卢凌宇

内容提要：国际关系理论创新包含两个层次，其中第一层次关联创新的有无，第二层次涉及创新的大小。第一层次的创新包括经验驱动性理论创新、理论整合性理论创新以及提出新概念或描述新事件/现象三种类型。第二层次的创新包括范式内创新和范式间创新两种类型。对于范式内创新而言，在其他条件不变的前提下，理论的自变量数量越多，外延就越小，相应地贡献也越小；在理论只有一个解释变量的前提下，理论的边际贡献就取决于该变量的空间性和时间性。范式间创新有两个亚型，其中第一个亚型是理论的结果变量相同(似)但基本假定不同，这样的理论可以时序优先性和理论内涵的大小作为评价标准；第二个亚型的特点是结果变量和基本假定均不同，这样的理论原则上不可比。作为一个变通，我们可以通过比较结果变量的“重要性”来推断理论的相对重要性，其主要指标是“基础性”。评估国际关系理论创新的参照系是专业文献，而非个体的直觉或见解。“众所周知”体现了评价主体良好的问题意识和文本较大的潜在理论贡献，但它本身并不是理论评价的科学标准。

关键词：国际关系理论；理论创新；方法创新；国际关系学科建设

Abstract：International relations theory can be evaluated at two levels. Whereas the first level relates to the presence or absence of theoretical contributions, the second gauges the size of a contribution. Theoretical breakthroughs at the first level may be empirically driven or theoretically integrated, or take the form of new concepts or description of new events or phenomena. At the second level, IR theoretical contributions consist of intra- and inter- paradigmatic innovations. For the former, ceteris paribus, the larger the number of independent variables, the smaller the extension of a theory, along with a smaller marginal contribution. Where a theory entails merely one explanatory variable, its marginal contribution rests on the spatial and temporal content of the variable. On the other hand, inter-paradigmatic innovations consist of two categories. Under the first subtype, IR theories, which are characterized by identical out-come variable and differential assumptions, can be gauged in terms of the timing of publication and the richness of their connotations. The second subtype theories differ both in the dependent variable and underlying assumptions. Such theories are non-comparable in principle. As an alternative, we can attempt to infer the relative importance of IR theories by comparing the "importance"of their outcome variables, whose principal indicator is social"fundamentalness". The frame of reference for evaluation is not individual hunch or insight but the professional literature. That "the theory is well-known”reflects the excellent question awareness of an evaluator, as well as the potential large contribution of a theory, but is not in itself a scientific yardstick for IR theory evaluation.

Key Words：IR theory, theoretical contributions, method innovation,IR disciplinary construction

本文载于《世界经济与政治》2023年第8期