Individual Conversions to Judaism in Late Antiquity:
Reconsidering the Epigraphic Evidence

di Joseph Sievers

1. Biblical "proselytes"

In the Hebrew Bible we read of two «converts» who may be considered emblematic. One is Ruth, the Moabite woman who chooses to leave her country, her kin, and her gods, to follow Naomi. Ruth says to Naomi: «Where you go, I will go; where you lodge, I will lodge; your people shall be my people, and your God my God» (Ruth 1:16). She leaves her earlier community and becomes an integral, even vital, part of the community of her choice.

The other case, differentiated by gender, social status and relation to the community, is that of Naaman, commander-in-chief of a Syrian army. After having been healed of a skin disease by the prophet Elisha, and when the latter refuses any gift, Naamansays: «... your servant will no longer offer burnt offering or sacrifice to any god except the Lord. But may the LORD pardon your servant on one count: when my master goes into the house of Rimmon to worship there, leaning on my arm, and I bow down in the house of Rimmon, when I do bow down in the house of Rimmon, may the LORD pardon your servant on this one count». Elisha replies simply by saying «Go in peace». (2 Kings 5:17-19 NRSV).

The stories of both Ruth and Naaman are masterpieces of Biblical narrative, in which unexpected reversal of fortunes is a central device. Historicity in the usual sense is not their main concern, but they do show some of the range of causes and consequences of what may be understood as religious conversions.

How does such narrative art compare with the prosaic evidence of those brief inscriptions that alert us to the real life and death of more ordinary proselytes? I will make some rather «lapidary» comments about the latter.            


2. The Epigraphic Record

Figueras has fairly recently attempted to give a complete inventory of inscriptional evidence for proselytism1. He lists fifteen inscriptions mentioning proselytes, plus the three persons included as proselytes in lists of (about) 125 people from Aphrodisias. To these should be added inscriptions from Caesarea2, Masada3, and Jerusalem4. One Jerusalem ossuary inscription, included by Figueras and others, seems to have a strong chance of being a modern forgery5. We are left with slightly over twenty cases in which the Greek term prosélytos or its Hebrew, Aramaic, or Latin equivalent appears in inscriptions from around the Mediterranean world and as far as Dura Europos.

A first question concerns terminology. The standard Greek dictionary by Liddell-Scott defines prosélytos as «one who has come over to Judaism, convert, proselyte». But in the second century C.E. and later the term occasionally refers to converts to Christianity6. Bagatti and Testa have claimed that the cemetery at «Dominus Flevit» and other cemeteries in Jerusalem were used by Judaeo-Christians and that the term on the ossuaries there and elsewhere refers to converts to Christianity7. This claim has met with strong opposition8. Clearly, Testa's explanation of many features of the ossuaries as Christian symbols is not convincing. The reference to literary use of prosélytos is interesting, but although Justin refers to «Christ and his proselytes» (ton Christon kai tous prosélytous autou, hémas ta ethné), he clearly uses the term primarily with reference to converts to Judaism9. Also, the specifically Jewish names are no obstacle to assuming that these proselytes were converts to Judaism, since in many other clearly Jewish proselyte inscriptions, either a name change or simply a Jewish name is recorded. There remains the fact that at least one of the ossuaries from the Dominus Flevit excavations seems to bear a chi rho sign. If correctly read and not added later it suggests a Christian connection10. For the moment, I do not feel in a position to exclude the possiblity that the Dominus Flevit site was used by Jewish Christians and thus consider inclusion of these ossuary inscriptions somewhat tentative.

More recently, Martin Goodman has claimed that at least in Matthew 23:15 the term prosélytos may refer to association with a particular Jewish group such as the Pharisees11. This interpretation, however, whatever its merits, does not apply to inscriptions, where, incidentally, Pharisees, Sadducees, Zealots, Essenes or other similar Jewish groups are never mentioned by name.

It is also noteworthy, that in inscriptions, even those that include multiple names, every proselyte is individually identified. As far as I know there is never any plural reference to proselytes in the epigraphic record. As is well known, the term proselyte is almost absent from the papyri12. Rights, and limitations of rights, of proselytes appear in several Qumran documents, but such questions are outside the purview of this paper.

It is impossible to guess how regularly proselytes had their status indicated in their epigraphic record. There were times, such as under Hadrian, when becoming a proselyte could lead to prosecution and punishment, although it might have been relatively safe to record the fact on one's tombstone13. Nevertheless the commemorator may have had to consider possible consequences. A widespread reluctance to make a conversion to Judaism public, may be suspected. The number of inscriptions is not great, but still substantial. I have found 28 possible proselytes. Of these, 14 are women and 14 are men. Nine women and all the men are designated by the term proselyte or what seem to be its various equivalents (although we should keep in mind that one of the men is mentioned in an inscription that may be a modern fake). In five cases the word for proselyte is not complete or its spelling is abnormal and could be understood differently. Thus we are left with 17 certain proselytes. Of these three are known from the Dominus Flevit necropolis and might be Christian.

