Egypt and Poli-Sci US academia
Issandr El AmraniAndrew Exum touches on an academic issue here worth mentioning: that the events in Egypt have been poorly predicted by North American academia, perhaps because political science departments largely focus on quantitative analysis. Andrew, as ever (and I blame living in Washington as well as his southern roots for this), is very polite about not bashing the "quants", as he calls them.
Personally, I would be more blunt. Quantitative analysis and the behaviouralist approach of most American PoliSci academics is a big steaming turd of horseshit when applied in the Middle East. Statistics are useful, yes, when you are in a country that has relevant statistics or where polling is allowed. But things like electoral statistics tell you very little about the political reality of dictatorships, because the data sets are inherently flawed, since they're either unavailable, fraudulent, or irrelevant.
I remember reading a year or two ago some bright young thing (who is now on tenure track at Harvard I believe) wrote some turgid paper about electoral participation and vote buying in Egypt. I won't name the person in question, since you have to do this kind of thing for your career, but the paper relied on a mishmash of social statistics (literacy etc.) and grand assumptions about the behavior of electors whose votes are bought, as well as electoral results. This is nonsense, because while the social indicators can be taken at face value, the rest (results, turnout, etc.) is most probably the work of the fertile imagination of an Amn Dowla officer. So focusing on the details of elections that are fraudulent may as well be an endorsement of fraud.
I'm a bookish fellow and long considered pursuing a PhD, but was turned off by this American approach, which is not only largely useless and dreadfully boring, but actually positively dishonest and misleading. Turn to the more historically-minded British analysts (and those few Americans who eschew QA) for real insight to the Middle East. In the meantime, it's sad to think generations of political scientists are still fed this claptrap.
Reader Comments (21)
Teehee. I agree. But can I also say that almost no-one saw this coming because no-one could have predicted the demonstration effect from Tunisia? Yes, we all knew there was discontent and a growing opposition movement and so on but the regime was just so darned good at keeping it all down.
May I also note that the quants are on their way out in popularity and the likes of our Dr. Josh are a sterling example of GOOD academic analysis?
There in lies the problem. In this country we have too many bean counters (accountants) who are focused on statistical models and not real people. My father was such a person who didn't realize the personal costs until he was laid off after 39 years of service.
I can only hope for a shift in our analytic methods here.
I think the important thing is not so much that Egypt could have been predicted as that QA work yields little understanding of political and social realities.
I think your analysis of American political science shows a lack of understanding of the field. Aside from the individual you mention (hint: Harvard doesn't have a Middle East person aside from Dr. Masoud at the KSG, so she's not there) and perhaps 2 or three others, formal and quant analysis are still largely lacking among Middle East regional specialists, who are more theory oriented and use ethnography and qualitative methods. This hasn't served us any better in predicting anything, nor has it served your British academics.
I completed my PhD in a US institution in Politics and Middle East Studies, and must say that our department was not overly weighted toward Quantitative analysis. My approach was qualittative, and I felt no bias agianst the work. I think the key for some basis of predictability is following events and currents closely. I was shocked when Condoleeza Rice said that the Administration had no idea that Hamas had so much support and the PA had lost popular sentiment, after the election in 2006; and I am surprised that the Administration really has been caught flad-footed these past few weeks. That is not to say that events have been predictable in their timing and scale, but something had to give at some point.
I agree wholeheartedly with this article. I finished my bachelors with an intensive focus on the 20th century Middle East last year at a prodigious university and was turned off time and time again by misleading quantitative analysis. However, I split my degree between the poli sci and history departments which yielded fruit when I combined historical driven approaches with newer constructivist/non-positivist poli sci techniques. I wish I could be back in school right now...
Stanford, not Harvard, ya rais.
Your follow up re QA not giving understanding is stronger than complaining it was not predictive. I don't think any Poli Sci can predict such moments.
On the other hand, I largely agree that much of Amer. Poli Sci quant / statistical analysis is pure bollocks when extended outside of data rich environments (and oft bollocks there, as the vision is too narrow).
Unequivocal, particularized and specified analysis (using quant or qual models) is fraught with complications and uncertainty. How to factor in chaos when chaos reigns? Many in PolSci have predicted that the regime structured governments are unsustainable in the region and that a combination of the youth bulge, high unemployment and lower living standards had to give at some point in time.
Yes - I got the university wrong. It's certainly not Masoud whose new book actually both contains rare data on elections and good analysis of the Ikhwan.
I would add that you also got the sarcastic usage of "bright young thing" wrong, too-- not only in the sense that (1) this person may still have plenty of interesting insights despite your personal issues with the field of American political science (which don't seem terribly well founded); but (2) I am curious to know if you'd also address your male peers in similar language.
Issandr,
Would you refer to a male as a "young bright thing"?
AMP - as someone who has heard scholar in question present a paper on what Issandr discusses, I can assure you there were no new insights - it was rehashed data and frankly a rehashed translation of research that had been published in Arabic years before, simply in quant lingo.
And if you are so convinced that quant analysis has served our understanding of the Middle East well, please offer some examples of research you like.
