There is a particular kind of courage in deciding that the best response to catastrophe is a closer reading of the text. After 70 CE, the Temple in Jerusalem lay in rubble, the sacrificial cult had ended, and the axis around which Jewish life had turned for centuries was simply gone. The rabbis did not abandon the law that had governed that cult. They moved it into a different room - the study house - and began to read it again, verse by verse, more carefully than before.
Two works produced in the land of Israel by about the middle of the third century CE are the clearest evidence of what that reading looked like in practice. The Mekhilta de-Rabbi Yishmael is a running legal and narrative commentary on most of Exodus, from the account of the Exodus itself through the laws of Sinai. The Sifra is an equally dense commentary on Leviticus, the book that had once governed the Temple's priestly rites. Together they represent the genre the rabbis called midrash halakha - exposition of the law - and together they show how a tradition that had lost its central institution converted that loss into intellectual energy.
The record
The Mekhilta de-Rabbi Yishmael is a tannaitic midrash on Exodus, covering large stretches of the book in a structure that alternates legal analysis with narrative homily. Its received attribution links it to the school of Rabbi Yishmael ben Elisha, a major legal authority of the late first and early second century CE, though scholars including Jacob Neusner and H.L. Strack have argued that the text reached its edited form in the third century, likely in the land of Israel. The Sifra - also called Torat Kohanim, "the law of the priests" - is the companion work to Leviticus, and its received attribution connects it to Rabbi Yehudah bar Ilai's school, also of the tannaitic period. The critical edition of the Sifra by Louis Finkelstein (Jewish Theological Seminary of America, 1983) remains the standard scholarly text.
What these texts actually do is remarkable once you slow down to watch it. A verse from Exodus says that a Hebrew slave who loves his master and does not wish to go free "shall serve him for ever" (Exodus 21:6). That single word, "for ever," sets the Mekhilta's interpreters arguing: does it mean literally until the end of time? Clearly not - the jubilee year releases all slaves. So it must mean "until the jubilee." But does that also apply to a Canaanite slave? The argument opens outward, drawing in other verses, other rulings, other tannaitic voices. The text records the disagreement without settling all of it. That is the whole method: any phrase is an invitation to press harder, to ask what the word means, what cases it covers, what limits it implies. The verse does not close. It opens.
The Sifra does the same for Leviticus but with particular intensity around the sacrificial system. Since the offerings described in Leviticus could no longer be brought - the altar was rubble, the priesthood scattered - the Sifra's analysis is by the time of its editing an exercise in careful memory and exacting legal imagination. The rabbis knew the rites in precise technical detail. They argued about which defect in an animal disqualified an offering, which ritual sequence was indispensable and which was not. These were not idle puzzles. They were acts of preservation: if the Temple were ever rebuilt, the law would be ready. And in the meantime, the study of the law was itself counted as a religious act of the first order.
The record
A passage in the Babylonian Talmud (Menachot 110a) cites a tradition in the name of Resh Lakish that "anyone who engages in the study of the order of sacrifices, it is as though he built the Temple and offered its sacrifices." This principle - that Torah study substitutes for and even reconstitutes the Temple service - is the theological engine that drives texts like the Sifra. The Mekhilta is similarly explicit about the weight of textual study: its commentary on Exodus 12:2 (the first commandment to Israel as a people) opens a discussion of whether commandments given before Sinai are binding in the same way as those given at Sinai, with named tannaitic authorities disagreeing. The Soncino translation of the Babylonian Talmud (1935-1948) and Finkelstein's critical edition of the Sifra remain standard reference points alongside Lauterbach's edition of the Mekhilta.
The Mekhilta and Sifra belong to a group of early rabbinic law-commentaries that also includes the Sifre on Numbers and Deuteronomy. Collectively, these texts show the tannaitic movement doing something precise: attaching rabbinic legal tradition to the scriptural text itself, so that the law did not float free of its biblical foundations. The Mishnah, compiled around 200 CE under Rabbi Yehudah ha-Nasi, had organised rabbinic law by topic. The midrash halakha texts work from the scripture outward, following the Torah's own sequence. The two approaches are complementary, not competing. Taken together, they mean that any serious legal question in rabbinic tradition can be approached from both ends - from the organised ruling in the Mishnah, or from the expounded verse in the Mekhilta or Sifra. That double anchor has held for seventeen centuries.
The record
The oldest surviving manuscript witnesses to the Mekhilta de-Rabbi Yishmael date to the medieval period, with important fragments preserved in the Cairo Geniza (discovered 1896; material now held at Cambridge University Library, the Jewish Theological Seminary of America, and other institutions). The standard printed edition of the Mekhilta de-Rabbi Yishmael with English translation is that of Jacob Z. Lauterbach, published by the Jewish Publication Society in three volumes between 1933 and 1935, revised in 1976. Lauterbach's introduction documents the manuscript tradition and the history of printed editions back to the Constantinople printing of 1515, which is the first known printed text of the Mekhilta.
A tradition that can argue about the meaning of a single word for long enough does not need a building to survive in. The Mekhilta and Sifra are not substitutes for the Temple. They are something the Temple could not provide: a form of the law that needs no location, no season, no priesthood, and no altar - only a text, a question, and someone willing to press the question harder than it has been pressed before. That is a durable piece of engineering. It has now outlasted the empire that destroyed the building by more than fifteen centuries.
