Cursing the Christians? History of the Birkat HaMinim

I suggest that we need to address this evidence with immense caution, admitting that while it is abundantly clear that the birkat haminim eventually was, in its medieval forms, a curse of Christians, we simply cannot document this in the period of its putative origins in the late first or early second century CE. It therefore cannot be identified with any certainty as a player in the “partings of the ways,” although it certainly served later to reinforce the boundary between the two communities.

See Also: Cursing the Christians? A History of the Birkat HaMinim (Oxford University Press, 2011)

By Ruth Langer
Associate Professor of Jewish Studies
Theology Department
Boston College
January 2012

The birkat haminim has been an enigma for centuries and an issue for Christians from at least the time of the Church fathers Epiphanius and Jerome. In the late fourth century, they both describe Jews cursing Christians (or Jewish Christians) three times a day in the synagogue, using words that are obvious cognates of the Hebrew terms minim and notzerim. Particularly after the publication in 1898 of the oldest-known texts of this prayer from Cairo (now dated early second millennium CE) that actually named notzerim (Christians or Jewish-Christians), there has been a positive fascination with this prayer. Where did it come from, and did it play a role in the parting(s) of the ways between Judaism and Christianity?1

According to the Babylonian Talmud (Berakhot 28b), the rabbinic patriarch Rabban Gamliel II, i.e., c. 100 C.E. (+/- 20 years), called for the institution of this prayer as part of the central element of the liturgy then being instituted to compensate for the lack of Temple sacrifices; error in this particular prayer by the precentor was particular cause for concern because it suggested heretical leanings. According to later traditions, the motivation for this addition was challenges generated by minim to the wellbeing of the Jewish community. Because a precentor had to recite this curse precisely, one who was a heretic or sectarian would not only curse himself, but the “amen” of the gathered community would affirm the curse. The effect was to exclude these minim from serving as precentor and hence from participating in the prayer community. Drawing on the work of J. Louis Martyn,2 a generation of New Testament scholars accepted that this was the scenario behind the descriptions of Christians being evicted from the synagogue in the Gospel According to John 9:22 (and 12:42, 16:2).3

But is this story correct? All of the rabbinic texts report that it hails from centuries after the purported date of this event. It presupposes a sufficiently significant Christian presence in the time of Rabban Gamliel to generate such a rabbinic response, and it presupposes a sufficiently influential rabbinic presence at that time to immediately influence actual live synagogue practice, not just in the centers of rabbinic life, but elsewhere as well. In my recent book, Cursing the Christians?: A History of the Birkat HaMinim,4 I suggest that we need to address this evidence with immense caution, admitting that while it is abundantly clear that the birkat haminim eventually was, in its medieval forms, a curse of Christians,5 we simply cannot document this in the period of its putative origins in the late first or early second century CE. It therefore cannot be identified with any certainty as a player in the “partings of the ways,” although it certainly served later to reinforce the boundary between the two communities.

What can we say about its early history?

