sociology/psychology

OVERVIEW

In this chapter, we provide an informal introduction to ethnography, a complex and multifaceted

methodology, largely but not exclusively qualitative in nature, with a long and rich history in the

social and behavioral sciences. Ethnography is a scientific and artistic approach to studying

human societies, and it resembles the ordinary person's self-reflexive and systematic

approaches to learning about the world around them, particularly when confronted with a new

cultural experience.

Following the informal introduction, we present some of the core elements of the

"ethnographic imagination" (Willis 2000), the specific tenets that distinguish this logic of inquiry

from others discussed in this book.

AN EXAMPLE-IN SEARCH OF RESPECT:

SELLING CRACK IN EL BARRIO

Philippe Bourgeois lived for three and a half years in East Harlem in order to document and

understand the local microcosm of crack users and dealers who live in a poor and marginalized

community. In his book In Search of Respect: Selling Crack in El Barrio, Bourgeois boils down

hours of interviews and observations into a multidimensional and complex portrait of men and

women coping with desperate conditions. Violence, incarceration, misogyny, heartbreak, hopes,

and betrayals are elements of the story, not only in the narratives about individuals but also in

the depiction of a culture as a whole.

Bourgeois didn't just drop into the neighborhood for a few days to interview passersby or take a

peek into bars and crack houses, and then return quickly to a safe, orderly middle-class world.

He immersed himself (and his wife and child) for a long time in the "other country" at the end of

a short subway ride. His conclusions are based on months of fieldwork, reported in extensive

quotes from interviews and conversations with people he came to know well, and documented

by descriptions of recurrent actions and situations that he saw and recorded. The story he tells

does not have a Hollywood ending nor does it offer simple conclusions and optimistic

"problem-solving" interventions. He probes "the human condition" in one specie context, writing

a dialogue between the voices of the people he met and his own understanding

of their conditions.

Another edgy, high-risk ethnography is Greg Scott's "It's a Sucker's Outfit: How Urban Gangs

Enable and Impede the Reintegration of Ex-Convicts" (2004). Scott not only describes the

setting, actions, and words of the participants in a vivid and carefully documented way; he also

systematically relates his data to an analysis of incarceration and the gang as an organization.

Few social scientists are as prepared to take risks as Bourgeois and Scott, but ethnography is

always a powerful, intense, and time-consuming design. Understanding the culture of others

requires preparation, systematic ways of recording and producing data, reflexive choices in the

writing. and a sociological imagination to create a coherent, insightful account.

AN EXAMPLE-LIQUIDATED: AN ETHNOGRAPHY OF WALL STREET

Only a few miles away from the fieldwork site where Bourgeois lived a short ride on the New

York subway-Karen Ho found a job in investment banking and entered the world of Wall Street

as an ethnographer to capture its corporate culture, a system of values and behaviors based on

a sustaining belief in smartness, exploitive hours of detail-oriented work for the young analysts,

absolute job insecurity, and the worldview that risk, impermanence, and the single-minded

pursuit of money should become the sole principles of global society. Like Bourgeois, Ho

immersed herself with no holds barred in an extreme setting (Ho 2009).

ETHNOGRAPHY IN EVERYDAY LIFE

In some sense, everyone is an ethnographer. Even the most ordinary life ushers in a parade of

ethnographic projects. As we encounter new and sometimes strange situations, we adopt an

ethnographer mind-set and course of action, although we rarely think of ourselves as

ethno-graphers. Our most ethnographic endeavors stem from major transitions or dislocations,

such as changing residence, traveling, starting a new job, being hospitalized for a long period,

beginning a residential drug or alcohol treatment, moving in with a lover or spouse or stranger,

or getting folded into a new circle of friends. At these junctures, we adopt the ways and means

of the

"professional stranger"; we approach life as an ethnographer approaches a research setting.

But what does this mean?

Literally, "ethnography" means "writing people" (graph = to write/depict; ethno = people).

This definition is dense with meaning and implications. Ethnography is the art and craft of

writing people. Performing ethnography successfully at the professional level demands a

combination of science, artistic sensibility, and vocational skill and training. It is part science,

part aesthetic creation, and part craft. In this chapter and in Part III "Focus on Ethnography,' we

offer insight into a complex methodology with a long and storied history in the social and

behavioral sciences.

Professional Strangeness

Ethnographers set out to study culture. More specifically, they seek to systematically investigate

and then accurately represent (through their prose and/or pictures and/or films, etc.) the culture

with an emic, or "inside-out perspective. This contrasts with other forms of doing sociological

research on groups or communities or organizations, where the researcher obviously assumes

an

"outside-in" (etic) vista and represents the issue or group from an outsider's perspective.

Chapter 7 • Ethnography: A Synopsis

Ethnographers strive to understand a given culture on its own terms, and on the terms of the

people who occupy it, dwell within it, and produce and reproduce it on a daily basis. They "write

people" from a perspective as observers who participate to some extent in the lives of the

people about whom they write.

The discipline of anthropology gave rise to ethnographic research but does not enjoy a

monopoly of this versatile methodology. In classic anthropological studies, the ethnographer

examined a primitive tribe in a foreign country, a tribe marked by sociocultural and geographic

isolation (an erstwhile "fact" debated hotly in retrospect). Such "pristine" and self-contained

cultures hardly exist these days, if they ever really did, and ethnographic studies in sociology

are increasingly domestic. Contemporary ethnographers are just as likely to be studying

"cultures within cultures" as they are to be studying indigenous cultures in some other land.

Today, ethnographers come from many different disciplinary backgrounds sociology,

anthropology, economics, medicine, law, communications, international studies, American

studies, and private-sphere commercial domains such as marketing/advertising and new

product research and devel-opment. Although each discipline has its own standards and

conventions for doing this kind of research, they share a common concern for "the cultural."

The ethnographer begins her study (which may mark a new phase of her life or serve as a rite

of passage) when she assumes the status of a "professional stranger" (Agar 1986). Over the

course of her time in an unfamiliar place, she gradually develops an insider's perspective.

Hardly ever does the ethnographer "go native," or become a full-fledged member of whichever

society she's studying.

Instead, she retains "critical distance," which ebbs and flows depending upon time and

circumstance.

In fact, it's almost impossible for the researcher to "go native." This contrasts with the popular

ethnographies we all perform as we encounter life-changing transitions. Most of us want to

become a full-fledged member of the culture and society into which we have moved

intentionally.

Solving Puzzles in Reverse

Ethnography at the professional and lay levels revolves around solving the problem of culture.

Think of culture as a completed jigsaw puzzle bearing an illustration of some sort. Let's assume

this particular jigsaw puzzle has 10,000 pieces; it's complicated. But unlike a real puzzle, the

picture we are striving to assemble is not shown on the box. As in everyday life, you must begin

to assemble the puzzle, piece by piece, through trial and error, and as you slowly develop some

idea of what the big picture looks like, you begin making decisions about how to put it together.

However, culture is more complicated than a static illustration. Each of the 10,000 individual

pieces continually changes its shape, and the total number of pieces keeps changing increasing

and decreasing without warn-ing. And even more problematic is that the ultimate illustration also

keeps changing because in the end the illustration you create is but one of many that could

have been created.

In your daily life you don't think about the culture in which you live. Nor do you spend much time

trying to visualize the culture in toto, writ large. That's because you don't have to. One of the

greatest privileges in life, arguably, is the freedom from having to figure out what your culture

consists of, what it "looks like," and how to operate fluently within it. Familiarity breeds

"functional ignorance" in many respects. The more familiar with your culture you become, the

less you will notice or see it, and the less capable of explaining it you will be. Ignorance of one's

own culture is a grand luxury, for it derives from many years' worth of assimilated, embodied,

rarely-if-ever challenged knowledge about oneself and one's place in the world.

When you find yourself in a new and foreign situation, you experience not only fear, trepidation,

and anxiety but also a sense of excitement and thrill, and these contradictory result from your

being culturally disembodied. You have yet to embody the knowledge and skill necessary to

function as a competent cultural actor. In short, you're an alien. Every day ushers in numerous

problems that need solving. Because you have spent so many years participating and observing

your own life, you know what many of these "problems" will be: feeling "at home" in your new

residence; moving through the new neighborhood in which you now live; obtaining a sufficient

supply of satisfactory-tasting food; getting to work or school; meeting and greeting strangers

and then negotiating your way into more intense relations with them (e.g., friendship).

During these moments in your life you pay very close attention to the little things to the things

people do, the words they say, the gestures they make, and how they carry themselves in

public.

You study the environment around you, and you try to make sense of things--you try to discern

patterns in how people behave as individuals and in groups.

You do all of this in an effort to figure out what's going on, to estimate what all of these details

add up to, how they are a product of and also express the "bigger picture" of life in your new

community. If, say, you're living and going to school in a new city or town, you cannot get to

know the entire community, the whole place. So you focus your efforts of knowing on the

immediate circumstances, contexts, and problems. A relatively narrow frame of reference

guides what you do, where you go, decisions you make, actions you take or don't take, and so

forth. Your hope is that by becoming acclimated to this small sector of the larger community, you

will become a competent member of the broader setting/scene/population as well. And if you

competently master the local lived experience, chances are you will succeed in the eyes of the

"host" society of which the local scene is one of many parts.

At some point down the road the adventure will end. You may well stir be living in this place that

used to seem foreign to you, but to some extent the thrill (and the need for) of concerted

discovery is gone. You no longer tap into that higher level of consciousness that reflects on the

rationale or strategy behind your actions or the actions of others. Rather than asking "how do I

live here" while simultaneously living there, you simply live. Living in this once-foreign place has

become second nature to you. As a member of the scene, even if some or many still consider

you to be an outsider of sorts, you embody the practices, values, and expressions of the

setting-you fit in. No longer do you think about daily life as a series of puzzles. Moreover, it's

very likely that you have come to believe that how you now do things is the only way you could

be doing them. But that isn't true.

Over time, you have participated observantly in your own life; you necessarily have been

immersed in your own life (this may seem obvious, but it's not), and you have sought answers to

pressing questions about your navigation through the world. Through trial and error, you have

developed solutions to a good many of the problems you face. And the solutions have synched

so well with the lives of others with whom you must interact that they have come to seem

"natural?

But they are anything but natural. You have developed them through social, economic, political,

and cultural interaction with others. In short, you have constructed your life ethnographically.

You have written yourself.

One of sociology's greatest thinkers, Everett Hughes, wrote extensively about the similarities

between ethnography and everyday life. For him, the two orientations had more in common than

not. In both domains one's success boils down to the capacity to assimilate information and

move forward by advancing educated guesses regarding what to do next. Just as learning to

live successfully and happily isn't easy, neither is learning to ethnography:

The problem of learning to be a field observers is like the problem of learning to live in society. It

is the problem of making enough good guesses from previous experience so that one can get

into a social situation in which to get more knowledge and experience to enable him to make

more good guesses to get into a better situation, as infinitum.

The point here is that ethnography is something we do every day. But our lay version is less

systematic, more taken-for-granted, and less reflexive (or critically contemplated) than the

professional version. Producing a good ethnography means learning the art and craft of

knowing, thinking about the logic of reaching conclusions about culture, and familiarizing

yourself with practices that will help you observe and write about culture.

ETHNOGRAPHY: A LOGIC OF KNOWING

Michael Agar (2006), a long-time ethnographer whose work we highly recommend you to read

(particularly The Professional Stranger, 1996), puts it best: "Ethnography names an

epistemology a way of knowing and a kind of knowledge that results--rather than a recipe or a

particular focus"

(2006: 57).

Ethnography is a design, an overall plan for understanding how a group exists. It generally calls

for the adoption of multiple methods and then bringing the resulting data to bear on a question

about a particular group's culture. Ethnography is a way of combining and using various

methods, both quantitative and qualitative, and so in a strict sense, it is neither a quantitative

nor a qualitative method; in any case, the guiding principle is that the ethnographic project

strives to answer questions about how a group of people gets along in the world. And you, the

ethnographer, assumed the responsibility of depicting how the group exists.

The Culture Question: "How?"

"How" is arguably one of the main focal points of ethnography. It may well be the central

question. As we've said already, we're talking about the study of culture. But what is culture?

This is another taken-for-granted, oft-tossed about term whose actual meaning we rarely make

explicit enough for agreement on what we're discussing. While hundreds of definitions exist,

we're going to adopt a relatively simple one for the purposes of this chapter: Culture consists of

a given group's members relatively well-shared, acquired knowledge for the purpose of

organizing behavior and also interpreting the behavior of others (on a small or very large scale).

