Tag: Methodology411

Methodology411: feminist methodology?

Other duties have prevented me from posting as much as I’d like to for the past couple of months, but now that the summer is really upon us I am going to try to make more regular appearances. I received a query from a reader that seemed quite appropriate for the Methodology411, so here we go:

Mada asks: “Is there a feminist methodology? And if so, what does it consist of?”

By way of kick-starting a discussion, I reply: You ask a deceptively complicated question, one that feminist scholars have been wrestling with for a long time. Part of the complexity stems from the fact that there are many different operational definitions of feminism, some of which lend themselves to methodological reflection and some of which explicitly reject any such reflection as inherently problematic. And to compound the issue, some extremely broad definitions of feminism — such as “the radical proposition that women are people” or “anything that helps to advance the status of women” — would more or less deliberately encompass a variety of methodologies, and thus answer your question by saying something like “there are many different ways to do feminist work, and any attempt to define a feminist methodology is likely to cut some of them off, so we should avoid the question altogether.”

That said, self-identified feminist scholars working in the social sciences usually share a rejection of neopositivist hypothesis-testing as the sole or even a preferred mode of knowledge-production. My colleague Charli Carpenter stirred up quite a hornet’s nest of controversy by using that kind of methodological approach to study questions about the impact of sex and gender on world politics; to her great credit, she explicitly framed what she was doing as a “non-feminist standpoint,” using a convenient shorthand made possible by the general feminist rejection of neopositivist hypothesis-testing throughout the social sciences, especially in IR.

What feminists often prefer are research techniques that stress personal experience rather than general abstraction as the foundation of valid knowledge-claims. “The personal is political” is a well-known feminist rallying cry, and a lot of feminist scholars take that to heart in seeking to ground their analyses in the personal experiences of their research subjects. This does not mean that feminist scholarship is somehow exclusively about personal feelings and impressions, but it does mean that knowledge that does not come from a personal standpoint is relatively valueless — especially when compared to that knowledge that can be gained by explicitly adopting the standpoint of the relatively marginalized members of a given society. Within IR, my favorite articulation of this is Cynthia Enloe’s essay “Margins, Silences, and Bottom Rungs” in the Smith/Booth/Zalewski edited volume from a few years ago.

Now, if we want to think about this methodologically, and not just in terms of methods or techniques, what is distinctive about placing personal experience at the center of one’s strategy of knowledge-production? After all, one could conceivably use the information gained by such experience-near modes of information-collection to code variables and test hypotheses the way that any neopositivist would, so it’s not the simple act of remaining close to the personal experiences of one’s research subjects that makes the difference. Instead, I’d argue, what makes feminist work methodologically distinctive is its emphasis on locating the researcher — the knower — within her or his specific social context. This reflexivity is obviously not just characteristic of feminist work; postcolonial theory and some strains of post-Marxist critical theory also share in this emphasis, along with sociology in the Bourdieusian mode. But in IR in particular, feminist scholars have been the most articulate and consistent proponents of this reflexivity.

So that’s what I would focus on as “feminist methodology” — not because there’s anything particularly or exclusively feminist about reflexivity, but because feminist scholars and feminist scholarship in the social sciences provide an especially good example of that reflexivity.

Share

Methodology411: Nemesis


Nemesis. No, I don’t mean the tenth Star Trek film, a film that many of us Trekkers would like to simply imagine never happened (and thanks to J. J. Abrams’ rebooting of the franchise, we now can). Instead, I mean the long-theorized stellar companion to our Sun — perhaps a brown dwarf star — the existence of which could perhaps help to explain cycles of mass extinction on Earth. NASA’s WISE satellite, presently conducting a survey of the entire sky in the infrared spectrum, might be able to provide photographic evidence of Nemesis’ existence, although we wouldn’t be able to confirm that until about 2013 because of the time needed to process all of the data.

