Andor, József - Pólya, Tamás (2001): A frame based, lexicalist approach to describing functions of the verbal prefix le- in Hungarian. In: Andor, J. - Szűcs, T. - Terts, I., eds., Színes eszmék nem alszanak... /Szépe György 70. születésnapjára/, Pécs: Lingua Franca csoport, vol I., 67-83. Pagination is not correct.
József Andor – Tamás Pólya (University of Pécs) A frame-based, lexicalist approach to describing functions of the verbal prefix ’le-’ in Hungarian
1. Introduction Most Hungarian linguists would agree that a thorough description of the behaviour, functions, semantic content and the part of speech relatedness or identification of morphemes called verbal prefixes poses many problems for the analyst and that not all of the problems have been solved or discussed extensively and satisfactorily and in the depth required. In our paper we would like to bridge some of the gaps related to clarifying the semantic functions and features of the prefix le- by outlining factors such as direction, motion, lexical content, frame relatedness, metaphorization, based on and supported by a body of experimental data gained from 50 adult native speakers of Hungarian. 2. Before addressing that issue, however, let us give a brief summary of some of the relevant critical views on the status of verbal prefixes taken from the literature. 2.1 The so-called Academic Grammar of the Hungarian Language (A mai magyar nyelv rendszere) edited by József Tompa (263-8.) states that concerning their part of speech relatedness verbal particles are more or less related to adverbs in their basic meaning, which meaning can be more or less marked, or it can even disappear and the prefix itself can then only refer to the mode of an action such as expressing its perfection (ex. megáll), duration (eltréfál), etc. This view is amply supported by historical evidence (J.Soltész, 1959). The Academic Grammar also states that verbal particles morphologically form a mixed category: dangling between free and bound morphemes, and that syntactically they function together with their verbs in taking arguments or adjuncts. All in all, it comprises a transitory, heterogeneous category both in its development, realization, and linguistic functions. Of the meaning relations expressed by or with verbal particles the AG gives the following list (1970, I:263): (a) directionality (locan specificity), eg. Ledob (except for the particle meg-) (b) perfection of the action (letelik) (c) starting of an action (ráböbben) – le- is not mentioned here (d) duration of an action (elvan) or recursion (átfest) – le- is not mentioned (e) mistaken nature of an action (which is a lexically characterized feature (elsóz) – le- is not mentioned here again The AG also mentions cases when using a verbal prefix is unnecessary due to its meaning implicitly incorporated in the verb itself (ex. kiértesít, beindít), hence le- is not mentioned here either. However, such prefixed formulations are rated unnatural in Hungarian. 2.2 Another authoritative, standard reference, Present-day Hungarian (A mai magyar nyelv) ed. By Benczédy et al. (1982:65-68.) in discussing verbal prefixes in its chapter written about adverbs, stresses their morphological and syntactic relatedness to suffices and adverbs. Summing up their meaning it mentions their
a) directionality (except for meg-) ex. lejön b) perfectivity, completion of the action, ex. leültet c) the starting of an action, especially with verbs expressing instantaneous actions, ex. felgyújt (le- is not mentioned) d) duration (most of the cases with el-, ex. elnéz) e) the resultative, successfully completed nature of actions, ex. levizsgázik f) the mistaken nature of actions (elsóz); no reference to cases with le- is given g) actions expressing processes of reduction or growth, ex. lesoványodik. The same source also mentions the important fact that different prefixes can be added to the same word depending on given situations, implicitly referring to frame dependence (ex. lemos – bútort, elmos – edényt, megmos – kezet, etc.), and this handbook, too, calls attention to the unnecessary occurrence of verbal prefixes with certain verbs, such as kiértesít, lefogad, etc., or cases of usage rated as improper, ex. bejósol. 2.3 In her seminal work written on the subject, published as early as 1959, Katalin J. Soltész, in discussing functions of verbal prefixes, identifies the following types: a) b) c) d) e)
