A Few Comments Regarding The First Woodcut Border Accompanying The Prague 1526 Haggadah

The Prague 1526 edition of the Haggadah is one of the most important illustrated haggadot ever published.  It is perhaps the earliest printed illustrated haggadah for a Jewish audience and served as a model for many subsequent illustrated haggadot.[1] The earliest printed haggadah with illustration was published in 1512 in Latin and for a non-Jewish audience. That haggadah contains six woodcuts, and was intended as a response to the infamous anti-Semite Pfefferkorn’s screeds against Judaism.[2]The woodcut accompanying the first page shows three Jews around the seder who have four cups in front of them.  Although the Talmud explicitly states that one is not required to have four distinct cups of wine, presumably the image is a crude method of indicating the four-fold nature of the wine during the seder rather than prescribing custom.

The Prague 1526 edition was published by Gershom and his brother Gronom Katz on Sunday, 26th of Tevet 5287 or December 30, 1526.[3] This detailed publication information does not appear on the title page, rather it appears at the end of the book and is  referred to as a colophon.  The colophon is a manuscript convention that was incorporated into earlier printed books. The Prague 1526 edition does not have a title page at all.  At that time, the usage of the title page was only in its early stages.[4]

1.          The Earliest Hebrew Title Pages

As with non-Hebrew titles, the title page developed over time, both in terms of content as well as usage.[5]   The first Hebrew title page is that of the Sefer Rokeah published in Fano in 1505.[6] But that title page is really one of the more basic forms of the title page, known as a “label title page” providing only the title and author and no other ornamentation or information.[7] In that same year, an edition of Abarbanel’s Zevah Pesach was published in Constantinople.  This edition was the first to contain a border with the title and author, but no place or date of publication.[8] The first Hebrew book containing all the elements of a traditional title page, border, title, author, place and date is likely the 1511 Pesaro edition of the Talmud published by Soncino.[9]

Traditionally, the Hebrew title page is referred to as a “sha’ar” or gate.  The theory behind this description is that many title page borders are comprised of “gates,” the most common are the pillars that adorn many Hebrew books and are assumed to be those at Saint Peter’s Basillica in Rome.  Their inclusion in Hebrew books is perhaps linked to the (discredited) notion that the Catholic Church maintains certain portions of the Jewish Temple, and these pillars were actually in the Temple.  The first Hebrew book to use an architectural border is Daniel Bomberg’s edition of the Jerusalem Talmud published in 1522.[10]

1. Illustrations in Hebrew Books

Returning to the Prague 1526 haggadah, as mentioned previously, this edition was copiously illustrated, including the first page of the book. This is not the first example of Hebrew printed illustrations.  The earliest illustration to appear in a Hebrew book is that of a lulav and a handful of other explanatory images accompanying the Rome edition of the Sefer Mitzvot Gedolot dated to before 1480.[11]

The first fully illustrated Hebrew book was published in the incunabula period as well, it is Isaac ben Solomon Ibn Sahula’s Meshal ha-Qadmoni, printed in Italy, circa 1491, by Gershom Soncino.[12]

The border surrounding the first page of the Prague 1526 incorporates both Jewish as well as non-Jewish elements.  First, it is obvious that a Jew had a hand in the border as, in the inset, it displays someone performing bedikat hametz (searching for the bread) where he is using the traditional implements of a candle and chicken feather.  The outside border is less Jewish, and as many have noted, appears to be a copy of Italian/German renaissance borders.  The two most likely candidates for models for Prague are the border first used in the 1518 edition of Sacri Doctoris by Raymond Lulli (available here) or a border first used in 1519 for Paolo Ricci’s, Lepida et litere in Augsburg and reused in an Augsburg 1522 edition of Erasmus, Ad reverendum (available here).  Although we cannot pinpoint exactly which of these, if any, served as a model, what is clear is that among the images included in this border are bare-breasted women.

The use of bare breasted women to illustrate the haggadah is not limited to Prague. Both Charles Wengrov and Elliot Horowitz have pointed to earlier manuscript antecedents for Prague’s usage of such illustrations.[13] Aside from the printed example of the Prague 1526 Haggadah, this convention continued in manuscripts as well illuminated after 1526.  There are at least four such 16th century examples.[14] Additionally, and contrary to Horwitz’s contention that Spanish Jews were less accepting of such displays,[15] the Sarajevo Haggadah, which originated in Spain around 1350, includes two panels of Adam and Eve both depicting a bare-breasted Eve.

Likewise, the Golden Haggadah, 1320 Spanish manuscript includes the same form of illustration of the Adam and Eve scene.  Additionally, the Golden Haggadah includes images of nude bathers when it depicts Miriam standing from afar to see what will become of baby Moses.[16]

III.            Censorship in Modern Reprints of Prague 1526

            These historical antecedents notwithstanding, recent reprints of Prague 1526 have not been as accepting.  This initial border has been altered to airbrush and removes the bare breasts.  In 1989, a facsimile edition of Prague 1526 was published with the commentary of the Prague rabbi, Rabbi Yehuda Loew (Maharal). This border has been “touched up.”

Similarly, in 1998, a colorized facsimile edition of Prague 1526 was published.  Although the publishers took great pains to provide color where before there was black and white, they also altered this border.

Oddly enough, although they found this image offensive, they decided to reproduce it in two other places in this reprint even though this image only appears once in the original. Here is the original:

Not only is the first border altered, but other instances of bare breasts have been removed; most notably the image accompanying the verse from Ezekiel 16:7 “I made you grow like a plant of the field. You grew up and developed and became the most beautiful of jewels. Your breasts were formed and your hair grew, you who were naked and bare.”