In essence, culture is a shared way of understanding the world and taking action within it. And

it's always relatively well-shared, meaning that not everyone complies with the dominant

culture's rules and forms, and most people don't conform all the time. But most of us comply

most of the time, but to what are we complying? Figuring out the "what" is essential: "What

comprises the culture I'm studying?" But to get to the "what?" question, you've gotta deal with

the "how": "How did this culture arise and how do its members keep reproducing, if continually

changing, it?"

At any given moment, we exist amidst a bramble of rules governing our behaviors, our actions,

prescribing what we can and should do and think, and proscribing the transgressive behaviors

whose carrying out we generally resist no matter how strong the urge. Some of the rules into

whose province we were born are formal and official; they are laws. Most of the rules, however,

are less formal and often not even spoken aloud; they are mores, also known as customs or

conventions. Rules and knowledge are often embodied, experienced as ways of moving,

sensing, and acting in the material world, not formed explicitly into words and ideas.

At the group- or community-level, we refer to the body of mores as "norms," patterns of

expected behavior. And we generally agree on, or at least recognize and to some degree

assimilate, the informal code of behavior, and it's quite obvious when someone breaks the code.

Every behavior, or every action, gives shape and reinforcement to these norms, mores, and

laws.

They act on us, and our resulting actions act back on them ... in many ways, we are the law, and

the law is us. No matter how much we like to think or tell ourselves (or others) that we are

original, unique, one-of-a-kind, the fact is that we are very much alike in nearly every respect

(physiologically and socially). Our behavior takes on ritualistic qualities, not necessarily in the

spiritual sense, but in that our behaviors are predictable and patterned and derive in part from a

commitment to conformity.

spiritual sense, but in that our behaviors are predica commitment to conformity.

EXAMPLE-ACCOMPLISHED INTERACTION: To summarize by way of illustration, let's take up

the issue of noninteraction (as inspired by the work of Goffman, Blumer, Hughes, et al.). In most

locations across North America, people commit informally and implicitly, without ever really

discussing it, to avoid interacting with each other in certain venues: waiting room at the doctor's

office, bus stop, train, car, elevator, and so on. These places all have one thing in common:

They are way stations transitional settings--places we occupy en route to a final destination.

There's no rule against talking in these places (for the most part), and there's no law against it.

And neither parents nor teachers ever tell children, "Don't talk to others when you're in the

following locations ...

" Curiously enough, though, nearly all of us subscribe to a deep and abiding

commitment to noninteraction. We accomplish mutual avoidance in such settings. We don't think

much about our accomplished noninteraction until someone breaks the rule for example, a

fellow passenger begins chatting us up, or a nervous fellow patient tries to initiate extended,

meaningful conversation as we wait miserably to see the doctor.

Remember, sociology is the study of people doing things, and/or not doing things, together.

Consistently avoiding interaction in certain specific places is something that we do, and we do it

together, and our doing of it derives from a mutual commitment to uncodified rules and a

reciprocal trust that we'll abide by the same unofficial code of conduct. Though perhaps trivial,

this example illustrates culture-the shared ways of understanding the world and organizing our

actions in it. And it offers some insight into the most important, or least most pervasive, rules

and boundaries that govern behavior the norms, mores, and customs about which we rarely

speak or even think. Finally, accomplished noninteraction, as a cultural phenomenon, tells us a

great deal about the ritualized quality of our public behavior.

Culture as Process and Structure in Context

Culture is a process. It's also a structure. And, more important in terms of discussing

ethno-graphy, it is an outcome, an effect, a derivative; it is, in short, a result of human activity,

both cooperation and conflict. Too often, sociologists (and journalists alike) talk about culture as

though "it" causes things to happen. Culture doesn't make things happen. Rather, "it" flows from

humans attempts to adapt to their environment, to the social system in which they find

themselves, to the political structures shaping their "placement" in the hierarchy of material or

symbolic wealth (the economy). We develop patterns of behavior, ways of being and doing, and

they have a particular substance and style. Writ large, these patterns of activity and styles of

doing and ways of thinking comprise culture.

Ethnographers generally set out to understand culture as a problem to be solved, a puzzle to be

understood, an enigma to be contemplated, dissected, and explained. In a word (and often-used

word, in fact), we "problematize" culture. How does this particular culture come about? How can

we explain its emergence? What are its various parts? How do they work together, or not work

together?

What sustains this culture? How did this or that given culture dissolve or expand or otherwise

change?

But ethnography doesn't always involve such big questions as these. Often it's just not feasible

to study or explain an entire culture. And so we pose midrange questions. Rather than trying to

explain, for instance, the entire culture of homeless or precariously housed people or the culture

of substance abusers throughout the United States, the ethnographer begins with a local

context and a local group, defined by its participants or by outsiders. Generalizations begin to

flow from the intense study of the local group in its local context.

This is exactly what James Spradley (1970) does in the classic ethnography (and a leading

study in urban sociology) titled You Owe Yourself a Drunk. Spradley examines closely how the

culture of drunks in Seattle is sustained, shaped, and generally affected by the community's

"legitimate" institutions such as law enforcement, business, and political governance. One of his

many findings was that the very institutions hell-bent on remedying the problems related to

public drunkenness and vagrancy were the core factors responsible for sustaining these

"problems." Spradley's "field" was defined and delimited by his concern for the points of contact

between the culture of drunks and the institutional structures and policy enactments of the social

control organizations in the community.

Depth versus Breadth

Ethnographic logic also emphasizes depth over breadth. Earlier in this book, we argued that the

primary difference between quantitative and qualitative research lies in their respective

approaches to the creation of knowledge concerning some aspect of social life. Quantitative

researchers generally examine only a handful of variables but a great number of cases. Their

research questions lead them to administer surveys, for instance, wherein they ask a small

number of questions to a large number of people. Then they analyze the responses and try to

explain variance. Qualitative researchers, on the other hand, ask questions that demand deeper

investigation. The answers to their research questions tend to involve a large number of factors,

which means they can focus only on a relatively small number of cases. Ethnographers, much

like qualitative researchers, also work with a small number of cases (sometimes only one case)

and a huge number of variables (although many of them prefer not to use the term "variable,"

which sounds too much like traditional quantitative research). These variables include all the

characteristics of the social context and all the social forces that operate within it. Ethnography

is the study of a single culture or the comparative study of several cultures. In both cases,

however, one examines many situations, individual experiences, and factors (or

"variables") within each culture in order to develop a holistic, comprehensive understanding of

how the culture operates.

A holistic approach requires the ethnographer to study cultural processes in situ, or in the

natural setting of the people who occupy the culture. This approach differs markedly from typical

quantitative (and many qualitative) studies, wherein the researcher first identifies a phenomenon

or problem (as social science tends to be obsessed with problems and pathologies) and then

isolates it, extracts the problem from the conditions that produced it, attempting to quantify or

qualify the problem, and finally attempts to examine recurrences of the problem throughout a

population, with a focus on explaining how the problem's distribution varies. Ethnography

focuses on cultural processes in context and attempts to explain how cultures operate, how they

produce their own problems and solutions, how they relate to other cultures (whether hostile or

friendly), and so forth. Occasionally ethnographers will isolate a particular instance and build an

ethnographic study around it, but this is relatively rare. In general, the ethnographer goes for

depth and complexity, and ultimately attempts to understand culture itself as "the problem"

which is to say that the ethnographer asks the question, "What gave rise to this culture, and how

did it become what it is?"

Falsification

Counterintuitively, perhaps, the ethnographer achieves the ability to tell a holistic story about a

culture by continually attempting to undermine the story while she tries to assemble it.

Falsification is the heart of ethnographic logic. The ethnographer must continually attempt to

falsify every emergent conclusion she reaches. Here, we encourage you to refer back to the

"falsification principle" presented in our discussion of qualitative methodology's relationship to

the traditionally accepted doctrine of the scientific method. This operation is like trying to make a

puzzle piece fit into a spot and setting it aside if it is not a good fit, to continue our puzzle

analogy; if it is not a good fit, we do not insist on cutting or bending it to force it into place.

Ethnography's focus on falsification is influenced by the work of Karl Popper (1963), who

believed it is the basis of all scientific inquiry. But ethnography presents real problems to the

process of falsification because we don't use the methods of the natural sciences and the

quantitative social sciences such as probability sampling, double-blind testing, and tests of

statistical significance. In ethnography, we are constantly tempted to create data that support

our initial conjectures.

Regarding ethnography's near-obsession with falsification, Michael Agar (2006: 103) writes,

Ethnographers, by and large, suffer from a disease called "chronic Popperism," if I can refer

tongue in cheek to the writings of philosopher Karl Popper (1963). We keep looking for evidence

that what we think is going on in fact is not. We are ambulatory falsification machines. And the

falsification we seek relies heavily, though not exclusively, on questions about meaning and

context. (2006: 103)Rather than assuming you're right, even at the outset, assume that you are

wrong about the reason you're seeing, what you're seeing, or hearing what you're hearing.

When I (Scott) first began working with street addicts (heroin and crack cocaine), I assumed that

I could attract them to my interviews with any incentive payment (honorarium) of $10 or more.

But I was wrong. Many local addicts chose not to agree to be interviewed. They were in the

same economic situation as those who agreed to the interview, and their drug-addiction levels

were roughly the same. So what gives? Why did some participate, while others did not?

The answers are many, but one of the most powerful explanatory factors was their relative

connectedness to non-drug-using family members. No matter how badly they needed the

incentive payment, they would often decline the interview out of fear that their family might read

about them in my published ethnography. This obviously contravened my assumptions about

their level of economic need, which was intertwined with my assumption that they couldn't care

less about what others thought of them. After all, they were street junkies how much worse

could their social situations get. But many of them had too much pride and too much personal

capital at stake to participate in my work.

Chronic Popperism is the science of ethnography, and it's an element that clearly distinguishes

ethnography from journalism. Journalists don't spend a year, or often even a month, trying to

figure out how they're wrong about the conclusion they've reached concerning the relationship

between X and Y; instead, journalists often collect anecdotal information (which is not data,

even in the plural), draw conclusions that they consciously or subconsciously believe will appeal

to their editors (whose audience ultimately consists of those who provide the revenue

stream--the advertisers), and then seek out confirmatory evidence.

Once they've submitted their "copy" (the story), someone called a "fact checker" goes through

the material, follows up with sources to check the accuracy of quotes, consults authoritative

texts to verify dates and other factual aspects of the piece, and so forth. But the "fact checker"

checks "the facts"; they don't check the interpretation, and they certainly don't go looking for

contravening facts! In news organizations, there's usually no one who holds the responsibility of

challenging the story, providing alternative and competing explanations or narratives. Once copy

is submitted, everyone's mission becomes one of confirming and ensuring the narrative's

publication.

As ethnographers, we're looking out for disconfirming evidence, but with an eye toward

developing an explanation for how certain events fit a certain pattern, and how many other

events don't fit the pattern, and finally, why some events fit the pattern and many other events

don't. Out of all this, we hope to develop a relatively coherent, holistic account of the culture

we're studying and of the specific events, occasions, rituals, and interactions we've chosen to

focus on in our study.

A word of caution: Doing ethnography can drive you crazy. With its commitment to holism and

context (everything is connected with everything else) and its compulsive falsification attempts,

the life of an ethnographer can be maddening. Michael Agar calls it "heartbreak of the holistic

mind" (2006: 120).

Questioning Reality

Notwithstanding the ethnographer's fascination with complexity, holism, contingency, context,

situatedness, and interminable interconnectedness, the methodology's logic calls for simple,

though not simplistic, explanation. Ethnography entails questioning "reality, repeatedly and

crit-ically. What everyone assumes to be real by their implicit collective agreement not to

question it is exactly what the ethnographer sets to question. Through chronic Popperism, the

ethnographer arrives at an answer- -not inductively or deductively, but abductively.

Abductive logic (discussed in detail in a later chapter) means openness to both deduction and

induction (the traditional logics of social inquiry) and above all, an openness to imaginative

leaps. But in the end, the ethnographer strives to articulate the most economical, parsimonious

explanation possible, one that identifies the culture itself as problematic ("How is this culture

possible, and how does it work?") and offers the simplest possible explanatory framework. It

integrates the view from the inside with the view from the outside, making sense of why insiders

see the world as they claim to do.

As John Dewey (1938/1998) wrote, "Reality is what we choose not to question at the moment."

