Patrick Manning | Interactions and Connections: Locating and Managing Historical Complexity | The History Teacher, 39.2

Interactions and Connections: Locating and Managing Historical Complexity^*

Patrick Manning
Northeastern University

MORE THAN EVER, history courses are crossing boundaries. For example, the college-level African Diaspora course I have been teaching is a case in point: it surveys interactions linking the African continent and its Atlantic diaspora over the past 500 years. Other examples of old and new boundary-crossing courses include surveys of world history, Western Civilization, and such thematic courses as environmental history and international relations. Courses in national history also partake of boundary-crossing. For instance, within United States history, courses which address multiculturalism, the American West, or interactions of the colonial era must cross boundaries. In teaching and scholarship, historians today are working to show students how to view the past as more than localized narratives, more than comparisons of isolated experiences.

Teaching at this breadth, however, brings problems of its own. In some cases, despite the hopes of the teacher, the available course materials and texts continue to organize the past into discrete localities and time periods—leaving students with most of the work in making connections across boundaries. In other instances, where course materials provide a rich array of interactions and perspectives, students may feel deluged by alternative interpretations and freeze up in the face of so many choices. At either pole, the difficulties presented by course materials may lead students in the direction of making oversimplified interpretations, neglecting the complexities that underlie the lessons of history. How can the teacher best convey complexity in history? How can we encourage students to embark on the voyage of discovery that will lead them to encounter the interactions in history? Once they embark on that exciting but perilous voyage, how can we enable them to steer safely between the Scylla of reductively oversimplified views of the past and the Charybdis of incoherently multivariate interpretations?

The most obvious problem in student interpretation of history is oversimplification.¹ Simple interactions—the impact of A upon B—may be adequate to explain some processes in history, but not others. Perhaps the term "impact" is sufficient to explain such processes as "the impact of Mongol armies on Central Asia" or "the impact of Captain Cook and his crew on the Pacific islands." On the other hand, the contact between European mariners and Pacific islanders may have involved more than just the "impact" of visitors on the islanders. ² If so—if Pacific encounters included complex elements of mutual influence or development of new traditions—we shall only recognize them if we have a terminology and a conceptualization enabling us to visualize complex interaction as well as simple impact. It is difficult to sustain a view of the complexity if we can't name it. Students, if they lack clear alternatives, are tempted to describe the past as a list of facts and interpret it as one thing after another, or simply to focus on their favorite part. The opposite extreme in student interpretation is incoherence: listing so many factors in history that none of them fall into any pattern or reveal any logic. Teachers sometimes receive student work containing overly long lists of factors contributing to a past development; more commonly, students hesitate before taking on such complexity and retreat to oversimplification. In both of these cases, what is missing from student interpretations is a clear interpretation of historical interactions.

In the several times I have taught the African Diaspora course at Northeastern University, I have had consistent difficulty in getting students to treat Africa as a recipient of and participant in Atlantic influences, and not just as a source of the people who populated the diaspora.³ The students, with their North American framework of interpretation, implicitly treated Africa as outside of the analysis: for instance, they saw the Atlantic slave trade from 1500 to 1850 as bringing an impact of West Africa on the Caribbean, and nothing more. In class presentations, I sought to show that the connection between the regions was more complex. The involuntary African settlers in the Caribbean interacted with Amerindians and Europeans, and developed new cultures based on their old ones. In addition, the simple fact of the departure of slaves changed population and society in Africa. Further, in the course of this interaction, the foods of the Caribbean—including peanuts, maize, and manioc—spread to West Africa. West African cooking styles continued to come to the Caribbean, and now included ingredients of Caribbean origin. Notions of racial hierarchy, developed in the Caribbean out of the population mix, spread to Africa. In outcome, the two regions each became more cosmopolitan and more hierarchical as a result of their connection. This approach reveals a historical complexity that is very different from the notion of the impact of West Africa on the Caribbean.

Students in my class were interested in such an interpretation but had difficulty in reproducing or extending it. After a particularly disappointing 2003 class (in which students both white and black were more interested in the multicultural experience of American hip-hop than in broader connections across the diaspora), I resolved to formalize the notion of historical connections. I had been telling students about the importance of two-way connections in African Diaspora history and had given them examples, but had not given them the tools with which to construct their own examples. In the 2004 class I tried out the materials presented in this article, and found that the students not only adopted immediately the language of interactions and connections but also developed and debated sophisticated interpretations with these tools. Not every class will be as strong as the group I taught in 2004, but the experience has led me to conclude that explicitly modeling processes of interaction in history can strengthen students' ability to articulate the historical connections they see.

The result of my experience, presented here, is an analytical model for describing interactions and analyzing connections in history. The first objective of this model is a procedure to describe interactions in the past. This first step is entirely descriptive and focuses on identifying complexity. Students, rather than leaping immediately to conclusions and interpretative summaries, can gain a sense of the texture of history by lingering over descriptions and acknowledging the complexity of the past. Nonetheless, identifying complexity in the past brings up the problem of managing it: students still face the task of balancing the big picture and the specific details. My second objective in this study, therefore, is a procedure to analyze connections in the past. This second step involves managing complexity—selecting, out of all the possible interactions between historical situations, those which are most significant. The objective is that students should be able to analyze complexity in an orderly fashion, rather than be overcome by the chaos of too many causal factors. Each of these steps is developed through treatment of two historical situations in interaction with one another. My third objective is to extend these procedures for description and analysis to the interaction of more situations: the third step presents techniques for global description and analysis.