(As discussed previously here a later, Venice 1609, edition also altered this page.)  Again in 2001, a facsimile of Prague 1526 “published by the Religious Council of Efrat in honor of the settlement’s twentieth anniversary . . . . two illustrations are surreptitiously deleted: the bare breasted woman” accompanying the verse from Ezekiel.[17] Most recently, in 2009, the airbrushed image of this woman was reproduced in The Schechter Haggadah: art, history; commentary.[18]

[1] There may be an earlier illustrated print haggadah, however, only 12 leaves of this haggadah are extant making it difficult to date (or identify the country of origin).  For a bibliography regarding this fragment haggadah see Y. Yudlov, Otzar haHaggadot, Magnes Press, Jerusalem:1997, entry 9; and most recently, Eva Frojmovic, “From Naples to Constantinople: The Aesop Workshop’s Woodcuts in the Oldest Illustration Printed Haggadah,” in The Library, Sixth Series, Vol. XVIII, No. 2, June:1996, pp. 87-109.

[2]  See R. Cohen, Jewish Icons: Art & Society in Modern Europe, University of California Press, CA:1998, pp. 21-22; see also Y.H. Yerushalmi, Haggadah; History: A Panorama in Facsimile, Philidelphia:1978, plates 6-8.

[3] Aside from the edition discussed herein, there is at least one copy of another haggadah that is also published by Cohen in Prague that same year.  While both versions are substantially similar, some of the images and borders have been changed.  Relevant for our purposes, is that the image accompanying “Sefokh” which is a full
page border with images of bare breasted women, has been replaced with a more innocuous border found elsewhere in the haggadah.  See Yudlov, Otzar, entry 8; A. Ya’ari, Bibliography of the Passover Haggadah,
Bamberg & Wahrman, Jerusalem:1960, entry 7; Rabbi Charles Wengrov, Haggadah & Woodcut: an Introduction to the Passover Haggadah Completed by Gershom Cohen in Prague, Shulsinger Bros., New York:1967, pp. 78-9.

[4] The first Prague imprint to include a separate title page is in a 1526 edition of Yotzrot published by Cohen.  See Wengrov, Haggadah & Woodcut, p. 82 n.238.

[5] Although, with regard to the adoption of the title page, Jews appear to adopt this convention at or near the time as society at large, that was not the case with other literary advances.  While the majority of the western world adopted the codex and discarded the scroll some time in the third century, the first recorded Jewish reference to the codex does not occur until the late
eighth or the early ninth centuries.  See Anthony Grafton, “From Roll to Codex: A Christian Initiative,” in Crossing Borders, Hebrew Manuscripts as a Meeting-place of Cultures, ed. Piet van Boxel & Sabine Arndt, Bodleian Library:2012, pp. 15-20.

[6] A.M. Habermann, Title Pages of Hebrew Books, Museum of Printing Art, Safed:1969, pp. 8-9.

[7] For a discussion of the development of the title page as well as the different types, i.e., label, border, end-title, see M.M. Smith, The Title-Page its early development 1460-1510, The British Library & Oak Knoll Press, London & Deleware:2000.

[8] See M.J. Heller, The Sixteenth Century Hebrew Book: An Abridged Thesaurus, Brill, Leiden & Boston:2004, vol. I, pp. 6-7.

[9] The border used for the Pesaro Talmuds first appeared in Decachordum Christianum (The Christian Ten String Harpsichord) published in Fano, 1507 by Gershom Soncino.  See M.J. Heller, Printing the Talmud, pp. 104-117  Additionally, see Heller’s discussion, id. p. 113, regarding Soncino’s reuse of the Dechachaordum‘s frames.  In reality, although the frames were originally cut for Decachordum, they were first used on Gershom’s edition of Bahya ibn Pakua’s commentary on the Torah, published four months prior to Decachordum.  See M.J. Heller, The Sixteenth Century Hebrew Book: An Abridged Thesaurus, Brill, Leiden & Boston:2004, vol. I, p. 41; Smith, The Title-Page, supra, pp. 47-59 (discussing the use of the blank title page).

[10]  See A.M. Habermann, “The Jewish Art of the Printed Book,” in Jewish Art, An Illustrated History, ed. Cecil Roth [revised ed. by B. Narkiss], New York Graphic Society Ltd., Connecticut:1971, pp. 167-68.  In Habermann’s earlier work, Title Pages of Hebrew Books, p. 9, he erroneously asserts that the earliest works to include architectural title pages were Soncino’s Melitza le-Maskil and Bomberg’s Tanach, both published in 1524/25.

[11]  See Joshua Bloch, “The Library’s Roman Hebrew Incunabula,” in Hebrew Printing & Bibliography, ed. Charles Berlin, New York:1976, p. 140.  For a description of this work see S. Iakerson, Catalogue of Hebrew Incunabula from the Collection of the Library of the Jewish Theological Seminary of America, JTS, New York & Jerusalem:2005, entry 7.

[12] See A.M. Habermann, “The Jewish Art of the Printed Book,” supra, at 169.  Habermann appears to argue that this is first printed Hebrew book containing illustrations, is incorrect. As discussed above, the first was the Rome edition of Sefer Mitzvot Gedolot. He is not alone in this error.  Ursula Schubert makes the same error.  See Ursula Schubert, Jewish Book Art, From the Renaissance until Emancipation, [Hebrew], Kibbutz hami-Uchad, Tel Aviv:1994, p. 27.  Habermann, id., notes that the great Hebrew bibliographer, Mortiz Steinschneider, was tricked with regard to one of the illustrations contained in Meshal
ha-Qadmoni.  Steinschneider, in a discussion about the alleged Christian origins of these illustrations, called attention to the fact that in one of them contains a monk wearing a crucifix. But, in the interim we have learned “that this last embellishment was a
practical joke played by a Christian scholar. . . [the crucifix] having been added by [a later] hand!”  Regarding the history and origins of the images included in Meshal ha-Qadmoni, see César Merchán-Hamann, “Fables from East to West,” in Crossing Borders, pp. 35-44; Ursula Schubert, Jewish Book Art, pp. 27-8.

[13]  Rabbi Charles Wengrov, Haggadah & Woodcut, p. 47 nn.112-13;Elliot Horowitz, “Between Cleanliness and Godliness,” in Tov Elem: Memory, Community & Gender in Medieval & Early Modern Jewish Societies, ed. E. Baumgarten, et al., Bialik Institute, Jerusalem:2011, *38-*39.

