versions found in the Cairo Geniza. That is different from the fluidity we find in R. Judah he-hasid’s Sefer Hasidim and his other pietistic writings or in some of R. Eleazar of Worms’s compositions.2
When an author writes multiple parallel versions of a single composition or names them differently or gives the same title to a single work and to a collection of related works, we need to rethink what we mean by a text. We would normally be inclined to try to sort out what appears to us as bibliographical confusion by assigning a unique title and author to each work, but we should not project our ideas of bibliography, derived from familiar single-authored unique compositions, back into medieval Ashkenaz where ideas about an author and a text were often more fluid.
Among examples of textual fluidity in R. Judah he-hasid’s writings is Sefer ha-Kavod, a composition he may have written more than once and that overlaps in some ways with Sefer Hasidim.3 Jacob Freimann showed that manuscripts of Sefer ‘Arugat ha-Bosem, written by R. Judah he-hasid’s student R. Abraham b. ‘Azriel, contain quotations from a Sefer ha-Kavod attributed to R. Judah he-hasid.4 Two Oxford manuscripts that contain a work called Sefer ha-Kavod and a truncated second discussion also about the Divine Glory (kavod) do not contain the passages quoted from “that work” in Sefer ‘Arugat ha-Bosem. The two Oxford manuscripts do, however, contain passages attributed to R. Judah he-hasid’s Sefer ha-Kavod that are found in R. Moses Taqu’s Ketav Tamim, a work that polemicized with Judah’s writings.5
Joseph Dan presents as a dilemma having to decide either that one of the two medieval authors is wrong in attributing his quotations to Judah’s Sefer ha-Kavod or that Judah wrote two versions of Sefer ha-Kavod. He considers which possibility is “more reasonable” (sevirah yoter). Although he thinks choosing between the two possibilities is “somewhat arbitrary” (yesh bah middah shel sheriratiyut), he decides in favor of Moses Taqu’s reliability, with the result that the Oxford manuscripts containing the quotations R. Moses Taqu attributes to Sefer ha-Kavod is in fact Sefer ha-Kavod. However, in light of Ashkenazic book culture’s fluidity, Dan’s second hypothesis, that Judah wrote more than one Sefer ha-Kavod, is very probable.6 R. Abraham b. ‘Azriel had one version of Sefer ha-Kavod, possibly lost, and R. Moses Taqu had a different one, that which was found in the Oxford manuscripts.7 There was no author’s single original text of Sefer ha-Kavod any more than there was one of Sefer Hasidim.8
The multiple character of Hebrew book composition in medieval Europe also explains some apparent anomalies in the sources such as a reference to “Sefer Hasidim” in a paragraph in Sefer Hasidim.9 The author did not think of a title as belonging uniquely to only one edition of related passages that he strung together more than once.10
Aside from Sefer Hasidim and Sefer ha-Kavod each consisting of more than one edition, the two works themselves seem to overlap. We see this from quotations of Sefer ha-Kavod found in Sefer Hasidim. Sometimes the author of Sefer Hasidim refers the reader to Sefer ha-Kavod for a longer treatment of something he mentions briefly in Sefer Hasidim. This would mean that a version of Sefer ha-Kavod already existed before that part of Sefer Hasidim was written. But in other passages in Sefer Hasidim, Judah indicates that more will be written in Sefer ha-Kavod. Here it seems that at least parts of Sefer Hasidim are earlier than a version of Sefer ha-Kavod.11
That such an overlapping relationship between Judah’s different writings exists is reflected in a comment R. Eleazar of Worms makes in his legal code Sefer ha-Roqeah. He says that he has taken passages on the topic of caring for the dead “from the notebook of R. Judah hasid,” but then he continues, “and you will find proof of his words in Sefer ha-Kavod” (mi-mahberet r. yehudah hasid u-ve-sefer ha-kavod timza re’ayah li-devarav).12 These comments suggest that Judah was writing passages in one “book” with the other “book” in mind; or, we might even think of them as two forms of his short pietist writings that he incorporated into differently named works, some as paragraphs of Sefer Hasidim and the other as paragraphs of Sefer ha-Kavod. The evidence suggests that he wrote some passages more than once and put them into separate works that had different titles.