prefixes can express the direction of the action they express the aspect of the verb they modify the meaning of the verb they can change the syntactic role of verbs (whatever this means for her) they also serve as a means of verb-formation morphologically
She stresses that the above types are closely linked and related to one another (1959:155). J. Soltész’s monograph is the authoritative source in which we can find an extensive listing and discussion of the particular functions or senses of particular verbal prefixes, among them those of le-. In giving a long list of the semantic content carried by le- as a prefix, however, she fails to synthetize or identify the major tendencies of the pathways of meaning, moreover, she does not systematically discuss or describe the processes of metaphorization. It is in this plain that the present paper intends to provide major insight. 2.4 Before entering into details, however, let us finally refer to Komlósy’s work on verbal prefixes (1992:494-506). In discussing aspects of the status of verbal prefixes, he draws parallels with the occurrence of other types of words in positions that can be taken by verbal prefixes to point to the blurred nature of characteristically verbal prefix-like syntactic behaviour. He points to the close, semantically based relatedness between verbal prefixes and verbs, reflected – as he mentions – by native speakers’ intuition, and to the role of the argument structure of the verb (taken together with its prefix as a single lexical item) in setting constraints on representation. He states that both verbal prefixes and verbs can serve as arguments of one another, however, the syntactic head in such relations – he states – is always the verb. Thus, he claims, verbal prefixes can be identified as free modifiers, i.e., they have verbs as their heads syntactically, but they can also dominate verbs expressed by their meaning as free modifiers. In agreement with Komlósy’s views, but not entering into further argumentation in discussing the status of verbal particles, let us turn to analyzing and systematizing the lexical semantic functions of le- as a prefix.
3. Based on our data collection, the following semantically based groups and their analogies have been identified: (1) The prefixed verb expresses downward spatial motion or direction intransitive: ‘the <S> (or his voice, facial look, etc., so to say) moves to a place <down or downward> in the way expressed by the
’ transitive: ‘the <S> gets the to move to a place <down or downward> in the way expressed by the ’ (1.a) The verb expresses some kind of motion (or it expresses a certain sense of motion) le- vhova fárad szédül csókol eszik, iszik harsog, kiabál, ordít lök, pofoz, rugdos zavar vhol <út, ösvény> is pragmatically involved) le-; <S> disappears, is removed completely, it ceases to exist; <S> was part of a larger structure’ transitive: ‘the <S> makes the take a flatter position in the way expressed by the ; disappears, is removed completely, it ceases to exist; was part of a larger structure’ le-; <down> means some specific quality of the <S>’ transitive: ’the <S> makes the move spatially or metaphorically to a place <down> in the way expressed by the or <S> takes the position by moving or acting in the way expressed by the ; <down> here means some specific quality of the ’ (3.a) <down> = ’fewer, less (quantity)’ (from change in physical conditions such as le- csökken, hajlik, hűl törpül fárad vki disznóz, marház, szid éget, ír, kicsinyel, néz, szól, tesz húz <előadást> <játékosnak> vki <játékosnak> irányít, küld vki vmit vhova (in a negative sense) passzol cserél, vált becomes perceivable’ le- diktál, fest, fényképez, ír, gépel, másol fest hoz néz, olvas vmit vhonnan (pl. eredményt) vhol vki (köröket, adósságot) vki <előadást/előadót> vki vmit <mérkőzést> vki vmit (távot, folyadékot, szobát), not <down> is expressed) <játékosnak> fejel <játékos> <játékosnak> ? igazol <játékost> 3. With the aim to support our groupings by gaining empirical evidence, we performed tests using 50 adult native speakers of Hungarian as experimental subjects. The experimental subjects were asked to ragte whether 59 given to them as a list could be prefixed with le- and be considered as correct verbs of Hungarian. They were also asked to give data of their judgment whether the verbs so prefixed could be used in the abstract sense, metaphorically, or in both ways. The results of the test are given in Table 1. We purposely included a number of verbs in the list given which we thought our experimental subjects would find extremely difficult to interpret and accept.
