An Illustrated Haggadah for Sefardi Exiles in Istanbul

Abstract

The first illustrated haggadah of the print era was published around 1505 by David and Samuel ibn Nahmias in Istanbul (henceforth “Istanbul Haggadah”). It was embellished with woodcuts that had been commissioned in 1492 in Naples. This paper approaches the Istanbul Haggadah as a cultural product of the early Sefardi Diaspora. A comparative iconographic method reveals idiosyncrasies in relation to the tradition of medieval manuscript haggadot, which are then contextualized within the cultural ambience of the early Sefardi Diaspora in Naples, where Don Isaac Abarbanel played a central role as a spiritual and communal leader. My analysis is based on three types of information and sources: Abarbanel’s post-expulsion writings, among others a commentary on the haggadah; book-historical data on the early phases of printing; and historical information on the lives of the refugees. Most Sefardi printing projects from the post-expulsion years were aimed at meeting the spiritual needs of the community of exiles. The Istanbul Haggadah, and particularly its illustration program, was a fitting compliment to these endeavors.

1. Introduction

The first illustrated haggadah of the print era was published around 1505 by David and Samuel ibn Nahmias in Istanbul. Extant only in fourteen Genizah fragments, it is significantly less well known than other early printed haggadot. Suggestions by earlier scholars about its attribution and date of printing are not based either on a colophon or a title page but on a series of careful examinations of the remaining pieces. However, these attempts say nothing about its history or the role this book may have played in the lives of those who purchased and used it. Twenty-five further fragments are extant, the remains of a second edition printed in c. 1514. Whereas most of the fragments of the first edition are full pages or even bifolia and, thus, fairly imaginable as a book, most of the remains of the second edition are tiny shreds of torn paper (for the shelf numbers, see Appendix A; for links to the websites of digitizations, see Bibliography). Three fragments are in the Library of the Jewish Theological Seminary in New York (JTS); the rest are found in the collection of the Taylor-Schechter Genizah Research Unit in Cambridge (T-S) (for details, see my attempt to reconstruct the book structure of the two editions, Kogman-Appel, in prep., chp. 6).

On the pages to follow, I first briefly survey the early scholarship that eventually led to the printshop of the Ibn Nahmias brothers in Istanbul (for background on the brothers, see Yaari 1967, pp. 17–21; Ben Na’eh 2001, p. 79; Hacker 2012, pp. 22–24). Based on these results, in the main part of the paper I look at the Istanbul Haggadah as a cultural product of the very early Sefardi Diaspora. Produced by exiles from Spain, printed in Sefardi types, and decorated with woodcuts that presumably were brought to Istanbul from Naples, in a way the Haggadah testifies to the routes some of the exiles had taken.

Around the same year that the Istanbul Haggadah was printed, Isaac Abarbanel’s Commentary on the Haggadah, Zevah Pesah (Passover Sacrifice) appeared in Istanbul (on 6 November 1505; Yaari 1967, no. 3; Hacker 1972, no. 3). Completed in 1496, this book counts among Abarbanel’s post-expulsion writings, which are all key documents regarding the ways the exiles experienced the trauma of migration. These circumstances suggest that the decision to print a haggadah at a time when stand-alone haggadot were in fact a rarity can be associated with the genesis of Zevah Pesah (henceforth ZP), with the experiences of Sefardi migrants, and with Abarbanel’s role in early post-expulsion society and culture. My argument, then, focuses on the imagery in the Haggadah and the way it relates to Abarbanel’s discussion of the Passover festival. A copy of ZP is now in the collection of David Jeselsohn in Zurich. Since a critical edition of the Commentary is wanting, and since it is this text that was accessed by the Ibn Nahmias brothers, I shall use this edition in my following discussions.

Historiography: Attribution

In December 1926, the Soncino Society in Berlin published two folios from the Haggadah as a Hanukkah gift for its members just after they had been discovered together with a third, unpublished, folio in Elchanan N. Adler’s collection housed in the Jewish Theological Seminary in New York (JTS). The pages were printed as a facsimile without any commentary (Haggada 1926). In an unpublished bibliography of early Hebrew prints extant in the Library of the Hebrew Union College at Cincinnati, Moses Marx attributed the fragments to the Ibn Nahmias brothers in Istanbul and dated them to 1513 (Frojmovic 1996, Note 21 referring to an unpublished list of early prints, Marx (1947)). In his 1961 Bibliography of the Haggadah Abraham Yaari voiced doubts concerning Marx’s attribution and dating, noting that “it cannot be excluded that the Haggadah was printed either in Spain or in Portugal before the expulsions” (Yaari 1961, no. 5; see also Mann 1992, cat. no. 51). Two years later, upon presenting an additional fragment from the Taylor-Schechter Genizah Research Unit in Cambridge (T-S), Abraham Habermann observed that some of the types were worn, which suggests that the Haggadah was printed using old Iberian types in the very early days of Jewish printing in Istanbul (Habermann 1962–1963). Five more pages from Cambridge surfaced in 1965 and were published by Alexander Scheiber. Given that one of them has a duplicate in New York, Scheiber realized that the fragments must come from two different copies (Scheiber 1965). In 1972 when Habermann discovered further fragments, he noticed that the qiddush page, which is decorated with an illustration of a seder table (Figure 1), also displays two images of a rabbit, a motif found in several books published by the Soncino family of printers, who set up a press in Istanbul during the later 1520s. Habermann thus suggested advancing the date of the Haggadah to c. 1527 (Habermann 1972). Habermann also attempted a reconstruction but was misled by the fact that he knew some of the pieces only from Scheiber’s earlier paper, which reproduced the pages of the bifolia separately. Subsequently, Habermann also did not realize that there were two editions. Either way, doubt was shed on his assertion ten years later when Scheiber discovered the printer’s flag of Judah ben Joseph Sasson (a lion rampant) on one of the pages from what now turns out to be the second edition (Figure 2). Judah Sasson is known to have been in the print business since 1514 as a partner of Samuel ibn Nahmias, the younger, a son of David ibn Nahmias and a nephew of the elder Samuel ibn Nahmias (Scheiber 1982, followed also by Heller (2021); Heller, however, was not familiar with Frojmovic (1996); on the partnership with Judah Sasson, traceable between 1514 and 1516 and on the flag in his books, see (Yaari 1967, pp. 18–19; Hacker 1972, pp. 465–66)). To the far left, the remains of a third motif are discernible and Eva Frojmovic, while admitting that no design can be made out, assumes that it is the right-upper corner of yet another print of the rabbit motif (Frojmovic 1996, p. 89, Note 8). A close-up of the motif, however, puts this assumption into much doubt.

