TAWATUR MA'NAWl

To stress the crucial importance of multiple transmission must not imply that individual instances of reporting are of no signif­icance. Indeed, there is a fundamental epistemological value in solitary reports.

article). The immediate knowledge which the tawatur engenders in the intellect eliminates, as we have already seen, any possibility of inference, thus connecting the original mahsus with the comprehension and sense-perception of the hearer. Therefore, when one hears a mutawatir number of identical reports transmitted, the knowledge that accumulates therefrom carries with it the actual original experience, almost as if it were the direct experience of the hearer himself. In tawatur, knowledge of past events stands on the same footing with particular sensory experiences, such as seeing a bird fly or experiencing pain when your Anger touches Are. See Abu Ishaq al-ShirazT, APTabsirah ft usul al-fiqh, ed. M. H. Haytu (Damascus: Dar al-Fikr, 1980), pp. 291, 11.4-6, 293, 1.6. The difference between empirical experiences and hearing a report may perhaps lie in the designation of certain senses as means of assimilating tawatur knowledge (e.g., hearing), while assigning to other senses the function of attaining tajrtbi knowledge (e.g., smell). Fakhr al-Dm al-RazT, Mahsul, 1.1:100-101, observes that knowledge obtaining through rational faculties along with the sense of hearing is tawatur, but that which occurs through rational faculties along with other means of sense-perception is inductive (tajribt). Hasan ibn Zayn al-Din al-Shahld al-Thani, Ma‘alim al-usul, ed.

⁴⁴ SunnT jurists are unanimous with regard to the zannt status of ahad reports. The statistical probability of zann is certainly above half. According to GhazalT, zann occurs when the soul is inclined to believe something but senses the existence of its opposite (“wa-yakunu lil-nafsi sukunun ila al-shay’i wa-al-tasdlqu bihi wa-hiya tash‘uru binaqidihi... wa-hadha yusamma zannan”). See his Mustasfa, 1:44, 11.3-4. Therefore, the lowest degree of probability in zann is greater than 0.5. A degree of probability equal to 0.5 is termed shakk. Jurists also recognized degrees of probability higher than 0.5 but lower than 1.0, which they called according to an ascending order different transmitters support a particular point or issue, their probability together increases to a significant degree. That is, if we assume that two solitary reports possess in common a given theme, and the probability that each report being true is, say, 0.51, the aggregate probability of their being true is increased to a degree significantly higher than 0.51 but lower than 1.0. When a greater number of ahad reports, all having independent chan­nels of transmission, possess in common one theme, the knowledge of this theme becomes conclusive as well as immediate.

Although the justification of al-tawdtur al-ma‘nawi ultimately reverts to the same principles which underly al-tawatur al-lafzi, those Muslim jurists who admitted the authoritative character of the former (and they were the great majority) seem to have stressed the element of inductive support as the cornerstone of its validity. The emphasis here is placed on the differences among the reports insofar as the chains of transmission and the verbal contents are concerned. Given the common theme that they con­vey, these reports cannot but enhance the argument that this common theme is unquestionably true. The degree of proba­bility attached to them individually is immediately eliminated once they are grouped together as one aggregate.^[248] In other words, the possibility that these reports are separately false is soon dismissed when, taken all together, they testify uniformly to a particular issue. An argument in favor of the certainty yielded by such corroborative evidence is the reputed generosity of Hatim, about whom there is no single mutawatir report. The multitude of the solitary reports which testify to his generosity, INDUCTIVE CORROBORATION, PROBABILITY, AND CERTAINTY 21 however, collectively engender in us an immediate, conclusive knowledge.⁴⁷

The process by which knowledge obtains in the mind through al-tawdtur al-ma‘nawz is similar to that which obtains through al-tawdtur al-lafzz. Without being aware of the actual process of reporting, the intellect accumulates knowledge until the point when it becomes entirely certain of the information reported. The process is purely corroborative and cumulative. It is likened by jurists to drops of water or small pieces of bread; individually they are insufficient, but when they are continuously consumed they will eventually quench the thirst and fill the stomach.⁴⁸ An outstanding example of the conclusive corroboration and complete confirmation that al-tawdtur ahna^awl lends to what may first appear as a probable issue is the authoritative basis of Sunni con­sensus.

