Methodology
Introduction
Anthropology
has developed some excellent methods over time, and so have other social
sciences. Not using these is comparable
to an astronomer using a spyglass instead of computer-integrated information
from modern telescopes, or an anatomist using a paleolithic handaxe instead of
a scalpel and microscope. There is
simply no excuse for doing poor work, especially on a genuinely valuable
project, because of failure to learn a few simple methods.
1. General Background
Technically, a methodology is a
suite of methods entailed by a particular theory. One uses these methods because they are the
proper or best way to test hypotheses generated by the theory.
A theory, in turn, is a general
assumption (or set of interconnected assumptions) about how things work. (The best account of such matters is Kitcher
1993). The theory may be just guesses,
like string theory, or may be very obvious statements that need formalization
and extension, like the theory of gravity. Newton did not discover that things
fall down instead of up; his genius was to explain why they did, as well as
could be done at the time, and to state it mathematically.
A theory should lead to
hypotheses (predictions or similar bets) that can be tested; otherwise it’s too
vague to count. Many theories get along
without making clear testatble statements, though, in spite of positivism. Still, if the theory doesn’t make you
formulate some sort of testable hypotheses,
it’s a waste of time. Marxism and
capitalist economics are both famously untestable bodies of theory, but do lead
to testable statements. The failures of
the USSR and Mao’s China show that, whatever Marxism-in-general has to offer, some orthodox Marxisms don’t work. The Great Depression and the world recession
of 2008 show that capitalism doesn’t always work, either. Many do not count Marxism as a body of
theory, however.
Some theories are
disproved and are essentially dead. The
most famous of these is Galen’s theory of humoral medicine, which guided medical
science throughout the Old World for centuries.
Usually, however, a theory does not totally die; it generates a few
useful formulations that go on and on.
And even a bad theory can generate useful hypotheses and
conclusions. Galen’s ideas about moderation
in diet and exercise are still with us, since he was perfectly right about
them, though for the wrong theoretical reasons.
A theory differs from
several theory-like formulations, all of which can be useful but are not really
theories. Orienting statements are one
type. An orienting statement gives you a
general way of looking at things, but is too general and abstract to test or to
suggest testable statements. Recent “theories”
about globalization, for instance, direct us to look at global-scale phenomena,
but usually do not make testable claims about those. Often an orienting statement is a moral
claim, and therefore untestable because it is about what we should do, rather than what we do.
Another shaky type of
“theory” is the banal, trivial sort of statement for which certain branches of
sociology are infamous (Mills 1959).
Saying that humans are social, that society requires organization, and
that organization requires leadership is too bland to be worthy of the name
“theory.” Theory begins when we make
claims about how organizations form, how leaders come on board, and what form leadership structures take
under given social circumstances.
Another, and much
worthier, alternative to true theory is interpretation (Geertz 1973). Interpretation is, by definition,
unprovable. It can range from my
idiosyncratic take on something to a generally accepted understanding, but it
is not provable in the scientific sense.
We find it most frequently in literary studies. Science cannot prove that one or another
understanding of the Bible or Hamlet is the “right” one, or that Beethoven’s
Ninth is noble and imposing, or that Dutch still-life paintings were comments
on the transience of life. We do not
have the creators of these works around to ask.
Yet, it is well worth while to talk of such matters and speculate about
them, and anthropology would be immeasurably poorer without such discourse.
One goal of theory and
interpretation is to “tell the story behind the story”—i.e., to figure out what
is actually causing the events we see.
In social science, theories often divide into broad categories according
to what is assumed to be the main
cause of action. Economists tend to
assume people want money or material goods.
Sociologists often assume social solidarity or social position are
especially important. Political
theorists, including “critical” thinkers like Foucault, often assume power is
the most basic thing (though they often have a hard time defining it). There are other possibilities. The wise social scientist will keep an open
mind, and see how all factors play in a given situation.
Finally, we have
philosophy, classically defined by Plato as the study of “the true, the good
and the beautiful.” Neither science nor
interpretation will ever tell us what those are, but the human race cannot stop
speculating and arguing about them. We
are better and nobler for doing so, in spite of the ultimate hopelessness of
the task.
Hopefully, all this will
save readers from the all-too-common tendency in anthropology to write a fun
story about one’s field work, and then—after the fact—hang some sort of
“theory” on it because an editor demanded same.
A decent anthropologist goes to the field with a body of theories, or
interpretive ideas, or philosophic concepts, and expects to test them, or at
least learn something important that is relevant to them.
Methodology comes in as a way of
testing the hypotheses and examining the theory. It can also greatly sharpen, expand, and
improve the quality of interpretation and philosophy.
Usually, we
in anthropology do not follow the rigorous positivist rule that a given theory
must call forth a specific methodology and a given method-set must be
theory-driven. (Some anthropologists,
especially in archaeology, do follow the positivist rules on this.) We use the term “methodology” to refer to
methods in general. Moreover, all the
methods I describe below can be used with almost any theory, though a
particular mix of them may be appropriate to only one body of theory. However, it is well to remember the connection
with theory. Most current cultural
anthropology is weakly theorized; at worst, it is mere travel writing. So-called “theory” is often no more than a
positive attitude toward the people studied and a negative attitude toward
outsiders that have an effect on their lives.
This is bias, not theory.
Physical anthropology uses
Darwinian theory, archaeology often uses ecological or processual or
post-processual theories, but cultural anthropology currently uses actual
theory rather sporadically. Theories of
the past (Boasian, Durkheimian, Marxian, etc.) are now used only in a rather
loose or general way. Some ecological,
linguistic, and economic theories are still used, but are often dated by
now. The theories of mid-20th-century
writers like Michel Foucault remain valuable, but often used vaguely or
loosely. This has produced a situation
in which much of anthropology reads like poor-quality journalism—a situation in
serious need of correction.
It is
extremely valuable to go into the field with a full tool kit of theories and
methods. No theory is adequate by
itself. Even the most comprehensive
social-science theories need major supplementation. The more you reject theories, the more
limited and hard to use your results will be.
