AND PERSONAL OPINION 125

of the other. Others regard everything that goes beyond the explicit text of Koran and sunna and is only its equivalent as qiyās. Shāfi'ī considers the conclusion a maiore ad minus 'a binding rule of qiyās' (Tr. VIII, 12), but in most cases where he draws it he does not call it by this name.

The element common to the original and to the parallel case on which a qiyās is based Shāfi'ī calls either informally ma'nā 'idea',¹ or more technically aṣl 'basis';² he does not use the later term 'illa. In the case of organs of the body, this common element is supplied by their common names; for example, the common name 'lip' justifies the award of the same compensation for injuries to the upper and to the lower lip, and Shāfi'ī states explicitly that 'the weregeld is based on names and not on the degree of usefulness'.³ But in another case he avoids reasoning 'based on the similarity of names', because it would lead him into a dilemma.⁴ If a ruling covering two species of a genus is to be extended, by analogy, to another species, it ought to be extended consistently to all species of that genus, or not at all (Ris. 27). The substitute (badal) must be treated in analogy with its original (mubaddal 'anhu) (Ikh. 97).

As a general safeguard against arbitrariness Shāfi'ī insists that analogy must start from the outward and obvious meaning (​​ẓāhir) of the passages on which it is based. This consideration, which corresponds to Shāfi'ī's rule of interpreting traditions according to their outward meaning,⁵ occurs in numerous passages, and is set forth in detail in the first part of Tr. VII (pp. 267-70).⁶ The whole of law, Shāfi'ī points out, is concerned with the forum externum; he proves this from passages in the Koran and from traditions from the Prophet, and gives examples.⁷

We have noticed cases where Shāfi'ī's qiyās falls short of his own