Of course I do not claim that setting forth these objectives is sufficient to solve the problem of how to teach about complexity in the past without losing clarity of interpretation. I do, however, claim to have found that offering students a statement of this problem and a terminology for engaging it can result in direct and sometimes sophisticated student responses. To explore this claim, therefore, I suggest that the reader evaluate the remainder of this text on two levels: first, as a heuristic device for thinking about complexity in history and, second, as a set of specific guidelines for classroom activities. The heuristic presentation is the principal emphasis of the present article. On this level, I invite the reader to assess the logic of the categories of describing interactions, analyzing connections, and conducting global descriptions and analyses. Is it logically consistent and relevant to students' interpretation of the past? At the level of practical teaching guidelines, the reader may consider whether the specifics of this reasoning can be applied usefully in classroom work. To document the practical value of this approach, in the concluding section of this article I offer an extended example of student interpretation and reference to some associated student activities.

As a straightforward metaphor for the logic of historical interactions and connections, within the experience of most of us, I have encouraged students to think about describing the relationships of siblings. Each sibling is born with his or her own characteristics, but becomes different in interaction with brothers and sisters. In describing the interactions among siblings, one may observe that they create games and stratagems to achieve dominance, or devices to win attention of parents, or decisions to remodel one's self and become less like the other. The resulting description is complex, but shows that siblings commonly become different through interaction with each other.

Extending the metaphor to the analysis of connections, one may seek to explain the siblings' differentiation. It may help to make the analysis more global, extending it beyond the siblings to include their parents. For instance, one might choose to focus the interpretation on the issue of sibling order. The resulting interpretation might highlight the expectation that the older sibling is more secure and the younger sibling is more rambunctious. In this case, making the analysis more global provides a way to simplify the interpretation, turning the appearance of complex sibling rivalry into patterns of behavior by sibling order.⁴

The core of this article is organized into three sections or steps, which focus in turn on describing interactions among pairs of situations, analyzing connections among pairs of situations, and extending to a global level the logic of describing interactions and analyzing connections. Each of these "steps" gives a condensed description of the procedure, structured so that it might be used as a handout for students. Throughout, the basic purpose of my argument is to show that students can benefit from learning how to identify and explicate what I will call "two-way interactions." If students are able to understand and document two-way interactions in analyzing human society, they are prepared to connect directly to the complexity of the past. A second and underlying purpose of this presentation, meanwhile, is to provide students with some initial guidelines for analyzing connections in history, peering into the complexities of the past to highlight some of the logic of historical change. While the details of my argument were prepared with a college-level student audience in mind, I believe that the argument is relevant for high-school student audiences as well.

Step 1. Describing Interactions: Influences and Outcomes

This section summarizes techniques for identifying interactions in history. It begins with definitions of aspects of the past that may be observed and described through the historical record. Then it applies the terms as defined: the result helps to create a description of interactions and complexity in the past. At this stage, the procedure includes almost no assumptions about cause and effect in history.

Defining situations, influences, and outcomes. By a historical situation is meant a region, a community, or some other definable unit at a given time.⁵ Each situation (which will be labeled A or B) may be assumed to be static and unchanging, or it may be assumed to be evolving and undergoing its internal processes of change and development. In addition, a situation may send influences to other situations and receive influences from other situations. The influences could consist of people, material objects, ideas, or practices of any sort. Further, each situation may be described in terms of an outcome, a description of its characteristics as they change over time.

Influences: Identification and Comparison. The first stage in description of historical interactions is to list the directions in which influences flow among situations.

Case 1. For the basic one-way influence, there are two situations and a flow of influence from one to the other. Basically, this is "the impact of" A upon B (Figure 1). A and B can be assumed either to be static or in evolution: it is best if the assumption on this point is stated explicitly. The basic description of a one-way influence is the identification of the two situations and labeling the influence passing from one to the other. This prepares the way for discussing the resulting change in the second situation.

Figure 1. One-way interaction

An example of one-way influence, for the history of the African diaspora, is the influence of African slaves on the Caribbean (the contribution of their labor and the effects of their culture), combined with the assumption (probably implicit) that Africa experienced no change as a result of slave trade.⁶

Case 2. In a basic two-way influence, each of two historical situations sends influences to the other and receives influences from the other (Figure 2). A and B can be assumed either to be static or in change: as before, it is best if the assumption on this point is stated explicitly. The basic description of a two-way influence is the identification of the two situations and labeling the influence passing from each to the other. This prepares the way for discussing the resulting change in both situations.

Figure 2. Two-way interaction

In describing a two-way interaction, as one traces influence from A to B, one should check for the possibility that there was also influence from B to A. It can be labeled simply as "mutual influence." As such it provides a much more general model of interaction than "impact of." This mutual influence can be complicated or simplified to make sense of various historical situations.