Both the “postmodern” anthropology that rejected science or even
systematic data collecting and the hyper-“scientific” work of the early
optimal-foraging-theory days have turned out to be too limited to use for any
purpose except stimulating others to go beyond them. A simple theory is always a good starting
point, but anthropology by 100 years ago had reached a stage where truly
simplistic theories were known to be inadequate (see e.g. Lowie 1937).
Many
anthropologists over the years have found great consolation in T. C.
Chamberlin’s classic essay “The Method of Multiple Working Hypotheses,”
originally published in Science in
1890 (republished 1965) and now available online on many websites—just search
the title. Chamberlin, a geologist,
learned to go into the field with multiple theories and hypotheses available
for every observed event. His
explanation of how and why to do this has never been surpassed. I have actually found this method the most valuable
I have ever used. It means you have to
be familiar with the widest possible range of high-level and mid-range
theories, from functionalism and structuralism to Foucaultian ideas and
Darwinian biology.
Thus, you
might think of using some or all of the methods below, so as to get at least
some real control on data.
Anthropology
is based on a methodology consisting of three fundamental approaches:
–Extended
field work, usually lasting at least a year, with a particular community. The preferred method is “participant
observation,” in which one lives as much as possible in the way the local
people do. Of course, really living as
the locals do is possible only if one is a local; many anthropologists study their
homelands, but most go to some less familiar group, which involves adjustment
and makes participant observation a rather qualified matter.
–A
holistic approach, which involves taking into account ecological, economic,
technological, social, psychological, and political factors.
–Cross-cultural
comparison, which involves comparing as many different cultures as possible, to
establish or disprove generalizations about people.
This
methodology was devised by Lewis Henry Morgan, the father of American
anthropology, in the 1850s and 1860s. I
think of it as the three stones that hold up the cooking pot— a metaphor used
for social categories (rather than anthropological methods!) by indigenous
peoples from the Toba Batak of Sumatera to the Maya of Quintana Roo.
The leading methods book for anthropology
is Russell Bernard’s classic Research
Methods in Anthropology (now in its 4th edition, 2006). This is a genuinely great work, a real Bible,
and must be kept at hand in field work and analysis.
The only other work I consider
indispensable for all ethnographers is Charles Frake’s Language and Cultural Description (1980), which contains several
essays on methods that are vitally important.
These essays include especially the classic descriptions of frame elicitation (see below).
There are specialized journals
devoted to field work and methods.
2. The Question of
Interpretation and Reality
The key thing anthropologists can do is find
out about the local culture. This
does not mean “getting inside the heads” of the locals or “finding out what
they think”; it means finding out about what they share. As an outsider, you will not have the level
of access to that shared knowledge and behavior that an insider has, but by
using specialized anthropological techniques you can get very close. You can learn just as well as any immigrant
and almost as well as any child. Frake
gives excellent discussions of what the ethnographer can and cannot do. You can’t read the local minds, but neither
can the locals; they have to infer rules, structures, and understandings, just
as you do. The goal of the field worker
should be what Frake (1980) calls “appropriate anticipation”—be able to
predict, more or less as well as the locals do, what will happen in a given
situation. The goals of the
ethnographer, again following Frake, should include telling the reader enough
that the reader could act appropriately if s/he were there. Think of a language textbook: it should, at the very least, tell you what
to say in given situations. Similarly,
an ethnography of local religion should at least tell you how to act and what
to expect if you go there and are asked to a ceremony. (Of course, a work on general theory, or on
comparative mythology, or on demographic history, will probably not have such
instructions. We are discussing
ethnography, specifically, in this case.)
On the one hand, this means you can
learn the culture, and claims that the locals have some mystic telepathic
sharing denied to you are just silly. On
the other hand, it means you should be exceedingly modest about
“interpretation”—even if you are a local!
Geertzian “interpretive anthropology” (Geertz 1973) and its ancestors
(“national character” studies, etc.) have a dubious record. Unless you are a cultural insider, you will
not normally share individual or collective experiences
of war, genocide, bias, or for that matter the joys of good harvests or
religious ceremonies. It is wise to
simply quote the locals, extensively, on such matters. Let them do the sophisticated interpreting.
In short, you
should do everything possible to find out shared knowledge and shared behavior,
but you should be appropriately modest about your ability to understand
personal experiences of particularly intense, evocative states and situations.
There was a
major debate within anthropology in the 1960s over whether we can get at “what
people think.” Marvin Harris (1968) took
an extreme view on the “no” side. He
maintained that we can record only behavior, and cannot trust what people say,
let alone our interpretations. People
lie, misrepresent, misunderstand their own motives, etc. At the other end of the scale,
interpretivists like Geertz and cultural psychologists like Rick Shweder
(1991), without making a huge point of it (as Harris did), took relatively
strong “yes” positions. Geertz and
Shweder implied that understanding what is in people’s heads is relatively
unproblematic, at least if one uses modern methods of finding out. Geertz
is modest about his interpretations, leaving the possibility of other
interpretations quite open. Shweder,
and others, have been more assertive.
The field basically solved the
problem by voting with their feet for the latter position. I do not know of anyone maintaining Harris’
position today. All anthropologists now
infer, to varying degrees, “what people think.”
All anthropologists admit that people do sometimes say what they think,
and that by careful cross-verification and other techniques (see below) one can
get at, or at least approximate, truth.
Even archaeologists are increasingly confident in their ability to infer
at least some simple, straightforward ideas from material remains and
ethnographic parallels, though this is a tricky game.
There is, however, a huge
range. Some extremely careful
anthropologists use a whole armamentarium of techniques to establish
meticulously a few rather simple understandings; this would include many
cognitivists, who work hard to find the meanings of “simple” plant and animal
names, food lore, kinterms, landscape terms, and other straightforward terms
that can be grounded in visible reality.
(I am in this category.) Others
make really quite wild assumptions about their ability to understand in depth
the most arcane and abstruse religious and philosophical ideas. This is obviously a dangerous game, since
even the locals may not share abstruse ideas very widely.
