Sociocultural Data to Accomplish Department of Defense Missions: Toward a Unified Social Framework: Workshop Summary (2011)

In contrast to its role in conflict environments, the U.S. military is also often assigned to provide support and assistance to cooperative governments. This may happen in a postconflict situation, for example, or after a civil war or a natural disaster. The missions may include such things as peacekeeping, treaty enforcement, providing security for refugee camps, and evacuating noncombatants. In such situations, it is not uncommon for the local population, or portions of it, to be uncooperative or not receptive to the American personnel, even if the national government has requested their presence or at least established a cooperative relationship with the United States.

To accomplish their mission, military personnel must be able to convey security and safety concerns in a way that is culturally sensitive and mutually acceptable to everyone. Thus, understanding the contextual, interactive, and communication dynamics of these cooperative relationships is vitally important to success. It is often the case that military personnel put into these situations are not trained or oriented to take on these perspectives; yet, if they are provided with the appropriate knowledge and tools, they can adapt to these potentially dangerous situations and accomplish their missions.

As Andrew Imada, the panel’s moderator, noted in his opening comments, “The guiding question for this session was, What sociocultural knowledge will enable Department of Defense personnel to work with cooperative partners to make local populations feel safe?” Three presenters addressed that question: (1) Robert Rubinstein, a political and medi-

cal anthropologist whose work focuses on cultural aspects of conflict and dispute resolution, including negotiation, mediation, and consensus building; (2) Alan Fiske, a psychological anthropologist who conducts research on social relations, including the models in which people characterize their social interactions; and (3) Donal Carbaugh, an expert on international and intercultural communication.

In the first presentation, Robert Rubinstein of The Maxwell School of Syracuse University came to grips directly with one of the key themes of the workshop: How much can sociocultural models do, and what are their limitations? In particular, he said, he is fascinated by the question of what data and principles exist that might be used to create a general predictive model of cooperative social behavior.

For a number of years, Rubinstein said, he has worked with United Nations’ peacekeeping forces, so he tends to think about cooperation in that context (for more about his work, see Rubinstein, 2008). In particular, he thinks about cooperation “in terms of the question of interoperability and the way culture plays into interoperability.” He said that interoperability can be thought of as “the ability of people and organizations to work together smoothly.” Ultimately, he is interested in building models of the sociocultural aspects of interoperability.

Sociocultural models have varying levels of complexity, Rubinstein explained. The least complex models are what might be called “traveler’s advice.” They include the sort of information given to someone who is traveling abroad, mainly advice about the things that one should not say or do so as to avoid a social blunder. “It is a very simple model. It works. People manage to get through a different culture and not make any mistakes.” Otherwise, however, it is not a particularly useful model, as it offers no insights in how the people in that culture generate new behaviors.

A somewhat more complex class of models can be described as “stereotyping.” As an example, Rubinstein presented a list of generalizations concerning cultural differences between the military and nongovernmental and other civilian organizations. For example: military organizations are hierarchical, whereas civilian organizations are decentralized; military organizations are culturally insensitive, whereas civilian organizations are culturally sensitive; and military organizations appreciate precise tasks, whereas civilian organizations thrive on ambiguity. Such models have a variety of weaknesses, Rubinstein said, including the fact that they assume that all military organizations and all civilian organizations are homogeneous and that each has a stable and static culture. “They give us a very false sense of understanding those groups.”

A third class of models looks at cultural styles and dimensions. These are generally empirically based, so they are more sophisticated than the stereotyping models. They classify cultures according to various empirically determined traits. For example, some cultures are said to be direct, valuing self-expression and verbal fluency, and others are described as indirect, with their members tending to use ambiguous language and to avoid saying “no” in order to keep things harmonious. In these kinds of models, some groups are characterized as relying more on linear thinking, putting their faith in logic and rationality and regularly seeking objective truths, and others are nonlinear, with indirect reasoning processes and no attempt to find external truths.

