are not found in Demosthenes, namely those which refer to the rôle of the ‘cousins’ in accepting ‘appeasement.’ We must not, however, conclude that the cousins had ceased to have a voice in ‘appeasement’ in the time of Demosthenes, or from the year 403/2 B.C. onwards, or in Solon’s time. We are convinced that the omission is due either to the negligence of a scribe or to the deliberate excision by Demosthenes of unnecessary elements of law in a legal quotation which included extracts from different laws, most of which are only remotely relevant to his main purpose in the speech. It would be absurd to suppose that a legal innovator jumped from the ‘small family’ to the neighbour-brethren (phrateres) and ignored the cousins in an enactment involving the transfer of property which constituted ‘appeasement.’ Surely if any change were made in the personnel of the recipients, the ‘neighbours’ would have been first omitted. And we cannot suppose that cousins had become obsolete since Dracon’s time!

The formal proclamation of a charge of manslaughter against the accused was the initial act of the ‘prosecution’ which, after a period of inquiry, after examination of witnesses, and after various other formalities, ultimately culminated in the formal trial of the accused at the Palladium court. But, as it stands, this quotation from Demosthenes suggests, prima facie, that trial could be dispensed with if the deceased had near relations who unanimously consented to accept ‘appeasement’: and that it was only in the absence of relatives that a trial took place, after which the phrateres, who were merely neighbours, negotiated the appeasement. But this prima facie inference arises from the clumsy and unscientific wording of the law. That the inference is logically invalid is obvious from the simple fact that, in the absence of relatives of the deceased, the slayer could not be tried at all! When the law says ‘if there are none of these,’ it must be taken to mean ‘if none of the groups which are privileged to decide about appeasement can be brought to unanimity.’

It is an extraordinary thing, that in this Demosthenic citation of the law relating to manslaughter there is no certain reference to the penalty of exile. Are we to assume that such a penalty was not legally compulsory, that it was merely a fortuitous eventuality which depended entirely on the attitude of the relatives to ‘appeasement’? Are we to suppose that if all the relatives concerned agreed to be ‘appeased’ immediately after the trial and the verdict, the manslayer could have remained at home precisely as in the old wergeld days? We have no doubt that so far as the relatives of the slain were concerned, he could have remained at home. But could he have been admitted to purgation? Was he not ‘polluted’ if the dying man did not ‘release’ him? Could he have ignored the anger of the gods and of the slain? The laws of Dracon do not directly assist us in answering these questions: on the contrary, by their obscure wording they suggest frequently the wrong answer. But we have seen[64] that these laws can only be explained as a ‘compromise.’ In the wergeld system of tribal Greek societies in pre-historic days, there was a regular and scientific method of ‘appeasement’ which, in most kinds of homicide, was recognised as a solvent of the feud. But in the Draconian code ‘appeasement’ appears in a degenerate and insignificant aspect. It is subordinated to other penalties which are not stated with any degree of emphasis, for the simple reason that they were universally familiar. All the arguments which we have put forward in support of our theory of a ‘compromise’ in Attic law compel us to assume that exile was an essential ingredient of the penalty for manslaughter. Such an assumption is implied in the reading ἐσέσθων (let them permit to return) occurring in the Draconian inscription. Demosthenes, unfortunately, has αἰδεσάσθων, which refers merely to ‘appeasement.’ As we should have expected, Glotz[65] and Müller[66] interpret this Demosthenic reference as if it were a logical scientific document: and they accept the prima facie inference that a person accused of manslaughter could, as soon as he was publicly proclaimed and banned from all public and religious intercourse, avoid the ordeal of a trial and the punishment of exile by simply taking some money with him to the house of the father, brother and sons of deceased; if he succeeded in securing a ‘settlement’ and procured a ‘legal release,’ he could have quietly resumed his ordinary occupations! This interpretation, which we have already rejected,[67] is inconsistent with other passages in Demosthenes and in Plato which we shall now discuss. While we admit that this law of Dracon does not, unfortunately, mention the exile penalty for manslaughter as an obvious and incontrovertible fact, yet we insist that it does mention trial as a normal concomitant. The Ephetae are there, first and last. The Ephetae must decide the degree of guilt: they must decide that the slayer slew involuntarily: they must in the absence of relatives or in the event of their disagreement select the ‘phrateres’ according to birth or merit. This at least is very different from ‘private settlement.’

