STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world byJSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.istor.org/participate-istor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. Classical Philology Volume XI October igi6 Number 4 EHISKH^IS AND THE AIKH ^ETAOMAPTTPIflN By George Miller Calhoun Within a comparatively short time four important studies of the h'\.Kt\ xf/evdofiapTvpiav, those of Rentzsch, Bonner, Leisi, and Lipsius, have made their appearance. The diversity of the views advanced by these able scholars suggests that a fresh examination of the evi- dence may be productive of result. Furthermore, the valuable information concerning the prosecution of perjury in Alexandrian law that is contained in the Halle papyrus lends new interest to the investigation of certain points.^ At the same time, although the importance of this new document for the understanding of Greek law can hardly be overestimated, extreme caution must be observed in making it a basis for conclusions regarding the details of Attic procedure.^ ' Dikaiomata. Auszilge aus Alexandrinischen Gesetzen und Verordnungen (Berlin, 1913). The text (cited as "P. Hal. 1 ") is found on pp. 15-24, and the portion dealing with the prosecution of perjury is reprinted on p. 48. The commentary and explana- tory essays (cited as '' Dikaiomata") contain scholarly comparisons of Alexandrian and Attic procedure, but an exhaustive examination of the Attic sources in the light of the new fragment naturally did not come within the scope of the editors' work. San Nicolo ("Einiges aus dem P. Hal. 1," H. Gross' Archiv, LIII [1913], 342-61) has endeavored with but doubtful success to develop new conclusions on some points of the Attic law {infra, passim, and Lipsius, Das Attische Recht und Rechtsverfahren [Leip- zig, 1905-15], pp. 987-88). His treatment does not indicate that he has attempted an adequate examination of the Attic sources. Lipsius also has given the new docu- ment careful consideration in discussing the effect on judgments of the SIkti) \j/e\iSa- /xaprvplav, but on other points has had to content himself with the brief comment that could conveniently be included in his "Nachtrage" (op. cit., pp. 955-60, 987-88). 2 For example, the use of iirCKaii^AveuBai. instead of iiriffK^TrreaSai warns us immediately that the law before us has drawn upon other sources than the Attic. Consequently no detail of the Alexandrian process can with safety be assumed for [Classical Philology XI, October, 1916] 365 366 George Miller Calhoun This discussion will in general be restricted to matters which are still subjects of controversy or, in the opinion of the writer, require further consideration. The nature of iiriaK-q^is. — Any investigation of the problems connected with the prosecution of perjury under the Attic law must begin by defining the term imaK-q^is and determining precisely what was the relation of the proceeding so designated to the biKj} \pevbo- ixaprvp'uav} The view of Hermann,^ that iiriaKriii/is was merely the announcement of an intention to bring an action for perjury and not the filing of the formal charge, was generally accepted until the appearance of Leisi's admirable monograph " Der Zeuge im Attischen Recht."' Leisi argues from the mention of written ^TTKr/cij^eis in the Laws* "dass die ericxKri^is schriftlich war, und da die Klage- schrift bei der dUr] ^pevdonaprvpicov sehr kurz war, so scheint es uns wahrscheinlich, dass die hr'i.(TKi}4'i-s nichts anderes ist, als die Ein- reichung der Klage selber."* The conjecture is undeniably an attractive one. But infer- ences in regard to the minutiae of Attic procedure based upon the Attic law unless there be supporting evidence. On the other hand, if some reasonable and advantageous provision which makes for simplification of procedure or scientific definition of perjury is clearly seen not to have formed a part of the Alexandrian law, we shall scarcely expect to find it in a system, earlier and less highly developed, which undoubtedly influenced Hellenistic procedure. But here also there is need for caution, since in this instance the transcript did not contain the entire law, but only such por- tions as were pertinent to the matter in hand (cf. the editors' notes to lines 34 ff. and 75 ff., Dikaiomata, pp. 55-56, 62-63). Furthermore it is probable that some details of the process were covered by general laws which regulated the procedure for entire classes of actions (cf. line 46: (carA t6 Sidypan/ia, and the editors' note, Dikaiomata, p. 58). ' That ixl(!KT]^pi.i is properly not synonymous with SIkt) xj/evSoiMpTvplioii, first noticed by Hermann (De vestigiis institutorum veterum imprimis alticorum, per Platonis de legibus libros indagandis [Marburg, 1836], pp. 69 f.), has long been recognized (see MSL, p. 491 ; Leisi, Der Zeuge im Attischen Rechi, pp. 125 ff.). But it could of course be loosely used as a virtual equivalent of 8Jki) 'j/evSofrnpTvpluv, as is probably the case in [Dem.] 47.51; Isaeus 4.17; Aristotle Pol. 1274 6 7. This may account for the confusion of the two terms by the lexicographers (Harp., s. ji. ivfaK^aTo; Et. Mag. 355. 35; Poll. 8. 