One
necessary part of this is getting a thorough sense of what words mean. You don’t have to be totally fluent in the
local language, though it helps.
Systematic questioning, coupled with lots of listening and observation
of how words are used in actual conversations, is necessary. (See Frake, again.) Using the words yourself is obviously
desirable—you’re sure to misuse them in interesting ways, thus producing
innnocent amusement for your subjects as well as a learning experience for
yourself. (Every ethnographer has a
favorite story. Mine is: when first in Hong Kong I had to buy water
from a local standpipe. People would
give me the standard greeting, “Where are you going?” I would answer “I’m going to buy water.” After a couple of shocked looks, I realized
something was wrong, and found out that the phrase “buy water” is used only
when you are getting water to wash the corpse of a family member! Just one of those idioms….)
3. Techniques
Field work by cultural
anthropologists usually involves participant observation (DeWalt and DeWalt
2001; Spradley 1980), lasting at least six months and usually a year or
more. Serious comprehensive ethnographic
research requires this. However, for
many reasons, we also do quick visits, long-distance studies (using other
people’s findings), straight interviews, visual studies, library and
documentary research, and cross-cultural comparative studies, among other
things. Limited projects (e.g. to find
out about one narrow subject—say, fish names or vegetable marketing) can be
completed in a few weeks, especially if one is familiar with the area and
people.
One valuable technique is rapid
rural assessment (RRA), which is a specialized interview-and-observation
technique that allows very rapid discovery of a lot of data (Gladwin 1989
covers it; there are more up-to-date, complete sources). Related is participatory rural assessment,
which involves organizing local people to do their own fact-finding and
synthesis. In participatory rural
assessment, local people set their own goals, map their communities, figure out
what they need by way of development or problem-solving, figure out what
resources they have, and so on; the anthropologist guides the approach and sets
the tasks.
Getting
started: Every community has somebody
who knows everybody. Frequently, this
individual is a minor government functionary in a “helping” role (as opposed to
a person keeping the place in line). The
local postmaster filled the role in American small towns. So did the waitresses at the local coffee
shop. Sometimes the village storekeeper
is a contact person, but sometimes he is seen as the village skinflint. Check around!
Then,
wander around the community being very nice to everyone, greeting them,
learning their names, introducing and explaining yourself. Become a local fixture to the point where you
are semi-invisible—just the local foreigner.
A census is
a good way to start serious work and get to know everyone. Ask very nonthreatening questions on an initial
census! See below on finding out about
local question etiquette.
Interviewing is the basic technique
in ethnography. This can mean anything
from applying a set questionnaire (closed-ended
interviewing) to free-ranging questions and discussion (open-ended interviewing). I get best results with a semi-structured questionnaire, one that
you memorize thoroughly before the interview and then apply in a rather
improvisational manner—not letting the interviewee escape without getting all
the questions answered, but letting some free play happen, so the interviewee
can get clear about meanings, discuss points, clear up ambiguities, etc. See any good book on social interviewing, as
well as Bernard.
Keep working on the language—we
could all use better fluency. I am a
terrible linguist, but I try.
The whole issue of how to interview
and ask questions is the first thing to address when you get to the field. Cultures differ dramatically as to what types
of question are acceptable. Many
Americans are astonishingly open about sex but hate to disclose their
income. Chinese (at least the ones I
worked with) are the reverse. Americans
also hate to admit they are racist. A
colleague of mine was amazed at how little racism his students found in our
city of residence. I asked him if it had
occurred to him that having bright young university students doing the
interviewing might possibly bias the responses.
“Why, no….” Another colleague was
similarly surprised by how carefully people were shopping in the supermarket—I
was less surprised, since he and his co-worker had followed shoppers around
with a videocamera. Having (again)
bright young university students watch every move would make anyone more
careful! Such examples are so obvious as
to be funny, but the danger is in far more subtle matters, especially when one
is translating a perfectly innocent question in English into what may be a
subtly leading question in Spanish or Chinese.
See also The Long Interview (McCracken 1988) and James Spradley’s The Ethnographic
Interview (1979). Others recommend (but I have not seen) a book
by Charles Briggs called Learning How to
Ask (1986), one by Meyer and Booker (1991) on interpreting interview data,
and a book on “active interviewing” by Holstein and Gubrium (1995). It is also very worthwhile to spend a while
with reporters finding out about journalistic methods of interviewing and
getting data.
One absolutely critical
interviewing technique that nobody covers well is depth interviewing. This is a
2- to 4-hour interview in which the ethnographer probes deeper and deeper into
the interviewee’s emotions, feelings, and personal stories. A good interviewer tries to keep questions
down to a minimum, and usually just makes encouraging noises (“and then…?” “mm-hm?”). The interviewer must appear relaxed but
thoroughly engaged—completely present, interested, and supportive. A good interviewer will appear not to “pry”
or “apply pressure” but will be sympathetic and concerned and genuinely
interested. This involves being
comfortable with silences—Native American informants in particular often remain
silent for a minute or even several minutes during such conversations. On the other hand, very gentle questioning of
the type “How did you feel about that?” and “are you comfortable talking with
me about that?” is necessary. In such
cases, DO take “no” for an answer; be comfortable with letting the interviewee
set limits.
Almost anyone loves to talk about
almost any subject, if they are given this level of genuine concern. (Be prepared for tears and other emotional
releases.) Such interviewing is an art
form, though it is basically developed from what close and empathetic friends
and family members do for each other all the time. It is
also so intensely personal that unless you are genuinely concerned and caring
about the interviewee, YOU SHOULD NOT ATTEMPT IT.
Depth
interviewing is necessary in many, many ethnographic cases, especially in
interviewing about tragedy and major stress.
It is astonishingly rarely taught or used. Psychotherapists are supposed to learn it but
often do not. The literature that
alleges lack of mother love and lack of regret for dead infants in certain
societies is evidently based on lack of familiarity with this interviewing
technique. I know this not only from the
literature but more directly from my own field work in at least one society
where such lack was widely alleged by superficial ethnographers, but instantly
disappeared under depth interviewing, when grief could come out openly.
