Huron, David. (1999). On Finding Field-Appropriate Methodologies at the Intersection of the Humanities and Sciences. Ernest Bloch Lecture, University of California, Berkeley, 1999. Full text.

I. Introduction

1. "Many scholars presume that methodological differences reflect fundamental philosophical disagreements concerning the nature of scholarly research. I think this view is only partly correct. As I will argue in this lecture, in most cases, the main methodological differences between disciplines can be traced to the materials and structures of particular fields of study. That is, differences in research methods typically reflect concrete differences between fields rather than reflecting a basic difference in philosophical outlook. (2)"
2. Philosophy of knowledge and methodology inform each other.
3. Research in the humanities often differs from research in the sciences in both philosophy (with postmodernism dominating the former and empiricism dominating the latter) and in methodology.
4. Methodology, more than philosophy, should be continually considered in both the sciences and in the humanities.

II. Philosophy of Knowledge

1. Empiricism
   • Problem of Induction
     • conventionalism - Duhem / instrumentalism
     • falsificationism - Popper
     • positivism
2. Postmoderism
   • Feyerabend - methodological "anything goes"
   • Kuhn - normal science vs. revolutionary science, incommensurability
   • Radical opposition to induction
     1. No privileged interpretation.
     2. All interpretations are equally valid.
     3. (#2 above does not necessarily follow from #1; methodology does not necessarily establish truth)

III. Philosophy of Methodology

Methodological differences that characterize different fields of study:

1. False-positive, theory-discarding skepticism ("There is insufficient evidence to support that.") vs. false-negative, theory-preserving skepticism ("There is insufficient evidence to reject that.")
   • Theory-conserving: tend to "open outwards," working with smaller amounts of data, include theories that either usurp other theories or coexist with them.
   • Theory-discarding: tend to distinguish between competing hypotheses, working with larger amounts of data.
2. High-risk vs. low-risk hypotheses
3. Retrospective vs. prospective data
   • Retrospective data susceptible to post-hoc theorizing
   • Prospective data includes experimental data (with data manipulated; causality can be established) and correlational data (with data unable to be manipulated; causality cannot be established)
4. Data-rich vs. data-poor fields
   • Positivist fallacy
   • Parsimony vs. pluralism
   • Reductionism/holism (often corresponding to humanism and mechanism)

IV. Conclusion

1. Methodologies should be chosen with their characteristics in mind.
2. "There is no methodological algorithm that ensures the advance of knowledge. Methodology consists primarily of a set of pointers that warn scholars of previously encountered pitfalls. Methodologies are extended and refined in the same manner as other theories. (32)"
3. "The principal impediment to careful selection of field-appropriate methods is the methodological inertia of a given discipline. (26)"