< Retour

In the Rhetoric (1.2.1355b35-39), Aristotle divides the means of persuasion (pisteis) into two broad categories: those requiring no skill (atechnoi) and those requiring skill (entechnoi). The former category includes the testimony of witnesses (martyres), information obtained by torture (basanoi), and written documents (syngraphai). In his book Témoigner et convaincre, N. Siron analyzes the role of witnesses in Athenian trials and the ways litigants used their testimony to convince judges. By neglecting the role of written documents, however, Siron offers a limited perspective on the methods of persuasion used in the Athenian courts. Previous studies of witnesses have concentrated on the procedural aspects such as who could testify and how testimony was submitted. By contrast, Siron takes a more cognitive approach (16) examines “le dispositif qui permet l’adhésion à une déclaration” (18. Cf. 30: “le système cognitif sur lequel s’appuient les orateurs pour valoriser leurs déclarations ou au contraire dévaloriser celles de leurs adversaires”). His study therefore concentrates on forensic oratory rather than deliberative and epideictic oratory but immediately runs up against an obstacle: we have only copies of speeches delivered in court, but in most cases do not know the outcome of cases nor the reasons why the judges voted to convict or acquit (25-26). This is actually too pessimistic: the litigants themselves often describe the reasons for verdicts – see my ‘The Athenian View of an Athenian Trial’[1] with a list of more than one hundred and fifty accounts of verdicts. To overcome this problem, Siron analyzes the lieux communs made by speakers in court about witnesses to elicit standard views about witnesses and their role (27-29).

The book is divided into three sections: The first chapter (“L’ère des témoins”) begins with a lexical study of the Greek terms μάρτυς, μάρτυρος and related words, which are used to designate people who remember information and provide it when requested as early as Homer and Hesiod. Outside forensic oratory, these terms are relatively rare until Xenophon and Plato (41‑48). In courtroom speeches, litigants make an implicit distinction between formal witnesses summoned to be present at legal transactions and accidental witnesses who are present at an event without being summoned (ἀπὸ τοῦ αὐτομάτου) (49-53). The terms associated with witnesses can also be applied to written documents (54). This analysis leads to the unsurprising conclusion that «Le lexique grec paraît très proche des mots français”témoigner”, “témoin”, et “témoignage”» (56-7). A large number of witnesses was considered an advantage (58-67) and the absence of witnesses a disadvantage (67-73), which is what one would expect. The table on page 67 is very misleading. The table lists cases in which one to six witnesses are called at a time, but misses several cases in which more than six are called (Andoc. 1.46 [prytaneis, who could be up to fifty]; Aeschin. 2.46, 107, 127 [probably eight ambassadors] 2.85-6 [proedroi, ten generals, representative of allies]). The assertion of Siron that the testimony of witnesses was considered superior (61. Cf. 87: “les témoins sont le moyen principalement choisi”) is also inaccurate, a point to which we will return. In modern trials, the opposing sides question witnesses to find the truth, but in Classical Athens litigants make assertions about the facts and call witnesses to confirm these statements (73-78). Siron briefly outlines the debate about the role of witnesses and notes that Humphreys and Todd believe that the contents of their testimony was less important than their identity[2]. As Rubinstein has shown[3], this is contradicted by the fact that roughly half of the witnesses are never identified and when they are identified, what is stressed is their ability to know the facts and the content of their statements. As Siron shows, witnesses are usually introduced by a phrase such as “to show that I am telling the truth, please read the testimony.” (79-87). The views of Humphreys and Todd are also contradicted by the existence of the private action for false testimony (dike pseudomartyrion), which shows that witnesses were duty-bound to tell the truth and could be penalized for lying, which is hard to explain if the role was only to show support[4]. Even informants who were granted immunity could be executed if their information turned out to be false (Andoc. 1.20), which shows that the contents were more important than the identity of the informant.

