In a recent book review in Perspectives on Politics, Chris Blattman thinks aloud on the moral problems of doing research on violence and conflict. The dilemma he describes, indeed, appears to be crucial: as scholars we should strive at keeping a critical and reflective distance to the subjects studied and this might, in the case of particularly mediatized cases like child soldiering, appear as cold and cynical reaction to what is clearly morally most appaling and horrible: “You catch yourself speaking of genocide or child abduction as variables and forgetting they are atrocities” (Perspectives on Politics, 403).
This sounds so right and very reflective and, yet, there’s something misleading with the assertion that the ethical problem of conflict analysis would be the gap between the inhumanity of the violent crimes and the coldness of the objective researcher, alone. This would imply that the moral motivations of the researcher might be questionable but, honestly, this is not the case, is it? We can safely assume that anyone who is interested in disentangling reasons and causes for war’s miseries does so, exactly because it is such a horror, and for this motivation the method by which research is undertaken, whether by statistical variable approaches or through qualitative ethnographic approaches, does not really matter. Nobody would suggest either that a medical researcher who treats cancer as variable does so out of disrespect for people dying from the disease; everyone sensibly assumes that researchers do research because they want to find answers and eventually cures.
In conflict analysis it is not so much the moral integrity of the researcher that is at stake but rather something more profoundly disquieting, namely the moral integrity of research itself, particularly of this kind of research. It is crucially the problem that our entire scientific, rational thinking and our desperate search for certainty might be less a solution than a part of problem of contemporary large-scale, collective violence.
Superficially, one could think of the obvious problem that research and research findings might be wrong or wrongly used. All research risks abuse, misunderstanding, and distortion through practice, and at the simple sight of much research in conflict analysis, particularly of “peace studies”, one could easily believe that the main ethical challenge lies in giving the “right” and to avoid “wrong” advice to practitioners and policy-makers. For instance, the decade-long obsession with “greed” and “warlords” in conflict analysis has heavily influenced major development agencies, from the World Bank to DAC passing by national agencies like DFID, before it became evident in research as well as in policy-making (although at a much slower pace) that there is more to collective violence than diamonds.
Yet again, focusing on the results of research and its possible usages hides from sight the more fundamental epistemological problem that the way how scientific research organizes knowledge, perceives and conceives of its object and proceeds in treating knowledge matters as objects. It is, indeed, much more in the search for certainty and for control of the unknown, unknowable and unconceivable that social sciences (and humanities) come to their limits when they deal with human violence. This double face of reason, being the condition for understanding but maybe also the seed for inhumane destruction have been cogently analysed Theodor W. Adorno and Herbert Marcuse in their Dialectics of Enlightment and Negative Dialectics. They demonstrated with Kant’s distinction between pure and practical reason that it is exactly the pure reason of enlightenment that has produced a sense of rationality in which moral judgment (which is always linked to the practice of life, i.e. to practical reason) has disappeared. Reducing reason to pure reason and situating moral imperatives in the domain of pure reason, makes rationality become “purposeless efficacy” of which the subject is irrelevant in the sense that it is the efficacy that counts…whether the efficacy of Auschwitz or the efficacy of the categorical imperative.
This is so, not because of pure reason itself, but because of the way pure reason has been integrated and become the very rationality of the capitalist production mode. Adorno’s and Horkheimer’s analysis is firmly although critically rooted in Marxist theory. Adorno and Horkheimer saw inherent affinity between Kant’s pure reason which eliminates the raison d’etre of practice and the commodification of all spheres of life and humanity through industrial and capitalist production. Everyone and everything becomes a thing as everyone and everything can be produced, and everyone and everything can, categorically, be thought without the experience of the practical world devising which and who of these everythings or everyones is actually unique, for instance as mother, as childhood memory, as “magic object”. Proust’s madeleine does not exist, only the combination of butter, eggs, almonds and orange scent baked at a given temperature, reproducable and edible. Pure reason has stripped off meaning, most particularly moral meaning, and left only purpose.
Behaviouralist research and particularly liberal-economist models of behaviouralist research tend to fall into exactly that trap although it is, to repeat, not the quantitative, statistical method as such that is problematic. Obviously, treating atrocities as variables reifies them yet the essential problem is that it does so on the grounds of an epistemology that is utterly unaware of the logic of reification. The aim of treating atrocities (or humans or ideas or anything else for that matter) as variables is to stripp them off their particular meaning and extract them (abstract them, Kant would say) from their concrete context, the rational being that it is exactly this operation that will produce knowledge about the item at stake.
Of course, this already presents a source of error as apples might be compared to pears but again this is only the methodological challenge. The moral and ethical challenge is rather that the logic by which behaviouralist research transforms everything into variables remains deeply unreflective of its own conditions of existence, its own rules and its own, tacit, even unconscious assumptions.
This is particularly evident for the homo oeconomicus assumption, the point being not so much that it might be wrong (it actually is) but that it’s condition of existence is the requirement of pure reason that the condition of our insight is not the particular, practical, contextual being but the abstract, reasonable idea. The homo oeconomicus exists because it is logic of liberal economic deductive models that he/she must exist. The constrution of the model of analysis follows the logic of the model and not the logic of practical life, as the epistemological premise is that pure reason would allow us to construct a logically consistent, and therefore an inalienable true explanation. There are a number of other basic assumptions and basic logical operations of our scientific models which, similarly, are mainly consistent with the scientific models to which they belong but not with practical life; they are constructed on the grounds of the pure reason of the argument, commonly an economic argument, but not on the grounds of the practical contexts of the human lives they are applied to.
If morality is to be reintegrated into conflict analysis, this logic for the purpose of the model has to be abandoned. This means that the epistemological approach has to be changed to allow practice and context as basis of insight, alongside pure reason. This infers a triple challenge : one, to deconstruct the logic of such a pure reason epistemology; second, to reflect upon the practical implication of any modelling and epistemology of conflict analysis; third, to reconstruct research that is conscious of the ethical hurdles of researching violence.
Presenting purely phenomenological accounts of the impressions violence and war make on the observer — something of the kind that large parts in Carolyne Nordstrom’s books do — is an unsatisfactory answer to Adorno/Horkheimer’s challenge. It rather avoids addressing the ethical problem of doing research by trying to make the researcher become a pure canvas on which the stories of violence are written. The motivation to subordinate the researcher’s personality to the narrative seems largely guided by the remorse of the survivor for not having shared the pain and misery of those whose stories we are recording. The ethical dilemma is then one of injustice between the observer and observed (which nevertheless might have the beneficial effect of preserving the subjective agency of the observed).
The path out of this aporia seems rather to make the logic of investigation a part of the investigation itself; to reflect upon the place this investigation has in a wider social net of interacting ideas and behaviours by including the question “why do I as a researcher think about the violence I’m investigating in this way?” rather than asking only “why do they (the actors) do this?”. If we do want to maintain the premise of reason’s unicity then both quesetions need to part of the same investigation.