That's the ethnographer's focus everything that cultural insiders do not question.

And once asked, the question cannot be answered by insiders alone, for they rarely appreciate

the fact of having a culture. This is true for almost everyone. Can you tell me what your culture

consists of, exactly? How is your behavior an expression/product and producer of that culture?

You see how difficult it really is. You must offer your own explanation, one that accounts for w

insiders know but one that goes beyond their knowledge. But it should not be overly comp

Again, we're striving to meet the criteria of "good science."

Parsimony and Ockham's Razor

"Good science" also means following the premise of Ockham's razor: The theory that you

ultimately set forth should make as few assumptions as possible and the assumptions it does

include should relate directly to the observed events. The simpler the explanatory framework,

the easier it will be to test, and then to build upon- and if necessary, to discard. At the end, your

theoretical base will enjoy a foundation comprised of very few core explanatory factors, but they

will be intertwined with each other in infinitely complex ways. Hence, you need to develop

parsimonious postulates easily tested through observation.

The World in a Grain of Sand

Using this phrase, borrowed from William Blake's poem, as a touchstone, we conclude this

chap. ter by emphasizing again ethnography's concern with understanding thoroughly the local,

the particular, and the seemingly inconsequential. Life, death, the gods, and the devil are all in

the details (to modify an old adage). But although ethnography requires deep, long-term, always

taxing immersion, its critical that the ethnographer learn how to successfully step back from the

details and employ abductive logic. Many an ethnography has gotten lost in the middle of a field

study, so caught up in participating in the culture that the ethnographer never finishes the study,

fails to answer the research question, and so the job never gets done.

Ethnography Holds Up a Mirror

Seeing the world in a grain of sand means performing an ethnographic study that says some.

thing important about the culture under study. But ethnography also tells the outside world

something about itself. A great ethnography is a looking-glass experience for the readers. What

appears at first glance to be entirely foreign, exotic maybe, turns out to be a study that says

something about the "hormal" world in which the published ethnography circulates. If an

ethnographer can achieve this "natural generalizability," then she has accomplished a great

deal. A good ethnography of a "foreign," "exotic," or "outsider" culture reveals our own lives

and culture to us.

The ethnographer, whether doing a community-oriented study or a problem-centered study,

always has a "problem" (not in the sense of a social problem, but an intellectual puzzle): how to

explain something, how to make sense of patterns, how to accurately convey "what's really

going on?" Later in this book, we present several chapters on the specific techniques that

comprise

"doing ethnography. The endeavor may seem awfully mysterious at this point, and to some

extent there is a bit of mystery in this science, as in all sciences, even physics. Science and

mystery go hand in hand. As Albert Einstein (2011) remarked, "The most beautiful thing we can

experience is the mysterious. It is the source of all true art and all science. He to whom this

emotion is a stranger, who can no longer pause to wonder and stand rapt in awe, is as good as

dead: his eyes are closed." Ethnography is the science of mystery and of mystification. Indeed,

one might say that ethnography is the science of demystification done all-too-often mysteriously.

In this book, we hope to unravel some of this mystery of doing ethnography.

Many professions (and professionals) focus on writing people. Take journalism, for instance, a

field we've already invoked in this discussion. For hundreds of years, professional writers have

investigated the lives of humans, individuals and groups, and then rendered those lives (in

varying degrees of detail and accuracy) in public forums. They identify an issue or problem with

broad public appeal or relevance.

Background, or archival, research furnishes knowledge they will need in order to gain entry to

the community and to ask the right questions. Then they go out and try to find informants,

spokespersons, people whose lives bear directly on the issue at hand.

The journalist often attempts to build a trust-laden relationship with informants so that they will

share with the writer sensitive information and insight into how their lives bear on the issue, and

how the issue affects their lives.

profession by the early twentieth century, largely to shield them from World War I government

propaganda and the emerging advertising and public relations industry (Schudson 1978).

Professional bodies, such as the International Federation of Journalists, promulgate rules to

guide journalists who report "the news." Although many codes exist, they generally share a

concern for these common tenets of journalistic reporting: objectiv-ity, truthfulness, fairness,

impartiality, accuracy, and public accountability.'

In the 1950s, journalism spawned subfields bearing a striking resemblance to ethnography,

particularly a branch that came to be known as

"New Journalism" (also termed as "creative nonfiction" and "literary journalism"). Norman

Mailer's highly controversial 1957 portrait of the hipster as "The White Negro" sparked a

journalistic movement that wove together orthodox journalism's commitment to fact and

chronology, literature, creative writing's devotion to the crafting of lyrical prose, and

ethnography's obsession with "thick description" of people, places, things, and events.

With the publication of Truman Capote's book-length account of the murder of a family in rural

Kansas and the execution of the murderers (In Cold Blood, a title that refers to execution as well

as murder), the "New Journalism" genre achieved prominence. Tom Wolfe's later work on hot

rod culture, Hunter Thompson's dogged reporting on the 1972 campaign trail, and Joan Didion's

essays on emerging lifestyles and culture (Slouching Towards Bethlehem, 1969) knotted this

new strand of journalism firmly into the history of American prose. As the war in Vietnam

escalated, the New Journalists focused on the war and became more intense in their critique of

U.S. policies.

The New Journalists differed fairly dramatically from their more orthodox counterparts working in

newsrooms and broadcast stations around the world. The New Journalists rejected the principle

of "objectivity," arguing that in practice this fun damental tenet meant a bland, uncritical reporting

of interviews conducted on all sides of an issue, a refusal to engage on any issue, and

consequently a (partially unintentional) complicity in promoting the frames and agenda of

powerful elites. The New Journalists insisted that "objectivity" stems from being involved in, not

detached from, the action, the scene, the lives of the people about whom you're writing.

Reporting the truth required stating their own point of view. They also eschewed the formalistic

and staid aesthetic of most journalistic prose. Rather, they wrote lengthy accounts bursting at

the seams with singing, lively, lyrical prose.

If you read books like Thompson's Hell's Angels or Joan Didion's Slouching Towards

Bethlehem, you'll find a very different kind of writing from the increasingly standardized prose of

newspaper stories. In contrast to the rather predictable sentence structure and tepid description

of the standard news story, the New Journalists' report unfolds in scenes with full dialogue

captured, or re-created; they don't rely on decontextualized, practically disembodied quotes

from bystanders with a second- or even third-hand account of what went down. Re-created

dialogue is central to the New Journalists' writing tool-kit. At the same time, though, the stories

are self-consciously just that-stories. They are far less concerned with "facts" and far more

concerned with conveying the "vibe," the essence of the moment as they experienced it, the

expression of the writer's

"interiority," for New

Journalism balks at the factual, acknowledging that what is "fact" is always under contest,

forever in flux, and perpetually unsteady.

The New Journalist strives to construct a believable (if self-referential) story, whether or not she

gets the "facts" right.

The New Journalists developed the stories; they wrote people. Developing the stories meant

immersing themselves, sometimes for long periods, in the situations that intrigued them. But

they upheld the journalistic virtues of experiential accuracy (as opposed to factual accuracy),

fairness, and public accountability. But impartiality, they argued, is unattainable undesirable

even. Why would anyone want to be impartial toward important, newsworthy events and

people? After all, partiality taking a perspective is a product and producer of passion, and the

best-told stories are those told passionately about passionate experiences.

Much like ethnographers, the New Journalists paid close attention to "status details," the

features of the environment, the characteristics and mannerisms of the "characters" in the

stories, the little things that add up to a way of life for a person. The opening sentence of On

Boxing, by Joyce Carol Oates, illustrates the New Journalists' belief in the adage "the devil's in

the details":

The young welterweights are surely conscious of the chorus of jeers, boos, and catcalls in this

great cavernous space reaching up into the cheap twenty-dollar seats in the balconies amid the

constant milling of people in the aisles, the commingled smells of hotdogs, beer, cigarette and

cigar smoke, hair oil.

in a pattern familiar to qualitative researchers, the New Journalists move back and forth

between the concrete and the abstract, the details and the generalizations about the bigger

scene of which a given moment or event is but a part. Later in the same essay, Oates pulls

focus and brings the reader to a higher level of analysis, as she shares some of her inferences

concerning the sport, art, and craft of boxing:To enter the ring near-naked and to risk one's life is

to make of one's audience voyeurs of a kind: boxing is so intimate. It is to ease out of sanity's

consciousness and into another, difficult to name. It is to risk, and sometimes to realize, the

agony of which "agon" (Greek, "contest") is the root.

Tom Wolfe, one of this genre's founders, once remarked that the only way to figure out how

people see themselves is by scrutinizing the minutiae of their surroundings. In this way, he said,

we can perform a "social autopsy" of a given character in a story. The New Journalists' social

autopsy looks a great deal like the social autopsy method we describe in a later chapter.

THE JOURNALIST ANDTHE ETHNOGRAPHER.

The journalist and the ethnographer work in similar ways, especially if we're talking about the

New Journalists and the investigative journalists who produce multivolume series. But there are

some important differences between journalistic work and ethnographic inquiry.

First, there are differences in the time frame.

Ethnographers tend to cultivate their "stories" and do their investigative work and their writing

over a much longer period of time. While the journalist may consider two or three months to be

a huge commitment for a single story, it's not unusual for ethnographers to spend two to three

years in the field and another two to three years writing up their experiences and findings.

Second, journalists and ethnographers differ with respect to theory. By name and usually by

conscious practice, ethnographers work within the social sciences. And while they argue

strenuously over the politics, significance, purpose, and practice of ethnography, there is

general consensus that ethnography is more than storytelling.

Third, ethnographers are more interested in generalizing their findings. Beyond a commitment to

accuracy or "truth" in ethnographic representa-tion-making, the ethnographer strives to say

something about the so-called "human condition.”

The specific "case" at hand is certainly important and valuable in its own right; but the

ethnographer's real interest lies in examining and relating the case in ways that help readers

better understand the world in general, or at least some aspect of the more general world.

Scott's (2009) ethnographilm The Family at 1312, for instance, ostensibly focuses on a "fictive

kinship" system developed and perpetuated by crack smokers, prostitutes, and drug dealers. In

dissecting this so-called pseudo-family, Scott's film holds up a mirror in which the viewer can

see some of the fundamental dynamics that operate in the majority of the "normal" families in

contemporary America.

Fourth, ethnographers and journalists impose very different narrative structures on their writing.

Journalists like to organize their stories around conflicts, people dramatically taking opposing

sides on an issue, or heroic struggles against the odds.

Ethnographers do not look for or find this narrative structure in their observations of cultures in

context.

Nor do they organize their data to fit familiar plotlines like the quest or the Hollywood ending.

Most published ethnographies don't hang on a narrative arc, nor do their writers concern

themselves with presenting a coherent plotline, although they try to write in a vivid, readable

way and to present their "research subjects" as real and complete human beings, not cardboard

cutouts representing concepts. The plot of an ethnography, however, typically emerges not from

the events and lives within the culture under scrutiny; rather, the plot consists of the unfolding of

the argument presented by the ethnographer. And the argument generally derives from the

application of the ethnographer's conceptual and analytic frame to the evidence she produces

as a result of long-term interactions with the culture she has chosen to study. The narrative

structure of the ethnography is its rhetorical unfolding.

CONCLUSION

In this chapter, we have provided an informal introduction to the art and science called

"ethnog-raphy. At this point, you should have a pretty good grasp on how ethnography

compares with the kind of inquiry in which we engage when we find ourselves in new and

strange situations of everyday life. We examined the building blocks of ethnography's "logic of

inquiry," its focus on cultures, its attention to context, and its concern with falsifiability in a

process that does not conform exactly to the scientific method practiced in the natural sciences.

We also elucidated some of the similarities and differences between ethnography and

journalism, its closest "ally" in the professions that involve writing about people.

OVERVIEW

Historical-comparative (HC) qualitative research design is guided by macro-level research

questions, and in that regard it is different from most qualitative research, which leans toward a

microlevel of analysis and is based on observations that take months or years to unfold, but not

decades or centuries. In HC research, we explain large-scale differences among societies. The

unit of analysis (UAO) is usually a whole society or a type of society, such as capitalist societies

or feudal societies. For example, we might be asking why societies end up with different types of

political institutions, religions and values, economic arrangements, or class structures. We

would explain these differences among societies in terms of large institutions and structures.