33, 36; Suidas, s. v. iirlo)vriv ovdels avBpcoTrwv, ore ovtos rjywvl^iro, oim p,LKpav ovre HtyaX-qv, and a few lines on paraphrases by ovk ewecTKri^'afi.riP iyu rint Tots fjidpTvai Tols ravTa p.apTVpov avadexv'rai, 6 iKnapTVprj- ffas, iKiivos virobiKos ri rwv ^ev8op,apTvpiwv, iav 8i jxif avadixv'O'i') ol fiapTVpri<7avT€s ttjv eKfiaprvplav.^ It is difficult to understand how 1 Cf. Leisi, op. cit., pp. 126, 138. 2 5. 17. Even in P. Hal. 1, with its full legal phraseology, we find the same omission (lines 64 ff.) after the filing of the plaint has been mentioned in preceding paragraphs (lines 44 ff., 55 ff.). 3 Dem. 29. 7, 8; [Dem.] 47. 1; Isaeus 3. 11, 12; Din. 1. 52; cf. Plato Laws 937B; Ar. Cons. Ath. col. 35. 11 ff.; CIA, II, 609. 16. * [Dem.] 48. 31, 44-45. 5 [Dem.] 46. 7. The correct interpretation of this passage is found in the accurate translation contained in Sandys's note ad lac. The S.ixa here does not imply that there was necessarily one trial for both deponent and attestants, since it is used idiomatically with the preceding re ... . koX to strengthen the contrast with iirb ttj^ airiji iiri- (TK^^ews. Leisi's observations on this passage (op. cit., pp. 121-22) show an imperfect analysis of the situation. We are dealing, not with a single offense, but with two distinct offenses, as the writer has pointed out. 'Eirt(TK7j^ts AND THE Akij 'i^evdoixapTvpioiv 369 a single written charge could have been so framed as to be valid against two sets of defendants, the original deponent and the wit- nesses who attested his deposition, when the guilt of the deponent was contingent upon the innocence of the attesting witnesses, and, vice versa, the guilt of the attestants upon the innocence of the deponent. It would be difficult to adapt the document commonly cited as the official formula for a charge of perjury to this situation.' But if the iirL(JKr)^Ls is simply a preliminary impeachment, the prosecution of both deponent and attestants aTro ttjs ain-fjs hnvKr]- ■^/tus presents no difficulty. The intending plaintiff would simply denounce the evidence, or certain parts of the evidence, contained in the iKixaprvpia deposed to and attested by the parties therein named, and this single impeachment, whether oral or written,^ would give hina the right to proceed against the deponent and wit- nesses severally until the guilty person was brought to justice or all who were responsible were acquitted. Again, if iiri(rKr]\l/i.s in this passage means the formal plaint, virbdLKos must be understood in a very unusual sense, as it normally refers, not to one against whom a plaint has actually been filed, but to one who is "liable" to prose- cution.' Another difficulty is presented by the statement of Aristotle that actions for perjury which had their origin in trials before the Areopagus came within the jurisdiction of the thesmothetes.* There is no evidence that Athenian law permitted a plaint to be lodged with magistrates other than those in whose jurisdiction the action lay, or that the thesmothetes had deputies (irapedpoi) who were empowered to receive plaints in their stead. Consequently, if iwlaKrjxl/is was the filing of the plaint, we are forced to the conclusion that some of the thesmothetes, notwithstanding their manifold duties of importance, must have been constantly in attendance upon sessions of the Areopagus in order to receive perjury plaints. This seems unlikely. 1 Dem. 45. 46. Cf. Leisi, op. cit., p. 129; Lipsius, Rechl, p. 781. 2 Cf. infra, p. 374. '[Dem.] 47. 1, 3; Plato Laws 879A ; Lys. 10. 9; Plato Laws 954A; Isaeus 8.32; Dem. 21.33; 29. 16; 37.35; [Dem.] 33. 29; 40. 54; 46. 4, 10; 58.26. * Cons. Ath. 59 . 6. 370 George Miller Calhoun Again, it is difficult to reconcile with Leisi's theory the fact that the purpose of an €TriavTias could be employed against perjured witnesses as well as false accusers. What man on trial for his life would neglect eiri- (TKrj^'is, which meant a stay of execution and a chance of life, and trust to friends to bring a ypa(j>ri avKocftavrias, which would expose the perjury only tjj'ik' ovSev riv en TrXeov rots weirovdoaiv ? Neither of these attempts to solve the difficulty can be called convincing. They are in fact based upon the assumption that has itself given rise to the perplexity they seek to solve, namely, that the procedure in perjury actions remained unchanged during the period covered by the sources.^ It was left for Leisi to discover the essential fact that 1 5. 95. We cannot fail to note a difference between this positive statement that condemnation to death does away with any chance of accusing a perjured witness and the commonplace that men have often been convicted of perjury after those against whom they testified had been executed. In this connection it must be remembered that in Athens execution followed conviction with the most summary despatch (cf. Thonissen, Le droit pinal de la republique Athenienne [Brussels, 1875], pp. 95 f.). 2 Op. cit., pp. 36 ff. ' The condemnation of the generals after Arginusae by an unconstitutional pro- ceeding would seem to lend some support to this view. But the Antiphon passage (5. 