[14] Mendel Metzger, La Haggada Enluminée, E.J. Brill:1973, plate LIV, nos. 303-305; Chantily Haggadah, Musée Condé, Ms.
732, fol. 13, reproduced in Index of Jewish Art, eds. B. Narkiss & G. Sed-Rajna, Jerusalem-Paris:1976, vol. I, card no. 36.

[15] Horowitz, “Between Cleanliness,” at *38, (“One suspects that a Spanish Jew coming to Germany in the early 15th century would have been equally surprised to see an image of a naked woman” in a Hebrew manuscript.).

[16]  The Golden Haggadah, ed. Bezalel Narkiss, Pomegranate Artbooks, California:1997, figs. 17 & 24.

[17]  Id. n.37.  Horowitz only notes the 2001 example of censorship of Prague 1526, and apparently is unaware of the earlier
examples.  Additionally, he chastises the Efrat reproduction for erroneously indicating a 1527 publication date.  Efrat is not only in erring regarding the secular date.  The first facsimile edition (Berlin:1925), which includes a scholarly introduction, full title
is:  “Die Pessach Haggadah Des Gerschom Kohen Gedruckt Zu Prag 5287/1527.” Perhaps one can excuse the error as, in reality, it was completed on the eve of 1527, December 30, 1526.

[18] The Schechter Haggadah: Art, History & Commentary, [illustrations selected and annotated by David Golinkin], Schechter Institute of Jewish Studies, Jerusalem:2009, p. 36, fig. 16.1.

This post is part of a series of posts regarding illustrations adorning manuscript and print Haggadot. Our first post dealt with a new work on the topic and can be viewed here. In this post we will focus upon the some of the Halachik implications of these illustrations. In many Ashkenazic manuscripts, the Passover illustrations begin chronologically earlier than the Seder. Many begin with the preparation of the matzah. For example, in the Second Nuremberg Haggadah[1], (the manuscript is online here) a 15th century Ashkenazic haggadah, contains ten scenes devoted to the matzah process. A similar haggadah, likely illustrated by the same artist, also includes numerous matzah baking scenes. All of these, however, begin with the bringing of the grain to the miller. This is in contrast to today’s practice whereby the matzah producing process begins earlier, with the cutting of the wheat.
(All images may be clicked for larger viewing.)
That is, for many today, matzah shmurah means shmurah (watched) from the time of harvest and not from the time of grinding. But, in reality the reason for the illustrations beginning where these haggadot do, is simply because they reflect the practice in Ashkenaz, based upon the R. Ya’akov ben Asher, the author of the Tur, that only from the time of grinding is it necessary to “watch” the grain. Matzah Shmurah in Ashkenaz in the medieval period meant grain which had been watched from the time of grinding not cutting.[2] Turning to a Sefardic custom, the Barcelona Haggadah, produced after 1350, is the earliest record of the custom to place the Seder Plate on someone’s head during the recitation of Ha Lahma Ania (Ashkenazim remove the plate from the table).Only some three hundred years later is this custom mentioned in printed sources. Additionally, there is a difference between the Sefardic and Ashkenazic haggadot regarding what the Seder Plate actually was, with the Ashkenaz depicting a plate, whereas the Sefardic manuscripts depict a basket.[3]
When it comes to marror and what vegetable that is, we have at least two different types depicted in various manuscripts. In the Brother to the Rylands Haggadah, marror is depicted as an artichoke, as is in the case with the Sarajevo Haggadah.

While in many manuscripts, marror is a leafy vegetable.[4]
Setting aside the issue of what marror is, another custom related to marror can be found in both printed and manuscript haggadot. In the Prague, 1526, the first illustrated printed haggadah, there is a picture of a man pointing at his wife with the legend, “there is a custom that a man points to his wife when mentioning marror based upon the verse Ecclesiastes 7:26 “Now I find woman more bitter than death.”
A.Y. Hyman the scholar of Jewish liturgy was appalled when he came across this. In his autobiography, he claims that there is no basis whatsoever for this “custom.” Hyman is wrong.[5] If you look at the Brother to the Rylands Haggadah you can see that it shows this custom. As does the Washington Haggadah. Likewise, the Rothschild Miscellany shows the same custom.

It’s worth noting that the Rothschild Miscellany shows another custom at the time, mid-14th century, that of mixed dancing.

The mixed dancing is that of couples, husband and wives dancing with each other, and not that of unmarried men and women dancing[6] In Italy, where this manuscript was composed, mixed dancing was apparently common during this period.[7] Returning to the gesturing at one’s wife at marror, in the Hiluq and Biluq Haggadah this custom takes on a somewhat more humorous dialogue with the wife no longer passive but instead returns the compliment. In that haggadah it includes speech balloons and they record the following: The husband states “touching marror I must recall that this one, too is bitter [as gall].” To which the wife replies, “It is you [my husband] is one of the causes of bitterness as well.” After which, we have a play on the 13 attributes of Rabbi Yishmael and the haggadah provides that “the third comes between them [perhaps the marror itself] and makes a stink” – or in Hebrew ve-yavo ha-shlishei ve-yakhriach benehem.[8]

Similarly, in some Ashkenazic haggadot manuscripts, they show the the husband and wife pointing at one another.[9] Finally we get to a halachik error in a manuscript haggadah. The Washington Haggadah was written by a scribe calledJoel ben Simon. This haggadah was first printed as part of the Diskin Orphan House haggadah series in 1965. The Library of Congress didn’t publish its facsimile edition until 1991, and last year another facsimile edition of this haggadah was published as well.[10] Although this haggadah was written close to 300 years prior to Diskin publishing it, until that time a significant scribal error escaped notice. Specifically, in the text for eruv tavshilin rather than just saying “with this eruv I am allowed to cook from Yom Yov for Shabbat,” it continues and says “and on Shabbat for Yom Tov.”