  • Archaeological and textual evidence demonstrates that cursing itself was a widespread mode of communication in the ancient and medieval worlds and that curses served not only to invoke punishment on someone who had transgressed but also to warn people from potential acts of transgression. The ritualization of a curse would have seemed normal, not shocking, in those days.
  • Synagogues were widespread, even before the destruction of the Temple. However, their main function seems to have been the reading and teaching of Scripture (as is evident in the New Testament’s descriptions of them), not prayer. When most synagogues began to integrate rabbinic forms of regular daily prayer is unclear.
  • The central weekday rabbinic prayer, recited at each of the day’s three services and containing at least by the third century a prayer, concluded with a praise of God as one who “humbles the arrogant/insolent,” language that appears in every known form of the birkat haminim. However, this is the only language of the prayer that can be verified this early. This blessing/curse also functions within the section of the prayer that petitions God for the elements needed for restoration of an independent Jewish state, complete with a functioning sacrificial cult. In that messianic- redemptive scenario, the birkat haminim is, in known forms, the only curse asking for the removal of elements that prevent the messianic drama from occurring.
  • Epiphanius in his Panarion 29:9, written 374-377, describes the Nazoraeans, a group of Jews who accept Jesus as the Christ, as people whom the Jews curse and anathematize three times a day in the synagogue. This suggests the context in which the birkat haminim occurs. However, he does not mention that Jews curse gentile Christians in this way.
  • Jerome, in a series of letters and commentaries dated 404-410, describes the Ebionites, a Christian heretical group who identify themselves also as Jews, whom the Jews refer to as Minaeans or Nazaraeans and curse. This language apparently corresponds to the prayer’s curse of notzerim and minim. He also accuses Jews of blasphemously cursing all Christians as Nazaraeans three times a day.6
  • We know that the texts of the birkat haminim found in the Cairo geniza, a storehouse of worn Hebrew manuscripts the oldest surviving of which date from the late first millennium CE, consistently contain this curse of “notzerim and minim,” i.e., of (Jewish?) Christians and heretics/ sectarians.7
  • The geniza texts of the birkat haminim consistently begin with a curse of meshummadim, probably apostates, although rabbinic texts also indicate that the term can apply to people who simply reject elements of rabbinic authority. Curiously, there is essentially no evidence that this term ever applied to Jews who converted to Islam, and there is significant evidence that Arabic-speaking Jews understood it through a convoluted etymology to apply specifically to baptized Jews.8 This may indicate that the texts of the prayers emerged before the Arab conquest in the seventh century. Similarly, these texts all contain a curse of the “insolent/arrogant empire,” praying that it will be uprooted and otherwise destroyed.9 Were these elements part of this prayer in the time of Epiphanius and Jerome, one might reasonably expect that they would object to them as well, but we hear nothing about them from Christians until the late Middle Ages.

This suggests that the prayer was gradually evolving and did not reach the forms with which we are familiar until well after Christianity became an established political force in the Land of Israel and beyond. Indeed, the silence of Christian sources may even suggest that the more vehement forms of the prayer, none of which had a later presence in Europe, may have emerged under Islamic rule.

This hypothesis suggests that we cannot employ the evidence suggested by many for the earlier life of the prayer; they over read the sparse sources to construct a history that cannot be justified. The following points must be taken into account:

  • The story of the implementation of the birkat haminim under Rabban Gamliel appears only in the Babylonian Talmud, a text that was redacted half a millennium after the “event.” A shred of the story appears also in the fifth-century Jerusalem Talmud, but it lacks the critical details. Today’s students of rabbinic literature are very aware of the degree to which redactors reworked narratives to fit their own purposes. This particular story is very Babylonian, inserted to explain how the eighteen benediction prayers became one of nineteen benedictions; there is ample evidence that the Jews of the Land of Israel never developed a nineteen-benediction prayer, whereas the Babylonians (who became the dominant authorities for world Jewry) did. We must therefore be extremely cautious about presuming the historicity of this narrative.
  • We simply do not know who the original minim were. The word itself simply means “kinds” or “sorts,” and hence heretics or sectarians. A careful study of its usage in rabbinic literature makes it obvious that the meaning of the word shifted over time and with changing social circumstances. By the High Middle Ages, it definitely referred to gentile Christians, but it is not clear when this meaning developed. If it means “Christians,” then why do the geniza texts of the prayer consistently pair it with notzerim, which more clearly refers to Christians, or at least Jewish-Christians?
  • Similarly, a presumption that a rabbinic decree, if it is itself historical, could have been circulated and implemented immediately c. 100 C.E. presumes both recognized rabbinic authority and perhaps also modern communication systems. Although rabbinic texts portray themselves as having this authority, other sources, including archaeological evidence, are less clear. It is safer to assume that it took centuries for rabbinic teachings to shape Jewish life widely, especially outside their immediate environs.
  • If the Gospel According to John reflects a community in Antioch, as many suggest, there is no reason to presume that the synagogue there was rabbinic in the late first century. Indeed, several centuries later when John Chrysostom preached his infamous sermons against judaizing Christians, the events he described that were attracting Christians to the synagogue suggest that it was not rabbinic even in his day.
  • Many point to Justin Martyr’s critiques of Jews cursing Christians as evidence for the birkat haminim in the mid-second century. However, here we need to remember the ubiquity of cursing in the Greco-Roman world; it is wrong to assume that any undefined Jewish cursing refers to this specific prayer. Indeed, most of Justin’s descriptions are of Jews cursing Jesus, not their contemporary Christians, and of their doing so after their prayers, not in the course of them.10 A more cautious reading suggests that Jews were simply participating in the polemical context of their day (of which Justin’s Dialogue with Trypho, the locus of these references, is a prime example).
  • Many presume that the geniza text published by Solomon Schechter in 189811 represents the original form of the prayer. This presumption includes several problematic layers. First, it requires that rabbinic liturgy was deliberately composed and promulgated rather than developing in a more oral and organic manner. This model is very much a matter of dispute among scholars of Jewish liturgy.12 Second, this presumption retrojects the text itself about a millennium from the approximate date of the manuscript to its putative composition in the time of Rabban Gamliel. Given that the full evidence of the geniza has proven to be much more complex than that published by Schechter (with seven identifiably different versions of the prayer, two of which reflect a rite of the Land of Israel, the rest from Babylonia), it is difficult to claim that this text accurately bridges the centuries unchanged. Indeed, the evidence presented above suggests the opposite: that the various elements of the prayer emerged very gradually and that even once they became common, they continued to shift both in precise language and in the meanings associated with accepted words.

