Six days work shall be done and the seventh day shall be holy, a Shabbat to Hashem.... Take from among you a tribute to Hashem; all who are benevolent of heart shall bring a tribute [in order to build the Mishkan] for Hashem....

(Shemot 35:2,5)

The Mishnah (Shabbat 73a) lists a total of 39 categories of Melachah that are prohibited on Shabbat for this reason. According to the Gemara, the number 39 has a Biblical source as well:

[The number of prohibited Melachot on Shabbat] corresponds to the number of times that the words "Melachah," "Melachto," and "Melechet" appear in the Torah, which is a total of 39 times.

(Gemara Shabbat 49b)

Rebbi Yosi wondered, "Is the verse 'And Yosef entered the house [of Potifar] to tend to his Melachah (Bereishit 39:11)' included in the count... [or is the verse] 'The Melachah was sufficient [for the purposes of the Mishkan] (Shemot 36:7)' included in the count?"

(Gemara ibid.)

This source for the 39 prohibited categories can be found in the Talmud Yerushalmi (7:2; p. 44a) as well [1]. The Yerushalmi adds that the verse (Devarim 16:8) "Six days you shall eat Matzot, and the seventh [day of Pesach] shall be a holiday; do not perform any Melachah" -- is meant to complete the full 39 appearances of Melachah; it is the last time that the word Melachah is mentioned in the Torah. All appears to be neat and trim.

Actually, the Tosfos Yom Tov was not the first to note this difficulty. Well before the publication of the Concordance, one of the earliest Talmud commentators, RABEINU CHANANEL [3], dealt with this apparent inconsistency. He suggested the following solution:

"There are actually 61 places where the word Melachah is used. (1) Subtract from them the 3 times "Melachah" is mentioned with references to Hashem's creation of the world [4] (because Hashem simply spoke and created the world; no *labor* was involved -Ra'avan #350); (2) and the 4 of which it is written "You *should* do work" or "Work *should* be done" (during six days of the week -- these are not included since the verse is enjoining us to *perform* work, and not to refrain from it -Ra'avan); (3) and the verse "l'Regel ha'Melachah" [5] (because the word Melachah there means "possessions," and not "labor." (4) and the 13 [7] times that it says "You shall not perform Melechet Avodah" with regard to Yom Tov (since that does not refer to the broader category of labor that is prohibited on Shabbat, but to the more limited set of "laborious tasks" -- Melechet Avodah -- that are prohibited on Yom Tov because they do not usually provide immediate benefit.)

(Rabeinu Chananel, Shabbat 49b)

Rabeinu Chananel's words certainly are convincing. After all, it is clear from the Gemara that we only count the Melachah's that refer to labor of the type prohibited on Shabbos, and not to "business" or the like. Secondly, according to Rabeinu Chananel it is clear why the Gemara only mentioned "Melachah, Meleches, and Melachto," but made no mention of the *fourth* suffix for Melachah that can be found in the Torah. Both places in which the suffix "Melach*techa*" appear [6] are with reference to *performing* work during the other six days of the week. They therefore are not included in the count (Rule (2)).

However, a careful look at the Concordance shows that there are a total of *65* times that Melachah appears in the Torah. Since Rabeinu Chananel only succeeded in lopping off 21 from the total, we are still left with 4 Melachah's too many! We must look further to find a solution.

(a) TOSFOS YOM TOV suggests we are not to count any of the places where the word Melachah is mentioned in the context of prohibiting Melachah or of administering punishment for profaning the Shabbat. It is from the *other* times that Melachah appears that we learn the meaning of the word in the context of Shabbat and Yom Tov, therefore we only count the other times to find a total of 39. (The Vilna Gaon, in his commentary to Mishnayot Shabbat 7:2, suggests a similar approach.)

(b) The MAR'EH HA'PANIM (Yerushalmi 7:2) explains that the count does not include the appearances of Melachah with the connective suffix, "Melechet." Although his count actually turns out to be mistaken (there are actually 23 such appearances, as pointed out by Amudei Yerushalayim, Yerushalmi ibid.), another commentator, NETZACH YISRAEL (Harav Yisrael Segal, 1741) polishes up this suggestion. We are to ignore suffixes of *both* Melechet, and *Melachtecha*, to get a total of 40 Melachot. (It is not clear what logic would justify limiting the count in this manner.)

(c) HARAV MENACHEM KASHER (Torah Sheleimah Bereishit 39:96) suggests that we must only count the times that the word "Kol" ("all") is associated with the word Melachah. This leaves exactly 39 times [8].