The example of two-way influence in the African diaspora shows slaves and their music and religion moving to the Caribbean, and foods and ideas of racial hierarchy moving to West Africa. In cultural connections of more recent times, musical styles have moved to Africa (Cuban music and reggae), and African styles of dress have moved to the Americas (Kente cloth and embroidered shirts).⁷ In this description, culture on each side of the Atlantic has changed.

Outcomes: Identification and Comparison. The second part of a description of historical interaction is to focus on the outcomes in various situations, assuming these outcomes to be related in some way to the influences flowing among them.

Case 3. Identifying a basic parallel is to say that the outcomes in situations A and B are similar to each other, or that they are changing in a similar fashion. Such a statement does not assume that A and B are connected. Description of a simple parallel is the demonstration that two situations are similar in some respect or have changed in similar ways. This sets the stage for explaining the similarity. As elsewhere, one may assume that situations A and B are static or in change. That is, a situation may be influenced and changed by its own past, and not only by external factors.

An example of a parallel in religious history is the remarkable similarity in the pantheon of gods of the Yoruba- and Gbe-speaking peoples of West Africa in recent times with the Greek and Roman gods of classical times. In each case, a supreme god, plus gods of thunder, war, hunting, love, and others live as a supernatural community in periodic contact with humans.⁸ These situations are separated by more than a millennium; we know that the outcomes were similar, but debate continues on whether the two cases were historically connected.

Case 4. Identifying a basic divergence is to say that the outcomes in situations A and B are different from each other, or that they are changing in different directions. This instance is similar to the logic of "parallels," except that outcomes for A and B are different rather than similar. This identification sets the stage for explaining the divergence. The divergence of A and B may be because they are out of contact and influenced by different factors. But it could also be that A has influenced B and has brought B to become different.

An example of a divergence is that West Africa and North America, in the five centuries in which they have been in contact, have gone from a rough parity in their levels of wealth to a situation in which West Africa is in poverty and North America has become wealthy.⁹ Certainly there is a divergence in outcomes between the two situations, but it is not necessarily the case that the divergence was a result of the two regions' connections.

Internal Complications to Influence and Outcome. The categories of influences and outcomes listed above are the most basic. Beyond these starting points, we will see that the basic influences and outcomes may each include further complications—that is, further flows of influence and possible outcomes that result from the influences. Within a given situation, two major such complications are the autonomous evolution of a situation (local change without external contacts) and what I will call the sending effect (the change in a situation as a result of its having sent influence elsewhere). These are two types of influence, each of which can contribute to outcomes.

Case 5. For the one-way model of Case 1, the internal complications are of two types. First, the autonomous evolution of situation A brings change in A. Second, the sending effect of the departure of influences from A brings change in A (Figure 3).

Figure 3. Internal complications: one-way model

An example of such an internal complication is the case of a region sending migrants off to another region. Even if the migrants are never heard from again, their departure changes the shape of remaining families and may change the ideas and practices of the home population.¹⁰

Case 6. Analogously, for the two-way model of Case 2, there are internal complications for both situations A and B. First, the autonomous evolution of situation A brings change in A. Second, the sending effect of the departure of influences from A brings change in A (Figure 4). (Similarly, the evolution of B brings change in B, and the departure of influences from B brings changes in B). The influence flowing from A to B may thus encounter a dynamic situation in B; and vice versa. The difference of the two-way model is simply that one considers complications in at least two situations, not just one.

Figure 4. Internal complications: two-way model

For instance, the export of art work from Benin to Europe brings money to Benin, but this creation of work for a new market may also have led to change in domestic artistic standards in Benin. Similarly, the European purchasers may find that the act of buying imported art work brings not only prestigious imported goods but also a change to their own artistic standards.¹¹

External Complications to Influence and Outcome. This set of complications arises from connections among situations. In each case, it involves adding a third situation (C) to the dynamic of the flow of influences among A and B. (This is equivalent to adding a parent to the description of interactions among siblings.)

Case 7. The external complications to the one-way model of Case 1 arise as a third situation C sends a flow of influence to A, B, or both (Figure 5). For the first such complication, the basic description is the influence of C on B, as well as the influence of A on B. For the second such complication, the basic description is the influence of C on A, and a description of how it changes the influence of A on B. As always in the one-way model, the focus of attention is on the influences reaching situation B.

Figure 5. External complications: one-way model

External complications can be illustrated through the case of Captain Cook (A) and Tahiti (B), where the French (C) provide the complication.¹² In the figure at left above, the French provide a complication by landing at Tahiti and influencing the Tahitians directly. In the figure at right, the French complication is indirect: Captain Cook, knowing that the French may also come to Tahiti, behaves differently toward the Tahitians.

Case 8. For the two-way model of Case 2, the external complications all involve a third situation, C, and a more complete catalogue of its possible influences. First, C sends influence to B, where it adds to the incoming influence from A. Second, C sends influence to A, where it adds to the incoming influence from B. Third, C may send influence both to A and B, thus adding to the influences going in both directions for both A and B (Figure 6).¹³

Figure 6. External complications: two-way model

To explore a musical example: Cuban rumba of the 1930s and 1940s had great influence in Central Africa, and Cuban music drew in turn on drumming practices from Central Africa.¹⁴ A new influence was the rise of the electric guitar, first in the United States. It may be that the influence of the electric guitar went first to Cuba and then to Central Africa, or to Central Africa and from there to Cuba, or directly from the United States to both Cuba and Central Africa. In any case, the Central African music of today relies far more heavily on the electric guitar than does Cuban music.

Interplay among Influences. A further type of complication is the interplay among various types of influences as they produce an outcome. The details of interplay in any situation turn out to be the same for both one-way and two-way influences, because interplay is described within a single situation. For instance, in situation B (Figure 7), the four possible sources of influence include: internal evolution of B, the sending effects in B of sending influence from B to A, the influence received in B from A, and influence received in B from C. These four sources and types of influence may bring a complex interplay yielding an unexpected outcome in B.

Case 9. In situation A, the equivalent four possible sources of influence include: internal evolution of A, the "sending effects" in A of sending influence from A to B, the influence received in A from B, and the influence received in A from C (Figure 7). The full exploration of interplay among influences could be rich indeed, as there could be many types of interplay, of both internal and external influences.

Figure 7. Interplay among influences: two-way model

For an example of interplay among influences, we can consider changes in cooking that developed along the coasts of West and Central Africa during the eighteenth century.¹⁵ The underlying influences included new food crops arriving from the Americas (including maize, manioc, peanuts, and pineapple); the settlement of European merchants along the coast to conduct a slave trade; and the displacement of populations brought by the slave trade. Some African men became cooks for the European visitors, and learned to cook in European styles. African women cooked for African households (including those of the male cooks), and incorporated both the new crops and the new styles into their cooking.

Step II. Analyzing Connections: Interpreting Paired Situations

In this step we assume that interactions among a set of historical situations have been described, and turn to the related task of analyzing connections among them. Beginning with some definitions, the section turns to the steps of analyzing influences, outcomes, processes, and linking them together in a statement of historical connections.

Defining processes and connections. This step begins with the same definitions as in Step I. In addition, this step defines processes as patterns of interaction among influences and outcomes. Included among processes are some that are already defined (such as diffusion, syncretism, creolization, and equilibrium), but the historian may also define and label new processes. Connections, finally, are defined as a summary statement of the important influences, processes, and outcomes for a set of historical situations. That is, the statement of connections is an interpretation of the key influences, processes, and outcomes in a historical encounter.

Focus on influences: Prioritizing past events. The point here is to review the list of influences linking situations and to prioritize then. In contrast to the previous step of describing influences, which is straightforward and systematic, this task depends fundamentally on judgment and selection. The simple rule, when looking for connections, is to prioritize those influences which seem most connected even when they do not at first seem the most important factors.

Focus on outcomes: Explaining parallels and divergences. Similarly, of the variety of outcomes described for each situation, the task here is to prioritize them and select the most important. Then, in comparing outcomes of the situations under study, the task is to identify and explain the parallels and divergences among the situations. In so doing, one makes the judgment on whether the parallels or divergences are connected.

Approach 1. Parallels: connected or unconnected. Where situations A and B have similar outcomes or change in similar ways, they are "parallel." Having determined this, one explores the influences flowing between A and B to determine whether they are significant enough to label the parallel as "connected." To confirm that they are connected, one must be able to show how the influences and outcomes are linked by a process, as described below.

Approach 2. Divergences: connected or unconnected. Where A and B have different outcomes or change in different ways, they are "divergent." Then the analyst explores the influences flowing between A and B and the process linking them (see below) to determine whether the divergence can be labeled as "connected."

Too often it is forgotten that a divergence can result from a connection. Another issue in interpretation is the place of internal evolution of situations A and B in an overall interpretation. If two situations end up with parallel outcomes, it might be because they had the same internal process of evolution, or were affected by the same external influences, or both. In any case it is important to ask whether a situation is unchanging or evolving on its own.

Here are three examples of parallels and divergences from the African diaspora. First, influential movements of evangelical Christianity arose in Central Africa and in the American South (through the Kimbanguist Church and the Watchtower Movement) in the early twentieth century: these were parallel outcomes.¹⁶ The cause of these parallel outcomes is open to question: did they result from connections between the two situations or from independent influences? Second, one may note that racial categories in the United States emphasize a dichotomy between white and black, while racial categories in Brazil emphasize many gradations of color: this is a divergent outcome.¹⁷ Is the divergence disconnected? It is often explained as disconnected, arising from the different English and Portuguese styles of colonization. But it could also be argued that the two cases were connected by the institution of slavery: one then needs to describe the process by which slavery led to one racial order in the United States and another in Brazil. Third, a simpler instance of "connected divergence" is the case of contrasting sex ratios in Africa and in the Americas in the era of slave trade. African adult populations became dominantly female because of the loss of males to an overseas slave trade, while African-based populations in the Americas were dominated by African-born male slaves.¹⁸ The connection of the slave trade lay at the root of this divergence.

