Interpretivism is perverse positivism.

That's an interpretive statement, and a far more complex one than the parallel "Diane Wilson is badass." A proper name has a proper referent, "badass" is certainly open to question and disagreement, but I'm confident that 95% of readers would, after some further argument and thought, say "Yes, that is true, Diane Wilson is badass." But each of the three terms in the other sentence -- interpretivism, perverse, positivism -- begs for additional interpretation, and the interpretive differences will quickly add up and compound. Yet I'll insist that that simplest of copulas, is, identifying and equating the terms on either side of it, is just as deserved in that case as it is in the statement regarding Diane Wilson.

Consider each as a product of qualitative data analysis. They each need meta/data sets of varying size and components that make up their evidentiary "ground," to stay with the conventional but suboptimal metaphorics. And, in qualitative data analysis as conventionally conceived and practiced, each of their respective grounds of meta/data requires some kind of analytic linkage, some procedure or protocol of reasoning that is embodied in a set of interrelated "codes" resulting from expert "coding" practices. More and more, qualitative researchers -- social scientists -- are making their qualitative meta/data open and accessible, and making their coding structures and practices also accessible. Like all scientists these days, qualitative researchers enact an ethos of "open data" and "analytic transparency" that underwrite the trustworthiness and reliability of that knowledge. Many social scientists follow the natural scientists even further in this ethos, maintaing that all this works toward and is intended to guarantee the reproducibility of that knowledge. Anyone who accesses the meta/data and recreates the analytic procedures will arrive at the same conclusion, the same is, with no additional (interpretive) analysis necessary. This is a view of knowledge, of the natural or the social world, simplified here as positivism.

Interpretivism, to continue the simplifications, is something other than positivism -- even if it's just "post-positivist," as some would have it. (I am leaving aside all the different versions of "positivism," past and present, and all the different interpretations of "interpretivism," that are slung around in academic pie fights.) One difference, maybe the difference, is that positivism removes the knowing person or subject from the equation, or at least tries or pretends to. This works best when all your meta/data is quantitative. But even researchers who work with qualitative meta/data and analysis will maintain that they at least want to minimize the effects (almost always regarded as negative) of the knowing subject -- hence the importance of archiving the codes and coding systems, and making them accessible. In such a conception, qualitative analysis counts as positivist, leaving as little room as possible for interpretation -- or rather, filling up all the room that interpretation needs with codes and the codes connecting the codes, so that any one knowing subject's analysis can be reproduced by any other knowing subject.

But our evaluations of each -- are they true, trustworthy statements of knowledge? -- abide by the same set of logics, take place in the same analytic context.

Interpretivism imitates positivism, even if it is a somewhat paler imitation.

PECE shares these shared contexts and logics. It's why we've sometimes been accused or positivism. But where we are fully committed to the principles of meta/data and analytic transparency and have built this infrastucture to support them, we do not hold to or pursure an ideal of reproducibility for our avowedly interpretive statements.

PECE shares these logics and contexts -- indeed, we cannot not share them. It's why we've sometimes been accused of being secret positivists, or pursuing some kind of positivism. But while we are fully committed to the principles of meta/data and analytic transparency -- open science -- and have built this infrastructure to support that project for qualitative analysis, we do so not in pursuit of the positivist reproducibility for our avowedly interpretive statements. We are keen to show our analytic work and our meta/data to anyone who wants to see it. But we don't expect them to reproduce our truths; we expect them to take hold of the disseminating forces of meta/data, their archives, and their interpretive re-codings and produce something different and new. But this does not entail some kind of souped-up interpretivism, outside of the dominant positivist knowledge culture. We are going for a third position that is both/and and neither/nor, as represented in the analytic claim, "Interpretivism is perverse positivism."

I wouldn’t regard that analytic claim itself as “reproducible.” I would not expect some other social scientist or empirical humanist, as we prefer to call ourselves, to pore over and analyze the same materials I have, follow the same lines of analysis, and arrive at that same analytic claim. But I also refuse to reduce it to "my interpretation," since it is not actually "mine" and we've yet to figure out what makes a statement "interpretive." Here is where we will need to shift from an epistemological register to an affective one.

