Notes from Limmud 2010

The Story of the Aleppo Codex

Bernard Schneider

The Aleppo Codex is a complete codex of the Bible, with diacritical markings and annotations. It was written ca. 930 in Tiberias. It's about 26.5 by 33 by approximately 20cm. It's not particularly well known. It's in the Shrine of the Book in Jerusalem. (And it doesn't look that impressive compared to the Dead Sea Scrolls upstairs.) But it is one of the most important documents in Jewish history—together with the Dead Sea Scrolls, one of the most important Biblical manuscripts we have.

Adoption of the codex and development of the Massoretic notation

A codex is a bound book (written before the late Middle Ages), as opposed to a scroll. Codices were invented by the Romans in the first century CE, and in wide use by the fourth century. They were a major technological innovation. A scroll can only be written on one side, and hence takes twice the room it needs; it's also not random-access—you need to spend time scrolling it if the part it's opened at is not the part you want to read.

Jews didn't adopt this new technology very quickly; this was because the rule had already been laid down for writing sifrei Torah on a scroll. The first evidence of codices being used by Jews is from the eighth century, the first extant Jewish codex from the earlier tenth century CE.

By this time, Aramaic had displaced Hebrew as the primary language of the Jews. There was a decline in knowledge of Hebrew. There is a dispute in the Gemara about how well the scholars of Galilee preserved their knowledge of Hebrew.

The scribes were aware of this issue, and came up with several solutions. The earliest, mentioned in the Gemara (Gittin) is סירוגין (serugin) texts, which would list for each verse the first word, and then anything non-obvious the writer of the text thought you should know about how to read it. This was good as far as it went, but you needed a full text to accompany it to make use of it.

Another problem in Hebrew was the lack of indication of the vowels. You were supposed to know how to pronounce it. But you don't always know, which leads you to a lot of homographs—words written the same but not pronounced the same. אמהות קריאה (matres lectionis) were intended to solve this. But this was also of limited use: matres lectionis were not permitted in any scroll intended for liturgical use. (Many rabbis say this applies to נ״ך too, including reading Haftarah.)

Starting in the seventh century, more complete notational systems were developed, in Babylonia, in Israel, and in Tiberias. The most complete, the latest to be finished, and the one we use this day is the Tiberian system. It includes representation of cantillation. The version we use today is simplified—there were probably different versions—and adapted to the pronunciation (and trop) we use today.

The culmination of all this development is found in the Aleppo Codex, also known as the Crown of Aleppo. It uses the Tiberian system of vocalisation, the Tiberian system for noting cantillation, and incorporates the highly detailed and consistent Ben Asher masorah.

Page from the Codex

The Masorah is divided into the Lesser Masorah, written on the side and referenced with little circles over the word, to indicate, for example, that this word is spelled this way here only. The Greater Masorah is amplification of these. There were two schools, named after scribes: Naftali and Ben Asher. The differences between them were relatively minor, but of course with תנ״ך every letter counts.

Who wrote the Codex?

The text was written by Shlomo ben Buya'a, with the diacritics and notations by Aharon Ben Asher, the fifth and last of the Ben Asher line. We don't know much about them; this is because we do not have the colophon for this manuscript. These books were looked upon almost like a Sefer Torah, and had to be written according to very strict rules. There is room for precision, but not for creativity. The colophon would have been typically where the scribes would have identified themselves and said why they wrote the book. It's almost inconceivable that the colophon would not have been written; both because all books had one, and because the book was designed to be an important text.

How do we know the Codex is their work? The Ĉūfrut-Kale codex in St Petersburg is clearly in the same handwriting, and is identified in its colophon as having been by ben Buya'a. The support for Ben Asher is a little less tight, but it still correlates about 93% with other texts of his. And of course we have the tradition identifying him as the co-author.

After this, the creative period of the Tiberian Masoretes ended.

Later History of the Codex

The Rambam specifically cites the Aleppo Codex as authoritative. He says he wanted to write a Sefer Torah, and he reviewed various texts, and found they were all over the map in terms of what paragraphs were open and closed, etc. "And I used this codex," he wrote, "which was edited by Aharon ben Asher, which was in Egypt."

The Codex was already accepted, and cited as a primary text long before Maimonides. For example the scribe who wrote the Leningrad Codex (also in the Russian National Library), written in Cairo in 1008, said he copied off the Ben Asher codex. It's clear the Aleppo Codex was designed to be a core text, because of the work they put into it. They went over it with a fine-toothed comb, and there are almost no mistakes (in terms of lack of internal consistency.) It's been checked and cross-referenced, and checked again.

