Methods of Discovery

There is a pretty sizable “academic self-help” literature out of there of books designed to help you learn the “tricks of the trade” or “finish your dissertation in 10 minutes a day” which vary wildly in quality. I just finished Andrew Abbott’s “Methods of Discovery: Heuristics for the Social Sciences”: which I enjoyed and which is wrapped in a shroud of mystery. With little in the way of an online-preview and a not-too-informative table of contents (what are “search heuristics?”) the book could get overlooked. But at fifteen dollars a pop I think it is worth a look — while I’m not sure how a student learning how to do research would view it, as a professor who does research and teaches I found it very enlightening.

The book is divided into three sections: chapters one and two, which cover the “aims, means, and assumptions of social science research”, a long middle section of chapters three, four, and five, which is about ways to think about research — roughly, “how to have a new idea”. The final two sections form a sort of coda — in chapter six Abbott pushes his own theory of ‘fractal heuristics’ and in the last chapter he talks about more abstract life issues that help one become better at figuring stuff out, like sharing thoughts with colleagues and reading widely.

The book is very clearly written in Abbott’s distinct voice and also very schematic. Thus in his overview of how social science works — which is nice to crib off of in case you hadn’t put a lot of thought into that lately — there are three types of explanations social scientists seek (pragmatic, semantic, and syntactic), four methods of gathering data (ethnography, surveys, record-based analysis, and history), and three scales of cases that are considered (the case-study, small-N, and large-N). As a result its readable and skimmable.

Throughout the book he introduces each of these typologies, then amplifies on them, and then gives examples of papers or books that exemplify them — the result that really lets you get your head around each of them and opens up your imagination through imagining different combinations (what would large-N pragmatic explanation ethnography look like?). As someone with a history of thinking about Chicago Sociology Abbott does not slight ethnography, which is nice for anthropologists.

His discussion of different basic positions in social science is also worthwhile because he actually covers them. Too often books like this tell students that there are ‘schools’ of thought like ‘poststructuralism’ or ‘interpretive’ anthropology, but with a very few execeptions few of these terms designate actual bodies of scholars with an agreed set of principles or methods. Instead these ‘schools’ are a strange mix of disciplinary history and preconceptions about how research is done which typically lack an actual rigorous empirical examination to summarize them. Abbott’s ‘great conflicts in approaches’ might seem experience-distant to some of us (are you an ’emergentist’? A ‘neocontextualist’?) but the categories do actually describe tendencies or modes of thought and are very useful to think through.

The core of the book are the three chapters on how to have new ideas. The goal, for Abbott, is to move from a position where you don’t have anything to say because you are ungrounded, to one where you have a certain ‘comfortable one-sidedness’ and have gotten good at using one position to illuminate data, to recognizing there are other viewpoints out there, to finally being able to generate your own distinct viewpoint by playing other viewpoints against one another. For example, he thus moves from ‘additive’ methods where you use commonplaces or a list of topics to execute a particular research program (feminism is an example: what about a gendered concept of X? What about women in relations to Z?) to generating new ideas through ‘fractal heuristics’ where you play off classical social science standpoints against one another.

Between these two extremes are some useful recommendation in chapters four and five. Behind the unintuitive names are some very good ideas. ‘Search heuristics’, for instance, is a fancy way of saying ‘poaching new ideas from others’: either by making an analogy (Chicago School sociology, for instance, involved making an analogy between biological and social systems and imagining the city as an ecology) or by borrowing a method (what if we used network analysis on people?). ‘Argument heuristics’ is ways to turn ideas on their heads to get a new purchase on them: problematize the obvious, reverse an argument, and so forth.

It is hard to tell whether a book is ‘good to teach’ just because it does intellectual work for you. My point here is just that Abbott’s book is much more down to earth and useful than the scarce preview material might suggest. For people looking for a bit of inspiration or trying to figure out what exactly they are doing in their dissertation, its a useful (and affordable) resource that I’d recommend.


Alex Golub is an associate professor of anthropology at the University of Hawai‘i at Mānoa. His book Leviathans at The Gold Mine has been published by Duke University Press. You can contact him at

One thought on “Methods of Discovery

Comments are closed.