In 1996 Frojmovic not only found two more tiny fragments in the boxes of the T-S, but also discovered a watermark on one of the pages, which we now know belongs to the first edition (NS 168.5A). She identified it as the same mark that appears on the paper that was used in 1505 by the Ibn Nahmias brothers for Abarbanel’s Zevah Pesah (Abarbanel, ZP 1505; Frojmovic 1996), confirming, thus, Marx’s earlier assumption. Frojmovic’s main argument, however, concerns the woodcuts, which she attributes to a Neapolitan printshop owned by the humanist scholar Francesco del Tuppo (d. after 1498). Her conclusion is based on a stylistic comparison with the woodcuts in del Tuppo’s 1485 edition of Aesop’s Fables. Cecil Roth had previously observed that the borders of del Tuppo’s Aesop edition reappeared two years later in books printed by the Soncinos in northern Italy (Roth 1953; see also Habermann 1961; Offenberg 1996), attesting to earlier connections between del Tuppo and Jewish printers. In short, it is likely that the Ibn Nahmias brothers printed the Haggadah in Istanbul in 1505 or 1506 using wood blocks cut in del Tuppo’s workshop. Frojmovic argues convincingly that these had been commissioned as early as in 1492, apparently shortly after the brothers’ arrival in Naples, presumably in the late summer of that year. The Ibn Nahmias, however, soon moved on to Istanbul, and may have taken the blocks with them (or perhaps they were shipped to them at some later stage). However, apparently owing to political circumstances, Hebrew printing in Istanbul ceased for more than ten years and resumed only in 1505. That the Haggadah was printed by the Ibn Nahmias once they were able to resume work, can, finally, be confirmed by the appearance of the brothers’ printer’s flag (Yaari 1943, no. 4), designed as a six-pointed star formed by floral elements. It appears next to Sasson’s flag (Figure 2) and is also known from a 1493 edition of Arba’a Turim (Figure 3, for the date of this edition, see Offenberg 1996) as well as from a Pentateuch printed by the brothers in 1505. Similar floral elements, albeit not in the shape of a six-pointed star, appear beneath the three matsot (Figure 4). These, clearly, were not cut in del Tuppo’s workshop (Frojmovic 1996, p. 106), but were apparently undertaken in 1505/1506 by a member of the Ibn Nahmias workshop. In terms of style and quality, they share nothing with the figural illustrations.

Frojmovic’s attribution makes a lot of sense. The blocks with the rabbits associated with the Soncinos may have come into the possession of the Ibn Nahmias while they were in Naples. The brothers are also known to have used Soncino types (Hacker 2012, p. 23). The rabbit images are not a printer’s flag meant to reveal the identity of a printer but simply a decorative motif, and the blocks may have been sold to the brothers together with the types. More difficult to explain is Judah Sasson’s flag because he is known to have been in the printing business only since 1514. The initial relevant scholarship spoke only of one edition, but in 1965 Scheiber pointed out that at least two copies of the Haggadah survived. A careful examination of the fragments reveals that not only are there two copies of one edition, but there must have been two editions. With the help of the surviving bifolia and given that the text pages contain an average of 100 words each, I attempted a tentative reconstruction of the two slightly diverging editions and their quire structures. Both editions survive in at least two fragmentary copies (hidden reference). The first edition was, thus, apparently printed in 1505 or 1506 by the Ibn Nahmias brothers, whereas the second, the one that carries Judah Sasson’s flag, may have been printed around 1514, when Samuel ibn Nahmias the younger (David’s son) entered into a partnership with Sasson.

2. Materials and Methods: A New Pictorial Program

The qiddush on the first page is accompanied by a depiction of a seder table with four men (Figure 1). Three folios later a five-man seder table composition accompanies the narrative of the five Rabbis in bnei braq (Figure 5; this is the folio on which Frojmovic discovered the watermark). Between the bnei braq narrative and the account about Rabbi Eleazar ben Azariah, we find another print of the four-man seder table (Figure 6), repeated yet again on the opposite page introducing the question of the Wise Son (Figure 7). As in virtually all illustrated haggadot the matsah and maror illustrations are juxtaposed with the sayings of Rabban Gamliel (Figure 4).

Biblical imagery is apparent in depictions of five of the Egyptian plagues: blood (Figure 8), frogs, lice (Figure 9), boils, and hail (Figure 10). This short series is inserted between the end of Rabbi Yose the Galilean’s words and the beginning of Rabbi Eliezer’s section on the plagues. The images of blood, frogs, and hail still have captions (דם, צפרדע, ברד) near the top or bottom of the frame. Those for the lice (כינים) and the boils (שחין) were lost.