Demosthenes[68] quotes another law of Dracon regarding manslaughter, as follows: ‘If anyone shall pursue or plunder beyond the civic boundary any of those slayers who have gone into exile and whose property is not confiscate to the State, he shall incur the same penalty as if he did so inside our boundaries’ (ἐν τῇ ἡμεδαπῇ). Fortunately we possess Demosthenes’ explanation of this law which, because of its peculiar expression, requires some such explanation. The word ἐπίτιμα, in reference to property, is opposed to ἄτιμα and means ‘not confiscated.’ Hence, the phrase ‘Slayers whose property is not confiscated’ must refer, says the orator, to ‘involuntary slayers,’ because the property of wilful murderers is confiscated to the State. Thus this Draconian law, instead of employing the adjective ‘involuntary’ (ἀκούσιος) as a predicate of ‘slayers,’ uses two clauses to describe what a single adjective would have described. Are these two clauses, then, to be regarded as definitive; as concerned with qualities which normally and universally characterised involuntary slayers? Are involuntary homicides, as a class, defined as ‘those manslayers who have gone into exile and whose property is State-guaranteed’ (ἐπίτιμα)? Or are we rather to suppose that there were two classes of involuntary homicides, and that this law refers to only one of these classes—that in some cases, as Glotz and Müller conceive the matter, the slayer bribed the relatives of the slain, and avoided all further trouble; and, in other cases, he went into exile? In our opinion this quotation suggests that all involuntary slayers went into exile for a period of time. Müller holds[69] that the duration of this period of exile was not fixed by any law: that the slayer remained in exile until such time as the relatives accepted ‘appeasement.’ We shall discuss this opinion more fully later, but we may say here that it seems very strange that the State should have guaranteed protection for the property of the slayer, and should, at the same time, have had no voice in determining the limits of his period of exile, no influence in constraining the relatives of the slain to accept ‘appeasement.’

Speaking of involuntary homicide, in another passage, Demosthenes says[70]: ‘If the accused be convicted and be found to have done the deed, neither the prosecutor nor anyone else has control over him, but the law alone. And what does the law command? That a person convicted of involuntary homicide shall on certain stated days leave the country by an appointed road and remain in exile until he has appeased certain[71] of the relatives of the slain (τινα τῶν ἐν γένει τοῦ πεπονθότος): then it permits him to return, not anyhow, but in a particular manner, ordering him to sacrifice and be “purged” and giving other directions which he must carry out. Rightly, men of Athens, does the law prescribe all this. It is just to make the penalty of involuntary homicide less than voluntary, and it is right to prescribe exile guaranteeing (a person) a secure exodus, and for the returning exile to free himself from tabu and be cleansed by customary rites; above all it is right that the laws should control everybody and everything.’ In this passage we find the usual obscurity of language and even apparent discrepancies.

Is it suggested that if the manslayer is not accused and convicted, the law has no control over him? Glotz and Müller would find in such quotations a proof of their theory of the legality of ‘private settlement.’ But it is absurd to examine as it were microscopically such passages as this. They must be interpreted, as far as possible, in the light of other parallel references, and accepted or rejected according to the criterion of consistency. We admit of course that Demosthenes is not always consistent; he was essentially an orator, and as an orator he placed rhetoric before logic, persuasion before truth. But in legal quotations he had to respect the legal knowledge of his audience. Hence such quotations contain of necessity an important element of truth. In the passage which we have just cited there is an apparent discrepancy which militates somewhat against its logical value. We may ask: ‘How can the law be said to be master of everybody and everything if it guarantees to the relatives of the slain the right to refuse “appeasement,” even if there be only one dissentient?’ A law of Dracon prescribed that ‘all must agree or let one objector hold the field.’ Was not this objector, then, κύριος τοῦ ἀνδροφόνου? What control had the law over such an objector? On the very face of it, therefore, this statement of Demosthenes seems inconsistent with itself! But perhaps Plato will help us to solve the problem.

