Jump to ContentJump to Main Navigation
Cursing the Christians?A History of the Birkat HaMinim$

Ruth Langer

Print publication date: 2011

Print ISBN-13: 9780199783175

Published to Oxford Scholarship Online: January 2012

DOI: 10.1093/acprof:oso/9780199783175.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2016. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 08 December 2016

(p.197) Appendix 2 Evidence for the Birkat HaMinim in the Pre-Sephardized Rites of the Muslim World

(p.197) Appendix 2 Evidence for the Birkat HaMinim in the Pre-Sephardized Rites of the Muslim World

Cursing the Christians?
Oxford University Press

Vast numbers of the Jews expelled from Spain and Portugal in 1492 and 1497 made their way to the more welcoming Islamic lands of North Africa and the Middle East, most of which were, or soon became part of, the Ottoman Empire.1 Unlike previous Jewish migrants who usually conformed to the liturgical rites (and halakhic norms) of their new homes, these Iberian Jews arrived with an understanding that they should maintain their identity and their received rites.2 The advent of print flooded the market with prayer books according to the Sephardi rite—very few of the local rites of their new homes were ever printed, and most early Hebrew printing emanated from Italy. Thus, in the Ottoman Empire, the Sephardi rite gradually overrode that of the indigenous Jews. As a result, for the most part, manuscripts of the local rites had no practical usefulness, and they were neither preserved in ways that we have yet discovered nor copied. Evidence for the local, pre-Sephardi rites is therefore mostly sparse and difficult to present here with certainty.3 Nevertheless, what does exist is important as the direct continuation and broadening of the picture presented by the geniza evidence; a comprehensive history of the birkat haminim also cannot confine itself to Europe.

We can clearly identify manuscripts of the pre-Sephardi liturgical rites of three communities: Yemen (which is also the rite of Maimonides, probably in Egypt), Aleppo, and Persia/Bukharia. Of these, only the Yemenite rite is richly documented, although not by early manuscripts. In addition, a few manuscripts appear to reflect rites of these genres, but we lack the evidence to identify their provenance. Aleppo is, of course, a single city; Yemen and Persia are larger and more diverse geographical areas. Yet, even the evidence for the Aleppo rite does not show an absolutely fixed set of details. The Persian texts show some similar variety, but this may be due to our inability to distinguish between inner-Persian regional rites. The Yemenite texts show great regularity, but this is because their liturgical rite closely follows that of their halakhic authority, Moses Maimonides (d. 1204).


The text of the birkat haminim in Yemen was and is:

Maimonides’ Text4


Yemenite Tikhlal6

למשומדים אל תהי תקוה

May there be no hope for apostates;

למשומדים אל תהי תקוה

כל המינים כרגע יאבדו

may all the minim [and the informers/mesorim] immediately perish;

כל המינים והמסורים כרגע יאבדו

ומלכות זדון תעקר ותשבר מהרה בימינו

and may You uproot and smash the empire of insolence speedily in our day.

ומלכות זדון תעקר ותשבר מהרה בימינו

ב"א ה' שובר אויבים7 ומכניע זדים.‏

Blessed are You, Eternal, who breaks enemies and humbles the insolent.

ברוך אתה ה' שובר אויבים ומכניע זדים.‏

(p.198) This is a very simple version of the birkat haminim, one that is structured like the geniza’s Babylonian Version 5, but lacks the line cursing enemies. The Yemenite text found from its earliest preserved manuscripts and maintained over the centuries is that of Maimonides,8 with the addition of a single word of substance, mesorim (informers).9 T. Sanhedrin 13:5 (and its talmudic citations) had included this category of miscreants with the meshummad and the min as one of those who will spend their eternity in Gehenna, so it is a natural addition to the birkat haminim. What is curious is that, while Maimonides spent his most influential years in Cairo, his version of the birkat haminim does not appear in the liturgical texts of the Cairo geniza, even though it was actively collecting documents during and after his tenure there. What then does this version represent? It is not likely his native Andalusian rite. Our evidence from the Iberian Peninsula, Maimonides’ place of birth, all of it much later, shows a more elaborate text that does include a curse of enemies. Daniel Goldschmidt struggles with this question, but in the end proposes only that this must be a rite current in Fustat that mixes various traditions. He observes also that the scribe of the manuscript wrote down a rite, presumably his own, which was a bit different from that of Maimonides. Maimonides himself subsequently corrected this manuscript; the Yemenite rite consistently follows this correction.10


We find many more surprises when we turn to our other exemplars. The Aleppo rite was printed in Venice in 1527; two earlier manuscripts further attest to the basic accuracy of this text as a representation of the apparent uniqueness of this rite. However, the manuscripts make a direct reference to Christians that was not printed.

