Tractate Berakhot of the Talmud of Eretz-Israel: 0004

1:
The Gemara had quoted the view of Rabbi Abba ben-Pappa to the effect that "One star definitely day, two stars possibly night, three stars definitely night." [See 0003] The Gemara now challenges part of this definition. Rabbi Abba says that there is a time when it is neither definitely day nor definitely night; that time is defined when two stars are visible. This raises awkward questions regarding a person's responsibility for his actions. The Gemara illustrates this by proposing a possible scenario.

2:
A person performs a melakhah. A melakhah is an act — simple or complicated — whose performance is forbidden on Shabbat. So, if a person does a melakhah late one Friday afternoon when one star is already visible he is not liable because according to the definition of Rabbi Abba it is still day, Shabbat has not yet begun. If that person performs the same act on the same day but a little later when two stars are visible, according to Rabbi Abba he would have to bring a special sort of offering to the Bet Mikdash because it might still be Friday or it might already be Shabbat. (This offering is obligatory when it is possible that one committed a sin but does not know whether he sinned or not: the sin is definite but whether it was committed or not is unknown.) If that same person performs the same act a little later when three stars are visible he has definitely committed a sin because it is already dark and therefore it is certainly Shabbat. (Such a person must bring a guilt offering.)

3:
We now come to the second stage of the argument. instead of it all happening late one Friday afternoon let's say that it happens late one Saturday afternoon. When one star is visible it is still Shabbat and therefore the act is forbidden; when two stars are visible it may still be Shabbat and it may not, so the person would incur the offering to be brought when one is not sure that a sin had been committed; when three stars are visible Shabbat has definitely ended so there is no liability.

4:
Having expounded the situations Rabbi Yosé now propounds the question. Let's imagine a situation in which a person performs a melakhah on Friday when only two stars are visible. In other words, at a time when it could be day or it could be night. Now the same thing could also happen on Saturday night. If we were to say that it is still day on Friday (and therefore he is not liable) we would also have to say that it is still day at that same time on Saturday (and he would be liable). This is because on Friday it is not yet Shabbat while at the same hour on Saturday it is still Shabbat. We could also apply the opposite argument that if we were to say that it is already night on Saturday and therefore that person is not liable we would have to say that at the same time on Friday, when it is already Shabbat, he would be liable. Thus you have an anomalous situation in which different situations would apply on two different occasions when only two stars are visible.

5:
Rabbi Yosé now brings another scenario. On Friday, when only two stars are visible, a person plucks a piece of fruit from a tree. The size of the fruit is half the amount that constitutes a melakhah. On Saturday morning he does exactly the same thing, thus completing, as it were, a melakhah. Alternatively, On Saturday morning he plucks a piece of fruit from a tree and that fruit is halof the amount that constitutes a melakhah. He does the same thing on Saturday afternoon when only two stars are visible. In either case, if we were to say that it was still day in the former case (and therefore no complete melakhah was performed because only half the act was performed when it was Shabbat) then we must say the same thing regarding the latter case. And, of course, one could also reverse the judgement, starting with Saturday night instead of Friday night.

6:
The Gemara answers these conundrums by repeating and re-emphasizing what has already been said. The problem that has been raised could only come about if you assume that there is halakhic standing for the visibility of two stars. But we have already said that the first star doesn't count! Therefore the second star in the above scenarios is really the first star that counts towards nightfall, and it is not yet night. Only when three stars that are visible only at night (and none of them are visible by day) can we say that night has begun. In order to clarify the situation Rabbi Yosé explains that when one is counting three stars one must see three stars that can only be seen at night, excluding any star that can be seen while there is still light.

Marty Berman has a question concerning the actual text of the very begining of our tractate. He writes:

In your version of the Yerushalmi you have:

Rabbi Ḥiyya [also] teaches 'from the time when people go in to eat their bread on Friday Nights'.

I have not found this girsa but instead:

Could you please explain.

I respond:

The difference between the two versions of the text is that the version I offered has an additional clause (which is not really material to the discussion). I have no explanation! I have checked all the books that I have available to me and have found that they all include that extra clause. But that does not mean very much!

The text of the Yerushalmi that has come down to us is quite late. No manuscripts have survived from antiquity. The first ever printing of the Yerushalmi was produced in Italy in 1523. It was, according to the printer, based on four manuscripts. However, only one of those manuscripts has survived and is to be found in the University of Leyden, Netherlands. It was written in 1289 by Rabbi Yeḥi'el ben-Yekuti'el ha-Rofé. That manuscript is available for viewing online and I have checked our text: it accords with the version that Marty has supplied. A very important online edition is being prepared which not only has explanations but also a very carefully corrected text. Several tractates have already been made available, but unfortunately, Tractate Berakhot is not one of them.

My suspicion is that the difference between the two versions derives from which "modern" printing the text was derived. There was a printing done in the Polish town of Krotoschin in 1866 and another which was done some years later by "The Widow Romm and Brothers" in Vilna, which is now considered to be more or less standard.

I'm sorry that I can't be more helpful.