“These models are very useful for some purposes,” Rubinstein said. “They have had an illustrious career in cross-culture business consulting. They have had an illustrious life in intercultural communications. They are useful for what they do.”

One problem with these models, he said, is that they can create a false confidence so that people act as though knowing how a group typically responds allows them to predict how an individual person will respond—but they do not. “When we proceed from a dimensional model, it is all too easy to think about how a person is going to respond [in terms of how] the group would respond.” Although these models do a very good job of characterizing general group traits, they do not actually provide any information about individuals, nor about how behaviors are generated.

A more sophisticated approach can be found in what Rubinstein referred to as “cultural models.” He illustrated this approach with a diagram that includes such considerations as language, symbols, rituals, and behavioral models (see Figure 4-1). “If you take a look at this diagram,” he said, “you will see that the way people interact, how they talk, what their styles are—they don’t just do it. They do it because they have a particular reason to do it, because it is connected in some way.”

One of the things to keep in mind about cultural models, Rubinstein said, is that each community will have multiple models that are all “grammatical”—that is, that are accepted by the community as ways in which one decides what is suitable behavior. The particular circumstances in which a person finds himself or herself will affect which model the person chooses to use.

As an example, Rubinstein referred to the “disproportionate revenge among the Khmer Rouge in Cambodia.” The anthropologist Alex Hinton, author of Why Did They Kill? Cambodia in the Shadow of Genocide (2005), identified two contrasting models in the Cambodian culture, one for disproportionate revenge and the other for compassion and forgiveness.

His second example comes from Mary Catherine Bateson’s studies of Iranian theater before the Iranian revolution (e.g., Bateson, 1997). She

FIGURE 4-1 Culture: Levels of analysis and observation.

NOTE: Cultural models give a sense of the range of possible “grammatical” actions and responses. They do not predict which model will be actualized.

SOURCE: Rubinstein (2003, p. 33). Reprinted with permission from John Wiley and Sons, http://interscience.wiley.com [January 2011].

identifies two models of what it means to be a good man, Rubinstein said. In one model a good man is an entrepreneur, and in the other he is deeply religious and principled. Examining these two models in the Iranian cinema, she found a shift in the popularity of the models over the years.

As the two examples illustrate, a culture can have contrasting models. Still, Rubinstein said, “they are what legitimates or gives moral force to actions. They are what is grammatical. Just as with language, they provide

lots of ways to say things that are understandable, that are acceptable.” But, he added, they do not predict actions, any more than the grammar of a language predicts how someone will choose to say that he would like to go get lunch.

This points the way toward even deeper, more complex models of culture—models that take into account the context in which various actions are taken. Various factors influence how people approach a situation, and one of the most important is a person’s motivation. “The same person in the same circumstance may react in different ways, depending on the motivational state,” Rubinstein said.

As an example, he mentioned that humanitarian workers often say that they don’t want to have anything to do with the military, because having soldiers around can influence how the people who need their help will view them. But when some humanitarian workers find themselves in a difficult situation, they may very well ask the military for help. In the same interaction they respond differently, depending on their context and motivational state. By taking context and motivation into account in this way, a cultural model becomes “a dynamic system of meaning,” Rubinstein said.

So is it possible to create a sociocultural model of cooperative behavior that can be used to make predictions? Rubinstein discussed briefly some of the features that might go into a model of interoperability. First, there are different levels of interoperability, ranging from operating in a completely unified way to operating completely independently. Second, there are a variety of culture issues that come into play, such as communication style, operating style, and level of trust. The model becomes exceptionally complex.

But no matter how complex a model is, it is never complete. “Modeling is for understanding selective features of the world,” Rubinstein said. “It is a heuristic, and heuristics all carry biases.” Thus the match between the model and the real world is always by necessity a partial match, not a complete one. In particular, attempting to move from a model that captures group elements of culture to one that attempts to predict individual responses requires data that are simply not accessible, because individual behavior is an emergent phenomenon, one that depends closely on the interaction among individuals, and that will vary according to each individual’s motivation and emotional state.