What remains of the second edition does not include any fragment with seder table compositions, but the depictions of the matsah and the maror are still extant and also show the printer’s flags mentioned above (Figure 2). Of the biblical scenes, only the plague of lice with its caption was inserted into the dayyenu poem after it had been detached from the plague of frogs as a separate, single wood block.

Frojmovic suggests that the series with the plague scenes in the first edition preceded the text, in line with a tradition established in a group of Iberian Passover compendia from the late thirteenth and the fourteenth centuries (Frojmovic 1996, pp. 88–89), but this is impossible for several reasons. The page with the plague of blood has part of the maggid section on its verso; apart from that, it is part of a full bifolio, whereas its complementary folio is the very last page of the Haggadah (Figure 8). It also appears that the other four images are part of surviving bifolia and that those pages are connected to portions of the hallel (Figure 9 and Figure 10). This means that all the plague depictions must have been part of the last quire. I am very grateful to the staff of the T-S, and above all to Julia Krivoruchko for identifying the bifolio with the blood image. This leaf was published by Habermann with its two folios photographed separately and without references to their shelf marks (Habermann 1972, fig. 4, Misc. 30.8a), which led scholars astray (among them Yerushalmi 1975b, pp. 5–6). Hence, the arrangement of this Haggadah did not and was not meant to follow the earlier Iberian tradition, but its editor inserted the selected plague scenes near the mention of the plagues in the text.

These observations suggest that the Istanbul Haggadah was not meant to follow in an earlier pictorial tradition found in Iberia. The history of illustrated haggadot from fourteenth-century Catalonia (and other areas of the Iberian Peninsula and southern France) constitutes one of the most celebrated chapters in the annals of Jewish visual culture. Several of these volumes, mostly produced in Catalonia, were particularly lavish and were in fact not simply haggadot but extensive Passover compendia or Passover mahzorim containing the piyyutim for the synagogue services (or parts of them) for the entire Passover week. As I have shown elsewhere, they were intended for or commissioned by a particularly wealthy elite of courtier scholars (Kogman-Appel 2006). That book genre ceased abruptly in the late fourteenth century, and while several copies of illustrated haggadot appeared in fifteenth-century Ashkenaz and northern Italy, in Catalonia, and elsewhere in Iberia and southern France the genre virtually disappeared. There are only a handful of haggadot that do not fall into the category of luxurious compendia. Whatever the reason for this cessation, it is obvious that the function the Passover compendia had fulfilled since the late thirteenth century was no longer relevant for the ritual practices common in Iberia in the late Middle Ages.

Scholars pointed at the relatively high degree of continuity between handwritten and early printed books (Sabar 2019 with regard to Iberia; Kogman-Appel 2017). While this is true in general, it applies less to early printed haggadot. The first printed Sefardi haggadah in fact appears in striking contrast to the lavish fourteenth-century compendia. Published in c. 1482 in Guadalajara, perhaps by Solomon Alkabez, it is but a small booklet of twelve pages without any kind of embellishment (Yaari 1961, no. 1; unicum at the National Library of Israel, NLI). It was produced at a time when most siddurim and many mahzorim still included the haggadah, and it appears that most of the households everywhere in the Jewish world did without individually bound haggadot to be used during the ritual family meal on the eve of the 15th of Nissan. The incunable period saw only one more Sefardi edition of the haggadah, apparently from Portugal, which was printed between 1477 and 1487 (Schaeper 1999, p. 630). A likewise undecorated Passover compendium was printed in Rome in 1508 (Vatican, Biblioteca Apostolica, NLI, microfilm 93 F 1.).

When the Istanbul Haggadah finally appeared, in terms of its layout as well as its illustrations program it represented an entirely new type of book. But there are more idiosyncrasies in comparison to the earlier tradition of illuminated manuscript haggadot. The four-man seder composition (Figure 1, Figure 6 and Figure 7) is unusual and the fact that no women attend the seder is even more so. Seder table depictions are well known from illuminated medieval haggadot, whether from Ashkenaz, Iberia, or Italy. They usually show crowded family gatherings with women, children, and youths, and occasionally even show a dog beneath the table (Figure 11).

At first sight, one might think that the Christian woodcutter in Naples misunderstood the ritual but, as Frojmovic points out, there are other details that in fact seem to demonstrate a real concern for accuracy regarding Passover customs (Frojmovic 1996, p. 107). The scene includes a table on which we can see a pitcher, some plates and cutlery, and a large bowl with the ritual foods—a shank, an egg, and a bunch of leaves are clearly visible. As required by the ritual law, all four men are reclining to the left, their elbows resting on special cushions. Hence, the Christian artists must have received clear instructions or model drawings from the commissioners. This notion is supported by the Hebrew captions integrated into the plague panels, which also attest to the involvement of the Ibn Nahmias brothers or someone on their behalf in the design of the blocks (Kogman-Appel 2024).

One of the men depicted is raising a cup in his right hand while performing a gesture of blessing with his left; another is drinking from a goblet; the man in profile to the right is chewing on a vegetable, and the fourth man is reading from a book. Instead of representing a crowded family meal, the Istanbul image of the four men seems not to be taken from real life (when any seder would be attended by the whole family), but, rather, to represent four ritual aspects, each with a certain performative dimension: blessing the wine, drinking the wine, consuming a ritual food, and teaching the next generation. Moreover, as we shall see further on, this imagery could also be transposed to represent the Four Sons and their questions.