95), as has been pointed out, evidently has in mind not such sporadic cases but the regular procedure in capital trials. The recurrence of this riiros in a time from which so few speeches remain is in itself an indication that the conditions described are not exceptional. * Recht, p. 782. While sycophants no doubt were skilful perjurers and perjurers could be loosely classed with sycophants (cf. the writer's Athenian Clubs in Politics and Litigation [Austin, 1913], pp. 79-81), the theory that a ypadvTi)S is obviously used in the loose sense just referred to. * An instance of this tendency to assume that Demosthenic or Aristotelian accounts of Attic procedure hold for earlier times is to be found in the fact that it was left for Bonner (and after him, independently, Leisi) to discover that written evidence was not always the rule. The existence of any law or process at a time not established by examples is always open to question. 'EirlffKrjxl/is AND THE Akrj '^evhoixa.prvpiwv 373 there was a change in the working of the SIkj; \pevSonapTvpio3v which made possible a stay of execution. Unfortunately, Leisi has obscured this simple but important truth by a number of conclusions of which the chronology and certain details will not bear examination. His erroneous conception of eirto-KJ/i/'is^ causes him to regard it apparently as a new judicial pro- cess which replaced an earlier type of perjury action, and to assume that it was inseparable from the stay of execution; in other words, that when there was no provision for a stay of judgment in capital actions iwla-Kri^' is could not have existed. Its introduction he puts in 403.^ Now the commonplace concerning the conviction of per- jured witnesses after the execution of the men against whom they had testified is found not only in the defense On the Mysteries but also in a speech of Lysias which was delivered probably fifteen years after the archonship of Euclides.^ 'Eirl(TKr]\pi.s we know was prac- ticed in the year 399.* We have then a period of not less than eleven years during which there was eTrl ■rapaxpijliO' Sray ij yywau dyayyaxTB^i. 376 George Miller Calhoun at the preliminary investigation when the witness deposed. What may be taken as a typical instance is described by Isaeus.^ Prosecution vnthout impeachment not allowed. — ^There is no specific statement in the sources that eirl<7Kr]\pt,s was an indispensable step in an action for perjury, but this has generally been taken for granted in studies of the subject, though no one apparently has considered the problem with a realization of its true importance. Such phrases as ovK eire(TKi]^aTO ov8' eire^epxeTaL^ and ov8' .... eirecTKriil/aTo ovde T€To\firiK€ dMKeiv^ might seem to point to the contrary con- clusion were it not obvious that they are merely the idiomatic negative of the idea eirLaKrjwTeffOai, /cat eTri^ipxicii, and that ^ttIo-kj/^u was a new pro- cess which replaced them, does not accord with the facts, as has been pointed out.* Reasons have already been given for rejecting the view of Lipsius that the ypaii avKoavTias could be brought against perjured witnesses.* Rentzsch, adopting a conjecture of Boeckh, believes that in prosecuting perjury either a public or private action, according to the circumstances of the case, might be employed: "Ita viro illustrissimo [Boeckh] placuit controversiam disceptare, ut in eo delictorum genere ponat falsum testimonium, quae per actionem et privatam et publicam vindicari possunt Prorsus ita ego sentio. Sicut enim in causa kXott^j, utrum genus adhibeatur, discrimen facit locus, tempus, furtum ipsum, sic in hoc delicto vindi- cando, puto, genus causae principalis, unde actio secundaria fluxit, ita tamen, ut pro 'Ypa4>rj yj/evbonaprvpuav, uti in kXot^s, actori licuerit adhibere Skijj' ipevdop,., non tamen pro SIk^ ypa^yliv."^ This, in the opinion of the writer, is substantially the true solution, though we may reserve judgment in regard to the somewhat complicated ques- tion of classification' and put the proposed conclusion in this simpler lAnd. 1. 7 =Lys. 19. 4; cf. also Ant. 5. 95. On these passages see Rentzsch, op. cit., p. 37; Lipsius, Recht, pp. 777-78. It would be interesting in this connection to know whether defendants convicted leWas were brought from prison to court and allowed to conduct their perjury prosecutions in person or were obliged to remain in prison and entrust the conduct of their cases to others, as the context seems to indicate (Dem. 24. 131). 2 Op. cit., pp. 89-90. ' Leisi points out (op. cit., p. 120, n. 2) that Callistratus, although he was acting in collusion with Olympiodorus, was still formally a claimant for half of the estate, and was therefore legally entitled to the privilege of iwtffKtfl/is. Professor Bonner, who has kindly communicated his later views on this case to the writer, no longer holds that it establishes the right of persons not parties to the action to prosecute. He is inclined to believe, however, that Callistratus had not carried his claim into court, but means that, if he had not been in league with Olympiodorus, he would have presented his case to the jury and would have concluded by impeaching the witnesses, which he would in that case have been in a position to do. * Cf. supra, p. 373. ' Cf. supra, p. 372. « Op. cit., pp. 37-38. ' To determine whether the SIkti '^evSonaprvptav was a "public" or a "private" action, or was sometimes the one and sometimes the other, or combined characteristics of both classes, is a problem which has given rise to prolonged controversy. It is of more importance to establish the details of the process. 