We have already[72] quoted Plato’s account of the penalty for manslaughter. The legend, which he mentions, ‘of priests of old’ concerning the temporary anger of the dead shows the religious significance which the exile penalty possessed for Plato: he understood the meaning of the ‘customary rites’ of cleansing and purgation which the manslayer had to perform on his return. In his penal code, Plato differentiates between different degrees of guilt in involuntary homicide: and it is significant that the penalties vary correspondingly—not in the extent of the ‘appeasement,’ but in the duration of the period of exile. Thus he says[73]: ‘If anyone kills a freeman in a passion, let him be of necessity an exile for two years.’ In this case there is an element of guilt, but there is no deliberation or intent to kill. He goes on to say: ‘He who in a passion but with a certain degree of intent (μετ’ ἐπιβουλῆς) slays a person, ... let him be an exile for three years ... being punished during a longer period because of the greater seriousness of his passion.’ ‘It is difficult,’ he continues, ‘to give laws on such matters with accuracy. Of all such matters, therefore, it is right for the guardians of the laws to have cognisance: and when the period of the exile shall have expired for each offender, it is right to send twelve judges to the civic boundaries who having considered still more clearly meanwhile the condition (or conduct—πραξεῖς) of the exiles, will be the final arbiters (δικαστάς) of the “appeasement” and their return home from exile: and let them abide by the decisions of these magistrates; and if, after returning from exile, anyone of these commits again the same offence, let him be exiled and never return: if he returns let him suffer in the same way as if a stranger returns’ (κατὰ τὴν τοῦ ξένου ἄφιξιν). Here we have a very different picture from that which the theories of Glotz and Müller and some Demosthenic passages suggest. There is question of manslaughter, but there is no reference to the power of wrangling relatives to prevent the exile’s return. On the contrary, it is stated that the ‘appeasement’ was controlled by judges who may have been phrateres, but were probably not kinsmen of the slayer. The last line in the passage refers to a law which we have already[74] mentioned, namely that which decreed perpetual exile for manslaughter committed between strangers in any given State. The penalty for ‘returning’ in such a case, that is, for rupture de ban, was death.

In a passage which refers to a case of kin-slaying, in which the dying man ‘forgave’ his slayer (without, however, granting a ‘release’), Plato says[75]: ‘If any person of his own accord absolves anyone for such a deed, let the purgations be made for the slayer as if his act had been involuntary, and let one year be the term of his absence from the country according to law.’ The theory of the legality of ‘private settlement,’ before or after trial, cannot be reconciled with this quotation. The phrase ‘according to law’ suggests that Plato refers to actual Attic law, and not to an ideal law of his own creation.

Plato adds that in such a case the slayer can never resume his ordinary domestic life, even though he recovers his civic status. Similarly, for the slaying in a passion of a husband by his wife, or of a wife by her husband, the penalty prescribed is three years’ exile, but such persons, even though not akin in blood, cannot return home to share in common domestic rites with their children, or to eat at the same table. In this law we see clearly the operation of a local or domestic ‘pollution’ which debars the slayer from his family hearth, and which is quite distinct from the civic pollution which debars him from certain definite States. It is important to observe that with the local or domestic pollution no civic or international law has ever interfered; whereas civic pollution has been regulated by law according to the varying degrees of guilt, and the claims of the relatives to ‘appeasement.’

Plato implies that one year was the normal period of exile for manslaughter. The Greek verbs ἀπενιαυτίζειν[76] and ἀπενιαυτεῖν convey the same implication. Can this fact be reconciled with the law of Dracon? We believe that it can, but only by distinguishing between theory and practice, between local and central courts, between local and central religion.