However, none of these seem to be viable solutions. The solution of the Mar'eh ha'Panim (b) is contradicted by the Gemara in Shabbos, which specifically says that the word "Melechet" *is* included in the count. (To justify his solution, he emends the text of the Gemara!) As for the others, the Yefeh Einayim points out that, as we mentioned, the Yerushalmi includes in its count Devarim 16:8 as the last of the 39 appearances of Melachah. That verse is indeed stating a prohibition (invalidating (a), the answer of the Tosfos Yom Tov), and it does not mention the word "Kol" (invalidating (c), the answer of Rav Kasher). If so, we must still seek a working solution to our problem.

(A) Rabeinu Chananel instructed us to omit the "4" places in which it is written "You *should* do work" or "Work *should* be done." However, there are actually *5*, and not 4, places, in which this type of expression is written [10] (Yefeh Einayim; Amudei Yerushalmi)! If we subtract the fifth one as well (which Rabeinu Chananel didn't mention, for some reason) we are down to 22 omissions.

(B) We do not count times when Melachah means just "possessions." (Rule (3)). If so, it is obvious that we should not count the two times where Melachah appears in Mishpatim, "she'Lo Shalach Yado b'Melechet Re'ehu," [11] since Melechet, in this verse, clearly means possessions (see Rashi ad loc.). Rabeinu Chananel perhaps did not see the need to mention this more clearly since it is self-evident.

(C) Similarly, when the verse states that the fats of a dead animal "may be used for all Melachah," [12] it clearly is not referring to "labor." Rather, the word Melachah is being used in its broader meaning of "usage." In keeping with the Gemara's own pointer, it obviously should not be included in the count.

We have thus minimized Rabeinu Chananel's Melachah count by the requisite 4, all the time sticking to the guidelines laid forth by Rabeinu Chananel [13]! Every single one of the remaining 40 Melachah's in the Torah was either said in reference to the Mishkan or to Shabbos, or to actual labor (such as the verse about Yosef -- Potifar's slave -- and the verses in Vayikra 11:32, 13:48, 13:51).

One more important inconsistency must be resolved, though, to fully explain Rabeinu Chananel's words. So far, we have taken for granted that, as Rabeinu Chananel wrote, there are 13 instances in which the Torah refers to doing "Melechet Avodah" on Yom Tov. However, as the Yefeh Einayim and Amudei Yerushalayim point out, there are actually only *12* such verses! The remaining 2 verses, which discuss the holiday of Pesach (Shemot 12:16, Devarim 16:8), mention only not to do "Melachah," without mentioning Melechet Avodah! Where is the 13th Melechet Avodah that Rabeinu Chananel left out of the count?

The answer to this question is obvious. In Shemot 12:16, the verse indeed does not mention Melechet Avodah. Nevertheless, it is clearly referring only to the narrow set of "laborious work" which is prohibited on Yom Tov, for the Torah immediately clarifies its intention in the very same verse: "no work may be done during the 1st and 7th days of Pesach *however* that which is eaten by people *may* be prepared on Yom Tov." The other verse that discusses Melachah on Pesach (Devarim 16:8), on the other hand, carries no clear hint that it is referring to the more limited group of Melachah, that is prohibited on Yom Tov. (That is, although it is true that only the more limited Melachot are prohibited on this Yom Tov as well, nevertheless since *this verse* does not say so, its Melachah is "Parve," and it can be included in the count. This is what the Yerushalmi means by saying that the Melachah of Devarim 16:8 was deliberately written in a strange way (without mention of Melechet Avodah), in order to "complete the 39-count!"

(a) The Yerushalmi explains that the 39 categories correspond to "the 39 times that the words "Avodah" or "Melachah" are written with regard to the Mishkan." In fact, it is exactly 39 times that the words "Melachah" or "Avodah" (including any prefixes or suffixes) occur from Parashat Terumah until the end of Parashat Pekudei!

(b) Rav Menachem Kasher (Torah Sheleimah vol. 16 Milu'im #1), though, cites a Midrash (Mishnas Rebbi Eliezer ch. 20) that finds the 39 exclusively in the sections discussing the building of the Mishkan itself. One simply adds up not only the permutations of Avodah or Melachah, but also the phrases that begin with "Ma'aseh" (such as Ma'aseh Choshev; Ma'aseh Rokem).

(c) Another Midrashic anthology (Midrash ha'Gadol, Shemot 35:11-19) cited by Rav Kasher suggests a slightly different twist in finding the 39 Avodah or Melachah that were required for the Mishkan. There are 39 times, he explains, that the verse relates that Moshe was *commanded to make* ("v'Asita") an object for the Mishkan in Parashat Terumah-Tetzaveh. (The Midrash provides the full list.)