A third text attributed to R. Judah he-hasid can overlap with Sefer Hasidim or with Sefer ha-Kavod. R. Judah he-hasid’s list of Commands (Zava’ah) is preserved in scores of manuscripts in different combinations of up to seventy short paragraphs.13 The title is often mistranslated as an ethical “will,” but the term refers to a series of discrete religious demands. The text was very popular. In the first edition is an introduction that says Judah was writing his rules for three different audiences: his family, other Jews, and the whole world. The “will” includes: “If there are two weddings the same week, one will become poor or go into exile and die. As to gentiles, they should not make two knights (shenei parashim) on the same day, but don’t tell them or they will be careful to avoid doing it”!14
Sometimes the Zava’ah was copied into codices that also contain parts of Sefer Hasidim.15 The two texts were often cited interchangeably, especially paragraphs of the Zava’ah as “Sefer Hasidim.”16 For example, in MS New York, JTS Mic. 5252 f. 93r: “I found in writing (something) copied from Sefer ha-Hasidim from the Commands of Rabbeinu Juda he-hasid, may his memory be for a blessing, and they are seventy commands” (mazati katuv ve-hu mu‘ataq mi-sefer ha-hasidim mi-zava’avot (?) rabbeinu yehuda [alef at end] hehasid…ve-hem shiv‘im zivuyyim).”17 Both texts have the same attributed author. Both are written in numbered paragraphs and command pietistic behavior. And both were often copied in the same manuscript codex or printed together.
There is also overlap between Sefer ha-Kavod and Judah’s Zava’ah. R. Jacob b. Moses Ha-Levi Mölln (Maharil) refers to seeing Sefer ha-Kavod, and that it included Rabbi Judah’s Zava’ah.18 Consider again the passage on mourning customs that is found in R. Eleazar of Worms’s Sefer ha-Roqeah (par. 316, end) that begins: “Copied from Sefer ha-Kavod written by the great man, R. Judah Hasid m[ay] the m[emory] of the r[ighteous man] (be for a blessing)” (ne’etaq mi-sefer ha-kavod she-yasad he-ish ha-gadol r. yehudah hasid zz” l). The text quoted is a very close version of the first nine paragraphs of Judah he-hasid’s Zava’ah even though it refers to the text as “Sefer ha-Kavod.”19
Apart from the compositions that R. Judah he-hasid wrote more than once, he also wrote multiple versions of short texts on the Divine Unity (shirei ha-yihud), and he wrote more than one commentary on the prayer book.20 Other parallel passages are to be found in other compositions that Judah wrote, such as his writings about nature, Zekher ‘asah le-nifleotav.21
R. Eleazar of Worms’s Fluid Writings
R. Eleazar of Worms also produced more than one edition of some of his many different writings. He wrote several short penitential texts about how to atone for one’s sins, not just one that he revised. This act of recomposing resulted in related penitential texts with different names such as two versions called Hilekhot Teshuvah, and others called Moreh Hata’im, Yoreh Hata’im, and ‘Isqei Teshuvot. They all contain overlapping parallel passages in different sequences as well as unique ones. One composition cannot be reduced to the others as the product of the author’s revision of an original text or of scribal errors. The textual evidence argues for Eleazar of Worms’s composing each of them.22
Overlap also exists in Eleazar’s halakhic writings. He composed a specific halakhic work that he called “Sefer ha-Roqeah,”23 but that title can also refer to a different halakhic text whose modern editor named it “Ma‘aseh Roqeah” (the Work of Roqeah), but medieval authors cited that text simply as “Roqeah.” Eleazar wrote a third legal text that is also sometimes referred to as “Roqeah.”24 Thus, when medieval authors refer to “Sefer