THE RIGID OBLIGATION that is imposed upon the sofer to safeguard the integrity of the text of the Torah scrolls he writes derives from the biblical statement "Thou shalt not add thereto, nor diminish from it" (Deut. 4:12, 12:32). From antiquity to the present day this passage has been interpreted by the scribe as a direct command to refrain from making any changes whatsoever in the sacred text and to be as careful as is humanly possible to eliminate all copying errors. The passage has also become one of the foundation stones upon which rabbinical authorities have throughout the centuries erected a corpus of laws governing the multitude of steps the scribe must take in the writing of a Torah scroll. Supplementing these laws, moreover, is the great body of tradition that has evolved touching the minutiae of the scribe's work, the scribe being required to adhere to this tradition just as though it were law.

It must be understood, however, that while certain soferic procedural rules and customs have come down to us from ancient times unchanged, others have been been altered over the course of the centuries. Moreover, not all these venerable rules and customs are universally observed. Hence, although the text remains absolutely the same from age to age and from region to region, a number of the physical aspects surrounding the writing of a Torah may differ somewhat. The calligraphic style displayed by a Pentateuch written in the Sefardic tradition is thus not quite the same as that of a scroll written in the Ashkenazic tradition; and the medieval sofer was compelled to make do with less durable quills than those that were to became available when, as a consequence of the discovery of the Americas, the turkey was introduced to the Eastern Hemisphere.

The preparation of a Sefer Torah, it must also be understood, has in itself come to be considered a religious act, one which demands not only complete accuracy on the part of the writer in his transcription of the text, but also the performance of certain carefully prescribed rituals as he proceeds with his work.

The scribe is expected to be a learned man of impeccable character. Upon awakening each morning he is required to purify himself by partaking in a ritual bath (mikveh) and to recite his morning prayers. Only then may he begin his scribal duties.

The parchment used to make a Torah scroll is prepared from the skin of a kosher animal that has been treated by soaking in limewater and other chemicals, and then scraped free of hair. Quills are made from a kosher fowl, usually a goose or turkey, the latter being preferred because its points last longer. The ink, which must be pitch black, is prepared in spoonful quantities, so that it is fresh when used, from mixtures of gallnut or tannic acid, gum arabic, copper sulfate, and water. (In ancient times it was made of soot, wax, gum from certain trees, honey, gallnut, and water.)

Before starting his day’s work, the sofer intones the blessing: "I am hereby writing this Sefer Torah for the purpose of proclaiming the sanctity of the Sefer Torah." He will, if at all possible, complete his daily stint with a joyous and optimistic verse.

The sofer usually scribes twenty-two uninked horizontal lines and two vertical lines for each column of text. A Torah written in the past century or two will more often than not contain 248 columns, with each skin being made up of from three to eight columns. Each line of text is written so that its letters are suspended from the scribed horizontal lines. Generally, the top margin is spaced about three inches high, while the bottom spacing ranges from four to five inches.

As noted above in Part 1, Chapter 2, there are 5,845 verses in a Torah scroll, these being comprised of 79,976 words, which are in turn made up of a total of 304,805 letters.

The sofer tests his quill by writing the word "Amalek," the name of the traditionally hated foe that had wantonly attacked the Israelites as they made their way across the desert, and whose very name had become anathema to them. The reason for choosing this particular term for the testing of the quill stems from the command in Deut.. 25:19, "Timkhah es zakhar Amalek" — "Thou shalt blot out the remembrance of Amalek." The sofer is now in a position to test his quill and at the same time comply with the scriptural injunction. Once he is satisfied that the quill is in good condition, he crosses out the word with one bold stroke and vigorously scrapes both the stroke and the offending name from the parchment.

Not even a single character of the Torah may be written from memory. Every letter must instead by copied directly and carefully from a Torah scroll that is known to be letter perfect or from an unquestionably accurate Tikkun Sofer (a soferic manual). The scribe is required to read each word in the master text aloud before writing it down. When the text calls for the insertion of the Divine Name, he is obliged to raise his quill before writing the last letter of the word immediately preceding the Name and inspect the tip of the quill, making sure that it is not defective. He then recites the traditional blesssing, "I am now writng this Holy Name in order to sanctify the Divine Name," and without pause writes the Name itself.

Since there are some 2,000 occurrences of the six versions of the Divine Name in the Torah, it is obvious that the opportunities for error in the writing of these Names are quite high. Should the scribe make such an error, he is not permitted to erase and correct it (as he may do with other words). Instead, the damaged skin is consigned to a genizah for storage, and its contents are rewritten on a blank, pristine skin. (Erasures are permitted, however, when the god referred to in the Torah is a pagan divinity.)

The absolute importance that is ascribed to the work of a sofer engaged in writing a Torah may be gauged from the stipulation in the ancient treatise Sefer Torah which forbids a scribe who is in the midst of writing the Holy Name from returning the greeting of a passerby, even if that passerby happens to be a king. For that matter, every king of Israel was not only expected to write a Torah scroll upon his accession to the throne (as commanded in Deut. 19:18-20), but also to study its text assiduously each day. Moreover, in order to ascertain that the royal scroll was free of error, it was proofread and corrected by each of the seventy-two learned men who constituted the Bet Din ha-Gadol (the High Court).