The geographical distribution of the 28 individuals included is as follows: 7 from Rome, 6 from Jerusalem, 4 from Aphrodisias, 2 from Sitifis in North Africa, and one each from Masada, Wadi Muraba'at, Caesarea, Dura Europus, Smyrne, Termessos in Pisidia, from Cyrene, Venosa (southern Italy), and from Hungary.

The group is not large or homogeneous enough to do much statistical analysis, although several factors are striking. Of the seven proselytes recorded in Rome, five are women against only two men and five inscriptions are in Latin, whereas in general over 75% of the Roman Jewish inscriptions are in Greek, and almost everywhere men are more likely than women to appear in ancient records14.

Leon remarks:

Since five of the seven proselyte inscriptions are in Latin, whereas some three-fourths of all our Jewish inscriptions are in Greek, we may possibly draw the inference that proselytes were more frequent in the more Romanized element of the community. The examples are, however, so few that the preponderance of Latin inscriptions of proselytes may be due only to chance. The same may be said of the fact that five of the seven are females15.

3. Proselytes and Their Social Context

The factor I would like to concentrate on for the rest of this paper are the relationships maintained by the group of proselytes here presented. These relationships are of different categories. They may be expressed in words of affection, in references to family relations, to patronage. Relationships may also become apparent through the use of symbols, through the general context, and through loose association.

For example, it may be significant that an ostracon from Masada among a list of seven men who presumably belonged to the sicarii who held the fortress until it fell to the Romans in 73/74 C.E., lists a proselyte, the last person on the list. Of course one should not overinterpret such a fact.

More amenable to closer study are the famous lists of about 125 men16 from Aphrodisias in southern Turkey. The first list contains among many others also the names of three proselytes. All of them bear Biblical names: Samuel, Joses, and Joseph17. This last person also has his father's name listed as Eusebius. As is well known, there are numerous references in rabbinic literature to the opinion that a proselyte is like a newborn child, loses all previous family ties, and has no heirs18. Consequently scholars have tried to square the epigraphic evidence with these halakhic requirements19. While it is theoretically possible that Eusebius converted after Joseph was born but before he became a proselyte, such a situation is quite unlikely.

Several other cases suggest that family ties were not necessarily broken by the conversion to Judaism of one family member. Perhaps the clearest case is that of Cresce[n]s Sinicerius, a man who died at age 35 and who was called Iudeus proselitus on a tombstone set up by his grieving mother for her sweet son. Noy comments «Although the deceased man adopted Judaism, his mother did not adopt Jewish epigraphic style»20. Ross Kraemer has pointed out that a similar situation may be reflected in an epitaph set up by a mother for her daughter Septimia Maria Iudaea. The dedication to D(is) M(anibus) «the gods of the lower world» sets a pagan element at the beginning, which may occasionally occur in Jewish inscriptions. Since the mother does not identify herself as a Jew, neither by her name, nor by the use of Jewish symbols or phrases, it is likely that she is not Jewish, that her daughter instead is considered Jewish and that they nonetheless did not break their relationship21. The case of Aurelia Artemeis Ioudea seems to be analogous22. Her father Marcus Aurelius Ermaios, son of Keues, (himself) son of Keues, set up a funerary urn for his daughter, by herself. Also a brother/sister relationship seems to have survived one sibling's conversion. Mannacius commemorates his sweet sister C(h)rysis as a proselyte. Figueras23 suggests that Mannacius too had converted, but while this is possible, halakhic considerations should not be considered operative in these instances, unless proven otherwise.

More complicated is the situation in the family of the Avilii in Sitifis in Mauretania. Assuming that M. Avil(l)ius Ianuarius is indeed the same person in both inscriptions24, it is odd that he commemorated both his wife and his daughter as Iud(a)ea. He himself does not call himself Iudaeus but gives his title in one inscription as «pater synagogae». Perhaps the term Iudaeus/a or Yehudah was applied to persons who were not Jews by birth. As Tal Ilan has correctly noted it is the most frequently used Hebrew name of (male) proselytes. In our sample of 14 men it occurs 4 times. Of the 14 women in our sample, six are possibly called Iudaea, though only one fragmentary inscription refers explicitly to proselyte status at the same time. Ross Kraemer has cautiously made a good case for at least considering the possibility that the added name Iudaea may refer to proselyte status, especially where strong ties to the non-Jewish community can be detected, such as in the case of Rufina25. That women, including proselytes, became strong supporters of synagogues and other Jewish institutions may be inferred in the case of Rufina. It is quite evident in the case of Beturia Paulla (Paucla?), who became a proselyte at age 70, assumed the name Sarah, and was called mother of two synagogues or congregations26. Clearly as a proselyte she was well integrated into the wider Jewish community.