HISTORY OF HISTORICAL-COMPARATIVE RESEARCH

You may remember from Chapter 2 (history of qualitative research) that HC strategies originated

with Weber and other classical theorists. Certainly Marx and Engels were also engaged in HC

analysis as they tried to understand the origins of capitalism and its global impact. In the late

nineteenth and early twentieth centuries, a favorite explanation of societal and cultural

differences was based on the notion of race; scholars as well as the general public in Europe

and the Americas believed that humanity was divided into a limited number of races, each with

distinct mentalities as well as physical features, and that these mentalities determined the type

of society that each race formed. The great classical sociologists-the theorists of the period

whose work we still read today all rejected racial explanations and explored societal differences

in terms of historical development of institutions and culture, not biologically inherited

mentalities. In the view of the classical sociological theorists, small initial differences in social

arrangements in ancient times (possibly related to climate and other environmental conditions)

produced cumulative and expanding effects, creating major societal differences by the dawn of

the modern period around 1500 C.E.

Weber and the other great classical sociological thinkers (most notably Marx and Engels, and

Durkheim) introduced HC design as a two-step process. We begin by defining a situation or

social arrangement as an ideal type, an abstract and extensive definition, and we then seek to

understand where it occurs (in what historical periods and contexts) and why its actual

manifestations (as opposed to its definition on a printed page) differ from the ideal type. To do

this well, we need to immerse ourselves deeply in historical knowledge. HC research requires

training in historical analysis rather than field experiences.

For example, we might define capitalism as an economic and social system characterized by

private ownership of productive property, markets, and free wage labor (not slavery or serfdom);

and then explore where, when, and how it emerged and what accounts for variation among

capitalist societies. For example, we could try to explain why markets are more regulated in

European countries than in the United States.

Let's look at several examples of historical-comparative research. The examples of HC research

will clarify the logic of the design.

Our first example is Eric Wolf's Peasant Wars of the Twentieth Century. Wot looked at six major

upheavals of the twentieth century: the Mexican Revolution that began in 1910 and son tinaed

until the 1930s the Russian Revolution of 1917; the Chinese Revolution that began in the 19205

and consolidated a regime by 1949; the Algerian Revolution against French colonial rule in the

late 19505 and early 1960s; the Vietnamese Revolution that began as a movement against

French colonial rule, continued as the United States' "War in Vietnam," and finally consolidated

a new unified state in 1975; and the Cuban Revolution (1999).

'These upheavals were violent, multifaceted changes in political systems, class structure,

culture and ideology, and economic organization. All of them took a long time to unfold and

involved millions of participants. They were not just quick military coups. Journalists and social

scientists would call them "revolutions" indeed many of them are always labeled with that word-

but for Wolf the word is only the beginning of the analysis.

Wolf's research question is focused on similarities: What led up to the armed struggles and the

overthrow of the central political institutions (whether it was a state or a colonial regime)?

What caused the revolutions in the first place? Was there a key precondition that all of these

cases have in common, despite the fact that they are located across the globe in very different

cultures?

This question could be answered in many ways. We might look at the organization of the

revolutionary movements, weaknesses in the military and administrative apparatus of the

central government or colonial regime, or cultural change such as the impact of Western ideals

of democracy and political rights, inadvertently introduced by Western powers as they directly or

indirectly controlled these regions. These explanations would have emphasized the role of elites

-emerging leaders who formed revolutionary movements and established new national identities

and the political, military, or cultural authorities of the collapsing regimes who lost control. Wolf

does not dismiss these explanations but he subordinates them to another perspective that looks

at the underlying cause that precedes these political and military shifts.

Wolf answers his research question by looking at what happened to peasant communities as

capitalism penetrated the countryside. The transformation of rural social relationships into

capitalist ones happened as a result of land becoming a commodity available for private owner

ship, growing and fluctuating markets in agricultural products, and wage labor as a replacement

for traditional obligations. Wolf argues that this economic, social, and cultural transformation

created the conditions on a vast scale for the uprooting of traditional peasants, their molding into

a proletarianized labor force, and the consequent onset of social instability in the countryside.

The destruction of traditional social relationships in the countryside was a necessary and

sufficient condition for the mobilization of revolutionary movements and armies.

Notice that capitalist penetration of rural areas was the common denominator of all six cases;

whatever their initial cultures and agrarian social arrangements, all six were impacted by

capitalist agrarian development, whether it was introduced by a colonial regime (as in Algeria

and Vietnam), by foreign companies (as in Cuba), or by national capitalists and landowners

turned capitalist, orienting production toward global markets (in Mexico during the presidency of

Porfirio Diaz, in Russia, and in China). This similarity explains the similarities in the outcome of

all six cases.

Capitalist penetration of the countryside is the independent variable and revolution redefined as

a peasant war is the dependent variable, if we reformulate Wolf's analysis in these traditional

quantitative terms. And they are very big variables. Notice how these variables are not defined

in measurable quantitative terms but as ideal types.

The reader may at first glance think that Wolf's hypothesis, deeply rooted in Marxist theories of

capitalism as a global system, cannot be falsified. But in HC research, alternative explanations

and hypotheses are just as possible as in quantitative designs or ethnographies based on

participant-observation in the field.

REVOLUTIONS Theda Skocpol asked a similar research question but proposed a different

answer (1979). Skocpol's question was this: Why did France, Russia, and China experience

massive revolutions, upheavals marked by both transformations of the political system and class

conflict on a very large scale, with the annihilation of aristocracy and landowners as a class?

Notice that her cases are not identical to Wolf's; she includes a case the French Revolution-that

took place before capitalism became a global economic system, and she leaves out the more

recent, anticolonial cases.

Skopol's answer is that the old regimes, protecting an increasingly oppressive agrarian class

structure dominated by a landed class, not weakened in military engagements. As she words it,

The revolutionary crises developed when the old-regime states became unable to meet the

challenges of evolving international situations. Monarchical authorities were subiected to new

threats or to intensified competition from more economically developed powers abroad. And

they were constrained or checked in their responses by the institutionalized relationships of the

autocratic state organizations to the landed upper classes and the agrarian economies. Caught

in cross-pressures between domestic class structures and international exigencies, the

autocracies and their centralized administrations and armies broke apart, opening the way for

social-revolutionary transformations spearheaded by revolts from below. (1979: 47)

Her methods included careful examination of secondary accounts written by historians.

Her conclusions offer a different view than Wolf's, encompassing Weberian as well as Marxist

perspectives: The "ancien régimes" as political structures failed to protect the class structures;

political actions undertaken by the regimes specifically failed military operations -weakened the

political structures that protected and defended the underlying class structure.

Goodwin and Skocpol (1989) also posed a challenge to Wolf's work in their study of Third World

revolutions, a list that overlaps Wolf's although focused entirely on upheavals that

culminatedafter World War It. This list does not indude the Mexican and Russian revolutions and

for all praci. cal purposes the Chinese Revolution, which was essentially complete by the end of

World War I1.

Capitalist social arrangements had appeared virtually everywhere in the world after Word War Il

except in the soviet bloc and China. Markets, wage labor, private ownership of productive

property, and commodity production were dominant or at least major institutions of all non.

Communist societies, even if not of communities in remote regions. So why did some regions

and nations experience revolutions, while others, equally changed in their social structure, did

not Goodwin and Skocpol argue that although capitalist relations of production may have been a

necessary condition, they were not a sufficient condition for revolution and did not in any

automatic or direct way produce a revolution. They also argue that revolutions appeared only

where significant parts of the population were excluded from power; it was exclusion from

power, and not capitalism alone, that was the precipitating condition for revolt, armed struggle,

and regime change, Goodwin and Skocpol identify the types of societies in which exclusion was

the common experience and they claim there were two of these: one was the "family-style"

dictatorship in which a single extended family and its friends and cronies dominated the

economy and political system, such as the Trujillos in the Dominican Republic, the Somozas in

Nicaragua, and the Haile Selassie regime in Ethiopia. Segments of the population that were not

connected and subservient to the dominant family were excluded from power, even if they were

educated and prosperous.

The second type of regime that was toppled by successful revolutionary movements was direct

colonial rule, a form of colonial administration in which little effort was made to co-opt traditional

indigenous authorities such as chiefs, kings, or rajahs. These exclusionary colonial regimes did

not create figurehead intermediary authorities and institutions, legitimated by the fiction that they

represented the legacy of traditional rulers and authorities. The use of traditional or

pseudo-traditional authorities to run a colony is called "indirect rule." The English generally

favored indirect rule and were clever at promoting and inventing "native potentates" and creating

symbols of traditional authority (visible not only in their overseas colonies but in the Scottish and

Welsh areas they occupied in the formation of Great Britain- regiments wearing kilts, indigenous

titles for English rulers of Celtic territories, etc.) (Hobsbawm and Ranger 1992). The French

favored direct rule and made little effort to co-opt indigenous rulers, thereby creating a feeling of

exclusion.

Two of Wolf's "peasant war" cases- -Algeria and Vietnam--were French colonies; Cuba could be

considered a family-style dictatorship in which Batista favored his cronies. Wolf's other cases lie

somewhat or entirely outside of the period considered by Goodwin and Skocol.

Remember also that Goodwin and Skopol are looking at successful regime changes; so a

protracted armed struggle like the one in Guatemala does not fit into their set of cases because

it was not won by the armed insurgency.

Have Goodwin and Skocpol "falsified" Wolf's hypothesis that capitalist penetration of the

countryside is the precondition of Third World revolutions? Or, have they only entered an

intervening variable the presence of an exclusionary regime into the equation, at least for the

post-World War I cases? Note that they have also changed the theoretical foundations, from a

predominantly Marxist perspective to one that is more closely linked to Weberian conflict theory

with its emphasis on political action as a force that is co-equal in explanatory power with the

economic base.

SOCIAL MOVEMENTS A couple of examples of HC analysis of social movements may provide

useful illustrations of this design.

Benedict Anderson (1999) asks the research question what precipitated nationalist movements

and his answer is that the causal forces were western education, literacy, and awareness of

common cultural interests- all inadvertently fostered by colonial rulers as they developed strata

of native elites and intellectuals in the process of administering their colonies.

In his study Imagined Communities, Anderson's dependent variables are nationalism as an

ideology and nationalist movements; these elements of collective action are the outcomes of the

process he analyzes. His independent variable is communication in the colonial territory that

helps to form in the minds and discourses of educated indigenous strata "an imagined

community"-a cognitive map of a single nation. Nationalism is not a "natural" or "instinctual"

sentiment, but a discursive and cognitive process that takes place among people with certain

types of education.

Anderson's cases are located in Latin American and Southeast Asia. In Latin America in the first

decades of the nineteenth century, creole (native-born of European ancestry) elites led

independence movements. In southeast Asia, nationalist movements began in the twentieth

century and achieved their aims in Indonesia, Vietnam, and the Philippines.

Unlike Wolf, Goodwin and Skocpol, Anderson uses not only secondary accounts but relies

heavily on memoirs- life narratives of nationalist leaders, a method we will discuss in Part III.

Martin Riesebrodt (1998) also looks at similarities in movements, in this case two recent

movements, the Islamicist movement that came to power in Iran in 1979 and political religious

fundamentalism in the United States. At first glance, these two movements seem quite different:

one is Islamic and successfully formed a regime, while the other is Christian and a political force

in the United States but not a governing power. It is not even clear that "fundamentalism" is an

appropriate label for the radical Shi'a Islamic movement at all. But in Weberian fashion,

Riesebrodt argues that they both represent an underlying ideal type: a stance of ambivalence

toward modernity that embraces modern technology and the form of the mass media, and even

the ideal of the democratic state, while rejecting modern secular forms of social relationships,

gender relations, and the family. Riesebrodt argues that both movements (and by implication,

similar conservative currents in Islam, Judaism, and Christianity) are propelled by the goal of

reestablishing patriarchal gender and family relationships. Riesebrodt is following the lead of

Weber here, in looking at religion, ideas and values, and political action as major forces in social

outcomes. Weber was just as interested as the Marxists in economic and class relationships,

but he was more inclined to think of religion and politics as independent variables than Marxists

generally do (Zeitlin 1997: 197-207).

The recent "transition to democracy" that took place in many nations in the last decades of the

twentieth century offers many opportunities for HC analysis. In this period, a number of

countries "returned to democracy" after a period of military or one-party rule. This widely used

expression immediately challenges us to consider whether we can establish an ideal type of

"return to democracy"- Is it indeed a single phenomenon and if so, to what other social

arrangements is it related?

