In my opinion one splits up the third Pasuk of Bereishis into 4+2. This is because what determines where one splits up a Pasuk - in case of need where there is no other option - is not that there should necessarily be an equal number of words in each half, but rather where the "Esnachta" (the musical note in the middle of the verse which indicates where the pause is) is. In Bereishis 1:3 the Esnachta is after "Or" which is the 4th word.

(After I wrote the above I saw that TESHUVAS CHASAM SOFER OC #10 DH Ve-HAYAH and DH U-PE'ER does not seem to agree with this. The discussion there is why in Kidush on Friday night do we start by saying "And it was evening and it was morning, the 6th day", which is starting in the middle of Bereishis 1:31? Chasam Sofer cites a contemporary of his - MAGEN GIBORIM - who seems to say that when there is an Esnachta or a "Zakef Katon", there is no problem with splitting up a Pasuk, and since "And it was evening..." comes after Esnachta, there is no problem in splitting up the Pasuk. Chasam Sofer strongly disagrees with the above Sevara and argues that it is contradicted by our Gemara, because the Gemara should not have had a question on Shmuel that he is breaking up a Pasuk, since he only broke it up after an Esnachta.

However I think that Chasam Sofer is not actually a contradiction to my answer, because even he will agree that according to Shmuel, who maintains that in some special cases one is allowed to split a Pasuk (for instance, in the case of our Gemara where there is a different problem of "Nichnasim ve'Yotzim"), in such cases it is preferable to make the break after the Esnachta, since this is after all a natural, logical pause.

KOL TUV

D. Bloom

I found a totally different approach in Talmud YERUSHALMI here (cited by YEFEI EINAYIM). According to Yerushalmi the posuk which is split is not the 3rd posuk of Bereishis but the 5th posuk. This is split so that "And it was evening, and it was morning, one day" becomes a verse on it's own. The reason that this is done is because we find in verse 13 that "And it was evening, and it was morning, the third day" is actually a complete verse according to our division also. KORBON HaEIDAH DH VAYEHI writes that since exactly the same kind of verse on the third day is considered a complete verse, then this verse of the first day is also considered complete, if it is cut at "And it was evening...". However this opinion (which is Shmuel's opinion in Bavli, and either Rav Cahana's or Rav Asi's opinion in Yerushalmi) agrees that one must not split up any other posuk.

However MAREH HaPONIM on Yerushalmi points out that this does not agree with Bavli. Rashi here DH POSEK writes that the Cohen reads 2 and a half verses and the Levi completes the half-verse of the Cohen plus another 2 verses, which suggests that it is the 3rd verse of Bereishis which is split, not the 5th.

KOL TUV

D.Bloom

1. I am not aware that there is a 2-word pasuk anywhere in the Torah. However I would like to throw the ball back into your court and ask: how could one split up verse 3 of Bereishis into 2 equal halves of 3+3 as you are suggesting? If we did this the first half would read "And Elokim said let there be...." and the second half ".....light, and there was light." These 2 halves could certainly not qualify as a pasuk, in my opinion, because each half is not coherent on its own.

Therefore I argue that it is better to have a coherent pasuk of 2 words (i.e."and there was light") than a meaningless pasuk of 3 words. Even if you want to claim that a 2-word pasuk is not considered a pasuk (which one is not necessarily forced to say because even though we never encountered a 2-word pasuk before, possibly one can argue that there could always be a first time), at any rate since the Gemara says that our case is "Lo Efshar" - "not possible" (i.e.the Pesukim here do not meet all the qualifications that one usually requires), at least in a Lo Efshar case, 2 words can be sufficient.

(In addition see TESHUVAS CHASAM SOFER OC #10 DH U-PE'ER who writes that Shmuel maintains that 2-and-a-half-word pesukim are equivalent to 3-word pesukim. According to this, Shmuel can argue that a pasuk of 2 words is not actually a complete pasuk, but it is a half-pasuk, and if you already have 2 complete pesukim, it is sufficient that the 3rd "pasuk" is actually only a half-pasuk.)

2. (a) If verse 3 is split into two, the Kohen reads verses 1 and 2 and verse 3 up till the first "Or". The Levi reads "Vayehi" and the second "Or" at the end of verse 3, and then reads verses 4 and 5. The Yisrael reads verses 6-8.

(b) If one does the division according to the YERUSHALMI, the Kohen reads verses 1-3, the Levi reads verses 4 and then verse 5 until "Laila" which is considered a complete verse. Since verse 5 from "Vayehi" to "Echad" is also considered a complete verse, this means the Levi has now read 3 verses. The Yisrael reads verses 6-8.

YEYASHER KOACH

D. Bloom

I think our messages "crossed in the air" and you will see that I cited TOSFOS KIDUSHIN 30a DH YESER as implying that 2 words cannot be considered as a pasuk, as you point out.

However a colleague pointed out to me that there is another way of understanding this Tosfos. It could be that Tosfos agrees that it is possible theoretically that a 2-word pasuk could exist. (Incidentally in the 10 commandments, according to the way we read them on Shavuos, there are 2-word pesukim - e.g. "Lo Tirtzach" - do not murder, so there is an example where 2 words can make a coherent pasuk). However the reason that Tosfos only mentions 3-word Pesukim in Tehilim is because usually 2-word Pesukim have no meaning. This is a similar to what we cited last time in the name of PISCHEI TESHUVAH YD 284:1 (I accidentally wrote last time PISKEI instead of PISCHEI) that one cannot tell if a 2-word Pasuk is expressing something mundane or holy.

I can mention another example of this. The MISHNAH in YOMA 37a relates that Queen Heleni made a gold tablet on which the Parshah of Sotah was written. The Gemara in Yoma top 38a states that this was written "Be-Serugin". RAMBAM, in his responsa PE'ER Ha'DOR #7, writes that this means she wrote 3 words on one line, and then wrote another 3 words on the next line, so that the 2 sets of words could not be read together.(This is cited by SHULCHAN ARUCH YD 283:3). The Rambam does not mean that the 3 words she wrote have a Din of a Pasuk, but rather that the reason she wrote 3 words at a time is because less than this, generally speaking, has no meaning.

Therefore the reason Tosfos did not mention 2 words is not necessarily because 2 words cannot be a Pasuk, but rather because David Ha'Melech, when composing the Book of Tehilim, would not have used Pesukim of less thah 3 words, because this is too short to be coherent.

YEYASHER KOACH GADOL

D. Bloom

1. I found a support for my first answer that one should break up the third Pasuk of Bereishis where the Esnachta is. This is from the TESHUVAS TORAH LISHMAH #374 (most scholars agree that this was written by the author of the BEN ISH CHAI - R. YOSEF CHAIM of Baghdad) who writes that his teachers ruled that when there is an Esnachta this is equivalent to a "Sof Pasuk" - "end of verse"- and there is no prohibition of "every verse which Moshe did not split up we must not split up" because since there is an Esnachta this means that Moshe Rabeinu himself split it up.

The Torah Lishmah writes that according to this one may say before Kidush on Shabbos morning "Therefore Hash-m blessed the Shabbos day and made it holy" even though this is only half a pasuk in Shemos 20:11 because there is an Esnachta before "Therefore". He also writes that this is the source of our custom to say "May fear and awe fall upon them, with the might of your arm they will fall like a stone" in Kidush Levanah even though this is only half of Shemos 15:16, because there is an Esnachta after "stone".

2. However I also found that the author of the Ben Ish Chai, in his TESHUVAS RAV PE'ALIM OC #11, cited a nice proof for your argument that two words cannot be a pasuk, from the Gemara in Gitin 6b which states that one may write two words from the Torah without "Sirtut" (underlining the words of the Torah) but one may not write three words without Sirtut. The Rav Pe'alim explains with this the source for the custom in Kidush on Friday night to say "Yom ha'Shishi" - "the 6th day" - before Kidush, even though this is only part of a verse. He argues, on the basis of Gitin 6b, that two words is not even considered a cut-up Pasuk. See the PISKEI TESHUVAH YD 284:1 who also writes that if there are only two words, one cannot distinguish whether this is a holy statement or simply mundane words. This strengthens your argument that two words cannot be considered a Pasuk.

