The Gemara rejects this parallel: According to the opinion of Rabbi Eliezer, everyone agrees that he holds that the priesthood is voided. When they disagree, it is according to the opinion of Rabbi Yehoshua: The one who says the priesthood is dead holds in accordance with the straightforward understanding of the opinion of Rabbi Yehoshua. And the one who says the priesthood is voided can also hold in accordance with his opinion and explain that Rabbi Yehoshua states his opinion only there, with regard to Temple service, as it is written: “Bless, God, his property [ḥeilo], and accept the work of his hands” (Deuteronomy 33:11). The term ḥeilo is interpreted homiletically to mean that even the offerings of those disqualified from Temple service due to flawed lineage [ḥalalin] are accepted after the fact. But here, with regard to the status of the priest, even Rabbi Yehoshua concedes that the priesthood is voided retroactively.

§ The mishna teaches: If the verdict of a murderer was decided at a time when there was no High Priest, and likewise in the cases of one who unintentionally killed a High Priest and in the case of a High Priest who killed unintentionally, the unintentional murderer never leaves the city of refuge. And one who is exiled may not leave the city at all; even if the Jewish people require his services, and even if he is the general of the army of the Jewish people like Joab ben Zeruiah, he does not leave the city of refuge ever. Rav Yehuda says that Rav says: Joab made two errors at that moment, when he fled from Solomon, as it is written: “And Joab fled to the Tent of God and grasped the horns of the altar” (I Kings 2:28).

He erred in that only the top of the altar provides refuge, and he grasped its corners. Moreover, he erred in that only the altar of the eternal House, i.e., the Temple. provides refuge, and he grasped the altar at Shiloh. Abaye said: It is with regard to this that Joab also erred, as the altar provides refuge only for a priest who grasps the roof of the altar and his service is in his hand, and Joab was a non-priest.

Apropos errors, the Gemara cites that Reish Lakish says: The angel of Rome is destined to make three errors, as it is written: “Who is this who comes from Edom, with crimsoned garments from Bozrah?” (Isaiah 63:1), which is a parable for God’s arrival after killing the angel of Rome in Bozrah. The angel of Rome will err in that it is only the city of Bezer that provides refuge and he exiled himself to Bozrah; he will err in that it provides refuge only to an unintentional murderer and he was an intentional murderer; and he will err in that it provides refuge only to a person and he is an angel.

§ The Gemara resumes its analysis of the mishna. Rabbi Abbahu says: Cities of refuge were not given for the purpose of burial of unintentional murderers within them, as it is written with regard to the Levite cities: “And their open land shall be for their cattle, and for their property, and for all their beasts [ḥayyatam]” (Numbers 35:3), from which it is derived: For life [leḥayyim] they are given, but not for burial. Even Levites who reside in these cities are buried beyond the open land surrounding the city. The Gemara raises an objection to this from the mishna’s interpretation of the term: “That he fled there” (Numbers 35:25), from which it is derived: There shall be his dwelling, there shall be his death, there shall be his burial. The Gemara answers: A murderer is different, as the Merciful One revealed concerning him that he is to be buried there. That does not apply to the other residents of the city.

The mishna teaches: Just as an unintentional murderer is admitted to the city of refuge, so is he admitted to its outskirts, located within the Shabbat boundary. And the Gemara raises a contradiction from a baraita with regard to that which is written concerning the unintentional murderer: “And he shall dwell in it” (Numbers 35:25), from which it is inferred: “In it,” but not within its boundary. Abaye said: This is not difficult. Here, the mishna is referring to the unintentional murderer being admitted to the city, which will provide refuge from the blood redeemer, who may not kill him there. There, the baraita is referring to the place where it is permitted for the murderer to dwell, i.e., within the city itself and not on its outskirts.

The Gemara asks: Why is it necessary to state that the murderer may not dwell on the outskirts of the city? Let him derive that halakha from the fact that one may not render the field of a Levite city an open space, nor an open space a field, nor an open space part of the city, nor the city an open space. Apparently, the outskirts of the city, whose status is that of an open space, may not be utilized for residential purposes. Rav Sheshet said: It is necessary to state this halakha only for tunnels. If a murderer excavated a tunnel on the outskirts of the city, although he did not violate the prohibition against ruining the fields of the city, he may not reside there based on the halakha by Torah law that he must reside inside the city.

