QUESTION: The Gemara quotes Rebbi Yehoshua ben Levi who says that in 24 places the Torah refers to Kohanim as "Leviyim." RASHI in Chulin (24b) suggests that they are called "Leviyim" because they perform the Avodah in the Beis ha'Mikdash, as the word "Levi" also means "those who help serve" (as in Bamidbar 18:2). The Gemara cites one example of such a verse (Yechezkel 44:15).

What are the other 23 places in which the Torah refers to Kohanim as "Leviyim"? The Acharonim discuss this question and have great difficulty identifying all 24 places.

The BEN YEHOYADA in Bechoros (4b) writes that he searched and found only eleven places where the word "Leviyim" refers to Kohanim and not to Leviyim. However, RAV DAVID COHEN shlit'a (in OHEL DAVID, end of volume 1) writes that he counted the places where the word "Leviyim" refers either to Kohanim by themselves or to Kohanim and Leviyim together, and he found 76 places where the Torah refers to Kohanim as "Leviyim."

Rav David Cohen asks further that there can be no set number of places in which the word "Leviyim" means both Kohanim and Leviyim, because the Gemara here cites a dispute about whether the verses in the Torah which discuss giving Ma'aser to a Levi include Kohanim as well. Accordingly, the number of verses in which "Leviyim" refers to Kohanim is subject to this dispute and cannot be the same according to all Tana'im. Why does Rebbi Yehoshua ben Levi assume that everyone agrees that there are exactly 24 places in which the word "Leviyim" refers to Kohanim?

Rav David Cohen also points out that Rashi in Yehoshua (3:3) records a different count in the name of the Midrash (Bereishis Rabah). According to the Midrash, there are 48 places in which the word "Leviyim" refers to Kohanim. (The Midrash, however, does not quote this statement in the name of Rebbi Yehoshua ben Levi.)

How is the statement of Rebbi Yehoshua ben Levi to be reconciled with the empirical count?

(a) The MEGADIM CHADASHIM (Berachos 51a) asserts that the Gemara does not intend to give an exact number, but rather it means to say that there are many places in which the Torah refers to Kohanim as "Leviyim." The number 24 is an exaggeration.

According to this approach, when the Midrash asserts that there are 48 places, it is simply choosing a different number for the exaggeration.

(b) RAV DAVID COHEN suggests that the Gemara indeed counts the places where "Leviyim" refers either to Kohanim or to both Kohanim and Leviyim. The reason it counts only 24 instead of many more is that it does not count "Pesukim" (verses), but only "Mekomos" -- "places," or topics. There are 24 basic sections in the Torah which refer to Kohanim as "Leviyim," while each section may contain many instances of the word "Leviyim" as a reference to Kohanim. Rav Cohen attempts to define what is considered a "place" or topic-section but does not reach a definitive conclusion.

(c) The Ben Yehoyada reached the number eleven by counting only the places where the words "ha'Kohanim ha'Leviyim" are used to refer to the Kohanim. However, he overlooked three instances in which that phrase refers only to Kohanim (besides the places he intentionally omitted, where the verse refers to both Kohanim and Leviyim). Moreover, he counted one place that should not have been included in the count because it refers to Kohanim and Leviyim (as the RADAK explains in Divrei ha'Yamim I 9:2). Hence, the total count is thirteen.

Additional instances are cited by other commentators. RASHI in Devarim (18:7) comments in the name of the Sifri that two verses there use the word "Levi" to refer to the Kohen and not to the Levi.

The RADAK in Yechezkel (48:11) and Divrei ha'Yamim II (5:4) proves that the word "Leviyim" refers only to Kohanim.

The METZUDAS TZIYON explains that the word "ha'Leviyim" in Yechezkel (44:10) refers to Kohanim, as is evident from the context there. The same is true of the verse in Yirmeyahu (33:22), as is evident from the context there (see verses 18 and 21). These instances bring the total to nineteen.

The RAMBAM (Mitzvos Aseh #34; see RAMBAN, Shoresh #3) writes that it is a Mitzvas Aseh for Kohanim to carry the Aron, even though the verse specifically says that the Leviyim (Benei Kehas) shall carry it. The Rambam says that the command to the Leviyim to carry the Aron applied only in the desert, where there were only three Kohanim (who could not carry the Aron by themselves). For all ensuing generations, the Kohanim are enjoined to carry the Aron.

However, in five places in Nevi'im and Kesuvim the verse states that Leviyim carried the Aron, as the RAMBAN asks (Shoresh #3). The Acharonim answer that according to the Rambam, the word "Leviyim" in those five places refers to Kohanim and not to Leviyim. Accordingly, there are exactly 24 places in Tanach in which the word "Leviyim" refers exclusively to Kohanim! (The Ramban will have to rely on the first answer above, that the number is an exaggeration.)

When the Midrash mentions 48 places, perhaps it includes the places in which the word "Leviyim" refers to both Kohanim and Leviyim (and it follows the opinion of Rebbi Akiva in Yevamos here who says that Kohanim are not entitled to eat Ma'aser Rishon along with Leviyim). There remain 24 places where the word "Leviyim" refers to both Kohanim and Leviyim. Those 24 verses mentioned by the Gemara in which Kohanim are referred to as "Leviyim" are as follows:

(1) Devarim 17:9 (2) 17:18 (3) 18:1 (4) 18:6 (5) 18:7 (6) 24:8 (7) 27:9 (8) Yehoshua 3:3 (9) 8:33 (10) Shmuel I 6:15 (11) Yirmeyahu 33:18 (12) 33:21 (13) 33:22 (14) Yechezkel 43:19 (15) 44:10 (16) 44:15 (17) 48:11 (18) Divrei ha'Yamim I 15:12 (19) 15:15 (20) 15:26 (21) 15:27 (22) Divrei ha'Yamim II 5:4 (23) 23:18 (24) 30:27

- The eleven places which the Ben Yehoyada counts are numbers 1, 2, 3, 6, 8, 9, 11, 16, 23, 24. He also includes Divrei ha'Yamim I 9:2.

- The five places which discuss Leviyim carrying the Aron are numbers 10, 18, 19, 20, 21.

- With regard to number 9, the Yerushalmi (cited by Tosfos to Sotah 37a, DH Iy) asserts that the words "ha'Kohanim ha'Leviyim Nos'ei Aron Bris Hash-m" in the verse refer to the Kohanim only and not to the Leviyim. This is consistent with the opinion of the Rambam, who says that only Kohanim were allowed to carry the Aron. The Bavli (ibid.), however, might disagree about this point. In any case, the Rishonim point out that the Bavli (Sotah 33b) presents a strong challenge to the opinion of the Rambam that only Kohanim are permitted to carry the Aron. (M. Kornfeld)

(This Insight is based primarily on Rav David Cohen's exhaustive lists of the appearances of the word "Levi" which might refer to Kohanim, and his insightful suggestions recorded in the end of Ohel David, volume 1.)

QUESTION: The Gemara proves that Kohanim are also called "Leviyim" from the verse, "veha'Kohanim ha'Leviyim Bnei Tzadok" (Yechezkel 44:15). Although the Gemara says that there are 24 places in Tanach in which Kohanim are called "Leviyim," the Gemara cites this verse alone, apparently because it is the clearest proof that the word "Leviyim" may refer to Kohanim. In all of the other places, it is only the context of the verse which indicates that the word "Leviyim" refers to Kohanim, but here the verse explicitly refers to the sons of Tzadok -- who was the Kohen Gadol -- as "Leviyim."

The phrase "ha'Kohanim ha'Leviyim" ("the Kohanim, the Leviyim") which appears in other places is not sufficient proof because those words can be interpreted to mean "the Kohanim and the Leviyim." Indeed, the RADAK in Divrei ha'Yamim II (5:4) points out that when the verse there (v. 5) uses the phrase "Kohanim ha'Leviyim" it must mean "the Kohanim and the Leviyim," because the parallel verse in Melachim I (8:4) reads, "ha'Kohanim veha'Leviyim" -- "the Kohanim and the Leviyim."

However, RAV DAVID COHEN (Ohel David, end of volume 1) asks that the Gemara could have cited an earlier verse as proof that "Leviyim" may refer to Kohanim. An earlier verse in Yechezkel (43:19) states, "ha'Kohanim ha'Leviyim Asher Hem mi'Zera Tzadok." The proof from this verse is just as clear as the proof which the Gemara cites. Why does the Gemara cite the later verse and not the earlier verse?

(In the end of Ohel David, volume 4, Rav David Cohen quotes a colleague who suggests that the later verse is quoted because it appears in the Haftarah of Parshas Emor and thus is well-known. However, this explanation is unsatisfactory, because the earlier verse in Yechezkel also appears in the Haftarah of Parshas Tetzaveh!)

ANSWER: The RAMBAN (Milchamos, 80a) explains that when a verse, Mishnah, Beraisa, or Amora lists several items and, at the end of the list, makes a statement about them, the statement may refer only to the earlier items and not to the later ones in the list. For example, Rebbi Avahu there (80a) lists "the signs of a Seris, and Aylonis, and a child born after eight months," and then makes a statement that "we cannot rely upon those signs until the person reaches the age of twenty." The latter statement refers back to Seris and Aylonis, and not to an eighth-month baby.

Based on this premise, the reason why the Gemara cites the later verse in Yechezkel is understood. The earlier verse, "ha'Kohanim ha'Leviyim Asher Hem mi'Zera Tzadok," may actually be a list (Kohanim, Leviyim) and a statement ("who are from the descendants of Tzadok") which refers back only to the first item in the list, Kohanim. As such, it provides no proof at all that the word "Leviyim" refers to Kohanim.

In contrast, the verse which the Gemara cites, "veha'Kohanim ha'Leviyim Bnei Tzadok" -- "the Kohanim, the Leviyim the sons of Tzadok," makes no independent statement about the aforementioned items. It does not refer back (with a pronoun) to anything in the list and say, "they are...." Rather, "Bnei Tzadok" ("the sons of Tzadok") is an inherent part of the phrase, "ha'Kohanim ha'Leviyim." Since it is one, undivided phrase, the Leviyim clearly are the ones referred to as "Bnei Tzadok." (M. Kornfeld)