Przeworski (1991) argues that we can indeed do this. His research question is what happened

in a number of countries by the 1990s, and his cases include both Eastern European countries

formerly in the Soviet bloc and under single-party state-socialist rule and Latin American

countries (specifically Chile, Argentina, and Peru) that "returned" from rule by right-wing military

regimes.

He argues that in both sets of cases, the political transition to democracy is accompanied by an

economic transition to markets (or to freer markets, in so far as the South American countries

were capitalist societies throughout the period of the dictatorships). Notice the very large

variables in this analysis democracy (and its absence), the presence and structure of markets,

the concept of a return to democracy, and the behaviors of institutional political actors such as

labor and capital.

Kenneth Roberts (1998) takes a different methodological approach to a historical-comparative

research project with a similar research question as Przeworski's. He explored the prospects for

and barriers against a "deepening of democracy" in Peru and Chile. By deepening of

democracy, he means extending participation in politics to a larger public and making it more

meaningful and more responsive to the needs of citizens. In other words, he means citizen

involvement that includes the minimal requirements of civil rights and liberties and regular

competitive elections, but goes further to include ways of bringing ordinary citizens into public

discussion and decision making. He examined two cases in which the minimal forms of

democracy were reestablished after military rule Chile was ruled by a right-wing military regime

from the coup carried out by August Pinochet against Salvador Allende on September 11, 1973,

until the late 1980s when the Pinochet regime was forced to begin making concessions to

opposition forces. Peru was dominated by an elected authoritarian government under Alberto

Fujimori who had restricted democratic freedoms after consolidating his power in what many

observers termed an "auto-coup," a coup against his own democratically elected presidency; his

authoritarian rule was finally brought to an end in public exposure of corruption and repressive

policies. The choice of these two cases creates a different logic of analysis from that of

Przeworski's analysis of a return to democracy: Przeworski's cases are mostly countries that

were substantially industrialized and modern with a largely literate and westernized population,

both in eastern Europe and in the

"southern cone" of Latin America. By comparing Chile with Peru, Roberts created a very

complex set of comparisons and contrasts for analysis: Peru is a much poorer country than

Chile and is characterized by a large indigenous population and a considerably lower level of

industrial development. His analysis therefore must take into account cultural differences and

very different class structures, despite a degree of similarity in the return of the political system

to democracy.

Indeed Roberts found major differences in the way democracy deepened but only within definite

limits in both countries. In Chile, movement organizations on the radical left were increasingly

forced into the model of standard electoral politics-playing by the rules of electoral competition

typical of developed capitalist countries; if they failed to enter this game, they recognized (or

believed) that they would not be able to play an important role in Chilean politics in the period

during which institutions were gradually restored to democratic functioning. On the other hand,

in Peru, large parts of the left refused to play the "bourgeois game" of participating in electoral

politics and instead concentrated on organizing grassroots community groups in poor

neighborhoods, hoping to create social movements that could survive outside of the

well-defined limits of electoral politics.

A very brief summary would be that in Chile democracy was not deepened because there were

pressures on radical democratic groups to join the game of electoral politics, whereas in Peru

there were also problems in deepening it because too many radical groups insisted on

developing relatively limited and often competing and even antagonistic grassroots

organizations leading to a disconnect between radical groups and the electoral system. Roberts

examines the political choices made by the groups and parties as collective actors in the

political arenas of the two countries. He interviewed political leaders and tends to interpret the

outcomes in terms of their choices within the global system; these choices are constrained but

not simply determined by these larger social forces, such as the collapse of the Soviet Union,

which meant that the Chilean left found itself with little support for more intransigently left

options. Ultimately, as in Skopol's work, the guiding theory is probably Weberian, an emphasis

on the action of individual and collective actors and the expression of their purpose in political

organization; it is also compatible with Bourdieu's concept of the political field (hancock and

garner 2009) through Roberts does not down much time on these underlying theories.

CHARACTERISTICS OF HISTORICAL-COMPARATIVE KESEARCE

Let's sum up what all of these studies have in common:

They are about very large units of analysis- generally whole societies, political systems,

capitalism as a global system and other types of social formations such as feudalism and

tributary kingdoms, class structures, and so on. Individuals are not the unit of analysis.

• The analysis covers long time spans, processes that take decades or even centuries..

•They focus on relationships among big variables, such as the type of economic and social

system, revolutions, the actions and ideologies of movements, and global political change.

These variables are almost always defined in terms of ideal types and the values of the

variables are identified categorically, at a nominal or ordinal level of measurement, not with a

quantitative metric.

• The studies examine the variables in their complex contexts; variables are not "pulled out" and

measured separately.

The studies use small samples (small N), generally less than 30 cases, and sometimes as few

as two.

The researchers examine large volumes of secondary sources, including writings by historians,

eyewitness accounts, life narratives and memoirs, and documents; only in a few of the HC

studies are interviews or observational methods used. Often "nonobtrusive measures" are

needed because the actors are now all dead and not available for interactive methods.

Weber's concept of the ideal type guides research, with the analysis focused on why and how

similar institutions or transformations emerged in different situations and on reasons for

variation.

• HC research faces challenges in establishing falsifiability. It is not impossible to do so, but a

straightforward hypothesis-testing format is rarely employed because the situations are

"large," unique in key respects, and historically situated. The large number of variables and their

highly contextual character creates problems for conventional hypothesis-testing procedures,

and so does the necessity of using nonrandom samples. Instead of falsifiability based on

statistical tests of significance, the emphasis is more on plausibility and the absence of better

explanatory frameworks; identifying preconditions rather than verified causes is the goal of the

research, a situation that is prevalent in many forms of qualitative research. Short of recruiting

research assistants among celestial beings who can test events in parallel universes, it is

difficult to use designs such as experiments and multivariate analysis to determine causality in

historical events.

It is hard to say that the French king's problems in the French and Indian war against the British

in North America and later his generous support of the American Revolution- generosity that

doubtless also offered a chance to get back at his traditional British antagonists-"caused" the

weakness of the French monarchy and precipitated the French Revolution. To truly test this

proposition, we would have to run a historical experiment in which the Louis XVI wisely refrains

from this foreign policy venture in one version of reality while engaging in it in the other version-

or better yet, observe his choices in a statistically robust set of parallel universes to see if the

choices make a significant difference in the course of French history-truly the stuff of science

fiction, not sociology.

• The research questions link the big variables in big hypotheses, such as the one that argues

that new, capitalist socioeconomic relations in the countryside were the precondition of the

emergence of revolutionary movements in the twentieth century.

Theories are strongly integrated into the project, and it is usually foundational theories that drive

the research questions and guide the interpretation of data; and these theories are often based

on Marxist or Weberian conflict theories.

• HC research leans toward a nomothetic orientation, and in that regard differs from some

historical research which is more ideographic; in other words, although the canons of the

scientific method have to be modified to study historical events and trends, social scientists in

HC remain confident that generalization is possible and that general patterns can be discerned.

This position is the nomothetic one--the one that searches for recurrent patterns and predictable

relationships among phenomena, as opposed to the ideographic view that all historical events

are unique and can only be recounted as interesting narra-tives, without the effort to find

underlying patterns or regularities.

WHEN AND HOW TO USE HISTORICAL-COMPARATIVE ANALYSIS

Now that we have looked at a number of examples of the logic of HC studies, we can say a bit

more about the choice of methods and the ways in which evidence is amassed, the way the

data are grown to use our farming metaphor.

• A small number of cases formed the basis of the sample--the researchers treated societies as

the units of analysis,

• Primary and secondary documents can be important sources of information.

• Many social scientists rely on historians to do the "grunt work" of historical research, the actual

labor in archives. They rely on their understanding of the quality of historical research and

interpretation--for example, the quality and reputation of journals of historical research- and

come into relatively little contact with primary data.

• Interviews and direct observation can be used only when participants are still alive, as was the

case in Roberts research; even in that situation, observation may be very limited if the

movements or political groups are no longer operative or the political situation has shifted

dramatically.

• Life narratives and memoirs were used by several researchers, such as Anderson.

CONCLUSION

Doing historical-comparative research is hard work! Although operating at the macro-level

research in terms of the dominant UOA, the research must attend painstakingly to the details

within and across cases. Moreover, the researcher's job is to identify and account for their

similarities and differences, and how various factors intersect, combine, converge, or otherwise

"interact." In the end, the researcher must craft a persuasive and well-evidenced argument

concerning how different cases came to produce similar outcomes and/or how very similar

cases ended up with very different outcomes. This chapter has presented a variety of examples

of HC research, and through these illustrations we have revealed the basic architecture of HC

research designs (in an abbreviated manner, of course). This kind of research can be very

rewarding, but at the same time frustrating because of the number of variables, the complexity

of their interactions, the challenge of selecting cases, the "wiggle-room" for alternative

interpretations, and the volume of potential data, whether from archives, secondary sources, or

participant narratives. In the end, however, HC scholars produce some of the most influential

and widely read texts in the field of sociology.

OVERVIEW

In this chapter, we present the "social autopsy," a design that dissects social causes and

consequences of adverse events in order to learn more about societies, institu-tions, and

organizations. We provide several examples of autopsies in order to illustrate the goals and key

components of this design. We also offer some practical guidance on designing and carrying out

your own social autopsy.

CONTEMPORARY LIFE AND THE DEMAND FOR AUTOPSIES

In the autumn of 2005, Americans watched with disbelief (and horror in many cases) as the city

of New Orleans was lashed by a hurricane and a flood, thousands of its residents were herded

into the Superdome and dispersed to all corners of the nation, whole neighborhoods became

uninhabitable, and about 1400 people died.

Hurricane Katrina and the flooding that accompanied it seemed to rip the cover off of

uncomfortable truths the poverty of so many residents of New Orleans, the poorly maintained

state of the levees and canals, the confused response of both state and local government, the

callous treatment of survivors, and inadequate leadership at the top level of FEMA, the federal

disaster management agency. Many political leaders and pundits referred to Katrina as a natural

disaster- an "act of God"-but critics pointed out and continue to point out the fact that human

failures contributed to the differential vulnerability of New Orleanians and to the fumbling of

agencies entrusted with disaster management and citizen security.

Katrina is certainly not the only instance of a disaster leading to exposures of institutional failure,

organizational and governmental mismanagement, social disparities, and inadequate planning.

In some cases, the disaster is largely natural even though the responses are shaped by social

institutions, in other instances, the disaster or adverse event is a social product from start to

finish. The analysis of institutional failures has come to be called "social autopsy" and it is the

basis for a special type of case study. The disaster uncovers underlying prob-lems, and social

processes that are normally concealed or opaque become visible and transparent.

Therefore, it is a powerful method for studying organizations, institutions, and whole societies.

The term "social autopsy" came into use especially with Eric Klinenberg's (2002) Heat Wave,

one of the most comprehensive and widely read sociological dissections of a disaster.

Klinenberg examined the distribution and causes of the deaths attributable to the 1995 heat

wave in Chicago. He identifies the logic of a social autopsy in these words:

What makes the heat wave such a meaningful event is that it represents the exemplary case of

what Marcel Mauss [a French social scientist and student of theorist Emile Durkheim] called a

total social fact, one that integrates and activates a broad set of social institutions and generates

a series of social processes that expose the inner workings of the city. (2002: 32)

TYPES OF EVENTS: NATURAL DISASTERS, ACCIDENTS, AND INTENTIONAL ACTS

The events that can be autopsied include natural disasters, accidents in high-risk systems, and

intentional acts. Hurricane Katrina, the Chicago heat wave of 1995, the 2010 floods in Pakistan,

the earthquake in Haiti, and the December 2004 tsunami in the Indian Ocean are examples of

natural disasters. The social autopsy focuses on the populations that were at risk and the

response of governments and relief agencies.

A second type of "adverse event" that can be autopsied are accidents, such as coal mine

disas-ters, the sinking of the Titanic, the nuclear power plant accident in Chernobyl, Ukraine, the

chemical leak in Bhopal, India, the BP explosion and leak in the Gulf of Mexico, and the

February 2008 recall of meat from "downer cattle" animals too sick to stand. These incidents

were certainly not intentional, but neither are they "natural disasters." They are the result of

decisions made in the management of an organization or errors in organizational practices and

routines. The organizations are involved in a high-risk technology or are operating in a

complicated natural setting, such as an iceberg-filled ocean or the depths of a mine. The

organization or system in which the accident takes place operates within a larger social

context--a local and national culture and a political and governmental system. For example, in

some countries a high rate of coal mine accidents is considered "normal" or "acceptable," while

in other countries the national culture and a strong union movement resist this naturalization of

mine accidents and insist on stringent safety standards. The autopsy addresses all of these

levels.