Needless to say this did not escape the eagle eyes of some who feared that someone may use this haggadah (we note that contrary to the other reproductions mentioned, the Diskin version is a poor copy) and inadvertently think it is permissible to cook on Shabbat. So, the ever wise Aggudat ha-Rabbonim took out ads in the Yiddish daily, Der Tag, and the Forward to let its readers know of this error. The publishers took this one step further and mailed out a letter, with the provocative title, “Heresy or Blunder,” after Passover indicating the error and also included a letter from Cecil Roth, who had written about manuscript haggadot.

In his letter he indicates that indeed this was most likely inadvertent and that Joel did not have a different tradition regarding eruv tavshilin. Indeed, we know from Joel’s other manuscripts, where he records the correct blessing, that the Washington Haggadah’s version was simply a scribal error.

This is not the only error related to halakha and haggadah illustrations. R. S.H. Kook, criticizes two aspects of illustrations that appear in the Prague 1526 Haggadah. Both of these issues center around how the wine glass is depicted. Specifically, he takes issue with the fact that in many of the illustrations show the wine glass in the left hand and not the right. Additionally, he complains that the illustration show the holder grasping the glass at the stem and not at with his fingers cupping the bottom of the glass.
Regarding the first issue, that of left handedness, this anomaly may be attributed to the fact that this was the first completely illustrated woodcut haggadah. And, as it was the first, it was not necessarily perfectly executed. But, before we continue we must digress and explain about woodcuts in order to get to the left handedness. A woodcut when inked and put on paper produces a mirror image of whatever the woodcut depicts. Thus, if the woodcut was of a right hand, when pressed on paper would produce a left handed image. Additionally, when copies were made, the copyist were not careful and would reverse the images. That is, they were working off the printed image and would copy it directly rather than accounting for reversing the image to ensure that when it was used it would produce the same and not a mirror image. So, as has been explained, during the early history of woodcuts “copies [of woodcuts] were constantly being made, with or without leave, for copyright hardly existed, and the same printer would often have to replace worn cuts by new blocks in successive editions. It is not always an easy task to distinguish copy from original. Immediately recognizable as from different blocks are subjects which appear in reverse directions, for the copyist who does not take the trouble to reverse his drawing from the original print, will make a block that will print the subject reversed.”[11] We can now explain why the images are left handed, presumably, they were copied and the woodcutter was not careful to reverse the image, thus producing a left handed image. As an aside, another example of copying which reverses the images may be seen when the Prague 1526 Haggadah was itself used as a model for a woodcut. The border surrounding Shefokh was reused in the Levush. As is apparent, the images are that appear on the right in the Prague 1526 Haggadah appear on the left in the Levush. And, those on the left appear on the right.

Regarding the second point, the failure to cup the glass rather than grasp the stem, this can be attributed to an error on R. Kook’s part and not the woodcutter. While today it may be commonplace to cup the glass, this was only popularized by R. Yeshaya Horowitz, in his book, Shnei Luchot ha-Brit, (Shelah). R. Horowitz wasn’t born when the Prague 1526 Haggadah was printed – he lived between 1565-1630. Moreover, his book was published posthumously, in 1648, more than 120 after the Prague 1526 Haggadah was published. Thus, it is unremarkable that the Prague 1526 Haggadah failed to account for a custom that didn’t exist at that time. This is another example of why bibliography is important, for other examples see here. Finally, we conclude this part of the series regarding halachot and Passover in general, and specifically, the notion that on Passover we are stricter than normal. At times it appears that some go overboard with the various humrot on Passover as well as cleaning all sorts of items that seemingly don’t require cleaning. But, from the evidence of manuscript illustrations, this notion of stringency is not a new one. The Golden Haggadah, circa 1320, includes this very nice image of cleaning and searching for leaven. If one looks closely they will note that the woman on the left is apparently sweeping the ceiling! Thus, indicating that perhaps going overboard has been the case for some while. Also of note is that the father is performing bedikat hametz and he is bareheaded (as is the son).[12]
[1] Its title is a reference to the fact that from the mid-nineteenth century until 1957 it was housed in Nuremberg, after which it moved to the Schocken Library, and then to a private collector. The reason it is the the Second, is because Nuremberg also had another manuscript haggadah – now known as the First Nuremberg Haggadah. It too is no longer in Nuremberg. Today it is in the Israel Museum.
[2] See Steven Fine, “The Halakhic Motif in Jewish Iconography: The Matzah-Baking Cycles of the Yahuda and Second Nurnberg Haggadahs,” in A Crown for a King, Gefen, Jerusalem-New York: 2000, pp. 106-07. [3] See Evelyn M. Cohen, “Seder Foods & Customs in Illuminated Medieval Haggadot,” in The Experience of Jewish Liturgy, D.R. Blank ed., Brill, Leiden:2011, 24-25. We note that while this article provides a summary of some of the images and text accompanying medieval haggadot, the article provides little context for various practices. Indeed, the article fails to provide sources which support many of the illustrations and texts and instead merely parrots what the the manuscripts say or depict. [4] On this issue of which vegetable is preferable for marror see Zohar Amar, Merorim, n.p., 2008. [5] His comments appear in the Misrad Hasikon 1965 reprint of the Prague 1526 Haggadah at pp. 14-15. [6] See Therese & Mendel Metzger, Jewish Life in the Middle Ages, Chartwell Books, Inc.:1982, 216-17. [7] See the sources collected by R. Henkin, Shu”t Benei Banim, vol. 1, Jerusalem: 1998, no. 37, esp. section 5 where he discusses Italian sources. Thanks to R. Weinfeld for bring this source to our attention. [8] See Bezalel Narkiss, “Art of the Washington Haggadah,” in The Washington Haggadah, Commentary, M. Weinstein, ed., Washington D.C.: 1991, pp. 73-75, discussing manuscripts that contain the marror/wife images, as well as the source in the following note. [9] See R. Yisrael Mordechi Peles, “Controversies Regarding Customs That Can Be Gleaned from Haggadot,” in HaMaayan, Nissan: 5771 (51,3), pp. 13-14, available here. [10] The 1991 edition was accompanied by a commentary volume. The 2011 edition also includes some articles with the color reproduction of the haggadah. [11] Arthur M. Hind, An Introduction to the History of Woodcut, Boston: 1935, vol. 1, pp. 284-85, quoted in R. Charles Wengrov, Haggadah and Woodcut, New York: 1967, pp. 87-88. In general Wengrov’s book provides a wealth of information regarding the images contained in the Prague 1526 Haggadah.[12] Mendel Metzger, Jewish Life in the Middle Ages, supra, p. 148 discussing generally medieval manuscripts and depictions of headcovering or lack thereof.

