It is commonplace in the Jewish law of marriage and divorce, as well as Jewish commercial law, that many transactions can be effected by an appointed agent, who acts on behalf of the principal party. Betrothal and divorce may be done directly by the man and woman involved, or either may be represented by an agent who acts on their behalf.
In Chapter 2 of Kiddushin, the rabbis seek to find a basis for the principle of “shelihut” (agency) in the written Torah. They adduce three possible sources:
(1) In the law of the Paschal sacrifice, the Torah stipulates that a company shall collaborate, obtaining a single lamb which is slaughtered, roasted and eaten by all members communally. But the act of slaughtering can only be performed by an individual. In such a case, the individual is therefore acting on behalf of all members of the group, and they are performing the “mitzvah” of slaughtering by implicitly designating him as their agent.
(2) In the law of Terumah, every farmer is responsible for designating a certain portion of his produce as a gift to the priests. But a landowner may (according to common practice, codified in rabbinic law) assign one of his servants or workers to separate the “terumah” from the crop. This servant or worker is then operating as an agent on behalf of the landowner. The rabbis deduce this provision from the text of the law: “You—even you (gam attem)—shall raise up from your produce a portion and give it to the priest.” The redundant words gam attem are interpreted to refer to doing it through an agent.
(3) In the law of divorce, the Torah uses the word ve-shillach in designating the act by which the husband “sends forth” his wife to be free, no longer married to him. It also uses the term ve-shillechah (“and he shall send her forth”), where the final “hei” with the aspirative mappik indicates grammatically the direct object of the verb (“her”). But one can also creatively read the word ve-shillecha (without the mappik), whereby the final “hei” would refer to a feminine subject (“and she shall send”). Though this is not the standard reading of the text, the rabbis nevertheless derive from this the principle that the woman, as well as the man, is entitled to act through appointing an agent to act on her behalf. (Indeed, taking the term ve-shillach meaning "he shall send forth [his wife to be free]" in the sense of "he shall send [an agent to perform the handing-over of the get on his behalf]" is another act of creative interpretation that would send a peshat reader of the text into a panic!)
The second and third arguments are representative of a kind of rabbinic argumentation that bothered me greatly when I was young, and stood in the way of my accepting rabbinic Judaism in its standard, received form as authoritative at that point in my development. How is one to know, in the case of a textual redundancy, what “additional” legal stipulation is to be taken as implied by the text? Maybe (in the case mentioned) gam attem is to be understood as referring to some other member of the household? Maybe it is emphatic, and meant to restrict who can perform the action (“you, yes you, nobody else”)? If there is a secret meaning not expressed directly by the words, how on earth can anyone presume to know what that meaning is? Many secret meanings are possible, not all of them in accord with rabbinic law!
And the third case was even more offensive to my critical young mind. Any word can be creatively read in any number of possible ways to suggest other meanings. But this is implicit in language itself. If such creative reading is permitted, nobody can ever speak or write unambiguously to rule out such readings if they are not intended. But then no text is determinate in its meaning. If God wanted to say X and not Y, then given this feature of language, it would be impossible for Him to do so, for some rabbi would take it in his head to read Y into the text anyway, and the divine intention would be frustrated.
The first argument is far more plausible. The law of the Paschal lamb clearly implies that several people will eat together, but only one will perform the crucial act of slaughtering on behalf of the group. Agency is built into the situation. There is indeed an additional step of generalization (called binyan av in rabbinic methodology) in inferring that if agency is allowed here, it is allowed in other situations as well. But this is not nearly so offensive to reason as the procedures in #2 and #3.
I will speak generally of “paralogic” as referring to arguments of Type #2 and Type #3, where the rabbis use procedures of inference that would not be allowed in normal logic.
In Heschel’s book Heavenly Torah, which I helped to translate, he distinguishes between the methods of Rabbi Ishmael and Rabbi Akiva in interpreting the laws of the Torah. #1 is characteristic of Rabbi Ishmael’s method. #2-3 are characteristic of Rabbi Akiva’s.