Finally, never underestimate the
value of “deep hanging out”—an excellent phrase used by Clifford Geertz to
describe everyday ethnography. Just
hanging around keeping your eyes and ears open remains the best of all field
techniques. I have found I talk less and
look more every time I do field work.
Etics and
emics: Kenneth Pike liberated the
linguistic endings from “phonetic” and “phonemic.” He meant something really creative: Etics involve studying a system by applying a
universal metric or analytic system—in
the case of phonetics, the international methods of studying sounds, via the
sonagram and other mechanical/impersonal techniques. Emics involves studying a system by finding its internal structure and the
units that make that up—in the case of phonemics, the sounds recognized by
speakers of the language as making meaningful contrasts.
Etics does not mean “outsider’s view” and emics
does not mean “insider’s view,”
contra the sloppy usage in many anthro books (including Conrad Kottak’s widely-used
textbooks). Using the terms this way
loses all their value. Both emics and
etics can be done by either outsiders or insiders, but only when trained in
structural analysis. In language, for example, any trained insider can use a
sonagram as well as any outsider; conversely, most people cannot provide a
phonemic analysis of their own language—only trained linguists do that.
A good
ethnographer, whether outsider or insider, will study both etics and emics,
just as any decent linguist will record both the phonetics and the phonemics of
a language. Consider food: a good ethnographer will do a nutritional
analysis and some kind of optimal foraging model or Bayesian-optimizing model
(all these are etic), but will also find
out what the locals call their foods, how they classify them, how they
structure them in terms of nutrition and social use, and other emic matters. Neither of these has anything to do with
outsider vs insider per se. (The typical
outsider’s view of local foodways is “yuck!”
The typical insider’s view is “yum!”
This does not get us far analytically.)
Stories and
texts: These were the bread-and-butter of old-time
ethnographers, and often is to this day.
Nothing beats collecting stories—personal stories, stories about the
community, about the origin of the world, about the economy, anything. People love telling stories. In most cultures, stories are teaching
devices; people teach their children and each other through this medium. Any and all texts and accounts are
valuable. Record them and transcribe
them. A particularly good authority on
working with stories is Julie Cruikshank (1998, 2005).
Decision-making is also a very
important, basic approach, best introduced in Christina Gladwin’s little
booklet Ethnographic Decision Tree
Modeling (1989, Sage) is basic. A
classic study, with methodological reflections (especially in the 2nd
edn., 1994) is James Young and Linda Garro: Medical
Choice in a Mexican Village. Shankar
Aswani has done some good work on decision-making in fisheries, and
thoughtfully related that to more purely economic and biological methods (e.g.
Aswani and Weiant 2004). Basically, the
idea is to ask people in detail about the steps that they went through to make
a particular decision—what crop to plant (Gladwin), what to do when someone in
the family is sick (Young and Garro), what to do about fishing and fish
conservation (Aswani), and so on. This
technique assumes that decisions can be broken down into ordered sequences of
yes/no answers: Can I get the seed for
this crop? Can I get fertilizer for it?
Can I get enough water for it? And so
on. People usually do decide that way,
at least in clear-cut matters like crop choice, and even if they don’t you can
break down decisions into yes/no or more-versus-less choices. But sometimes people decide on impulse, or
subconsciously integrate several factors at once. Careful questioning allows you to deal with
such cases, and continue to use decision tree analysis. It is a particularly powerful technique,
especially for decisions that are important but that involve well-known, rather
routine choices, like agricultural decisions.
A farmer or gardener normally knows exactly what crops she can plant and
how to grow them, and how to get information if she does not know enough about
something. Decisions about what to do in
an unforeseen new emergency are less clear-cut and consequently harder to
analyze, but in principle can be covered the same way.
Another
absolutely essential technique is the focus
group, in which the interviewer recruits 4-6 people or so and gets them to
talk about the specific subject under investigation. This has turned out to be a major winner as a
research method for political researchers and marketers as well as for
anthropologists. See David Morgan
(1996).
Another universally used technique
is the Likert scale, that little
scale where you get to rank things from “agree strongly” to “disagree strongly”
or “most liked” to “most disliked,” as on student evaluations, political
surveys, etc. It works well only if you use 5 or 7 cells. 5 is generally better.
A large range of personality tests and other
psychological tests is available. In
general, I advise against using these, because they rarely work in local
conditions—the local worldview and language are probably too different from the
testmakers’. But they may be useful
where this does not apply and where you can get a psychologist to help
administer them.
Other formal techniques include frame elicitation. This is best explained by Frake (see above),
but basically it consists of looking around and asking everyone “what’s
that?” When you have names, you sit down
with a consultant and ask “what kinds of X are there?” till there are no more
divisions. Then you can work up: “Is X a kind of…?” Beware, though; this can force a spurious
level of systematization on your consultants.
Better to do all this informally in the field, one question at a time,
and to use focus groups to get people talking about how they conceptualize
things. Carefully used—with much asking,
pointing, and walking around, rather than mechanical frame interviewing in a
house—this is the most valuable of all the analytic or specialized techniques.
Related are card-sorts and
pile-sorts, in which names of things are written on cards and sorted into piles
according to whatever criteria you want to study. See Bernard.
Walking around in the fields and
woods, asking about everything, remains the best of all techniques for finding
out about names, categories, and ethnobiological knowledge. A formalization is a “nature trail,” in which
the investigator lays out a short set course with known plants along it. Then the investigator can walk this trail
with different subjects, seeing how many plants they can name. This is particularly useful with children—one
can see how much they know at what age.
(Brian Stross, Gene Hunn, Rebecca Zarger, and J. R. Stepp, studying
children in the south Mexican highlands, have worked particularly with this
technique.)
Child-following
is used to advantage in such situations, and in nutrition research. You just follow a child around, seeing what
she does. It’s the only way to find out
what children actually eat, as memorably shown by the late Christine Wilson in
her field work.
