This week’s summary is written by Yoni Zolty and Yakov Ellenbogen
In several places, the Gemara suggests using the claim of “adata dehachi” as a way of retroactively disqualifying a marriage or divorce contract (gittin). Thus, for instance, a husband can claim that when he divorced his wife, the divorce rested on certain conditions he had assumed at that time. While those conditions were unstated at the time of the divorce, the husband now claims that because they failed to materialize the divorce should retroactively be invalidated.
The claim “adata dehachi” allows marriage contracts to conform to the expectations of the respective partners. Marriage partners can now be assured that the legal agreements they form–in this case, a divorce contract–will take into account their conditions and reservations. Thus, for instance, the Gemara in Yevamot 65a presents a case of a man who divorces his wife because he believes that his wife is barren, but would prefer to remain married otherwise. Although, he may not have stated explicitly his assumption and qualification to this divorce document, the claim of “adata dehachi”–that he would not have divorced her had he known that she was able to reproduce–provides him a legal mechanism of expressing his regret and retroactively annulling the divorce. Thus, on the one hand, the advantage to the claim of “adata dehachi” is its ability to allow one to express his unstated qualifications for a divorce document.
However, these qualifications were unstated at the time of the divorce. They may have been implicit and obvious to the husband, but they were not clear to his wife or to the Beit Din. This raises several critical issues. First, there is an epistemological problem–how do we determine what were these implicit conditions? We obviously can’t just take the husband at his word, so how do establish these qualifications? The rishonim discuss determining a person’s implicit conditions based on a presumption of the husband’s motives for the divorce and debate in which scenarios these presumptions are justified; i.e. what type of umdanas (established guesses) can be used?
Moreover, the bigger issue in the sugya is that claims raised at a later date which would retroactively annul a divorce, may have disastrous consequences for the woman and her potential children from a second marriage. The Gemara, astounded with the possible implications, asks whether her children would be mamzeirim (bastards)?
Thus, these two opposite perspectives to the claim of “adata dehachi”–its potential benefits and disadvantages–affect the borders and limits to the claim and determine the situations in which the Gemara, and subsequently, the rishonim, are willing to allow the claim.
These issues may likewise determine whether the claim of “adata dehachi” can be used in other scenarios besides marriage contracts. Can we extend this legal principle to general contract law? In general, the sugyot try limiting the use of “adata dehachi” in gittin as much as possible. The question is whether these limitations would also extend to contract law where there may be more advantages to not requiring all tena’im (conditions) to be stated explicitly and the potential negative consequences may not be as calamitous.
The sugya surrounds four scenarios where a husband divorced his wife and would like to annul the get after discovering that he gave it on false premises:
A. His wife is barren and he later discovers that she is capable of childbirth (Yevamot 65a).
B. His wife is incapable of having regular menstrual periods and she later begins to have regular periods (Nidah 12b).
C. His wife has a “shem ra”–a bad reputation, and he later discovers that her shem ra was baseless (Gitin 45b).
D. He believes his wife is an aiylonit (incapable of undergoing puberty) and she later undergoes puberty
The gemara presents different ways of preventing retroactive annulment of divorce for each of these cases, and it is not clear why these differences need to exist. The rishonim struggle to explain these distinctions and devote much ink explaining why the gemara gave different explanations as to why the divorce remains valid in each situation.
One useful tactic to better understand the distinctions between the above cases suggested by the rishonim is to create a useful theoretical framework to analyze the cases. Rabbi Klapper suggested that there are three ways to categorize information and when one has access to it:
- The information is true now and accessible, but you personally are not aware of it
- The information is true now but is currently inaccessible and can only be accessed at a later time.
- The information is not true now, but will be true later.
Thus, for instance, in the case of the woman who after the divorce begins to have regular periods, the information that she will have regular periods in the future is both inaccessible at the time of the divorce, and is also nonexistent (it will only become true in the future). It therefore belongs properly within category (3).
