Now that I am a professor part of my job is trying to explain to students how to do the ‘lit reviews’ that are a part of several of the mandatory genres that they must learn to write. When I was in grad school lit reviews came without saying because they went without saying, so when it came time to make my expectations for this genre explicit, I hit the books. My project: a meta-project reviewing the lit on lit reviews. I found out three things: first, anthropologists do a lousy job of reviewing their literature. Second, I am not a big fan of how other fields review their literatures.
How do anthropologists systematically review the things that other anthropologists have written? Do we in any way systematically read, summarize, and synthesize the things others have written in order to draw generalized conclusions from them? In the case of sociocultural anthropology, I think the answer is generally: no. I mean, we have Annual Review articles and introductions to edited volumes, we have landmark collections which deal with particular topics or ethnographic areas — its not like we have no interest in figuring out where the field is going and what others are doing. But we don’t have a methodical means of working through our materials.
I think partly this is due to the history of our discipline — the time when you could read everything by everybody was not that long ago. When was the last time biologists could say that? The past couple of decades have seen a tremendous growth in the field as well as its professionalization — although perhaps not in all areas.
But there are other reasons that anthropologists don’t systematically summarize their literatures — reasons that become clear when you look at the fields that do. The most rigorous method, metanalysis, is frequently used in STEM (science, technology, engineering, and medicine) to compare studies, tests, clinical trials, and what not. A lot of the books I’ve discovered in my meta-lit review were actually delightful how-to guides written for nurses on understanding the medical literature. Rigorous metanalaysis involves using mathematical analysis to normalize studies and make them comparable even when the number of patients involved, duration of the clinical trial, amount of medication given, etc. vary across them.
There are a boatload of reasons that sociocultural types can’t do this: at the most basic level, we just don’t follow the same generic format in writing our articles: you can’t quickly jump between the findings, discussion, and conclusion of our articles because we don’t write them all in the same order. It’s just this simple. And then of course there are the methodological issues: as in, we don’t have any. Or at least one one susceptible to any degree of mathematical tweaking to make our findings (which we may also be lacking) comparable with those of our fellow ethnographers.
To be honest, I’m not so sure this is a bad thing. I think of anthropology a frying pan full eggs that someone tried to make into an omlette, but didn’t quite manage to flip over and as a result it is sort of half-omlette, half-scrambled eggs in a bizarre and slightly runny mixture which still tastes perfectly delicious despite looking really wierd. Even that half of our discipline that has aspiration to be something more than poetry greatly benefits from the other half that has a practicum in its class on shamanism that requires students to become Possessed By The Beaver Spirit. We gain an insight and vantage point that is valuable because it has not solidified. Of course the flip side is the constant feeling you get that we are reinventing the wheel.
The second method of literature review, the ‘matrix method’, is really just a barely-structured way to impose some order on your notes: you create a matrix with authors on one axis and the topics or themes they treat on the other, and then you fill in the boxes in the resulting table: Foucault says this about power, Durkheim says this about power, Weber says this about power. Foucault says this about socialization, Durkhein says this about socialization, Weber says this about socialization. Etc.
I think this could be a useful method for some things — especially ethnographic details — but not for others. In the example I just gave, for instance, what would record for Foucault’s take on socialization or Durkheim’s take on power? Some would argue these are the same things, while others would say that their viewpoints are so incompatible that comparison of this sort would require a metalanguage so vague as to be useless. Or — more to the point — you’d have to have read the articles already in order to construct the metalanguage that you’d use as you read the articles. In a field with many alternate vocabularies and a penchant for induction, this is a deal killer.
I am not sure if the lack of fit between anthropology and these forms of lit reviews indicates a fundamental problems with the discipline or the scholarly method. It is surely telling, however, that anthropologists can’t easily use either of these methods.
What about encyclopediae? Where do they fit into a lit review? Are they a meta-analysis? I recently discovered that encyclopediae still exist, even since the advent of wikipedia made the general encyclopedia extinct. The Encyclopedia of Global Archaeology, for example, will cover the archaeology of the entire world – each article (some long, some short) aims to give the relevant literature for a particular topic (time/location/method/theory). I suspect this would be difficult to do for, say, an entry on Postprocessual Archaeology, and I suspect that different people will think that different research is important to include.
There is an Encyclopedia of Anthropology, although I’ve never had occasion to consult it. Is this a useful form of lit review? For whom is it useful? Should we continue to update encyclopediae, or are they becoming obsolete?
I think many of the articles in the Annual Review are quite good literature reviews, but then many of them are quite bad. I sometimes complain to journal editors, ARA editors, etc. when they publish bad review articles, but–surprise, surprise–the editors usually don’t agree with me. So then I may vent on my blog. But to me, a bad review article focuses on a small part of the topic at hand and covers only a narrow band of relevant sources. A bad review article makes a big attempt to prove a pet theory, putting partisan argument over objective consideration of the topic. A bad review article is poorly organized. Etc. etc.
I don’t think there is anything inherent in the nature of anthropology that prevents good review articles. Perhaps some theoretical perspectives don’t lend themselves well to succinct synthesis and analysis, though.
@Kristina – Encyclopedias are the trash-heap of scholarly publishing. I am baffled that publishers are still publishing these things. A very small minority are very good, but most are worthless (including most of those that I have contributed to).
Two observations that don’t relate to one another, but speak to the important issues Rex raises regarding information overload and “the literature” in contemporary anthropology:
-For grant proposals, especially student grant proposals, my impression is that a major function of the literature review is demonstrating competence within a particular community. You show that you’re reading works that your imagined reviewers think are cool and important and, in doing so, make readers think you’re going to produce something cool and important. Much of the work of “doing the lit review,” therefore, is becoming aware of the images or associations that certain authors, kinds of literature have and trying to have a sense of how you are presenting yourself—as well as to guard against what you might inadvertently be “giving off” in the Goffmanian sense. Perhaps I’m being too cynical, not talking about specific findings, methodologies, approaches, etc.–but as Rex also notes, this sort of thing also tends to be lacking in much of cultural anthropology today. And perhaps it’s not cool to reference Goffman.