For that matter, following adults is necessary too, but has
to be done with more circumspection.
There are also censusing, surveying,
survey design, optimal foraging study and modeling, GIS and GPS, statistics,
economic data management, and other formal techniques; Bernard covers all of
them adequately, though if seriously interested in optimal foraging or in
economics you will need supplementary reading on these (they have a large,
specialized literature).
Surveys often involve poorly
designed questions that lead to misleading results. Most people agree with both “Individuals are more to blame than social conditions for
crime” and “Social conditions are more to blame than individuals for crime”—depending
on which one you ask (Radwin 2009:B9).
In other words, people love to agree with any old statement. It’s all in the way you phrase it. Question order, bias words, and so forth all
influence the result. Some questions are
so poorly worded that a large percentage of the respondents cannot figure out
what is being asked. This is
particularly common when a questionnaire is translated from one language to
another, as very often happens in anthropological research, so watch out;
pre-test questionnaires for comprehensibility.
Remember to avoid leading questions
(now often called “push questions”):
questions that imply you want a certain answer. Indeed, avoid everything that might be taken
as implying you want to hear a particular kind of answer. Find
out what counts as leading questions in the culture you are studying. Many questions that are perfectly innocent
and non-leading in English turn out to be strongly leading when translated into
Chinese. I found out the hard way—but at
least I learned it fast—that “how are you?” was interpreted as “you look sick,
what’s wrong?” Similar pitfalls occur in
other languages.
Response bias can enter quite
dramatically. Surveys of food
consumption in the United States correlate very well with sales figures at
stores, but sales figures of liquor consumption can be up to five times what
the surveys show! People may understate
consumption, but more important here is the fact that an extremely high
percentage of liquor is drunk by relatively few people, and those few are
rarely in any condition to answer a survey.
Finally, people lie, almost always
to give the socially “correct” response.
Many more people say they voted in the last election than could actually
have done so (Radwin 2009:B9). And
almost no racists exist in the United States—if you believe the survey
results!
Anthropologists also do a great
deal of visual anthropology:
photography, recording, film and videotape work, and other methods of making a
permanent and more-or-less-objective record of what we find. There is also ethnomusicological recording to
worry about. There are specialized works
on this. My experience is that it is
difficult to do quality visual work and quality interviewing or other
talking-ethnography at the same time.
One can work as a team, or do the interviewing first and visuals
later. Some geniuses can do both at the
same time, but I am far from this level.
I’m not an expert, and will not
push this one, but a useful tip from Douglas Medin (presentation at Society for
Anthropological Sciences, 2010) is that there are four general ways to do a
picture: directly on (the usual
approach—“voyeur”), embodied (shows hands working, from the viewpoint of the
worker—as if you and the camera are doing the work with your hands), over the
shoulder (of your main subject—so you are standing behind her and seeing what
she sees), and “fourth wall” or “breaking the wall,” in which case the people
in the photo are all looking at you (as in a standard group shot). Doug showed that Native American children’s
book illustrations (drawings, not photos) have much more of the last three
types of pictures than Anglo ones, a culturally very interesting
observation.
This brings home the point that
interpreting others’ photos and pictures is a major part of visual anthro. Both interpreting cultural representations
(pictures, etc.) and getting your subjects to take photos for you are standard
techniques and very effective if well done.
Another under-taught topic is
historical research–documentary, archival, and text work. Historians learn as a kind of second nature how
to evaluate a document—how much to trust it, how to cross-check, how to allow
for biases, etc. The best way to find
out about this is to ask a historian.
They have their own books, but an hour with a seasoned historian will
give you a good enough start.
At the very least, read the major
anthropological and historical works on your area! I am appalled at the illiteracy of some
graduate students. What were their
professors thinking? Egregious mistakes
even get into the published literature.
This is inexcusable. Much more
common is the field worker who misses a great deal for lack of knowing the
questions to ask, the deeper matters to look for, and the contexts to use in
interpretation.
How to take
and keep field notes is the subject of Roger Sanjek’s Fieldnotes and a lot of journal articles. Some other useful lore is in Tony Robben’s Ethnographic Fieldwork: An Anthropological Reader; Joseph
Casagrande’s marvelous and far too neglected anthology, In the Company of Man; and Michael Agar’s classic The Professional Stranger. See also LeCompte and Schensul, Designing and Conducting Ethnographic
Research (2nd edn. 2010).
Multi-sited
ethnography: this has been advocated by
George Marcus and many others. Obviously
it’s appropriate if you’re studying mobile, transnational, or migrant
populations. It isn’t if you’re studying
people who stay put, unless you want to do systematic or controlled comparisons
(very valuable, but a different issue). Use common sense and don’t feel compelled to
do it just because it was a buzz word for a while. If you have only a year, as most of us do,
it’s better to stay put. Finding out
much about even a very small community in a year is already challenging enough. The great transnational studies, like Michael
Kearney’s (see Kearney 1996—a “must read” if you’re working with this), were
30-year or 40-year projects.
Teamwork: The day of the lone field worker who found
out “everything” about the Trobriands or the Nuer is most emphatically
gone. Do what you do best, and
collaborate with other people who do what they do best.
Work with biologists, political
scientists, photographers, anyone that has expertise you need. Many ethnographers go in as part of a
team. I find it more useful to work with
people on the ground. Local scholars generally
need and appreciate the opportunities.
(On the other hand, many see outside scholars as a threat to their
monopoly and their status. Be careful
about this.)
One type of
“teamwork” is working as a family.
Fortunate is the anthropologist who has a spouse who can work with him
or her. Alas, field work is not always
the easiest posting, and some spouses do not adjust well. Most valuable of all is working as a family
with children. Children disarm
suspicion, attract friendly and solicitous attention, evoke stories, and allow
study of child-training practices. Also,
when they are old enough, they are born ethnographers. They are curious about everything and are
amazingly quick with social cues and social learning. (They are wired for
it. The human animal evolved as a social
creature, and social learning is a child’s main occupation.) However, working with children is reasonable
only if you are near a good hospital.