While it’s interesting in itself to think that we might be living in a binary-star system rather than in the single-star system that we’ve all been taught about for generations, what’s even more intriguing to me here are the curious methodological issues that the whole question of Nemesis’ existence or non-existence raises. Since we can’t see this hypothetical brown dwarf with any of our human-normal senses, any discussion of Nemesis necessarily takes place in the shadowy realm of the unobservable — which is a realm that anyone who has been following debates about scientific realism in IR or in the social sciences more generally has heard a lot about in recent years. In particular, we are often told that social structure, being unobservable, implies a scientific realist ontology in order to really make sense as a scientific concept.

I’m not sure that this is true either of social structure or of Nemesis, but not for the same reasons. There’s a key ambiguity in the notion of an “unobservable” that is sometimes exploited in scientific realist arguments in IR, in that the arguments often equivocate between things that theorists positing them say that we simply haven’t observed yet (like Nemesis) and things that theorists positing them say that we can’t possibly observe (like social structure, especially in the hands of scientific realists). Research implications follow, but first we have to be clear on the conceptual complexity involved.

First, a little basic ordinary-language philosophy. The most usual sense of the notion of “existence” involves something that we have direct sensory evidence of: I know that the book on my desk exists because I can see it, and can pick it up, and so forth. There are a whole series of conceptual calisthenics associated with teasing this notion out in a consistent way, many of which involve optical illusions or dreams or other hard cases, but I think that the basic point holds as far as ordinary speech is concerned and as long as we’re dealing with physical objects (and not definitionally transcendental objects like God or the soul; to say that such things do or do not exist gets us into a very tricky metaphysical realm that I want to avoid for the present discussion). If I don’t have direct sensory evidence of some physical object — if I haven’t actually seen a unicorn — it’s difficult for me to claim that it exists and to have those words mean what they conventionally mean in everyday speech.

In other words, our usual everyday notion of existence is pretty empiricist, to the extent that it relies on empirical evidence as the final court of appeal. Now, two caveats apply more or less immediately. First, relying on someone else’s direct sensory evidence does not seem to be a particularly complicated warrant, philosophically speaking; there are all kinds of practical or technical questions regarding the identification of reliable witnesses, but we do this all the time in everyday life so I see no reason that this ought to present any special conceptual challenge. Second, and a little more problematically, the boundary defining the things that are taken to be “direct sensory evidence” seems to be historically mutable. Most famously, the invention of the telescope did not immediately result in people regarding what one could see through a telescope as having the same epistemic status as what one could see with the naked eye; instead, it took time for the telescope to become popularly and philosophically regarded as a way of augmenting human senses such that telescope-mediated visual evidence was basically on par with direct sight. But once this was done, the telescope in effect ceased to present a perceptual problem, and looking through a telescope thereafter becomes a form of “direct sensory evidence.” Repeat the basic outlines of this story for the photographic camera, the scanning electron microscope, radar and sonar, etc., and we have a kind of “robust empiricism” which can deal with augmented human senses pretty easily.

The robust empiricist answer to the question of whether Nemesis exists, then, is pretty straightforward: look for it, and if you can see it, it exists. [I am not going to go into the various historical reasons why sight almost invariably gets privileged as the source of evidence in these discussions, but just realize that a) it does and b) that it does so is somewhat philosophically problematic, even though it doesn’t challenge robust empiricism to shift from sight to smell or touch.] One of the commentators on io9’s coverage of this story — which emphasizes the point that Nemesis is conjectured to cause periodic extinctions by disturbing the orbits of comets in the Oort Cloud and sending them speeding into the inner solar system — spells out the empiricist position on the issue quite well:

. . . we’re not even sure if the Oort Cloud is even there. Proof comes from observation, and not once have we EVER seen a comet at the distance required to prove the Oort Cloud is really there. We know the Kuiper Belt is there, because we can directly observe Kuiper Belt bodies. Not so with the Oort Cloud.

“Proof comes from observation” is the money-quote here, along with the implication that you could only really know that the Oort Cloud existed if you could get sufficient distance to actually see it and the comets that supposedly populate it. Otherwise, the Oort Cloud (the existence of which is accepted by basically all contemporary astronomers), and the brown dwarf Nemesis that might be affecting it, remain “theoretical” entities.

