The results of the analysis are the following.

1. "Noda" sentences and negative potential sentences are politer than negative sentences regarding the parameters of consideration, indirectness, familiarity, and formality.
2. In comparison to negative potential sentences, "noda" sentences show a higher degree of consideration, indirectness, and familiarity. However, in terms of distance and formality, negative potential sentences show a higher degree of politeness than "noda" sentences.
3. "Noda" sentences show a higher degree of familiarity than negative potential sentences. This can be interpreted as a sign of positive politeness. The degree of distance and formality, however, is higher in the case of negative potential sentences than in the case of "noda" sentences, which could be interpreted as a sign of negative politeness.

Firstly, I pointed out that these expressions vary and will be classified into 8 types:

1. a clause + a simple word
2. a clause + a compound word
3. a phrase + a simple word
4. a phrase + a compound word
5. a word + a simple word
6. a word + a compound word
7. ø + a compound word (which has more than 2 word constituents)
8. ø + a compound word (which has 2 word constituents)

1. ``VICTIMS'' appear more than ``ASSAILANTS''.
2. ``TIMES'' mostly appear in noun modifications, seldom in compound words.
3. Although ``STAGES'' appear in noun modifications, in (i),(ii), (iii), (iv), (v) and (vi), they appear in compound words in (vii) and (viii)
4. ``CRIMINAL ACTS'' mostly appear in (ii), (iv) and (vi), seldom in (vii) and never in (viii)
5. In many cases, only ``CRIMINAL ACTS'' appear in compound words in (ii). Moreover, in noun modification clauses, there is no conspicuous difference between (i) and (ii).

1. younger people tend to misuse polite expressions more than elder people,
2. most people tend to misuse polite expressions using the (``go'' + word) form more than they misuse those using the (``o''+ word) form, and there was less tendency to make functional misuses (e.g., misuses regarding the use of respectful expressions vs. that of humble expressions) for expressions using the (``go'' + word) and (``o'' + word) forms.

1. LINGUISTIC FACTORS
   Primary determinants were conjugation types and conjugation forms (i.e., affirmative vs. negative) followed by the articulation of the stem-end consonant and the stem length.
2. EXTRA-LINGUISTIC FACTORS
   1. The long potential is considered as more canonical than its shorter counterpart.
   2. The frequency of the short potential verb use depends on the situation and the addressee.

I chose 14 novels written by Haruki Murakami, 23 translated novels (14 novels originally written by Raymond Carver, and 9 novels originally written by others) and 39 novels written by other Japanese writers and then calculated the rates of appearance of parts of speech (RAPS), the rates of appearance of them just before the ``Toten'' which is the pause in the middle of sentence in Japanese(RAT), the rates of particles just before the ``Toten''(RPT), and 10 indexes (the rates of a noun, the rates of demonstrative, the length of sentence, the rates of conversation sentence, etc.)

According to the analysis by using RAPS the differences could be observed in the style of sentences between original and translated ones. However according to the analysis by using RAT and RPT the differences couldn't be observed except for novels written by Raymond Carver. In conclusion, there are some differences about RAPS between original and translated novels, while how to use the ``Toten'' is dependent on the consciousness of the translator.

Analysis by using 10 indexes was effective to grasp the characteristics of sentences about Haruki Murakami, but it wasn't effective to distinguish between original and translated novels.

There are three purposes for this research:

1. To quantitatively investigate error corrections of keigo verbs made by intermediate Chinese learners.
2. To hypothesize the learners' interlanguage of keigo verbs by constructing a model.
3. To quantify the learners' interlanguage by applying the results of the investigation to the hypothesized model.

1. The interlanguage of sonkeigo and teichogo is quantitatively greater in sociolinguistic competence 1 than in morphological competence.
2. The interlanguage of kenjogo is quantitatively greater in sociolinguistic competence 2 than in sociolinguistic competence 1 and morphological competence. Sociolinguistic competence 1 is equivalent to morphological competence in the quantified interlanguage of kenjogo.

The usage investigation was done by using written materials which quoted spoken language. Furthermore, the questionnaire investigation for the young generation groups was done in four areas. Conclusions are as follows:

1. Though -masen is considered as proper form, -nai desu is used very often in spoken language.
2. With i-adjective, noun, and non-existence expression, -nai desu is used with comparative frequency. With verb, -masen is used abundantly, but the rates for -nai desu increase somewhat with verbs which express capability, with the verb wakaru and with -teiru form.
3. When the final particle is added, -nai desu is often used. In the young generation, the addition of the final particle makes the degree of an approval of -masen a little lower.

In the case of Korean, the vertical relation by age or the social status is most thought as important in ``the way of speaking regarded as right.'' The point of thinking the vertical relation by age or the socia1 status as important is almost the same also in ``a usual way of speaking.'' The influence of uchi-soto to the degree of politeness is not seen at all.

On the other hand, in the case of Japanese, in ``the way of speaking regarded as right,'' the vertical relation by age is thought most as important, also taking into consideration a uchi-soto relation and a social status. However¡¤by ``a usual way of speaking,'' a uchi-soto relation and a relative degree of intimacy relation serve as a key factor which determines the degree of politeness.

As a result, it became clear that:

1. The relative usage of the two forms differs to suit different text types and styles, depending on the parts of speech of the lexemes prefixed to naru.
2. The relative usage of the two forms differs to suit different text types and styles, depending on the differences in meanings of [noun] ni + naru and the differences in functions to main clauses.