“What I think,” Rubinstein ended by saying, “is that trying to find a general predictive model of the social and cultural elements of cooperative behavior is really—let’s just be provocative—a fool’s errand, and not something that is very helpful to try, because it generates technique without validation against the real world. Things are much too dynamic, and moving from group elements of culture to trying to map individual responses doesn’t make any sense.”

In the next presentation, Alan Fiske of the University of California, Los Angeles, described four models of social relations that, when understood properly, can help improve the effectiveness of cooperative relationships. “The argument I am going to make,” he said, is that “social coordination in virtually all domains of social life in all cultures is organized out of four basic structures, or four relational models.”

At first glance, he acknowledged, it might seem hard to believe that in every culture people coordinate everything they do using just these four relational models in various combinations and implemented in culturally diverse ways, but there is a great deal of evidence that this is the case. Relational models theory is based on a synthesis of classical theories along with ethnographic fieldwork, ethnological comparisons across many cultures, and analyses of research in social and cognitive psychology. It has been tested in many studies with many methods and has been applied to understand a variety of domains of sociality in many cultures. At last count, he said, more than 200 articles, chapters, and books have been published by more than 100 scholars to test, apply, or extend relational models theory. It is, in short, a well-established area of anthropology.¹

The four models that relational models theory uses to understand cooperative behavior are (1) communal sharing, (2) authority ranking, (3) equality matching, and (4) market pricing, Fiske said, and he gave a brief description of each model.

In a communal sharing relationship, people feel connected to each other, they feel that they belong together, and they identify with each other. “People feel that they have something essential in common,” Fiske said, “and that differentiates them from other people who don’t have that.” Furthermore, everyone is seen as being the same. “In any culture, when people are eating communally, sharing food and drink, when they are being generous and kind to each other, they have a communal sharing relationship.”

There are many different types of communal sharing relationships, and individuals may be part of several of them at the same time. “You may be deeply in love with somebody,” Fiske said. “You may have a communal sharing relationship with other people in your discipline or other people in your work group.” These relationships can differ in intensity, from the highly intense connection between lovers or among members of

a platoon in combat to the more diffuse sense of commonality and sharing among members of an ethnic group, a nation, or even all of humankind.

In mathematical terms, Fiske said, the communal sharing relationship can be thought of as an equivalence relation. “It is a nominal scale of measurement in which the social world is divided up into categories of people, and everybody within each group is treated as equivalent and different from people in other groups.”

The second type of relational model is a hierarchical one, in which people’s positions in an authority ranking are clear. The authority is seen as legitimate, with those in higher positions expected to provide guidance and leadership to subordinates, who in turn are expected to be deferential and respectful. “I want to emphasize,” Fiske said, “that this is a relationship in which those above are perceived as legitimately, naturally, necessarily entitled to deference and respect from subordinates, but the leaders are also expected to lead, to guide, to stand up for, to speak up for, to protect, and in general to take care of their subordinates.” In particular, this is not a relationship in which control relies on force or coercive power. “I am talking about the kind of relationship that would be described in East Asia as filial piety, or the kind of relationship that you would have in the military where you believe that your superiors are entitled to obedience and deference, and where in turn the leaders are expected to look out for their subordinates.”

In mathematical terms, the relationship is an ordinal scale, in which relative positions are well defined but the differences among the rankings are undefined. “So you know that a general is higher than a lieutenant, and you know that a lieutenant is higher than a private,” Fiske explained, “but you can’t exactly say what the distances are, and you can’t exactly compare the difference between a lieutenant and a private or the difference between a lieutenant and a general. The order is clear, but the distances are not.”