למינים אל תהי תקוה

May there be no hope for minim;

והמוסרים [בכ”י: והנצרים] והמלשינים כולם כרגע יאבדו

and may the informers (moserim, as a substitute for noẓerim, as found in the manuscripts) and informers (malshinim) immediately perish;

וכל אויבי עמך ישראל וצורריהם ומבקשי רעתם כולם מהרה יכרתו וישמדו

and may all the enemies of Your people Israel and their oppressors and those who seek evil for them all speedily be cut off and persecuted;

ואל תתן תקומה ואחרית לכל אויבי נפשינו

and do not give recovery or an afterlife to any of the enemies of our persons.

אנא שבור עול הגוים מעל צוארינו ובכף אויבינו אל תמסרנו

Please, break the yoke of the nations from our necks and do not consign us to the hand of our enemies;

ומלכות זדון מהרה תעקר ותשבר בימינו

and may You speedily uproot and smash the empire of insolence in our days.

ברוך אתה ה' שובר אויבים ומכניע זדים.‏

Blessed are You, Eternal, who breaks enemies and humbles the insolent.

(p.199) The Aleppo rite reflects the five-fold structure of the mature geniza texts and the organization of the lines found in Version 5, that which became its exclusive form in Europe but, as we shall see, not in the east. Unlike the geniza models,11 though, this rite addresses the first line to the minim and makes no mention of meshummadim.12 Does this reflect anything more than a sense that a text called the birkat haminim ought to mention minim prominently? There are hints of a similar phenomenon in some Italian manuscripts and some manuscripts from unidentified Arabic speaking Jews.13 As we see in Chapter 5,14 there was also a call in the modern period among Sephardi Jews to restore minim to this blessing, with the result that many rites introduced the term into places where it had never been, at least in a Sephardi rite. It is possible that this rite or one like it served as a model.

Most unique in the post-geniza era is the retention in Aleppo of an explicit reference to noẓerim (Christians).15 Likely, this responds to the significant ongoing presence of Christian communities in Syria. Apparently, the Bomberg press in Venice found itself unable to print the term in 1527 and substituted a near homonym, moserim, which itself is a synonym of the next word, malshinim.16 More commonly in the rabbinic language of “mesorim” or “mesorot,” this category of people appears also in versions of the blessing in the Romaniote and Sephardi rites. However, the form moserim substitutes much more perfectly and subversively for noẓerim. Although this term appears only in scattered Babylonian-rite texts from the geniza, there are more substantial medieval parallels to the inclusion of malshinim in the blessing in its early lines, especially in Italy. Finally, the Aleppo text for the segment of the blessing addressed to enemies is the most elaborate and the most vehement of all known rites, retaining all the elements attested in the geniza’s Version 4B and then adding to them with quite explicit language.17 After all this, its curse of the empire of insolence is relatively restrained,18 and its concluding blessing is normal.19

(p.200) Persia

It is unfortunate that we have no evidence for the post-geonic Babylonian rite except for what found its way to the Cairo geniza. The evidence from Persia suggests that the earlier organization of the blessing found in the geniza’s versions 3 and 4 may have remained the norm there, albeit in its more elaborate forms. The overall organization of the blessing in Persia is mostly that of Version 4, with the curse of the empire appearing in the second line. Two manuscripts of the birkat haminim survive that reflect a pre-Sephardi Persian rite; however, they are far from identical, containing substantial variation especially in their curse of enemies. The more legible, if somewhat later, text reads.20

למשומדים ולמינים ולזדים ולכופרים ולרשעים אל תהי להם תקוה ותאוה

May there be no hope or desire for apostates and for minim and for the insolent and for the heretics and for the evildoers;

ומלכות זדון במהרה תעקר ותשבר ותכניע ותאביד

and may You speedily uproot and smash and humble and cause to perish the empire of insolence;

וכל אויבי עמך ישראל וצרריהם במהרה יכרתו וישמד ויאבדו

and may all the enemies of Your people Israel and their oppressors speedily be cut off and annihilated and made to perish;

ואל תתן תקומה לכל אויבי נפשינו

and do not give recovery to any of the enemies of our persons.

בא"י שובר אויבים ומכניע זדים.‏

Blessed are You, Eternal, who breaks enemies and humbles the insolent.