Focus on process: Modeling. To link influence and outcome, the analyst must articulate a "process" or "model" that explains how influences lead to outcomes. Such a model must explain how influences and outcomes interact to bring about change in historical situations—or indeed, to reproduce an unchanged situation for a later time. We can distinguish two basic categories of explanations. In cause-and-effect explanations, some factors are seen as causal and others are seen as consequences. This is the simplest and most common type of explanation in science and in social science. In feedback explanations, any factor can be seen as both cause and consequence. Feedback explanations are becoming more common in studies of complex systems. Here are further distinctions between these approaches:

Approach 3. Cause-and-effect explanations (one-way explanations). If the strength of the Mongol army is seen as the cause of Mongol conquests of Eurasia, this is a one-way, cause-and-effect explanation.¹⁹ (It is thus parallel to a one-way interaction.) That is, for a pair of situations, one aspect of a situation is seen as the cause and another aspect of either situation is seen as the effect.

Approach 4. Feedback explanations (two-way explanations). These are two-way explanations, and are parallel to two-way influences.

As a straightforward example of feedback in migration in the Atlantic slave trade, early English planters in South Carolina found that rice would be a good crop to grow in the wetlands, because it could be sold to Barbados and elsewhere as provisions. Slaves born in Sierra Leone had grown rice at home, and enabled rice cultivation to prosper. As a result, the demand for Sierra Leone slaves grew in South Carolina, and with it rice output grew.²⁰ Each region was changed through feedback from the other.

The work of modeling processes in history is open-ended, so that the analyst may decide on the most appropriate of several approaches. Here are four more options. First, one may adopt any of several processes or dynamics which have already been developed and labeled: these include such processes as diffusion, syncretism, fusion, creolization, melting pot, assimilation, and survival. Second, one can reverse the direction of influence, such as by observing that an outcome may generate influence rather than simply be the result of influences. For instance, the development of new religious systems such as Vodun and Santería in the Caribbean (an outcome) eventually brought their spread to other regions (an influence). Third, one can reverse the direction of attack on the problem: I find that students sometimes develop successful interpretations by working back from an assumed change to a hypothesized cause. For instance, if nationalist sentiment is assumed to have risen throughout the African diaspora in the twentieth century, one may ask whether Pan-African activists or the success of European nations are best seen as the cause.²¹ In this case an outcome tells us where to look for connections; changes tell us where to look for similarities. Fourth, there is the approach of brainstorming: comparing the situation under study to other cases in search of inspiration and new ideas.²²

Summarizing historical connections. The summary statement on historical connections is the objective of this overall exercise. A complete statement on historical connections should include a description of influences and outcomes as well as the process linking them. Too often, as in historical interpretations focusing on cultural diffusion or religious syncretism, these processes are invoked in a void, without any clear statement of the influence or outcome. Often, as in assertions of the diffusion of rock music or fast foods around the world, what is described is an apparently parallel outcome, tied to an unverified assumption that the outcome results from a one-way diffusion of influence from the United States.²³ The discussion above shows how many other possibilities must be considered before leaping to such an interpretive conclusion and labeling it as "diffusion."

Step III. Interactions and Connections at a Global Level

The discussion in Steps I and II has been at the comparative level, describing and interpreting situations one or two at a time. This third step goes beyond the comparative level of the two previous steps to address the global level—that is, usually, interpreting more than a pair of historical situations.²⁴ We should look for patterns and characteristics in a global context that cannot easily be seen by studying individual situations within it. I will focus here on the assumption that we view the world by aggregating observations of its parts (though it is also possible, in some circumstances, to observe the global situation directly).²⁵

Defining "the world." First we define the limits of the world we are exploring. It might be the combination of A and B and their interactions; or it could be the combination of A, B, and C (see examples of each in Figure 8).

Figure 8. Global influences

Describing global influences. The first point here is that describing influences globally requires identifying two-way influences rather than one-way influences. One must look for two-way influences to avoid misperceiving global connections.

Note how many influences must be considered when treating A, B, and C equally as parts of the world, as shown in Figure 8. Adding a fourth situation to the world increases the number of influences even more, and so forth. For this reason, while the student must consider all the possible influences, it is quite impossible to take them all into account. What is needed is a procedure for simplifying the full set of interactions, and selecting those which are most significant. Global analysis, therefore, requires criteria for selection and a procedure for aggregation. At the global level, because there are so many influences, outcomes, and processes, it is necessary to be selective in description as well as in analysis.

Here is a preliminary list of techniques for simplifying the long list of all possible influences. Most simply, one could select the most prominent single flow of influence. Similarly, one could select, along with the main flow of influence, the various flows and countercurrents associated with it. On the other hand one could assume that there were contrary flows of influence, and select two major flows to show their contrast and their interplay. Further, one could seek out examples of major interplay among influences as highlights to the analysis. All of these approaches involve neglecting large numbers of the patterns of connection. But at least, after such a review, one has selected the main element's influence and interplay based on a thorough review of the possibilities. I think the objective in world historical analysis is to get beyond the simplest single-factor approach yet still present a coherent and logical analysis.

Describing global outcomes. For outcome as for influence, it is necessary to use two-way approaches in order to catalogue the possibilities. To construct a view of a global outcome, one needs a procedure for selecting outcomes in the various situations, and a procedure to aggregating them into a picture of global outcome. Describing global outcomes does not include as many factors as describing global influences, but it is still important to include a balance of differences among situations, their parallels and divergences.