380 George Miller Calhoun form: any citizen in full enjoyment of the franchise had the right to prosecute witnesses who perjured themselves in public actions. This view is advanced by Rentzsch simply as a conjecture intended to explain the apparent contradictions of the sources, and Lipsius is quite correct when he says that there is no direct evidence to support it.' But it is entirely reasonable, it offers a satisfactory explanation of the facts, and no definite objection to it has yet been advanced. Furthermore, it accords with the attitude of the Athe- nian courts and public toward perjury. As the writer has elsewhere observed,^ perjury in private suits was thought of primarily as an infringement of the personal rights of the individual rather than an offense against the majesty of the law, and it was left to the indi- vidual litigant, with the encouragement of the state, to initiate measures of redress and punishment. But perjury in public actions, whether it was committed by witnesses for the defense or prosecution, affected the state more directly, and, as in the case of other offenses which touched closely the interests of the state, it would be reason- able for any citizen to constitute himself the public representative and initiate a prosecution. Under ordinary circumstances the parties to the original action would be the persons most likely to avail themselves of this right, and in actual practice the percentage of prosecutions instituted by citizens not parties to the action was probably small, if we except those cases in which convictions had made it impossible for defendants to prosecute.' We have noted that Aristotle makes no distinction between public and private suits.'* This would indicate that even in public actions, though any citizen could prosecute by virtue of an impeach- ment, custom if not law restricted the privilege of instituting the impeachment to the actual litigants.* It is unlikely that in private 1 Recht, p. 778. ' "Perjury," p. 7. The importance of allusions in perjury prosecutions to deceit practiced upon the court is sometimes exaggerated (e.g. Rentzsch, op. cit., pp. 38-39). Athenian juries were never unfair or careless; they were always "deceived" by unscrupulous rascals. 3 Cases are cited on p. 371, n. 4; cf. Bonner, op. cit., p. 90. « Cf . supra, p. 378. * The general negative formula oiSeli ^irecrK-iJ^aro is not sufficient ground for con- cluding that any person could impeach a witness at the trial of a public action. '^icKfiiffis AND THE Mktj '^(vSoiiapTvpiav 381 actions the right to prosecute secured by impeachment could be transferred.' Both the Laws and the Alexandrian SiKaiii/xaTa are in substantial agreement with Aristotle, for in the one it is explicitly provided that the litigants are to impeach,* and the other proceeds clearly upon the assumption that impeachments will originate with litigants.' Both, like Aristotle, fail to make any distinction between public and private actions, though the Alexandrian law seems to contemplate only the latter class of actions.* The abandonment of perjury prosecutions. — Rentzsch observes that there is nothing to indicate that a litigant who had impeached was forced to prosecute, and he regards it as likely that the failure to do so entailed no penalty except the loss of court fees.* Leisi apparently concludes from Isaeus 5. 18 that all prosecutions for per- jury could be dropped at any stage without incurring a penalty.* The question perhaps merits further consideration. Prosecutions for perjury committed in private actions could unquestionably be dropped, apparently at any stage, without penalty. For we hear of impeachment not followed by prosecution,'' of prosecutions abandoned before coming into court,* and in one instance of a com- promise reached at the moment when the jury had finished voting and the ballots were about to be counted.* These cases represent both ' In Isaeus 5. 9 the speaker intimates that the death of Polyaratus before he could prosecute the witnesses he had impeached did away with any chance of reopening the case. But this may have been because there was no male old enough to undertake the prosecution, and in fact the witnesses were prosecuted (12), apparently many years later (35; cf. Wyse's note to 12). While the case is too complicated and the details too obscure to afford safe ground for definite conclusions, the writer is inclined to believe that in an inheritance case the impeachment of a witness by a single claimant would secure to the other claimants also the right to prosecute. Cf . infra, p. 383, 2 937B. ' P. Hal. 1. 44 f., 55 f., 64 f. * Cf. Dikaiomata, p. 50; Lipsius, Recht, p. 987. ' Op. dt, pp. 29-30. On Rentzsch's conjecture that court fees were deposited when the impeachment was announced, cf. infra, p. 386. « Op. cit., p. 124. ' Lys. 23. 