In this rite, the opening line lists a full five categories of unpleasant characters.21 This suggests that the terms meshummadim (and minim) may have lost their specific meaning, especially as they now appear along with some very abstract categories of malefactors.22 The word ta’avah (desire) in this line has no parallel in known texts and is difficult to explain.23 The curse of the empire remains in the second position, with the verb ta’avid (cause to perish) added to the list in anticipation of the primary verb of the coming line.24 That line, though, had been the one to curse minim (and noẓerim). Now, instead, the line curses just enemies; it conflates the verbs of the two lines, praying that the enemies (and not the minim) will “perish” as well as “be cut off” and adds an innovative “to be annihilated,” a verb with the same root as meshummad.25

However, it seems likely that this was a text in some flux. Our other witness to the Persian rite takes a different route and even adds to the vehemence of the prayer for the downfall of Israel’s enemies. It maintains the line that has dropped entirely in the text above, praying “ומינים כרגע יאבדו” (and minim will immediately perish) immediately after the curse of the empire (which now does not include (p.201) the verb “perish”).26 It then includes an extensive curse of enemies, which prefaces the text above with “ושבור עול הגוים מעל צוארינו ומעל צוארי כלל עמך ישראל” (and break the yoke of the gentiles from our necks and from the necks of the totality of Your people Israel), a version of which we encountered in Aleppo. Before the final petition that the enemies not recover, it then inserts a prooftext, Isaiah 14:5, “שבר ה' מטה רשעים שבט מושלים” (the Eternal has broken the staff of the evildoers, the rod of tyrants), which in its biblical context is part of the reproof to Babylon after the restitution of Israel. This thus makes the eschatological content of this blessing and its critique of the current ruling powers extremely explicit and even local to the Babylonian/Persian context. This version of the blessing also concludes with shover resha‘im (who breaks evildoers) instead of “who breaks enemies.” This of course echoes the language of the prooftext it cites, which in turn lends very specific meaning to the concluding benediction. The “evildoers” and the “insolent” both reflect back on lines of the prayer calling for the downfall of governing powers.

While the Persian texts are clearly related to each other, there are substantial differences between them. That the most substantial differences are in the places that we have noted changes in the geniza rites should not be surprising. The curse of the enemies is, in these rites, the latest and the least stable element. It is also among the easiest to apply to one’s current situation as its language, however harsh, is not historically specific. Enemies need not be as massive and mighty as empires either. We note also a tendency to move away from specific applications of the inherited language of the blessing. As far as can be determined, only the Aleppo rite continues to mention Christians by name. Meshummadim and minim have less and less presence or independent presence, perhaps because it was difficult to identify them in this context. It was illegal and hence somewhat unusual for Jews living under Islam to become Christians,27 so the received language, at least as it was interpreted, became a dispensable category.

Anomalous Texts from the Geniza28

Two sets of geniza fragments present texts that belong more obviously in the context of this appendix than anywhere else. They appear to be from eastern rites, but we have little basis for identifying them further. The first and better represented group (six manuscripts) comes from a rite that abbreviates the birkat haminim along with other blessings of the ‘amidah. It reads.29

ומלכות זדון מהרה תעקר ותשבר ותכניע בימינו

May You quickly uproot, smash, and humble the empire of insolence in our day;

והמינים כרגע יאבדו

and may the minim immediately be lost.

ברוך אתה ה' שובר אויבים ומכניע זדים

Blessed are you, Eternal, who breaks enemies and humbles the insolent.

(p.202) There are two main variants within this group: one manuscript, instead of “minim” has “nozerim and the minim”; and another lacks the second line entirely.30 Is it possible that the elimination of the opening line here reflects the identification of meshummadim with Christian baptism in Judeo-Arabic, and the reality that this was not a major category of Jewish experience in Muslim lands? This is coherent with the disappearance of noẓerim from all but one manuscript of this version as well as the complete elimination of that line in a second text.

The second version is represented in only three manuscripts, two of them from the geniza, and is a more complex text.31

המינים והמלשינים (כולם) כרגע יאבדו

May the minim and the informers [all] immediately be lost;

וכל צוררי עמך ישראל ואויביהם וקמיהם וחושבי רעתם (ומלשיניהם) מהרה מארץ יכרתו

and may all the oppressors of Your people Israel and their enemies and opponents and those who think ill of them (and their informers) be speedily cut off from the earth;

ישברו ויאבדו ותכניעם במהרה בימינו

may they be smashed and lost; and humble them speedily in our day.

ברוך אתה יי שובר אויבים ומכניע זדים

Blessed are You, Eternal, who breaks enemies and humbles the insolent.

Here too the references to meshummadim and noẓerim have disappeared, as has the explicit reference to the empire. Unlike the Aleppo rite, though, the first line has disappeared totally. The fourth line persists only in its verbs, now applied to the enemies listed elaborately in the previous line. That list of the enemies is reminiscent of that found in the Babylonian rite, Version 4 from the geniza.


(1) . Wikipedia’s article on the Ottoman Empire links to a map showing its growth through 1683. See http://en.wikipedia.org/wiki/File:OttomanEmpireIn1683.png (accessed December 27, 2010).

(2) . A major source for this was the ruling of Shmuel de Medina, Shu”t Maharashdam OH 35.

(3) . In addition to the evidence presented below, we should add the partial and anomalous text attributed to Rabbenu Ḥananel ben Ḥushiel of Kairouan (today’s Tunisia, d. c. 1050), cited in the ’Or Zarua‘ I, Hilkhot Tefillah #90 of the Ashkenazi R. Yiẓḥaq of Vienna. It reads: את המינים ואת מל' זדון במהרה תעקר תשבר ותמגר ותכניע אויבינו במהרה בימינו. בא"י שובר אויבים ומכניע זדים (Speedily uproot and smash and defeat the minim and the empire of insolence, and humble our enemies speedily in our day. Blessed are You, Eternal, who breaks enemies and humbles the insolent). Given that the rest of this passage apparently abbreviates prayer texts, it is likely that something has been eliminated here, too.

(4) . According to that published by Daniel Goldschmidt, “The Oxford Ms. of Maimonides’ Book of Prayer [Hebrew],” in On Jewish Liturgy: Essays on Prayer and Religious Poetry (Jerusalem: Magnes, 1980), 199, based on Ms. Oxford Hunt. 80, which contains a note signed in Maimonides own hand that he has checked this text (Goldschmidt, p. 188).

(5) . Words in square brackets appear in the Yemenite rite but not in Maimonides.

(6) . According to the earliest dated manuscript of this rite, Ms. Cambridge University Library Add. 1727, dated to 1446.

(7) . The scribe had originally written resha‘im (evildoers), perhaps according to his own preferred rite, but Maimonides corrected this manuscript to read ’oyvim (enemies). See Goldschmidt’s critical notes.

(8) . Goldschmidt, “Maimonides,” 187.

(9) . There is a wealth of preserved manuscripts of the Tikhlal, and the ones I have examined show essentially no variation in the wording of the text. Many pronounce the verbs in the curse of the empire as kal verbs, writing them with an added vav —more a change in emphasis than in substance. Some add a conjunctive vav at the beginning of the second line. More significant changes include the addition of a few quite normal verbs in the curse of the empire (cf. Ms. Cedarhurst New York M. R. Lehmann Collection 109, dated sixteenth–seventeenth century) or the indication of the object “them” as a suffix to these verbs (cf. Ms. Manchester John Rylands University Library Ms. Gaster 2042, dated seventeenth–eighteenth century)

(10) . Goldschmidt, “Maimonides,” 189–91.

(11) . See below for the anomalous texts from the geniza.

(12) . Censorship affected the Venice 1560 edition, and it substitutes malshinim here as is done regularly in Italy at this time. See Appendix 4, p. 222ff.

(p.342) (13) . For the Italian texts, see Appendix 3. See also Ms. Rome Biblioteca Casanatense 3085 (dated to the fifteenth–sixteenth century), which collapses the first two lines of the blessing (or skips the normal first line), beginning it והמינים והמלשינים כולם כרגע יאבדו (and may the minim and the informers all immediately perish), in all four versions it presents of the text (for weekday mornings and afternoons, Saturday evening, and the afternoon service before Yom Kippur). It is possible that this is the manuscript that Louis Finkelstein, “The Development of the Amidah,” JQR, NS, 16 (1925–1926), 140, rpt. in Contributions to the Scientific Study of Jewish Liturgy, ed. Jakob J. Petuchowski (New York: Ktav, 1970), 147, cites from “A Copy of a Manuscript Siddur of an Arabic Rite Which is in Rome [Hebrew],” ’Oẓar Tov [Hebrew appendix to the Magazin für die Wissenschaft des Judenthums] (1878): 10. However, this manuscript has not been identified as such in the catalogue of the Institute for Microfilmed Hebrew Manuscripts at the National Library in Jerusalem.

(14) . See pp. 140–49 and the details presented in Appendix 4, p. 239.

(15) . The word appears in the original in the text in Ms. Cincinnati 407 (colophon from 1410). In Ms. Oxford Bodleian Ms. Marsh. 90 (1146), dated to the fourteenth–fifteenth century, the word has been added in the margin as a correction. We should note that two versions of the prayer in Ms. Rome Biblioteca Casanatense 3085 (see the note 9) include the word והמוצרים, “and the moẓerim” in this line. As this is not a known Hebrew word and it otherwise makes no sense here and is unattested elsewhere, one wonders whether it is a (deliberate?) corruption of noẓerim. Its text of the birkat haminim shows some concern with the perceptions of the non-Jewish world as it omits explicit reference to the empire and preserves only the verbs of that line, now applied to Israel’s enemies.

(16) . Mesorim (or mesorot) appear in the list of those who will go down to Gehenna with the meshummadim and minim (T. Sanhedrin 13:5 and its later citations). The Venice 1560 edition was subject to formal Catholic censorship. Having already substituted malshinim for minim in the first line, it continues with the reference to moserim in the second line but substitutes zedim (insolent) for malshinim here to preserve the rhythm of the prayer. Thus, only the manuscripts of this rite predate the influence of European concerns.

(17) . Both manuscripts read veloḥaẓeihem (and those who oppress them) instead of the 1527 edition’s veẓorereihem (and those who persecute them). The difference in meaning is slight. Ms. Cincinnati 407 has a slightly shorter version, without the reference to ’aḥarit (afterlife) or the final phrase “do not hand us over to the hand of our enemies.” Here too the Venice 1560 edition shows the heavy hand of the censor. It eliminates the segments addressed to enemies entirely. As in other segments of the blessing, Ms. Rome Biblioteca Casanatensa 3085 is our only other witness to this sort of text, although essentially all of the elements it presents appear in one geniza manuscript or another of Version 4B. See Uri Ehrlich and Ruth Langer, “The Earliest Texts of the Birkat Haminim,” HUCA 76 (2006): 102. The continuing flexibility of this line is evident in the fact that in the four different appearances of the blessing in this manuscript, no two versions contain exactly the same list of kinds of enemies! Its most comprehensive version, for Saturday night, reads: וכל צוררי עמך כל בית ישראל ואויביהם וקמיהם וחושבי רעתם ומלשינהם ומשטינהם מהרה מארץ יכרתו (and may all the oppressors of Your people, all the house of Israel, and their enemies and those who rise against them and those who plot evil against them and those who inform against them and those who lead them astray, may they speedily be cut off from the earth).

(18) . However, it does not appear at all in Ms. Cincinnati 407. The Venice 1560 edition, again because of censorship, has eliminated the reference to the empire, leaving just a verb that otherwise does not appear in this rite, vetakhni‘em (and humble them). Because of the elimination of both the reference to the empire and the entire preceding segment, this verb now refers to the informers and insolent people of its version of the second segment as well (p.343) as, perhaps, the malshinim with which this version opens. This is not an uncommon result of censorship.

(19) . Uri Ehrlich, “The Text of the Amidah in the Siddur of R. Shlomo b. Natan and the Question of the Provenance of the Siddur [Hebrew],” Kenishta 4 (2010): 25, based on evidence for the influence of the rite of the Land of Israel, proposes that this twelfth century siddur originates somewhere near Aleppo. However, its text of the birkat haminim clearly belongs to the Babylonian rite version 4, containing few of the distinctive characteristics of the later Aleppo rite. It reads (p. 16): למשומדים אל תהי תקוה ומלכות זדון מהרה תעקר ותשבר בימינו והנצרים והמינים כרגע יאבדו וכל אויבי עמך וצורריהם מהרה יכרתו וישברו ברוך אתה יי שובר רשעים ומכניע זדים (May there be no hope for apostates, and may the empire of insolence speedily be uprooted and smashed in our day, and may the noẓerim and minim quickly perish, and may all the enemies of Your people and their oppressors speedily be cut off and broken. Blessed are You, Eternal, who breaks the wicked and humbles the insolent).