Children have died in remote field situations.
“Studying
up”: Laura Nader and others have
advocated studying the rich and powerful.
Unfortunately, I could never get a million-dollar grant for subsistence. More seriously, most anthropologists don’t
have the tools and training to do this effectively. If you want to study up, work with and learn
from political scientists and sociologists!
They have the methods and tools!
When faced with the need to find out what the powerful were up to, I
have worked with political scientists, and have also picked the brains of
anthropologists who had done that type of work and had learned the techniques
and methods.
There are lots of political
scientists, sociologists, historians, and others studying elites, but only
anthropologists study the people low on the political hierarchy. We thus best use our talents and training in
the latter cause. We are generally the
only people that can bring their words and concerns to a wide audience. Now and then we get the chance to help them bring
their own voices or causes to the wide arena—a blessed and wonderful chance if
carefully done. I thus strongly
recommend studying ordinary people and especially neglected and oppressed ones.
4. General
Philosophical Concerns
Completeness
and comparability are major concerns, and major problems with many field
projects. Be sure to get all the data possible on the subjects
under study. Be sure that interviews,
forms, and data recording makes findings strictly comparable between subjects and situations. The same information has to be collected in
the same way.
One word of philosophical guidance
about culture in general: Only real
people (or animals) do things. This
should be obvious, but anthropologists all too often fall into the social science
trap of saying that Capitalism, or The State, or the Flying Spaghetti Monster
did such-and-such a thing. No. They didn’t.
People did. Capitalism and the
Flying Spaghetti Monster don’t exist (the former is an analytical abstraction
that bears only some resemblance to any current real-world referent). The State exists, but if you think it acts or
is real by itself, look at Somalia, DR Congo, or Afghanistan. The State functions because the people in it
have decided that preserving it and working for it will best accomplish their
human goals. It becomes a true emergent,
like a kinship system or a myth, and thus has a genuine reality (unlike
capitalism). However—again like a
kinship system—it exists only as long as a lot of people buy into it and don’t
question it too strongly. Always study
emergents and recognize their reality, but remember they don’t really act by themselves. The ability of people to believe in such
things, and to believe they act on their own, is fascinating, and related to
the belief in supernatural beings.
As Andrew “Pete” Vayda has been
insisting for years (Vayda 2008), some background in the philosophy and history
of science (specifically, epistemology) is absolutely essential. This would include, at least, Thomas Kuhn’s
classic The Structure of Scientific
Revolutions (1962) and Philip Kitcher’s The
Advancement of Science (1993). See
also Ian Hacking (1999), Bruno Latour’s work (esp. 2004, 2005), Alison Wylie’s
work (2002, 2004), and Pete Vayda’s and others’ relevant writings. Some background in the history of
anthropology is essential (see many books by Adam Kuper and by George
Stocking). Theory and history are not
covered adequately in many anthro graduate programs, so read these on your own.
Perhaps the
most valuable thing one learns from these works is how to avoid mindless use of
current buzzwords. Buzzwords usually
start out as useful concepts, but lose it all when they become too widely
used. Go back to the original source and
read the full, properly qualified story.
Those of us who have checked are always astonished at how wrong even the
best secondary sources get the classic writers, to say nothing of slapdash
textbooks. Reading Durkheim, for
example, is a real revelation if you knew him only from even the best histories
of anthropology.
Ethics: Here again, one can start with Bernard, but
an excellent practical guide to working ethics has now appeared (Whiteford and
Trotter 2004) Many ethical questions
have been treated in detail in Anthropology
News over many decades. The American
Anthropological Association’s Code of Ethics is easily available online from
the Association, and is basic.
Always
be meticulous about touching bases in the field area. Contact local scholars, and promise to
help and work with them if possible. Go
through all the bureaucratic hoops uncomplainingly. Find people you can work with, institutes you
can collaborate with, and universities you can hang out at. Be humble; First World investigators are
threatening to many Third World bureaucrats and scholars. Many—if not all—Third World and indigenous
scholars have encountered arrogant, overbearing, and inconsiderate First
Worlders. These were not usually
anthropologists, but you will pay the price even if it was a diplomat or an
agricultural advisor that dissed the local scholars. Bear it and be genuinely polite. Save your hate for the diplomat or advisor,
not the locals.
In your community, similarly, get the official cooperation of the local
authorities—complete with signed permission to work. Share
your results, in so far as possible, when you do any writing up.
Questions that permanently concern
anthropologists include: Are we really
somehow ripping off the “natives” by finding out things? How does one collaborate? Coauthor?
How does one “represent the other” without being a mental
colonialist? How to get honest responses
and publish them? How much can one
publish the local dirt—corruption, conflict, sordid tales? (My recommendation is simple: don’t unless you have to.) How to
be tactful? How much to get involved in
local matters? How to avoid
factions? How to avoid local
entanglements? One would, for instance, think it unnecessary to warn people NOT
EVER to get sexually involved with people one is doing field work with! But I hear that some people do this—a good
way to get killed.
Err on the side of caution. Remember the first clause in the AAA’s Code
is that your most immediate duty is to the people you are working with. It is not acceptable to put them at serious
risk. It is not acceptable to exploit
them for money, e.g. by selling photos or writing a bestseller without cutting
them in on the profits. It is not
acceptable to use their words and information without giving full credit,
including coauthorship if their input is really significant. It is not acceptable to refrain from helping
them with medicines, etc., if you have the knowledge or connections; if it
messes up your medical anthro research a bit, too bad; their lives are more
important than any dreams of intellectual purity. Do not let yourself be exploited or “used,”
but be as helpful as possible when help is needed.
Avoid involvement with local
factions, no matter how right your favorite one seems to be. Involvement ruins your field work, endangers
your safety, and inevitably makes local politics worse. Let them sort it out.