§ The mishna teaches a dispute between Rabbi Yosei HaGelili and Rabbi Akiva with regard to a case where the unintentional murderer emerged beyond the Shabbat boundary of the city of refuge, and the blood redeemer found him there. The Sages taught that it is written: “And the blood redeemer finds him outside the border of his city of refuge and the blood redeemer murders the murderer, he has no blood” (Numbers 35:27): It is a mitzva for the blood redeemer to kill him, and if there is no blood redeemer available to fulfill this mitzva, it is optional for any person to do so; this is the statement of Rabbi Yosei HaGelili. Rabbi Akiva says: It is optional for the blood redeemer to kill him, and any other person is liable for killing him.

The Gemara analyzes the mishna: What is the reason for the opinion of Rabbi Yosei HaGelili? He says: Is it written: If the blood redeemer murders the unintentional murderer, he has no blood? It states: “And the blood redeemer murders the murderer,” indicating that it is a mitzva. And Rabbi Akiva says: Is it written: The blood redeemer shall murder, in the imperative? It merely states: “And the blood redeemer murders,” which is merely relating the scenario under discussion.

On a related note, the Gemara cites that Mar Zutra bar Toviyya says that Rav says: In the case of a murderer who emerged beyond the Shabbat boundary of the city of refuge, and the blood redeemer found him there and killed him, the blood redeemer is executed for killing him. The Gemara asks: In accordance with whose opinion did Rav issue this ruling? He issued it neither in accordance with the opinion of Rabbi Yosei HaGelili, who maintains that there is a mitzva for the blood redeemer to kill him, nor in accordance with the opinion of Rabbi Akiva, who says the blood redeemer has the option of killing him.

The Gemara answers: Rav states this ruling in accordance with the opinion of that tanna, as it is taught in a baraita that Rabbi Eliezer says with regard to the verse: “And the murderer shall not die, until he stands before the congregation for judgment” (Numbers 35:12): Why must the verse state this? It is necessary since it is stated: “And the blood redeemer finds him…and the blood redeemer murders the murderer” (Numbers 35:27). One might have thought that the blood redeemer may murderer him immediately; therefore, the verse states: “Until he stands before the congregation for judgment,” from which it is derived that the blood redeemer may kill the murderer after only he is convicted in court. Mar Zutra bar Toviyya says that Rav says that the blood redeemer is liable if he kills the murderer before he is convicted.

The Gemara asks: And as for Rabbi Yosei HaGelili and Rabbi Akiva, with regard to this verse: “Until he stands before the congregation for judgment,” what do they derive from it? The Gemara answers: That verse is necessary for that which is taught in a baraita that Rabbi Akiva says: From where is it derived in the case of a Sanhedrin that saw one kill a person that they may not execute him until he stands trial in a different court? It is derived from a verse, as the verse states: “Until he stands before the congregation for judgment,” meaning: Until he stands before a different court. Since they themselves witnessed the murder, they are no longer capable of considering the possibility that he may be innocent.

The Sages taught in a baraita with regard to the verse: “And if the murderer emerges [yatzo yetze] …and the blood redeemer murders the murderer, he has no blood” (Numbers 35:26–27): I have derived only that the blood redeemer may kill the unintentional murderer if the unintentional murderer emerges from the city intentionally. From where is it derived that the same applies if he emerges unwittingly? It is derived from this verse, as the verse states: “If yatzo yetze”; the doubled form of the verb serves to teach that this halakha applies in any case where the unintentional murderer emerges from the city of refuge.

The Gemara asks: But isn’t it taught in a baraita with regard to an unintentional murderer who emerged from the city of refuge unwittingly: And one who kills him intentionally is executed, and one who kills him unintentionally is exiled? The Gemara answers: This is not difficult, as this second baraita is in accordance with the opinion of the one who says that we say: The Torah spoke in the language of people, and no halakha is derived from the doubled form of the verb: Yatzo yetze, as it is merely a rhetorical flourish, and that first baraita is in accordance with the opinion of the one who says that we do not say: The Torah spoke in the language of people, and the compound verb was employed in order to derive that the blood redeemer may kill the unintentional murderer even if he emerged from the city of refuge unwittingly.

