Before I do that, however, I’d like to take a moment to point readers to Carole McGranahan’s 2012 post “What Makes Something Ethnographic?” There she provides a list of nine features generated by her class. One of the points of she makes is that these features are constantly changing and evolving. This is why, in defining ethnographic film, I chose to dodge the bullet by avoiding the question altogether! Letting others deal with that problem is the easy way out, I don’t deny it; but it also allows me to articulate a definition that can change along with the discipline. Looking back at previous attempts to define ethnographic film, many of them strike me as having been dated before the ink even dried on the paper. Hopefully this more flexible approach can avoid that fate.

And now on to the list! If you feel I missed an important feature, or overlooked something, please let me know in the comments.

1st Dimension: Discipline

This dimension includes features that connect ethnographic films to the discipline of anthropology.

Films made by anthropologists.

• Obviously, not all films made by anthropologists are ethnographic films. Home movies, or even research footage are not inherently ethnographic just because they were made by an anthropologist. But (as explained in the last two posts) when combined with the other features listed here, this feature gains newfound importance.

Films made in collaboration with an anthropologist.

• The nature of this collaboration matters a lot. Some anthropologists work closely with a filmmaker, while others are just “consulted” to bolster a film’s credentials. But anthropologists need not be behind the camera to help ensure that a film is ethnographic.

Films based on or inspired by anthropological research.

• To the extent that anthropologists write for a wider public, we should also hope and expect for this wider public to read and understand our work. Thus, it makes sense to place films inspired by such an engagement on the same level as those made by the so-called experts. (And similar to such works, they need to be evaluated by the full list of features associated with each film.)

2nd Dimension: Norms

This dimension includes features related to the norms and practices of ethnographic research.

Films made with an “ethnographic intent.”

• When anthropologists or their collaborators are working with the intension of making an ethnographic film, it is very likely that they will succeed – at least to the extent that they have the necessary skills and training to follow up on their intentions.

Films made in accordance with visual research ethics.

• I think a lot of anthropologists would agree that films made with the consultation and cooperation of the film’s subjects often are able to provide much more complex and nuanced ethnographic insight. The norms of research ethics are one area where ethnographic films have changed a lot in the past few decades, but it is also precisely because of these changes that classic works like Robert Flaherty’s “Nanook of the North” have been given newfound attention by visual anthropologists.

Films made following ethnographic methods.

• Intending to make an ethnography, and doing so ethically, doesn’t really count for much if you don’t know what you are doing. Some understanding of the tools of the trade is necessary, but still may not be essential if the film is not made by a professional anthropologist. But it is perfectly possible for amateur ethnographers to exceed the professionals in their grasp of these skills.

3rd Dimension: Subject

This dimension includes features related to the topics and peoples discussed in the anthropological literature.

Films about topics anthropologists study.

• This is a tough one, because today there is virtually no subject outside of the realm of anthropology. Still, there are definitely some subjects, such as kinship or magic, which anthropologists have claimed as their own. These include both classic subjects taught in intro courses, as well as hot new topics like the “Anthropocene” which are featured at recent anthropology conferences.

Films made in collaboration with people from groups that have historical been the subjects of anthropological research .

• Precisely because of the colonial history of anthropology, films about indigenous or nomadic peoples have long been a staple of the discipline. For this very reason, films made in collaboration with such subjects serves as an important commentary and perhaps even a critique of previous ethnographic films.

Films made by people from groups that have been subjects of anthropological research.

• While films made by indigenous or nomadic peoples are no more necessarily ethnographic than films made by anthropologists, the same thing said about anthropologists holds for them as well: when combined with other features on the list the fact that the films were made by members of such groups can become an important consideration. In my last post I brought up the case of Atanarjuat: The Fast Runner, which is a narrative feature film made by indigenous filmmakers. I would argue that despite being a narrative feature film, it shares many of the features listed here, including collaborating with the community, following the practices of ethnographic research, and abiding by the ethical norms of the discipline, etc. For this reason I don’t have much difficulty listing it as an ethnographic film.

4th Dimension: Genre

This dimension includes features related to the various styles associated with the genre of ethnographic film.

Films made in an established ethnographic style.

• There are a wide variety of styles that mark the genre of ethnographic film: observational, reflexive, sensory, etc. Not all of these styles are exclusive to ethnographic film, but some films are clearly marked as ethnographic by their stylistic choices.

Films made using multiple ethnographic styles.