Chapter 2 (“La fabrique de la preuve”) examines the role of witnesses and seals in verifying documents. Siron claims that documents provided information but witnesses provided confirmation but fails to make a crucial distinction between private documents held by individuals and public documents kept in the Metroon and by officials. One searches in vain for any reference to J. Sickinger[5]. Private documents could be forged and required verification by witnesses because the Athenians had no official like a notary public. Even seals attached to documents were not infallible, and private documents might be entrusted to third parties or verified by witnesses (110-20). But litigants often support their statements with public documents like decrees, laws, letters from kings and officials, denunciations by informers, and records of import and export duties, none of which are examined by Siron, who makes no use of the important discussion in C. Pébarthe[6]. None of these documents is ever accompanied by the testimony of witnesses to prove their authenticity, which completely refutes Siron’s assertion that the Athenians were suspicious of written documents (“la suspicion des tribunaux envers l’ écrit produit seul”) (92). As Aeschines (2.89; 3.73) twice states, the Athenians placed more trust in public documents than in the oral statements of politicians, two passages one cannot find in Siron’s Index des sources commentées. The account of the plaint (engklema) (104) contains serious mistakes, claiming that this document was presented to the magistrate at the public arbitration. This is not true: the plaint was submitted to initiate all cases, both public (where there was not public arbitration) and private (where there may or may not have been public arbitration). Because Siron does not mention the Metroon and its role, he fails to mention the key fact that the plaint was kept on file in the public archive after the trial and could be used without witnesses as evidence for a verdict (Dem. 32.37; 38.6). The conclusion to this chapter[7] may be true for private documents but is not accurate for public documents and in general highly misleading.

The second part on the foundations of persuasion contains two chapters. Chapter 3 (“Le sacrement du témoignage”) examines the role of oaths. Oaths were sworn by the accuser, the defendant, and their witnesses on sacrificial victims in trials for homicide but rarely if ever in other trials (129-38). Siron does not explain the difference in procedure: for the reasons see E. M. Harris[8], a work not cited by Siron. Women might swear oaths about paternity in out‑of‑court negotiations but did not testify in court (138‑44). Siron misses the fact that male minors could also testify[9]. Litigants could also present a summons to testify to a reluctant witness, who could comply or refuse on oath (exomosia) on the grounds that he was not present or did not know (144-50). Siron repeats Thür’s unconvincing view that the witness who refused on oath stated that the contents of the statement drawn up by the litigant were false (149), but misses the detailed refutation by G. Martin[10]. Siron also misses the similar procedure attested in the convention between Stymphalos and Sikyon (IPArk 17, lines 10-14), which provides comparative evidence against Thür’s view. Litigants could bring a private action for false testimony against a witness, which might lead to the original verdict being overturned and a payment of damages and a form of atimia for the witness after three convictions (150-57). In this way witnesses were responsible for what they said although supporting speakers (synegoroi) were not vulnerable to this charge (157-62). This action is very important for understanding the role of the witness, but Siron does not analyze in detail any of the speeches in cases brought for this action and the ways in which testimony could be attacked. Slaves could be tortured to obtain information even though all the challenges to surrender slaves for torture mentioned in the extant orations were refused (162-9). Siron claims that in certain cases one cannot tell the status of those tortured (167, note 245), but this is not true because in two cases those tortured were clearly free metics or foreigners (Aeschin. 3.223-4; Dem. 18.132-3). Siron lists Dem. 23.28 as a case of an individual tortured, but the passage is not relevant to the topic. Witnesses had to testify about what they knew or saw and could not present hearsay evidence (Dem. 57.4). Witnesses who could not be present however could present their testimony to other witnesses who would report it in court (ekmartyria) (169-73). Siron ends the chapter with some good observations about the absence of a secularization of legal procedure in the Classical period (173-4).

Chapter 3 (“Le savoir du témoin oculaire”) studies the role of the eye-witness. In the historians, primarily Herodotus, sight is considered more reliable than hearing (177‑81), and one finds the same preference for eye‑witness testimony in the orators (181‑8), where the emphasis however is more often on the presence of a witness than what he saw (188‑99). Messengers in tragedy and comedy play a similar role by providing eye-witness testimony (199-205). The judges themselves could not witness events so became the “witnesses of witnesses” (205-210). One misses here a discussion of the term ep’autophoro used in cases in which a thief was caught with stolen goods, which were tied to his back and brought with him to the Eleven ([Arist.] Ath. Pol. 52.1 (one cannot find this passage in the index) with see E. M. Harris[11]. Unlike judges, private arbitrators might be chosen because they knew the basic facts in a case, but they too might have to question the litigants about key facts (210-6). Siron (211) rightly sees that the document at Dem. 21.94 is not authentic but paradoxically claims that “il semble transcrire l’idée générale de la loi.” He does not give the text of the document nor analyze its contents. (In note 143 he incorrectly states that the author of the analysis of this document[12] is Mirko Canevaro; it is actually Edward Harris.)