Thus, the questions surrounding the origins and early history of the birkat haminim remain intriguing. Indeed, it may be true that this prayer functioned as an early rabbinic device to remove those who accepted Jesus as Messiah from the midst of the Jewish community. However, the evidence simply does not exist to justify making this claim with any certainty. What we can say is that by the time of the Christianization of the Roman Empire, the prayer had begun to serve this function: first as a curse of heretics/sectarians and (Jewish) Christians and then at some later date broadening to address Jews who accepted baptism, all Christians, and the governing powers who persecuted Jews. It is this prayer (minus the specific address to Christians as notzerim) that entered the liturgy of Jews living in Christian lands in the High Middle Ages and that eventually again came to the attention of Christian authorities. Beginning in the fourteenth century and regularly from the mid-sixteenth century, these authorities forced changes in the wording of the prayer, resulting in its eventual transformation into today’s versions that lack any concrete element of curse.


1 The text of the prayer itself was not static. By the medieval world, its body consistently had four elements: a curse of apostates that they would “loose (eschatological) hope”; a curse of minim (sectarians or heretics) that they would perish; a curse of Israel’s enemies that God would excise them; and a curse of the arrogant/insolent empire that God would uproot and otherwise destroy it. It concluded with a praise of God for breaking enemies and humbling the arrogant/insolent.

2 History and Theology in the Fourth Gospel, 3rd edition (Louisville, London: Westminster John Knox Press, 2003), 47-66.

3 The noted exceptions to this were the German scholars Günther Stemberger and Peter Schäfer. For details and a selection of the extensive bibliography of Christians drawing on Martyn’s work, see my Cursing the Christians?: A History of the Birkat HaMinim (Oxford University Press, 2012), notes to pp. 27-29 on pp. 266-267.

4 Chapter 1, “Origins and Early History.”

5 Until it was censored by the Catholic and then Protestant churches in the early modern period into new forms that gradually found acceptance among Jews through a complex set of circumstances; its texts today focus on abstract categories of evil, not specific human beings. See Chapters 4-5 of Cursing the Christians?

6 Texts collected in A. F. J. Klijn and G. J. Reinink, Patristic Evidence for Jewish-Christian Sects (Leiden: Brill, 1973), 200-1, 218-19, 220-21, 224-25.

7 There is the exception: one form of the prayer that we can fairly reliably date to the early tenth century and that is found in the prayer book of Rav Saadia Gaon of Babylonia, who includes only the two other lines discussed here in the body of his text. Why he omits what is likely an extremely early part of the prayer is an enigma.

8 See Cursing the Christians?, p. 48.

9 On these texts, see Cursing the Christians?, Chapter Two, or my article with Uri Ehrlich, “The Earliest Texts of the Birkat Haminim,” Hebrew Union College Annual 76 (2007): 63-112.

10 Dialogue with Trypho. Only four texts locate the cursing in the synagogue, and only one of these, 96:2, specifies that the Jews are cursing Christians. 137:2 places the cursing, apparently of Jesus (the king of Israel) after the prayers.

11 “Geniza Specimens,” Jewish Quarterly Review OS 10 (1898): 657. He describes a second text on p. 659.

12 For a discussion of this issue in one of its more recent iterations, see my review essay, “Revisiting Early Rabbinic Liturgy: The Recent Contributions of Ezra Fleischer,” Prooftexts 19:2 (1999): 179-204; and Fleischer’s subsequent response and my answer in Prooftexts 20:3 (2000) 380-387.

Comments (4)

I would be glad to see any account that softened the sense of two religious groups cursed, if that's the word, to be deeply antagonistic - I'd like, if scholarship were ready to gratify me, to hear less of Jewish curses, less of New Testament anti-Semitism. However, I think your argument as it stands has a weakness, in that the first point - that curses were a significant part of the discourse (the discourse of antagonism, I suppose) at all relevant times - works logically against the thrust of the other points, all tending to suggest that antagonism may have built up to the point of ritualised curses only slowly. Don't these arguments have opposite tendencies?

#1 - Martin - 01/24/2012 - 20:56

No, they do not necessarily have opposite tendencies. If cursing is part of my rhetorical culture, I use it rather freely, in answer to ad hoc needs, and not necessarily only by employing full received formulae. In this case, it also needs to be understood within the context of a larger liturgy that was also gradually taking shape and moving towards more and more fixed formulae. The birkat haminim cannot be understood in complete isolation from this context.

#2 - Ruth Langer - 01/24/2012 - 22:11

Thanks for reply!
If we find that we should be cautious in saying that there was a bad relationship between Jewish and Christian groups in the early days, that's fine and even a bit comforting.
But if we are to emphasise point 1, the idea that rhetoric of execration is something to be expected in the early period, then the claim of the later ritual formulae to differ from the words of the earlier period only by being formulaic, not by conveying different thoughts or attitudes is strengthened. So we would end up only with first or second generation Christians becoming aposynagogoi as a result of prevalent, though informal, rhetoric rather than as a result of forms of words fixed to the letter. This difference doesn't seem so comforting.
I'm not demanding to be comforted, only suggesting that there are some tensions, or opposite tendencies, in an argument that seems to imply both a slightly comforting and a distinctly uncomfortable conclusion.

#3 - Martin - 01/26/2012 - 15:18

What you write mostly agrees with my 1996 findings in "The Netzarim On: The Birkat ha-Minim, The Notzrim and Jerome" ( However, I think one statement merits closer examination:

"Indeed, several centuries later when John Chrysostom preached his infamous sermons against judaizing Christians, the events he described that were attracting Christians to the synagogue suggest that it was not rabbinic even in his day."

Rather, it would seem to suggest that these were Netzarim-Pharisee (rabbinic) Jews in a Pharisee (rabbinic) synagogue. Remember that (rabbinic) Pharisee Jews defended these Netzarim-Pharisee Jews, in Hellenist Roman courts, against the Hellenist Sadducees. Hellenist Notzrim (Christians) didn't become a recognizable entity until, according to Eusebius, after toppling and displacing the 15th Netzarim-Pharisee (rabbinic) Paqid with the first Hellenist gentile episkopos (bishop / pope) in 135 C.E. The error is in failing to distinguish between two antipodal opposites: Pharisee Netzarim Jews, the original followers of Ribi--rabbinic Pharisee--Yehoshua, as opposed to the 2nd-4th century and later Hellenist Christians.

Further, this reinforces the reason why the malediction would be expected to emerge coincident with "the time of the Christianization of the Roman Empire."

Paqid Yirmeyahu,

#4 - Paqid Yirmeyahu - 02/28/2012 - 12:27

Add new comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.