藉由批判性地回顧主流的實證經濟學之後設方法論一亦即，以 規律性為基礎的後設方法論一以及檢視總體經濟學的理論構作實作之一項案例，本文指出：與一般看法相左，「假設之操控」與「假設 其他情況不變定律的使用」這兩項策略，不僅對經濟理論構作而言是 重要的，甚且是方法上的要求。這些要求實乃源自於經濟學本身的性 格：經濟學是一門建構在經濟學家有關下列項目的檢視之上的學科——即針對「經濟變數的儲能」與「由這些變數所組成的起因結構」 所進行的檢視。在這項新的有關經濟理論構作的特徵化描述之基礎 上，本文進一步論述：公認的科學解釋之涵蓋律模型並不適用於來闡 明經濟說明的模式；因而，一項新的替代解釋模型一起因結構的模 型一就被提出來以進行這項闡明。本文提出下列的立場作為結論：經濟學的確可以被認為是一門實證科學，如果針對目標經濟現象所進 行的說明之模式是被放在起因結構模型的架構內被闡釋的話；而起因 結構模型本身，則是在歸納經濟學家在其理論建構的過程之中所真正遵循的眾多方法上的程序之後，而被推導出來的經濟理論構作之一般架構。

By critically reviewing the mainstream positive meta-methodology of economics—i.e., the regularity-based meta-methodology—and examining a case of the practice of theorizing in macroeconomics, this paper points out that, contrary to what is generally thought, assumption-manipulation and the application of ceteris paribus law are not only important in economic theorizing but are also methodological requirements. These requirements originate from the character of economics itself: economics is a discipline that is constructed on the basis of economists，examination of the capacities of, and the causal structure of, economic variables. Based on this recharacterization of economic theorizing, this paper further argues that the received covering-law model of scientific explanation is not an adequate model for explicating the pattern of economic explanation; consequently, an alternative new explanatory model—the causal structuralist model—is proposed. This paper concludes by maintaining that economics can indeed be regarded as a positive science if the pattern of explanation of the targeted economic phenomena is interpreted within the framework of the causal structuralist model. Moreover, this model, in turn, constitutes a general framework of economic theorizing that is inductively derived from the methodological procedures that are actually adopted by economists in the process of theory-construction.