One who has two houses opposite each other on the two sides of the public domain, and he wishes to carry from one house to the other on Shabbat via the public domain, he may place a side post from here, on one side of one of the houses, and an additional side post from here, on the other side. Alternatively, he may place a cross beam from here, from one end of one house, and an additional cross beam from here, from the other side of the house, and then he may carry objects and place them in the area between them, for in this manner he turns the middle area into a private domain. The Rabbis said to him: One cannot render a public domain fit for carrying by means of an eiruv in this manner. Apparently, there is no way to establish an absolute public domain fit for carrying by means of an eiruv.

The Gemara questions its previous conclusion: And if you say that it is only in this manner, by way of a side post or a cross beam, that a public domain cannot be rendered fit for carrying, but by means of doors it can be rendered fit for carrying. But this is not true, as didn’t Rabba bar bar Ḥana say that Rabbi Yoḥanan said: With regard to Jerusalem, were it not for the fact that its doors are locked at night, one would be liable for carrying in it on Shabbat, because its thoroughfares are regarded as a public domain? This shows that the presence of a door is not sufficient to render it permitted to carry in a public domain; rather, the door must actually be locked.

And similarly, Ulla stated: With regard to the city entrances [abbulei] of Meḥoza, which meet the criteria for a public domain, were it not for the fact that their doors are locked, one would be liable for carrying in them, because they are regarded as a public domain. Apparently, without the actual locking of doors it is impossible to establish a public domain fit for carrying by means of the symbolic partitions of a side post or a cross beam. If so, how can the Sages in the baraita argue about how to establish a public domain fit for carrying?

Rather, Rav Yehuda said: The wording of the baraita must be emended so that this is what it says: How does one render alleyways that are not themselves public domains but are open on two opposite sides into the public domain fit for carrying by means of an eiruv? He constructs an opening in the form of a doorway from here, on one side of the alleyway, and a side post or a cross beam from here, on the other side.

It was stated that the amora’im differed on how the halakha is to be decided with regard to this issue. Rav said: The halakha is in accordance with the opinion of the anonymous first tanna of the baraita, and it is sufficient to have the form of a doorway on one side and a side post or cross beam on the other side in order to render it permitted to carry in an alleyway that is open on two opposite sides to the public domain. And Shmuel says: The halakha is in accordance with the opinion of Ḥananya, following the position of Beit Hillel, who also require a door on one side.

A dilemma was raised before the Sages concerning the position of Beit Hillel: According to Ḥananya, in accordance with the opinion of Beit Hillel, must this door be locked or need it not be locked? Come and hear a proof from that which Rav Yehuda said that Shmuel said, who, as mentioned earlier, rules in accordance with Beit Hillel: The door need not be locked. And similarly, Rav Mattana said that Shmuel said: The door need not be locked. Some say that Rav Mattana said: A case involving this very issue happened to me, and Shmuel said to me: The door need not be locked.

They raised a dilemma before Rav Anan with regard to this issue: Need the door be locked or need it not be locked? He said to them: Come and see these city entrances of Neharde’a that open on two opposite sides into the public domain, the gateways of which were filled up halfway with earth, so that the doors themselves could not possibly be locked. Mar Shmuel regularly goes in and out through them, but has never said anything to the people of Neharde’a about them. This shows that it is not necessary for the doors to be locked.

Rav Kahana rejected this proof and said: Those doors in Neharde’a were partially blocked, and therefore there was no need to lock them, but in general, the door of an alleyway that opens on both sides into the public domain must be locked.

The Gemara relates that when Rav Naḥman came to Neharde’a, he said: Clear away the earth, so that the doors can be locked. The Gemara attempts to understand Rav Naḥman’s instruction: Let us say that Rav Naḥman holds that the door of an alleyway that is open on two opposite sides must be locked. The Gemara explains: No, this is not proof. In order for carrying to be permitted in such an alleyway, it is enough that the doors be fit to be locked, even if they are not actually locked.

The Gemara describes a certain crooked, L-shaped alleyway that was in Neharde’a, upon which they imposed the stringency of Rav and the stringency of Shmuel, and required it to have doors. The Gemara attempts to understand this ruling: The stringency of Rav, namely, that which he said, that an L-shaped alleyway is regarded like an alleyway that is open on two opposite sides. But this is difficult, for didn’t Rav say that the law follows the anonymous first tanna of the baraita, who says that even an open alleyway itself does not require doors, and that an opening in the form of a doorway suffices?

The Gemara answers: They required doors in accordance with the opinion of Shmuel, who said that the halakha is in accordance with the opinion of Ḥananya. But this too is difficult, for if the doors were required in accordance with Shmuel’s opinion, didn’t Shmuel say that an L-shaped alleyway is regarded like an alleyway that is closed on one side, which does not need any doors at all? The Gemara explains: The doors were required in accordance with the opinion of Rav, who said that an L-shaped alleyway is regarded like an alleyway that is open on two opposite sides. Therefore, they adopted the stringencies of both Rav and Shmuel: Rav’s stringency that an L-shaped alleyway is deemed an open alleyway, and Shmuel’s stringency that an open alleyway requires a door.

The Gemara poses a question: But do we adopt the respective stringencies of two authorities who disagree on a series of issues? Wasn’t it taught in a baraita: The halakha is always in accordance with the opinion of Beit Hillel, but one who wishes to act in accordance with the opinion of Beit Shammai may do so, and one who wishes to act in accordance with the opinion of Beit Hillel may do so. If he wishes to adopt both the leniencies of Beit Shammai and also the leniencies of Beit Hillel, he is a wicked person. And if he wishes to adopt both the stringencies of Beit Shammai and also the stringencies of Beit Hillel, with regard to him the verse states: “The fool walks in darkness” (Ecclesiastes 2:14). Rather, he should act either in accordance with Beit Shammai, following both their leniencies and their stringencies, or in accordance with Beit Hillel, following both their leniencies and their stringencies.

The Gemara first raises a problem concerning the wording of the baraita: The baraita is itself difficult to understand, because it contains an internal contradiction between its clauses: You first said that the halakha is always in accordance with the opinion of Beit Hillel, and then you reversed that and said that one who wishes to act in accordance with the view of Beit Shammai may do so.

The Gemara answers: This is not difficult. Here, the baraita’s statement that a person may act as he wishes was made before the Divine Voice emerged and announced that the halakha is always in accordance with Beit Hillel; and here, the statement that the halakha is always in accordance with Beit Hillel was made after the Divine Voice issued this ruling.

And if you wish, say a different answer: Both this statement and that statement were made after the Divine Voice announced that the halakha is in accordance with Beit Hillel,