• A film needn’t be restricted to one particular style, and some directors combine multiple styles or approaches into a single film.

Films that seek to comment on or subvert the genre.

• Films that are in dialog with the genre are ethnographic even if they deliberately avoid replicating aspects of ethnographic style. A great example of such a film is Trinh T. Minh-Ha’s Reassemblage which seeks to comment upon and undermine the norms of ethnographic cinema.

• Discipline: features related to the discipline of anthropology (e.g. films made by anthropologists)
• Norms: features related to the norms and practices of ethnographic research (e.g. research ethics)
• Subject: features related to the topics and peoples discussed in the anthropological literature (e.g. films by or about nomadic peoples)
• Genre: features related to the various styles associated with the genre of ethnographic film (e.g. “reflexivity”)

If I were to graph films using this approach it would look like what is known as a “spider (or radar) chart” with films rated along multiple axes, each representing a different dimension. Like this graph comparing monster trucks and drag racers along five axes:

Rather than sharing the full list of sub-features today, I’ll save that for my next post. Instead, I’d like to use these four dimensions to explicate how such an approach to defining ethnographic films might work in practice. Doing so will help avoid confusion over how the full list is meant to be used, and will also serve to highlight the strength and limitations of this approach.

Let’s start by comparing the films The Gods Must be Crazy and Atanarjuat: The Fast Runner. Both are fiction films by non-anthropologists and so would be excluded from traditional definitions of “ethnographic film.” Would they both be accepted under this new definition, since they both deal with indigenous peoples that have been much studied by anthropologists? Not necessarily. If that was all it took this approach to defining ethnographic film wouldn’t be of much use. For one thing, it doesn’t really conform to our own practices as anthropologists: Atanarjuat has been shown at ethnographic film festivals, but (to my knowledge) The Gods Must be Crazy has not. And secondly, if we left it at that, nearly every film ever made could be included as “ethnographic.”

The solution lies in the fact a film generally must contain features in more than one dimension in order to be considered ethnographic. A home movie of a child’s birthday party is not ethnographic even if it is made by a famous anthropologist. And a film like The Gods Must be Crazy is not ethnographic just because it is about the San. Without getting into a big discussion (see here and here if you’d like to know more), I will simply assert here that Atanarjuat also ranks highly along the second dimension of “films made in line with the norms and practices of ethnographic research” but The Gods Must be Crazy does not. It is for this reason, and not because of the subject matter, that we accept Atanarjuat as having ethnographic qualities, but reject The Gods Must be Crazy.

It is also worth mentioning that the way we evaluate each of these dimensions is constantly evolving. The discipline, along with its norms, subjects, and styles are very different today from the sixties. For this reason even though Dead Birds is one of the most celebrated ethnographic films of all time, a new film made in the same way today would probably be rejected by many festivals. (See discussed by Jay Ruby’s article on Robert Gardner if you’d like to get a sense of why that would be.) Similarly, contemporary ethnographic film festivals are more likely to include “sensory ethnographies” or films about the “the Anthropocene” than festivals organized by the same people just a few years ago.

Before wrapping up I’d like to point out one limitation of thinking of these four dimensions in terms of a spider graph (like that pictured above). Namely, we tend to rank these dimensions hierarchically, but this hierarchy isn’t well represented by the graph. To compensate for this oversight we could weight each of the dimensions differently. But this still isn’t good enough, because a weighting used to evaluate one movie might not be appropriate for another. For instance, I might not weight stylistic features as strongly for film by an academic as I would in considering a film by an artist. And I might not hold a journalist to the same exacting disciplinary norms as I would for an anthropologist. But what we want to avoid is making any single dimension definitive, for doing so would saddle us the kind of “closed” definition we’ve been trying to avoid. While it may not be ideal, the idea of adding weights to each of the dimensions on a case-by-case basis seems to be the best way to capture how anthropologists actually go about evaluating the suitability of any particular film for an ethnographic film festival.

In my next post I will elaborate in depth on each of these dimensions — breaking them down into their constituent features, but I hope that I have managed to make the case for the utility of this approach to defining ethnographic film and given readers a sense of how it would work in practice.

The best I could offer was “I know one when I see one” but this definition cost me dearly. We had over 1,500 entries for the festival, and it took a lot of work to weed out which of those films would go on to the judges and which would not. In the end about two thirds of the films were rejected in the first round. In many cases we only needed to read the film description or watch a few minutes to know that it wasn’t right for the festival. In other cases I ended up watching the whole film before deciding. It was a lot of work.