The image of the Rabbis in bnei braq, finally, seems to complement this approach. Whereas manuscript miniatures of the Rabbis, all of whom are Tannaitic sages, commonly show them engaged in study and discussion, without any suggestion of an actual seder ritual (Figure 12), the Istanbul Rabbis are all shown performing ritual acts associated with the Passover meal (Figure 5): two of them are eating vegetables; the one in the center is reaching toward a large plate with a vegetable apparently performing the dipping; the man next to him has two halves of a matsah in front of him; and the one to the far right is performing the ritual hand washing with the aid of a servant. In fact, nothing in the image hints at a scholarly discussion in an academic setting, as the adjacent episode implies.

In the following discussion, the idiosyncrasies pointed out in this section will be contextualized within the scholarship of Isaac Abarbanel and the historical circumstances of the early Sefardi Diaspora.

3. Discussion

3.1. The Rationale of the Illustration Program

The imagery in the Istanbul Haggadah is thus unique. It cannot be said for sure whether those who conceived the book to be and supervised the commission in del Tuppo’s atelier were not aware of the earlier traditions of manuscript haggadot or deliberately and consciously broke with those traditions. But the fact that the Ibn Nahmias brothers had already been experienced book-trade professionals back in Spain (Offenberg 1992, p. 121) sheds doubt on the former possibility and indeed suggests that they consciously created a new type of book, in a new medium, designed for a different kind of readership. The Haggadah was conceived as a modest book measuring 226 mm × 160 mm. Its plain script is consistent throughout, and with the exception of the qiddush page, there are no fancy fonts or initial decorations. All the texts including the vernacular instructions are printed in square type. The page layout is simple and does not feature columns or marginal texts.

The fact that the Haggadah and Abarbanel’s commentary were printed on paper with the same watermark and, thus, close in time suggests some links between them not only from a technical point of view, but also in terms of the emergence of these two projects. In his introduction to Zevah Pesah, Abarbarnel described the hardship of expulsion and migration in detail, whereas his occupation with the haggadah offered him consolation. Abarbanel and the Ibn Nahmias brothers belonged to the same community of exiles. It is certainly possible, then, that the brothers consulted with him about the imagery, and drafts of some portions of the commentary may have been available. I shall further elaborate on this setting in the next section, but first turn to explore these links as they are apparent in the imagery in the Haggadah.

Zevah Pesah was composed as a clearly structured tract with an introduction that explains what led Abarbanel to write the commentary. The introduction also includes a series of 100 questions about the haggadah and the rituals that in his view required elucidation. The main part of his work explains the haggadah verse by verse while referencing the questions posed at the beginning.

The episode of the Rabbis in bnei braq (Figure 5) raises, thus Abarbanel, several issues. The essence of the anecdote is the importance of the intensive narration of the Exodus:

A story about Rabbi Eliezer and Rabbi Joshua etc. The compiler of the haggadah (hamaggid) wanted to bring evidence for [the saying] that the one who adds to the narration of the departure from Egypt is to be praised; [this applies] even to scholars and wise men and those who know the entire Torah; this is what occurred to the five giants of their generation, to the Sages of Israel on the night of Passover when they reclined in bnei braq.

(Abarbanel, ZP 1505, fol. 8b)

Further, how exactly is “bnei braq” to be understood? Why did the Rabbis remain awake during the whole night? Why did they not feast with their families? Were the Rabbis staying in a town “baraq” and feasting with the citizens (“the sons”—bnei) of that town? Were they being hosted by different families?

After posing all these questions, Abarbanel did not read the episode as taken from real (Tannaitic) life but interpreted it allegorically. Based on the understanding of baraq as “shine” or “sparkle,” bnei b[a]raq are to be understood as “the shining vessels [on the ritual table] around which the Rabbis reclined. He continued to explain that the Rabbis made the said night a vigil in imitation of the Israelites remaining awake on the night of their departure from Egypt. In order to reenact the event of liberation from slavery:

[T]hese saints did as they (the Israelites) did; at the beginning they occupied themselves with the laws of unleavened bread and bitter herbs (מצות מצה ומרור) and the commemoration of the Passover, as did our fathers in Egypt, and then for the rest of the night they kept telling the story of departure; that is how they saw themselves as if they were departing [from Egypt].

(Abarbanel, ZP 1505, fol. 8b)

Hence, in contrast to the actual text of the haggadah, and with the help of allegorical interpretation, Abarbanel focused on the Rabbis performing ritual acts, with the understanding that the narration of the Exodus itself is a ritual act. Against this background, the depiction of the Rabbis performing the seder makes complete sense.

It is, in fact, likely that Abarbanel’s discussion of the Rabbis in bnei braq had an impact on the imagery of that episode also in later haggadot. While virtually all medieval depictions of the Rabbis show them engaged in study, conversation, or debate (Figure 12; in some cases just the portrait of a sage with a book illustrates the story), early modern images that postdate the Istanbul Haggadah, such as the second edition of the Mantua Haggadah and a series of haggadot printed in Venice between 1599 and 1605 show them at a seder table (Figure 13 and Figure 14). Hence, the medieval imagery did not imply that the Rabbis were narrating the Exodus on the occasion of the actual ritual, but rather that it was a generic scenario of study. In all these Venetian editions, the haggadah is accompanied by a commentary by Joseph ben Jacob Shalit of Padua, who adopted Abarbanel’s understanding of bnei braq as “shining vessels”. Zevah Pesah had become quite popular in Italy after the middle of the sixteenth century. After the first print in Istanbul, it was published in Venice in 1545, in Cremona in 1557, and in Riva di Trento in 1561. Copies were kept in many learned Jewish households. It is quite possible, then, that Abarbanel’s reading of the bnei braq episode also had an impact on the imagery of the story as a seder table scenario in early modern prints in Italy.

