Shevuot 4

On today’s daf, we find the Gemara still engaged in asking meta-questions about the mishnah that opens this tractate. That is, the Gemara is currently interrogating the structure of the mishnah; it will engage with the actual material in more depth later. Yesterday, we discussed why the mishnah contains so much material that has nothing to do with oaths (the primary subject of the tractate) and why it fronts that material. Today, the Gemara asks another meta-question: Whose opinion is represented by the mishnah? Specifically, does it accord with Rabbi Akiva or Rabbi Ishmael (both of whom headed important schools of rabbinic thought in the tannaitic period)? (Spoiler alert: It’s the former.)

In the course of this discussion, also on yesterday’s daf, the Gemara raised a contradiction between Rabbi Yohanan’s position that one cannot be flogged for a transgression that does not contain an action, and the position articulated in our mishnah, which suggests that one can be flogged for failing to fulfill a positive oath — which is usually an inaction. For example: If someone took an oath that they would eat bread the next day and they did not, that person would be flogged, even though this transgression consists merely of a failure to act. Since unattributed teachings are considered more authoritative, this is surprising: How could Rabbi Yohanan rule against an unattributed mishnah?

The answer, says the Gemara, is that Rabbi Yohanan is ruling according to a different unattributed mishnah that aligns with his position. Here’s that mishnah:

If one states: “On my oath I will not eat this loaf,” and states: “On my oath I will not eat it,” and then he ate it, he is liable for only one violation. This is an oath on an utterance for which one is liable to receive lashes for its intentional violation, and to bring a sliding-scale offering for its unwitting violation.

How does this second anonymous mishnah support Rabbi Yohanan’s ruling that one cannot be flogged for a transgression that does not contain an action? The Gemara explains:

It is specifically this case of an oath on an utterance for which one is liable to receive lashes for its intentional violation. But if a person stated: “On my oath I will eat,” and then he did not eat, he is not flogged.

Since the mishnah used limiting language — “this is an oath on an utterance for which one is liable to receive lashes” — the implication is that it comes to exclude another case: that in which a person violated an oath through inaction as opposed to action. In such a case, as Rabbi Yohanan holds, the person would be exempt from lashes.

Having found a different unattributed mishnah that supports Rabbi Yohanan’s view, the Gemara asks a logical follow-up:

What did Rabbi Yohanan see that he practiced in accordance with this unattributed mishnah? Let him instead practice in accordance with that unattributed mishnah.

The Gemara points out that Rabbi Yohanan’s ruling seems arbitrary. One supports his position, and one mishnah does not. Why did he choose to rule like the later mishnah, if both of them are unattributed (and therefore equally authoritative)? The Gemara brings an even thornier follow-up question:

And according to your reasoning, then with regard to Rabbi Yehuda HaNasi himself, how could he teach us as unattributed both this opinion in the mishnah here and that opinion in the mishnah there?

Why are there two contradictory mishnahs at all? Our frustration with Rabbi Yohanan is misplaced: The real challenge should be against Rabbi Yehuda HaNasi, the compiler of the Mishnah, who chose to include two unattributed rulings that contradict one another! Luckily, the Gemara resolves both problems:

Rather, it is apparent that initially Rabbi Yehuda HaNasi held that for a violation of a prohibition that does not involve an action one is flogged, and so he taught it as an unattributed mishnah. Later, he retracted his opinion and held that one is not flogged for a violation of such a prohibition, and so he taught that opinion as an unattributed mishnah. And he left the first mishnah as it was because a mishnah does not move from its place.

According to the Gemara, the first unattributed mishnah reflects Rabbi Yehuda HaNasi’s initial opinion. The second unattributed mishnah reflects his ultimate conclusion, after a change of heart. And while this may seem unremarkable, the Gemara is in fact suggesting a fascinating principle: The mishnah, as we have it, may indeed include rulings that were later reversed. But they were not erased or changed, or even marked as mistakes. Instead, they were kept because, “a mishnah does not move from its place.”

If we study the Mishnah, we find that there are in fact several times when an original tradition was later reversed, and the Mishnah says as much. But the notion that such reversals could be contained within our text without any indication is far more radical. We might be tempted to conclude that this is a post-facto way of explaining contradictions the editors of the Mishnah did not discover. However, Rashi takes a different tact, suggesting, quite simply, that once a tradition spread amongst Torah learners and was repeated by them endlessly within the beit midrash, it was nearly impossible to eradicate that tradition from learners’ rote memories. So it was left in by editors who had faith that “the wise person will realize the later teaching is the essential one and the first had been retracted.” It’s a gesture of enormous respect toward the learner of this material.

Read all of Shevuot 4 on Sefaria.

This piece originally appeared in a My Jewish Learning Daf Yomi email newsletter sent on May 5, 2025. If you are interested in receiving the newsletter, sign up here.