Marc Michael Epstein, The Medieval Haggadah, Art, Narrative & Religious Imagination, Yale University Press, New Haven & London: 2011, 12, 324 pp. Most discussions regarding the Haggadah begin with the tired canard that the Haggadah is one of the most popular books in Jewish literature, if not the most popular, and has been treasured as such throughout the centuries. Over sixty years ago, Isaac Rivkin noted that as a matter of fact, only since the 19th century has the Haggadah become one of the most printed Jewish books. Prior to the 19th century, the Haggadah is neither the most printed nor most written about work in the Jewish cannon.[1] Epstein does not fall prey to this canard nor any other of the many associated with the Haggadah. Dr. Epstein’s survey of four Jewish medieval manuscripts is novel, vibrant, and sheds new light on these manuscripts, as well as Jewish manuscripts and the Haggadah generally. Epstein covers four well-known medieval Haggadah manuscripts:[2] The Birds’ Head Haggadah, The Golden Haggadah,[3] The Rylands Haggadah,[4] and the Brother to the Rylands Haggadah. First, a word about manuscript titles. Sometimes manuscripts are referred to by the city or institution that houses or housed the manuscript, while in other instances, especially when a manuscript contains a unique marking or the like, that unique identifier may be used to describe the manuscript. The Rylands Haggadah (currently housed at the John Rylands Museum, Manchester, UK), is an example of the former, and the Birds’ Head Haggadah is an example of the latter. In the case of the Birds’ Head, most of the figures depicted in the manuscript are drawn not with human heads, but with birds’ heads. Similarly, the Golden Haggadah is another example which gets its title due to the proliferation of gold borders and filler. Finally, the Brother to the Rylands, gets its title from the similarly of its illustrations to that of the Rylands, indicating some connection or modeling between the two manuscripts. As alluded to above, Epstein is not the first to discuss these manuscripts. Indeed, in the case of both the Birds’ Head and the Golden Haggadah, book length surveys have already been published.[5] Epstein, however, differs with his predecessors both in terms of his method as well as what he is willing to assume. Regarding assumptions, previously, many would take the path of least resistance in explaining difficult images and attribute confusing or complex illustrations to errors or lack of precision of the illustrator. Rather than assume error, Epstein gives the illustrations and illustrators their due and, in so far as possible assumes that the images are “both coherent and intentional.” As an extension of his “humility in the face of iconography,” Epstein attempts “to understand how the authors understood it rather than assume that [he] must know better than they did.” He does “not fault the authorship for what” he, “as a twenty-first century viewer, might fail to notice or understand concerning the structure or details of the iconography.” Furthermore, engaging with illustrations not only from tracing the history of how the image came into being but, more importantly, how that image was interpreted and what meaning it carried for its audience throughout its transmission is also one of Epstein’s goals. In furtherance of these goals, Epstein is all too aware of his own limitations and throughout the book, Epstein willingly admits both where the evidence can lead and, what is pure speculation. All of this translates into a highly satisfying and illuminating (no pun intended) perspective on these and Jewish manuscripts in general. The book is divided among the four manuscripts, with each getting its own section, with the exception of the Rylands and its Brother that are included in a single section. At the beginning of each section, all of the relevant pages from the manuscript are reproduced. The reproductions are excellent. This is not always the case in other books that reproduce these images. Indeed, in Narkiss, et al. who compiled an Index of Jewish Art that includes detailed discussions regarding a variety of medieval Haggadah manuscripts, only reproduce the images in black and white.[6] Similarly, Metzger, in her La Haggada Enluminée, also only reproduces the images in black and white (and many times the images are of poor quality). Here, each page containing an image is reproduced in full, in a high quality format that allows the reader to fully appreciate the image under discussion. Appreciating that to obtain similar high quality images requires the purchase of an authorized facsimile edition, which in some instances can be cost prohibitive highlights the importance and attention to detail that characterizes Epstein’s work on the whole. The Birds’ Head Haggadah is the oldest illustrated Haggadah text, dated to around the early 1300s. This manuscript is not the only Jewish manuscript to use zoophilic (the combination of man and beast) images. Zoophilic images can be found in a variety of contexts in Jewish manuscripts. For example, in the manuscript known as Tripartite Machzor, men are drawn normally while the women are drawn with animal heads.[7] Or, the well-known manuscript illustrator Joel ben Simon playfully illustrates the prayer God should save both man and beast, which can be read as God should save the man/beast, with a half human-half beast:
When it comes to the Birds’ Head manuscript, a variety of reasons have been offered for its imagery, running the gamut from halachik concerns to the rather incredible notion that the images are actually anti-Semitic with a bird’s beak standing in for the Jewish nose trope. Epstein ably summarizes the positions and based upon a close examination of the illustrations as well as his stated methodology, dismisses much of the prior theories. His ultimate conclusion, which builds upon the halachik position, is more nuanced and, hence, more believable, than his predecessors. The Birds’ Head provides a striking example where Epstein’s unwillingness to simply ignore complexity by claiming error, demonstrates the interpretative rewards offered to a close reader of the illustrations. While most of the images carry a bird’s head, there are a few exceptions. Most notably, non-Jews, both corporal and spiritual do not. Instead, non-Jewish humans as well as angels have blank circles instead of faces. But, there is one scene that poses a problem. One illustration shows the Jews fleeing Egypt (all with birds’ heads), being pursued by Pharaoh and his army. But, unlike the rest of the figures in Pharaoh’s army, two figures appear with birds’ heads. Some write this off to carelessness on the illustrator’s part. Epstein, who credits his (then) ten-year old son for a novel explanation, offers that these two figures are Datan and Aviram, two prominent members of the erev rav, those Jews who elected to remain behind. The inclusion of these persons, and allowing them to remain with their “Jewish” bird’s head, may be a statement regarding sin, and specifically, the Jewish view that even when a Jew sins, they still retain their Jewish identity. Sin, and including sinners as Jews, are motifs that are highlighted on Pesach with the mention of the wicked son and perhaps is also indicated with this illustration. The illustrator could have left Datan and Aviram out entirely or decided to mark them some other way rather than the Birds’ head. Thus, utilizing this explanation allows for the illustrator to enable a broader discussion about not only the exodus and the Egyptian army’s chase, but expands the discussion to sin, repentance, Jewish identity, inclusiveness and exclusiveness and other related themes.
(click to enlarge)
Epstein’s discussions of the other manuscripts are similarly eye-opening. For instance, the Golden Haggadah is an example of the Sefard manuscript Haggadah genre. Manuscript haggadahs are placed in two broad categories, Ashkenaz and Sefard. The former’s illustrations appear in the margins and generally explain the text or refer to Pesach scenes such as baking matzo or looking for hametz. The latter’s illustrations appear before the text and are a series of illustrations, appearing either in two or four panels on a single page, depicting the beginning of Jewish history with Adam and Eve, or in the case of the Sarajevo haggadah, the actual creation sequence. The illustrations culminate with the exodus. But, unlike the Ashkenaz examples, the Sefard manuscripts generally do not illustrate the Haggadah text (with the exception of HaLachmanya, a picture of matzo or the like). The Golden Haggadah follows the Sefard conventions and includes the Jewish history scenes. Epstein demonstrates, however, that the images should not just be read chronologically. Rather, the Golden Haggadah illustrator subtly linked events that did not necessarily follow in time. For example, the placement of the water in a scene depicting Jacob’s blessing to Pharaoh is linked to the scene, occurring much later, to the boys being thrown in the Nile and is similarly linked by imagery to Moses being saved from the Nile, as well as Moses rescuing Jethro’s daughters. Epstein connects all of these scenes by noting the unique method and placement of the water in the scenes. But the linkage is not merely water, instead, this interpretation affords insight into God’s blessings, promises, the parameters and methods of His divine punishment of “measure for measure,” gratitude, and salvation. Again, this is but one example where close examination of the illustrations enriches the Haggadah discussion. All of Epstein’s discussions display his keen awareness and erudition regarding illustrations appearing in both the manuscript as well as print Haggadahs. Although the work employs end notes, which we find generally to indicate that the notes are unnecessary for the text, the notes should not be ignored. They are full of interesting sidebars as well as additional information on the illustrations discussed and the history of Haggadah illustration.[8] As a testament to the importance of this work, as well as its accessibility, the book was originally published after Pesach last year (hence our belated review) and, already, before even a single Pesach, its publisher is sold out. The work has already received numerous accolades from numerous others to which we add our small voice. This is an incredible work in terms of its insights, methods, and production values that is a welcome breath of fresh air to stale and repetitive Haggadah genre.

