The agreement of al-Madina had weight with Malik ibn Anas. The agreement of many divines and legists always had weight of a kind. Among lawyers, a principle, to the contrary of which the memory of man ran not, had been determining. But this was wider, and from this time on the unity of Islam was assured. The evident voice of the People of Muhammad was to be the voice of God. Yet this principle, if full of hope and value for the future, involved the canonists of the time in no small difficulties. Was it conceivable that the agreement could override the usage of the Prophet? Evidently not. There must, then, they argued, once have existed some tradition to the same effect as the agreement, although it had now been lost. Some such lost authority must be presupposed. This can remind us of nothing so much as of the theory of the inerrant but lost original of the Scriptures. And it had the fate of that theory. The weight of necessity forced aside any such trifling and the position was frankly admitted that the agreement of the community was a safer and more certain basis than traditions from the Prophet. Traditions were alleged to that effect. "My People will never agree in an error," declared Muhammad, or, at least, the later church made him so declare.

But ash-Shafi'i found that even the addition of agreement to Qur'an and Prophetic usage did not give him basis enough for his system. Opinion he utterly rejected; the preference of Abu Hanifa and the conception of the common welfare of Malik ibn Anas were alike to him. It is true also that both had been practically saved under agreement. But

he held fast by analogy, whether based on the Qur'an or on the usage of the Prophet. It was an essential instrument for his purpose. As was said, "The laws of the Qur'an and of the usage are limited; the possible cases are unlimited; that which is unlimited can never be contained in that which is limited." But in ash-Shafi'i's use of analogy there is a distinction to be observed. In seeking to establish a parallelism between a case that has arisen and a rule in the Qur'an or usage, which is similar in some points but not precisely parallel, are we to look to external points of resemblance, or may we go further and seek to determine the reason (illa) lying behind the rule and from that draw our analogy? The point seems simple enough and the early speculative jurists sought the reason. For that they were promptly attacked by the traditionists. Such a method was an attempt to look into the mysteries of God, they were told; man has no business to inquire after reasons, all he has to do is to obey. The point thus raised was fought over for centuries and schools are classified according to their attitude toward it. The position of ash-Shafi'i seems to have been that the reason for a command was to be considered in drawing an analogy, but that there must be some clear guide, in the text itself, pointing to the reason. He thus left himself free to consider the causes of the divine commands and yet produced the appearance of avoiding any irreverence or impiety in doing so.

Such then are the four sources or bases (asls) of jurisprudence as accepted and defined by ash-Shafi'i—Qur'an, prophetic usage, analogy, agreement. The