To be honest, I don’t know if a better definition would really have helped. Festival submissions are free¹ and a lot of filmmakers don’t bother to read the rules before submitting. Many of the rejected films didn’t even meet the most basic entry requirements listed on the submissions page, and hundreds of them were clearly scripted dramas with no claims to being the slightest bit anthropological or ethnographic. Still, the whole process got me thinking about how I would go about trying to define ethnographic film. Here’s what I came up with. I’m posting this in two parts. Today I’ll set out my goals for such a definition, including my overall approach. In a later post I plan to actually sketch out what such a definition might look like.

Previous definitions of ethnographic film seem to fall into two camps: On the one hand there are “open” definitions that tautologically define anything anthropologists like to watch or talk about as an ethnographic film. On the other hand, there are “closed” definitions that try to very narrowly define ethnographic film in terms of some kind of ideal which would effectively exclude most films shown at ethnographic film festivals. (For a good overview, see Matthew Durington’s Oxford Bibliography entry for “Ethnographic Film.” It’s paywalled, but the relevant section is right there in the preview.) Neither of these are very useful for actually running an ethnographic film festival. Too open a definition and we run the risk of becoming just another documentary film festival, one of dozens held in Taiwan each year. Too narrow a definition and we end up with a bunch of boring films by anthropologists that nobody (not even most anthropologists) wants to see. So what to do?

I think it is useful to start by grouping films into three categories. At one end are films like Indiana Jones and the Temple of Doom which everyone agrees is most certainly not an ethnographic documentary. Which isn’t to say that Indian Jones couldn’t be used by anthropologists in the classroom, or as a tool to study American popular culture in the 1980s, but that doesn’t make it an “ethnographic film.” At the other end of the spectrum are films that nobody doubts are ethnographic. These are the films that we now consider to be part of the ethnographic film canon. Films like Trobriand Cricket, or Dead Birds that have been seen by generations of students taking anthropology 101. But while these films may be uncontroversially ethnographic, a film made in 2017 that was just like Trobriand Cricket or Dead Birds might not be accepted into an ethnographic film festival. Changes in the discipline as well as in the medium have changed our expectations of what ethnographic films can and should look like. But this means that there is an increasingly large grey zone between those films we know to be ethnographic and those we know are not. A good definition will never eliminate this grey zone, but it should give us some signposts so that we can better navigate our way through it.

To do this I find it helpful to draw on the model Umberto Eco used to articulate his famous definition of “fascism” (which Matt blogged about in 2013). This approach was itself drawn from Wittgenstein’s “notion of a game.” Rather than listing a set of proscriptive features, such definitions rely instead upon a “family of resemblances.” One unique feature of this approach is that while an item might be excluded from a strict definition for lacking even a single required feature Wittgenstein allows two items to be included in the group even if they share no features in common — as long as they share features with other intermediate elements in the series. Confused? Eco describes it better than I can. He asks us to “consider the following sequence”:

[a b c], [b c d], [c d e], [d e f]

Suppose there is a series of political groups in which group one is characterized by the features [a b c], group two by the features [b c d], and so on. Group two is similar to group one since they have two features in common; for the same reasons three is similar to two and four is similar to three. Notice that three is also similar to one (they have in common the feature c). The most curious case is presented by four, obviously similar to three and two, but with no feature in common with one. However, owing to the uninterrupted series of decreasing similarities between one and four, there remains, by a sort of illusory transitivity, a family resemblance between four and one.

This is perfect for “ethnographic film” which includes wildly disparate types of works: films by anthropologists, films made in collaboration with anthropologists, films by indigenous filmmakers, films made in collaboration with indigenous filmmakers, films about subjects traditionally considered the domain of anthropology, films made by non-specialists that have an anthropological sensitivity, etc. Not to mention films made in very different styles: reflexive, observational, experimental, sensory, ethno-fiction, etc. We need a way to include all of these films together while simultaneously excluding the vast majority of documentary and fiction films which are not ethnographic. And we need to do it in a way which isn’t overly restrictive so that we can actually have a film festival and not just restrict ourselves to watching The Wedding Camels over and over again. It would certainly be impossible if we tried to adopt a closed or proscriptive approach, but I feel confident that something along the lines of a “family of resemblances” might actually work. Later on I hope to finish working out a draft of what such a definition might look like and share it in a follow-up post. In the meantime, feel free to leave your suggestions in the comments!