Abaye said: It stands to reason that the halakha is in accordance with the opinion of the one who says: The Torah spoke in the language of people, and the blood redeemer is liable for killing an unintentional murderer who emerged from the city of refuge unwittingly, in order to ensure that the ultimate punishment of the unintentional murderer, when he emerges from the city of refuge, will not be more severe than his initial punishment, when he is sentenced in court. Just as with regard to his initial punishment for murder, if he killed intentionally he is executed, and if he killed unintentionally he is exiled, so too, with regard to his ultimate punishment, if he emerges from the city of refuge intentionally he is killed by the blood redeemer, and if he emerges unwittingly he is returned to exile in the city of refuge.

§ It is taught in one baraita: In the case of a father who killed his son, his surviving son becomes his blood redeemer and may kill him. And it is taught in another baraita: His son does not become his blood redeemer. The Gemara suggests: Let us say that this baraita, which states that his son does become his blood redeemer, is in accordance with the opinion of Rabbi Yosei HaGelili. Since there is a mitzva for the blood redeemer to kill the murderer, this mitzva applies equally to a son. And that baraita, which states that a son does not become a blood redeemer to kill his father, is in accordance with the opinion of Rabbi Akiva, who holds that the blood redeemer has only the option, not a mitzva, to kill the murderer.

The Gemara rejects that understanding. And how can you understand it in that manner? Both according to the one who says that there is a mitzva for the blood redeemer to kill the unintentional murderer and according to the one who says that it is optional, is it permitted for a son to do so? But doesn’t Rabba bar Rav Huna say, and likewise the school of Rabbi Yishmael taught: With regard to all transgressions of the Torah, even if the father is liable to receive lashes or be ostracized, a son does not become an agent of the court to flog his father or to curse him, apart from the case of a father who acted as one who incites others to engage in idol worship, as the Torah states in his regard: “You shall neither spare nor conceal him” (Deuteronomy 13:9)?

Rather, the Gemara suggests that the apparent contradiction between the two baraitot is not difficult, as this baraita, which says that the son does not become a blood redeemer to kill his father, is referring to his son, and that baraita, which says that the son does become a blood redeemer, is referring to the son of his son, who can become a blood redeemer to kill his grandfather, as the grandson is not required to honor his grandfather as he is required to honor his father.

MISHNA: The previous mishna teaches that the halakhic status of the outskirts of the city is like that of the city itself in terms of the unintentional murderer being provided refuge there. The mishna adds: With regard to a tree that stands within the Shabbat boundary of a city of refuge, whose boughs extend outside the boundary, or a tree that stands outside the boundary and its boughs extend inside the boundary, the status of the tree, whether it is considered inside or outside the boundary, in all cases follows the boughs.

GEMARA: And the Gemara raises a contradiction from a mishna (Ma’aser Sheni 3:7) taught with regard to second tithe. Second-tithe produce must be consumed within Jerusalem or redeemed outside of Jerusalem: With regard to a tree that stands within Jerusalem, and whose boughs extend outside the city wall, or a tree that stands outside the city wall and whose boughs extend inside the wall, the principle is: The halakhic status of any part of the tree that is above the wall and inward is that of an area within the wall and the halakhic status of any part of the tree that is above the wall and outward is that of an area outside the wall. Apparently, the trunk does not follow the boughs and the boughs do not follow the trunk. The status of each part of the tree is determined by its position relative to the wall.

The Gemara rejects the parallel between the cases. Are you raising a contradiction between the halakha of second tithe and the halakha of cities of refuge? With regard to the halakha of second tithe, the Merciful One made the status of the tree dependent on the wall, and with regard to cities of refuge, the Merciful One made the status of the tree dependent on dwelling. One can dwell in its boughs, but one cannot dwell in its trunk. Therefore, with regard to cities of refuge, the halakhic status of the tree is determined by the boughs.

And the Gemara raises a contradiction between the previous baraita with regard to the halakha of second tithe and another baraita with regard to the halakha of second tithe, as it is taught in a baraita (see Ma’asrot 3:10): In Jerusalem, follow the boughs in determining the status of the tree with regard to consuming second-tithe produce, and likewise, with regard to cities of refuge, follow the boughs in determining the status of the tree with regard to providing refuge for an unintentional murderer. Rav Kahana said: This apparent contradiction is not difficult, as this baraita is in accordance with the opinion of Rabbi Yehuda, and that baraita is in accordance with the opinion of the Rabbis. As it is taught in a baraita with regard to partaking of second tithe in Jerusalem: