| While the canonical position of the Japanese kedo (though) clause is said to be before the main clause, it occurs after the main clause in spoken data. In order to consider the characteristics of postposed kedo clauses, we collected tokens of kedo clauses in both preposed and postposed position from spontaneous conversations and examined them in terms of what type of recipient’s reaction (Gardner, 2001) which the turn featuring a kedo clause receives. The assigned values are the six categories: (i) strongly positive reaction, (ii) sentence co-production (Lerner, 2002), (iii) turn taking, (iv) backchannel, (v) no overt reaction, and (vi) strongly negative reaction. We found that, while a construction featuring a preposed kedo clause tends to receive a positive reaction, a construction featuring a postposed kedo clause tends to receive a negative reaction. This may mean as follows. Since a kedo clause conveys contrastive/concessive information relative to the main clause, preposing such information “enhances the projectability” (Lerner, 2002) of what comes next in the ongoing turn. This feature of preposed kedo clauses gives the recipient an opportunity to share the main point or climax with the speaker. This is why a preposed kedo construction tends to receive a positive reaction. On the other hand, postposing kedo clauses serve to counterbalance the main clause, so that a speaker can offset a potential exaggeration or offensive assertion. This effect allows the recipient to react negatively to the turn. This is why a postposed kedo construction tends to receive a negative reaction. Moreover, a speaker can postpose a kedo clause as soon as (s)he recognizes that the hearer is not reacting positively. Thus, counterbalancing can be seen as both a result of a hearer’s negative reaction and as a cause for it. |