TAWATUR AND CONSENSUS

The central role that the corroboration of multiple transmis­sions plays in bringing about certainty is obvious in the theory of

⁴⁷ Mu waffaq al-Dm Ibn Qudamah, Tahrim al-nazar fi kutub ahi al-kalam, ed. and trans. George Makdisi (London: Luzac, 1962), p. 39, par. 68 (English trans, pp. 26­27); AmidT, Ihkam, 1:164, 232-233, 237-238. See another example given in my article “On the Authoritativeness of Sunni Consensus,” International Journal of Middle East Studies 18 (1986): 445.

⁴⁸ Qarafi, Sharh, p. 350; SarakhsT, Usui, 1:295.

⁴⁹ For a detailed analysis of this problem see my “On the Authoritativeness of Sunni Consensus,” pp. 427-454, especially pp. 441-450. Also see Ibn Qudamah, Rawdah, pp. 118-119; QarafT, Shark, pp. 338-339. the mashhur report, a report which originated as ahadi but ac­quired the status of tawatur subsequent to the demise of the Com­panions and the Followers.⁵⁰ This report yields certain, though mediate knowledge. The assumption that the Companions and the Followers could not, by virtue of having lived in so early and pristine a phase of Islam, have lied or conspired on a forgery, pre­cludes the possibility that this ahadi report is probable.

Despite their full awareness of the probabilistic nature of the early stages of the mashhur's transmission, the jurists who up­held this theory insisted on the impossibility of error or forgery when a sufficient—mutawatir—number of people were involved in its later stages. The mere fulfillment of the condition of tawatur number could, one is compelled to conclude, ensure infallibility and, consequently, certitude. Tawatur number as such can there­fore be used interchangeably with consensus. Indeed, commenting on the mashhur report, ‘Ala’ al-DTn al-Samarqandl observes that “the mashhur tradition is that which the scholars [of the age] have favorably received, so that there exists a consensus of the people

⁸⁰ For what seems to have been the beginnings of the notion of mashhur, and for a highly interesting but unrepresentative view of tawatur, see the early treatise of Hasan ibn Sahl ibn Ghalib, Al-Tasdiq, with a French translation by Marie Bernand- Baladi, “Des criteres de la certitude: un opuscule de Hasan Ibn Sahl sur la credibilite du dire transmis par un grand nombre,” Journal Asiatique 257 (1969): 102-138. The translator’s remarks in the introduction (pp. 98-100) to the text concerning the circularity of hadlth, and consensus should be read with great caution.

⁵¹ SarakhsT, Usui, 1:291-292; Nasafi, Kashf al-asrar, 2:6-7. While the majority of the jurists who recognized the category of mashhur held that the knowledge resulting from this report is certain though mediate, a minority, such as ‘Isa ibn A ban, argued that such report can yield only a high degree of probable knowledge (tuma'ninah).

of the age to accept it. Its status is thus that of consensus, and since the latter imposes certain knowledge, so does the former.”^[249]

In explaining the rationale behind the certitude of consen­sus, SarakhsT insists on the significance of the number of peo­ple involved. Attaining certainty on the level of the community is tantamount to a combination of a great number of degrees of probability. Identical instances which are probable will conjoin to create certainty. In the section dealing with consensus he asserts that “what is established in sensory and legal matters by an aggre­gate of things (ijtima‘) cannot be established by individual things severally (infirad). Individuals cannot separately carry a heavy log; but if they group together they will be able to do so.... [This principle] we have already expounded when we dealt with the mutawatir."^[250] The very principle of tawdtur then constitutes the rationale for the certitude of consensus.^[251] The association of con­sensus with tawdtur corned out quite clearly in the debate about the number of qualified jurists (mujtahids) who make up consen­sus. While such legists as ‘Abd al-Wahhab al-Subki refuse to accept as consensus any agreement from which even a less than mutawatir number of mujtahids are absent, some Malikis and early Mu‘tazilTs accept any consensus as valid as long as the number of mujtahids who do not partake in it is less than mutawatir.^[252] In either case, a mutawdtir number of mujtahids opposing consensus will certainly nullify that consensus.

If consensus grounded in tawdtur acquires legitimacy, then a consensus in which all the scholars of the community partake will, a fortiori, lead to certitude. In such a consensus the enumera­tion of probable instances goes beyond the already conclusive mu­tawdtir to encompass each and every instance, thereby reaching a conclusion squarely based upon summative induction. Each con­firming instance, represented by a legal opinion of an independent mujtahid, is a member of a class. The mujtahid upholds an opinion which is admittedly probable. But in reaching this opinion, he, as an acknowledged authority on the Law, establishes himself as a self-contained agent; that is, he himself must reach the opinion through a direct analysis of the authoritative legal sources, and thus his opinion represents the ruling which was probably decreed by God. If all the other mujtahids who reached their opinions through similar, independent means agree on the same point of law, then they will be corroborating each other as well as con­clusively demonstrating that the ruling of that particular case is certain. Here we attain a full enumeration of all the members of the class. The certitude resulting from this enumeration hardly needs any justification, especially in a system which is endowed with divine providence against error once tawdtur obtains.