In the dominant epistemic culture, in our dominant scientific culture, in the dominant positivistic knowledge culture of (to be as crude as possible about it) "the West," an interpretive statement is Other to a scientific truth. They are supposed to be mutually exclusive: scientific, positivist truths are not subject to or open to interpretation. Scientific, positivist truths have no essential attachment to a knowing subject, and they are closed to any difference or dissent. But positivist and interpretivist statements are not just different, not just Other to each other, not just made in different sectors of the university campus according to different sets of rules and concepts. I've been involved in too many discussions of these matters that leave it at that, as simple difference. But shift to a more analytic register, and it becomes clear that not only are positivist statements regarded as having greater epistemic value and cultural authority, they are better. Positivism is more virtuous than interpretivism. Positivism is the ideal and, most importantly, the norm, and interpretivism is at best a pale imitation ---wannabe positivism -- and more often just...well, not normal. (Not that there's anything wrong with that...)

But we've seen that "interpretation" is not some kind of disqualifying qualifier that "real sciences" have no need of. We've seen that data is always meta/data, always divided and attached to an Other to be itself, always "about" itself, always having an "about" that needs to be read. Do you really think the situation isn’t at least a little more perverse than the normally operative oppositions would suggest: real or interpreted, quantitative or qualitative, normal or perverse, hard modernist positivist rationality or squishy postmodern depositivist bullshit? I know that it is, truly, a little more perverse than that, and rather than pathologizing that perversity and trying to purify it out of the system, we should be listening more carefully to it, having richer and more open-ended conversations within it, and even encouraging it

Positivism or interpretivism? The perverse reality is that each needs the other, relies on the other, is supplemented by the other. Interpretivism is not the other of positivism, but its perversion.  Not nonoverlapping magisterial of knowledge, but one  not two cultures, and not some bullshit dialectic third culture, but  Cultures

PECE is a collaborative effort to build digital infrastructure to better foreground and explore these paradoxes and contradictions that inhabit and indeed drive every complex analytic knowledge system, interpretive and/or positivist alike. We have our doubts about it all – how could we not?  Doubt is, after all, perversely integral to the truth-making practices that we, as cultural anthropologists, both document and study among environmental, health and data scientists, and are ourselves informed and traversed by. I try to articulate some of them here—fulfilling another part of what I think my charge was—while embedding those misgivings, properly if paradoxically, within our unshatterable commitments to data availability, transparency, and reliable and relevant truth—and their difficult perversities.

Positivism is anxious about non-reproducibility, it’s most dreaded perversion. But this retroping gives insight I think: non-reproducibility is not so much a failure of positivist knowledge as its ineradicable or irreducible limit.  The perversion of non-reproducibility will always be part of doing science, but instead of normalizing that, our culture pathologizes it. No amount of availability or transparency will fully ward it off or exclude it.  No matter how big you make Big Data, as a protective fetish against reproductive failure, it can’t be big enough.

Take the perverse/normal couplet in this quote from anthropologist-turned-psychoanalyst Muriel Dimen:

Perversion and that inadequately specific term normality construct each other…[P]erversion and heteronormality constitute each other’s limits. Perversion marks the boundary across which you become an outlaw. Normality marks off the territory that, if stayed inside, keeps you safe from shame, disgust, and anxiety.” (Dimen 2001: 838)

And substitute the interpretivist/positivist couplet:

Interpretivism and that inadequately specific term positivism construct each other…Interpretivism and positivism constitute each other’s limits. Interpretivism marks the boundary across which you become an outlaw. Positivism marks off the territory that, if stayed inside, keeps you safe from shame, disgust, and anxiety.”

Doesn't that help you begin to make a different kind of sense of C.P. Snow's famed "Two Cultures" of the sciences and the Humanities? Can we start to see positivism and interpretivism not as two cultures, but not as one, either? Can we start to make sense of them as "the √2 cultures," with an infinity of interminable differences within?