Modeling global processes. How is a global process different from a local process? The only real difference is the obvious one: a global process is bigger and more complex. When thinking in global terms, one must remember to look for more possibilities—interactions among parts of the system or "world," influences from larger-scale factors outside that world, and even linkage to small-scale factors within the system. Nonetheless, the actual process for a historical analysis is limited by the availability of data and by the reach of the human mind.

Analyzing global connections. How are influence and outcome linked? At the global level as at the comparative level, this is the most complex and most indeterminate aspect of the analysis of connections. Here are several examples of global interpretations of connections, some more oversimplified than others.

One example of oversimplified global analysis is the listing of a variety of outcomes of different situations, and the assertion that they add up to a pattern. With no explanation of how the situations were selected or how they relate to each other, this would be a weak statement indeed of global patterns: one needs to identify the process, not just the linked influences and outcomes. A slightly stronger approach could be based on a scattered set of one-way influences: such an interpretation would be of exaggerated significance if it did not take account of other possible influences. A step further could be sequential one-way influences: A impacts B, B impacts C, C impacts D. Such a chain of influence could be plausible, but is weakened if it ignores most other possibilities. On closer investigation it may be simply a repetition of diffusionist "impact-of" reasoning.

As an example of a global connection in the Atlantic world of the eighteenth century, here is an interpretation of the growth of patriarchy through interaction of regional situations. In Britain, we know the story of the development of factory production, the separation of male and female workers, and the development of the "family wage," which paid men higher for factory work, and restricted women increasingly to work in the home. This was one type of patriarchy. In the Americas of the eighteenth century, a system of slavery developed two sorts of patriarchy: the superiority of free people over slaves, and the control of white men over both free and slave women. In Africa, the eighteenth century brought large-scale export of slaves, especially males, and many women who remained were held in slavery, resulting in another development of patriarchy. These three regional systems of patriarchy were all distinctive, but they arose mainly because of the interconnection of the regions, involving the movement of European goods to Africa, African captives to the Americas, and American produce to Europe. The regular maritime connections among these regions surely reinforced the idea of patriarchy—that men were above women and that some men were above other men—but the particular forms of patriarchy differed in the three regions not only because of local habits but because of their place in a system of global connections.²⁶

Narratives of Interaction and Connection in History

Students who have been introduced to the steps above seem to me to give clearer statements of complexity in the past. To document this argument, I conclude with this discussion and assessment of presentations by students I have taught. Among the assignments in my courses on the African Diaspora are group presentations on various media in popular culture. Groups of two to five students prepare and deliver forty-minute presentations (often with PowerPoint) on such issues as literature, film, music, and architecture. They are urged to present a mix of materials on the African continent and the African diaspora, to identify interactions, and to address a significant range of historical time. These presentations, if successful, convey a mix of the continuities, specificities, transformations, and connections in the culture of Africa and the diaspora.

I thought the strongest presentation in my 2004 course was on cuisine. In it, four students presented cuisine of several areas of Africa and, in the diaspora, of the Caribbean and the American South.²⁷ The opening sections of their presentation corresponded to what I have labeled above as "describing interactions." They began by listing a range of the ingredients in African cuisine—starches (including yams, millet, maize, and African rice), proteins (fish, fowl, domestic and wild meat, beans), greens of several sorts, condiments (peppers, ginger, onion), oils (palm oil, peanut oil, sesame oil), fruits, nuts, vegetables, and more. Then they turned to methods of preparing main dishes. Commonly, to a starch base are added greens, protein, and condiments, often boiled as a stew, the starches, greens, and seasonings varying with regions of the continent. Starch known as foofoo in parts of West Africa is a porridge of beaten yam paste; other starches are millet or sorghum flour, maize paste, couscous made from wheat or millet, and manioc pounded as paste or heated to form granules of gari. Various preparations are boiled as stews, steamed, fried, roasted, or slow-roasted as barbeque: students gave quick summaries of distinctive aspects of cuisine in North Africa, Central Africa, East Africa, West Africa, and Southern Africa. Having surveyed characteristics of African cuisine, students turned to the movement of ingredients over time and space. They identified ingredients that have entered Africa—from South America (peanuts, cocoa, manioc), the Caribbean (chili peppers), the Mediterranean (olive oil), the Mid-East (fava beans), India (chutneys), Southeast Asia (bananas, coconuts). In opposite directions, they listed foods that have come out of Africa—black pepper went from Africa to Europe; okra, black-eyed peas, and various greens went to the Americas. In a more complex trajectory, they argued that peanuts moved from South America to western Africa, and then moved to North America along with enslaved migrants.

