“And you shall remember all of the way (i.e. road) that Hashem your G-d made you travel these forty years in the desert, in order to afflict you, to test you, to know what is in your heart, whether you will keep His commandments or not” (Devarim 8:2). The above translation follows the “k'ri,” the wording as it read; had I translated it according to the way it is written instead, the “k'siv,” there would be one slight difference: Instead of being His “commandments” (plural) that we are supposed to follow, it would be His “commandment” (singular), as the third to last word in the verse has no letter “yud,” but we “pretend” that there are six letters in the word instead of five, adding a “yud” before the last letter.
I could try suggesting a reason for this discrepancy. For example, even though the test is whether we will keep all of G-d's commandments (not just one), hence the added “yud,” since Rashi tells us that the crux of the test is whether we will do so without complaining, it is in essence this one commandment (not to complain) that is the focus of the verse. In order to teach us that we are being commanded (singular commandment) not to complain while keeping all of the commandments (plural), the Torah embedded both the singular and the plural forms into the same word.
However, rather than taking a closer look, this week I'd like to take a broader look, and discuss the very nature of “k'ri/k'siv.” Why are there words in the Torah (and throughout Tanach) that are written one way but are supposed to be read a different way? There are even instances where there is no word at all yet our tradition says to pretend a specific word is there and read the verse as if it contains that word, and instances where the text contains a word that our tradition says to skip when reading the verse. Why is this so, and how did these “discrepancies” get there?
As I alluded to above, it can be suggested that each version (the version we read and the version in the text) has something to add. The Malbim (Ayeles HaShachar #247) says that the way things are read (the “k'ri”) represents the plain meaning of the word/verse while the way it is written (the “k'siv”) represents its exegetical meaning. The Vilna Gaon (in his commentary to Mishlay 16:19) says that the “k'ri” usually reflects the “revealed” meaning while the “k'siv” reflects its “hidden” meaning. This approach can be applied to narrative as well. For example, Sefer Devarim is primarily the words Moshe spoke to the nation shortly before his death. When Moshe said these words, he either said commandment (singular) or commandments (plural), but could not have said both. Nevertheless, when G-d told Moshe to write down the words he (Moshe) spoke to the nation and include them in the Torah, He (G-d) could easily have told him to write the word in its singular form but pronounce it as if it's written in the plural form. And this is what the Talmud (Nedarim 37b) seems to say happened, as “[words] read but not written and [words] written but not read are “halacha l'Moshe mi'Sinai,” laws taught to Moshe at Sinai.
However, this is problematic, as these words were said by Moshe at Arvos Moav, 38+ years after the nation left Sinai; how could G-d tell Moshe at Sinai to write them down if they weren't said yet? (This issue applies to all of the narrative that occurred after they left Mt. Sinai.) But even if we want to “translate” the word “Sinai” liberally, and include any direct communication between G-d and Moshe (since there is no practical difference whether this communication took place atop Mt. Sinai, in the Mishkan while at Sinai, in the Mishkan at any of the stops along way, or in the Mishkan at Arvos Moav), calling it “halacha l'Moshe mi'Sinai” is still problematic. The examples the Talmud gives are from the books of Shemuel, Yirmiyah and Rus, all of which were written well after Moshe died! How could any “k'ri/k'siv” contained in them have been taught to Moshe?
Ben Yehoyada drives this point home (without discussing it) by saying that there was never a change in the text, as all of it, including how it should be written and how it should be read, was said to Moshe at Sinai, “and so it was with the verses of Nevi'im (Prophets) and Kesuvim (Writings); this was how it was revealed to us (i.e. with both the “k'ri” and the “k'siv”) by the prophets that wrote them.” How this qualifies as “halacha l'Moshe mi'Sinai” is not explained.
The Rosh (Hilchos Mikva'os 1, after Meseches Nidah) explains “halacha l'Moshe mi'Sinai” in a very non-literal way, understanding it to mean a law that is as clear (uncontested) as a law told directly to Moshe at Sinai. In this context, when the Talmud uses the expression “halacha l'Moshe mi'Sinai” regarding how the text of Tanach is written, it would mean that these aspects (the three categories included by the Talmud) are not later additions, but were part of the original text (and how it should be read). Nevertheless, there is much written to explain what “k'ri” and “k'siv” represent, including how those in Nach (Nevi'im/Kesuvim) can be classified as “halacha l'Moshe mi'Sinai.”
Radak (in his introduction to Sefer Yehoshua and on Shemuel II 15:21), gives his take on “k'ri” and “k'siv:” “It seems [to me] that during the first exile (after the first Temple was destroyed) the books (containing the Biblical text) were lost and moved, and the sages that knew the text died. And the Men of the Great Assembly, who returned the Torah to its original state, found discrepancies in the texts that were available, and (when reconstructing the authoritative text) followed (for each discrepancy) the majority, based on their opinion/knowledge. And in place[s] that their knowledge/opinion did not reach a point of full clarity (i.e. they were not completely confident that they had restored the text to its original form), they wrote one [version] but did not vocalize it, or they wrote it outside (the margins) but not inside (the body of the text), or wrote one [version] outside and one inside.” Radak is not the only one to suggest this, although it seems that the others who do so are following his lead (i.e. Meiri, in his introduction to Kiryas Sefer, who quotes Radak's language).
