Tadmin, "implication of meaning," in medieval Arabic.

• Author: Gully, Adrian

1. INTRODUCTION

   The concept of tadmin as a semantic, and to a lesser extent grammatical, phenomenon has received only sporadic attention by Western scholars. In 1934 the Arabic Language Academy of Egypt (henceforth referred to as ALA) devoted considerable time to a discussion of tadmin in an attempt not only to unravel some of the complexities of the term, but also to establish guidelines for its legitimacy as a linguistic category and for its future employment within the Arabic language. Since that time, its importance has been dealt with cursorily by Stetkevych (1970: 13, n. 36), Monteil (1960: 167ff.), Wehr (1943: 24, 33), and in more detail by Massignon (1954: 14-16). Although Massignon devotes more attention to the etymological significance of the term than its actual employment, he outlines the eight fundamental interpretations of tadmin as discussed by the ALA, which based its debate primarily on the medieval perception of the term. Massignon's understanding of tadmin as "l'involution semantique du concept" (1954: 14) is particularly enlightening, and this idea is cultivated by Gardet (1986: 70-72), particularly in relation to its counterpart term takhrij "extraction, sortie de" (1986:70-71) as part of the dialectic "bi-polarite" of Muslim thought.(2) This paper sets out to examine what exactly the medieval grammarians, the rhetoricians, and even the legal theorists understood by the term tadmin. A brief assessment will also be made of the relevance of tadmin for the language today.

   Our working translation for the term tadmin will be "implication of meaning" (cf. Monteil's "contenu implicite" [1960: 168]). In a discussion of three key elements of semantic evolution in the Arabic language, i.e., qiyas, majaz, and tadmin, Monteil describes the latter as the most innately Arab concept of the three. There is certainly sufficient evidence in the medieval literature to justify this view; Ibn Jinni (d. 392/1002), for example, deemed it a particularly delightful device, which could be employed and understood only by someone possessing a degree of mental dexterity and familiarity with the concept (Khas., 2: 310).(3)

   Scholars of poetry, meanwhile, will perhaps be more familiar with at least two other important usages of the term which are worth noting. The first definition, which is perhaps less relevant to this discussion, is "enjambment" (Van Gelder 1987: 28; Heinrichs 1969: 13; cf. al-Jurjani's definition, Ta rifat, 62), as deployed in Ibn al-Mutazz' work of the late third century A.H., Kitab alBadi). However, the other poetry-related use of tadmin is particularly interesting. Bonebakker translates it as "incorporation" (1966: 39, n. 41), and defines it further by saying that "it depends on incorporating a well-known proverb or line of poetry in one's composition and employing one of the words that occur in this line of poetry in a different sense (my italics)" (1966: 86).(4) Now, we shall see shortly that, in very broad terms, the grammarians' understanding of tadmin rested either on a verb (or in some cases a particle or noun) replacing and conveying, acceptably, the meaning of another one used in a specific context, but without affecting that meaning, or a verb becoming transitive by a particle by which it would not normally become transitive (Tahanawi, Istilahat, 4: 896). However, there was also a Kufan understanding of the term that required that a given verb be interpreted according to its opposite meaning. This would seem to be particularly relevant to tadmin as described by Bonebakker, above.

   As far as dating the first usage of the term tadmin, in the sense "implication of meaning," is concerned, it is very difficult to say whether the poets influenced the grammarians and rhetoricians in adopting the term or vice-versa, especially since some features of the language which came to be identified as characteristic of tadmin were sometimes discussed within categories of another type. We know that as early as the late second century, the Kufan grammarian al-Farra (d. 207/872) availed himself of the term with regard to Qur anic interpretation.(5) The employment of tadmin by al-Zajjaji (d. 337/949 or 340/951) in a sense similar to the one under discussion here is also particularly interesting, as is the more esoteric and varied usage by Ibn al-Sarraj (d. 316/928). A detailed analysis of its occurrence in these and possibly other works, however, awaits further research.(6)