13—14. On impeachment as a means of putting a witness on oath, cf. tupra, p. 370. » Isaeus 5. 13. In [Dem.] 47. 64, 74-76, a similar release is evidently allowable, even immediately prior to the trial of the perjury actions. • Isaeus 5. 17-18. 382 George Milleb Calhoun formal compromises and simple abandonment;* some are ordinary per j my actions, while others are dtafiaprvpla cases.^ But in these instances we have only prosecutions arising from private suits. Consequently they do not justify us in assuming that all perjury actions could be abandoned without penalty. That a distinction was made is a priori not improbable. Unfortunately, however, no arguments in prosecutions arising from public actions have been preserved and there is no evidence on which to base an opinion. In this connection a further question suggests itself. In the event that a litigant impeached a deposition attested by more than one witness, was he not required, if he prosecuted one, to prosecute all who had deposed to the same statement ? In Aphobus v. Phanus is found a case in point. Aesius, the plaintiff's brother, was alleged to have attested the very deposition which was the occasion of the prosecution against Phanus and to be ^(vbonaprvplav inrhSiKos. At the trial of Phanus he was acting with his brother and denied point blank having attested the deposition to which his name was appended.' Had Aphobus been under any legal obligation to prosecute him with the other witnesses, Demosthenes would cer- tainly allude to his failure to do so, for it would strengthen his case materially. But there is not the slightest hint that Aphobus is either required or expected to prosecute him. Again, in the litiga- tion over the estate of Dicaeogenes, Menexenus, who had prosecuted the witnesses to an alleged will, released some of the defendants after one of them had been convicted.* It is worthy of note that in the Halle papyrus we find the express provision that the litigant who impeaches a witness is to impeach " all who gave the same testimony."* Neither here nor in the Laws^ is anything said regarding the abandonment of actions against witnesses. 1 In Lys. 23. 13-14, Panoleon simply failed to prosecute the witness whom he had impeached. 2 In Lys., loc cit., and in Isaeus 5. 18 we are dealing with Sianaprvpla cases. ' Dem. 29. 15-18. As has been observed (supra, p. 377, n. 1), the name of Aesius was appended to the deposition (cf. 15: irfypa^a), and the question was whether or not he had attested it when it was read in court. * Cf. infra, p. 391. 5 27-28. The motives for this provision suggested by the editors (Dikaiomata, p. 54) and by San Nicol6 (op. cit, p. 347) doubtless all entered into consideration. Cf. infra, p. 392. ' Loc. cit. 'ETl(r/o;^ij AND THE Akij "^evSonapTvp'uav 383 Time limit for filing plaint. — It is obvious, as Rentzsch observes/ that depositions which had been impeached could not have been preserved indefinitely by officials, and that a reasonable time must have been allowed for filing plaints, at the expiration of which the right to prosecute lapsed. What this was cannot be determined. In Isaeus 5. 9 we are told of a litigant who died ov xoXXqj XP^''V vdTipov utTO. rfiv diKr]v .... Trpij' hre^ekdeiv ols lKi Cf. Isaeus 5. 16-17 for what is apparently the customary procedure. • Lys. 23. 14 indicates that the filing of the plaint at the hearing was not obliga- tory, unless we assume that Pancleon changed his mind during the proceedings before the magistrate. But here again lTeirj\9ey is rather indefinite and may include a long train of legal steps. » Cons. Ath. 59. 6. *MSL, pp. 59, 488-89; Rentzsch, op. cit., p. 29; Leisi, op. Hi., pp. 124-25; Lip- sius, Recht, p. 777. 6 Cf. MSL, pp. 485-86. « Cf. infra, p. 387. ' For example, Aristodemus v. Leochares ([Dem.] orat. 44) and the Sia/uipTvpla cases in Isaeus, while they are technically perjury actions, are in reality suits for inherit- ances, and the questions of law and of fact which they involve belong strictly within the jurisdiction of the archon. In Lys. 23. 14 a case of another kind illustrates the same principle. Had Pancleon brought the witness of Aristodicus into court, the question at issue would have been one of jurisdiction, and the court of the polemarch would have passed upon it just as if a irapaypa Loc. cU., pp. 29-30. *Loc. dt. This conjecture is adopted by Lipsius (Recht, p. 783). ^ MSL, p. 489; Rentzsch, op. cit., pp. 41 ff.; Bonner, op. dt., pp. 91-92; Leisi, op. dt., pp. 131-32; Lipsius, Recht, pp. 782-83. « 5. 19. ' 10. 22. « Ant. 23. 7; Dem. 29. 16, 50. ' MSL, loc. dt.; Rentzsch, op. dt., p. 49; Bonner, loc. dt.; Leisi, op. dt., p. 132. Rentzsch (pp. 41 ff.) reviews the whole controversy at great length. 'Eiria-Krjyl/is AND THE AUi} '^evSonaprvpiuv 387 but Lipsius, in his latest discussion of the subject/ observes that this explanation is extremely dubious "weil dies der einzige Fall ware, in dem Atimie nicht kraft des Gesetzes, sondern richterlichen Spruches eintrate." He accounts for the allusions to arifua as a penalty by the desire of the speakers to emphasize the gravity of the offense, and accepts Meier's solution of the Isaeus case, that the sum pro- posed was so great that the defendant's property would not suffice to pay the judgment, he would find himself meprjiiepos, a biKij i^ov- Xtjs would be brought against him, and &nnla would be the result.