@article{oai:shobi-u.repo.nii.ac.jp:00000102, author = {大木, 啓介 and OOKI, Keisuke}, journal = {尚美学園大学総合政策研究紀要, Bulletin of policy and management, Shobi University}, month = {Mar}, note = {コンテクスト問題は従来、比較研究者を執拗に悩ませてきた。本稿では、比較政治分析の戦略として近年再解釈されつつある「中範囲比較」を取り上げて、この研究設計がコンテクスト問題をいかに克服しようと意図しているかを吟味する。概して言えば、この研究設計は普遍主義的渇望から戦略的に後退し、比較の範囲を限定することと引き替えに現象のコンテクスト性を考慮に入れてコンテクスト問題に対処する。この分析方針には、在来型の方針と比べると、理論を展開し検証し洗練するという比較の役割のあり方に関しても、政治の世界における因果関係とは何か、また因果推論はどうあるべきかに関しても、大きな転回が見て取れる。本稿では、こうした認識論上、存在論上の諸前提をも射程に入れて中範囲比較の輪郭を描出し、この比較戦略がその推論の説得力を範囲条件の設定と因果メカニズムの分析から引き出すことを突き止めると共に、コンテクスト問題に有効に対処し得る可能性が少なからずあることを論証する。, The Context Problem is a crucial and perennial issue in comparative inquiry. This paper discusses the arguments founding an ideal-typical `middle-range' small-N comparative research approach called Mid-Range Comparison and examines how its recent style as distinguished from the old one copes with the Context Problem. Firstly, the Context Problem is considered with particular reference to the problem of comparability. Secondly, through examination of its epistemological and ontological backgrounds, it is shown that Mid-Range Comparison recently reconceptualized is the mode of comparison in which scholars analyse multiple cases in the effort to formulate causal generalizations about those cases by taking the contextuality of the phenomena into consideration. And then, drawing upon G.Sartori's `ladder of abstraction' scheme, the author demonstrates that the approach represents an ideal-typical center between nomothetic quests for universally applicable causal law and idiographic, context-bounded narratives and draws a great deal of its inferential power primarily from setting temporal and spatial scope conditions and analyzing causal mechanism. The paper concludes that there is a good possibility of Mid-Range Comparison overcoming the Context Problem through constructing concepts which allow at the same time comparison and empirical measurement and therefore the approach could contribute to the store of substantive knowledge in ways that large-N quantitative and decontextualizing approaches may not., 1, KJ00005551719, 論文, Article}, pages = {1--21}, title = {中範囲比較の論拠とコンテクスト問題}, volume = {16/17}, year = {2009}, yomi = {オオキ, ケイスケ} }