The function of patrons for their former slaves is highlighted in two inscriptions from Rome. Niketes was commemorated by his patroness Dionysias, whereas Felicitas, whose further name is not entirely clear, received her burial by her unnamed male patron. In other cases, the relationship of the deceased with the family that took care of their burial is ambiguous. From Cyrenaica we have an inscription commemorating several members of a Jewish family (one child is named Joses), last of whom is mentioned Sarah a proselyte, 18 years old. Perhaps she was a slave or had been adopted as a child, receiving a new name.

The most difficult case in this respect is Irene, a three-and-a-half year-old foster-child, called a proselyte. It seems that at least her mother was Jewish, but her father is mentioned too. It is not entirely clear if these are the foster-parents or the natural parents. About six different interpretations of the central part of this short inscription have been offered, but none is entirely convincing27. Family situations in antiquity could get almost as complex as nowadays and this inscription might be too brief to let us grasp that complexity.

We have only one case in which a woman is commemorated as a proselyte together with a man, presumably her Jewish husband (Aster and Paregorius). Certainly conversion of the wife on the occasion of marriage was a frequent occurrence in many places. Goodman, and even Feldman at one point, acknowledge that marriage may have been the main cause of conversions to Judaism. Yet our epigraphical record is nearly silent on this score, or we have not yet been able to ask the right questions. The work of Kraemer and of Ilan will certainly be an important basis for future research in this area.

4. Conclusions

I have tried to trace briefly some of the relations that become evident from a close reading of the somewhat meager epigraphic material for the question of conversion to Judaism. In my opinion it seems evident that many proselytes did manage to become active and accepted members of a Jewish community. This did not mean, however, that in every case all bridges to the past were cut. While the cases of rejection, that certainly occurred, leave practically no trace in the epigraphic record, the cases of conversion with continuing family relationships may have been due in some incidents to nearly simultaneous conversion by more than one family member. But in some families, different religious affiliations or at least different levels of religious commitment seem to have been accepted.


1. P. Figueras, Epigraphic Evidence for Proselytism in Ancient Judaism, in «Immanuel», 24/25 (1990), (Festschrift Flusser) pp. 194-206.

2. B. Lifshitz, Inscriptions grecques de Césarée en Palestine (Caesarea Palaestinae), in «Revue Biblique» 68 (1961) pp. 115-116, no. 2 (with photograph).

3. Masada I. The Yigael Yadin Excavations 1963-1965 Final Reports: The Aramaic and Hebrew Ostraca and Jar Inscriptions by Y. Yadin and J. Naveh, Jerusalem, Israel Exploration Society, 1989, n. 420, pl. 24.

4. T. Ilan, New Ossuary Inscriptions from Jerusalem, in «Scripta Classica Israelica» 11 (1991/92) pp. 149-159; ossuary 007.

5. E. L. Sukenik, Jüdische Gräber um Christi Geburt (Jerusalem 1931) p. 18, pl. 3 = Corpus Inscriptionum Judaicarum 1385. Milik (B. Bagatti - J.T. Milik, Gli Scavi del Dominus Flevit, vol. 1, La necropoli del periodo romano, Jerusalem, 1958, p. 84) claims that it is a forgery because the lettering is based on Greek print characters. L. Y. Rahmani (A Catalogue of Jewish Ossuaries in the Collections of the State of Israel, Jerusalem, The Israel Antiquities Authority, 1994, p. 12 n. 6) acknowledges Milik's doubts concerning the inscription's authenticity, without taking an explicit position on this point. Other authors accept the inscription as genuine, apparently without being aware of Milik's appraisal. So Figueras («Epigraphic Evidence» no. 5), Ilan (Scripta Classica Israelica 11 [1991/2] 154; P. W. van der Horst, Ancient Jewish Epitaphs, Kampen, Kok Pharos, 1991, p. 72; G. Mussies in Jan Willem van Henten and Pieter W. van der Horst, Studies in Early Jewish Epigraphy, Leiden, 1994, p. 260.