The wider question of advocacy is
more serious. Anthropologists almost
always find that their groups are getting a raw deal, because we usually study
small, less-than-affluent communities who are low on the political
hierarchy. Serious advocacy is often
desirable, but should take the form of “speaking truth to power” as the phrase
goes. It is not usually appropriate to
get off into strong statements or political action in the field site On the other hand, it sometimes is appropriate, e.g. in cases of
outright genocide. Generally, the very
best thing is to carry local voices to the wide world—if you can do it without endangering your subjects. For example, giving quotes that can be traced
to an individual is not a good idea
in a state that is persecuting that community. Confidential reports to trusted government
people who can really help your community are sometimes desirable. The best thing is usually to do the best job
you can at getting the facts right and producing a scholarly book. Do what is morally right, but in the most
cautious and least overstated way.
Think seriously about who can hear
your message and use it. I did one
substantial piece of field work in a really dangerous situation. I never published or disclosed the worst and
most hidden material. I got the rest of
the really touchy material to people in the government whom I knew I could
trust and whom I knew would use the information wisely. I kept everything else on ice for years,
until the situation changed and I could safely publish the less touchy chunks
of it.
Anthropologists
are driven almost mad by the steadily increasing obsessiveness of institutional
review boards (IRB’s, a.k.a. Human Subjects Review Committees). They exist to prevent lawsuits over problems
arising in sensitive, invasive, or dangerous medical and psychological research. Thus they are often inappropriately
restrictive for anthropological field work.
We cannot always get signed, detailed protocols proving that our
subjects know every possible risk they are incurring. And we may have to take photographs and films
of large ritual or market situations where we cannot possibly get signed
permissions from every man, woman and child.
And we rarely do anything that puts
subjects at any real risk. The main
exception, and it is an important one,
is research in or on military, criminal, or other genuinely dangerous matters. For these, the investigator does need to worry about the full IRB
panoply of concerns. For the full story,
see the fall 2007 issue of American
Ethnologist, which has a whole excellent section on IRB’s.
Applied
anthropology is a whole separate area that I do not want to cover here; suffice
it to say that the same general moral rules apply. Do what will actually help and what will
actually not be undercut by someone else.
Collaboration
with local communities in getting particular projects done is another
enormously complex and involved topic, beyond my range here. See the journal Human Organization—just search back through it.
The
question of objectivity always surfaces.
No, you aren’t totally objective; you’re involved with your
subjects. But, as one anthropologist
wrote, “just because you can’t maintain pure asepsis doesn’t mean you can do surgical
operations in a sewer” (Geertz 1973). Be
solidly grounded in facts and establish everything as solidly as possible. I once worked with an anthropologist, a
superb field worker, who collected over 50 detailed stories of the same
event—and stayed perfectly neutral and calm through it all, properly writing
down everything, though the stories wildly disagreed about basic details! Then we sat down to analyze why the stories
were so different.
The idea is
to be as factual, or objective, as possible, but be open about your biases,
too. Self-awareness is important.
5. Final Tips
Field work is lots of fun, but one
thing we face is culture shock, a useful term coined by the
Finnish-Canadian anthropologist Kalervo Oberg.
This condition is not confined to anthropologists. What typically happens, when Person A goes to
live in Society B, is that the first 3 to 6 months are a sort of honeymoon
period. After that, Reality hits, and it
can hit pretty hard. A period of painful
adjustment follows—the 6th month is usually the hardest! It’s good to plan a brief vacation from your
field work at that time. After the 6th
month, things get easier. Students are
familiar with a mild form of this from adjusting to college (dorms, roommates,
classes…). Adjusting to marriage or any
other life-and-residence change is comparable.
There is a honeymoon period, a let-down period, and then adjustment,
hopefully peaceful and contented.
Once you have adjusted to a new
community, adjusting back to your own home typically involves some “reverse
culture shock.” Do not be surprised at
this; it’s normal.
Field work is normally one of the
least dangerous activities on earth. I
have always been healthier in the field than at home. Forget the poisonous snakes and scorpions of
the travel books—you won’t see any, or if you do they will be the least of your
worries. (I have had to kill more than
one cobra in my field dwellings. They
don’t usually strike.)
However, don’t take insane
chances. Take a first-aid kit and standard first-aid medications, notably
general antibiotics that will quickly knock out skin infections, traveler’s
diarrhea and food poisoning (Salmonella,
Shigella, etc.), and the like. Be
sure to take the proper anti-malarial medicine in malarial areas; the medicine
of choice varies from region to region.
Use tough, sturdy shoes or boots if
in a literal “field” situation. Today,
the “field” is often an urban neighborhood, but some of us still work in actual
fields. You are far more likely to come to grief from wearing inadequate shoes than
from all those poisonous critters put together.
Take sunblock and suchlike things as appropriate. DO ask people who have been in the area you
are going, and DO read the Lonely Planet guides, or similar guides for active
and enterprising travelers.
Don’t worry about the local food,
including “street food.” It’s safe enough
if cooked at high heat. Any fruit or veg
with a tough peel (bananas, mangoes…) is safe if the peel isn’t broken. In most of the tropics, the water is still
dubious, however, and so is raw seafood.
The only time I got really sick in 2 years of field work in Mexico was
from eating undercooked oysters in a fancy restaurant.
The fad for “reflexive” ethnography
a few years ago gives you lots of accounts to learn from. Many are far from exemplary. One particularly candid account of a
particularly intelligent, sensitive researcher’s first taste of the field is
found in Eric Mueggler’s The Age of Wild
Ghosts (2001). I could name many
others. A nice balance of
self-revelation with consultant’s own stories is Zapotec Women by Lynn Stephen (1991). She has her opinions and experiences; she
also gives the facts; and she gives the women’s own testimonies, which often
disagree with her interpretation. Stephen
makes their lot sound very bleak, but the women she quotes sound decidedly more
happy. I visited her field site and did
some field work myself to understand this.
It turned out that Stephen emphasizes the hardship which is indeed the
lot of most Zapotec women, but the women she quoted in the book were a
relatively more successful group who were generally more upbeat on their
situation. Also, Mexican women are
taught to aguantar—bear
uncomplainingly. They don’t expect as
much from life as an elite American academic does.
