How to Tell Good Ideas from Bad Ideas (according to Andrew Abbott)

This is my third post on Andrew Abbott's excellent book, Methods of Discovery: Hueristics for the Social Sciences. Previous posts focused on:

• The major debates in the social sciences. Abbott really clarifies theoretical divisions in the social sciences (and thus in archaeology), a discussion I have found tremendously useful.
• Take Andrew Abbott's "taxi driver test."  This device is about being able to explain your research in a couple of minutes to a taxi driver. If you can't do this, then you don't really understand your research.

Here I will discuss some material from Abbott's chapter 7, "Ideas and Puzzles." Much of his book is concerned with how to come up with new ideas. In this last chapter of the book, he talks about some ways to tell the good ones from the bad ones. Here are some of the ways to do this.


(1) Test your ideas. 

"Obviously, the first test of an idea is to try it out, to run it past some data." (p.213). If you are going to test your ideas, they "must be framed in such a way that they can be wrong" (p.215). This sounds pretty basic, but Abbott observes, "It is quite surprising how many researchers -- even graduate students in their dissertations -- propose arguments that can't be wrong." (p.216). Such arguments are often classifying phenomena, or simply illustrating a theory. (Note, in Gabriel Abend's 2008 typology of theory, this kind of theory-illustration is typically associated with theory types 3, 5 and 7). Abbott goes on to state (p.218):

A good idea, then, ought to have some referent in the real world. This is not to deny the utility of pure social theory, but the vast majority of social theory consists of relabeling. All real theory arises in empirical work, in the attempt to make sense of the social world, no matter how abstractly construed. A student is well advised to stay clear of writing pure theory. It's an open invitation to vacuity. 

(Note: by "real theory" Abbott is probably referring to Abend's theory types 1 and 2). Abbott goes on to argue that Bourdieu's concept "practice" is just a relabeling of what others in sociology have long called "regular behavior."


(2) Get opinions from other people.

Get feedback from friends, colleagues, and others. "an idea that requires a huge amount of explanation is probably not a good idea." (p.222). Don't get mad at negative feedback. It often does not mean your idea is bad, but rather that you did not state it well. This is where he brings up the "taxi driver test." And in reference to colleagues, Abbott suggests that "the ability of others to restate  your idea clearly is the watershed." This signals that you have a good idea and that you have explained it well.


Abbott goes on to discuss other ways to evaluate ideas, including how they fit in the literature, good taste (that is, good taste in scholarship and academic phenomena), and personality. This is a fascinating chapter, full of useful insights.


Abbott, Andrew
2004    Methods of Discovery: Heuristics for the Social Sciences. Norton, New York.

Abend, Gabriel
2008    The Meaning of "Theory". Sociological Theory 26:173-199