In contrast, category (2) occurs when the information was true at the moment of the divorce, but was simply inaccessible and unknowable. Thus, if a woman is considered to be infertile but after a divorce has children, this new information was always true (the woman was always fertile), but simply unknowable at the time of divorce.
In category (1) the knowledge was accessible and existent at the time of the divorce, but the husband was not aware of them; for instance, in the case of a husband who does not wish to investigate whether his wife was an aiylonit. The information was accessible–he could have investigated whether she could undergo puberty, he just didn’t put the effort necessary to find out.
This conceptual framework can be used to understand the chilukim suggested by Tosfot (Yevamot 65a s.v. iy ihi) between the cases. In the first case, Tosfot wishes to distinguish between the case where the husband assumed that his wife was incapable of bearing children (case A) and the case where the husband assumes his wife is an aiylonit (case D).
וי”ל דהתם … כיון דלא חשיב לבדוק יפה בסימני אילונית גמר בדעתו לגרש בכל ענין … אבל הכא אין בידו לבדוק
Tosfot suggests that in the case of an aiylonit the husband could have investigated whether she was indeed an aiylonit. Had he really wanted to divorce her only if she remained an aiylonit then he should have first checked to verify that she indeed was an aiylonit before divorcing her. The fact that he didn’t check, implies that he in fact reneged on this qualification and was willing to divorce her irrespective of her status as an aiylonit. Conversely, in the case of the woman who is unable to bear children, because the husband had no way of verifying this information, his implied condition is considered valid and can be used to undo the divorce.
The difference between these two cases can be better appreciated by placing them within the categories developed within our conceptual framework. The case of an aiylonit is the type of information which one could have had right now but was too lazy to check–type (1), while the case of a woman unable to bear children is the type of information which was impossible to ascertain at the time (type 2). Tosfot is effectively arguing that information of the type (1) is considered invalid grounds for retroactively dissolving a divorce, while information of type (2) is a valid reason for retroactively undoing a divorce. Since in case A (that of a woman unable to bear children) the divorce can be retroactively annulled, the gemara must give another reason for why the divorce is not retroactively cancelled, namely that ‘[we claim] she has only now been healed [while at the time of the divorce she was indeed unable to have children]’.
Likewise, Tosfot makes a similar distinction between the case of an woman who is unable to bear children (case A) and a woman who has a bad reputation (case C).
דהכא לא שייך קילקולא דאמרינן השתא הוא דברייתא ובשעה שגירשה לא היתה ראויה אבל בשם רע ונדר ואילונית איכא קלקולא כשנתגלה לבסוף שהשם רע היה שקר ולא היתה אילונית גם מתחלה דאם היתה לא הוה בריאה לעולם ונדר נמי כיון שיכול להפר כאילו לא היה נדר
Tosfot argues that one cannot annul a get in the case of a woman who is unable to bear children because she has undergone a physical change and while she is now capable of giving birth now, she truly was not capable of doing so before the divorce. Conversely, in the case of the woman whose bad reputation is proven to be baseless, even originally the bad reputation was baseless–there was just no way of ascertaining that until the future, and therefore one can annul the get.
Once again, these cases can be placed within our conceptual model. The case of a woman who had undergone a physical change and can now give birth is of type (3). The information that she was able to give birth before the divorce was not only inaccessible but false–she truly was unable to give birth at that time. In contrast, in the case of the woman who is believed to have a bad reputation, the fact that the bad reputation was baseless was always true and therefore is of type (2). Tosfot is thus arguing that information of type (2) is considered valid grounds for divorce while information of type (3) is not. (The reason that the case of the woman who was unable to give birth switched from type 2 to type 3 is because the Gemara changes its assumption about the details of that case. This is related to the gemara’s answer of ‘she has now been healed’ mentioned above.)
Thus, using this conceptual model we were able to explain some of the distinctions that Tosfot makes between the cases. These distinctions are important because they provide guidelines as to when claims of implied conditions are considered valid and for what types of information we allow implied conditions instead of requiring that they be expressed formally and explicitly during a divorce.