III.2. Interpretive methods
To participate, you may either post a contribution to an existing discussion by selecting the thread for that topic (and then click on "Post Reply") or start a new thread by clicking on "New Topic" below.
For instructions on how to follow a discussion thread by email, click here.
William J. Kelleher, Ph.D.
- Posts: 19
- Joined: Thu Apr 07, 2016 4:38 pm
Critique of the Self-Defeating Concessions Made to Positivism in the QMMR Spring 2015 Issue
I applaud the editors and contributors to that issue, and those who have set up this site for further discussion so as to include a broader number of participants. This is very democratic!
Since the time for discussion is winding down, I would like to briefly mention some matters that seem to me to have been over-looked so far. In their efforts to be comprehensive, reasonable, and co-operative, the editors of QMMR15 have conceded positions to what I regard as the positivistic adversaries of interpretive political science. As something of a Kuhnian, I define this situation as one of paradigm politics in which the identity of the profession is at stake. Making the concessions I will discuss can only encourage the advocates of the dominant paradigm to stay the course.
First, the editors seem to accept the DA-RT proponents’ terms of the debate. They write that the three principal forms of transparency identified in the APSA 2012 ethics guidelines are: production transparency, analytic transparency, and data access (p3). While expressing reluctance about doing so, they also accept the definitions of “qualitative” and “quantitative” used by the DA-RT positivists (see p58f). Then they concede to the usage of “transparency” as a “meta-standard” for both of the two approaches (p2). In my view, the editors don’t sufficiently stress the point that the aim of the transparency ideal is to facilitate the replication of studies. The positivists believe that replication is the surest way to falsify or verify the results of a study.
Not a Meta-Standard
These concessions to the positivistic terminology needlessly burden the presentation of the interpretive political science point of view by forcing it to express itself in alien concepts. In my proposed Community Transparency Statement (CTS) for Interpretive Political Science I have explained why the notion of transparency-for-replication is utterly irrelevant to interpretive political science (on this QTD website, at viewtopic.php?f=21&t=195#p800 .)
In short, unique persons conduct unique studies of unique subjects. As a matter of fact, none of those elements can be replicated. Hence, transparency-for-replication cannot be a “meta-standard” for a research community to which it is inapplicable and irrelevant. (Also see Kathy Cramer’s insightful critique of replicability on this site at
To call the DA-RT proponents’ categories “meta-standards” legitimizes the implication that those standards are relevant for interpretive political science, and thus places interpretivists in a position of subjection to the already dominant paradigm. Interpretive/qualitative political science can and ought to develop its own standards, based on its own self-awareness as a distinct methodology.
The editors show some awareness that interpretive studies are not replicable. But rather than rejecting positivistic concepts as inauthentic to the interpretive/qualitative methods, they offer an awkward verbal concession to the dominant paradigm. They call it “replication-in-thought.” Of course, there is no more replication in thought than there is in deed. Yet some guidelines for validating procedures that are applicable to interpretive studies could be useful.
I agree with the editors that validating studies generally requires “the provision of sufficient information to allow readers to trace the reasoning and analytic steps leading from observation to conclusions, and think through the processes of observation or engagement.” (p57) I also agree with their view that the convergence of findings following from similar studies would be an important consideration for measuring the degree of acceptability of the knowledge claims under scrutiny. (p57) Similarity of studies may be possible, and the methods and results of such studies can be compared and contrasted.
These are factors to be considered by the relevant community of experts when critically evaluating the validity of knowledge claims. So, instead of making a concession to the term "replication," why not create a more authentic category, like “Standards for Evaluating the Validity of Interpretive Knowledge Claims”? (I discuss the problem of validation in several other posts on this site.)
The editors concede the positivist definition of “production transparency” as “a full account of the procedures used to collect or generate the data” (p3). But this demand for “a full account of the procedures used” is a positivistic ideal, and impossible for interpretive political science to comply with. The ideal itself condemns interpretive approaches to a less than “scientific” status.
This ethical rule presupposes a kind of “objective” procedure, akin to an entomologist lining up a collection of insects in the course of categorizing them. But interpretive work, such as participant observation and case studies, is not at all like observing bugs in a lab. The participant observer is one of the subjects. Her or his mind is working in a way that cannot be set out step-by-step so that it can be replicated. The mind is alert, often absorbing details unconsciously. Generalizations can spring into consciousness after they have been formed below consciousness. Meanings can be understood intuitively, and in a very personal way. How can a “full account” of this be given?
Interpretive political science ought not submit itself to the domination of the positivist’s conception of “evidence gathering,” but should formulate a conception based on its own way of doing science.
Openness about method is important, and some degree of discussion about how a study was conducted ought to be recorded, so that light can be shed on how the knowledge claims were arrived at. But since a “full account” isn’t possible, none should be expected, and standards of reasonableness under the circumstances should be applied.
RE: the second rule – “In the qualitative tradition, analytic transparency typically means making verbally explicit the logical steps or interpretive processes linking observations to conclusions or understandings.” (p3)
Here, again, too much is conceded to the dominant positivism. For interpretation, “conclusions or understandings” come out of a person’s mind in ways that even he or she doesn’t fully understand. It’s a creative process. There are no “logical steps” to make “verbally explicit,” so that some positivist can come along and replicate them. The aim of validating interpretive knowledge claims must be satisfied in some other way.
What is required to validate interpretive knowledge claims is peer criticism based on articulated standards appropriate to the claims being made. As Easton has said, weighing causes, for example, is part of the political scientist’s work. What were the meanings, or reasons, to those who voted for Trump? Attempts may be made to answer this question with interpretive empirical evidence. Criticizing these efforts will require a combination of logic, informed judgment, and expert consensus to establish the degree of validity of the explanation, and to show how well it fits in the body of professional knowledge.
Thirdly, “data access” is supposed to promote “transparency.” Once again, positivistic presuppositions that are incompatible with the needs of interpretation ought to be highlighted and deleted, not conceded. The QMMR article states that, among other things, “Data access also allows other researchers to replicate and evaluate analytic procedures using the original evidence.” (p4) Of course, the notions of “replication” and of “using the original evidence” have no place in interpretive political science. The “original evidence” of Goffman’s study of inner city youths, for example, is gone. It was the people and conditions in which she immersed herself. A positivistic political scientist can no more replicate those conditions than he can replicate her. (See my discussion of Goffman’s study at
viewtopic.php?f=18&t=200&p=846&hilit=Goffman#p846 and other posts on this site)
However, questions can be raised as to whether or not, or to what extent, she wrote up a “faithful representation” of her experience. Like her methods, her interpretations can be criticized on the grounds of logic and reasonableness. Here, too, interpretive political scientists ought to focus on formulating their own standards for conducting studies and for validating knowledge claims, and not submit to the dictates of the currently dominant point of view.
The positivistic term “data access” skews the problem of validating knowledge claims. It implies that “data” comes in bits that can be examined by anyone to test the inferences drawn therefrom. The term’s quantitative bias immediately disadvantages meanings as the object of research, because meanings always occur in contexts. Meanings are lost if one focuses solely on their constituent parts. Just analyze the letters of a word, and see if the word thus loses its meaning.
The Extractive/Immersive False Dichotomy
The QMMR15 discussion suggests that “One might more broadly conceive of production transparency as a detailed account of the process of empirical engagement—whether this engagement is more immersive (in the mode of ethnography) or extractive (in the mode, e.g., of survey research or document-collection).” (p3)
First, what is interpretive “empirical engagement”? It is one unique person empathically interpreting the meanings of other unique persons in a fluid and unrepeatable context. As mentioned, how much detail is either necessary or possible to give depends on the context in which a study is done.
But accepting this distinction between “immersive” and “extractive” processes is also a concession to the positivists. As Timothy Pachirat points out, “extractive” is a mining metaphor which de-humanizes the subject, attempts to objectify the method, and reduces human meaning to a type of ore, or “data.” (See his insightful article at QMMR15, p27f)
Recognizing empathic interpretation as the way in which knowledge claims are arrived at acknowledges the common humanity of researchers and subjects. Only by assuming a human-to-human relationship can a researcher understand the meanings of his or her subjects as they engage in political behavior, or hold political beliefs and opinions. While “extraction” digs ore from the ground, or pulls teeth from a mouth, “empathy” is the entering into of another person’s feelings, meanings, and intentions. As Pachirat says, this is an intersubjective empirical engagement.
In actual practice, some of the spokespersons for positivism seem to be unaware of what they are doing when they empirically engage political subject matter. In my view, they begin their work in a condition of immersion, rather than being somehow above it.
When a study is conceived of, and designed, a researcher is already immersed in his or her understanding that other people have opinions and meanings which can be reached, or “extracted,” through quantitative methods. A researcher is steeped in intersubjectivity (what Polanyi calls “conviviality” in Personal Knowledge), from the beginning. The questions to be asked have been formulated in a researcher’s mind through the use of their own empathic understanding that there are particular kinds of meanings out there to be aggregated and quantified. Once the responses have been gathered, they are empathically interpreted as human responses by the interpreting humans. Again, the quantitativists is immersed in meanings.
Thus, to contrast the so-called “extractive” with “immersive” forms of research creates a false distinction. Quantitative studies necessarily presuppose the immersion of their designers in human meaning and in an awareness that other minds, like their own, exist.
Quants claim the so-called objectivity of methods that physics and chemistry seem to have. That is, the human physicist seems to stand without inter-relationship to inanimate objects and studies their behavior following physical causes. The chemist seems to be equally unrelated to the chemical compounds he analyzes. But the political scientists who portray themselves as thus detached are telling themselves a false story. Without the obscuring desire to be like natural scientists, the relation of the political scientist to his or her subjects can be seen as that of people-to-people. The positivist political scientist is a person whose mind is full of meanings, just as are the persons engaged in political behavior. To understand that behavior, the political scientist must empathically put himself or herself in the place of the actors under study. Intersubjectivity rather than objectivity is the relationship of the political scientist to his or her object of study.
Thus, all political science is interpretive. Were it not for the false self-understanding of the positivists, the qualifier “interpretive” would not be necessary. Only because they lack the self-awareness of their reliance on empathy and their human-to-human relationship to their subjects is there controversy over the scope and methods of political science. Whether designing a large-N study, or a mathematical model, they are presupposing other minds, and thus practicing interpretive political science, but they are in denial about doing so.
What is to be Done?
Because all political science depends upon empathic interpretation, more resources and attention ought to be invested in training graduate students in the exercise of professional empathy. The uses of empathy ought to be raised from the depths of denial to the level of conscious and critical attention.
Unfortunately, Quantitativists and Qualitativists are not in a “separate but equal” relationship. Positivists control the lion’s share of the profession’s resources. Recently, a whopping 27 journal editors signed the JETS agreement to enforce positivistic standards over submissions. They have the status, prestigious positions, and institutional resources to attract graduate students to be indoctrinated into their false self-image. They are the paradigm powerful.
They justify their exercise of this raw power by telling themselves that their approaches are superior in “objectivity,” scientific method, and scientific knowledge. But the delusional character of these doctrines are also their weakness.
After chiding Israeli Prime Minister Benjamin Netanyahu for increasing settlement building in the West Bank, Secretary of State John Kerry declared at a press conference, “Friends need to tell each other the hard truths.” (News Hour, 12-29-16) And so, we must be willing to speak truth to power, and tell our friends and colleagues the hard truths. That seems to be the only way we have of resisting and perhaps reversing the paradigm domination of the elite positivist political scientists. We should make no concessions to their false self-image. That would only encourage them.
William J. Kelleher, Ph.D.