1. It is well-known that the Mishnah in the tenth chapter of Pesachim includes a set of mah nishtannah. But if one opens a standard printed Babylonian Talmud (Pes. 116a), one sees four questions[2] in the text of the Mishnah (matzah, maror, roast, and dipping), while if one opens a standard printed Jerusalem Talmud, one sees three (dipping, matzah and roast). Is this an instance of a disagreement between the text of the Mishnah preserved in Babylonia and the text of the Mishnah preserved in Palestine? There are such disagreements,[3] but this is not one of them. Here, the manuscript evidence points overwhelmingly to the conclusion that the original text of the Mishnah recorded only the questions of dipping, matzah and roast.[4] Moreover, if one opens up a standard Masechet Pesachim of the Babylonian Talmud and looks at the text of the Mishnah recorded in the Rif and the Rosh, one sees that they too record a Mishnah which included only the questions of dipping, matzah and roast. Also, Rambam utilized a text of the Mishnah which included only these three questions.[5] Let us see how the Encyclopaedia Judaica in its original and new editions deal with this issue. When Daniel Goldschmidt published his classic work on the Haggadah in 1960, after earlier German and Hebrew versions,[6] he clearly took the position that the original text of the Mishnah included only three questions.[7] It was therefore somewhat surprising when the “Mah Nishtannah” entry in the original Encyclopaedia Judaica (published in 1972 and written by the anonymous editors) stated that “[t]he Mishnah enumerates four questions.” In its bibliography, this entry had cited only Goldschmidt’s 1960 work, and one other less scholarly source.[8] We are now in the 21st century. There is now a widespread consensus among scholars that the original text of the Mishnah included only three questions.[9] But when one looks at the new edition of the Encyclopaedia Judaica, one finds only a reprint of the original entry![10] —— 2. A widely quoted explanation of the Vilna Gaon takes the positions that there were always four questions, and that, after the churban, the roast question was substituted by the reclining question.[11] We have already seen that the first of these positions is incorrect. The Haggadah fragments found in the Cairo Genizah show that the second of these positions is incorrect as well. Genizah material generally dates from the 10th through the 13th centuries.[12] It is reasonable to assume that this is roughly the period of the Haggadah fragments from the Genizah.[13] Of course, not all of the Haggadah fragments from the Genizah span the mah nishtannah section. But of those that do, many include the roast question.[14] (The survival of the roast question for 1000 years, post-churban, has been implicitly acknowledged in the Haggadah shel Pesach, Torat Chayim, published by Mossad ha-Rav Kook, at p. 29, n. 59.) [15] Furthermore, although most of the mah nishtannah Haggadah fragments found in the Genizah record four questions the way they are asked today,[16] we also find the following:[17] -Several record three questions: matzah, dipping, and roast,[18] just like the original text of the Mishnah. Most likely, the fragments with these three questions reflect the original Palestinian tradition of the Haggadah,[19] which followed the text of the Mishnah, and the fragments with four questions reflect the Babylonian tradition,[20] which gradually penetrated into Palestine and its surrounding areas.[21] -One records the following three questions: dipping, matzah and reclining.[22] -One records five questions: dipping, matzah, roast, maror, and reclining.[23] (See the photograph in M. Kasher, Haggadah Shelemah, p. 93.) In this fragment, the roast question is phrased:… צלי כולו המקדש בבית אוכלים היינו הזה והלילה. This is the only fragment in which the roast question is phrased with such a qualification. [24] -Two record only the questions of dipping and roast.[25] (But there does not appear to be any reason why the matzah question would have been intentionally discontinued. Perhaps the matzah question was accidentally dropped by a scribe in one source, and further copies were later made from that source. It would have been easy for a scribe to accidentally turn his eye to the wrong mah nishtannah line in the source he was copying from, and thereby omit a question.) – One records only the questions of dipping and matzah. I would like to focus on this last source, which is not actually a Haggadah fragment, but is a section of an anonymous Geonic responsum which includes an outline of the procedures at the seder. It can be deduced that the responsum was composed in Babylonia because it includes avadim hayyinu, which was not a part of the Palestinian seder ritual in this period.[26] This responsum was published by Louis Ginzberg, in his Ginzey Schechter, vol. 2, pp. 258-260. [27] Theoretically, it is possible to argue that the author of this responsum gave only an abbreviated version of the mah nishtannah, and listed only the first two questions, even though his practice was four. This seems to be the understanding of Ginzberg, who writes (p. 259, n. 1): הראשונות השאלות ‘ב אלא כאן נאמרו ולא נתקצר נשתנה מה נוסח. B.M. Lewin included this responsum in his Otzar ha-Geonim, Pesachim (pp. 154-55), and he seems to agree with Ginzberg. (See p. 154, n. 13: נתקצר נשתנה מה נוסח.) But this interpretation seems very unlikely. The whole purpose of the responsum was to spell out the procedures and text of the seder. Abbrevation here would have defeated its purpose. The Safrais take a different approach to this responsum in their monumental work, Haggadat Hazal. They write (p. 64, n. 53) that the third and fourth questions are העמוד בסוף חסרות, implying that these questions were originally included in this responsum but were cut off. Moreover, in their chart on p. 266, they put brackets over a supposed area of the third and fourth questions in this responsum, implying that these questions were once there. They take this approach so that the set of questions in our responsum could then parallel the set of questions found in the other known Babylonian Geonic sources of the Haggadah text: Seder R. Amram Gaon, Siddur R. Saadiah Gaon, and the Haggadah text published in 1984 by M. Lehman.