Orit Malka: Response to Law and History Review Forum on “Disqualified Witnesses between Tannaitic Halakha and Roman Law: The Archaeology of a Legal Institution”

Orit Malka

Orit Malka is a Postdoctoral Research Fellow at Stanford University.

Ed. note: Orit Malka’s groundbreaking article, “Disqualified Witnesses between Tannaitic Halakha and Roman Law: The Archaeology of a Legal Institution,” appeared in Law and History Review, vol. 37, no. 4 (November, 2019) alongside commentaries by Christine Hayes, Amalia Kessler, and Paul Du Plessis. Below, Malka offers her response.

My article, “Disqualified Witnesses Between Tannaitic Halakha and Roman Law,” was fortunate to receive the attention of several prominent scholars in responses published in Law and History Review (vol. 37, no .4, 2019, pp. 937-959), and I am greatly indebted to all of them. The expert contextualization they so generously outline—with respect to the study of Jewish law vis-à-vis Roman law in the case of Christine Hayes and Paul du Plessis, and regarding the history of evidence rules and their rationale in the case of Amalia Kessler—is an extraordinary gift for which I am deeply grateful. The critical points they raise also present me with an invaluable opportunity to sharpen my argument and clarify its thrust. I take this opportunity to address here some of these points relating to core methodological aspects of the study of ancient procedure and the links between ancient legal regimes. There surely cannot be a better forum to discuss these fundamental questions of legal history, and for this I am sincerely thankful to the editorial board of Law and History Review and The Docket for hosting this discussion.

In my article, I proposed an analogy between early rabbinic laws of disqualification for testimony and the Roman legal concept of infamia. While Christine Hayes approves of this suggestion, Paul Du Plessis objects to the analogy in light of the differences between the two legal institutions. In particular, he highlights one difference relating to the scope of disqualification for testimony. As he writes: “[T]here was no blanket exclusionof a person from acting as a witness in a Roman court by virtue of that person being subject to infamia […] neither litigants (nor their representatives), nor witnesses were ever excluded from access to a Roman court on account of being infames” (953). By implication, Du Plessis seems to suggest this is an important feature differentiating Roman law from tannaitic Halakha, which enumerates a list of people disqualified for testimony and other witness roles—the list which is the subject of my original article.

Before one can evaluate the force of this objection, it is important to note that what is phrased by Du Plessis as a descriptive statement may be more accurately characterized as an interpretive conclusion based on the surviving evidence—one that depends on a particular, contingent understanding of the materials. Moreover, this understanding rests on certain presuppositions regarding the role of witnesses in legal procedure that are seldom made explicit in the scholarship and deserve critical examination. According to these implicit presuppositions, when reading and interpreting ancient laws relating to testimony, one should assume that witnesses  fulfill a purely probative role in the legal procedure. However, it is precisely this assumption that I hoped to call into question in my article. Kessler clarifies this point in her response (p.938-939):   

The historiography of modern Western legal systems […] is structured around a narrative in which earlier, ritualistic conceptions of litigation and procedure […] gave way to a modern conception of these devices as aimed at deciding the case in accordance with the law and the verified facts in dispute. Moreover, as this modern conception emerged, the role of witness testimony came to be defined in narrowly probative, evidentiary terms […] The development of evidentiary rules that are designed to constrain the role of witness testimony to that of supplying the relevant facts is part and parcel, in short, of the rise of modern courts devoted to the legal (rather than political) function of applying law to the verified facts […] Accordingly, there are grounds for concern that the effort to trace the origins of modern rules and practices may at times lead to anachronism, obscuring the range of different functions once served by evidentiary and procedural devices, and indeed, more broadly, by litigation itself.

In other words, to avoid anachronism in interpreting ancient sources, one ought to question any presuppositions regarding the role of witnesses in ancient litigation. The objection put forward by Du Plessis seems to me to be grounded in such presuppositions, against which I wish to argue. These assumptions are embedded in his reading of the sources, as I will explain. But in order to do so, I will first take a step back to deal with the fundamental question raised by Du Plessis: whether Roman law did or did not disqualify people subject to infamia for court testimony.