(See also the TOSFOS in KIDUSHIN 30a DH YESER who seems to imply also that the minimum that could be considered a Pasuk is three words). However I still contend that one must say that even though normally two words have no importance, nevertheless Shmuel maintains that where there is no other alternative, it is better to divide pasuk three of Bereishis into 4 + 2, because that way both halves have some meaning, than dividing it into 3 + 3, in which case the first half would have no meaning at all.

KOL TUV

D. Bloom

In previous replies we mentioned that MAGEN GIBORIM and TORAH LiSHMAH maintain that where there is an Esnachta this is equivalent to the end of a pasuk, and one may break a pasuk there.

This opinion seems to be contradicted by the Gemara here which states that Rav maintains that one is not allowed to split up the 3rd pasuk of Bereishis because Moshe did not split it. The question is, that since there is an Esnachta in pasuk 3, why did Rav not split it there?

Magen Giborim 61:3 (written by the author of SHOEL U-MESHIV) answers this question and asserts that according to Rav there is actually no prohibition against splitting the verse at an Esnachta, but on the other hand if one does so one does not gain an extra pasuk. According to this, the Gemara which says "we do not split the verse" means that we cannot split it up and thereby make it into two verses, because the division of the verse is not complete(but not that there is actually a prohibition to split it).

MG writes that Rav is consistent with his opinion in NEDARIM 37a and 37b, that one is not allowed to be paid for teaching "Pisuk Ta'amim" because the Halachos of where the verse end are deoraisa. Shmuel however maintains that Pisuk Ta'amim is only derabonon but mideoraisa one may break up the verses of the Torah wherever one desires. Therefore in our Gemara, which is "Lo Efshar", Shmuel maintains that one can break up the verse in the middle, and thereby gain an additional verse. Similarly, to teach young children, who are not capable of reading an entire verse, Rabbi Chanina HaGadol ruled that may one revert to the deoraisa din and stop in the middle.

KOL TUV

D. Bloom

1. I was very happy to find that our discussion has already been mentioned by CHIDUSHIM U-BIURIM by Rabbi Chaim Greineman Shlita of Bnei Brak, to Megilah 22a. (The Gemara there also cites the dispute between Rav and Shmuel).

(a) He writes that it would appear that according to Shmuel the Kohen stops after "Or" (the 4th word in the 3rd verse) and the Levi starts from "Vayehi", because otherwise the meaning will be broken off in the middle. This implies that the Levi only reads 2 words from the 3rd verse, but nevertheless this is sufficient. R. Greinemann refers us to Yerushalmi (so far this is my argument). However he concludes "Tzorich Iyun" i.e . this matter requires further research (he is referring to your argument, that it may be that if the Levi only reads 2 words of a verse, this is invalid).

(b) I also found in Chidushim U-Biurim to Kidushin 30a that he again refers to our discussion. TOSFOS there DH YESER (which you cited) writes that even if the verses of Tehilim would only contain 3 words, they would still not be as many as in the Chumash. R. Greineman writes that it might be that in Tehilim any piece which is coherent if read on it's own, may be considered as a pasuk. (In fact this argument has already been made there by TIFERES SHMUEL to MAHARSHA #2 who writes that in their Tehilim the verses were extremely short, contrary to Tosfos' reasoning, i.e. the verses contained less than 3 words). So Chidushim U-Biurim's argument is that the verses were lessthan 3 words long, and therefore there were more verses in Tehilim than in Chumash. However he concludes by referring us to what he writes on Megilah 22a, which presumably means that he is doubtful here whether one can have a pasuk of less than 3 words.

(c) In addition CHIDUSHIM U-BIURIM to Gittin 6b writes that if someone writes 2 words on parchment this does not possess the holiness of Sefer Torah etc.

2.I also found in the commentary of RABEINU AVRAHAM MIN HaHAR (by one of the Rishonim from Provence, at a similar period and location to the authors of Tosfos) to Megilah 22a, that the Kohen reads the first half of verse 3 and the Levi the second half. One could argue that this is a proof for your position, that each one reads exactly half, i.e. 3 words, but I could refute this proof, that he does not mean exactly a half, but rather each one reads a part, and it could still be 4+2.

KOL TUV

D. Bloom