We can not apply the Rosh's understanding of “halacha l'Moshe mi'Sinai” to the Radak's explanation, and it would be difficult to suggest that Radak is disagreeing with the Talmud's understanding of the origin of “k'ri” and “k'siv” (especially since Meiri repeats it in his explanation of the Talmud, which also precludes the possibility that “k'ri” and “k'siv” being “halacha l'Moshe mi'Sinai” is only a minority opinion). I would therefore suggest something similar to what I wrote regarding the Talmud's assertion that Esther was given at Sinai and that the Aggadic literature was given at Sinai (see http://rabbidmk.posterous.com/shuvuos-5770); Moshe was taught the system to reconstruct the Biblical text (when necessary) at Sinai, including what to do if unsure exactly what the original text stated.
Abarbanel (in his introduction to Sefer Yirmiyah) rejects Radak's approach, for several reasons. For one thing, if a Sefer Torah is missing even one letter it is invalid, and if “k'ri” and “k'siv” are based on discrepancies that were unresolved, all of our Sifri Torah would likely be invalid. I'm not sure why this would be so; as long as we follow the proscribed procedure for writing/fixing the text, the halacha (law) would be that it is a valid Sefer Torah.
Abarbanel's next issue with Radak's approach is that there is a “guarantee” that the Torah would always be with us, which Abarbanel understands to mean with no doubts about what the original text was. Included in this concept (Abarbanel continues) is the comforting thought of the consistency of the Torah (its text) no matter how harsh the trials and tribulations we are forced to withstand are (keep in mind that Abarbanel had been a royal officer and had to flee Spain during the Inquisition, an upheaval that affected him greatly, including his having to rewrite much of his commentary due to the loss of his original manuscripts). This is why (according to Abarbanel) one of the fundamentals of our faith (referring to Rambam's 8th principal) is that the text we have now is the exact same text that was given to Moshe. A close reading of Rambam's 8th principal, however (see Igros Moshe, Yoreh Daya 3:114), indicates that Rambam did not mean (or at least focus on) whether the text we have is the exact same text given to Moshe, but whether any part of the text was written (or added) by Moshe on his own, without having been dictated to him (or to Yehoshua, according to the opinion that the last eight verses of the Torah were written down by Moshe, see Bava Basra 15a). Rambam does not mean that no textual doubt will ever arise, but that nothing in the text was purposely added; it all came directly from G-d. Reconstructing the text is not the issue, since the reconstruction is an attempt to determine what G-d had dictated, not a suggestion that the original came from another source. (Whether anyone disagrees with Rambam regarding this being a fundamental is a separate issue; I am only trying to clarify Rambam's opinion, since this is what Abarbanel is basing himself on and what the traditional community has accepted.)
The notion that we will always have the exact text that was dictated to Moshe is difficult to sustain. Aside from Radak (et al), Chikuni (Beraishis 15:5) says that the dots over certain letters or words in the Torah were added by Ezra to indicate which letters he was unsure were in Moshe's original text. (Although some attribute Avos d”Rav Nosson 34:4 as being Chizkuni's source, the context makes it clear that Avos d'Rav Nosson understands the dots to be a means of conveying additional messages from the text, not an indication of not being sure whether that part of the text really belonged.) Conceptually, the Chizkuni is very similar to Radak; Chizkuni uses the concept to explain the dots while Radak uses it to explain “k'ri” and “k'siv.” In addition, there are numerous places where the text used by the Talmud and Midrashim does not exactly match our text (see Ran on Sanhedrin 4a). Rashba (Responsa Attributed to Ramban 232) acknowledges that there are differences between the sages that excelled at exegesis and those that excelled at maintaining the text, and between “the sages in the east and the sages in the west” (which I understand to mean those in Bavel and those in Israel) regarding what the text actually is, making it impossible to say that we will always know what the exact text was. [The differences in the text are very few, and very minor, and almost always have no practical difference. In almost all cases, the exegesis involved did affects Jewish law, only Aggadic interpretation. Since the point of these interpretations is the message they carry, and the message was carried via Chazal, we can trust the message to be a valid one. The one instance where the difference between the texts affects Jewish law (and is only relevant to a rare circumstance, one that may never have occurred, as part of the service in the Temple in Jerusalem), is the subject of Rashba's Responsa. Conceptually, this is not really in issue either, and can be compared to different communities having different customs, or following the decision of their local Poskim (decisors of Jewish law); when it comes to maintaining the text, we follow the decisions of those whose expertise is in maintaining the text, and when it come to laws deduced through exegesis, we follow the decision of those who are experts in that area. Besides, at the time this “difference” could have been relevant, there was a Sanhedrin available to make the final determination.] Chazal (Meseches Sofrim 6:4, Yerushalmi Taanis 4:2, Sifri on Devarim 33:27) tell us that there were three Sifray Torah found in the Temple that had (slightly) different texts, and (in order to reconstruct the correct text) they used the concept of “majority rules” (Shemos 23:2), with the final text not being exactly the same as any of the three they had started with. This is not the same as Radak or Chizkuni, as once the text was reconstructed there was no indication included in the finished product that there had ever been a doubt, but it does show that discrepancies can arise, and at least until the reconstruction process is completed there could be doubts about exactly what Moshe's original text was.
It is unfair to mandate to others what should provide comfort, and practically speaking it is impossible. Nevertheless, I don't take any less comfort in knowing that we have a system in place (with divine approval) to reconstruct the text than if we had a guarantee that we would never have any doubts about what Moshe's text was, exactly, to the letter.
Another issue Abarbanel has with Radak's approach is determining which version to include in the body of the text (the “k'siv”) and which to put in the margin or the vocalization (the “k'ri”). It would seem, though, that even if Ezra and the Men of the Great Assembly weren't confident enough in one version to totally disregard the other, they could still choose which version they thought was more likely to be correct. In any case, what their “doubt” was needs to be explored, as if there was a majority of manuscripts that indicated one way, why didn't they just go with the majority (as stated in Meseches Sofrim, et al)? Did Radak think that there were exactly the same number of manuscripts for both possibilities for every “k'ri” and “k'siv?” That is rather unlikely. Quite possibly, there were reasons why they thought certain manuscripts were less reliable, and even though, when all the manuscripts were counted, there was a majority indicating one version over the other, since these less-reliable manuscripts created the “majority,” they didn't want to disregard the version contained in the reliable manuscripts. Whatever the situation was, it does not seem strange that they would prefer one version over the other, and indicate as much by making one the “k'siv” and the other the “k'ri.” [It should be noted that this discussion is not necessarily affected by the Talmud's discussion (Sanhedrin 4a) regarding whether the way things are written is primary (“yesh aim l'moseres”) or the way they are read is (“yesh aim l'mikra”), as in those cases the differences are not between the “k'ri” and the “k'siv,” but between how it could be written based on how we read it vs. how it could be read based on how it is written.]
The final issue Abarbanel challenges Radak with is the strongest. Many of the differences between the “k'ri” and the “k'siv” occur numerous times with the same word (although, ironically enough considering Abarbanel's take on the possibility of differences in the text arising, if you compare his examples with our text, they don't add up). If the “k'ri” vs. “k'siv” differences are the result of copyist errors, how likely is it that the same exact error happened almost every time a word was written. (Even though one version of the Chazal relating the three Sifray Torah found in the Temple has the third discrepancy occurring numerous times, it seems more likely that this is an inaccurate version of the Chazal rather than the same error being made so often.)
After rejecting Radak's approach, Abarbanel reiterates that the “k'siv” was never in doubt, and says that Ezra added a “commentary” to those words that would not have been understood properly (usually because the way it was written has hidden meanings) in the form of a way to read the word that brings out its simple, straightforward meaning. (This is consistent with the Malbim and Gra quoted above.) Although Abarbanel does not directly address how the Talmud can say this was “halacha l'Moshe mi'Sinai,” he does say that Ezra being able to explain the text this way was “without a doubt received from the prophets and the sages of the generation that preceded him.” If we take it back further, and say that the authorization to clarify the text this way was given to Moshe at Sinai, the approach of the Abarbanel can fit with the Talmud as well.
Maharal (Tiferes Yisroel 66) takes very strong issue with both Radak and Abarbanel's approaches, with one of the main reasons being that they go against Chazal telling us that “k'ri” and “k'siv” are “halacha l'Moshe mi'Sinai.” (As we have discussed, this is not necessarily the case.) According to Maharal, the text as written contains the deep wisdom appropriate for divine works, while the way it is read reflects its plain meaning. Maharal does address how “k'ri” and “k'siv” can be “halacha l'Moshe mi'Sinai” if they appear throughout Tanach, with the “halacha” being to hint to the deeper meaning in the written text but to read it in a way that the plain meaning is understood.
That the “halacha l'Moshe mi'Sinai” is the system of using “k'ri” and “k'siv” rather than how each individual word should be written/read is evident from the Talmud itself. Aside from the fact that the examples given are from books written after Moshe had died, one of the other categories said to be “halacha l'Moshe mi'Sinai” is reading “eretz” as “uretz” (changing the vowel under the first letter from a “segol” to a “kumatz”) when it is the last word of a verse (or verse segment). This occurs for proper names as well (such as “Lemech” becoming “Lumech,” see Beraishis 4:18 and 5:25). If the Talmud is saying that Moshe was taught how to pronounce the specific word (“eretz/uretz”), it would be no different than his being taught how to pronounce every other word in the Torah as G-d dictated it to him (see Ritz, quoted by Shita Mekubetzes). If however, the Talmud means that Moshe was taught the rule of how to vocalize these types of words in these types of situations (the “rule” rather than each word by itself), we can understand which “halacha” Moshe was taught independent of being taught the entire text. And just as the “halacha” regarding “mikre sofrim” refers to the rule, not the specific words the rule applies to, the “halacha” regarding “k'ri/k'siv” refers to the rule, not how to write/read each specific word.
No comments:
Post a Comment