To be sure, there is no explicit statement tying together disreputable occupations or other infamous practices and disqualification for testimony in the extant Roman legal materials. There is no doubt about that. Nevertheless, scholars studying infamia have reached the conclusion that this status resulted in disqualification for court testimony, for reasons that deserve further attention. Notably, Roman jurists attest to an explicit debarring from testimony of people who are elsewhere subjected to infamia – although the sources do not link these two facts explicitly. Thus, Callistratus notes that the lex Julia de vi mentioned a class of individuals who were forbidden from giving evidence under this law; his list includes, inter alia, anyone who had been condemned in a iudicium publicum and had not been reinstated, anyone who had hired himself out to fight beasts, anyone who had been convicted of taking money for the purpose of giving or refusing to give evidence, and any past or present female prostitute (D.[1] This list includes “a class of individuals, closely resembling those of the Lex Julia Municipalis and the praetor’s Edict”, where their classification as subject to infamia is clear.[2] Are these two facts interconnected? Most scholars believe they are, despite the fact that no explicit mention of infamia is made in the clauses disqualifying those people from testimony. The same is true with regards to people condemned under Lex Julia Repetundarum and Lex Julia de adulteriis: they were sanctioned with infamia and, likewise, disqualified for giving testimony. Again, these facts might very well be connected, and not coincidental.

One reason to assume that these sanctions are indeed related to the status of infamia is the connection drawn by several sources between the disqualification for testimony in courts of law and the ban on serving as witnesses to wills. Such a link is made by Paul, when discussing the sanction placed on a person found guilty of corruption, who “cannot witness a will or give evidence” (D.22.5.15). Papian draws a similar connection between the two sanctions when referring to the legal outcomes of adultery (D. 22.5.14). In the statements of both jurists, the two forms of disqualification of witnesses seem to go hand in hand. Now, let us recall that, in several sources, ill repute is connected explicitly with the sanction prohibiting serving as witness to a will—unlike the implicit connection between infamia and disqualification for court testimony described above. Thus, according to a provision preserved by Aulus Gellius, already in the Twelve Tables we find that:

Whosoever shall have allowed himself to be called as a witness or shall act as one who balances the scales (libriprens), if he refuses to provide his testimony as a witness, he must be dishonored as morally corrupt (inprobus) and incapable of acting as witness (intestabilis) in the future (Gell. Att. 15.13.11).[3]

The sanction of being intestabilis, i.e., disqualified for witnessing a will,[4] is intrinsically connected here with the title improbos, one of many ways of indicating social dishonor of the kind that later became more standardly known as infamia.[5] This association appears to be stable over the years, as we find it again in the Justinian Institutiones (Ιnst. II.10.6). Notably, the fact that we find here “improbus” as the label denoting social disgrace and dishonor, and not, say, “infamis,” is an important reminder of the terminological indeterminacy of this civil-legal degrading, referred to in scholarship under the unifying title of infamia. In fact, several other terms are also used in similar contexts: “variations of improbus, probum, ignominia, nota, notare, and notatio, as well as many appearances of fama in its negative form”.[6] This is not merely a linguistic issue: “The diverse terminology reflects the lack of a clear and consistent legal definition or meaning for the concept”.[7] Given this feature of infamia, one should not be surprised that the indications of a connection between this status and disqualification for testimony are partial and blurred. The entire mechanism and its legal meaning were not ‘strongly defined’ (in the mathematical sense); it was more of a living idea than a fixed institution. Therefore, to determine that infamia was linked with disqualification for court testimony, one may rely on tacit and complex indications, if their conjunction produces a sufficiently unambiguous picture.

On this point, Abel H.J. Greenidge writes: “We shall find many civil disabilities imposed on individuals in consequence of some stain on their moral character or some defect in their social standing; but how far the persons so enumerated are coincident with the list made out by the praetor and developed by the imperial legislation is in most cases a matter of doubt […]. But it is clear that a civic disability, recognized by the State, and resting on moral grounds, is an integral part of the history of Civil Honour, and as such these cases require discussion, whether we believe that, in treating them, we are dealing with the infames proper or not.”[8] This is the kind of logic that leads Greenidge to determine that infamia—or whatever we choose to call the general phenomenon of degrading a person’s legal status due to moral deficiency—was intrinsically connected with disqualification from court testimony.


Du Plessis seems to view the conclusion reached by Greenidge and others as mistaken, in light of the following general statement by the third-century Jurist, Modestinus:

The value of testimony depends on the dignity, faith, morals, and gravity of witnesses. Hence, those who depart from their previous evidence are not to be listened to (D.22.5.2).