The third part examines “La transparence des tribunaux”, which is not directly related to the topic of witnesses. Litigants often appealed to the knowledge of the judges (“L’appel au témoignage des juges”) both about contemporary events and events in the past (225-57). In this way Athenian judges were unlike modern jurors, who are selected because they do not know about the case. Unlike witnesses, judges in large courts could not be held accountable. They swore an oath, which called down the wrath of the gods if they disobeyed their promises (257-63). Private arbitrators gave decisions after swearing an oath but could also not be indicted (263-5). The final chapter (“L’évidence des faits”) (267-309) discusses more topics not directly relevant to witnesses. Litigants often claim that “the facts speak for themselves” as a way of diverting attention from their rhetorical stratagems (270-5). Siron observes that litigants often accuse their opponents of digressing but misunderstands this criticism by failing to connect with the plaint (engklema), which listed the legal charges, and the promise in the judicial oath to vote only about the charges in the plaint. As Plato notes in the Theaetetus (172d-e), it was not the laconic style of speakers but the plaint that compelled litigants to “keep to the point.” Siron’s anthropological approach blinds him to the way in which legal procedures shaped discourse in court. Siron claims that the concise style of Socrates owes something to the courts but misses the way the Platonic Socrates criticizes forensic oratory (282-4). The rest of the chapter collects passages in which speakers claim to be inexperienced in rhetoric (284-91) or in trials (291-305) or avoid public life (305‑8). A brief conclusion summarizes the main points of the book (311-4).

There are two appendices. The first lists all cases of witnesses called in the extant speeches of the Attic orators with an indication of whether the witness statement is preserved in the manuscripts. Siron (329-30) claims that the documents at 23, 25, 28, 32, 34, 40, 47, 48, 61, 70, and 123 in the speech of Apollodorus Against Neaira are judged authentic by the majority of commentators, but never identifies these commentators and gives no bibliography. This is at best misleading, at worst very inaccurate. Westermann[13], considered all the documents in the speech forgeries, while J. Kirchner[14], followed by Drerup, considered the witness statements genuine because they contained names found in contemporary inscriptions. More recently, C. Carey[15], considered most of the documents genuine, but his analysis is very superficial, and some of his arguments from names are not convincing (see my review in Classical Review 44 [1994] 21-23). In the most thorough recent analysis, K. Kapparis[16], considers the documents at 23, 25, 28, 32, 34, 48, 54, 61, and 84 forgeries with good evidence. One might add that all the documents in the speech lie outside the stichometry, which Canevaro (op. cit. supra) has recently shown is good grounds for considering all of them forgeries.

The second appendix (315-38) lists all the speeches of the Attic orators with their proposed dates but does not provide any references to sources such as The Second Letter to Ammaeus of Dionysius of Halicarnassus. There is an “Index des sources commentées” (355-76) but no subject index, which makes the book hard to use.

In general, the book is well produced; I noticed only a few small errors: on page 93, the Greek word symbolaion is mistranslated as “contract”; on page 161 the accent on ἒλεγχος is wrong; on page 199 “manteamême ux” should have been corrected; on page 320 Apatouriois should be Apatourios; on page 353 the page numbers for Siron’s essay in Gagliardi and Pepe eds.[17] should be 267-89 and not 283-285.