* No one who has studied this much-mooted passage has taken into consideration the fact that the case is a diafiaprvpia, in which the deposition of the witness who was put forward was regarded, in practice if not in strict law, as the substantial equivalent of a plead- ing or claimant's affidavit.* It is extremely unlikely, in the opinion of the writer, that in a Siaixaprvpia the jury would have assessed so severe a penalty that Leochares could not have paid it, for he was evidently the possessor of considerable wealth.* This is Hkewise a serious objection to the view that the dnjuia in this case would be imposed by xpoo-ri/iT/o-ts, for it is equally improbable that in a Sia- IxapTvpia the jury would impose so grave an additional penalty. By far the simplest solution of the whole problem is that here we have a case in which the defendant has already been twice convicted. Although this explanation is not even suggested by most scholars and is barely mentioned by Rentzsch, there would seem to be no » Recht, pp. 256, n. 54, 783. Lipsius approves the opinion of Lelyveld concerning inula which Rentzsch {op. cit., p. 50) rejected. ' Lipsius, after having once rejected this solution as "zu kUnstlich" (MSL, p. 219, n. 56 fin.), now cites in its support Isoc. 16. 7 (Recht, p. 256, n. 54 fin.). ' That the witness in a Siafiaprvpla occupied an exceptional position in this respect can hardly be doubted. It is clearly established by the arguments in Sux/iaprvpla cases, which naturally are not concerned at all with the witness's means of knowing the facts, his character, or possible motives for perjury, as in ordinary prosecutions, but merely with the general soundness of the claim which his deposition formally maintains. At the same time the penalty attached to a third conviction was a wholesome check, for it tended to deter unscrupulous persons from repeatedly lending their aid to purely obstructive tactics. * The estate of Dicaeogenes produced the exceptionally large revenue of eighty minas annually (Isaeus 5. 11, 35), and Leochares was accepted as joint surety with one other for the two-thirds interest claimed. He must therefore have been a man of large means. 388 George Miller Calhoun valid objection to it.* It affords an adequate explanation not only of this case but also of that mentioned by Lysias, which Lipsius does not even attempt to solve. Furthermore, it is in no wise incon- sistent with the other passages in which artiua is said to be a penalty of perjury, as has been pointed out.* But it may be remarked that a careful review of the controversy on this point makes very evident the dangers of advancing positive assertions where the evidence is so slight. The view that more severe penalties could be imposed in prose- cutions originating in public actions than in those which arose from private suits is based on insufficient evidence, and the arguments urged in its support involve misunderstanding of the sources as well as questionable logic' It had been generally assumed that the fine went to the prose- cutor, as in all private suits, until Bonner expressed a doubt "that the fine was ever paid to the prosecutor in the way of damages."* While no very pertinent or convincing objection to Bonner's view has as yet been advanced,' his arguments scarcely justify us in regarding the matter as finally settled. But the question he raises ' Rentzsch, op. cit., p. 45: "Nisi vero arbitramur his locis de tertia \ji€vioiMprv- ftluv condemnatione cogitandum ease, quamvis nulla eius mentio iniciatur." The absence of a specific allusion to three convictions is sufficiently accounted for by the fact that Athenian audiences were thoroughly familiar with the circumstances under which drifJo entered into the punishment of perjury. It may be remarked further that in all but the Isaeus passage, and possibly Dem. 29. 50, any such allusion would have been a distinct blunder. In Dem. 29. 16 and Ant. 28. 7 we have a rhetorical topos on the dangers of committing perjury, and in Lys. 10. 22 the admission that Dionysius had twice before been convicted would have been a sad anticlimax to the pathos of the passage. ' Cf . the preceding note. ' The conjecture of Meier (cf. MSL, p. 488, n. 33 fin-) that an action i-rar-^atut could be brought against those who gave false testimony in public actions ia properly rejected by Lipsius {iUd., p. 426). Thonisaen, in his attempt to maintain it (op. dt., pp. 386 ff.), attaches undue importance to the denunciations of perjury which were a commonplace in pleading, applies to witnesses remarks made by Isocrates with reference to false accusers, and confuses false "informations" with false testimony. Rentzsch attempts to prove that more severe penalties could be proposed in the hypothetical "ypai) ^evSo/jutpTvpluv" by arbitrarily applying to perjury actions what is known of the ypaii fevSokXtiTelas, an entirely different process (pp. dt., pp. 50 ff.). * Op. dt., p. 92. ' The fact that the SUri tl/evSo/Mprvptuv was an dYiiii- Tt^tjrAs (Leisi, op. dt., p. 131, n. 1) proves nothing. While the observations of Lipsius (Recht, p. 783, n. 19) on the absence of fine distinctions in Attic law are sound, they are not applicable to Bonner's position, which is that the fine was never paid to the prosecutor. 