社会科学中的客体：从本体论到认知论和方法论

一个简略的说明

唐世平

博主按：（陈）定定老师在一个人人的状态中提到“实验方法”在社会科学中的作用的问题。这个问题其实非常复杂。而诸多讨论的问题在于：许多社会科学过去和现在的关于认知论和方法论的辩论和讨论都是因为首先没有搞清楚社会科学的客体（objects，或者说是研究对象）的本体论区别。因此，我的这篇文章“The Objects of Social Sciences: Idea, Action, and Outcome (From Ontology to Epistemology and Methodology) 【社会科学的客体：观念、行动、和结果（从本体论到认知论和方法论）】”旨在现在本体论水平上澄清，然后再讨论认知论和方法论层次的问题。

【基于文章的PPT在我的个人网页上。】

这背后的根本问题是：社会科学的研究对象【或者“客体（objects）”】对我们的研究方法选择有很大的限定作用。简单地说，我们研究的客体有三个大类：

思想（idea, 比如，直觉、特定的观点、具体的科学理论或者洞见）、

行动（action; 或者intentional behavior），包括个人水平的行动以及集体水平的行动

社会结果（social outcomes，比如，革命、战争、族群冲突、经济增长、合作、民主化、民主巩固、民主崩溃、总体的社会流动pattern）这样的宏观结果，也有微观的结果，比如婚姻、获得某个职位或者地位、某个公司的兴衰等等）。

这里特别要强调，尽管思想和行动都是社会结果，我除了思想和行动之外的所有其他社会事实，均作为是“社会结果”。这样的一个划分是最为简洁但又是对我们理解认知论和方法论问题最为有用的划分。

我的基本结论是：一旦我们理解这些不同的客体的属性，对我们的研究是大有帮助的。这是因为不同的客体，需要不同的认知论和方法（论）。不同的客体需要不同的方法（通常是方法组合），而有些方法在理解某些客体上有帮助，可在理解其他一些客体上则帮助很小，甚至是误导性的。

因此，本文的讨论对于同学们选择学习什么样的方法技巧以及秉承什么样的认知论立场都是大有帮助的。本文是我最近在社会科学哲学方面做出的最重要的工作。

基于这项研究的讲座最开始在在我的《研究设计》课程里介绍过。改进之后，在上海财大、华东政法、北京大学也介绍过。感谢同学们以及其他地方的同仁和同学们的批评和指正。

核心观点提示（来自文章的Introduction）

Social scientists and philosophers of social sciences have been engaging in an ongoing and heated debate on what constitute the proper epistemologies and methodologies in social sciences. These debates have greatly improved our understanding of the tasks and the challenges of social sciences. This article contends, however, that something far more foundational has been missing from these debates: existing debates have failed to adequately grasp the key objects or explananda of social sciences at the ontological level and hence their implications for epistemology and methodology. As a result, social scientists may have missed some clear-cut solutions to some seemingly intractable impasses with epistemologies and methodologies and many issues have remained muddled.

I advance three key arguments. First, at the most foundational level, there are only three types of object or explananda in social sciences: idea, action, and (social) outcome. These three objects are ontologically different. Second, explaining these three different objects requires different epistemological perspectives and methodological tools, because these three explananda are ontologically different. As Bhaskar (1976[2008]; 1979[1998]) explicated a quarter of a century ago, ontology first, epistemology second, and methodology third: “though there isn’t necessarily a perfect correspondence between ontology, episte­mology, and methodology, they do constrain each other.” (Chatterjee 2013, 73; see also Bunge 2006, 59) Third and unfortunately, much, if not most, of the existing discussion on epistemology and methodology implicitly, if not explicitly, assumes that explaining the three objects requires roughly the same set of epistemologies and methodologies. This invalid assumption has been a major cause behind the intractability and much confusion within the major debates between different epistemologies and methodologies. As such, sorting out some of the thorny issues behind these three objects paves the way for further scientific progress.

Grasping the three key objects or explananda of social sciences and the three objects’ implications for epistemology and ontology adds not only critical empirical value to the pursuit of knowledge in social sciences but also critical pedagogical value. Nowadays, most graduate programs of social sciences demand students to master at least one, if not two or more, methods. Implicitly, graduate advisors and students take these methods to be of roughly equal value to our pursuit of knowledge and students’ career. My analysis shows that this implicit assumption is misplaced: some very fashionable methods are of very limited value whereas some are of more value, and yet none of them is omnipotent. Worse, some methodologies are ill-suited for certain objectives even if they are implemented flawlessly. As such, my analysis provides graduate students with a base for picking methodologies to be trained when they face the inevitable crunch of “too little time and too many things to learn.”