We do something very unusual in Kiddush Levana. After reciting the first half of Exodus 15:16 three times, “tipol aleihem eimatah va-fachad, bigdol zeroacha yidmu ka-aven,” we recite these same words backward, three times: “ka-aven yidmu…aleihem tipol.” We are never else instructed to recite a verse or section of a verse backward in our prayers. What is going on here? An article by David Farkas published several years ago in the Journal Ḥakirah (vol. 7) explores this custom and I would like to share it.
A basic text of Kiddush Levana is found in the Talmud at Sanhedrin 42a. But over the centuries, many additional customs were added to what is found in the Talmud. That is what we are dealing with here.
Aside from the backward recitation, our recital of Exodus 15:16 is also unusual because we are reciting only a part of the verse. The balance of the verse is “ad yaavor amcha YHVH, ad yaavor am zu kanita.” There is a statement in the Talmud (Megillah 22a) that implies that we should recite only full verses.
Farkas observes that our present practice (first half of the verse—forward three times, then backward three times) has its origin in the Tur (d. 1340), sec. 426, and is then codified in the Rama. But earlier than the Tur, he can only find a reference to the practice in the Sefer Ha-Rokeach of R. Eleazar of Worms (d. 1238), sec. 229.
Interestingly, R. Eleazar of Worms instructs the recital of the entire verse, forward and backward, three times. (That the entire verse should be recited is also the view of Magen Avraham, who cites R. Eleazar.)
But going back to our fundamental question, where did R. Eleazar of Worms get the idea that any part of Ex. 15:16 should be recited forward and backward? Farkas directs our attention to Masechet Soferim. (According to the most recent scholarship, this work was completed in the ninth or 10th century.) The relevant passage in Masechet Soferim has several instructions regarding Kiddush Levana that are not found in the Talmud but that became part of our ritual. It includes the following instruction: “ve-omer shelosh pe’amim: 1) ke-shem she-ani roked….2) tipol aleihem eimatah va-fachad, u-le-mafrea, amen amen amen selah halleluyah.” (This passage is in the 19th chapter of M. Higger’s scholarly edition of this work, but is in the 20th chapter of the standard printed edition.)
Farkas suggests that the Tur was just passing down a tradition that originated with R. Eleazar of Worms, and that R. Eleazar of Worms misinterpreted the meaning of “u-le-mafrea” in Masechet Soferim.
Farkas writes: “It seems strange that Maseches Soferim itself would suggest reading a verse backwards…What might not be out of place in a kabbalistic or chassidic text seems decidedly out of place in Maseches Soferim. This work…is generally a sober halachic text governing the laws of scribes, Torah reading and liturgy. Although some aggadic passages are included in the work, nowhere else in Maseches Soferim do we find anything remotely resembling a suggestion to read a verse out of order.”
Farkas points out that R. Eleazar of Worms was a figure within chasidei Ashkenaz in Germany, a group that often gave mystical interpretations. Here it seems that R. Eleazar gave an unnecessarily mystical interpretation to the word “u-le-mafrea.”
Farkas admits that the hardest part of his task, after rejecting the interpretation of R. Eleazar of Worms, was to determine the correct interpretation of “u-le-mafrea” here. He observes that when the other Rishonim discuss Kiddush Levana, they either quote the relevant passage from Masechet Soferim without explaining how they understood the word “u-le-mafrea,” or they omit the relevant passage altogether.
Farkas offers a few solutions. First he cites the English translation of the passage in the Soncino edition of Masechet Soferim: “Let terror and dread fall upon them and may this be retrospective.” He suggests that the translator was expressing the idea that the punishment on the enemies of Israel, according to this passage, should also include retribution for prior attacks. (He also suggests creatively that since Kiddush Levana can be recited up to approximately the middle of the month, we might be asking for retrospective application of the punishment to date back to the first of the month!)
Then he suggests a more likely and entirely different approach. “U-lemafrea” really belongs to the subsequent phrase. The import of “u-le-mafrea” is merely “as to what has previously been recited,” add “amen, amen, amen, selah, halleluyah.” Why was the word “u-le-mafrea” needed here at all? Without it we might have misunderstood and thought that the phrase “amen, amen, amen, selah, halleluyah” should also be recited three times. Or the “u-le-mafrea” clarifies that these “amens” can be said “retrospectively” on the main earlier blessing of Kiddush Levana that was interrupted with the various three-time statements.
Of course, ordinarily we do not want people today to be reinterpreting texts and possibly changing practices. But this unusual custom cried out for re-analysis! (Farkas is an attorney who received rabbinical ordination from Ner Israel Rabbinical College in 1999.)
- The author mentioned that there were no other examples of reading verses backward in our prayers. But he does point out that in the bedtime Kriyat Shema there is a three-word verse from Gen. 49:18, “liyshuat’cha kiviti Hashem,” that is supposed to be read in various permutations. But this is a slightly different concept. (It surely has its origin in some mystical source later than Masechet Soferim.)
- One issue I did not clarify above is precisely how the recital is done three times. The Complete ArtScroll Siddur (p. 614) instructs us to say the “tipol” statement three times, and then the “ka-even” statement three times. But the Tur had implied that we first recite both the forward and backward verses, and then repeat this three times. (Similar was R. Eleazar of Worms.) This is what is done in nusach ha-Gra today.
- The next article in the same issue of Ḥakirah takes on a different issue involving Kiddush Levana: the recital of “shalom aleichem” to three different individuals. Masechet Soferim had instructed: “ve-omer le-chaveiro shalosh pe’amim ‘shalom.’” How did one friend become three friends, and how did what we say evolve into “shalom aleichem,” a greeting to each friend in the plural? I refer you to this article by Zvi Ron. (Articles in Ḥakirah are available online at hakirah.org. Briefly, Ron suggests that it was the influence of foreign languages, where the plural is often used as a sign of respect, that led to the shift to the plural “aleichem” from the earlier practices of “shalom” and “shalom alecha.”)
Why is there any kind of greeting to a friend or friends at all in Kiddush Levana? Masechet Soferim had not provided any explanation. I will mention two suggestions. Maharil (14th cent.) writes that since Kiddush Levana “is such a great mitzvah, and is considered like greeting the Shechinah, it is appropriate to greet one another out of joy and good feeling.” Another explanation (brought down in Mateh Moshe, 16th century, but found earlier as well) is that after wishing downfall on our enemies, it is appropriate to turn to our friend and say, “Not on you. To you, only peace and peace.”
By Mitchell First
Mitchell First is a personal injury attorney and Jewish history scholar. He can be reached at [email protected] He can relate to the concept of le-mafrea/backwardness since his last name is “First.”
For more articles by Mitchell First, and information on his books, please visit his website at rootsandrituals.org.