Plato shows how local judges would have solved the difficulty caused by recalcitrant relatives. We have seen[77] that Plato decrees perpetual exile for manslaughter between strangers. But exile from what State? Surely it was only from the State in which the deed took place: and the reason for this penalty was probably the fact that the relatives of the slain did not live in the State where the deed took place: and hence no ‘appeasement’ of these relatives could formally admit him to that State, though he could be admitted through ‘appeasement’ to his native State, if the slayer and the slain were both citizens of the same State. Thus the tendency of the pollution doctrine, apart from the claims of the relatives of the slain, was to exact perpetual exile for manslaughter. Plato decrees that any citizen had the right to prosecute a stranger for manslaughter, but not that he had a right to accept ‘appeasement.’[78] Hence, by a strange paradox, the relatives of the slain provided a medium by which the man-slayer regained his civic status. Yet, in the case of involuntary kin-slaying, the slayer could never re-enter his home! We believe that these decrees are not Platonic creations, but were found in Attic law, written or unwritten. Can they be reconciled and made intelligible?

We saw[79] that wergeld was not admissible for kin-slaying in the Pelasgian tribal system. Outside the kindred, however, wergeld permitted the slayer to remain at home or to return after a time, if he could not pay the full were. A comparison of such customs with the historical homicide code suggests quite obviously a compromise, in which the seventh-century pollution-doctrine failed to impose its will on the relatives of the victim because of a real or presumed ‘forgiveness’ on the part of the slain. Without the anger of the dead, the pollution doctrine could not operate.[80] Apollo himself could not enforce it. The relatives of the slain had a just claim to be regarded as the best interpreters of the anger of the dead. It was in this crevice, so to speak, in the doctrine of pollution that the kindred of the slain drove the thin end of their old tribal wedge. They claimed the right to determine the period of exile for manslaughter, but for manslaughter only: for in such cases the anger of the dead could not be regarded as perpetually implacable. In theory, then, these relatives had the right to consent to ‘appeasement’ at any time; but in deference to the dead their consent could not become effective before a year had passed. They could in theory delay their consent indefinitely, but delay was less probable in local than in central jurisdiction. They were compelled by law to prosecute the manslayer in court if the slayer denied his guilt; but if he admitted guilt, no trial was necessary; and it was only in such a contingency that ‘appeasement’ could occur without trial: nevertheless a year’s exile was still necessary before the relatives could accept ‘appeasement’ and finally remove the barriers to ‘purgation.’ The fact that the involuntary kin-slayer could never re-enter his home we attribute to the tradition of Pelasgian domestic religion.[81] This solution reconciles, we think, the law of Dracon, the code of Plato, and most of Demosthenes’ references. It is also in harmony with our general theory[82] of the compromise between ‘pollution,’ tribal wergeld, and State law, which is expressed in the murder-code of historical Greece.

Justifiable and Justifiably Accidental Homicide

In our analysis of the Attic laws concerning justifiable homicide, we will begin by drawing a distinction between three possible contingencies. First of all, we can conceive that blood has been shed without any intent to kill, but with a certain element of neglect (ἀφυλαξία), which has however been expressly mentioned and declared to be justifiable in law. Secondly, we may suppose that there was a certain degree of intent to kill and a certain amount of deliberation, but also that there was an extenuating element of impulse or passion which has been decreed guiltless, in certain circumstances, by the law. Thirdly, we may suppose that the person slain was an outlaw or a State-criminal, whose life was forfeit by the laws of the land, and whose citizen-slayer was declared to be justified in advance.

Homicide of the first class has so much in common with ordinary accidental homicide that we think it probable that they were often confused in Greek thought, if not in law. The words ἄκων and ἀκούσιος which, we have seen,[83] were applied indiscriminately to denote cases of different degrees of guilt in accidental slaying and in manslaughter, were also used to denote such forms of accidental slaying as were expressly ‘justified’ by law. Perhaps this confusion may help to explain still further the apparent discrepancies in Demosthenic references to ‘release’ and ‘private settlement.’ For the case which we are now discussing, there was no penalty, no exile, or loss of property, not even a fine. Pleas of justifiably accidental homicide were doubtless frequently made in answer to charges of manslaughter or of wilful murder. The King-Archon (and perhaps also the Tribe Kings) had to decide between the merits of the ‘charge’ and of the ‘plea.’ Obviously, it was always as a result of a ‘plea,’ never as a result of a ‘charge,’ that homicide cases were referred to the Delphinium court.

