The Source that a Reshus haRabbim Requires 600,000 People
The following is based on an article by Rav Mordechai Yehuah Leiv Zaks in no. 27 vol. 6 (1953) of HaPardeis. (H/T Eruv Online blog.)
All of our community eiruvin rely on Rashi’s opinion that if the area does not have ששים ריבו, sixty myriads (=600,000), people using the street, it is not a true reshus harabbim (public domain) according to the Torah, only a karmelis (an in-between domain) which is only prohibited derabbanan if it has no eiruv.
The Ramban (twice: Shabbos 57a, Eiruvin 59a), the Ritva (Eiruvin ad loc.) the Maggid Mishnah (Hil’ Shabbos 14:1) and the Rivash (#405) all ask where Rashi gets this from. They knew of no source from Chazal.
Well, Rabbi Zaks believed he found the source!
The Me’iri (Shabbos 6a) has a line in the gemara we don’t have. ” “לפי שאמרו בכאן אין רשות הרבים בבבל — because they say that here [in Bavel] there is no reshus harabbim.” The Ravyah (#201) quotes a different variant of the missing line, “אמר עולה אין דרך רשות הרבים בבבל — Ula said: there is no route that is a reshus harabbim in Bavel.”
Notice that this alone implies there are more criteria than we have in the gemara. As those who don’t use community eiruvin note, going just from the sources we have, few neighborhoods could have an eiruv. What would have excluded every city and town in Bavel, unless there are other requirements for a reshus harabbim?
Rav Zaks actually found a manuscript with the missing line from Ulla. In the Vatican Library (Bibliotheca Apostolica, Ebr. 127), paralleling our Shabbos 7a.
אמ’ עולה אין רשו’ הרבי’ בבבל
אמ’ רבה בר בר חנה ירושלים אילמלא דלתותי נעולות בלילה חייבי משו’ רשו’ הרבי.
Ulah said: there is no reshus harabbim in Bavel.
Rabbah bar bar Chanah said: Yerushalayim, if its doors weren’t closed at night, would be obligated as a reshus harabbim.
(This second line is in our gemaros in Eiruvin 6b, and the Vatican manuscript has a second copy there as well.)
So Ulla believes there is neither a 600,000 person crowd in Bavel nor a reshus harabbim there.
Someone may have inserted the line in a margin note in the Munich codex (Bayerische Staatsbibliothek, Hebrew MS. No. 95) copy of Shabbos as well. The text is barely visible, though. See the area in the rectangle.
So the only question left is what requirement would Ula have that would rule out the existence of a reshus harabbim in Bavel, but doesn’t apply to Yerushalayim?
It is not unreasonable to believe it is related to another statement by Ula of the form “there is no ____ in Bavel”. As in this line that is found in Berakhos (58a), which states:
אמר עולה נקיטינן אין אוכלוסא בבבל
תנא אין אוכלוסא פחותה מששים רבוא
Ula said there is no ukhlusa in Bavel.
Mishna: there is no ulukhsa smaller than sixty myriads [of people].
Voila! Ula, as taught in these versions of the gemara, appears to be the source of saying that a reshus harabbim is only an area trafficked by 600,000 people and since there is no such areas in Bavel, there is no reshus harabbim in Bavel.
I want to highlight the sources of the different variants of the Talmud Bavli in question.
The Vilna edition of shas that we all use derived from the Bomberg shas (1519-1523, Venice) in having the same dapim, Rashi on the outside, Tosafos on the inside, and is the primary source for the specific text as well. Bomberg, in turn, based his version of the text on that of Joshua Solomon Soncino (1483, in the town of Soncino, which is in the duchy of Milan), who also had the same layout of Rashi and Tosafos, but not our current pagination. All in all, the text we use has an Italian heritage.
I would therefore highlight the origins the Vatican manuscript, which was produced in Germany in the late 14th century and of the Munich codex (1369), which based on a mid-9th century manuscript. The text or texts Rabbi Zaks found are both from Medieval Germany. They are more likely to be similar to the text Rashi studied from. Rashi is simply ruling like Ula, as he appears in Ashkenazic tradition.