The third type of relational model, equality matching, is one in which people are careful to balance things. “If you invite me to dinner, I owe you one,” Fiske said. “If I haven’t invited you back, but you have invited me again, I owe you two. Now I invite you back, and I still owe you one. So you can add and subtract. We know what would balance the relationship.” The democratic idea of one person, one vote, is an example of an equality matching relationship, as are most games and sports, in which the rules specify equal numbers of players, equal numbers of pieces, taking turns, each side defending one half of the field, and so on. “An equality matching relationship is one in which people attend to the additive differences with reference to even balance.”

Like every relational model, equality matching can take on both positive and negative forms, Fiske said. “It can organize violence as well as

cooperative relations. An eye for an eye, a tooth for a tooth—that kind of vengeance, which has been widespread in history, is violence organized in an equality matching way.”

Mathematically, equality matching can be thought of in terms of ordered Abelian groups, which have addition and subtraction that obey associative and commutative principles. As a measurement type, it corresponds with interval scaling.

The fourth type of relational model, market pricing, is any kind of interaction that is organized with reference to ratios, rates, or proportions. The most familiar types of market pricing relationships involve money, but money does not have to be involved per se. “The money can be absolutely anything,” Fiske said. “It is an arbitrary symbol.” Any time that cost-benefit ratios are calculated or efficiency is analyzed, that is a market pricing relationship. Another example is proportional justice—the idea that the punishment should be proportional to the crime. “If you get a parking ticket and the judge sentences you to 40 years of hard labor, what is the matter with that?” Fiske asked. The problem is that the punishment is disproportionate to the crime. Similarly, when people expect their rewards to be proportionate to their effort and their contribution, that is market pricing.

In mathematical terms, a market pricing relationship can be expressed in terms of an Archimedean ordered field, a set of entities that can be multiplied and divided as well as added and subtracted, in which there is a zero point, and every entity can be expressed as a multiple of every other nonzero entity. Multiplication and division are meaningful in this coordination structure, which is homologous to a ratio scale.

These four relational models are the ones that people use to organize virtually everything in virtually every culture, Fiske said. Furthermore, they are highly moralized models. “People think that these are right, that violations of them are transgressions, that morality itself is based on these. People have very powerful emotions about these things.”

So what does that mean for social cooperation? It implies, Fiske said, that to cooperate with people in any given group, one needs to answer two basic questions: What relational model are they using, and how do they implement it?

Consider, for example, a humanitarian mission to distribute food. “First, you need to find out what model other people think would be appropriate for food distribution,” Fiske said. Is it an authority ranking model? A communal sharing model? And, once it is known which relational model the people use, it is important to figure out exactly how that model is put to work. “Let’s say they use authority ranking. You need to know who is in what position in the hierarchy. Who has the authority to make decisions? Then how are decisions announced and transmitted?

Hierarchies … can be implemented in innumerably different ways, and you can’t function just knowing that authority ranking is the model unless you know who fits in where and how they use authority ranking.”

“Or suppose they use communal sharing to make decisions about food sharing. How do they make the decision? Is it a Quaker meeting where everybody pitches in? How do they decide whether a consensus has been reached? And so forth.”

Furthermore, once they have decided how to share the food, they may use a different relational model in the sharing itself. They might decide communally to allocate it equally. Or they might vote—an equality matching method—to distribute the food communally, with each person helping himself or herself, or they might allocate the food in such a way to benefit the group as a whole.

The bottom line, Fiske said, is that “if you know which relational model people use and how they implement it, you will have a very, very good chance to being able to coordinate effectively with them, and to understand their judgments and emotions, the motivations behind their actions, what actions constitute transgressions of their models, and how they sanction transgressions. So to coordinate, cooperate, and engage with people in any culture, you need to discover what relational model they are using, how they implement the model, and then you have to invoke that model, make people feel committed to it, and you have to commit yourself to it.”