This whole argument is related to the question of whether the law of agency is to be regarded as de-oraita (legislated by the written Torah, and therefore attributable to God directly) or de-rabbanan (legislated by the rabbis). All three arguments seek to prove that the law of agency is de-oraita, and is therefore a fundamental, unalterable category of Jewish law.
I must mention at this point a book by Jay Harris: How Do We Know This? The central argument of his book is that the rabbis of the Talmudic period used “paralogical” methods as part of the organic style of their thought, without it occurring to them that such interpretation of text was unnatural. The medieval commentators were more logical in their approach. Especially the peshat (plain-sense) commentators of the Torah, such as Abraham Ibn Ezra and Rashbam, would say in a case such as this that the rabbis made their own laws, but used paralogical arguments from the written Torah text as asmakhta (casual support). By doing so, they were performing a symbolic gesture, affirming that though the law evolves through addition, it is to be regarded as a single legal tradition, and the later developments have their roots in the ancient precedents.
If Jay Harris’s book had been available when I was younger, it might have spared me some of my agonizing and ambivalence toward the tradition.
I have another angle, however, to contribute to the issue just presented. There is a point of view from which such basic laws as that of “agency” can be regarded as de-oraita despite the spuriousness of the rabbinic “proofs” offered for them. Talmudic law knows of a category of laws derived from sevara (reason or common sense). The 19th-century Talmudist Zvi Chajes (in The Student's Guide Through the Talmud [Mebo Ha-Talmud, East and West Library, London, 1952], Chapter 4) argued that laws based on sevara were on a par with laws specified in the Written Torah. They are thus similar (though not identical) to those in the category halakha le-Moshe mi-Sinai. But whereas those in the category halakha le-Moshe mi-Sinai are often arbitrary (example: tefillin should be black in square boxes), those based on sevara are so common-sensical that the alternative is less reasonable by comparison. (Example: “the burden of proof falls on the claimant.”)
This is not to say that a sevara rule is so universal that no society is thinkable without it. In the case of the rule of agency, the Encyclopedia Judaica [article: "Agency"] maintains that Talmudic law was in advance of contemporary Roman law. Whether a particular transaction may be done through an agent is a specific, contingent fact of the legal custom of a given society. But any society, at any point in time, must have an accepted practice in such matters, whether X is customarily allowed or not. If effecting commercial transactions through agents was accepted practice by the 2nd century, it did not require a specific rabbinic enactment to bring it about. Once the conveniences of such a practice are established, it is hard to turn the clock back and disallow it. It then becomes, naturally, part of the accepted fabric of social practice and thus naturally gets ratified when statutory law is codified (as it was in the Mishnah).
But from an intellectual standpoint, rabbinic thought strove for unification. The Written Torah was considered to be the primary authoritative legal document. Wherever possible, accepted legal practices ought to be justified by “finding” their sources in the written text. (The Talmudic question “Minayin? From whence?” is the source of the title of Harris’s book How Do We Know This?”)
If we were to write a “Logic” of the Talmud, based on this argumentative practice, it would have to be along the following lines:
- Whatever is written in the Torah, is law.
- Whatever can be derived from the written Torah by strictly logical methods, is law.
- Whatever can be derived from the written Torah by “paralogical” methods is not necessarily law. (How could it be? This method proves too much!) However, paralogical methods are a permitted move in order to justify a law that has independent authority on other grounds. (“Other grounds” can be: a Mishnaic statement, or sevara, or universal common practice.) Typically, rabbis will disagree on the derivation of a law from the Written Torah, while agreeing on the basic law itself (though the difference of derivation may be connected with differences on detail of understanding the basic law).
One more caveat: The rabbis did not tell us everything that was on their minds. We may be rightly suspicious of the presentation of a law, whose only support is a “paralogical” derivation. A good rule of thumb in such cases is: Accept the paralogical derivation as a decorative embellishment, and ask: What could have been the real reason for this law?
The rabbis were no fools. They knew what they were doing. They were well aware that the paralogical methods, pushed to their extremes, could be abused. But those methods gave them freedom, to make their best considered judgment of what was necessary and just, given the social reality of their time, and present that judgment, cloaked in the time-honored garb of interpretation of the Written Torah.