Agar, Michael.
1985. Speaking of
Ethnography. Newbury Park, CA: Sage.
— 1996.
The Professional Stranger: An
Informal Introduction to Ethnography. 2nd
edn. New York: Academic Press.
Aswani, Shankar, and Pam Weiant. 2004.
“Scientific Evaluation in Women’s Participatory Management: Monitoring Marine Invertebrate Refugia in the
Solomon Islands.” Human Organization 63:301-319.
Bernard, H. Russell.
2006. Research Methods in
Anthropology. Lanham, MD: AltaMira (Rowman and Littlefield).
Bernard, H. Russell (ed.).
2000. Handbook of Methods in
Cultural Anthropology. Walnut Creek,
CA: AltaMira.
Briggs, Charles.
1986. Learning How to Ask. Cambridge:
Cambridge University Press.
Casagrande, Joseph (ed.).
1960. In the Company of Man: Twenty Portraits by Anthropologists. New York:
Harper.
Chamberlin, T. C.
1965 (orig. in Science, 7 Feb. 1890).
“The Method of Multiple Working Hypotheses.” Science 148:748-759.
Cruikshank, Julie.
1998. The Social Life of
Stories: Narrative and Knowledge in the Yukon Territory. Lincoln: University
of Nebraska Press.
Cruikshank, Julie.
2005. Do Glaciers Listen? Local Knowledge, Colonial Encouinters, and
Social Imagination. Vancouver: University
of British Columbia
Press.
De Munck, Victor C., and Elisa J. Sobo (eds.). 1998.
Using Methods in the Field: A
Practical Introduction and Casebook. AltaMira.
Denzin, Norman, and Yvonna Lincoln (eds.). 2005.
The SAGE Handbook of Qualitative Research. Sage.
DeWalt, Kathleen, and Billie DeWalt. 2001.
Participant Observation: A Guide
for Fieldworkers. Walnut Creek, CA: AltaMira.
Frake, Charles.
1980. Language and Cultural
Description. Ed. Anwar S. Dil. Stanford:
Stanford University Press.
Geertz, Clifford.
1973. The Interpretation of
Cultures. New York: Basic Books.
Gladwin, Christina.
1989. Ethnographic Decision Tree
Modeling. Newbury Park, CA: Sage.
Hacking, Ian. 1999.
The Social Construction of What?
Cambridge, MA: Harvard University
Press.
Holstein, J., and J. Gubrium. 1995. The Active Interview. Newbury Park, CA: Sage.
Kearney, Michael.
1996. Reconceptualizing the
Peasantry: Anthropology in Global
Perspective. Boulder, CO: Westview.
Kitcher, Philip.
1993. The Advancement of
Science. New York:
Oxford University Press.
Kuhn, Thomas.
1962. The Structure of Scientific
Revolutions. Chicago: University of Chicago Press.
Latour, Bruno.
2004. Politics of Nature: How to
Bring the Sciences into Democracy. Tr.
Catherine Porter. Cambridge,
MA: Harvard University Press.
Latour, Bruno.
2005. Reassembling the
Social: An Introduction to
Adctor-Network-Theory. Oxford: Oxford
University Press.
LeCompte, Margaret, and Jean J. Schensul. 2010.
Designing and Conducting Ethnographic Research. Lanham, MD:
AltaMira.
Lowie, Robert.
1937. The History of Ethnological
Theory. New York: Rinehart.
McCracken, Grant.
1988. The Long Interview. Newbury Park, CA: Sage.
Meyer, Mary A. and Jane M. Booker. 1991. Eliciting and Analyzing Expert Judgement: a Practical Guide. London: Academic Press.
Mills, C. Wright.
1959. The Sociological
Imagination. New York:
Grove Press.
Morgan, David.
1996. Focus Groups as Qualitative
Research. Newbury Park, CA: Sage.
Mueggler, Erik.
2001. The Age of Wild
Ghosts: Memory, Violence, and Place in Southwest China. Berkeley: University of California Press.
Radwin, David.
2009. “High Response Rates Don’t
Ensure Survey Accuracy.” Chronicle of
Higher Education Review, Oct. 9, pp. B8-B9.
Robben, Antonius C.
2006. Ethnographic
Fieldwork: An Anthropological
Reader. New York: Wiley-Blackwell.
Rubin, Herbert, and Irene Rubin. 2005.
Qualitative Interviewing: The Art
of Hearing Data. 2nd
edn. Sage.
Sanjek,
Roger. 1990. Fieldnotes:
The Making of Anthropology.
Ithaca: Cornell University Press.
Shweder, Richard.
1991. Thinking Through
Cultures: Explorations in Cultural
Psychology. Cambridge,
MA: Harvard University Press.
Spradley,
James. 1979. The Ethnographic Interview. New
York: Holt,
Rinehart and Winston.
— 1980. Participant Observation. New York:
Holt, Rinehart and Winston.
Stephen, Lynn.
1991. Zapotec Women. Austin:
University of Texas Press.
Vayda, Andrew P.
2008. “Causal Explanations as a
Research Goal: A Pragmatic View.” In Against
the Grain: The Vayda Tradition in
Ecological Anthropology, Bradley Walker, Bonnie McCay, Paige West, and Susan
Lees, eds. Lanham, MD: AltaMira (division of Rowman and
Littlefield). Pp. 317-367.
Whiteford, Linda M., and Robert T. Trotter II. 2008.
Ethics for Anthropological Research and Practice. Long
Grove, IL: Waveland Press.
Wylie, Alison.
2002. Thinking from Things: Essays in the Philosophy of Archaeology. UC.
Essays; 514 pp.
Wylie, Alison.
2004. “Why Standpoint
Matters.” In The Feminist Standpoint Theory Reader:
Intellectual and Political Controversies, ed Sandra Harding. London: Routledge.
Pp. 339-352.
Young,
James Clay, and Linda Garro. l994.
Medical Choice in a Mexican
Village. 2nd edn. Boulder,
CO: Westview.