Siron provides a good study of the use of witnesses in the Athenian courts and some standard commonplaces of forensic rhetoric. But témoigner was not the only way to convaincre used by litigants. In public cases litigants often rely on documents from the Metroon or from the archives of officials to prove the facts. Already in the fifth century Andocides in On the Mysteries and the accuser in Against Agoratus of Lysias (13) rely more on written documents than on witnesses. In speeches against inappropriate laws, the case is built entirely on written documents (Dem. 20 24). In Against Aristocrates there are also more documents than witness statements (Dem. 23), and at the trial of Aeschines in 343 both litigants place more weight on documents than witnesses (Dem. 19; Aeschin. 2). In his prosecution of Ctesiphon in 330, Aeschines (3) uses only documents and does not call a single witness. If there are many more witness statements in private speeches, it is because there were no public documents available for litigants to cite though in one rare case a litigant placed a copy of a will in the hands of the astynomos (Isaeus 1.14-15). In the first part of the euthynai procedure ([Arist.] Ath. Pol. 54.2. Cf. Aeschin. 3.9-31) officials submitted their written accounts to the logistai, who examined them carefully. To ensure accountability, extensive financial records were inscribed on stelai, many of which have been preserved in whole or in part. As Pierre Fröhlich[18] has shown, this practice was widespread throughout the Greek world. But one cannot find any reference to this book or to any inscriptions in Siron’s study. And in the absence of documents or witnesses, orators also used sophisticated arguments from probablity to convince judges (Aeschin. 1.90-92), a topic neglected by Siron. As a result, readers will get a misleading impression of the Athenian courts, which in Siron’s account seems closer to pre-literate societies in its dispositif de la vérité. Such are the risks of an anthropological approach to ancient Greece.


Edward M. Harris, Durham University

Publié dans le fascicule 1 tome 123, 2021, p. 284-288

Réponse de Nicolas Siron à la recension d’Edward M. Harris


[1]. E. M. Harris in C. Carey, I. Giannadaki, W. B. Griffith éds., The Use and Abuse of Law in the Athenian Courts, Leiden 2018, p. 42-74.

[2]. See S. Todd, “Witnessing in Athenian Law is a ritualised socio-political act of support”.

[3]. “Main Litigants and Witnesses in the Athenian Courts: Procedural Variation” in R. W. Wallace, M. Gagarin. éds., Symposion 2001, Vienna 2005, p. 99-120.

[4]. See C. Carey, “Legal Space in Classical Athens,” Greece and Rome 41.2, 1994, p. 183-184 : “the emphasis on matters of fact.”

[5]. Public Records and Archives in Classical Athens, Chapel Hill 1990.

[6]. Cité, Démocratie, et Écriture: Histoire de l’alphabétisation d’Athènes à l’époque classique, Paris 2006, p. 147-243.

[7]. “les documents n’ont aucune efficacité sans les témoins, mais aucune déposition n’aurait d’intérêt sans la référence préalable aux actes écrits”.

[8]. “The Family, the Community and Murder: The Role of Pollution in Athenian Homicide Law” in C. Ando, J. Rüpke éds., Public and Private in Ancient Mediterranean Law and Religion, Berlin-Munich-Boston 2015, p. 11-35.

[9]. See E. M. Harris, Democracy and the Rule of Law in Classical Athens, Cambridge-New York 2006, p. 355-364.

[10]. “The Witness’ Exomosia in Athens” CQ 58, 2008, p. 57-62.

[11]. Democracy and the Rule of Law in Classical Athens, Cambridge-New York 2006, p. 373-90.

[12]. In The Documents in the Attic Orators: Laws and Decrees in the Public Speeches of the Demosthenic Corpus, Oxford 2013, p. 231-233.

[13]. Untersuchungen über die in die attischen Redner eingelegten Urkunden, (Abhandlungen der Sächsischen Gesellschaft der Wissenschaften. Phil. Hist. Klasse), Leipzig 1850, p. 114-129.

[14]. “Zur Glaubwürdigkeit der in die [Demosthenische] Rede wider Neaira eingelegten Zeugenaussagen,” Rheinisches Museum 40, 1885, p. 377-386.

[15]. Apollodorus, Against Neaira, Warminster 1992.

[16]. Apollodorus, Against Neaira, Berlin-New York 1999.

[17]. In Dike: Essays in Greek Law in Honor of Alberto Maffi, Milan 2019.

[18]. Le contrôle des magistrats dans les cités grecques, Paris 2005.