2. THE GRAMMARIANS AND RHETORICIANS ON TADMIN

   This paper is in some senses a logical extension of Gully (1994), where the question of synonymy, or more specifically, substitution (niyaba), amongst the particles (particularly the prepositions) was discussed.(7) Perhaps this is a sound vantage point from which to initiate the main lines of the present inquiry. That the term tadmin, with the sense of "implication of meaning," was not employed by the Kufan grammarians is fundamental to this discussion. They adhered unreservedly to the view that substituting one preposition for another in a given context was acceptable, provided it conveyed the same meaning. Moreover, they held that instances of an intransitive verb, as such, becoming transitive, such as rahubatkum al-dar "the house was spacious for you," or a transitive verb with a particle becoming transitive without the particle, as in tamurruna l-diyara "you pass by houses," in which the verb marra normally requires the preposition bi, could only be accepted as part of "poetic licence" (darura); otherwise they would be considered anomalous (shadhdh) (Majma, Majalla, 185-86). They believed that the nonposited meaning (al-ma na ghayr al-wad i) of a verb (or its derivatives) was derived from the substitution of one preposition for another rather than an element of expansion (tawassu) in the verb itself. In other words they held that each particle possessed more than one posited meaning (Majma , Majalla, 184). Therefore, when a verb such as radiya "to be pleased" appeared with the preposition ala instead of the normal an, in, for instance, the following line of poetry,

   idha radiyat alayya banu qushayrin . . . when the Qushayr tribe is pleased with me . . .

   the Kufans would relate this to the category of al-haml ala diddihi, "agreement according to its opposite,"(8) to account for the correspondence with its opposite verb sakhita "to be angry" which normally became transitive in the presence of the preposition ala. Likewise they referred to a direct substitution of one verb (or noun) for another as al-haml ala nazirihi, "agreement according to its counterpart" (Ibn Jinni, Khas, 2:311).(9) However, in contrast to this, the Basrans interpreted the verb radiya as implying the meaning of the verb atafa which conveys the same meaning and takes the same preposition (cf. Ibn Hisham, Mugh., 1: 143).(10)

   Another important reason why the Kufans preferred not to use the term tadmin in this context was to avoid any confusion with an earlier employment by which certain nouns assumed the meaning of certain particles. At this point it is worth introducing one of Ibn Hisham's (d. 761/1359) descriptions of the communicative meaning (fa ida) of tadmin. He notes that it occurs when one word conveys the force of two (yudallu bi kalima wahida ala kalimatayn) (Mugh., 2: 530),(11) which could be exemplified by the conditional and interrogative nouns; for instance, when the particle ma in its capacity as an undefined noun (ism nakira) implies the meaning of the interrogative particle with the sense of "what thing" (ayyu shay), as in ma hiya? "what is it?" (Mugh., 1: 298). According to the earlier grammarians tadmin was used to describe a noun which signified, by convention, a meaning which should have been signified by the particle, as in the case of the conditional or interrogative nouns - this was one of the bases of non-inflection (ilal al-bina) (Khafaji, Tiraz, 19). Therefore, they held that the conditional noun haythu, for instance, was uninflected because of its implying the meaning of the conditional particle in. In other words, in addition to its posited adverbial sense, haythu possessed a partial meaning (ma nan juz. i) normally conveyed by a particle, in this case the conditional element conveyed by the expression in.(12)

   To sum up this particular perception of tadmin in relation to Ibn Hisham's description of its communicative meaning (fa ida), what is at issue is the capacity of the nouns of "interrogation" and "condition" to convey not only their nominal meaning, but also the meaning conveyed by the particles of those categories. In the case of the interrogatives they also convey the (underlying) meaning of interrogation expressed by the hamza, whilst the conditional nouns convey the (underlying) sense of condition expressed by the conditional particles such as in.(13) Although it is fair to say that the grammarians and rhetoricians discussed this aspect of tadmin less than that of the acceptability of one verb replacing another in a given context, it was still an important feature of the debate. Ibn Hisham's description at that point in the Mughni was potentially misleading because it did not include all the other possible interpretations suggested by a word conveying the meaning of two words. We could tentatively suggest, therefore, that his original description took into account the more grammatical connotations of the term, and that it was only later, when he produced a definition of tadmin according to which "they (i.e., the linguists) might inculcate an expression which has the meaning of another expression, thereby giving it its status" (qad yushribuna lafzan ma end lafz akhar fa yu tunahu hukmahu) (Mugh., 2: 685),(14) that he invoked the semantic connotations of tadmin as understood by the rhetoricians and the legal theorists. At that point he also described the communicative meaning of tadmin in a manner similar to its earlier use, viz., when one word conveys the import of two words...