A third type of autopsy dissects an event or situation that is entirely "human made"_-for

instance, the torture and humiliation of prisoners that took place in the prison at Abu Ghraib in

Irag under U.S. occupation, the Watergate break-in planned by the Nixon administration, or

instances of ethnic displacement and genocide. Here the analysis links the actions of lower level

personnel to the larger organizational, political, cultural, and international context.

Examples of Social Autopsies

THE BUFFALO CREEK FLOOD Sociologists and social psychologists had studied disasters for

decades during the period after World War Il, often focusing on collective behavior and social

disorder that emerged during the event or on the traumatic experiences of survivors. But one of

the more comprehensive, critical, and socially contextualized studies of disaster was written by

Kai Erikson (1976) about a disaster that occurred in West Virginia in 1972, the Buffalo Creek

flood. His study already had many of the characteristics of the complete social autopsy.

The flood raised the question what do we mean by "natural disaster" or "act of God"?

Buffalo Creek flowed through a valley in West Virginia. Pittston Coal Company, one of the large

mining companies in the region, had created a series of sludge ponds high in the valley, pools

that formed amid the piles of material removed from the ground during the mining process.

These sludge ponds and the surrounding mud and debris became highly unstable; in a season

of heavy rains the sludge ponds finally burst loose and flooded the valley with 130 million

gallons of mine waste water and debris. As the water, mud, and rocks rushed through the valley

it picked up additional debris--trees and houses torn from the ground. When this terrifying mass

hit homes it swept them up and killed their inhabitants, drowned in the raging waters or crushed

by swirling logs, housing materials, rocks, and mud. By the time the flood was over, 125 people

were dead and 4000 were homeless. Many of the survivors had seen loved ones killed and had

come near-death themselves. Was this a "natural disaster" or a disaster caused by human

negligence?

Kai Erikson traveled to Buffalo Creek to gain an understanding of the impact of the flood on the

community and the survivors. The law firm that represented survivors in a suit against Pittston

Coal funded Erikson's research. His focus was not primarily on the actions of owners,

managers, and employees of the coal company that had created the sludge ponds, but the

experiences of the survivors; in that regard it was somewhat different from later social autopsies

that focused on conditions and decisions that caused a problem in the first place. Erikson

emphasized the need to understand the impact of the flood in the specific cultural and social

context of Appalachia. Although it was one of the poorest regions in the United States, its

people were independent and self-reliant. They not only had strong social bonds with each other

but also held to a value system of autonomy and freedom, which they linked to the rugged

terrain of the Appalachian Mountains.

Erikson believed that these cultural traditions, somewhat paradoxically perhaps, made it

especially difficult for the people of Buffalo Creek to cope with the disaster. Their dispersed

living pattern was disrupted as they were forced into temporary housing, and they felt

increasingly dependent on government aid for day-to-day survival. They were reluctant to ask

for or receive government support and equally reluctant to leave their valley to start a new life

where there were more jobs and economic opportunities. Their ethos of self-reliance excluded

the very "help-seeking" behavior necessary for them to obtain relief. Erikson's book takes a

pessimistic view of the long-term impact of the flood on community and identity.

Diane Vaughan's (1996) The Challenger Launch Decision is closer in aim and methods to

Klinenberg's definition of a social autopsy. In January 1986, on a very cold day in Florida, NASA

launched the space shuttle Challenger. Across the world, hundreds of millions of television

viewers watched excitedly as the shuttle began its thunderous liftoff. Within a few seconds, the

craft exploded, killing its seven occupants.

The ensuing investigation quickly focused on damage to the O-rings, large gasket-like seals

between the rocket sections, as the technical cause of the explosion. It turned out that NASA

had experienced a series of problems with this component. The question that Vaughan (and

others) wanted to answer went beyond the technical details of exactly how and why the O-rings

had failed-the question was why the decision to launch was made when there were indications

that there might be problems with the O-rings, especially on such a very cold day. What

sequence of behaviors, inter-actions, and decisions had led up to the launch decision? Vaughan

was able to interview managers and engineers at NASA and at Thiokol, the contracting firm,

and had access to key documents.

Vaughan proposes that the work groups charged with the development of the shuttle began to

form a distinctive culture. In interaction, people create norms about how to handle ambiguous

and potentially risky situations. The need to manage risk and accomplish goals leads to

breaking the formal rules and developing informal rules that allow a higher level of risk-taking.

These informal emergent norms are part of the routinization of deviance.

We can see this process in many organizations hospitals and police departments, for example.

If everyone "worked to rule" following every formal rule to the letter, it would be literally

impossible to get anything done in most organizations; so formal rules are shaded and informal

rules are followed that enable workers to accomplish their job. Recruits are quickly socialized to

understand these informal rules, though they are not spelled out explicitly. According to

Vaughan, at NASA and its contracting firms a "culture of production" was pervasive, and many

people-managers perhaps more so than engineers were prepared to take some risks in

ambiguous situations in order to keep the mission on target. Furthermore, the organizational

culture emphasized hierarchy, and roles were defined in a way that discouraged subordinate

employees from speaking out. Thus, the culture and structure of the work groups created a

situation in which managers were willing to take risks that eventually led to the needless deaths

of seven people.

Vaughan's book is a detailed study, carefully based on interviews and documents. But it is not

the only direction that a social autopsy of the launch decision can take. For example, Edward

Tufte (1997) in an article of only a few pages to Vaughan's 575- argues that it was poor graphing

and data analysis that led to an underestimation of the risk. He provides a dramatic graph,

plotting O-ring damage as a function of temperature, which would suggest to even a complete

novice that a launch at a freezing temperature might be a very bad idea. Upon seeing this

graph, one of the authors' (Roberta) entire social research methods class not a single one of

whom was an engineer or rocket scientist- immediately said that it was clear that there was a

relationship between low temperatures and O-ring damage. Yet this simple graph plotting O-ring

damage as a function of temperature for all the launches had never been developed by the

space shuttle launch team. Tufte concludes that the problem was a failure of analysis and

cognition, not work group culture and the routinization of deviance.

Charles Perrow (1999), author of a massive study of high-tech accidents, suggests yet a third

explanation- closer to the ones in the popular press and public opinion at the time of the

disaster.

He argues that NASA allowed production pressures to override safety concerns and specifically

that upper management under pressure from the Reagan administration (eager for a

spectacular launch) suppressed safety concerns that were being voiced by engineers alarmed

by the very low temperatures. Rather than the result of a subtle process pervading work groups

throughout the organizations (NASA and its contractors), the risky launch decision was a blatant

mistake at the top, caused by external political pressure. Perrow warns that we should never

confuse culture and power-"We miss a great deal when we substitute culture for power" (1999:

380).

THE CHICAGO HEAT WAVE Klinenberg's social autopsy of the Chicago heat wave focuses

both on the distribution of deaths and on the social and organizational processes that allowed

the deaths to happen. Unfortunately, many readers do not read far enough into the book to

understand the logic of the entire design. At the start of the analysis, Klinenberg looks at who

died: the victims were largely poor elderly people who lived by themselves and were isolated

from their families and social support networks. They were disproportionately white and African

American, and relatively less likely to be Latino or Asian American. At this point some readers

stop reading, satisfied that their stereotypes are confirmed: Latinos and Asians value family and

take good care of their elderly.

Of course, this is not the end of Klinenberg's study; he then analyzes death rates and the

circumstances of death in two neighborhoods and provides a larger social context for the

individual deaths. In a poor neighborhood with high commercial and community activity, death

rates were relatively low, whereas in an even poorer neighborhood with high crime rates, vacant

lots, and thinned out social activities, death rates were particularly high. The first neighborhood,

South Lawndale, was predominantly Latino, but poor elderly whites living there also enjoyed the

protective factor of a vibrant community life. The second neighborhood, North Lawndale, was

predominantly African American, but Klinenberg emphasizes that it was not race per se, but the

social processes of crime and urban demolition that had diminished community life, which gave

it a higher death rate. People who lived there were afraid to leave their homes, there were few

retail areas or safe, attractive public spaces, and social bonds had become weaker.

Klinenberg then moves to a decisive institutional analysis, offering yet a larger frame of analysis-

-municipal and national social policy. He argues that in the period of neoliberalism since the

1980s social services have not only been cut back. They have also been reorganized on a

"smart consumer" model of social services in the neoliberal era which requires complex choices

in a quasi-market situation. Public services have been restructured to create a welter of choices

among many competing agencies. Finding and contacting the right agency or nonprofit provider

is difficult and complex; simple entitlements to one-size-fits-all public services have ended. This

reorganization has happened in many developed capitalist societies and major cities and in

social service institutions from Medicare drug plans to school systems. Smart young people

looking for magnet schools for their children may be able to navigate this new terrain of

public-sector consumer choice with ease and assurance. Poor ill old people, living in isolation in

frightening high-crime neighborhoods, were not able to get the help they needed. Someone who

is very ill, isolated from family, too old to be in contact with friends and social networks, and too

afraid to open the window is not likely to be making effective choices among a welter of

hard-to-contact government and nonprofit agencies who might be able to provide help.

At the end of the book, Klinenberg analyzes the way the media and the public relations

agencies of the city minimized the disaster and chose not to reveal the failure of social services.

Social autopsies offer a range of approaches and perspectives. They can highlight one or more

of the following:

• The structure and practices of organizations designed to manage risky technologies.

• Organizational culture that permits or even encourages taking

vent safety procedures.

"shortcuts" that circum-

Power and hierarchy and the role of organizational power holders in encouraging risky

behavior-for example, the pressures in NASA to launch the shuttle.

• Organizational routines and work group norms that permit high-risk behavior.

•Problems of resource allocation and the impact of diminished or misallocated resources in

precipitating an accident or preventing effective response to a natural disaster- for instance, the

poorly maintained state of levees and canals in New Orleans.

Disparities among the groups that are exposed to or affected by the accident or disaster, which

in turn reveal underlying inequalities_-for instance, the vulnerability of poor people, people of

color, and populations in poor countries to both natural disasters and to the siting of high risk

and hazardous technologies.

The response of governments and other institutions to the event, revealing their effectiveness

and concern, or lack thereof.

• The representation of these events in the media, and the consequent formation of public

opinion and public concern toward the event, the survivors, and the institutions involved.

Let's sum up the logic of social autopsies as a method of social research:

• Social autopsies are a special type of case study.

• Social autopsies are a way of studying organizations, power and decision making, and

inequalities in the impact of disasters, accidents, and intentional acts.

• Normally secretive or "opaque" practices and organizations are exposed to view in the social

autopsy.

Social autopsies are potentially contentious; the researcher may "step on the toes" of people

with power, raise questions about the functioning of institutions, or bring attention to disparities

and inequalities. In some cases, the researcher may be an advocate from the outset.

Social autopsies are highly contentious interpretations. They tend to antagonize those whose

action and/or inaction contributed to the disaster's onset, unfolding, and/or adversity of

aftermath. All steps in the analysis, from the discussion of disparities and vulnerable populations

to the identification of organizational and institutional contexts and analysis of media coverage

will cause controversy.

We strongly suggest to readers a little exploration of this on the Internet, using "Chernobyl" as

the search term. "Chernobyl" was a town in Ukraine (part of the Soviet Union at the time -1986),

in which a major accident took place in a nuclear power plant. A number of people died directly

as a result of the accident and a large cloud of radioactive material was released and floated

over northern and eastern Europe. Communities near the plant had to be evacuated, and a

fairly large zone remains closed off and uninhabited. Beyond this very terse description of the

incident, there is little that can be said about Chernobyl that finds consensus.

Explorers online will find vastly different estimates of the toll. At one extreme, websites suggest

that about 60 people died, most of them firefighters who heroically gave their lives battling the

fires in the days immediately following the accident. They died within a few days or weeks of the

accident.

These websites and many of them are associated with the nuclear power industry--claim that

there is little evidence of longer-term effects on larger populations. At the other pole of the

argu-ment, sites that call for continued aid and support for populations in Ukraine claim that

100,000s of people especially children--are suffering from adverse effects and illnesses such as

leukemia and thyroid cancers and that everyone in the region will continue to be at risk for a

long time to come.

What is the quality of the evidence? Does the truth lie somewhere in between? How can we

decide?

