the real philosophy of social science puzzle

There is an intrinsic interest in the philosophy of social science. Ideally, we all want well motivated and logical explanations for how we should do our professional work. However, there is usually one question that you don’t hear much about – why does scholarship seem to progress in the absence of a well motivated philosophy? In other words, doctors probably have a bad philosophy of science, but I don’t see philosophers refusing the services of their physicians.

I don’t have an answer to this, I’ve only started to think about this issue. But I raise it in the shadow of our debate over critical realism and the earlier debate over post-modernism. The claim of some supporters is that social scientists really need a new theory of social science (e.g., critical realism) because social scientists rely on a flawed positivist theory. It may be true that positivist social science is wrong and that we should adopt a newer theory. This view does not take into account two issues: (a) The cost of adopting a new theory is steep. If Kieran can’t quite get critical theory after reading it for 18 months, then I sure won’t get it. (b) A new social science that proceeds along new rules of engagement may not generate enough differences to make it worthwhile. For example, now that Phil Gorski has adopted critical realism, how would his book, The Disciplinary Revolution, be written any differently? Not clear to me since  a lot of what Gorski does in that book is apply a specific theoretical lens in reading various developments in state formation. He might sprinkle a discussion of “multiple levels of causation” at the top but then he’d probably proceed to make similar arguments with similar data.

The ultimate puzzle, though, is in areas that seem to make progress even when practitioners work with a bad philosophy. This suggests that the demand for better foundations simply isn’t important for generating knowledge. Another datum is that advances in science, or social science, rarely require entirely new foundations. Take sociology. I don’t need to adopt anything new to, say, appreciate Swidler’s attack on functionalism. And I seem to be able to understand most feminist sociology by using meat and potatoes positivism. The bottom line is that, at the very least, there needs to be an explanation for the ubiquitous disjuncture between foundations and practice.

50+ chapters of grad skool advice goodness: From Black Power/Grad Skool Rulz 

Written by fabiorojas

March 13, 2014 at 12:30 am

8 Responses

Subscribe to comments with RSS.

  1. I’ve always been somewhat fond of pragmatism, but I like falsification, too. Generally, we do a lot more describing than predicting, so it probably doesn’t matter what our philosophy of science is (or that we don’t have any real consensus on the issue).



    March 13, 2014 at 4:16 am

  2. But how we know what progress is if we can’t agree on the epistemological foundation of the claims or study we’re evaluating? What you define as progress in our knowledge may be confined within a specific epistemology. For example, we might claim that our understanding of the causes of war have progressed because we know have better external validity with new datasets and larger sample. But the notion of external validity itself is rooted in a specific epistemology. A critical epistemology might argue that progress should be measured by emancipation and not generalizability of arguments. So while debating new or old epistemology may not lead us to converge on a singular form, we do need to clarify the epistemological foundations of our research to ensure coherence and consistency–and allow others to adjudicate our work better.


    Evan Laksmana

    March 13, 2014 at 1:36 pm

  3. @evan: I think you put your finger on the sticking point, Evan: are coherence and consistency within the discipline important? Will they make the field more relevant and effective in its contributions to the greater good? Those who think yes are dissatisfied with the current state of affairs, hence the calls for a philosophy of science to replace the reliance on positivism (or pragmatism, for that matter).



    March 13, 2014 at 4:43 pm

  4. A sociologist without a philosophy is like a fish without a bicycle.



    March 13, 2014 at 6:06 pm

  5. I’m leaning on the side of yes. Coherence and consistency is important if we want to hold on the notion that every social science endeavor is a collective project of building a house brick by brick. Without epistemological clarity, one scholar believes it’s okay to build the house with bricks while another believes sponges is more crucial to the house. But that doesn’t mean we should only have or build one type of house for all the neighborhood (i.e. one epistemology to rule them all). Epistemological pluralism is now a fact of our profession. All that’s left is for us to make our foundations clear and allow our work to be assessed, or as we evaluate others’, within those foundations. Otherwise, we’ll keep talking past each other. Critical theory work cannot be satisfactorily evaluated within a Humean epistemology and vice versa. So no, we don’t necessarily need a new epistemology to unify the social sciences but we do need to clarify our foundations if theory adjudication or evaluation remains one of the core tasks of our profession.


    Evan Laksmana

    March 14, 2014 at 12:14 pm

  6. Great post Fabio. I like Evan’s comments a lot too.

    Let me try one way of directing this conversation in an empirical direction, could we evaluate the predominant philosophy of science guiding a discipline, or sub-discipline, or scholar, and then use that to predict something, e.g. the continuation/growth of that research area, the citations that area receives from other areas, measures of that area’s influence on the larger culture?


    Michael Bishop

    March 18, 2014 at 6:36 pm

  7. Given our current databases and data management software, I think that’s possible. But most published empirical social science work are not explicit about their epistemological foundations (hence why theory adjudication is a marginalized task). So we would need to create a “coding rule” based on a detailed typology of existing social science epistemologies. This is as far as classifying the existing work goes. Measuring impact as a consequence of epistemological coherence would be trickier but might be an interesting project indeed.


    Evan Laksmana

    March 18, 2014 at 9:47 pm

  8. The most unrealistic thing to me about Anathem was that the science-monks in the novel were primarily organized by epistemological assumptions rather than substantive objects of inquiry. Relative to that all the multiverse stuff and neologisms made total sense.



    March 19, 2014 at 6:09 pm

Comments are closed.

%d bloggers like this: