Literary critic Adam Kirsch is reading a page of Talmud a day, along with Jews around the world.

For a man to betroth a woman, we have learned in Tractate Kiddushin, requires two things: a declaration of intent, and the transfer of money or an item of value worth at least one peruta. This seems straightforward enough, but as always in the Talmud, the rabbis are alert to possible complications. There are many ways that betrothal can go wrong, due to a flaw in either the declaration or the purchase. This week’s Daf Yomi reading explored a number of these possible errors and in doing so entered into fairly remote regions of the law. Because halakha is one vast system, you can easily start out talking about betrothal and end up talking about tithing or agricultural laws.

At the time the Talmud was written, Jews were polygamous, which meant that it was possible for a man to betroth more than one woman at a time. However, it was against biblical law for a man to marry two sisters or a mother and a daughter. (These prohibitions were discussed at length in Tractate Yevamot, when the rabbis laid down the complex rules governing levirate marriage.) Say, then, that a man betroths two sisters at the same time. How does the law deal with his action? Is he effectively betrothed to one of the sisters but not the other, and if so, how can you tell which one? Or is the betrothal entirely invalid?

Or, perhaps, the man can actually be betrothed to both sisters, even though this is against the law? In a secular legal system, this ought to be impossible: The law could simply declare the man’s actions null and void, just as, in America, a marriage ceremony performed by a man who is already married has no legal standing. But in the Bible, Rava observes, the law against marrying two sisters is followed by the threat, “Whoever shall do any of these abominations, even the souls that do them shall be cut off from among their people.” This is karet, the cutting-off of the soul, which is one of the most terrible punishments in Jewish law. But it makes sense to punish such a forbidden betrothal only if that betrothal is a meaningful action. As Rava says in Kiddushin 50b, “if betrothal does not take effect with her, what renders him liable to receive karet?” Evidently the double betrothal is a real action with real effects, or else it wouldn’t deserve a real punishment.

So, is a forbidden betrothal a betrothal or not? This is the subject of a long dispute between Rava and Abaye, two of the most important amoraim in the Talmud, who were friends and frequent debate partners. What is really at stake, they say, is the status of a betrothal, which, if consummated sexually, would involve the partners in a sin. Such a relationship is what the Gemara calls “betrothal that is not given to consummation.” In a sense, such a betrothal could be compared to a contract committing the participants to commit a crime. Can such a contract be enforced, or is it ipso facto invalid?

The sages disagree about its status: “Abaye says it is betrothal, Rava says it is not betrothal.” This dispute, as recorded in the following pages of Gemara, is a classic example of a Talmudic argument, drawing on biblical interpretation and analogies with other areas of halakha; so it is worth following step by step. First, Rava supports his position by citing a verse from Deuteronomy: “When a man takes a woman and engages in sexual intercourse with her.” The conjunction of taking and intercourse here suggests that betrothal, which as we have seen is legally a form of acquisition, is essentially linked with sex. It follows that a man cannot “take” a woman with whom he is legally forbidden to have sexual relations.

The Gemara counters by raising an important technical distinction. If a man says to a pair of sisters, “You are both betrothed to me,” then according to the mishna his words are invalid and he is betrothed to neither. If, however, he says, “One of you is betrothed to me,” then he is in a legal limbo: He is in fact betrothed to one of the sisters, but it is impossible to know which one. As a result, he cannot have intercourse with either of them, because he may theoretically end up sleeping with the one who is his sister-in-law, not the one who is his wife. In other words, the man has entered into a betrothal that is not given to consummation—something that Rava said was impossible. Evidently, Rava was wrong and Abaye was correct.

But Rava has a counterargument prepared. The mishna tells of an episode in which a man betrothed a group of five women by giving them a basket of figs, and among the five were a pair of sisters. According to the mishna itself, the man was effectively betrothed to the three other women, but not to either of the two sisters. Apparently, then, the law agrees with Rava: A betrothal that is not given to consummation is not a betrothal. However, what if the man did not intend to betroth all the women, but just one of them? In that case, he would be partly betrothed to both sisters and would have to divorce both of them in order to regularize his situation. Read this way, the case supports Abaye’s stance, that a betrothal not given to consummation does have legal standing.

The Gemara now goes on to raise a series of challenges to Rava, each of which he parries. (As always in these exchanges, we are not necessarily reading a historical record of a real debate: rather, later editors and commentators are imagining what Rava could have said in response to various objections.) Say, for instance, that a man has “two groups of daughters from two women,” and he announces that he is betrothing “my elder daughter” to someone. Does he mean the elder daughter of the first group , who is older than all her full and half-siblings, or the elder daughter of the second group, who is older than all her full siblings but younger than her half-siblings? Or, indeed, could he mean any of the daughters of the first group, since all of them are older than all of their half-siblings?

Because of this ambiguity, Rabbi Meir says that all the daughters—except the very youngest—requires a divorce, in order to clearly establish that they are not engaged. But if they require a divorce, this could only be because the betrothal was effective. This is true even though it was a betrothal not given to consummation, since it made all of the sisters sexually prohibited to the groom. This means that the law is on Abaye’s side against Rava after all.

Or is it? What if, the Gemara replies, the case is understood differently: The father actually did mean to betroth a specific one among his daughters, but he then forgot which one. In this case, the initial engagement was a valid one, and it is only subsequently that it became confused. This means that Rava could still be right about betrothals not given to consummation, since the betrothal in this instance did not belong in that category at all. But finally, the pro-Abaye side plays its trump card. This is yet another case in which a father of five sons met a father of five daughters and declared, “One of your daughters is betrothed to one of my sons.” In this case, it is clear that the identity of the son and daughter was never established; yet here again, all five daughters require bills of divorce from all five sons. This can only be because a betrothal not given to consummation is a valid betrothal; therefore Abaye is right and Rava is wrong, QED. Only a reader who can find the interest, even the excitement, in this kind of dialogue will be able to savor the true pleasure of reading the Talmud.


To read Tablet’s complete archive of Daf Yomi Talmud study, click here.