But note that this kind of robust empiricism has a directionality to it: even if we can’t presently have direct sensory evidence of the existence of the Oort Cloud or of Nemesis, there is nothing to prevent us from a) speculating about how its existence might help to solve certain puzzles and b) engaging in a more or less direct search for direct sensory evidence, which in this case means building a better piece of sensory augmentation equipment (the WISE satellite, which is basically designed to provide us with a better picture of the whole sky than we could get just by looking with our human-normal eyes — such survey-mapping is, so to speak, robust empiricism par excellence). Of course, we can’t say that something that we haven’t yet observed actually exists — or, better, we can’t say with any certainty whether it does or does not exist. But the point is that determining whether it does exist or not is a relatively straightforward matter of getting ourselves into the right position from which to observe it. Indeed, we might easily conclude that the point of scientific research is precisely to get ourselves in positions from which to observe as much as possible, and thus to steadily eliminate “theoretical” entities from our conceptual inventory by replacing them with observed ones.

In other words, Nemesis is what we might call an unobserved observable: an object that we could in principle observe under the proper circumstances, including the use of the proper sensory augmentation equipment. The appropriate research project for such an object, as I have suggested, is to find or build some way of observing it. Though there might be technical challenges or political obstacles to surmount in doing so, there is no reason in principle to suggest that the object couldn’t be observed, and hence no theoretical barrier to trying to do so. Matters are quite different with objects that the very theory and theorists that posit them declare to be in-principle unobservable: quarks, very high-energy fundamental particles like the Higgs boson, black holes, or social structures (at least as conceptualized by certain kinds of social theorists, in particular Marxists and feminists). The fact that the former three are thought to be in-principle unobservable because of physical laws, while the latter is thought to be unobservable because structures are dispositional conditions of possibility rather than entities reducible to their observed effects, is in this case immaterial; what matters is that all four of these objects are very different kinds of “unobservable” than Nemesis is.

As such, the terminus of research into any of these four objects cannot be a direct observation of them. According to the confinement principle of quantum chromodynamics, quarks cannot appear singly. The Higgs boson only exists at such high energy-levels that if one were to be created it would immediately decay into other particles (much like other fundamental particles, actually), so the best we can do is to indirectly detect such particles (which is why the massive machines built for the purpose are called particle detectors, and not particle observers). Black holes capture all of the energy within their event horizon, so all we can do is to infer their existence indirectly. And while we can observe what social structures make possible, we can’t build a structure-o-scope that would allow us to simply view capitalism or patriarchy — though we could and do measure the effects of those structures.

There’s more to say here, of course, particularly about how one might ever know that an in-principle unobservable object exists, and why it matters a lot if the unobservable in question is detectable or not — but I’ll save that for my next installment. For now, it suffices to conclude that whether or not Nemesis exists is a relatively straightforward question, easily answerable within the bounds of a slightly elaborated kind of common-sensical everyday empiricism, and the kind of fantastically impressive sensory augmentations that it easily accommodates.

Share

Methodology411: science and truth

Fascinating little op-ed over at The Guardian Online today (h/t Jesse Crane-Seeber) called, provocatively, “Scientists should stop deceiving us.” The punchline of the piece is that “in holding that the intellectual aim of science is truth alone, scientists seriously misrepresent its real, problematic aims, and thus prevent urgently needed critical assessment by scientists and non-scientists alike.” Chief among these “problematic” aims, we learn, is the preference for unified theories “that attribute the same laws to all the phenomena to which the theory in question applies”; coming in a close second is the “humanitarian or political dimension” of science that leads it to produce knowledge that can be used “to enhance the quality of human life.”

The tone of righteous indignation in this op-ed is kind of startling to me; “deceiving” is a strong accusation, and the implication here is that scientists have sceret value-laden agendas that are somehow affecting their work and their results. To which I say: okay, sure, maybe there’s a great conspiracy to foist unified theories (!) on an unsuspecting public, but if scientists were to be “deceiving” the public I’d expect it to be for something a bit more nefarious than a formal principle like this. (Parenthetically, the author is just wrong when he suggests that a preference for unified theories is “a substantial thesis about the universe independent of evidence”; it’s a methodological principle, not an empirical claim, and methodological principles are pretty much by definition independent of evidence since they structure what counts as “evidence” in the first place.)

Indeed, this op-ed seems to me to be precisely the kind of misunderstanding of science that results when people operate with an unreflective neopositivism as their tacit philosophy of science. Neopositivists — the linear descendants of Vienna Circle positivists via a Popperian shift in the United States during the 1960s — hold that knowledge is the truest (having the most “verisimilitude”) representation of mind-independent reality that can be achieved through the process of successively proposing and falsifying hypothetical generalizations. This is entirely empiricist inasmuch as no lawlike generalization ever has anything but provisional certainty attached to it; there’s no way past experience to grasp causal powers of mechanisms in any kind of definite manner. In such a conception of science, “values” can only appear as distortions — as obstacles to clear representation, since they would presumably not be falsifiable propositions (and if they were then it’s unclear why we’d call them “values” in the first place). And operating with such a conception of science, the discovery of the fact that scientists operate with non-falsifiable notions, both formal epistemic principles like “unified explanation” (which is, parenthetically, not characteristic of all “science” — even the physicists referenced by the author would agree that unified explanations break down in very high-energy situations, or in the smallest incremments of time following the beginning of the universe) and purposive epistemic principles like “enhancing human life,” looks scandalous.

News flash: it isn’t. Scientific explanation hasn’t been about a classically objective representation of mind-independent reality since the invention of quantum theory, and arguably even before that. Contemporary mind-world dualists — scientific realists, we call them — accept the intertwining of the observer and the observed in ways that would make neopositivists shudder (if there were any left to listen to them: there basically aren’t neopositivist philosophers of science any longer, and so neopositivism lives on only among the laity, or among those social scientists who desparately want to emulate what they think that physics was doing in the time of Isaac Newton . . . but that’s another rant for another post). And contemporary mind-world monists — pragmatists, analyticists, etc. — are not only happy to acknowledge the central role of value-commitments in the production of facts, but they would and do argue that only value-commitments permit the production of facts, because without such commitments there’d be no way to focus on or isolate anything in particular in order to study it. There was this German social theorist named Max Weber . . .

Anyway, the main point of the op-ed seems to be that we should indignantly demand that scientists come clean about their value-commitments. No objections there, as long as we all understand that such demands are unlikely to disclose some kind of worldwide conspiracy to enslave the human race to the tyranny of a secret cabal of pro-environment activists. (The op-ed seems to taking some odd swipes at climate science, or at least participating in a debate about it.) I don’t even have any objection to the notion that there ought to be public debates about the value-commitments appropriate to scientific research; indeed, I’ve argued elsewhere that the central contribution of scientific research to public debate is to systematically work out the implications of particular value-commitments in practice, and in so doing make those value-commitments available for a more robust debate and discussion.

But accusing scientists of “deceiving” the public? That’s uncalled for, especially since the public still seems to operate with the old neopositivist notion that scientific objectivity and value-commitments are somehow incompatible. We ought to be trying to educate people about how science actually works, not fanning the flames of the kind of sweeping anti-intellectualism and no-nothing-ism that transforms every question into a matter to be answered by private opinion. If there’s any deception going on, it’s the deception perpetrated by those who erroneously think that science provides classically objective answers to difficult questions and thus relieves us of the burden of dealing with them in our usual messy, practical way. Including the author of the op-ed, who should have chosen a different target for his ire.

Share

Methodology411: defining “methodology”


Vikash’s comment on my announcement post suggests to me that it might be useful to have a less elliptical and, so to speak, inside-baseball-y declaration of intent for the Methodology411. Even though I’ve written about and railed against it on numerous occasions, I often forget how firmly ingrained the equation of the terms “methodology” and “statistical methods” is in anglophone Political Science and IR — and that equation (combined with my reference to fantasy baseball) would, I now see, make my initial announcement of the Methodology411 look like a call for exclusively discussing statistical techniques.

Therefore, to set the record straight: Methodology411 is about methodology rather than method, and it is about social-scientific methodology broadly defined rather than about statistical (or, better, neopositivist) methodology in particular. I envision a forum for the discussion of the logic of inquiry understood in a pluralist sense. This means that although there is no single universally correct form of social-scientific practice, there are concrete implications and conclusions for the conduct of research on the social world (and on world politics in particular, since there is a general but not exclusive commitment to “IR” around here) to be drawn from different stances on or wagers about the epistemic status of worldly knowledge-claims. One need not be a statistician to have interesting and valid things to say about world politics, but given the unreflective dominance of neopositivist methodology throughout our field, it is especially important to be clear about what one is if one is participating in a different tradition of inquiry. Not that neopositivists shouldn’t have to be equally clear, but the plain fact is that they all-too-often are not, relying instead on the already-established common-sensical nature of their methodology between and among the members of the audience to whom they are normally speaking and writing.

The in-your-face version of this point that I have used as the summary paragraph for talks that I’ve given on this goes as follows:

For decades, IR scholars have labored under a delusion: that there is One True Scientific Method, and that strict adherence to that method will guarantee the scientific status of their empirical researches and pronouncements. Even a cursory examination of the literature in the philosophy of science shows us that this is simply not the case — and yet the cultural valence of the notion of “science” remains, making it imperative that the field have some kind of answer to “the science question.” I present a pluralist solution, one that acknowledges the existence of significant differences between philosophical ontologies (ways of thinking about the mind-world hook-up) and the methodological perspectives to which they give rise, but organizes that diversity so as to promote internal consistency, public discussion, and worldly insight as the hallmark of a scientific study of world politics.

Now, the implication of this kind of pluralism is not that we shouldn’t do neopositivist work that is designed to test hypothetical propositions about systematic cross-case covariations between variables. It is, rather, that we should regard neopositivism as one methodological perspective among others, enjoying no special relationship with “scientific knowledge” per se. And in consequence, scholars utilizing every social-scientific methodology ought to be relatively explicit about the philosophical presuppositions of their approach to knowledge-production, striving to be internally consistent and coherent when they apply and enact those presuppositions in conducting their substantive work.

Methodology411, then, is animated by this kind of commitment to pluralism, and will be a place for the discussion of what pragmatists might call the “cash value” of varying ways of producing knowledge: what our philosophical ontologies get us in practice. The precondition of such a discussion is, of course, that practitioners are clear on the logic of their approach and its implications, and part of the Methodology411’s mission is to help promote that kind of clarity regardless of which methodological approach one is utilizing or of which tradition of inquiry one is working within. Internal consistency and global diversity within a broadly social-scientific frame, where “social science” means something like “systematic efforts to produce worldly knowledge about the social, and to so do in such a way that the results are publicly evaluable according to standards shared by a community of researchers.” So: neither art, nor politics, nor ethical critique, but the pursuit of factual knowledge with the clear understanding at the outset that there is more than one way to skin a fact, er, a cat.

Puns, alliteration, and ultra-geeky sci-fi and baseball references (and the occasional song-lyric from a progressive rock tune) are a bonus feature. With that, open wide the floodgates, and let the dance begin!

Share

Announcing “Methodology411”


Spring is around the corner, which means that baseball season is finally getting underway. Amidst the spring training games and the speculations about the upcoming season, one can find the clear signs of the truly obsessed — the fantasy baseball players — ramping up their debates about how to select players in their fantasy league drafts, weighing the relative importance of slugging percentage versus base-stealing ability, and generally working to find the best strategy for success in this bizzare kind of second-order competition.

Yes, I’m among them. Guilty. I am a big fan of second-order competitions, second-order debates, and second-order concerns — heck, my new book is about philosophy of science and its methodological implications for the study of world politics, and how much more “second-order” can you get than that? But there is an art to conducting such debates well, to be sure a different art from the actual playing of the game itself, but an art — a practice, a sensibility — nonetheless. In the fantasy baseball world, there are a lot of charlatans peddling snake-oil, which makes a source of good strategic information all the more valuable. One that is a particular favorite of mine is the Fantasy411, which started life as a radio show and podcast and now also includes a blog and even some TV appearances. Fantasy411 consists of a small group of folks who weigh in on issues related to the playing of fantasy baseball, and who take questions from their listeners and viewers and answer them on the air or online; even if one doesn’t agree with the proffered advice, the discussion is usually engaging, and disagreeing can teach you as much if not more than agreeing can.

So it occurred to me that what works for fantasy baseball might work well for our own peculiar kind of second-order concern with social-scientific methodology: how do we produce knowledge about world politics? Sure, there are all kinds of neat little research manuals out there, but they are generally pretty narrow in their orientation, and even the best of them don’t really provide the same opportunity to collectively wrestle with the complex issues of research design appropriate to specific projects. And sure, all of us with Ph.D.s in Political Science had to suffer sit through some kind of methodology course in graduate school, but I am going to bet that it was a) boring; b) more of a class in technique than a class in research design; and c) dominated by some fairly easy statistical methods. (Yes, there are exceptions to that claim, but from the people I’ve talked to over the years that seems to be a pretty accurate description of what passes for a methodology course in most Political Science departments.) Besides which, who actually remembers what they learned in their graduate school methodology course that was helpful to them once they got into actually doing their own scholarly work? Anyone? Anyone?

Okay, so there’s an opportunity here. Duck gets a fair number of readers within the IR profession, and we’ve had some discussions about these issues here before — and I’m usually involved in them, because as anyone who knows me or reads my posts knows, I actually don’t really care as much about contemporary politics as I do about theory and methodology (and science fiction). So in establishing the Methodology411 here on Duck, I am merely giving formal expression to something that has been the case for a long while: I am most interested in and indeed passionate about those parts of the scholarly IR enterprise that bore most people silly. And I am also taking advantage of the fact that we have excellent people here at the Duck who have far more interesting things to say about actual contemporary world politics than I do most of the time — like I said, second-order concerns, that’s my thing — so I don’t feel compelled to try to generate posts about things that, frankly, aren’t my main interest.

Here’s how this is going to work: I have cleverly grabbed a gmail address for the Methodology411; you can probably guess what it is, but if you can’t, let me spell it out in such a way that trawling bots and spiders might miss it: methodology411 at-sign gmail dot com. Send me your queries and issues, send me things that you’d like to see answered or discussed, and I’ll post them once or twice a week with a few of my own thoughts and open a discussion-thread about them. If it’s a slow week I’ll look for something else to comment on from this point of view, like Andrew Exum’s little manifesto on the quantitative analysis of war, a manifesto which — quite problematically, in my view — combines rules appropriate for good statistical inference with indefensible assertions about the inherently non-statistical nature of the phenomenon of war. [The problem here, in a nutshell, is that “statistical” names a way of studying the world, and is not a commentary on the world itself: it designates a methodology, and says nothing whatsoever about the inherent character of any objects thus analyzed. The only limitation to studying something statistically and quantitatively is the imagination of the researcher; anything can be quantified, which doesn’t mean that it should be, but it does mean that if one wants to quantify or doesn’t want to quantify one has to actually provide reasons for doing so — reasons that aren’t about the nature of the object, but which are about the kind of research question that you’re asking and the techniques appropriate for answering it. Methodological reasons. And let’s not even get into the conflation of “statistical” and “quantitative” implicitly operative in many manifestos like this.]

See, it’s all too easy for me to go off about these things, because I actually get quite worked up about them — much more so than I do about contemporary politics. So in the interest of being a force for good, I offer the Methodology411 as a place where you can bring your questions for discussion, and I can geek out about philosophy of science and research design to my heart’s content. And hopefully, we just might learn something by talking together about these issues — something that might help us to produce better social-scientific knowledge. So bring your queries, and we’ll see what happens.

Share

© 2020 Duck of Minerva

Theme by Anders NorenUp ↑