After the fifteen minutes of overview just summarized, the presenters moved on to the main part of their presentation, in which the students gave further details on cuisine of five regions of Africa and two regions of the Americas. This corresponded to what I would call "analyzing connections." The presentations emphasized the overall importance of marinades and spices in developing the taste of preparations, but showed the varying regional applications of these principles. For Northeast Africa, ancient traditions of bee keeping and herding cattle and sheep were supplemented by the arrival of cornbread from the north, garlic from Syria, banana paste from the east, and later by influences from France, England, and Italy. In West Africa, cocoa, bananas and oranges arrived from Brazil, lemon marinade arrived from France and combined with chicken in Senegal, and the Yam Festival continued as before. For Southern Africa, a main dish of Madagascar—chicken in coconut milk, served over rice with lemon juice, pepper, tomato, onions, and ginger—was in some ways the equivalent of a Zambian dish, Nshima, of maize and cassava served with a sauce of peanut paste and greens. In each case, a real African meal was made of ingredients that had been brought to the continent over time. For the Caribbean, students argued that Arawak and African traditions of barbecue combined to produce jerk (marinated, roasted meat); in the nineteenth century, ingredients from China and India joined the region's cuisine and European migrants modified the food of Cuba. In the American South, the fine line drawn between Cajun and Creole was partly that between country and city. In slave households, potatoes took the place of yams, while collard and mustard greens replaced African greens; gumbo (a stew) and barbeque retained their importance. A post-emancipation tradition of large Sunday dinners sustained the big meals of earlier times, and this tradition became celebrated as Soul Food.

I thought that this group of students worked skillfully to include a great deal of specific information in a framework that conveyed the character of cuisine and the connections among regional food-ways. The presentation treated Africa and the African diaspora as a global unit, and considered interactions among its subunits and influences from beyond it. The students showed substantial change over time, though rarely with specific chronology. They addressed 500 years of interaction over large areas, so it was hard for them to pin down the order of changes. They basically treated ingredients as influences and cuisine as outcomes; they showed both parallels and divergences in regional cuisine. They responded to my question about whether they had shown examples of one-way or two-way interactions, especially two-way interactions over time. They made some errors on directions of movements—thinking bananas and oranges moved from Brazil to Africa, when the direction was more likely the opposite. The arrival of new ingredients had presented endless external complication for them and they hinted at internal complications through the interplay of ingredients.

Overall, I was pleased with their work in identifying and managing complexity in the history of cuisine. They had conveyed the sense of a global culinary system with regional subsystems, connected by many feedback loops in ingredients and methods of preparation. As a bonus, they succeeded in conveying how it is that people can come to feel very strongly about the uniqueness of their local cuisine, even when most of its materials and techniques have been borrowed. For myself, I felt able to taste the meals they described.

In a second presentation, two students focused on literature, with one of them emphasizing literature of Western Africa and the other centering on black authors of North America. In each case, they emphasized the substantial grounding in oral literature on which the written texts relied. They each noted that written works expanded in quantity especially in the late nineteenth century. Rather than a single dominant emphasis within each literary tradition—though religious themes were often prominent—the students noted the numerous shifts in outlook and genre within each tradition. Writers of both Western Africa and North America gained wide audiences in the late twentieth century, and this led to a Nobel Prize for literature on each side of the ocean—for Wole Soyinka of Nigeria and Toni Morrison of the United States. Insightful as this presentation was, I found it to rely more on comparisons than on interactive descriptions. Their comparison did, however, reveal the parallels in the development of African and African-American literature. African-American literature in English began earlier than in Africa and sustained a wider readership, though the reverse may have been true for literature in French. In response, I was able to use the language of my pedagogical framework to provide a critique of this presentation, pointing out its comparative structure and asking what interactions among literary trends they might have been able to discover. Meanwhile, presentations by other groups of students addressed the culture of the African Diaspora through the issues of religion, visual art, and textiles and dress.

In order for students to develop facility with the logic of interactions and connections in history, I believe they will need practice. For this reason I have developed a set of exercises for students to work through.²⁸ Instructors wishing to rehearse students in this reasoning should be able to develop additional exercises with ease.

While this essay has addressed a wide range of points in historical description and analysis, I want to end as I began with two basic points. My principal purpose in this study has been to strengthen the tools with which students and teachers work to get beyond oversimplified, diffusionist thinking in interpreting the past. Students at all levels should be able to recognize and describe complexity in a historical connection when they see it. Students, textbooks, and scholars too often assume one-way influences and connections, when they ought to be looking for two-way influences and connections.

My second purpose has been to show ways to manage the interpretive complexity that comes from considering the details of historical connections. The study of "connections" includes exploration of the whole complex of influences and outcomes in linked situations. Analysis using these considerations should help locate the connections that have been most significant in any historical situation. The point is that the analyst should be aware of the possible connections, and know how to look for them. In a world-historical analysis it is necessary and desirable to simplify, but one should first explore comprehensively the possible relationships, then set criteria and choose what appear to be the best simplifications.

Notes

^* The author expresses appreciation to Tiffany Trimmer, Sharon C. Cohen, and three anonymous reviewers of this journal for helpful comments on previous versions of this essay.

^1. This problem is inherited from the earlier tradition of discrete local or national histories, but it is reinforced by the desire to identify a clear narrative in border-crossing histories.

^2. Jack Weatherford, Genghis Khan and the Making of the Modern World (New York: Crown Publishing, 2004); I.C. Campbell, "The Culture of Culture Contact: Refractions from Polynesia," Journal of World History 14 (2003), 63–95.

^3. I taught the course in 2000, 2001, 2003, and 2004; it has been cross-listed in the departments of History and African-American Studies (meaning students can take credit in either department). Students have ranged from freshmen to seniors, though most have been in the last half of their undergraduate programs; just over ten percent of them have been majors in History or African-American Studies. For course syllabi, see the teaching section of the author's website at www.worldhistorynetwork.org/manning.