Justifiable homicide of the second class has close affinities with extenuated manslaughter, or slaying in a passion. The essential difference lies in an express legal justification in one case, and the absence of such a justification in the other. When we come to analyse the Oresteian legends of Attic tragedy we shall find[84] that the close affinity which exists between these two legal conceptions caused considerable confusion in the legends—caused Orestes to be immune from punishment, from one standpoint, but liable to a period of exile, from another. The oracle of Apollo, which commanded him to slay his mother, should naturally have been accepted as a complete justification. Some legends took this view.[85] But such a contingency was not expressly mentioned in the Attic laws concerning justifiable homicide. Plato assures us[86] that under no circumstances, not even in self-defence, was it lawful to slay a parent. Hence it became necessary to regard Apollo as the divinely immune cause of guilt; and Orestes, as his blind, obedient instrument, became liable to a merely nominal charge of manslaughter or extenuated matricide! His mother was almost compelled to ‘forgive’ the deed! In one legend[87] Apollo commanded the Erinnyes to withdraw from pursuit, and drove them from his temple; in another he decreed that they should pursue Orestes for a year![88]

Our third classification includes cases in which only two issues could be raised—namely, lawful homicide or wilful murder. From this point of view the act of Orestes could also be discussed, and an Athenian court could find it interesting to discuss and difficult to decide at what precise time in the post-Homeric social evolution did private vengeance become illegal! The trial of Orestes at Argos, which Euripides describes in the Orestes, seems to depend upon this legal difficulty.[89]

We will now illustrate these cases by relevant quotations.

Aristotle’s[90] account is a mere extract, or rather a kind of summary: ‘Homicide,’ he says, ‘admitted and alleged to be lawful, as of an adulterer caught in the act, or of a friend mistaken for an enemy in war, or of an antagonist in an athletic contest, is tried in the Delphinium.’ The first clause has been included in our second classification, the rest of the cases belong to our first category. Demosthenes quotes[91] the Draconian laws, and his account is almost identical with that which Plato[92] gives. To facilitate comparison, we will quote both accounts together in parallel columns:

We have elsewhere[98] cited a Delphic law which concerned justifiable slaying, and which, in its completeness, we regard as the common parent of both these legislations. These excerpts are strikingly similar, especially if we remember that they are a number of extracts which we have put together with the object of eliciting a complete list of cases. No better proof than this could be adduced of the ‘historicity’ of Platonic legislation[99] regarding homicide. Antiphon[100] refers to the legal immunity of physicians, and we may therefore consider this Platonic law to be also Draconian (or Solonian).[101]

To both the above quotations our triple classification of cases of justifiable homicide can without difficulty be applied. Confining our attention to the Demosthenic account, we may point out that the reference to accidental slaying in a contest, ambush, or battle, is covered by our first category: the infliction of death for adultery or seduction is included in our second category: and the slaying of unjust aggressors, and of murderers en rupture de ban, in our third category.

In regard to adultery, we have already[102] suggested that the right to kill in flagrante delicto must not be regarded as a relic of a primeval custom which decreed the death penalty for adulterers in all circumstances. The Gortyn laws and the Homeric customs which are denoted by the word μοιχάγρια support our view that the right to kill in flagrante delicto was an innovation of the period of synoekism and of centralised government. Philippi[103] thinks that there is a strong probability of correctness in Pausanias’ view[104] that the age of Theseus represents the point at which the distinction of justifiable homicide could be applied in such matters. But our reading of Pausanias suggests that in his view the distinction was first applied to adultery in the time of Dracon.[105] We believe the distinction, thus applied, was as late as the seventh century, and we agree with Pausanias that the penalty thus decreed was severe.

Wilful Murder