Du Plessis takes this quote to indicate that Roman law knew no rules of inadmissibility, no “blanket exclusion,” as Du Plessis calls it, of infames’ testimony. Whether this is the correct understanding of Modestinus’ note is questionable (is it possible that, in his opening statement regarding considerations that should be taken into account when evaluating testimony, Modestinus only means to introduce the relevant background for his conclusion—namely, that one who departs from a previous testimony is not to be listened to—with  no intention of establishing any general claims regarding inadmissibility rules? The phrasing of the quote might suggest that). Nevertheless, even if we accept this reading, what we are left with is an apparent contradiction between this statement by Modestinus and indications of a priori disqualification of certain people from giving testimony (whether due to infamia or not), mentioned by Callistratus (D., Paul (D.22.5.15), Papian (D. 22.5.14), Marcellus (D. 1.9.2) and Venuleiuss Aturninus (D. What is the meaning of this apparent contradiction? It seems to me that the best way to understand the relationship between the statements by different jurists is to posit development in the conception of the role of witnesses in Roman law.

As explained carefully by Kessler in her response, the rationale of court testimony is too often regarded as self-evident, stable and unchanging, and the possibility that it might have undergone radical transformation over time or in different cultures is seldom given appropriate consideration. However, some important works have pointed to ancient contexts in which the role of court witnesses must have been different from the conception common today. In these ancient depictions, witnesses seem to exercise an authoritative function that includes making decisions, rather than an instrumental role in which they merely provide information to a third party—a judge—who is authorized to render a decision.[9] If we take these depictions seriously, we ought to conclude that a conceptual evolution necessarily took place in the meaning and function of court testimony, either at this point or at another, and the history of this shift needs to be told. In the context of Roman law, this very change was already alluded to, albeit briefly, by Yan Thomas. In an article exploring the legal status of women in Roman society, Thomas describes what seems to be such a transformation in the conception of witnesses’ role:

In the earliest period of Roman law women could not make wills, because to make a will one had to belong to a comitia, or political assembly. Nor could they testify in court. Under archaic procedure testifying was a virile office because the entire body of citizens had to vouch for a claim in order for it to stand up in court…women were of course excluded from this public function. They ceased to be excluded when the meaning of testimony changed from vouching for the existence of a claim guaranteed by all citizens to simply providing evidence. Women could testify because their testimony no longer had the abstract force of a kind of mediation, the general import of an officium.[10]

Here Thomas draws attention to a shift in the meaning of court testimony in Roman law, from one associated with an office—an authoritative political standing—to that of mere evidence, or the provision of information.  Notably, the earlier meaning of testimony portrayed by Thomas echoes the ancient conception of testimony mentioned above, whereas the latter, in which testimony is grasped as simple provision of evidence, is similar to testimony as conceived of today. This conceptual transformation resulted, according to Thomas, in at least one change in the laws governing inadmissibility of witnesses’ testimony: women were barred from court testimony as long as it was considered a virile office, but they were permitted to testify when testimony came to be perceived as mere evidence.[11] Could it be the case that the same change also affected other norms related to inadmissibility, like those regarding infames? Is it possible that they, too, were initially disqualified for court testimony, in the same way as they were excluded from witnessing wills and from serving in political office, but once court testimony was no longer perceived as bearing authoritative standing, their status changed, so that the prior blanket exclusion was replaced with judicial discretion of the kind portrayed by Modestinus?

Scholars have pointed to additional indications suggesting that a change occurred in the Roman legal conception of witnesses’ role roughly during the first century CE,[12] and the debate regarding this question is far from exhausted. Until we have a better picture of this shift and its causes, we ought to be careful which presuppositions we take for granted when dealing with ancient sources on witnesses and testimony. A clear-cut separation between the rules regarding court testimony and those relating witnessing documents, among them wills, might be guided by just such an anachronistic presupposition regarding the role of witnesses in courts. After all, what is the justification for assuming that the role of witnesses in the two procedures was different? Such an assumption is not supported by any terminological distinction either in Latin or in Hebrew.[13] Obviously from the perspective of a modern conception of witnesses’ roles, the function of these two types of witnesses is different; today, we think of witnesses in court as merely providing information to the judicial authority—and not as bearing any substantial authority of their own—whereas with regards to authorization of documents, we are less invested in this claim. Du Plessis openly admits to holding a modern perspective on this point, writing that, “as in modern law, the value of the evidence [in Roman courts of law] was judged with reference to its credibility.” (952; emphasis mine). His response does not touch at all on the disqualification of infames or others from serving as witnesses to wills, implicitly assuming that this should be regarded on different grounds. But what if this division is only the product of a conceptual change in the role occupied by witnesses in court, as Thomas suggests? It is my contention that we ought to avoid presupposing this conceptual separation when it comes to the ancient sources, especially when these sources indicate a connection between the two witness functions, as is the case in the writings of the classical Roman Jurists. If we consider court testimony and the role of witnesses to wills (and other documents) in tandem, we obtain a picture that clearly connects infamia with disqualification for witnesses’ roles, in their various forms.    