Thus, the Rabbis are shown at the seder in some analogy to the four-man meal setting. As noted, these images do not reflect scenarios taken from life. Each of the men shown in both compositions represents one particular ritual act. The two images taken together basically address all the principal rituals to be performed: blessing the wine, drinking the wine, reclining, dipping the vegetable into vinegar or haroset—shown twice—eating a vegetable after dipping it, narrating the story from a book, washing the hands, breaking the matsah, and eating maror. Abarbanel’s approach to the careful elucidation of the ritual acts of the seder one by one was somewhat similar and might also have inspired that imagery.

As mentioned, the table scenario with the four men is repeated twice in the Istanbul Haggadah. Reusing wood blocks in one and the same volume was common procedure in early print culture, and all the more so in a project that was planned quite hastily within the setting of refugee life and political turmoil. The option of repeated use of imageries was in fact one of the advantages of printing which was all about efficient production methods. It is first inserted between the story of bnei braq and the words of Rabbi Eleazar ben Azaria (Figure 6). Abarbanel explained that Rabbi Eleazar’s story is the second piece of evidence that the maggid (the assumes “author” of the haggadah) presented to show the importance of the narration of the Exodus (Abarbanel, ZP 1505, fol. 8v). The four men at the seder appear again on the opposite page as an illustration of the four sons (Figure 7), between “the Torah speaks of four sons […]” and the question of the wise son. It is likely that the four-man composition was inserted there to illustrate the four sons as a group, as is implied in the introduction to the question (Habermann 1972; Frojmovic 1996, n.6; rejected by Yerushalmi 1975a, Plate 1).

Traditional iconography of the four sons suggests that the wicked son is meant to represent (gentile) hostility. Most often shown as an armed warrior and in a few examples as a fancily dressed but likewise armed man (not following Jewish sumptuary laws), indicating that he has left the Jewish community and joined the gentiles (Figure 15). He is, thus, a convert, an outcast. His appearance as a warrior referencing hostile actions against Israel was designed to ‘other’ the wicked (Metzger 1973, pp. 149–63; Harris 2005; Barlow 2017 from a postcolonial angle). Daniel Boyarin demonstrated that modern visualizations occasionally show the wicked as a “Muscle-Jew” or as a soccer player considered alien to the Jewish tradition (Boyarin 1997, pp. 52–78). Similar interpretations are also found in literature on the Passover ritual, and it was suggested that originally the section on the wicked son was meant to address the late antique Jewish-Christian community and the necessity to refute Christian positions (Bloch 1978, pp. 152–63; Yuval 1999, p. 161).

However, none of these interpretations is implied in the woodcut, where the four sons are seated around the table, and it is not at all clear as to who is who. The image of the seder table from the first page of the text was thought to be good enough and clear enough to serve here as well. Clearly the wicked son is participating in the seder and is, thus, neither a gentile nor an outcast. In fact, the traditional reading of the haggadah section on the four sons does not necessarily imply such an understanding. David Biale even goes as far as to argue that the visualization of the wicked son as a knight or armed warrior in illustrated haggadot actually suggests that Jews must have borne arms (Biale 1986, p. 73). From this perspective, and as implied by several haggadah commentaries, the wicked son is rebellious, he challenges the law, and his question makes no mention of God (Eleazar ben Judah of Worms [d. 1232], Perush haroqeah 1984:69; Zedekia ben Abraham, Shibole haleqet, Seder Pesah 218; see also Tabory (2013, p. 174) on the Mishnah, Pesahim 10:7).

Abarbanel’s discussion of the four sons’ questions is very detailed and includes various suggestions toward interpretation. However, none of them implies that the wicked son is to be understood as a gentile, a convert, or an outcast. Abarbanel first suggested that the maggid meant to infer a social distinction: the wise son demonstrates his wisdom; the wicked doubts the law, is filled with the spirit of heresy, and, thus, vexes and annoys (מקנטר). Hence, he is an enemy from within. The simple son realizes his own lack of knowledge and tries to gain wisdom, and the fourth son is not able to ask questions and he has no knowledge at all. Abarbanel continued and became yet more abstract. Another solution he offered was to interpret the four questions as referring to the material qualities of things, their shape, their function, and their goal. Although the four portraits cannot necessarily be firmly linked to any of these interpretations, the fact that the traditional visual association of the wicked son with the gentiles does not reflect Abarbanel’s readings can explain why they can all appear at the same table.

In short, the two types of seder table representations in the Istanbul Haggadah, so familiar at first sight from the illuminated manuscripts, and yet quite different, were not conceived as scenes taken from life as was the case in their medieval predecessors. Rather, in a compact fashion, they visualize a series of ritual acts to be performed without any visual association with actual family settings. Traditionally, it is a man who performs these tasks or takes the lead in performing them, which explains the absence of women in these abstract settings.

The rest of the surviving images depict a selection of five plague scenes inserted into the Tannaitic discussions of the plagues. The reconstruction of the second quire of the Istanbul Haggadah indicates that the depicted scenes were meant to be a selection and not a full series. ‘Arov, murrain, locusts, darkness, and the death of the Egyptian firstborn are not shown. Two questions arise: what determined the selection of the plagues to be shown? Why is the plague of blood singled out as a full-page image?

The rabbinic tradition discussed various ways of categorizing the plagues. One way to do so was to distinguish between the seven plagues brought about by actions performed by Moses or Aaron and the three that were caused by God alone: ‘arov, murrain, and the death of the firstborn. None of these latter three are shown in our series. All those that are depicted belong to the first category, and in all the images Moses and Aaron are shown prominently performing the acts commanded by God. The first three images all show Aaron with his rod touching the river first to turn water into blood (Exod. 7:17–22; Figure 8), then to let frogs jump out of the waters (Exod. 8:2–4; Figure 9), and finally to bring up lice from the soil (Exod. 8:12–14; Figure 9). In the image of boils, the right section is missing; however, it is likely that in analogy to the other compositions, it showed Moses and Aaron throwing ashes toward heaven (Exod. 9:8–9; Figure 10). The last image shows Moses stretching out his arm (with a rod) and bringing about hail (Exod. 9:22–23).

