Reduplication of nouns as a mode of expression of quantitative meaning in the Japanese language

Analyzed in this article is reduplication of nouns in the Japanese language. The aim of this study is to reveal whether Japanese native speakers can or cannot estimate reduplicated nouns as forms with quantitative meaning. There was created a questionnaire that covered various ways of expression of quantity in Japanese with the part of it concerning reduplicated nouns. Questioning was carried out anonymously in December 2014. The respondents were students and teachers of Waseda University including teachers of the Japanese language for foreigners (non-professional volunteers) of social organization Hosien, Waseda University, 110 people (30 men and 80 women), 20-65 years old. The questionnaire contained 15 nouns of wago, kango and gairaigo origin. Reduplicated nouns were presented in the written form accompanied with the question «What do you think about these words?». Three variants of answering were given: - «right», X - «wrong», △ - «doubtful». Surprisingly, one more variant emerges in the already obtained data, where respondents leave blank spaces, which is difficult to explain. The first stage of analysis concerning gender preferences did not show great difference between «right» or «wrong» estimation of reduplicated nouns and can be left unattended. On the whole, it is noticed that women were more often reluctant to answering than men. Practically all respondents unanimously deny the possibility of making reduplicated nouns from kango, giving answer «wrong». The problem was with estimation of reduplicated wago nouns. Reduplicated nouns shimajima «islands», kumogumo «clouds», tanidani «valleys» and sakanazakana «fishes» caused the biggest difficulties in answering. Probability of appearance of «right»«wrong» answers (provided that the number of «right» answers makes less than the half of the retrieved data) was calculated by means of the method of constructing of confidence interval for the parameter of binominal distribution. The results are given in the table form. Impossibility of reduplication kango and gairaigo is recognized very confidently with the probability not less 0.83. The result for wago is dependent on the word analyzed. Presuming that respondents form sufficiently representative sampling, it is arguable that Japanese native speakers, at least for some groups of wago, estimate the phenomenon of reduplication uncertainly. For example, the probability of more than 54 % of native speakers know about impossibility of reduplication of sakanazakana «fishes» is not more than 0.1. The article explains why really non-existing reduplicated nouns were estimated as «right». In private conversation with respondents the next explanations were presented: habit ( pettopetto «pets»), admiration of something nice ( mushimushi «insects»). One more explanation may be given in terms of analogy (PETPET → pettopetto ). This hesitation in estimation of reduplicated nouns is reflected in linguistic papers dedicated to the problem of reduplication. Lists of really existing reduplicated nouns varies from paper to paper. It is resumed that reduplicated nouns are not clearly understood as plural forms by the Japanese native speakers. Not all the Japanese speakers can estimate reduplicated nouns. Somebody doubts or denies answering. Thus, in the minds in native speakers, there is no understanding of reduplication and the rules of making reduplicated nouns as a mode of expression of quantitative meaning.