(20) . Ms. New York JTS 4522 ENA 23, published in facsimile by Shlomoh Tal, The Persian Jewish Prayer Book [Hebrew] (Jerusalem: Ben-Zvi Institute, 1980), 84. The manuscript cites the Shulḥan ‘Arukh and thus must date after 1564. Its hand is Bukharian (and E. N. Adler purchased it there), suggesting that it may come from the Persian diaspora of this time of persecution in Persia proper (Introduction, 35, 40–41). Adler’s description of his trip to Persia and neighboring countries to purchase manuscripts (and of the state of western collections of manuscripts from this region) is enlightening and helps explain the paucity of evidence from this ancient community. See his “The Persian Jews: Their Books and Their Ritual,” JQR OS 10 (July 1898): 584–625. Walter Joseph Fischel, “Persia,” Encyclopaedia Judaica, Second Edition, ed. Michael Berenbaum and Fred Skolnik (Detroit: Macmillan Reference USA, 2007), XV:787 speaks of the persecutions of the Jewish communities under the Shi’ite Safawids in the seventeenth century as including not only forced conversions but also a “crusade against Hebrew books,” which may also help explain the paucity of surviving manuscripts.

(21) . Ms. New York JTS 5432 (which may be the manuscript that Tal could not locate there, although its number is different than the one he gives) contains a list of similar length, though two of the nouns are not fully legible. The dating of this manuscript is somewhat uncertain. The Jewish Theological Seminary proposes a dating of the thirteenth–fourteenth century; the catalogue of the Institute for Microfilmed Hebrew Manuscripts at the Jewish National Library lists it as fifteenth–sixteenth century. There are no other extant Persian manuscripts reflecting this version of the birkat haminim.

The text of the birkat haminim that appears in the published Siddur Rabbenu Shelomo ben Nathan of Sigilmassa, ed. Shmuel Ḥaggai (Jerusalem, 1995), seems to be of this type. It addresses its opening line to meshummadim, minim, and zedim. However, this text derives not from the primary manuscript of this siddur (Ms. Oxford Bodleian Poc. 262, completed in 1202, which is missing a substantial segment from the middle of its first chapter), but from the slightly later and obviously less reliable Ms. Vatican Biblioteca Apostolica 497, dated mid-thirteenth–mid-fourteenth century. If the provenance of this siddur is itself a problem (see the literature that Ḥaggai references in his introduction, though he does not take it deeply into account), the provenance of this particular manuscript is even more so. This parallel with the Persian rite does suggest an origin in the east, but we should note that the colophons of this manuscript are written in Arabic, not Persian (for what is, of course, a Judeo-Arabic text).

(22) . A phenomenon that also appears in Europe. See Chapter 3, p. 81.

(23) . Ms. JTS 5432, our other exemplar of this rite, reads ’aḥarit (end), which is coherent with reading lack of “hope” as meaning a loss of afterlife, whether after death or at the end of time.

(p.344) (24) . This verb apparently does not appear in the difficult to decipher JTS 5432 manuscript, which presents its verbs in a different order and may conclude with vatemagger (and defeat), a verb common in European rites but not elsewhere in this context. It does appear in the Vatican manuscript of the Siddur Rabbenu Shelomo.

(25) . Does this reflect the religious persecutions suffered by the Jews in Persia in the period of this manuscript? See Tal’s introduction, 13–18.

(26) . Compare the Vatican text found in the printed Siddur Rabbenu Shelomo, which maintains both lines but interchanges their objects and verbs and places the prayer that zedim (not minim) be “cut off” in the midst of the curses of the enemies. It reads: …אויבינו כרגע יאבדו והזדים מהרה יכרתו ואל תתן תקומה לכל אויבי נפשנו וצררנו (may our enemies immediately perish, and the insolent quickly be cut off, and do not grant recovery to any of the enemies of our persons or our oppressors … ).

(27) . Sarah Stroumsa, “On Jewish Intellectual Converts in the Early Middle Ages Under Islamic Rule [Hebrew],” Pe‘amim 42 (1990): 64.

(28) . This discussion appeared in Ehrlich and Langer, “Earliest,” 70 n.30, 106, based primarily on Ehrlich’s research. Its substance is reproduced here with permission.

(29) . The manuscripts preserving this text are Cambridge, University Add. 3160.8; Cambridge, University Or. 1081 2.77A; Cambridge, T-S 8 H 10.20; Cambridge, T-S AS 109.83; Cambridge, T-S NS 122.50; Cambridge, T-S AS 105.136.

(30) . Ms. Cambridge, Add. 3160.8, and Ms. Cambridge, T-S AS 109.83, respectively.

(31) . Mss. Cambridge, T-S AS 102.60 and Cambridge, T-S NS 278.151. See also Ms. Rome Casanatense 3085.