ABU ISHAQ AL-SHATIBl

Partly grounded in what may be described as a subjective the­ological transcendentalism and partly in rational-cum-empirical justification, this theory of inductive corroboration stood as the underlying logical-methodological foundation for a number of ma­terial and theoretical legal principles, ranging from the various types of Prophetic reports to consensus. The theoretical elabora­tion and development of this inductive doctrine as part of legal theory came about after the introduction of Greek logic into kalam and usul al-fiqh. Whether or not there existed a necessary connec­tion between the introduction of Greek logic into the domain of law and the development of a theory of inductive corroboration, it remains true that it is only during the fifth/eleventh century and INDUCTIVE CORROBORATION, PROBABILITY, AND CERTAINTY 25 thereafter that this theory began to surface in the works of the ju­rists. In fact, it was not until the seventh/thirteenth century that logic, whether formal or inductive, made itself an integral part of an important segment of usul al-fiqh works, and affected in one way or another the construction of theoretical legal precepts.^[253] While the majority of usu/ists confined the use of the theory of inductive corroboration to problems such as those discussed above, we find that some others have utilized this theory even more ex­tensively. In his Al-Muwafaqat ft usul al-ahkam,^[254] for instance, Abu Ishaq al-Shatibi (d. 790/1388) went so far as to anchor his entire theory of usul al-fiqh in inductive principles.

ShatibT begins with the fundamental premise that general the­oretical legal principles and the sources of law are firmly grounded in certitude and that they derive their authoritativeness from God, for should there be any degree of probability concerning these principles and sources, there might follow the ominous conclu­sion that such probability (zann) may well decline to a degree of doubt (s/iafcfc),^[255] thereby rendering the Shariah, the decree of God, mutable. The certitude surrounding the general principles and sources of the law must then be derived either from conclu­sive pieces of textual evidence (adillah) which, ShatibT admits, are rare if they exist at all, or through an inductive survey of the mul­titude of probable pieces of evidence supporting these principles.^[256] ShatibT argues that the aggregation of such pieces of evidence is perhaps the main source of certainty in law. The tawatur, whether lafzior ma‘nawi, derives its certainty from this principle. Each in­dividual chain of transmission is undeniably probable, but when a sufficient number of transmitted reports are heard, certainty of the contents of the report obtains. He asserts that both the conclusive certainty concerning the Five Pillars of Islam and the indubitable authoritativeness of consensus and qiyas are established in this manner. Likewise, such fundamental juridical principles as the natural right to have a religion, to own property, to procreate, and so on, are advocated by the SharPah in no uncertain terms, terms that are individually probable, but in their multitude they corroborate and affirm the validity of these principles beyond any doubt.^[257]

The evidence that may be utilized in proving the certainty of legal principles may not be confined to the formal verbal expres­sion contained in the Prophetic reports and the Qur’an. Rather, these principles derive from the meaning which may be found, by means of induction, to permeate the entirety of shar'i material sources. Induction in Shatibi’s theory is not merely an exhaus­tive account of the reports that pertain in their totality to a par­ticular issue, but rather a thematic induction (istiqrd’ ma'nawi) of the spirit and the letter of the Shari*ah. The evidence may not be especially decreed for a particular case, or may not even directly touch upon the issue in question, but its indirect relat­edness to, and subsidiary support of, the issue yields certainty in the event of obtaining a sufficient number of corroborative instances.^[258]

In other words, the certainty of the general legal principles results from the cumulative corroboration of statements and in­dications found in passages and contexts that are not as a whole necessarily relevant to these principles. Corroborative pieces of evidence may appear in passing, or as minor points in a larger body of evidence.^[259] The conclusive knowledge of Hatim’s gen­erosity, for instance, obtains upon hearing countless anecdotes in all of which there exists, irrespective of the differing nature and contents of these anecdotes, a common theme to the effect of his generosity.^[260] It is therefore the recurrence of a meaning or a theme in a Prophetic report, Qur’anic verse, or any cir­cumstantial evidence, which will inevitably lead to certainty in the mind of the hearer. Shatibi plainly states that widespread and constantly recurrent themes or statements (“ma hasala fihi INDUCTIVE CORROBORATION, PROBABILITY, AND CERTAINTY 27 al-takrar wa-al-ta’kid wa-al-intishar”) rank equally with decisive and conclusive texts.⁶⁴ In ShatibT’s theory then, all mutawatir and mashhur reports have a force equivalent to that of summa- tive induction.⁶⁵ But what are the logical properties and logical justification of this induction?