Going back to the comparison between rabbinic and Roman rules of disqualified witness, it should be noted that the disqualification of the four characters enumerated in the rabbinic context is by no means limited to court testimony—it applies similarly to serving as a witnesses to legal documents. The Hebrew word ‘edut [עדות] refers equally to both witness roles – a point I wish I had emphasized more in my article. Thus, the tannaitic norm echoes the Roman prohibition on infames from witnessing wills. Moreover, women are disqualified from serving as witnesses to formal documents, according to both tannaitic and Roman rules. Therefore, even if there is a difference between the tannaitic and Roman contexts in the exact scope of disqualification of men subject to infamia, the broader picture shows a strong tendency in both normative systems to limit the capacity of such men in the context of witness roles, and to associate their status with that of women. This is all that I have argued for in my article. This shared tendency should be examined alongside other similarities between the tannaitic and Roman mechanisms, in order to evaluate the argument for a possible historical connection between them.

Here I come to a point raised by Kessler regarding philological methodology and the possibility of determining the grounds on which the rabbis disqualified women from giving testimony based on the sources. In my article I made use of Josephus’ reflections on the subject—but why should one consider Josephus a reliable source of information regarding the rabbinic view? Clearly, Josephus is inclined toward Greco-Roman ideas and ideals. This would be a valid criticism had I relied on Josephus alone in this matter, however it is precisely the philological method that changes the picture. I was only able to make use of Josephus as a source in this case because of the identical formulation used by both Josephus and the Tosefta in this context. The Tosefta, an internal rabbinic text, states that women are disqualified for testimony because they are suspected of testifying “out of temptation or out of fear” (מתוך הפיתוי ומתוך היראה). An almost identical phrasing, albeit in Greek rather than Hebrew, is used by Josephus with regards to the disqualification of slaves, who “do not bear witness to the truth, whether because of gain (διὰ κέρδος) or because of fear (διὰ ϕόβον).” This formulation may seem coincidental in a rabbinic work, but when we find it in Josephus we can safely see it as grounded in the Greco-Roman context of the ethics of self-control. Therefore, the philological examination of rabbinic texts is precisely what enables and informs the turn to Josephus for contextualizing the rabbinic stance.

To sum up, it seems to me that philological considerations suggest a close affinity between tannaitic laws of disqualification for testimony, the Greco-Roman political ideal of self-control, and the Roman legal mechanism of infamia. Clearly, I am not arguing for a direct parallel in the way that the Roman and tannaitic mechanisms operated on the ground, but rather for a strong literary resonance between the extant sources that relate to these mechanisms. For this argument to stand, a clear-cut, blanket exclusion of all infames for testimony, the absence of which is stressed by Du Plessis, is absolutely not necessary. Likewise, it would not be necessary for all sources to monolithically connect women and infames, nor for them to unanimously declare all women weak, in order to support the claim, established in previous scholarship, for a link between infamia, weakness and femininity. Rather, it is sufficient that some sources clearly make these connections, as my contention is merely that the rabbis interpret the legal institution of infamia as they understand it, probably based on its manifestations in aristocratic circles, while adapting it to their own needs. Given the structural, ideological and technical parallels between the two institutions (recall the phrase חזרה גמורה – which is a literal translation of restitution in integrum, suggesting that some reference to Roman legal institutions does hover in the background), I believe that the claim for a strong affinity between them, exceeding mere chance, still holds. But even if I am wrong on that point, I have hopefully demonstrated that the jurisprudential rationale of testimony in antiquity deserves further study, as it challenges common assumptions regarding the logic of testimony as mere evidence.

[1] Most scholars accept that the disqualification of witnesses according to this list is due to infamia. See, e.g., Jane F. Gardner, Being a Roman Citizen (London: Routledge, 1993), 123; Peter Garnsey, Social Status and Legal Privilege in the Roman Empire (Oxford: Clarendon Press, 1970), 231. Textbooks on Roman law regularly state that infamy resulted in disqualification for court testimony; see note 5 herein. I wish to thank Lorena Atzeri and Hartmut Leppin for their helpful advice in this matter, and for reading and commenting on the response. Of course, the fault for any mistake that might still be found in it is solely mine.

[2] Abel H. J. Greenidge, Infamia, its Place in Roman Public and Private Law (Oxford: Clarendon Press, 1894); 166.

[3] Translation from Jeffrey A. Stevens, Staring into the Face of Roman Power: Resistance and Assimilation from behind the ‘Mask of Infamia’ (PhD. Dissertation, University of California, 2014), 49.

[4] At least according to a limited interpretation. See Adolf Berger, Encyclopedic Dictionary of Roman Law (Philadelphia: American Philosophical Society, 1953), s.v. intestabilis. Some interpret it more broadly, as referring also to serving as a witness in court—see Gardner, p. 118. This might be the plain reading of Gaius, D.28.1.26.

[5] See the analysis of the connection between the different terms by Stevens, p. 49-54. And see also Paul Du Plessis, Borkowski’s Textbook on Roman Law (Oxford 2010), 106 : “Under the twelve tables, persons who where involved in discreditable conduct could not act as witnesses, either in litigation or in formal transactions (such as the making of wills […])”.

[6] See Stevens, 46n58, based on Greenridge, 2-6. -,.

[7] Stevens, 46.

[8] Greenidge,161-162.

[9] For a description of witnesses functioning authoritatively, as pseudo- judges, in an Ancient Near Eastern context, see Simonetta Ponchia, “Witnessing Procedures in the Ancient Near East: Problems and Perspective of Research,” in Witnessing in The Ancient Near East (Simonetta Ponchia & Nicoletta Bellotto eds., Padova, 2009) 225 -251, 242-243; cf. Shalom E. Holtz, Neo-Babylonian Court Procedure (Leiden: Brill ,2009) 296-300. Throughout most of his book, Holtz assumes that witnesses’ testimony is simply one kind of evidence, but in certain cases he is forced to withdraw from this assumption, when the procedure seems to be resolved by the presence of witnesses alone, with no judicial forum present. For the role of court witnesses in the Hebrew Bible, see Pietro Bovati, Re-Establishing Justice: Legal Terms, Concepts and Procedures in the Hebrew Bible (105 JSOTSup; Sheffield, 1994). Bovati depicts a system in which judicial disputes are resolved by a third party whom he describes as “witness-arbiter” who “does not pass a law court judgment, but the invocation of one has the same practical effect” (p. 82). In the Greek context see David C Mirhady, “Athens’ Democratic Witnesses”, Phoenix 56 (2002): 255-74, 264-265. Mirhandy describes certain circumstances in which witnesses must have acted as “surrogates for the court itself”. In the Roman context, see Alan Watson, International Law in Archaic Rome: War and Religion (Baltimore: Johns Hopkins University Press ,1993),10-19. ‏ According to Watson, in several texts the word testis means “judge” and not “witness”. He reaches this conclusion because of the authoritative function of testes implied by these texts. The scholarship in this field is far from exhaustive, and the significance of the accumulated findings from various cultural contexts requires further scrutiny.

[10] Yan Thomas, “The Division of the Sexes in Roman Law”, A History of Women: From Ancient Goddesses to Christian Saints (Cambridge MA: Harvard University Press, 1992), 83-138, 137.

[11] Contrary to Du Plessis’ statement that “gender was not a bar to acting as a witness” (953). One indication noted by Thomas for the claim that women were previously barred from court testimony is the fact that the vestal priests were granted the capacity to testify in what seems to be a special privilege denied to regular women who did not belong to this select group (ibid, n. 166). For a detailed discussion of the sources referring to this prerogative, see Robin Lorsch Wildfang, Rome’s Vestal Virgins, (London & New York: Routledge 2006), 67-69. Notably, the very same logic is used by Paul (D. 22.5.18) and Ulpian (D. to argue for the opposite conclusion, i.e., that women were, at a certain point, accepted as court witnesses. They note that certain women were disqualified for testimony according to the lex Julia de adulteriis, and this must mean that other women were allowed to testify. More important for our context, the very fact that Paul and Ulpian even address the question of women’s eligibility for testimony shows that it was not evident or obvious that women were considered legitimate witnesses in their time, again reinforcing the claim that the status of women must have changed around this period.

[12] Elizabeth A. Meyer, Legitimacy and Law in the Roman World: Tabulae in Roman Belief and Practice, (Cambridge UK: Cambridge University Press, 2004), 118-119 and notes 111-112. Meyer discusses the change in the conception of testes authorizing acts of private law, and their gradual merging with signatores of documents, around the first century CE (p. 158).

[13] In both contexts it is עד in Hebrew, testis in Latin. Moreover, in the Hebrew Bible, the language used to describe the authorization of legal transactions and the role of witnesses in judicial proceedings is the same (see Det. 19.16, יקום דבר, and cf. Ruth 4.7 לקיים כל דבר); Gene M. Tucker, “The Legal Background of Genesis 23,” Journal of Biblical Literature  85.1 (1966): 77-84, 83.