[28] All these sources record the standard four questions: dipping, matzah, maror, and reclining.[29] But I was able to view an image of the responsum (T-S Misc.36.179), and with the kind permission of the Cambridge University Library, have included this image at the end of this article. It is clear that the third and fourth questions were never there. The first side of the fragment ends with the last words of the matzah question (the last three words of this question were written below the last regular line), and the next side continues immediately with avadim hayyinu. It is clear from the image that there are no missing lines in between. Assuming we reject the unlikely interpretation of Ginzberg, this source records a two-question set in Babylonia.[30] The idea that we have now been able to “excavate” such a set, evidence of a period before four questions became the universal practice there, is truly remarkable. An issue remains whether the responsum dates to the period before the maror and reclining questions were added in any Babylonian communities, or whether it dates to a period after they were added, but was composed in a community which did not add them. The latter seems more likely. (The responsum does have at least one other unique aspect.[31] It includes the statement: חורים בני היום[ו] עבדים היינו אתמול. The Safrais discuss this statement at p. 111 and do not mention any other fragments from the Genizah with this language.[32] On the image at the end of this article, this statement is found on the first page, six lines above the last full line.) Regarding the issue of when the maror and reclining questions were added, the following are some reasonable observations that have been made by scholars to date: ◦ The reclining question was probably the last question to be added.[33] Unlike the maror question, it did not make its way into in any manuscripts of the Mishnah,[34] and in all communities, it is the last question of the set.[35] ◦ The maror question probably did not arise until after the text of the dipping question was changed in Babylonia (see Pes. 116a[36]) and the dipping question lost its connotation as a maror question. Once the dipping question lost this connotation, it was probably viewed as necessary to add a question relating to maror.[37] ◦ The reclining question probably originated in Babylonia as well.[38] It was probably added, after the maror question, due to a desire to fix the number of questions at four, parallel to the themes of four cups of wine and four sons.[39] Aside from Haggadah fragments found in the Genizah, the earliest sources which include the reclining question are: Seder R. Amram Gaon,[40] Siddur R. Saadiah Gaon,[41] and the Geonic Haggadah text published by Lehman in 1984. (With regard to Seder R. Amram Gaon, it is accepted that many additions and changes were made to it in the centuries after R. Amram’s death, c. 875. The three surviving manuscripts of the work are only from the 14th-16th centuries. See D. Goldschmidt, Seder R. Amram Gaon, pp. 11-13. With regard to the Siddur R. Saadiah Gaon, the material in our possession has traditionally been viewed as representing a generally reliable version of the original 10th century work by R. Saadiah.[42] With regard to the Geonic Haggadah text published by Lehman, see above, n. 28.) —— 3. A main issue of debate among scholars has been whether the mah nishtannah set of the Mishnah reflects questions from Temple times. If we focus on the roast question ( הלילה הזה כולו צלי ), a reasonable initial assumption is that this question is referring to the pesach sacrifice, since the pesach sacrifice was something that was required to be roasted. This would suggest that this question, and hence perhaps all the questions, were composed during Temple times. But many scholars argue that the roast question is not referring to the pesach sacrifice and that this question was composed after the churban (and hence, the other questions probably were as well.) They make the following arguments: 1. A practice arose after the churban, approved (or perhaps initiated) by Rabban Gamliel of Yavneh, of eating a gedi mekulas as a method of commemorating the pesach sacrifice.[43] See Betzah 2:6-7 and Eduyyot 3:10-11. (Although these two sources do not specify that the Rabban Gamliel they are referring to is Rabban Gamliel of Yavneh, it seems fairly clear, for a variety of reasons, that this is the case here.[44]) The meaning of the term mekulas is subject to debate.[45] Nevertheless, it is clear that the gedi mekulas was not designated as a pesach sacrifice, but was arranged on the spit during the roasting process in the same manner as a pesach sacrifice, and was roasted in its entirety just like a pesach sacrifice. See Tosef. Betzah 2:11 (איזהו גדי מקולס? כולו צלי ראשו וכרעיו וקרבו ) and Pes. 74a.[46] The Sages prohibited the gedi mekulas,[47] but it is possible that a large section of Jewry followed the leniency of Rabban Gamliel and engaged in this practice.[48] The roast question could have been composed post-churban in an area which followed this practice and could be referring to this practice.