"Chernobyl" not only exemplifies disagreement over disaster and accident impact; there are also

very large differences in the discussion of its causes. Some websites emphasize operator error-

-human error at the lowest rungs of the organization. The operators decided to run a test, but

neglected to follow basic safety rules concerning the water supply. Other accounts emphasize

that the design of nuclear reactors and power plants in the Soviet Union was faulty and highly

risky. In this perspective, any Soviet nuclear facility was an accident waiting to happen; they

were all poorly designed and failed to include safeguards for overriding inevitable human errors

or at least containing their effects. Yet other observers point beyond both operators and plant

design, instead placing the blame on the Soviet ideology of uncritical use of technology and

disregard for consequences to human beings and the environment. In this larger perspective,

the unconscionable draining of water from the Aral Sea to irrigate central Asian cotton fields is

part of the same disastrous overall culture, state policy, and practice as the Chernobyl accident.

Look at these contending websites for yourself. Is it clear which site has the best

evidence-based data on effects? Which interpretation of the causes of the accident seems more

persuasive? In any case, exploring these questions online will give you a fascinating glimpse of

why social autopsies can become one of the most contentious and highly politicized methods of

social research.

Social autopsies of disasters and accidents are not conspiracy theories; the researcher is not

claiming that some secret or shadowy "they" has made the disaster happen on purpose. The

Daley administration in Chicago did not cause the heat wave in order to kill old people. Pittston

Coal did not want the sludge ponds to flood the valley. Nobody wanted the Challenger to

explode. Even in the case of Katrina, stating that there was gross negligence in the upkeep of

the levees as well as inadequate evacuation procedures is not the same as believing that the

Lower Ninth Ward was allowed to flood on purpose. (You may want to watch Spike Lee's When

the Levees Broke, in which some of the people interviewed voice the opinion that the flooding

was intentional, but it seems fairly clear that the filmmaker is keeping his distance from this

position- -watch it and decide for yourself.) Nevertheless, social autopsies could also be carried

out for events that were intended-torture, genocide, and ethnic displacement. But even here the

focus is not only on the perpetrators but also on the organizational and institutional context, and

perhaps the culture and geopolitical situation as well, in which it was allowed to happen or that

enabled it to continue.

Social autopsies "do not reduce complex events to a single guilty actor or causal agent," as

Klinenberg notes (2002: 32). He even states that they cannot be reduced to a single social

force; they are usually the result of multiple intersecting causal processes- -a sort of "perfect

storm" in the social structure. For example, one could argue that the Challenger launch decision

involved both work group culture and pressure from the top and that "Chernobyl" included

operator error in the context of poorly designed systems and uncritical state policies of

technology development. "Katrina" was based on converging forces including poverty, racism,

incompetence at the apex of FEMA reflecting the effects of cronyism and fund-raising in the

federal government, deferred maintenance on public works, and failed emergency planning at

the state and local level.

The highly contentious nature of social autopsies suggests that the researcher must take

special precautions to limit bias once the research is underway and must be exceptionally

sensitive to ensuring that falsifiability is possible in the design. If the researcher was hired by

plaintiffs in a lawsuit or is working with an advocacy group, these concerns are particularly

pressing.

COMPONENTS OF A SOCIAL AUTOPSY

A social autopsy therefore is an overall research design, not a specific technique. It includes

quantitative as well as qualitative analysis, interviews and observation as well as unobtrusive

measures.

Here are components that might be included in a full-scale, multilevel social autopsy and a few

suggestions, or directives, for specific data collection methods.

Who Was Affected?

The method involves examining risks that disproportionately affect vulnerable populations, the

disparities and inequalities that contribute to heightened risk of disaster for some but not others,

the spatial distribution of risk and vulnerability, and the vast range of survivors' experiences.

VULNERABILITY AND RISK The focus of the study could be vulnerable groups and the

disparities in the impact of a disaster or adverse event. Here official data, available from

government sources, is invaluable. Klinenberg looks at the data collected by the Cook County

Public Administrator's Office as well as notes kept by police and left at the Cook County

Morgue.

Data from government agencies such as the EPA may be useful in analyzing how siting

decisions were made for dangerous processes, toxic waste sites, and other environmental

hazards. These are rarely placed in communities of wealthy and powerful people. It is not

surprising to find that a very poorly managed and monitored Union Carbide plant was located in

a community of poor people in a city in India-Bhopal. In his analysis of the social and spatial

impact of Katrina, John Logan (2006) made use of data collected by government agencies.

DISTRIBUTION OF THE IMPACT Who were the Individuals Affected? At this level, several

distinct sources of data and methods of analysis can be used. First of all, we might want to look

at the victims of the disaster. Who were they in terms of demographic variables such as age,

income level, poverty status, occupation, employment status (unemployed, underemployed,

employed), race or ethnicity, gender, and length of residence in the area? In this analysis,

characteristics such as income level, occupation, poverty status, and employment status are

indicators of the larger, more complex variable social class.

The immediate next step is to ask whether the victims of the disaster or accident were different

from other people. The analysis quickly moves to a discussion of disparities, and these in turn

point to underlying structural inequalities. In order to carry out this part of the analysis, it is not

enough to examine the data on victims (and survivors) but also to compare them to those who

were relatively unaffected. For example, Klinenberg found that people who died in the Chicago

heat wave were older, poorer, more isolated, and less likely to be Latino or Asian than other

Chicagoans. If carefully collected statistics on victims are available in public data sources, a

sophisticated quantitative analysis may be possible. A multiple regression model in which the

relationships among variables are considered may help us to understand the relative role of

age, living alone, race-ethnicity, and poverty as predictors of the outcome of death or survival, to

use the example of the heat wave. Other predictor variables would be selected for other types of

disasters or accidents, as appropriate to the specific situation. Since these characteristics may

be interrelated, the multiple regression model allows the effects of each one to be considered

separately while controlling for the others.

The quantitative analysis is complicated by the fact that in most disasters and accidents the risk

of dying or other adverse effects is quite low in absolute terms. Although the death toll in New

Orleans is quite horrifying and completely unacceptable, it is nevertheless the case that the vast

majority of New Orleans residents survived. Therefore in the quantitative comparison, the two

groups being compared- the victims and the nonvictims are very different in size and this

difference may create technical problems in the statistical analysis. Another way of saying this is

that because the overwhelming majority of Chicagoans did not die in the heat wave, it is in fact

difficult to compare the death rates of (for example Latinos and non-Latinos or the wealthy and

the poor all of them had a very low rate of absolute risk. Although the comparison of rates is the

ideal standard of analysis, in these cases, we may have to resort to comparing the demographic

profile of victims to the profile of nonvictims or of the population as a whole. We also have to be

careful in using expressions such as "risk factor," keeping in mind the very low absolute risk. For

instance, being white or African American was associated with higher death rates in the heat

wave, but we certainly do not want to jump to the conclusion that all whites and African

Americans were "at risk" in this disaster in any meaningful sense of the term

"risk factor."

SPATIAL ANALYSIS OF THE DISTRIBUTION OF THE IMPACT At the next substep, a spatial

analysis can be added to the model, still in the context of asking who was vulnerable. We

consider not only the characteristics of vulnerable populations compared to the relatively secure

or immune but also the overall spatial patterning and the effects of neighborhood or community

contexts. To once again cite Klinenberg's example, poor, elderly live-alone white people in the

predominantly Latino community of South Lawndale had a relatively lower death rate than

people with the same individual characteristics living in other neighborhoods. Spatial context

matters. Once we have determined that spatial context makes a difference in vulnerability, we

need to conceptualize the way it makes a difference. The data themselves may have little to

reveal about this -we need to turn to sociological theories and qualitative research to understand

why neighborhood contexts are important-exactly how they "work" to make some areas more at

risk than others.

We can map the impact of the disaster, identifying the spatial areas where people are most

likely to experience the impact. If we do that for Katrina's effects in New Orleans we very quickly

see that the most adverse consequences death, injury, long-term homelessness--were

experienced in poorer and predominantly African American areas of the city, especially the

Lower Ninth Ward.

In mapping the spatial dimensions of "adverse event" impact, we need to think about the

boundaries of our mapping. For example, if we study the impact of Katrina on the Gulf Coast as

a whole, we would find that many white people lost their homes. The racial disparities are most

evident within New Orleans and its immediate surrounding communities. How big we make our

map makes a difference in how we reach our conclusions about racial disparities in Katrina's

effects (Logan 2006).

This type of spatial analysis works not only for natural disasters and accidents but also for

intentional acts. Hannah Arendt (1992), in a controversial book on Nazi genocide, Eichmann in

Jerusalem, suggested that three countries in Europe Denmark, Italy, and Bulgaria- -had

considerably higher rates of survival among Jews than other countries during the Holocaust.

What accounted for the difference? Why were people in these three nations more likely to resist

the orders of the Nazis to round up and deport lews? Arendt suggests the reasons were quite

different and deeply rooted in national cultures as well as the specific circumstances of German

occupation.

Danes felt a sense of civic engagement and collective responsibility; although it had been

impossible for them to defend their small, flat country against the German invasion. They were

seen as "fellow Aryans" by the occupiers and not as closely and brutally monitored as "inferior

peoples."

As is well known, the Danish resistance was able to get most Jews in the country- citizens and

refugees from other countries alike onto fishing boats and smuggle them to Sweden, a neutral

country prepared to accept them.

In Italy, resistance to genocide was more likely to be based on individual, anti-statist values of

common humanity and concern; the Fascist government cooperated with the Nazis, but many

individuals risked their lives to help Jewish friends (and strangers as well) to escape. In

Bulgaria, the attitude at many levels of the society was one of "Nobody can come in here and

tell us what to do." When the Nazi occupiers ordered Jews to be rounded up throughout the

country and brought to the capital to be deported from there to death camps, the Bulgarian head

of the national security forces ordered Jews to be shipped out to remote villages in the

countryside so they could not be easily found and deported. In this example, the initial mapping

of the crime the identification of countries with high or low rates of Jewish survival is only the

first step toward the analysis of cultural, institutional, organizational, and military characteristics

that can be used to explain the distribution of the outcome.

SURVIVOR EXPERIENCES VARY GREATLY, BUT PATTERNS EXIST Still focusing primarily

on characteristics of survivors and the distribution of the event in terms of rates for demographic

categories and spatial areas, many autopsies include insights into the feelings of the survivors,

the impact on communities, and the human impact of specific public policies designed to

manage the disaster. In this case, interviews with survivors are obviously a valuable method,

and so is participant-observation in the communities that were affected by the disaster, as well

as among survivors who may be scattered far away from the disaster site. Because many

disasters disrupt communities and force people to seek new homes and new jobs, the

researcher might have to range far from the point of impact. For example, Erikson interviewed

survivors of the Buffalo Creek disaster not only to understand what had happened to them in the

flood but also to understand their feelings about temporary housing and relocation. Following

the tsunami, residents of fishing villages in southern India and Sri Lanka were relocated away

from the coast and faced great hardship in making a living.

There are difficult issues in both sampling and interpreting survivor interviews. Sampling is

difficult because in many cases there is no clear "sampling frame" that allows the researcher to

enumerate all possible survivors, though this problem varies with the nature of the incident. In

industrial accidents, for example, we can at least think about a list of all employees of a

factory-though it might be very hard to obtain that list from the company. In natural disasters or

ethnic displacement, the initial lists are hard to specify.

Homelessness and displacement are major consequences of disasters, large-scale accidents,

and policies of"ethnic cleansing" so it becomes difficult to find survivors. Those who can be

found--for example, those who stayed in their homes during a disaster or did not flee from ethnic

violence-may be quite different from the displaced, creating problems of generalizability for the

researcher.

Survivor accounts require careful interpretation. Survivors may have learned to tell a story that

will obtain support or win a court case. Survivors may therefore exaggerate the impact of the

incident or overestimate its long-term effects. Survivors who want to talk about their experiences

may be different in many ways from those who refuse to be interviewed or take measures not to

be identifiable as a survivor at all. Survivor interviews may be very similar to "testimonial

literature"_-the collection and editing of the accounts of survivors of a large range of terrible

events. This literature has been especially strongly developed in Latin America, where political

activists and intellectuals have dedicated themselves to collecting and disseminating the stories

of survivors of torture, mas-sacres, and "disappearances" under military regimes (in Guatemala,

El Salvador, Uruguay, Argentina, and Chile, for instance), as well as survivors of natural

disasters such as the earthquake that claimed between 10,000 and 20,000 lives in Mexico City

in 1985 (Poniatowska 1995).