People everywhere give shape and meaning to their life as they speak about it. These ways of speaking structure each person’s sense of who they are (their identity), how they act, how they feel about what they do, and how they dwell in places. According to Donal Carbaugh of the University of Massachusetts Amherst, the study of cultural discourses reveals the distinctive identities, actions, feelings, and local practices of people in places.

When people from two different cultures meet and converse, it is very easy for them to misunderstand each other because they come to the conversation with very different means and meanings for communicating. Carbaugh described a way of avoiding many of these misunderstandings through a systematic approach he refers to as cultural discourse analysis.²

As an example of the sorts of misunderstandings that can arise in cross-cultural conversations, Carbaugh provided an abridged transcript of a conversation that took place between a Nepali farmer and a worker from a nongovernmental organization (NGO) (see Box 4-1). After some initial pleasantries, the NGO worker suggests to the farmer that he try organic fertilizer on his fields. The worker asks the farmer what he knows about using organic fertilizer, but the farmer demurs, insisting that the worker should instruct him on how to use it. “What does an illiterate fellow like me know?” the farmer asks. “We only do things in an ad hoc way. You should tell. Educated and knowledgeable people like you know how to do things properly.”

It seems to be a straightforward conversation, but has the NGO worker really understood what the farmer is saying from the farmer’s point of view? Probably not, Carbaugh suggested. “One of the things

that the Nepali farmer knows is that, when interacting with outsiders, he needs to defer to them and their knowledge, as this is proper conduct in the farmer’s view. Also, the farmer may understand he needs to treat the NGO worker respectfully if he is going to get the resources that they have to give,” he explained. “What the NGO worker may not know is that the Nepali farmer has farmed these fields for decades, knows what works, has used organic fertilizers in the past, knows that they are expensive and he can’t afford them, even with the help of this agency, and so on.”

There is a whole system of subtexts underlying this conversation that “needs to be understood as a communicative dynamic that is active in that exchange between that farmer and that NGO worker,” Carbaugh said. “If we don’t understand that kind of dynamic, we are missing a lot of what is going on in these kinds of situations when we are trying to help and work with others.”

The key is to understand the ways people declare form and meanings in their lives through communication practices; people do this differently. It is important to understand what someone from another culture is saying and how they are saying it. “What are they doing and what does that mean? We can always supply meanings from our view, given our habitual ways of seeing and thinking, but we don’t always supply the meanings of the practices from the vantage point of those participants.” To understand the communicative means and meanings from the vantage point of the other people requires also learning to see oneself from the perspective of others.

To do that, Carbaugh and his colleagues have developed a five-phased theory and methodology—cultural discourse analysis—to communicate about matters from multiple points of view. “I want to mention at the outset that this perspective is based upon data analysis and modeling together over an extended period of time,” he said. “It is not something that an armchair theorist has just invented in order to come up with some flashy ideas.”

The goal of cultural discourse analysis is to “unveil the deep beliefs and values in local actions,” Carbaugh said. It is important not only to understand the ways in which people act but also to comprehend the meanings of those actions from the point of view of the people involved. It is also to be on the lookout for what Carbaugh called “slippages” or “miscommunication”—areas in which two different languages or two cultures do not match, creating the possibility of misunderstandings, misattribution of intent, and stereotypings.

Because cultural discourse is inherently an interactional concept—it arises from interactions among people rather than from individual actions—the data used in studying it must also be interactional. Researchers who study cultural discourse do so by observing interactional practices.

To illustrate how cultural discourse analysis can be used, Carbaugh described examples from two projects in which he has participated. “In every case,” he said, “we have to discover what the local communication practices are that people act through, that people use to express what they are about and who they are.” In discovering the local communicative means and their meanings, researchers must keep in mind that they—the researchers or helpers—also have their own local forms of cultural discourse in use. “Scientific and agency discourse is one member of this class,” he said. “It doesn’t stand outside or above the others.”