We know of no earlier complete codex of the Bible. [Though the codex in the British Library may be older, as it refers to Moshe Ben Asher, (Ben Asher's father) without "ז״ל", therefore as still alive. Its colophon too, is missing, though.] Given how many books have been destroyed, that doesn't mean there weren't any, though. It was already being called הכתר, or التاج (al-Tāj).

The Codex was written in Tiberias ca. 930 CE; by 1050 it was in Jerusalem. Though there was no colophon, there was a dedication, which said that in 1050 someone donated the Codex to the main Karaite synagogue in Jerusalem. In the eleventh century the Seljuks and Fatimids were fighting over Jerusalem; by 1099 the Crusaders had conquered Jerusalem, and there was a bloodbath. The Codex was ransomed to Fostat (near Cairo) in 1100, which is where Maimonides read it in 1170/1180. (We assume the references in the Cairo Geniza to the Crown refer to it.) After that, it arrived in Aleppo at some point, probably after 1204 and definitely before 1479, when Saadya ben Cohen ben David of Aden said he saw the Codex used by the Master [Maimonides].

It might have been brought by Maimonides' great-great-great-grandson (though it was never possessed by his family). Aleppo was a centre of learning. But it's a world-famous Codex in the Jewish world. The Aleppan Jews become very proud of their possession of it, but they were very paranoid about it; they hardly let anyone see it, including members of their own family, even though the dedication says it should be made available to any believer, Karaite or Rabbanite, who wants to check the text of their own Sefer Torah. It makes one wonder if it got to Aleppo legitimately. There was a tradition that if the Codex left Aleppo, the community there would fall.

The Codex in Aleppo

Aleppo was in existence not later than the eighteenth century BCE. There's a tradition that Abraham fed milk (בלח) to the poor there, hence its name حلب (Ḥalab). Or maybe the name came from the white chalk there, halaba. The Ottomans conquered it in 1517. It was on the trading route, and commercially significant; it had a large hinterland stretching into Anatolia. When the trade routes changed with the advent of the Suez Canal, though, and the Turkish Republic was declared with the border thirty or forty miles north of it, it was cut off from its hinterland and went into decline.

The Jewish community goes back to the first century BCE, when Syria [lacuna]

The original Great Synagogue was built in the fourth Century; Timur the Lame (Tamerlane) destroyed it, but it was rebuilt. There were a lot of scholars who came from Aleppo; like all Jewish communities their welfare depended on that of the city in general. In the late nineteenth century, Arabs became much more hostile towards Jews. Historically the Arabs Muslims had seen the (Arab) Christians as a big threat: a potential fifth column for the western powers. Once Zionism came about, that changed: the Jews became a perceived threat, and the Jews began to emigrate.

In the 1940s, when the Vichy regime had control of Syria, Jews were nervous about the fate of the community and the Codex. There were three trips in 1943, but the Aleppan rabbinical leadership absolutely refused not only to let it leave Aleppo, but even to copy it, or even to spend a lot of time comparing it. Umberto Cassuto, chief rabbi of Florence, a major scholar, got to see it for a couple of days. He was the last person to see the Codex complete.

Immediately after the UN vote to partition Palestine into Jewish and Arab states in November 1947, there was a "spontaneous" riot against the Jews of the city (and elsewhere)—so spontaneous that there were lorries with petrol to burn the Great Synagogue, which was gutted. The Codex went missing.

After a while, rumours came out that the Codex was not destroyed, but that they had had to say it was to get the Syrians off their tail. There were further plans to try and retrieve it, and then the State of Israel was declared, and all access was cut off. [For decades, Syrian Jewry was completely cut off; when I was an undergraduate UJS was campaigning to get them permission to emigrate, which they now mostly have.]

The riot was on 30 November 1948; the Codex, or part of it, did not appear again until 1957. There are at least seven contradictory accounts of who saved part of the Codex. No one seems to want to take credit. The sexton of the synagogue, Asher Baghdadi, apparently was never questioned.

There were actually four codices preserved in the Great Synagogue. Two survived, part of the Aleppo Codex and one other; two did not. So per[haps] [lacuna: possibly that the other two codices were sacrificed to allow the Aleppo Codex to be removed.]

Murad Mordechai Faḥam, an Aleppan businessman, came back to Syria after being expelled. He had a lot of money, and came back. He found his former business partners were told not to deal with him. He made plans to leave again, and was approached by the Haḥam, Moshe Tawil, to smuggle the Codex out of Syria. He wrote a specific letter lifting the curses on taking the Codex out of Syria, but even then it took several years to get the Codex out.

Faḥam smuggled it out to Israel, but who was he supposed to give it to? Faḥam to his dying day said, and said in court, he was told by Haḥam Tawil to get it out and give it to "someone whom you see fit to give it to". So he gave it to President Yitzchak Ben-Tsvi, who had been following this affair for years.

