不是所有的国内变量都是中介变量——新古典现实主义理论的中介变量与调节变量辨析

Not All Domestic Variables Are Mediating Variables:Distinguishing Between Mediating and Moderating Variables in Neoclassical Realism

新古典现实主义理论在结构现实主义理论因果关系的基础上简单地插入国内变量作为中介变量,导致该理论在研究方法上存在两点严重缺陷。一是混淆中介变量与调节变量,错误地将所有国内变量都视为中介变量,这也是新古典现实主义理论中的国内变量至今无法有效整合的直接原因;二是将体系因素与国内因素简单相加,没有论证体系自变量与国内中介变量之间的互动关系。作者从研究方法上厘清了中介变量与调节变量的定义及其在因果机制中的作用,提出新古典现实主义理论存在基于中介变量和基于调节变量的两条研究路径。为从实践上更好地理解中介变量和调节变量的区别,作者选取2008年以来的澳大利亚对华政策作为案例,构建了一个同时涵盖中介变量和调节变量的新古典现实主义分析框架,论证了新古典现实主义理论应该基于中介变量研究路径整合国内变量,从而使理论更加统一和简洁。

Neoclassical realism,which builds upon the causal relationships of structural realism,simplistically inserts domestic variables as mediating variables,resulting in two significant methodological flaws in the theory.First,it merely adds system-level and domestic factors without demonstrating the interaction between systemic independent variables and domestic mediating variables.Second,it confuses mediating variables with moderating variables,erroneously treating all domestic variables as mediating variables,which has been a direct cause of the ongoing difficulty in effectively integrating domestic variables within neoclassical realism.This paper clarifies the definitions of mediating and moderating variables and their roles in causal mechanisms from a methodological perspective,proposing two research pathways in neoclassical realism based on mediating and moderating variables.To better understand the distinction between mediating and moderating variables in practice,this paper uses Australia's policy toward China since 2008 as a case study,constructing a neoclassical realist analytical framework that incorporates both types of variables.Moving forward,neoclassical realism should integrate domestic variables based on the mediating variable research pathway,thereby achieving a more unified and streamlined theory.