Siman 361

A roof adjacent to the reshut harabim, meaning that the house does not stand in a courtyard far from the reshut harabim, but rather on the boundary of the reshut harabim like our buildings. And the roof from three of its sides is ten tefachim high as is the way of roofs, but from the side of the reshut harabim it is not ten tefachim high, such as where the ground of the street is higher than the ground of the courtyard, and from the Torah it is a complete reshut hayachid. Nevertheless, since it is within ten of the reshut harabim, and many people use it—for they place their hats and their scarves upon it, as when people are tired in the summer days they take their hats and their scarves from their heads until the wind blows through them (Rashi, Eruvin 94b). Therefore, it is forbidden for the owner of the roof to use it from the courtyard until he makes for it a fixed ladder from his courtyard, for by this he has revealed his intention that he has removed this roof from the power of the people of the reshut harabim, and his desire is to use it himself. But a temporary ladder is not a removal, for since he will eventually take the ladder away, it is as if it is not there. And what is a fixed ladder? Anything that he fixed there whether for Shabbat or for weekdays. But if he fixed it for Shabbat and not for weekdays, or for weekdays and not for Shabbat, this is not fixed. And it appears to me that a fixed ladder is larger than a temporary ladder, for so it implies explicitly in Eruvin (76b), and it implies ibid. (77b) that a fixed ladder must have no fewer than four rungs, which are steps, see there.

And if the roof is ten high also from the side of the reshut harabim—one does not need any ladder. And if from all sides it is not ten high—a ladder is of no avail, (and in the Magen Avraham, subsection 1, there is a printing error as the Machatzit HaShekel wrote) and even though the house within it is ten tefachim high, it is not called a reshut hayachid on the roof but rather a karmelit. (The HaGaon Rav Zalman) And if from the side of the reshut harabim it is ten high, and from the side of the courtyard it is not ten high—likewise one does not need a ladder. (Tosefot Shabbat, subsection 1, in the name of the Hagahot Ashri, but this is not found in the Hagahot Ashri; nevertheless, for the halacha it appears so. And all this is when many people shoulder loads on the roof, but if they do not shoulder loads—one needs nothing. And the Beit Yosef wrote that from the Tur it is implied even when they do not shoulder loads, but the Bach wrote that this is also the intention of the Tur, see there; and for Rashi in Eruvin 84b there is another shitah in this, but that which we have written is the shitah of the Rambam and the Tosafot ibid.).

A courtyard that was breached in its entirety or for more than ten cubits into a reshut harabim—the place of the partition is judged as a karmelit, for it is judged as the sides of a reshut harabim, and the sides of a reshut harabim are not considered like a reshut harabim. And also in the courtyard itself—it is automatically forbidden to carry in it more than four cubits, for it is breached in its entirety or for more than ten, but according to the Torah the entire courtyard, except for the place of the partition, is a complete reshut hayachid (Tosafot 93b, s.v. 'Chayav'), and so it is in the Yerushalmi there. However, the opinion of the Rambam is that the entire courtyard is considered a karmelit according to the Torah, as I wrote at the beginning of siman 345, see there.

But if the breach is only ten amot, and certainly less than this, it is a doorway, and it is permitted to carry in the entire courtyard. This is when the breach is along the length of the wall, but when the breach is at a corner—even less than ten forbids, for people do not make a doorway at a corner, as our Sages, of blessed memory, said (93b). However, if he made the form of a doorway, with mezuzot and a lintel, it is properly a doorway even at a corner (Tosafot 6a, s.v. "de-pitha"). But along the length of the wall, even a complete breach without fixing it with mezuzot or a lintel is considered a doorway within ten, and certainly less than ten, and in such cases it does not avail at a corner to be considered a doorway. The Tur and the Shulchan Aruch did not mention this because it is simple, for this is not a halacha le-Moshe mi-Sinai that there is no doorway at a corner, but rather such is the way of the world, and since this person took the trouble to make it a doorway, why should we care.

A house that stands on the boundary of the public domain or a karmelit, and the wall on the side of the public domain was breached—if the roof remains as before, even if the breach is more than ten cubits, and even in its entirety, this is nothing, because we say "the edge of the ceiling descends and closes," and it is as if the partition stands in its completeness; and this matter is a halacha le-Moshe mi-Sinai. And even if two of its walls that stand adjacent to one another were breached, such as east and south or east and north, and likewise west and south or west and north—we say "the edge of the ceiling descends and closes." But with two walls one opposite the other, we do not say "the edge of the ceiling descends and closes," because this is like a passageway. And the ceiling must be four by four cubits wide, for with less than four we do not say "the edge of the ceiling" etc., as I wrote in siman 358. And another of the conditions of "the edge of the ceiling" is that the roof must be level, but with a slope, we do not say "the edge of the ceiling," because it has no "edge." And even though in siman 358, section 24, regarding a karpef of three se'ah where one roofed a house-sized area within it, it was explained there that we say "the edge of the ceiling" even with a slope—there they were lenient, because a roofed area is not called a karpef. And that which we said there using the language "the edge of the ceiling descends and closes" was not precise, but rather it means that it diminishes it. But in a place where the primary permit is based on the reasoning of "the edge of the ceiling," we do not say "the edge of the ceiling descends and closes" with a slope (see Tosafot 90a, starting with "Ela," and so wrote the Tosefot Shabbat). And there is one who says that with a width of four cubits we say "the edge of the ceiling" even with a slope (Taz, subsection 2), but it is not compelling; and so ruled our teacher, the Beit Yosef, at the end of section 2, see there. And it is simple that we say "the edge of the ceiling" even at a height above twenty cubits. And it is not similar to the beam of an alleyway, which is invalid above twenty, for it is only a handbreadth and does not have the status of "the edge of the ceiling" but is merely for a recognition, and above twenty there is no recognition. But for "the edge of the ceiling," there is no limit. And just as we said regarding a courtyard, that people do not make an opening in a corner, so too regarding a house when it is breached in a corner—even with less than ten cubits, and the roofing was also breached with it until it remained diagonal—we do not say "the edge of the ceiling" there to permit carrying in the house that was breached into the public domain or a karmelit, for we do not say "the edge of the ceiling" in a corner. For the reason for "the edge of the ceiling" is also because of an "opening." That is to say, since we say "the edge of the ceiling descends and closes," it is only like an opening. And since they do not make an opening in a corner, consequently we also do not say "the edge of the ceiling" there. And there are other opinions on this matter, but what we have written is the view of the Tur and the Shulchan Aruch based on the words of our teachers, the authors of the Tosafot, at the end of the chapter "Kol Gagot," see there; and the words of the Rambam are unclear on this matter in chapter 16, see there.