The first project Carbaugh described involved a research team from the United Nations Institute for Disarmament Research (UNIDR) working in northern Ghana. Carbaugh was a member of its research advisory group.³ Because the northern area of Ghana had seen a great deal of violence over the years, including the beheading of one chief, the research group was tasked to conduct a security needs assessment of the local area. Their research involved a good deal of team-based fieldwork in order to develop an understanding of the local cultural discourse.

“One of the things that we found is that in the Dagbani language in northern Ghana there is no term for security,” Carbaugh said. So that created a difficulty they had to overcome if they were going to do a proper “security” needs assessment. Assessors address the questions “What are you doing? What indeed are you assessing? This is the kind of problem that demonstrates slippages that have to be addressed.” Further research found that the Dagbani language does include the idea of “protection,” so it was possible to use that concept instead of “security” when talking with the local people about what they needed (for a detailed report of the process, see Miller and Rudnick, 2008).

After working with the local people for a while, the research team heard from some of them that, although development and humanitarian workers had been in the area for decades, no one had ever asked the local people what they thought, and they were grateful that this group had actually thought to ask. “Thank you,” they said. “We think you understand something.” Talking to the local people and understanding their means of communication, through their eyes, is the only way to know how to help in ways that matter to them, Carbaugh said.

Carbaugh’s second example arose from work done by Elena Vladimir (Khatskevich) Nuciforo, a doctoral candidate at the University of Massachusetts Amherst, who is working in her homeland of Russia to reduce

alcohol use (Carbaugh and Khatskevich, 2008). Examining health programs developed for use in American communities, she found that they often did not work effectively in other communities, even in some American communities different from the original ones for which the programs were designed. Nuciforo set out to determine how a health program designed specifically for Russians might differ from ones designed for Americans.

“She observed,” Carbaugh said, “that many of the questions and the campaigns were designed—through a popular American discourse—according to the idea that health is a matter of personal choice and personal behavior, that ultimately it is up to you what you do with your body, what you put in it. This is a very American thing to say. It is part of popular American discourse, centered as it is upon the self.” She also found that health is thought of as a physical and a biological state. In the United States, health is often “medicalized.”

“But this is not the way health is thought about elsewhere,” Carbaugh said. He has done fieldwork in Native American communities, and that is not how they speak traditionally of health. It is also not how Russians express health issues, and presenting them in this way in Russia would be likely to alienate those one wants to help.

So what is the Russian discourse with regard to health? “Health is a matter of emotional well-being,” Carbaugh said. “It is a matter of having positive social relationships. It is a matter of morality, doing what is proper and good with others to cultivate good social relationships with them.” Thus, any efforts to convince Russians to change their health behaviors should be framed from this perspective—as an issue of positive social relations, of emotions, of morality, and not solely as an issue of personal choice and human biology. The American and Russian codes for thinking and talking about health are indeed quite different.

In order to deal with the problems presented by differences in cultural discourse, Carbaugh has developed an approach that involves “listening and understanding rather than entering a scene with judgments about it or with solutions that might be misplaced.” His method, which is based on cultural and communication (including linguistic) research, relies on a dialogue between experts and local people who serve as part of the research team. “There is a two-way flow of information,” Carbaugh said, with multiple data points based upon rigorous observations and interviews. “What this procedure produces is a map of the local discursive terrain, that is, the system of communication practices that people use and the meaningfulness of them to those people.”

With an understanding of the local discourse, Carbaugh works with agency personnel in order to map out exactly what it is they are trying to do to better the lives of the local people. Then he looks for “potential

gaps” between the discourses of the agency and the local people and designs practices for the agency workers based on that analysis. “This is a different way of working,” he said, “because it is focused on discourse and on intensive qualitative and interpretive analyses about practices and contexts and the meaningfulness of those to people.”

“The outcomes that we target are enhanced effectiveness, better relations with people, local ownership of projects and programs that they help to create, that are their own and are therefore more sustainable. In this way, we were helping to build bridges between these different discourse communities.”