-For introductions to recent edited volumes, I’ve noticed an important feature of the narrative structure of these literature reviews—mixed in with continental philosophy or praise about the distinctive features of ethnography and continued relevance of anthropology—is some kind of positive reference to pre-1980s anthropology. Phrases like “Anthropologists have long been interested in…” or “Since the Manchester School…” are quite frequent, even as these earlier anthropologists’ ideas are rarely engaged with in a serious way. I am not necessarily trying to critique this practice, but wondering how to make sense of it. Is it simply a matter of trying to do sound intellectual history? Is it about legitimizing a present line of inquiry as a logical outgrowth of earlier traditions and thus “still anthropology”? Can we speak of a shift from an uber critique mode of the 1980s and 1990s to a superficial and, at times, condescending embrace of earlier anthropologies?
It has long seemed to me that an anthropological subculture that prizes above all the individual vision quest has resulted in a field whose history is a midden of projects, once proposed and prototyped (an exemplary figure did what ever it was once), but subsequently never followed up. In this respect, anthropology is much like advertising, my other field of semi-expertise, in which nothing destroys an idea more quickly than “That’s been done before.”
This isn’t to say that exemplary figures never attract imitators. They do. But the imitations are so disregarded that, like dandelion fluff, they soon disappear. “Schools” are never more than two generations deep.
This is very interesting and I would agree we are probably still in the throes of growing beyond the time when most all of anthropology could be read and known by one person.
However, it would seem that even in fields like medicine, with quantitative clinical trials, there is a lack of review of previous work:
Trial in a Vacuum: Study of Studies Shows Few Citations
I’m not sure what anthropology should do about our own lit reviews, but it is important to let people know that other fields are not necessarily doing this any better.
This is a great conversation. There are a few things that I think are really important in lit reviews.
One of the things that I notice that drastically differs in what we offer up as lit reviews in Anthro from the STEM meta analysis that Rex discussed above is that we seem to pay far more attention to the context of production—that is who is engaging with whom about what. While a string of clinical studies on the same subject certainly develop in reaction to each other (e.g. they found an effect for 5 mg, let’s see if we can do it with only 3 mg adding vitamin C), the quantitative meta analysis ERASES this conversation by removing time and making everything commensurate. Yet it seems to me that one of the qualities of an excellent literature review is to actually rely the chronology of a developing conversation around a particular problem. Rather than X say Y about power, it’s X said Y about power, which prompted Z to think that it was actually Y’.
The complication in this quality of excellence #1 is how much of the conversation to cover—as indicated in JP’s “since the Manchester School” or Michael’s comment about review that are too small. To write a dissertation, you actually have to write out an overly extensive (and boring) account of the conversation to prove that you know it. Once you choose your audience, you then hack the conversation down to sufficiently get people on the same page and make the topic seem provocative.
This brings me to quality of excellent #2—causing wow. I really think that the point of a lit review is to show that other people have thought about really fascinating things that lead you to wonder about the world. Unfortunately quality #2 has been misrepresented as “finding the gap.” I really don’t care if there is a gap. What I care about is that you have convinced me that this paper that you want me to read actually has the potential to clear up a point that you, and now I, am wondering about.
Cultural anthropology as façon de parler, in a manner of speaking.
@MTBradley Nice.
But suppose we think of it this way. There are fields in which scholarly work resembles a jigsaw puzzle. It is easy to see what is already done and what pieces are still missing. There are other fields, of which cultural anthropology is typical, in which the scholar resembles an artist facing a blank canvas or shapeless pile of clay. There is a field in which the scholar competes, but whether the work done is memorable, acquires a small following, or is relegated to history’s ever-growing trash heap is hard to predict. The only thing we know for sure is that those who succeed once are more likely to succeed again—and even they may burn out. The problem is trying to apply metrics appropriate to jigsaw puzzles to work that is, in fact, more like art.
I have to say that I enjoy anthropology of the jigsaw variety. I think there is a lot of room under the roof of anthropology, but I do think people who decide to stay under the roof should realize that it is a house of social science, not of fine art or humanities. IMHO the anthropology you speak of is neither fish nor fowl and more dilettantism than art, particularly when compared to unequivocal humanities such as art history and folklore.
@MTBradley Alas, there is too much true in what you say. Too much of what now passes for anthropology is neither science nor art. For a bit of inspiration, I offer something that I just posted on OAC, a mediation inspired by an MIT Technology Reviewpiece on David Hockney’s most recent experiments.
“Mediation” [sic] should be “meditation.” The URL is
http://openanthcoop.ning.com/forum/topics/should-an-ethnography-have-a-focus
I was employed by a health faculty specifically to write lit reviews and results of research. The methodology contrasts significantly from the anthropological; my 5000 word lit review on Indigenous health for a grant application was whittled down to 250 words
Thanks for these comments. I suppose there is a distinction between the ‘lit review’ in a grant application and other sorts of genres. There really is an art to performing ‘I know about this and am centrally located in your social space’ in these grant apps, something I’ve never been good at and find fascinating.
I suppose I was also speaking about a second issue, which was viewing lit reviews as a way in to the question of whether or not anthropology should be cumulative, and how best to do it. It seemed to me that other disciplines were (or I thought they were) and that we lack institutionalized genres that would enable us to be — if that was what we wanted to do. Perhaps the grass is always greener on the other discipline.