Toward resolution of the controversy over the function and usage of the passive: Prashant Pardeshi, Qing-Mei Li & Kaoru Horie Voice phenomena, especially the relationship between active and passive has played an important role in the development of modern linguistics. Notwithstanding extensive studies on the passive construction across languages of the world (e.g. Siewierska 1984) , there seems to be little agreement among scholars on the issue of the function of the passive. For example, Giv o n (1979:186) claims gfocusing the themeh to be the function of passive. Givon (1981:168) further notes that passive serves three functions: (i) to assign of topic status to a non-agent, (ii) to suppress the agent and (iii) de-transitivization of the verb form. Shibatani (1985:830), admitting that the afore-mentioned three functions motivate passive clauses, argues gagent defocusingh to be the primary function of the passives. Kuno & Kubaraki (1977) conceive passives as reflection of gempathyh while Brown & Levinson (1987) advocate politeness as one of the functions of the passive. Experts on Chinese grammar like Li & Thompson (1981) identify gadversityh as a primary function of the Chinese passive and claim that the same traits are observed in other East and South-East Asian languages like Japanese, Thai, and Vietnamese. With regards to the usage of passives, Chafe (1982) claims them to be predominant in writing (academic papers) than in speech (conversation) while Poole & Field (1976) found them to be fewer in both writing (narrative) as well as speech (interviews). Biber (1986) claims that the econtradictory' results reported in Chafe and Poole & Field are in fact accurate given the text types used but the global claims based on them regarding speech and writing are wrong?academic prose have highest score and conversation the lowest; fiction has intermediate scores, as do interviews. Against the backdrop of the mosaic of such differing as well as complementary claims on the function and usage of passives we have undertaken a cross-linguistic parallel corpora analysis of the passive constructions in languages like Japanese, Korean, English and Marathi to determine the extent of applicability of the foregoing diverse claims and shed some new light on this conundrum. We used a Japanese novel called KOKORO written by a Japanese novelist Natsume Sooseki and its translated version in Korean, English and Marathi as the source of our data. This story is a first person narration and was chosen on purpose to assess the effect of the so-called gempathy hierarchyh proposed by Kuno & Kubaraki (op. cit.). As a standard practice in typological study we adopted prototype approach advocated in Shibatani (1985) to identify passive constructions in the original Japanese version. Having identified the passive tokens in Japanese we compared them with their counterparts in respective languages at hand. The token frequency counts attested in each language are tabulated below.
The remarkable difference in the token frequency of passives across languages suggests that identifying a battery of functions shared across languages may be futile. Nonetheless, zeroing down on a single common function is possible. A close scrutiny of the data reveals that gagent defocusingh is an omnipresent function shared across languages under discussion as envisaged by Shibatani (1985) while other functions proposed by various scholars seem to vary from one language to other or in other words are language-specific. Thompson ( 1994) voices similar views in his cross-linguistic study on passive and inverse constructions. Out of the total tokens of the passive in Japanese a third can be ascribed to empathy effect (involving the ego/narrator as the goal/patient). In Japanese it is almost obligatory to use passive when the ego/narrator or the members of his in-group are goal/patient of the state of affairs in question. Other languages at hand vary on this count considerably suggesting thereby that the empathy hierarchy cannot serve as a universal yardstick. This cross-linguistic contrast is arguably attributable to the higher degree of sensitivity of Japanese language to speaker's subjective viewpoint (perspective), which has been observed in other domains of Japanese grammar (e.g. choice of personal pronouns vs. names; cf. Uehara 2001). In contrast, other languages sampled appear to show lesser degrees of sensitivity to speaker's subjective viewpoint, with Marathi occupying the lowest end of the cline while Korean and English situated in between. Furthermore, the notion of advers ity is conspicuously at work in Japanese while it plays a relatively marginal role in Korean, English and almost no role in Marathi. With regards to empathy and adversity similar results were found in our questionnaire-based study of the passives involving the speaker as the patient/theme spanning over languages like English, German, Japanese, Korean, Mongolian, Thai, Vietnamese, Khmer, Hindi, Marathi, Nepali and Sinhalese. With regards to the usage, the results of our study seem to challenge Biber's claim that narrative (fiction) have intermediate scores in that the Japanese narrative under investigation abounds with passive while its counterparts in other languages don't. Our results suggest the possibility of intra-genre variation across languages owing to the variation in the function of the passive in a given language. Our ongoing project on non-Japanese novel translated into Japanese shall shed more light on this. It should be emphasized that the functions are scalar?languages vary in terms of the degree to which they are sensitive to the function in question. In sum, like that of function, in the domain of usage as well the common denominator across languages seems to be the written text wherein the effect of concealing the agency can best be achieved. Selected references Biber, D. 1986. Spoken & written textual dimension in English: resolving the contradictory findings. Language 62: 384-414. |