In contradistinction to the particularistic nature of positive legal rulings (/uru‘), legal theory entails the construction of gen­eral principles and universal truths (kulltyat). The great majority, if not the entirety, of these kulliyat are based, as ShatibI has al­ready intimated, on a multitude of probable instances or particular statements (juz’zyat) which corroborate each other to the degree of certainty.

ShatibT is acutely conscious of the basic rule of induction which premises that to attain certainty with regard to a mat­ter all the particulars or species subsumed under that matter must be enumerated and found to be mutually corroborative. If; on the other hand, one succeeds in reaching a kulltyah on the basis of enumerating the great majority of instances, then there should be no instance which will contradict the kulltyah. This curious assertion may be explained in light of his assump­tion that if a particular instance is proven to be in contradiction with the rest of the instances constituting a given kulltyah then one of the two, the instance or the kulltyah is invalid. But our author remains insistent upon the claim that once a kulltyah is reached on the basis of the great majority of relevant, corrobora­tive instances, then such kulltyah cannot be invalidated by what appears to be a non-conforming or non-corroborative instance. Indeed, upon examining such an instance one would find that its essence (dhat) differs from the essences of the other corrob­orative instances, a difference which justifies its unsubsumabil­ity under that kulltyah. The essences of the particulars under a kulltyah must be identical, and must not be subject to external influences (umur kharijah). Take, for instance, the essence of hu­manity, namely, acting with intent and premeditation. You survey a great number of humans who will indeed corroborate and ulti­mately confirm this assumption. You may, however, come across

⁸⁴ Muwafaqat, 3:192, 194, 11.17-18.

⁸⁸ Ibid., 1:13, 11. Iff.; 3:189, 11.4-9. an individual who does not act with premeditation because he is afflicted with a certain severe disease. This particular individ­ual will not invalidate the kulliyah which premises that humans act with premeditation, for illness has effected a change in his essence. Likewise, the general principles of contractual law will not be rescinded just because the contract of ‘araya conforms to a different set of contractual legal assumptions.^[261] Similarly, tayam- mum (ritual cleanliness with sand or soil) cannot abrogate the prescribed kulliyah of ritual purity (taharah) which God intended as a means of cleanliness.^{[262] [263]} These non-conforming instances each have an essence, designed by divine wisdom (hikmah), which is not identical to the essences of the instances subsumed under that particular kulliyah.

That a particular species is proven to be unlike other species forming a genus and thus is rightly excluded from enumeration, creates no difficulty in Shatibl’s theory. When a species is shown to have no common denominator with a number of other species, there would be, to begin with, no sense in utilizing it as a corrob­orative instance. But the logically problematic issue in Shatibl’s theory is that he accepts as conclusive a kulliyah even though there has not been a complete survey of all the relevant instances. “In the SharTah,” he says, “the great majority of particulars [con­stituting a kulliyah] are considered as tantamount to a conclusive general, since the instances diverging from a kulliyah cannot con­stitute another kulliyah which can then compete with the first established kulliyah.”^ Put differently, a truly authoritative as well as conclusive kulliyah will not allow for a significant number of diverging instances to constitute another equally authorita­tive kulliyah. Therefore, ShatibT argues, an already established kulliyah will not be undermined because a single instance stands INDUCTIVE CORROBORATION, PROBABILITY, AND CERTAINTY 29 in opposition to it.⁶⁹ Only a significant number of competing par­ticulars can weaken or undermine a kulliyah.™

In support of this thesis, Shatibi introduces the example of the solitary reports which are generally considered authoritative for practice (‘amal) although some of them are dubious. In com­parison with those sound reports, the few doubtful ones pale into insignificance. Similarly, in rational sciences a mistake remains iso­lated and unaccounted for, but once scholars continue to reiterate the same mistake, it becomes the rule rather than the exception. When exceptions are numerous in legal matters no general princi­ple can be derived. It is only when the number of exceptions is insignificant that a kullvyah may be reached. Shatibi’s awareness of his unconventional stance towards inductive logic seems to have compelled him to justify his thesis on the grounds that it is the only conceivable means by which general principles of law can be formulated.⁷¹