BACKGROUND RESEARCH The first step for the researchers is to immerse themselves in

available information. Often the social autopsy requires an understanding of complex scientific,

medical, or technological knowledge. The researcher can be begin with textbooks and research

articles concerning the topic or may be lucky enough to find a review essay or published review

of the literature. For example, Vaughan had to learn a large amount of information about shuttle

and rocket technology in order to understand what respondents said in interviews about safety

procedures and risk management. Klinenberg had to understand how people die in a heat wave

before he could conduct meaningful interviews with medical examiners. In order to be a credible

and effective interviewer, the researcher has to acquire much of this information and knowledge

before conducting the interviews.

INTERVIEWS WITH INDIVIDUALS IN ORGANIZATIONS AND INSTITUTIONS The researcher

might next turn to interviews with members of the organization or institution. The warnings and

strictures that apply to the role of guides in qualitative research apply to the respondents in this

situation as well. People who agree to be interviewed may be angry and marginalized within the

organization; they are eager to present their own behavior in a positive light (as a

whistle-blower, for instance), while accusing others in the organization of having acted

irresponsibly.

Alternatively, they may be "stranger handlers" in a role designed to minimize negative

conclusions about the organization and assist in a cover-up or "damage control."

Sampling respondents who were or continue to be in the organizations involved in the incident

presents complicated problems. Most organizations do not provide comprehensive public lists of

employees, so creating anything resembling a sampling frame and a random sampling

procedure is virtually impossible. If a researcher resorts to snowball sampling (asking each

respondent to identify other individuals who might agree to be interviewed) the practice is likely

to amplify any initial bias in the process with a repetition and confirmation of what the initial

respondent in the network claims is accurate. Sampling individuals who left the organization

after the incident is even more difficult and is almost certain to have to rely on snowball

sampling techniques tapping into networks of probably like-minded people.

In these complicated, complex, and organizationally contextualized case studies, interview

guides need to be developed that are worded nonjudgmentally. The interviewer must know

enough about the situation, including underlying technology, to be ready to pose focused

follow-up questions in an interactive manner; it is not enough to prepare a well-designed, initial

structured interview guide.

A naive front can be useful in the interview. Even though the interviewer must know something

about the topic of the interviewers, in some cases, it may be a good tactic to ask the respondent

to "begin at the beginning" and to explain the basic processes and terminology. The interviewer

can compare the way the respondents talk about the process with the way it is described in

textbooks or the research literature, and the comparison may be revealing, possibly containing

clues about shortcuts or crucial, risky decision-points.

It is important to remember that IRB approval almost always carries with it measures to ensure

confidentiality and anonymity. These safeguards may induce respondents to be frank in their

discussion of organizational problems.

Social autopsy includes the response of media and other organizations that specialize in the

representation of reality-public relations firms, advocacy groups, and so on. To gauge the way

the image of the disaster is formed and manipulated, the methods of content analysis and

discourse analysis are useful. For example, why was the phrase "refugees" used repeatedly

about people forced to leave their homes during Katrina? Were black people removing goods

from stores labeled "looters," while white people doing the same thing were described as

looking for items they needed to survive? Are accidents in non-Western countries given less

attention by the media than those in Western or developed nations? For example, the 4000

deaths and over 200,000 serious injuries resulting from a chemical accident in the Union

Carbide plant in Bhopal, India, seem to have drawn less attention than smaller accidents in

North America and Europe, such as the Three Mile Island nuclear accident in Pennsylvania in

which no one was killed or injured.

As terrible as Katrina was, in terms of deaths and displacement it was much smaller than an

earthquake in Pakistan that took place a few weeks later, in October 2005, and caused possibly

as many as 73,000 deaths and homelessness among over a million people. A student in one of

Roberta's classes a few months after this earthquake said he had never heard of it and excused

his ignorance by claiming he was "a victim of the media." Although the excuse is absurd, it is

true that the media coverage of the quake in the United States was terse and disappeared

quickly compared to Katrina coverage.

A few basic categories for content analysis could include:

•Typification of victims and victim behavior. For example, are the victims portrayed as innocent

or careless, irresponsible, incompetent, and even complicit?

• Typification of the rescue effort, first responders, and long-term aid . Who is interviewed and

the types of authorities and experts interviewed?

* Point of view. Does the coverage present different points of view on what happened? Does it

assign blame or suggest that what happened was "an act of God" or not preventable?

•Duration and positioning of coverage. How long is the story? Does it appear on the front page

or at the top of the news on TV? When does it fade from a leading spot and where does it go?

After an initial content analysis based on these simple categories, a more refined coding can be

developed and applied.

The social autopsy in its most developed form-in Klinenberg's book, for example- covers all of

these elements: the distribution of its impact, which in the case of larger disasters is rarely

random and generally linked to social disparities; the analysis of cultural, organizational, and

institutional processes that have set conditions for the adverse event or precipitated it; the

impact on survivors and people indirectly affected by it; the institutional response to it; and the

images and representations used to portray it. So it usually requires a multimethod approach

including interviews with survivors and with decision makers, archival records and government

data, quantitative and qualitative analysis, and possibly participant-observation.

Throughout this book, we encourage readers to connect research design to theory. Theoretical

foundations lead to a deeper analysis and help us to understand underlying processes, not just

the problems of the day. Social autopsies are very easy to link to theories here are a few

examples.

Klinenberg uses ideas from both conflict theories and Emile Durkheim in his analysis of the

Chicago heat wave. The two initial parts of the analysis seem strongly influenced by Durkheim's

work. As Durkheim did in Suicide, Klinenberg looks at the distribution of deaths-who was

vulnerable? The concentration of mortality among elderly people living alone also points to the

significance of Durkheim's view of society. The great French theorist believed that sometimes

society is dense, vibrant, and cohesive, but sometimes and increasingly so in the modern era- it

is "thinned out" and atomistic, a condition that puts elderly people at risk. This Durkheimian

analysis is developed more explicitly and intensely in Klinenberg's contrast of the two

communities.

The community with a dense, vital social life had better survival rates among at-risk categories

of people than the community with little public sociability and a high degree of fear and isolation.

Klinenberg's reference to Marcel Mauss (a student of Durkheim) and the phrase "total social

fact" signal his debt to Durkheim. But in the second part of the book, Klinenberg draws on

conflict theory the Marxian subtext seems unmistakable- to point out how the withdrawal and

reorganization of public social services contributed to the disaster. Heat wave deaths have to be

understood in light of the major transition from

'embedded liberalism" (characterized by

government provision of social services and many entitlements) to "neoliberalism" with

governments cutting services or reorganizing them to limit access, charging user fees, and

generally making it more difficult to get help. The media failed to analyze these problems in

social services in their heat wave coverage. Klinenberg's critical perspective on the media links

him to conflic

theories.

Durkheim and Marx are almost always good choices for sociological inspiration, but there are

many other theorists whose concepts can deepen a social autopsy. For example, Michel

Foucault was interested in regimes of surveillance and regulation; he used the quarantine as a

key example of institutional efforts to define and contain a problem. Max Weber and Pierre

Bourdieu, along with Marxists, are theorists of inequalities and stratification systems and their

work can be used to understand the disparities that become evident in many disasters.

Bourdieu's

"symbolic violence" and "misrecognition" are concepts that draw our attention to the

misrepresentation of events in the media. Weber as a theorist of bureaucracy, organization, and

instrumental reason contributes to understanding organizational decision making and in the

analysis of failures to follow procedures, distortions produced by hierarchical relationships and

the very workings of instrumental reason itself.

C. Wright Mills' (1940) concept of " vocabularies of motives"-the excuses and rationaliza. tons

that are used to explain behavior--as well as the symbolic interactionists' labeling theory help us

to see similarities in the way that many accidents are managed and represented--for instance,

the tendency to blame them on the lowest level of the hierarchy, such as the operators at

Chernobyl or the lowest-ranking soldiers at Abu Ghraib. The whole point of the social autopsy is

to understand mismanaged disasters, accidents, and intentional acts as produced by larger

forces than individual error or misconduct. Erving Goffman's analysis of teams in the

presentation of self and impression management provides insight into organizational

presentations and cover-ups often associated with accidents. Feminist theory encourages the

researcher to give voice to survivors who might otherwise remain silent or silenced and to turn

away from official histories of adverse events.

Finally, we can see a similarity between the multistage social autopsy and the

progressive-regressive method of understanding human action that was suggested by

existentialist philosopher Jean-Paul Sartre. Sartre began with a key event or situation in the life

of the individuals about whom he wrote and then traced the antecedents of this event- -the

regressive part of the study--and its consequences--the progressive part. The antecedents took

Sartre in the direction of preconditions and causes, ranging from childhood experiences to

situations later in life and the immediate precipitating circumstances; the consequences showed

the impact of the event on the life of the individual and the way it shaped his or her later actions

and choices. Similarly in the social autopsy, we look at both the impact on survivors and

communities and the causes and preconditions of the accident, disaster, or action.

So we can see--in this quick sketch--that autopsies can have deep roots in theory, as well as

applications in the "real world" of policy and social change which we will briefly sketch as

well.Readers of this chapter willalmost certainly be asking themselves how social autopsies

differ from investigative reporting. Indeed the two kinds of writing are closely related!

Here are a few of the differences:

• The social autopsy researcher generally has more time- a time frame of years to conduct

research, whereas journalists tend to be under time pressure of days or at best weeks to

complete their investigation.

• The social autopsy researcher feels freer to look at larger social and cultural contexts

Although good investigative reporters are not hesitant to "go after" higher levels of officials and

the upper reaches of organizational hierarchies, they may feel pressure to avoid social science

concepts such as "routinization of deviance." They may feel compelled to stick to simpler

explanatory frames focused on the misconduct of individuals. This is beginning to change,

however, and many journalists are comfortable using terms such as "organizational culture."

• Few reporters are eager to present readers with a complex statistical analysis.

. The social researcher must conduct the research with IRB approval, and this requirement

almost certainly means the anonymity and confidentiality of sources. Names not only aren't

revealed, they are also kept confidential. Journalists face the diametrically opposite problem;

they are encouraged never to publish blind interviews (i.e., ones in which names are withheld)

unless the interviewee is at great risk. So both professions face challenges in the management

of confidentiality and with it, the public's ability to verify the evidence, but the challenges lead to

opposite strategies for managing information about the respondents' identities.

Another related type of writing is the testimonial, which we have already mentioned. For an

example of a powerful and moving compilation of testimonials, see Elena Poniatowska's work

(1995) on the 1985 earthquake in Mexico City.

APPLICATIONS

There are many possible applications for social autopsies beyond the impartial search for

knowledge and a better understanding of what causes adverse events. Here are a few

examples of applications related to activism, advocacy research, expert witness, and the

development public policy:

Social autopsies may be part of a larger environmental justice research goal. The term

environmental justice means that there is a growing sense that environmental hazards are not

randomly or evenly distributed in most societies. Economic and racial stratification are related to

heightened risk. Toxic waste sites, pollution from industrial processes, truck traffic with

accompanying air pollution, hazardous mining operations, and landfills are more likely to be

situated in or near poor communities or communities of racially marginalized people (people of

color in Europe and the Western Hemisphere, or ethnic minorities in many parts of the world).

Garbage and toxic waste are exported to poorer nations. A social autopsy can pinpoint specific

harm caused to a community and trace the decision making that produced the outcome; for

example, identifying the sources of air pollution that affect communities and disproportionately

often low-income and minority communities (Ash et al. 2009).

Social autopsies can be a component of advocacy research, specifically focused on changing a

situation, pressuring policy makers, preparing testimony in lawsuits or legislative hearings.

Social autopsies can be used to improve organizational practices, both routine practices and

emergency practices, as well as technology.

Social autopsies can be used to change policies regarding access to social services.

Klinenberg's study made it clear that police had not been trained to identify and help elderly

people at risk and that the expectation that they would perform this function had to be backed

up with better preparation or assigned to other agencies.

Social autopsies can be used to call public attention to social disparities (and the underlying

inequalities) and contribute to social policies to reduce disparities.

The social autopsy can be a powerful tool for the sociological study of natural and human-made

disasters (with most "disasters" being a combination of both natural elements and human

frailty). It also represents a vehicle by which research can "go public" in a very controversial,

social change-producing way. In this chapter, we have discussed some of the most widely read

social autopsies, and from them we have drawn key lessons on how to design and implement a

social autopsy of your own.