(d) Finally, Rav Kasher (ibid. p 121) cites from Minchas Bikurim (Y. Sekili), an early Torah authority, a very original -- but confirmed -- new count. There are exactly 39 times that "Shabbat" (or "Shabbaton") is mentioned in the Torah!

==========

FOOTNOTES

==========

[1] The Yerushalmi, however, questions whether to include in the count the Melachah of Yosef -- as above -- or the Melachah of the verse "Hashem finished on the seventh day the Melachah that he performed [during Creation] (Bereishit 2:2)

[2] The 65 appearances of the word "Melachah" in the Torah: (1) Bereishis 2:2; (2) Bereishis 2:2; (3) Bereishis 2:3; (4) Bereishis 33:14; (5) Bereishis 39:11; (6) Shemot 12:16; (7) Shemot 20:9; (8) Shemot 20:10; (9) Shemot 22:7; (10) Shemot 22:10; (11) Shemot 31:3; (12) Shemot 31:5; (13) Shemot 31:14; (14) Shemot 31:15; (15) Shemot 31:15; (16) Shemot 35:2; (17) Shemot 35:2; (18) Shemot 35:21; (19) Shemot 35:24; (20) Shemot 35:31; (21) Shemot 35:29; (22) Shemot 35:33; (23) Shemot 35:35; (24) Shemot 35:35; (25) Shemot 36:1; (26) Shemot 36:2; (27) Shemot 36:3; (28) Shemot 36:4; (29) Shemot 36:4; (30) Shemot 36:5; (31) Shemot 36:6; (32) Shemot 36:7; (33) Shemot 36:7; (34) Shemot 36:8; (35) Shemot 38:24; (36) Shemot 38:24; (37) Shemot 39:43; (38) Shemot 40:33; (39) Vayikra 7:24; (40) Vayikra 11:32; (41) Vayikra 13:48; (42) Vayikra 13:51; (43) Vayikra 16:29; (44) Vayikra 23:3; (45) Vayikra 23:3; (46) Vayikra 23:7; (47) Vayikra 23:8; (48) Vayikra 23:21; (49) Vayikra 23:25; (50) Vayikra 23:28; (51) Vayikra 23:30; (52) Vayikra 23:31; (53) Vayikra 23:35; (54) Vayikra 23:36; (55) Bamidbar 4:3; (56) Bamidbar 28:18; (57) Bamidbar 28:25; (58) Bamidbar 28:26; (59) Bamidbar 29:1; (60) Bamidbar 29:7; (61) Bamidbar 29:12; (62) Bamidbar 29:35; (63) Devarim 5:13; (64) Devarim 5:14; (65) Devarim 16:8

[3] Rabeinu Chananel's comments on this subject are printed in the margins of the common printing of the Talmud, Shabbat 49b, and are quoted by many of the other Rishonim ad loc.: Ra'avan #350, Ramban, Rashba, Ritva, Tosfos Rosh, Chidushei ha'Ran, Radvaz (from manuscript, cited by Tosfos Chadashim).

[4] #(1-3) [5] #(4) [6] #(7),(63)

[7] The Ramban, Rashba, Chidushei ha'Ran, and Radbaz cite Rabeinu Chananel as saying that this phrase appears *14* (and not 13) times. However, our version of Rabeinu Chananel, which is also the version cited by Tosfos ha'Rosh and Ra'avan, appears to be more reliable, for otherwise Rabeinu Chananel will have subtracted *22* from the total and not *21*

[8] Rav Kasher's solution is problematic, as he himself points out, since we are looking for a count that leaves behind *40* and not 39 Melachah's. As the Gemara says, we are unclear as to which of *two* verses is the 39th "Melachah." See Torah Sheleimah ibid. for another, even more problematic suggestion. See also Yefeh Einayim, Shabbat 49b, who offers his own suggestion (based on a textual emendation). However, his words, too, seem to involve a miscount, since the word "Melachah," without any prefix appears *24* times in the Torah.

[9] Ra'avan (ibid.) actually counts out all of the 61 appearances, one by one. Unfortunately, his words appear to be riddled with typist's errors -- for instance, although he claims that there are only 61 Melachah's, he counts out a total of 62(!) places where it appears. Also, it is not clear that Rabeinu Chananel meant to count the same way as Ra'avan counted.

[10] #(7),(14),(16),(44),(63) [11] #(9-10) [12] #(39)

[13] Others have attempted to correct Rabeinu Chananel's count, but they have done so by suggesting various emendations to his words, and by creating somewhat forced limitations for the Melachah count. See Harav Menachem Gettinger in Hadarom, Tishrei 5726, and Dr. Dov Heiman in Talpiyot, Tishrei 5721.