Abarbanel addressed this distinction between the types of plagues briefly in Zevah Pesah (Abarbanel, ZP 1505, fol. 5a) and at greater length in his Commentary to the Book of Exodus:

[…] Of all the plagues that occurred in Egypt there were three that were performed by Aaron by means of the rod, and they are: blood, frogs, and lice; and three of them were brought forth by the Holy One Blessed be He, not with the aid of Moses and not with the aid of Aaron, and they are: ‘arov, murrain, and the death of firstborn […]. Three other [plagues] were performed by Moses, our teacher, be peace upon him, and they were: hail, locusts, and darkness, and one plague was performed by Moses and Aaron together: boils.

(Abarbanel, Perush hatorah 1997, vol. 2, pp. 124–25)

Abarbanel also elaborated on the observation that the haggadah singles out the plague of blood, as it is mentioned not only with the listing of all the plagues, but also in the context of the explanation of the miracles that were performed for the Israelites: “blood, fire, and pillars of smoke (דם, אש ותמרות עשן)” (Abarbanel, ZP 1505, fol. 4 on Questions 77 and 78). Another reason for the special status of the plague of blood is that it came as a punishment for the death of the Israelite infants (Abarbanel, Perush hatorah 1997, vol. 2, pp. 111–12).

The depictions also seem to echo Abarbanel’s discussions of the plagues with regard to specific iconographic details. Thus, unlike images of the lice in manuscript haggadot (Figure 16), which simply show animals and people affected, the latter scratching their bodies where the lice are crawling, the woodcut emphasizes that the soil actually turned into lice (Figure 9). The entire foreground of the picture shows lice creeping around on the ground. Abarbanel discussed this phenomenon at some length:

[…] The (plague) of lice was different from the (plague) of blood insofar as blood occurred (only) on those places that Aaron touched (with the rod). […] Rather, Aaron stroked the soil of the earth once and everywhere in Egypt the dust turned into lice, and that is why it is written: all the dust of the earth turned to lice throughout the Land of Egypt (Exod. 8:13). This means that with only one stroke that Aaron performed with his rod onto the dust of the earth lice came about everywhere in Egypt […] the rod in his (Aaron’s) arm hinted at the four cardinal points as well as up and down so that the plague would occur everywhere. […] This also explains, as I mentioned, that Scripture says: lice came upon man and beast; all the dust of the earth turned to lice (Exod. 8:13). This means that two different things happened at the time of this plague: firstly, lice came upon man and beast; note that here the word “all” does not appear, as in the phrase “all the dust of the earth”. […] This means that all the men and all the beasts that were near Aaron at the time the plague occurred were infested with lice, but not the rest of the men and beasts throughout the Land of Egypt; but this does not apply to the dust, because [the mention] of all the dust means that there were lice everywhere that dust is found nearby or far away […].

(Abarbanel, Perush hatorah 1997, vol. 2, p. 121)

The woodcut also emphasizes the conversation between Pharaoh and his magicians about the latter’s attempts to imitate the plagues. The plague of lice was the first that the magicians were unable to reproduce. Exodus 8:14–15 reads: “The magicians did the like with their spells to produce lice, but they could not; and lice were upon man, and upon beast; and the magicians said to Pharaoh: This is the finger of God” (translation: JPS 1985). Abarbanel’s Commentary to the Book of Exodus elaborates on this conversation, emphasizing that the magicians realized that Moses and Aaron did not perform sorcery but that the plague was brought about by divine power (Abarbanel, Perush hatorah 1997, vol. 2, p. 122). The image in fact shows Pharaoh gesturing with his left hand in what seems to be a rebuke to the magicians, whose statement about God’s finger turns here into an apology.

The selection of these particular plague scenes, finally, avoids visualizations of divine actions. Time and again Abarbanel set apart those plagues that were brought about by God and on the presence of the shekhinah (Abarbanel, ZP 1505, fol. 5 on Questions 74–76). If the Ibn Nahmias brothers as the commissioners of the woodcuts were inspired by Abarbanel’s readings of the haggadah, it is likely that they were well aware not only of his strong emphasis on God’s immediate involvement in the Exodus, but also of his general approach to the transcendence of the non-anthropomorphic Godhead, which, most likely, did not tolerate visualizations of divine acts. This approach had its roots in Abarbanel’s interest in Maimonideanism with its very clear views on anthropomorphism, which did not allow for even an imagined visualization of the divine (Halbertal 2006). Like many pre-expulsion Iberian Jewish intellectuals, Abarbanel was deeply influenced by Maimonides, while at the same time, he was often critical of some of his “master’s” ideas. (Lawee 2001, pp. 32–34, 41, 55–56). In fact, the issue of theophany and the possibility of corporeal apprehension of the divine was addressed as early as in Abarbanel’s first work, ‘Ateret zeqenim, in which he discussed the theophanic event at Sinai (Lawee 2001, pp. 59–62). Thus, for example, Abarbanel challenged Maimonides’s take on Genesis 1:27 on God creating man in his image. While their approach differs in several aspects, however, both scholars agree with regard to God’s incorporeality (Cohen-Skalli 2021, pp. 234–37).