'Ettio-kt/^is and the AUri ^evSonaprvpluv 389 is of particular importance as regards cases which originated in public actions, for in these if in any we should expect to find that the fines imposed went to the public treasury. But unfortunately here again the lack of evidence in regard to the prosecution of perjury in public actions makes impossible any certain conclusion.' Very interesting in this connection are the provisions of the Alexandrian law. If a defendant against whom judgment has been given convicts a witness or witnesses, the judgment stands annulled and the fine imposed upon the witnesses is put in execution.^ This fine is fixed by the terms of the law at one and one-half times the amount of the judgment in the ordinal action.' If a plaintiff who has failed to secure a verdict convicts in an action for perjury, the fine is entered against the witnesses as in the preceding case, and the defendant in the original suit is required to pay to the plaintiff the amoimt originally sought with an added penalty of ten or fifteen per cent.* If both plaintiff and defendant convict witnesses, the fine is entered against the witnesses and the original suit is retried, imless an appeal be taken.* In view of the sharp distinction here made between punitive fine and compensation, it can hardly be doubted that the fine went to the public treasury.' Unfortunately, here, as in the chief sources for Attic law, private suits only appear to be in question." In the Laws nothing is said regarding the punishment of perjury except that three convictions disqualify for witness service.' Effect on the original action. — Recent discussions agree that retrial was permitted, on proof of perjury, in all cases in which the pecuniary penalty attached to the SIkjj yf/fvSonapTvpluv and S'lKri KaKOT€X''nS>' did not constitute satisfactory indemnification; in other words, where the original action involved the loss of life, freedom, citizenship, family rights, etc.' As regards the conditions on which > Cf. supra', p. 382. « P. Hal. 1. 44 ff. ' Ibid., 30-31. It should be noted that in 53, as well as in 58 I. and 66 f., we are dealing with the fine in the perjury action and not with the judgment sought in the original suit. * Ibid., 66 ff. » Ibid., 64 ff. • Cf. Dikaiomata, p. 51; San Nicold, op. dt., p. 360. ' Cf. supra, p. 381, n. 4. » 937C. "Leisi, op. cit., p. 133; Lipsius, Recht, pp. 956-67. 390 George Miller Calhoun retrial was granted in these cases, there is some uncertainty of opinion. The passage of the Laws which treats of the prerequisites for retrial is understood to mean that over half of all the witnesses introduced by the successful litigant must be convicted.^ That this is not to be applied to Attic procedure is very generally agreed. Rentzsch and Leisi conclude from a statement of Isaeus that the conviction of a single witness was sufficient.'' But Lipsius very prop- erly observes that the phraseology is not so precise as to warrant a definite conclusion to this effect.' On the other hand, the argvunents by which Lipsius attempts to prove from the litigation over the estate of Dicaeogenes that the conviction of several witnesses was necessary cannot be regarded as convincing.* Even if Menexenus and his fellow claimants had convicted several witnesses, it would not prove that this was a legal prerequisite.* But Wyse is unquestionably right in holding that one witness only, Lycon, was actually con- victed.* The circumstances were as follows: Dicaeogenes got his judgment in the case on the strength of the second alleged will, and the witnesses of whose false testimony the speaker complains were the witnesses to this second will; they it was whose testimony had been impeached by Polyaratus and who were subsequently made defendants by Menexenus.' After one of these witnesses, Lycon, whose case came first to trial, had been convicted, Dicaeogenes, realizing that the conviction of the others and the consequent ' 937B-D. ' 11. 45: Kc\€ici S' 6 vS/ws, iii.v a\Q tis twv \pev5ofuipTvpluv, TiXiv ii ipxv^ eXvai vepl airwv t4s Xi^ieis. Cf. Rentzsch, op. cit., p. 62; Leisi, op. dt., pp. 134-35. The theory originated with Buermann, ' 'Die Unechtheit der dritten Rede gegen Aphobus," A^. Jahrb., CXV (1877), 586. ' Recht, p. 958; cf. MSL, p. 982, n. 614; Wyse, Isaeus, ad. loc. * Recht, pp. 958 ff. ' Even if the conviction of a single witness was sufficient to reopen the case, it would obviously be to the interest of the litigant to convict as many as possible before the retrial of the main suit (cf. Buermann, op. cit, pp. 586 ff.; Leisi, op. cit., pp. 134-35). No one apparently has noted that this is supported by the procedure of ApoUodorus in his litigation with Phormion, since he is evidently prosecuting a num- ber of witnesses (Dem. 45. 