Dear SP and others,
I don't think it makes sense to continue this pissing match. As someone who has criticisms for those who are too area focused and too quantitatively oriented, I encourage you to use your valuable skill-sets to challenge the work you feel is misguided. This would do the discipline (and policy-makers) a great service. But these silly personal attacks out of jealousy and/or resentment serves no one.
I am happy to clarify. The original post and comments have criticized this person's work on two fronts: (1) bad data, which is a legitimate criticism if it's true, and is an unfortunate aspect of working in our region regardless of your methodology; and (2) the use of quant methods. Bad data is one thing, but using it to bash quant methods is quite another. I realize that Mr. Issandr has qualified his statement by saying that quant methods is OK with good data, but the discussion has digressed in the comments section. So I'll take up point (2).
This debate, and the broader one involving Drezner, fundamentally misunderstands the distinct goals of quantitative versus qualitative analysis-- and by qualitative analysis I mean comparative historical work with some standards vis-a-vis case selection, process-tracing, and congruency testing, not the musings/diatribes/speculations of someone who has simply spent lots of time in the region-- an unfortunate presence in both Europe, the US, and Middle East scholarship.
The basic goals of quantitative analysis are to establish correlations with predictive capability. Advances in quantitative methods mean (1) better models, (2) better operationalization of variables, and (3) often new correlations. New data is good, but recycling data is also OK as long as you do something new with it. By contrast, the basic goals in qualitative analysis are to (1) describe (bring in new insights, as you say), (2) establish causal mechanisms through process-tracing and ethnography, and (3) theory build. Some people might quibble with my distinctions, but I think they generally hold.
Indeed, the prevailing belief among many methodologists/ philosophers of science is that you cannot explain something with statistics, and you cannot predict with qualitative work. The irony of the current debate about American quantitative methods is that (1) people are attacking it for not predicting Egypt, or Iran, or whatever case, when it truly is only quantitative analysis that has predictive power, even if it can't *explain*; and (2) the attackers of quant methods, such as Mr. Issandr here, seem to favor qualitative methods, which can do a hell of a lot in terms of ad hoc explanation but not much in terms of prediction. Show me a European qual who produced an empirically-based prediction of Egypt's or Tunisia's protests, their timing included, and I'll eat my words.
That said, I am inclined towards qualitative methods and theory-building myself (I'd rather not name names for fear of initiating some vicious discussion of their work on a blog). However, I recognize the different contributions of each methodology, and I don't try to measure the quality of one approach based on the standards of the other.
RL and AMP - I agree. As I said in the beginning, no one saw this coming. I was just curious about specific examples of quantiative contributions to Middle East studies - I can only think of very few (Ellen Lust, e.g.). On the qualitative side, Nathan Brown, Josh Stacher, Jason Brownlee, Maye Kassem and several others have done sterling work. It's not about Europeans either, most of these folks work in US academia.
I think the original gripe was that you don't get taken seriously in mainstream polisci if you can't talk the quant talk, and talking that talk is difficult when the area you work in doesn't have reliable quant data to begin with.
Yes I refer to anyone young as bright young thing, spare me the tiresome political correctness, I have always treated men and women in equally abusive manners. What's more, I never referred to the person's gender.
And to boot, the usage of the phrase if self-referencial and ironic, since I am myself a young thing, although starting this debate suggests I may not be too bright.
AMP — you miss my fundamental point, which is that there is almost no good data in authoritarian political systems.
AMP, I also find in ME Studies and otherwise, that if you ask questions based on what you can prove quantitatively, this tends to limit your research/questions to - umm - restating the obvious. I would love to be proven wrong so any quant research you have on the Middle East that says something really new, do pass it on, and perhaps I'll even renew my APSR subscription
Issandr, you are not being precise. I fully acknowledged your point that there is no good data in authoritarian systems-- see my last post. However:
(1) you could be wrong, if such an authoritarian system has an incentive to provide good data in certain areas (certainly the case for some of Egypt's economic data), or if the researcher conducts their own survey (see Melani Cammett's great new quant/geospatial work in service provision in Lebanon)
(2) the discussion on the comments list proceeded from this point to quant methods bashing.... and you trail in this direction as well in mentioning the Exum piece.
You also missed my point about quant vs qual methods. "Saying something new" can be both introducing news facts and theory (what you seem to want) OR providing a new model (what a lot of quants and methodologists want). Just because that "something new" is not of interest to you as a journalist and commentator does not invalidate it as "something new."
Your sentiment and analysis of what's wrong with PolSci and it's quantitative-driven results is shared my many. Here's an excerpt from Kevin Drum's book review of Daniel Drezner's Theories of International Politics and Zombies http://motherjones.com/media/2011/02/egypt-afghanistan-iraq-zombies
Drezner's real genius is that he's written a stinging postmodern critique of IR theorists themselves, applying the full force of their structured reasoning to topics as diverse as Michael Jackson's breakdancing zombies, Peter Jackson's lesser film canon (Dead Alive, a splendid Kiwi undead gorefest), and romantic zombie comedy flicks—"rom zom coms," as he puts it. It's both a pedagogical text and a lampoon of pedagogy.
TIPZ is a pretty good book. As Adam says, it's part mockery ("postmodern critique" wouldn't have occurred to me, but maybe it's that too) and part serious primer about the insights and weaknesses of various IR theories.