3.2. A Haggadah for Sefardi Refugees

Frojmovic’s observation that in all likelihood the Ibn Nahmias brothers commissioned the woodcuts in del Tuppo’s workshop in Naples suggests that the Istanbul Haggadah was among the first projects they embarked on after landing in Naples in the summer of 1492, as they were trying to set up a printshop there. When they approached del Tuppo, they must have had a clear concept of what the imagery was meant to convey and what shape the book should take. This concept, I contend, owed a significant debt to Isaac Abarbanel’s approach to the haggadah, which was crystallizing around the same time.

Adri Offenberg assumed that the brothers arrived in Naples as part of a group of immigrants that circled around Abarbanel and that the 1505 printing of Zevah Pesah should be seen as a by-product of this relationship (Offenberg 1992, p. 121; 1996, pp. 226, 231). Jonathan Ray shows, that in general, the refugees preferred to travel in relatively large groups (Ray 2013a, pp. 37–38). Following up on Offenberg’s suggestion, Cedric Cohen-Skalli argues that Abarbanel’s work after 1492 and the early printing projects by the Ibn Nahmias brothers in Istanbul reflect clear attempts to meet the spiritual needs of the community of exiles (Cohen Skalli 2007; Cohen-Skalli 2021, chp. 16). The most pressing projects concerned the ritual law (the 1493 Arba’a Turim), the liturgy including biblical readings (the Pentateuch with Haftarot and Abarbanel’s commentary to Pirqe Avot, all printed in 1505), and exegeses. The Haggadah was a fitting complement to this program, and the fact that it was printed on the same paper as Zevah Pesah confirms this link. The Zevah Pesah edition is larger than the Haggadah (measuring 290 × 220 mm) and uses cursive script types for the commentary along with square ones for the haggadah text. Furthermore, the breadth of a sheet from the ZP edition corresponds to the height of a haggadah sheet. For the included haggadah sections, it uses square types that are very similar, albeit not identical to and somewhat larger than those used for the Istanbul Haggadah. They are of the same size as the enlarged types used in the Haggadah for initials. All these circumstances taken together suggest that the links I noted between the Istanbul Haggadah and Abarbanel’s Commentary were not accidental.

But the Haggadah and Zevah Pesah served for more than just liturgical needs. As Abarbanel explained at length in his introduction to his Commentary, the Passover theme had great potential to offer consolation to a population that found itself in a state of crisis and turmoil—both physically and morally. He described in great detail the hardship these people experienced during travel and after their arrival: danger lurking everywhere, losses of family members, the threat of being kidnapped and forced into baptism, great poverty, and in general difficulties in getting settled (for other sources: Hacker 1980, 1992; Zeldes 2017, 2019). From his loneliness in exile and out of bitter distress, he recalled better times of family celebrations and, thus, decided to turn his back on depression and, instead, reflect on divine law:

[…] I said [to myself] it is time to prepare an encompassing commentary on the stories of the Passover [that deal] with the exile and its causes and with the various ways of redemption and its wonders, […] and I will entitle my tract “Passover Sacrifice”, because it is my sacrifice to God from a broken spirit […]

(Abarbanel, ZP 1505, fol. 2a)

The link between the story told in the haggadah and the fate of exiles recurs in one of Abarbanel’s slightly later reflections on the expulsion: his commentary on Ezekiel 20, authored in 1503, while he was in Venice. He again described the hardship of migration, while repeatedly using the expressions “with a strong hand, outstretched arm, and wrath poured out” (ביד חזקה וזרוע נטויה ובחמה שפוכה) associated with the haggadah. These terms, however, are understood here not only as divine tools of redemption, but also as tools of divine punishment for religious laxity and conversion. More importantly, the converted are still given to God’s judgment and to the hatred of the gentiles, whereas those who moved on to Islamic lands and remained Jewish are exempted from this kind of punishment. Hence, “‘with a strong hand’ applies to the committed Jews in the Land of Edom, ‘with an outstretched arm’ pertains to those in the Land of Ismael, and ‘with wrath poured out’ refers to those who changed their religion” (Abarbanel, Perush hanevi’im 2009). A clear and somewhat comforting message to the exiles who had avoided conversion by all means.

Obviously, the imagery of the woodcuts could not have been based on the final version of Zevah Pesah. In the last sentence of the commentary Abarbanel noted that “its completion occurred […] in Monopoli […]” on the eve of Passover 1496 (Abarbanel, ZP 1505, fol. 39r). Given the role this tract was supposed to play in the lives of the exiles, however, it is perhaps feasible to speculate that Abarbanel had made plans to write Zevah Pesah as early as when he was in Naples in the fall of 1492, which was around the same time that the Ibn Nahmias turned to del Tuppo. But when he left for Messina to accompany his patron, Alfonso II of Aragon, on an official mission, Abarbanel had to interrupt his work. We know that during his stay in Naples, Abarbanel was particularly active composing several texts, but most of them have not survived. While in Sicily he would not have found conditions suitable for further engagement in scholarship (for biographical details, see Lawee 2001, chp. 1; Cedric Cohen-Skalli 2021). “The completion occurred (היתה השלמתו)” is a standard formula that Abarbanel occasionally used in his colophons, which does not say much about when he began the work and when he penned drafts of it or first conceived of its contents. A similar wording appears at the end of his Mashmi’a yeshu’ah or his Merkevet hamishneh both also completed in Monopoli (Abarbanel, Mashmi’a yeshu’ah; Abarbanel, Merkevet hamishneh). If there was a scholarly discourse among exiled Sefardi intellectuals who landed in Naples in 1492, ideas about how to stabilize the community of refugees, ideas that eventually led to the completion of the Commentary could well have been part of it.

Cohen-Skalli elaborates on Abarbanel’s efforts to stylize himself as a spiritual leader of that community (Cohen-Skalli 2021, chp. 11–13). Not only did he turn to an enhanced literary activity immediately upon his arrival in Naples (geared to this image of leadership), but part of this effort was invested in preaching. Joseph Hacker has shown that sermons were central in the effort to establish a firm community of exiles (Hacker 1987). It is likely that in the setting of large numbers of refugees arriving in Naples and their need for leadership, his occupation with the haggadah came up in sermons, teaching sessions, and such. Given that Zevah Pesah was specifically geared to offering comfort to a community scarred by the hardship of expulsion and travel, it is conceivable that drafts of the text already existed. These efforts were soon interrupted again, and the days of Abarbanel’s departure to Sicily—after the French conquest in 1494—were days of chaos and persecution for the Jewish community, during which time, Abarbanel’s belongings, books, and writings were lost (Cohen-Skalli 2021, pp. 215, 218).

The nature of the Istanbul Haggadah as a modest and simple book breaking with the Sefardi tradition of luxurious Passover compendia for wealthy scholars seems to have been appropriate for addressing impoverished exiles. However, for the time being the Haggadah project was postponed. At some point apparently still in 1492, the Ibn Nahmias left Naples, perhaps to escape the outbreak of an epidemic (Lawee 2001, pp. 20–21; Lacerenza 2002, pp. 407–8). By late November 1492, the expectation of an epidemic began to be an issue. Already since May 1492 an increase in deaths had been recorded in the city, likely caused by some sort of contagious disease. A full outbreak of the plague, for which the refugees were eventually blamed, occurred in February 1493 (Lopez 1989, pp. 26, 91–100; Zeldes 2019, p. 235). The brothers moved to Istanbul, whereas Abarbanel remained in Italy (and Corfu for some time). There, in December 1493, the Ibn Nahmias first completed the print of Jacob ben Asher’s Arba’a Turim, which was at the time an indispensable guide for conducting a basic Jewish life. Soon thereafter Hebrew book production stopped, perhaps owing to the interruption of the paper supply in consequence of the war with Venice (1499–1503) and the tensions that led to it (Offenberg 1992, chp. 5; 1996; Hacker 2012, pp. 22–24). The publication of the Haggadah and Abarbanel’s commentaries had to be postponed yet again. However, after the French conquest of Naples and the subsequent expulsion of the Jews from there, it was clear that the project had not lost its applicability. Many of the Sefardim had moved on to Istanbul or Salonica (Rozen 2010, chps. 3 and 7). Hence, the final Haggadah project was addressed to the community of refugees in Istanbul and elsewhere in the Ottoman Empire.

It was there that the exiles would eventually build a new collective identity based on the notion of Sefardi Diaspora or double Diaspora as a collective catastrophe affecting the Iberian Jewish world. In fact, Abarbanel’s efforts to stylize his persona as a leader also implied the engendering of a collective identity, a notion that he shared with numerous other Sefardi intellectuals (Ray 2013b; Cohen-Skalli 2021, chp. 11). The majority of exiles were from Castile and used the Castilian vernacular, and that was the language chosen for the occasional instructions inserted into the Istanbul Haggadah. Eventually, once a collective identity had developed, that language would evolve into Ladino (Schwarzwald 1999; Ray 2013a, pp. 39, 136–38). Ray seeks to integrate the study of Sefardi refugees within current Diaspora Theory and thus detach the discourse on the Sephardim from the traditional concept of Jewish galut and focus on sub-ethnic diasporic experiences. Against this background Ray points out, that the emergence of a collective identity was complex, as the refugees first fostered Catalan, Aragonese, or Castilian identities (Ray 2008, 2013a, 2013b, 2021).

Sources describe not only the physical hardships the refugees had been through but also a state of spiritual demoralization in the face of losses of relatives, converts remaining in Iberia, a general anxiety about the fate of the Jewish nation, and, indeed, the need for spiritual guidance (Hacker 1989). In this vacuum and despite sometimes dire social conditions, intellectual life soon began to flourish in Istanbul, and religious conduct was considered a crucial factor in establishing stability and a firm identity (Ray 2013a, chp. 6). The elite sponsored numerous educational institutions. A lively book culture played a crucial role in the efforts to create a community with a shared identity and to consolidate ritual life while preserving specific Sefardi customs (Hacker 1987). Twenty-three out of thirty-two sponsors of printing projects undertaken until 1530 were Sefardim and the rest were Romaniote. Up to 1566 at least 126 Hebrew books were printed. Some of these were sold in small sections of single quires through a subscription system, so that they became affordable to wide circles (Yaari 1967, pp. 17–21, 103; Rozen 2010, pp. 251–70). Clearly this market addressed not only the elite, but also the middle class.

4. Conclusions

The Istanbul Haggadah, a small book of two quires with a few handsome woodcuts, one of them used three times, was a book for the middle class. My reading of its material and artistic features suggests that it was conceived as part of these scholarly efforts to trigger a collective identity among the refugees, to provide ritual tools to help conduct a proper Jewish life that would maintain that identity, and, in particular, to find consolation in the performance of the Passover ritual. It seems that the project was quite successful. Between 1492 and 1497 an estimated number of 2500–3000 Iberian Jews arrived in Istanbul (Hacker 1992). The Haggadah was printed in two editions. Scholars assume that liturgical incunables were printed in a few hundred copies (Schmelzer 2006), but that this figure grew after 1500 to 1000–4000 (Baruchson-Arbib 1989). It thus seems that the two editions provided the majority of Sefardi households in Istanbul and numerous copies could perhaps have been sold elsewhere in the Ottoman Empire. Some made it to Egypt and eventually to the Genizah in Cairo. Except for the few torn fragments found there, it is likely that all the copies were used up, were worn out, and perished.