The need for exercising all possible precautions to keep even the slightest copying error from creeping into a Torah scroll finds frequent expression in the literature of Judaism. In the Babylonian Talmud, Tractate Erubin (fol. 13a), for example, we are told of a brief conversation between two renowned scholars of the second century, the gist of which is that the responsibility the scribe assumes when he writes a Torah scroll is an awesome one. "My son," Rabbi Ishmael inquires of Rabbi Meir, "what is your occupation?" "I am a scribe," the younger man replies. "Be meticulous in your work, then," Rabbi Ishmael cautions him, "for it is heavenly work. Should you happen to omit or add even one letter [in the writing of a Sefer Torah] you could destroy the entire universe thereby."

What Rabbi Ishmael is suggesting here is that under certain con- ditions a scribal error involving as little as a single letter of the Pentateuch may become a vehicle for introducing blasphemy into the sacred text, an offense which by its very nature undermines the foundations of the universe. Two illustrations of the kind of error Ishmael may have had in mind are offered by Epstein in a gloss to the passage in Erubin which discusses the conversation between the talmudic sages. If a scribe were accidentally to omit the letter aleph from the Hebrew word for "truth" (,nt, aleph-mem-tov), Epstein points out, the meaning of that word would be distorted to "dead" (,n, mem-tov); and if he were to add a vav to the word for "and He spoke" (Rcshu, vav-yod-dalet-bet-resh) that word would become "and they spoke" (urcshu, vav-yod-dalet-bet-resh-vav). And, as Epstein then explains, "when such terms are applied to the Deity the scribe is, in the latter case, guilty of acknowledging polytheism, while in the former he denies the Living God." The indignant outburst of Schiller-Szinessy over the plethora of scribal errors which he found in Scroll 4 at Cambridge becomes readily understandable when viewed in the context of Rabbi Ishmael's dire warning against adding or deleting even a single letter of the Pentateuch.

Once a scribe has finished writing a Torah scroll it becomes the direct responsibility of the congregation which acquires it to correct any discrepancies that may subsequently be discovered in that scroll. Presumably, therefore, a much used Sefer Torah can tell us something about the level of Judaic knowledge of the congregation in whose synagogue it was read.

Given the critical importance that Judaism has always ascribed to the writing of Torah scrolls, one can understand the proud claim that has been made by Jewish scholars that no other document of substance has been handed down throughout the ages that is as free from error as a Torah scroll.

The Kaifeng exemplars, however, do not fall in this category, though only insofar as their orthography and their scribal styles are concerned, and not at all in their wording. As has been previously pointed out, the numerous spelling and stylistic peculiarities we note in their texts make it abundantly clear that the isolated community which created these was not nearly as familiar with either the Heb- rew language or the rules of Torah writing as were its sister communities in other parts of the world.

While it is true that there are several slightly differing and equally valid approaches to the making of a Torah scroll, these have to do with form only — that is to say, with calligraphic variants of the Ktav Ashuri, physical format, the treatment of the skins being employed, and so forth — but never with substance. It is well known, of course, that a number of problems regarding specific points in the texts of the scriptures remain unresolved, among them the vexing question of determining the precise spelling of certain words. When he comes across problems of this nature as he writes a Torah scroll, a sofer will normally handle them according to the custom prevailing in his community. In short, the wording of the text remains the same — regardless of who writes it, where it is written, or when it is written — but the manner in which various details relating to form and orthography are treated may be somewhat different from community to community and from generation to generation. Accordingly, when we discover a deviation from the norm in a Kaifeng Sefer Torah — in, let us say, the positioning of words within a particular passage — it behooves us to ask whether this merely represents an error on the part of the scribe or whether it was intentionally written in this manner. The question which then arises is: if the latter is the case, how, when, and from where did the tradition guiding the the writing of that Torah come to Kaifeng?

The balance of this study is devoted to the examination of a sampling of the orthographic, calligraphic and stylistic idiosyncrasies that mark the individual Kaifeng Sifrei Torah and to the discussion of the problems and conclusions that these raise. It will be demonstrated that some of the textual and stylistic minutiae that have been selected for analysis herein are carried through in all the Kaifeng Torahs I have seen. It will also be shown that others are not treated uniformly throughout all these exemplars, a fact which raises the possibility that the Jews of Kaifeng could have received their tradition, at least insofar as the preparation of Torah scrolls was concerned, from more than one source. And this may in turn suggest that the ancestors of these people came to Kaifeng (as well as, presumably, to the several other cities of China in which Jewish settlements are known to have existed) from more than one geographic area and, in some instances, perhaps in separate waves.

It must be emphasized, however, that the discussion which follows is intended to be no more than a preliminary overview of the problems posed by the Kaifeng Sifrei Torah. A definitive in-vestigation of these scrolls, one which will not only analyze them in much greater depth but will also compare them with earlier Sifrei Torah from Jewish communities outside of China, must await the attention of scholars whose specialties lie in the fields of Masoretic tradition and soferic law. Moreover, since a number of the smaller manuscripts, primarily the section books (parashiot), which were acquired by the Delegates from the Jews of Kaifeng in 1851 and are currently held at the Hebrew Union College and elsewhere are scriptural in nature, it may prove fruitful to compare the peculiarities which they exhibit with those of the extant Kaifeng Sifrei Torah, as well as with those of early Pentateuchs emanating from non-Chinese Jewish sources. The same, obviously, holds true with respect to the surviving Kaifeng Haggadahs and liturgical texts.