The Aleppan community in Israel was not so pleased, though, and said it was in trust for the Aleppan community. They were furious; they wrote letters and saidy no one had the right to give the Codex to anyone. More to the point, they sued Faḥam in the Beth Din in Jerusalem.

Faḥam was a stubborn character; he decided he was right and it should be the patrimony of the entire Jewish people and should be owned by the State of Israel. So he agreed to go to the Beth Din. A year later, Moshe Tawil came out from Syria and under cross-examination admitted he did not give any specific instruction to Faḥam to give it to the Aleppan rabbis. [ETA: Nope; this is Faḥam's version of the story. As the book The Aleppo Codex by Matti Friedman reveals, Tawil actually said nothing of the sort, but rather than Faḥam betrayed the trust put in him.]

The compromise worked out is that it's held by the Ben-Tsvi Institute, on permanent loan to the Shrine of the Book.

The state of the Codex today

Today, 295 pages exist of approximately 487 original. The ink has faded in many places (although it has been reconstructed). The bottom corners are rounded and discoloured by fungal damage in the past, but there is no evidence of damage from the fire in the Great Synagogue. (A conservator in Israel did an experiment and exposed a sheet of klaf to heat (not fire). It curled; but the Codex shows no signs of its sheets curling.)

What is missing? Most of Torah, up to Deut. 28:17; from the Song of Songs mid 3:11 to the end (including Ecclesiastes, Lamentation, Esther, Daniel and Ezra-Nehemiah). Also, fifteen additional pages in the middle from 2 Kings, Jeremiah, the Twelve Prophets and Psalms. Also the Introduction, Dedication and final Masoretic annotations. We have the text of these but not the original.

We have three photographs of pages from the missing sections: William Wickes photographed one page of Genesis in 1887 Joseph Segall photographed it in 1910; it was also photographed by Nissim Beḥar. There may be other ones too.

There are several texts which appear to have been copied from the Aleppo Codex over the centuries. Also rabbis sent Tenachim asking for them to be checked against it.

Hijari Torah in JTS. [The speaker was going too fast here for me to be able to take more than very cursory notes; I have no idea what this means.]
Menashe Sithon's responses to Yaakov... [This page explains: In the 1850’s R. Ya’akov Sapir, an Ashkenazi rabbi of Jerusalem, composed a long list of more than 500 remarks on the spelling, vocalization, and accentuation in various places in the Torah, the haftarot and the Five Scrolls. This list was sent to R. Menashe Sithon, a nasi in Aleppo, who was asked to examine the codex and write “thus” or “not thus” next to every item on the list.]
...

Significance of the Codex.

The spelling of דַּכָּא in Deut. 23:2, based on Masorah for Psalms 90:3: Is it with an א or a ה? With a ה in Ashkenazi sefarim. The Yemenites and some mizrachim write it with an aleph. They are very careful, subject to fewer influences from outside, and follow the Maimonidean tradition. [lacuna] says this word is written with an aleph in three places in Tenach; and one of them is this place.

The open and closed sections [i.e. paragraphs ending in either a new line or nine spaces] in Esther. Our tradition is that the Book of Esther is written only with closed sections, as we don't know where the open sections are. We don't have Esther in the extant Codex, but we do have annotations taken by other people (not the Rambam—he only did this for the Torah), and it turns out there are six open sections in the Book of Esther.

The proper format of פַּרְשַׁת הַאֲזִינוּ. The Rambam in Hilchot Sefer Torah says הַאֲזִינוּ is supposed to be written in sixty-seven lines; in some texts, it's seventy lines—and in the printed editions, it's always seventy lines. The main argument made for the Aleppo Codex not being the one the Rambam consulted is the fact is has sixty-seven lines. In Ashkenazi and also Mizrachi Sifrei Torah there are seventy lines. Cassuto and the rabbis then concluded this wasn't the Rambam's text.

But another scholar went through manuscripts of the Rambam, and it turns out the most reliable ones, including one in the Bodleian library with his signature saying "I proofed this text", say "sixty-seven lines".

This is the only codex from that period to have sixty-seven lines. It's apparently very difficult to write it in sixty-seven lines. So they either had a variant tradition and changed it to seventy, or they thought the Rambam must have made a mistake. (The Yemenites have sixty-seven lines too.)

So are our Torah and other scrolls, then, invalid? If we hold that one missing letter or added letter or incorrect paragraphing makes the scroll invalid—the position of the Rambam and most other people—and accept that the Aleppo Codex is correct—then the implication is that every Ashkenazi Sefer Torah today is פָּסוּל.

Jewish learning notes index