Midrash, Mishnah, and Gemara : the Jewish predilection for justified law

From the Introduction...

This book was written at a time when my critical commentary on the Talmud (called Sources and Traditions) had achieved half its goal: it covered half of the Talmud text. That commentary discusses in great detail more than a thousand different subjects and touches fleetingly in the notes on ten times that number. It avoids being disjointed because it follows the order of the Talmud and thus has a consecutive thread. It is further united by the several very important historical facts it implies, principally that the present text of the Talmud most often evolved from a different preceding text, and that in the process of evolution the present text absorbed both transmissional changes and redactional changes.

I became very interested in these redactional changes. Transmissional changes enter the text without the transmitter's awareness. In contrast, redactional changes are consciously made for the sake of improving the text, either contextually or aesthetically. Transmissional changes are understandable, though unpredictable. They are mechanical changes, made unwittingly by the transmitter. A person, for instance, may genuinely think he heard the word "can" and transmit it that way, whereas in fact the word "can't" was said. Not all mechanical changes are a result of faulty hearing; they may also result from faulty speech. The speaker may think he said "can't," but the word he actually spoke was "can." Transmissional changes are simply a part of human susceptibility to error. Redactional changes, on the other hand, are made purposefully by the redactors. When the purpose of these changes is to improve content or correct defects, the question arises: who is responsible for these defects? Did the original authors release defective texts? This is most unlikely; more plausibly, the texts became defective during the interval between the time of the authors and the time of the redactors.