7, 41 : Srar «' Isaeus 5. 12-13. 'Ibid., 14 ff. The correctness of Wyse's view that the plural (14: o! fidprvpct; 15: ol /MtpTupiJcraKTes) refers simply to Lycon, whose conviction is regarded as a vir- tual condemnation of all the witnesses to the will, is attested by 7i7>'o/i^i»ati' (15), regarding which Wyse observes: "no judgment had as yet been obtained against the other (will) , but the condemnation of one of the witnesses had damaged the credit of the document." The fact that Dicaeogenes thought it advisable to employ a Jio- imfiTvpla, instead of again putting forward the witnesses to the will, is of itself a strong indication that the witnesses were unwilling to face the conviction which was certain after the condemnation of Lycon. Lipsius, in his attempt to maintain the opposite view, is forced to take /«9' rjiiSv viXir lirpaTTe (14) of further actions against witnesses, when obviously it means no more than "was again on our side." 392 George Miller Calhoun lation if for the present we assume merely that the requirement was definite and quantitative.^ It is significant that Alexandrian procedure apparently was based upon the identical principle that has been developed from the Dicaeogenes case. The law says, "let the one who impeaches a deposition (6 ixaprvpias iirCkaixfiavdnevos) .... impeach all the witnesses who gave the same testimony (iwLKanfiaviadu de iravruv T&v ravrd. naprvpr)(Ti.vTcov), .... and let him be allowed to impeach even a part of the deposition" {i^iaru Sk Kai fxepovs rrjs ixaprvpias iiriXapiffOat.)-^ In the remainder of the law, and particularly in those sections which provide for retrial or the setting aside of a judgment, "the witnesses" {oi ndpTvpes, tuv ixaprvpuv, ktX.) are the defendants in the perjury action, the witnesses who attested the deposition or the part of the deposition impeached, and not all of the opponent's witnesses.' It is clear that the principle of the law is the one which has just been developed, that some portion of the evi- dence must be refuted in order to have a judgment set aside or a case retried. Furthermore, throughout that portion of the law which describes the customary steps in a typical perjury action, the singular p,ap- Tvpla, i) /xapTvpia, is used without exception;'' with the introduction of special classes of p.apTvpiai we find a sudden shift to the general- izing plural.' This insistent use of the singular strongly suggests that in the general run of cases a litigant was expected to impeach not more than one fiaprvpia, that the refutation of a single deposition, or even of a single point in a deposition, effected a setting aside of judgment. This is so manifestly just and reasonable, both theo- retically and practically, that one would be tempted to maintain it on a priori grounds even if there were no evidence in its support. For in many cases the issue turns upon a single fact, and even a single perjured statement can vitiate completely the justice of the decision which has taken it into account. And it is obvious that a litigant who had decided to bolster up his case by perjured testimony would 1 Cf. infra, pp. 393. ' P- Hal. I. 24-34. » Ibid., 44 ff., 55 ff., 64 ff. The provisions regarding the effect of perjury con- victions on the original suit have been given in substance in the discussion of the penalty (supra, p. 389). " 24, 29, 32, 39. ' 70 ff. '^iffiai\l/is AMD THE Alio; "^evSoftaprvpluv 393 not be so simple as to incur unnecessary expense and darker by introducing witnesses to swear to a number of comparatively imma- terial falsehoods. He would, if it were at all feasible, introduce in a single deposition, or even in one clause of a deposition, testi- mony to establish the essential point in his contention and would suborn witnesses to swear to that.' Only in the event that the nature of his case made necessary corroborative evidence would he run the risk of introducing several false depositions and incur the increased expense or trouble of securing numerous witnesses. In view of these general considerations and the phraseology of the Alexandrian law, the writer inclines to the opinion that in Attic law also the requirement for retrial was the successful prosecution of the witnesses to a single portion of the evidence. It has been assumed by some scholars that it was necessary, in order to secure a retrial, to show that the false evidence had influ- enced the verdict. Rentzsch believes that this question was passed upon in a separate action — Nullitaisklage — in which the court was petitioned to grant a retrial.* That no such special action inter- vened has been shown by Lipsius,' who in turn suggests, on the basis of Laws 937C i\l/ev8ij 8o^&vrwv fiaprvpeip Kal rijv v'ikijv t^ &<6vti ireiroMjKemi), that the presiding magistrate or the court empaneled for the trial of the perjury action pronounced upon this question.* But it is not Ukely that these words refer to a judicial decision, for if there had been a pronouncement of court on this point at the trial of the perjury action Plato would certainly not have made express provision for ai^ument and decision of the question in the retrial (aiMJ>i,icpiiTri