vekhatav lo et mishneh hatorah hazot (Deut 17:18) ('and he will write the mishneh of this Torah') The Sifre comments: Is it only mishneh Torah [the second version of the previous books of the Torah (Malbim), Deuteronomy - pointing out that mishneh is related to shenayim 'two' - that he writes], how [do we know that he writes] the rest of the Torah? We learn [in the following verse] 'to keep and to do all the matters of this Torah.' If so why does it state mishneh Torah? [The answer is] because it is destined lehishtanot ('to change') [pointing out that mishneh and lehishtanot are of the same root, that is: Shin, Nun, Heh]. Others say [it is because] on the day of hak'hel ('gathering') we read only mishneh Torah ('Deuteronomy'). The Gemara learns that mishneh hatorah itself indicates two copies of the Torah (Sanhedrin, 21b) without referring to the following verse.

Rashi writes: 'two sifrei ('scrolls of') Torah … and Onkelos translated patshegen ('repetition, copy, abstract' - Jastrow) understanding mishneh as being related to shinun and dibur ('sharp learning' and 'speech').' It would seem that Rashi first follows the Gemara and then points out that Onkelos understands mishneh differently.

Rashbam writes: My grandfather [Rashi] explained … it to mean two sifrei Torah and that its Targum is patshegen. This is a noncommittal reference. He then proceeds to disagree with Rashi's final point and says 'but it is not related to veshinantam (Deut 6:7) ('and you shall teach them').' He is saying that Rashi's suggestion that mishneh is related to shinun is not acceptable. This is because Rashbam had accepted the three-letter root theory, according to which the relationship between mishneh (root: Shin, Nun, Heh) and shinun (root: Shin, Nun, Nun) is not possible. Such a relationship is only possible to Rashi because he was an adherent of the two-letter root theory which would give the root here as Shin, Nun. Indeed Menachem ben Saruk in his dictionary Machberet, entry Shin, Nun, lists 'sharpening' as the second meaning, 'teaching' as the fifth, and 'two' or 'doubling' as the eighth. As in many cases, Rashi here is following Menachem.

R' A. ibn Ezra says simply 'a second version.' As both Rashbam and R' A. ibn Ezra were supporters of the three-letter root theory they saw the root as Shin, Nun, Heh, and no other meaning was acceptable to them.

The Dagesh in the Bet and related matters

shivti bevet H' (Psalms 27:4) ('that I remain in the house of Hashem') There is a rule widely reported by the traditional grammarians which has rhyme and is in a style reminiscent of the Mesora (the notes by the ancient scribes aimed at maintaining the text of the Bible, the rule is sometimes reported as being part of the Mesora, though its place is not given), stating that whenever the letters BG"D KF"T come at the beginning of a word which follows a word concluding with one of the letters Yud, Heh, Vav, Alef, [i.e. the vowel letters], the letters BG"D KF"T are soft (without a Dagesh); excepting 1) when the letters Yud, Heh, Vav, Alef, are pronounced as consonants; 2) if they are followed by a pause in the sentence; 3) if closely pressed; and 4) if the tunes are far apart (If you don't understand the last two you are in good company. They are thoroughly discussed by Prof. A. Dotan in" levayat dachik ve'atei meirachik," Proceedings of the World Congress of Jewish Studies 4 vol. 2 Jerusalem, pp. 101-105, 1969).

In our case the tune on the word shivti is a Munach - a service tune - not a pause (see R' S.Y. Weinfeld, Taamei HaMikra, Jerusalem, 1981, table at back). Clearly the first two exceptions do not apply. According to the traditional interpretation of the last two, they do not apply either. Rabbi Sh. Divlitzky points out that R' Zalman Hanau (Raza"h) reinterpreted the rule so that all the cases previously regarded as belonging to exceptions 3) & 4) came to belong to exception 3) and exception 4) was left empty for another exception which had previously stood alone without a name and outside the rule. This was the exception which says that when the letters Yud, Heh, Vav, Alef at the end of a word are followed by a word beginning with BG"D KF"T with a sheva and the same letter (or perhaps another letter of BG"D KF"T) follows it, it does take a Dagesh, and this explains bevet. Previously this rule, though known, had no home to rest in. By amalgamating exceptions 3) & 4) into exception 3), Raza"h created space into which he placed the rule about BG"D KF"T with a sheva and the same letter following it and this became exception 4)! Rabbi Divlitzky compares this to a certain grocer who had two similar varieties of beans in two bags which were partly full, and a third variety of beans which had no container - what did he do? He poured the two kinds of beans whose differences were so fine that no one would notice the differences into one bag, and put the beans which had no container into the empty bag. (R' Sh. Divlitzky, Chok yehu umvatlav Benei Brak, 1982, p.16). (Thank G-d, in this humdrum generation, we have a rabbi and grammarian with a sense of humor.) Whether or not the exception about words starting with BG"D KF"T with a sheva and the same letter following it belongs to the well known rule about BG"D KF"T following Yud, Heh, Vav, Alef at the end of a word, it is a known exception and explains the Dagesh in the Bet of shivti bevet H'.