6. See G.W.H. Lampe, A Patristic Greek Lexicon, Oxford, 1961, s.v. prosélytos.

7. B. Bagatti, Scoperta di un cimitero giudeo-cristiano al Dominus Flevit, in «Liber Annuus» 3 (1952-53) pp. 148-194; Id., Alle origini della Chiesa, vol. 1, Vatican, 1981, pp. 213-214; E. Testa, Il simbolismo dei giudei-cristiani, Jerusalem, 1961, pp. 426-513.

8. See P. Figueras, Decorated Jewish Ossuaries, Leiden, 1983, pp. 82-86; Id., Jewish Ossuaries and Secondary Burial: Their Significance for Early Christianity, in «Immanuel» 19 (1984/85) pp. 41-57.

9. Justin Martyr, Dialogue with Tryphon, 122.5, cfr. 122.1-123.2, 23.3. For Archambault, the editor of a critical edition of the Dialogue, the terms geióras (a transcription of Aramaic gyora') and prosélytos are synonymous (ad 122.1).

10. See R. S. Kraemer, Jewish Tuna and Christian Fish: Identifying Religious Affiliation in Epigraphic Sources, in «Harvard Theological Review» 84 (1991) pp. 141-162, esp. 160-161.

11. M. Goodman, Jewish Proselytizing in the First Century, in J. Lieu, J. North, T. Rajak, The Jews among Pagans and Christians, 1991, pp. 60-63.

12. P. van Minnen (Drei Bemerkungen zur Geschichte des Judentums in der griechisch-römischen Welt, in «Zeitschrift für Papyrologie und Epigraphik» 100 [1994] 254 n. 7) mentions an unpublished 3rd cent. B.C.E. papyrus from the Duke University collection that contains the phrase tines tón prosélytón, but no other indication of Jewish origin. There are no known proselytes in Greek papyri or inscriptions from Egypt.

13. For discussions of Roman legislation see L. H. Feldman, Jew and Gentile in the Ancient World, Princeton Princeton Univ. Press, 1993, pp. 385-397; A. Linder, The Jews in Roman Imperial Legislation, Detroit, Wayne State Univ. Press, 1987, and my paper, Lo status socio-religioso dei proseliti e dei timorati diDio, in «Ricerche storico-bibliche» 8 (1996) pp. 183-196, esp. 188-190.

14. Tal Ilan (Notes on the Distribution of Jewish Women's Names in Palestine in the Second Temple and Mishnaic Periods, in «Journal of Jewish Studies» 40 [1989] 186-200) finds a total of 247 named women as opposed to 2040 men, a ratio of 8.3 men to 1 woman.

15. H. J. Leon, The Jews of Ancient Rome, updated edition with a new introduction by Carolyn A. Osiek, Peabody, MA, Hendrickson, 1995, p. 256.

16. Iaél, normally a woman's name (Judges 4:17-5:24) and possibly the only woman in the entire inscription, heads the first list with the (masculine) title of prostates. J. Reynolds - R. Tannenbaum, Jews and Godfearers at Aphrodisias, Cambridge, 1987, p. 101, argue that this Jael is a man.

17. J. Reynolds - R. Tannenbaum, ivi, p. 5. Tal Ilan's statement that Judah and Benjamin are the only Hebrew names recorded for proselytes therefore needs correction, although these Hebrew names are recorded here in Greek (New Ossuary Inscriptions fromJerusalem, in «Scripta Classica Israelica» 11 [1991/92] p. 155).

18. E.g. Mishnah Bekorot. 8:1; Baba Batra 3:3; Baba Kamma 4:7; Babylonian Talmud Yebamot 62a.

19. So Mussies (note 5 above) p. 260 and others.

20. D. Noy, Jewish Inscriptions of Western Europe, vol. 2, The City of Rome. Cambridge, Cambridge University Press, 1995, p. 393, no. 491.

21. R. S. Kraemer, On the Meaning of the Term 'Jew' in Greco-Roman Inscriptions, in «Harvard Theological Review» 82 (1989), pp. 41-42, following A. Scheiber, Jewish Symbols in Hungary, p. 45.

22. R. S. Kraemer, 1989: 43-45.

23. «Epigraphic Evidence», p. 200, no. 13.

24. Corpus Inscriptionum Latinarum, 8.8423, 8499.

25. R. S. Kraemer 1989:45-46.

26. D. Noy, Jewish Inscriptions of Western Europe, vol. 2, The City of Rome, Cambridge, Cambridge University Press, 1995, pp. 457-459, no. 577.

27. See D. Noy, Jewish Inscriptions of Western Europe, vol. 2, pp. 390-391, no. 489; R. S. Kraemer 1989: 38-41; Figueras («Epigraphic Evidence», no. 8) offers a halakhic solution.