[49] 2. The Mishnah at Pesachim 4:4 records:מקום שנהגו לאכול צלי בלילי פסחים, אוכלים; מקום שנהגו שלא לאכול, אינן אוכלין. It is possible that this Mishnah was composed during Temple times and is referring to a practice of eating roast meat on the seder night outside the Temple. But just as likely, this Mishnah was composed after the churban and is referring to a post-churban practice of eating roast meat[50] as a commemoration of the pesach sacrifice. It is possible to understand the roast question in the Mishnah as composed, post-churban, in an area which followed this practice and as referring to this practice.[51] But (as pointed out to me by my friend Sam Borodach) the precise phrasing of the roast question does not support either of these interpretations. When read in light of the parallel matzah question, the roast question implies that the question is about a required behavior of the evening. While there were areas which had a custom on this evening, post-churban, to eat a gedi mekulas or to eat roast meat, these were not required behaviors. It is hard to imagine the roast question phrased the way it was by someone composing it with the above optional commemorative behaviors in mind. The language of the roast question, read in light of the parallel matzah question, fits the pesach sacrifice best.[52] Does it follow from this interpretation that the roast question was composed during Temple times? Not necessarily. Even if the roast question is referring to the pesach sacrifice, it very possibly could have been composed after the churban, as a simulation of a question that might have been asked during Temple times. Once the idea for a mah nishtannah question about matzah took hold, it would be have been natural and instructive to create a parallel question covering such an important commandment as the pesach sacrifice, even if the pesach sacrifice was no longer being offered. (All the more so if practices commemorating the pesach sacrifice were ongoing!) (It is also possible to understand the roast question as having been composed post-churban if the pesach sacrifice itself continued after the churban.[53] But the evidence for this is weak.[54]) The Mishnah that concerns us, Pesachim 10:4, is found in the last chapter of this masechet. This suggests that this Mishnah was composed after the churban, since it is reasonable to presume that mishnayot that are presented in the early parts of a masechet were composed first and that mishnayot that are presented later were composed later.[55] Moreover, there is specific evidence suggesting that this particular chapter was composed after the churban. Mishnah 10:3 includes the statement:ובמקדש מביאין לפניו גופו של פסח. This statement suggests that the normative statements in this chapter describe only post-churban practice.[56] Of course, the questions themselves could still reflect questions from earlier times, which were recorded in a chapter composed after the churban. Based on recent scholarly developments, a new argument can be made to support the idea that the questions were composed post-churban. The mah nishtannah is not included in the tenth chapter of the Tosefta. This chapter includes much material parallel to the tenth chapter of the Mishnah. In his Tosefta Atikta, published in 2002, Shamma Friedman studied the relationship between the Mishnah and Tosefta of Pesachim in the first four chapters and in the tenth chapter.[57] He came to the conclusion that the material in the Tosefta in these chapters seems to have originated earlier.[58] If Friedman is correct, this also tends to support a post-churban origin for the mah nishtannah. So far, I have not discussed the precise role that the mah nishtannah serve in the Mishnah. But most likely, the mah nishtannah of the Mishnah was not a mandated piece of liturgy to be recited by the father or a typical son. Rather, it was what the child who lacks understanding is taught to ask (…מה נשתנה אביו מלמדו, אם אין דעת בבן).[59] This also perhaps implies that the mah nishtannah text was a later development. ——– Note finally that the Talmud, at Pes. 70a, records and accepts a statement by R. Hisda that the roast question was authored by Ben Tema. Ben Tema took the position that the chagigah too had to be roasted. In the view of R. Hisda, the language of the roast question, הלילה הזה כולו צלי, fit the position of Ben Tema only. Ben Tema is only mentioned at Pes. 70a and Yoma 83a and there is not enough information to determine if he lived during Temple times. (Some suggest that he is to be identified with R. Judah Ben Tema, who probably dates to the second century C.E,[60] but this identification is only conjecture.) The above statement of R. Hisda is not recorded again at Pes. 116a, where one would expect it.[61] R. Hisda’s statement is also surprising, because it takes an overly literal approach to the roast question. There is also no statement in the Jerusalem Talmud expressing the view that the roast question was authored by Ben Tema.[62] ———- To summarize, it is possible that the mah nishtannah was composed during Temple times, but it also possible that it was composed post-churban. In support of the latter is that the mah nishtannah is found in the last chapter of the masechet, and in a chapter whose normative language suggests that it was composed post-churban. The inclusion of a question about roast meat is not inconsistent with this approach. This question could have been composed post-churban in an area which followed the post-churban practice of preparing a gedi mekulas or in an area which followed the post-churban practice of eating roast meat. More likely, if the question was composed post-churban, it is referring to the pesach sacrifice and was composed for educational purposes, as a simulation of a question that might have been asked during Temple times. The fact that the mah nishtannah is not included in the Tosefta may also support a post-churban origin for it. ——— Cambridge University Library T-S Misc.36.179 (reproduced by kind permission of the Syndics of Cambridge University Library)