^4. The metaphor can be extended in various directions, if one wishes, to make additional points about interactions and how to study them.

^5. The advantage of using the term "situation," I find, is that it tends to imply a restricted time period as well as a limited space. Equivalent terms that might also be used in place of "situation" include: "case," "entity," "place," and "phenomenon."

^6. For an interpretation of the African diaspora applying this sort of approach to cultural issues, see Robert Farris Thompson, Flash of the Spirit: African and Afro-American Art and Philosophy (New York: Vintage, 1984).

^7. John Storm Roberts, Black Music of Two Worlds, 2^nd ed. (Belmont, CA: Wadsworth, 2003).

^8. Geoffrey Parrinder, African Traditional Religion (Westport, CT, 1976).

^9. Joseph E. Inikori, Africans and the Industrial Revolution in England: A Study in International Trade and Economic Development (New York: Cambridge University Press, 2002).

^10. Sylviane Diouf, ed., Fighting the Slave Trade: West African Strategies (Athens, OH: Ohio University Press, 2003).

^11. On African art works and the market for them, see Frank Willett, African Art, 3^rd ed. (New York: Thames and Hudson, 2002); Sidney Littlefield Kasfir, Contemporary African Art (New York: Thames and Hudson, 1999); and Richard J. Powell, Black Art: A Cultural History, 2^nd ed. (New York: Thames and Hudson, 2003).

^12. Campbell, "Culture of Culture Contact."

^13. We could also consider the influences of A and B on C, but if our analysis began as a study of A and B, we need only consider the influences from C to complete that study.

^14. On early Caribbean musical connections with Central Africa, see Phyllis Martin, Leisure and Society in Colonial Brazzaville (Cambridge: Cambridge University Press, 1995); and John Storm Roberts, Black Music of Two Worlds: African, Caribbean, Latin, and African-American Traditions, 2^nd ed. (New York: Schirmer Books, 1998).

^15. For print materials, works in two series provide good information on cuisine that facilitates comparisons. In the "culture and customs" series, see Oyekan Owomoyela, Culture and Customs of Zimbabwe (Westport, CT: Greenwood Press, 2002); John Mukum Mbaku, Culture and Customs of Cameroon (Westport, CT: Greenwood Press, 2005). The Culinaria series includes a spectacular study of Caribbean cuisine: Rosemary Parkinson, Culinaria the Caribbean: A Culinary Discovery (Cologne: Könemann, 1999).

^16. Karen E. Fields, Revival and Rebellion in Colonial Central Africa (Princeton: Princeton University Press, 1985); Marie-Louise Martin, Kimbangu. An African Prophet and His Church, translated by D. M. Moore (Oxford: Basil Blackwell, 1975).

^17. Carl N. Degler, Neither Black nor White: Slavery and Race Relations in Brazil and the United States (Upper Saddle River, NJ: Prentice-Hall, 1971).

^18. Patrick Manning, Slavery and African Life: Occidental, Oriental, and African Slave Trades (Cambridge: Cambridge University Press, 1990).

^19. Weatherford, Genghis Khan; see also Carter Vaughn Findley, The Turks in World History (New York: Oxford University Press, 2005), 77–92.

^20. Judith A. Carney, Black Rice: The African Origins of Rice Production in the Americas (Cambridge, MA: Harvard University Press, 2002); Margaret Washington Creel, A Peculiar People: Slave Religion and Community Culture among the Gullah (New York: New York University Press, 1988); Peter H. Wood, Black Majority: Negroes in Colonial South Carolina from 1670 through the Stono Rebellion (New York: Norton, 1974).

^21. P. Olisanwuche Esedebe, Pan-Africanism: The Idea and the Movement, 1776–1991, 2^nd ed. (Washington, DC: Howard University Press, 1994).

^22. This is an approach that I have labeled elsewhere as "exploratory comparison." Manning, Navigating World History: Historians Create a Global Past (New York, 2003), 315–316.

^23. Benjamin R. Barber, Jihad vs. McWorld: Terrorism's Challenge to Democracy (New York: Ballantine, 1995).

^24. By "global level" I mean any level of analysis encompassing some collection of local situations. This "global" level could be defined within a region, or for the whole of the earth, or conceivably beyond its limits.

^25. Is it possible to describe the global situation all at once, and not treat it as the sum of parts? We view an organism as a whole rather than the sum of its parts, and a flock of birds as a whole rather than as a sum of individual birds. Can this take place in history? At the very least we can distinguish between instances where we observe the totality directly (as in "early humanity" or "civilization") and instances where we observe pieces and aggregate them.

^26. Patrick Manning, Migration in World History (London: Routledge, 2004), 137–138.

^27. The four students prepared this presentation over the course of two to three weeks, relying primarily on internet sites but also on print resources.

^28. "Exercises in Interactions and Connections," www.worldhistorynetwork.org/manning, "Teaching Materials."

Content in the History Cooperative database is intended for personal, noncommercial use only. You may not reproduce, publish, distribute, transmit, participate in the transfer or sale of, modify, create derivative works from, display, or in any way exploit the History Cooperative database in whole or in part without the written permission of the copyright holder.

Interactions and Connections: Locating and Managing Historical Complexity*