Conversion in English and its implications for Functional Discourse Grammar
Download 202.86 Kb. Pdf ko'rish
|
Conversion in English and its implicatio
Pseudo-instrumental verbs
, actually denoting manner of motion: (9) a. He hammered the desk with his shoe. b. He brushed his coat with his hand. c. I paddled the canoe with a board. d. String him up with a rope! e. Can you whistle with a blade of grass? f. The convict sawed off the bars with her dentures. g. She anchored the ship with a rock. h. We wedged the window open with a screwdriver. According to Kiparsky, the verbs in (9) describe a manner of motion which can, but need not, be performed by the related noun. Hence, it is possible for another object to appear in the sentence without producing an anomalous semantic effect unlike the examples in (8). Thus, he claims, the definition of a verb like hammer would be something like ‘to strike with the flat side of a heavy object’ and not ‘to strike with a hammer’. 4 Therefore, by dividing the category of instrumental verbs into those two classes, Kiparsky is able to prove the directionality of the process at least for the verbs in (8). Crucially, Kiparsky’s distinction seems to correlate with formal properties. Thus, true denominal verbs keep the same stress pattern as the input noun, as shown in (10a), whereas verb/ noun pairs may show a stress contrast as in (10b) 5 : (10) a. to pa´ttern/a pa´ttern (cf. * to patte´rn), to ı´ndex/an ı´ndex (cf. * to inde´x) b. to prote´st/a pro´test; to permı´t/a pe´rmit As noted by Don (2004:939) a natural explanation for Kiparsky’s facts should assume the existence of word classes and the directionality of conversion. Don also adduces facts from Dutch and German to support this view. As I will show in the following section, however, the different behaviour of instrumental verbs in examples (8) and (9) can receive a different explanation under Clark and Clark (1979) analysis. As for the notion of frequency, Adams observes that items like campaign, duel, experiment or gesture are normally treated as nouns and the corresponding verbs as derived since the verb is much less frequent. However, as in our previous reasoning, given the fact that they denote actions, it could be argued that they parallel nominalizations and, therefore, the opposite direction in the process could be defended. Frequency, then, is not a reliable criterion to establish the direction of the process either. What is more, in those cases in which both the verb and noun are equally frequent, the matter seems more difficult to solve. As for historical evidence, Plag (2003:108) cites the case of moan, first attested as a noun in 1225, and converted into a verb in the 16th century. Yet, the noun is today best defined as ‘the act of moaning’ which would indicate that it derives from the verb. Thus, even though Plag (2003:116) argues that the directionality problem can be solved in most cases through a D.G. Velasco / Lingua 119 (2009) 1164–1185 1169 4 As noted by Adams (2001:21) this has the consequence that a hammer might be defined as an instrument which is used for hammering and therefore, could be considered a deverbal noun, contrary to what is usually assumed in most descriptive approaches. 5 But note that, for Huddleston and Pullum (2002) , changes in stress placement are not part of the notion of conversion. Also, as pointed out to me by a reviewer, this might indicate that we are in fact dealing with two different lexical items. combination of historical, semantic, formal and frequential evidence, they can also produce conflicting results in some cases. To conclude this short review of conversion as zero-affixation, it should be noted that the analysis also suffers from an important theoretical drawback which has been noted by several scholars ( Huddleston and Pullum, 2002; Lieber, 1992, 2004; Plag, 2003 ). As Huddleston and Pullum (2002:1641) claim, in ordinary affixation we seem to have a number of different affixes which produce different semantic effects on the input bases. As conversion also produces numerous distinct semantic effects on bases, we should accordingly assume the existence of numerous ‘zero-affixes’ whose identification and description would be rather difficult, if possible at all. This theoretical problem has led some authors to propose an alternative analysis of conversion, to which I now turn. Plag (1999, 2003) and Lieber (1992, 2004) argue that conversion is not that similar to affixation as proponents of the zero-derivation analysis seem to believe. They claim that the semantic range exhibited by conversion to verbs in English is wider than that exhibited by the verb-forming suffixes –ize and –ify. Also, the distribution of derivatives is significantly different: whereas there are many instrumentals, performatives and similatives among the verbs formed by conversion, these categories seem to be rather scarce within the class of –ize/–ify verbs. By contrast, the latter contain more locatives, causatives and resultatives than we normally find among converted verbs. This leads Lieber (2004:94) to claim that ‘verbal conversion simply does not behave semantically like derivation with –ize, –ify or any other verb-forming affix’. Consequently, she claims, conversion is not a type of derivation, but, rather, a form of ‘coinage of novel lexical items’, which she names ‘relisting’. Lieber’s coinage seems to be a more or less intentional and creative process, subject to general principles of pragmatic interaction. To the extent that the new use becomes conventional through reuse, it is assumed to be lexicalized and thus relisted in the mental lexicon. Her analysis is based on the proposals presented in Clark and Clark (1979) , to which I will come back shortly. Suffice it to say for the moment that the relisting analysis accounts for the fact that the realm of conversion extends to any semantic category within a given word-class and is not restricted to semantic subcategories as in the case of derivational affixation. Further arguments are given by Clark and Clark (1979:781) , who note that conversion is semantically more specialized than derivation, usually creating idiomatic verbs whose meaning is not predictable. Admittedly, as Don (2005) rightly points out, the semantics of the converted verbs can always to a certain extent be predicted on the basis of the meaning of the base, as there are no derivatives totally unrelated in meaning to the base. At the same time, even regular affixation processes typically entail some degree of specialization of the derived item (e.g. a drinker is someone who drinks (usually too much) alcohol). 6 While this is undoubtedly true, it is also certain that the semantic distance between input and output is usually greater in the case of conversion than in the case of affixation. As the analysis offered by Clark and Clark (1979) will be crucial in the proposal to be defended in this paper, I will postpone a more detailed exposition for the following section and move now to the third analysis of conversion. The third main analysis of conversion is very aptly expressed by Bauer (1983:227) : The commonness of conversion can possibly be seen as breaking down the distinction between form classes in English and leading to a system where there are closed sets such as pronouns and a single open set of lexical items that can be used as required. Such a move Download 202.86 Kb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling